* [PATCH v13 00/23] Compile-time stack metadata validation
@ 2015-09-22 15:46 Josh Poimboeuf
2015-09-22 15:46 ` [PATCH v13 01/23] tools: Fix formatting of the "make -C tools" help message Josh Poimboeuf
` (23 more replies)
0 siblings, 24 replies; 38+ messages in thread
From: Josh Poimboeuf @ 2015-09-22 15:46 UTC (permalink / raw)
To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86
Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra,
Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen,
Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges,
Andrew Morton, David Vrabel, Konrad Rzeszutek Wilk,
Boris Ostrovsky, Jeremy Fitzhardinge, Chris Wright, Alok Kataria,
Rusty Russell, Herbert Xu, David S. Miller, Pavel Machek,
Rafael J. Wysocki, Len Brown, Matt Fleming
This is v13 of the compile-time stack metadata validation patch set,
along with proposed fixes for many of the warnings it found. It's based
on the tip/master branch.
v12 can be found here:
https://lkml.kernel.org/r/cover.1441749989.git.jpoimboe@redhat.com
For more information about the motivation behind this patch set, and
more details about what it does, see the patch 4 changelog and
tools/stacktool/Documentation/stack-validation.txt.
Patch 1 fixes the formatting for the "make -C tools" help message.
Patches 2-6 add stacktool and some related macros.
Patches 7-21 are some proposed fixes for several of the warnings
reported by stacktool. They've been compile-tested and boot-tested in a
VM, but I haven't attempted any meaningful testing for many of them.
Patches 22-23 are some additional warning fixes from Chris J Arges.
v13:
- LDFLAGS order fix from Chris J Arges
- new warning fix patches from Chris J Arges
- "--frame-pointer" -> "--check-frame-pointer"
v12:
- rename "stackvalidate" -> "stacktool"
- move from scripts/ to tools/:
- makefile rework
- make a copy of the x86 insn code (and warn if the code diverges)
- use tools/include/linux/list.h
- move warning macros to a new warn.h file
- change wording: "stack validation" -> "stack metadata validation"
v11:
- attempt to answer the "why" question better in the documentation and
commit message
- s/FP_SAVE/FRAME_BEGIN/ in documentation
v10:
- add scripts/mod to directory ignores
- remove circular dependencies for ignored objects which are built
before stackvalidate
- fix CONFIG_MODVERSIONS incompatibility
v9:
- rename FRAME/ENDFRAME -> FRAME_BEGIN/FRAME_END
- fix jump table issue for when the original instruction is a jump
- drop paravirt thunk alignment patch
- add maintainers to CC for proposed warning fixes
v8:
- add proposed fixes for warnings
- fix all memory leaks
- process ignores earlier and add more ignore checks
- always assume POPCNT alternative is enabled
- drop hweight inline asm fix
- drop __schedule() ignore patch
- change .Ltemp_\@ to .Lstackvalidate_ignore_\@ in asm macro
- fix CONFIG_* checks in asm macros
- add C versions of ignore macros and frame macros
- change ";" to "\n" in C macros
- add ifdef CONFIG_STACK_VALIDATION checks in C ignore macros
- use numbered label in C ignore macro
- add missing break in switch case statement in arch-x86.c
v7:
- sibling call support
- document proposed solution for inline asm() frame pointer issues
- say "kernel entry/exit" instead of "context switch"
- clarify the checking of switch statement jump tables
- discard __stackvalidate_ignore_* sections in linker script
- use .Ltemp_\@ to get a unique label instead of static 3-digit number
- change STACKVALIDATE_IGNORE_FUNC variable to a static
- move STACKVALIDATE_IGNORE_INSN to arch-specific .h file
v6:
- rename asmvalidate -> stackvalidate (again)
- gcc-generated object file support
- recursive branch state analysis
- external jump support
- fixup/exception table support
- jump label support
- switch statement jump table support
- added documentation
- detection of "noreturn" dead end functions
- added a Kbuild mechanism for skipping files and dirs
- moved frame pointer macros to arch/x86/include/asm/frame.h
- moved ignore macros to include/linux/stackvalidate.h
v5:
- stackvalidate -> asmvalidate
- frame pointers only required for non-leaf functions
- check for the use of the FP_SAVE/RESTORE macros instead of manually
analyzing code to detect frame pointer usage
- additional checks to ensure each function doesn't leave its boundaries
- make the macros simpler and more flexible
- support for analyzing ALTERNATIVE macros
- simplified the arch interfaces in scripts/asmvalidate/arch.h
- fixed some asmvalidate warnings
- rebased onto latest tip asm cleanups
- many more small changes
v4:
- Changed the default to CONFIG_STACK_VALIDATION=n, until all the asm
code can get cleaned up.
- Fixed a stackvalidate error path exit code issue found by Michal
Marek.
v3:
- Added a patch to make the push/pop CFI macros arch-independent, as
suggested by H. Peter Anvin
v2:
- Fixed memory leaks reported by Petr Mladek
Cc: linux-kernel@vger.kernel.org
Cc: live-patching@vger.kernel.org
Cc: Michal Marek <mmarek@suse.cz>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Pedro Alves <palves@redhat.com>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Bernd Petrovitsch <bernd@petrovitsch.priv.at>
Cc: Chris J Arges <chris.j.arges@canonical.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Chris J Arges (2):
x86/uaccess: Add stack frame output operand in get_user inline asm
x86/stacktool: Ignore head_$(BITS) files.
Josh Poimboeuf (21):
tools: Fix formatting of the "make -C tools" help message
x86/asm: Frame pointer macro cleanup
x86/asm: Add C versions of frame pointer macros
x86/stacktool: Compile-time stack metadata validation
x86/stacktool: Add file and directory ignores
x86/stacktool: Add ignore macros
x86/xen: Add stack frame dependency to hypercall inline asm calls
x86/paravirt: Add stack frame dependency to PVOP inline asm calls
x86/paravirt: Create a stack frame in PV_CALLEE_SAVE_REGS_THUNK
x86/amd: Set ELF function type for vide()
x86/reboot: Add ljmp instructions to stacktool whitelist
x86/xen: Add xen_cpuid() and xen_setup_gdt() to stacktool whitelists
x86/asm/crypto: Create stack frames in aesni-intel_asm.S
x86/asm/crypto: Move .Lbswap_mask data to .rodata section
x86/asm/crypto: Move jump_table to .rodata section
x86/asm/crypto: Create stack frames in clmul_ghash_mul/update()
x86/asm/entry: Create stack frames in thunk functions
x86/asm/acpi: Create a stack frame in do_suspend_lowlevel()
x86/asm: Create stack frames in rwsem functions
x86/asm/efi: Create a stack frame in efi_call()
x86/asm/power: Create stack frames in hibernate_asm_64.S
MAINTAINERS | 7 +
Makefile | 5 +-
arch/Kconfig | 6 +
arch/x86/Kconfig | 1 +
arch/x86/boot/Makefile | 3 +-
arch/x86/boot/compressed/Makefile | 3 +-
arch/x86/crypto/aesni-intel_asm.S | 21 +
arch/x86/crypto/crc32c-pcl-intel-asm_64.S | 8 +-
arch/x86/crypto/ghash-clmulni-intel_asm.S | 5 +
arch/x86/entry/thunk_64.S | 4 +
arch/x86/entry/vdso/Makefile | 5 +-
arch/x86/include/asm/frame.h | 51 +-
arch/x86/include/asm/paravirt.h | 9 +-
arch/x86/include/asm/paravirt_types.h | 18 +-
arch/x86/include/asm/stacktool.h | 45 +
arch/x86/include/asm/uaccess.h | 5 +-
arch/x86/include/asm/xen/hypercall.h | 5 +-
arch/x86/kernel/Makefile | 1 +
arch/x86/kernel/acpi/wakeup_64.S | 3 +
arch/x86/kernel/cpu/amd.c | 5 +-
arch/x86/kernel/reboot.c | 7 +-
arch/x86/kernel/vmlinux.lds.S | 5 +-
arch/x86/lib/rwsem.S | 11 +-
arch/x86/platform/efi/efi_stub_64.S | 3 +
arch/x86/power/hibernate_asm_64.S | 7 +
arch/x86/purgatory/Makefile | 2 +
arch/x86/realmode/Makefile | 4 +-
arch/x86/realmode/rm/Makefile | 3 +-
arch/x86/xen/enlighten.c | 4 +-
drivers/firmware/efi/libstub/Makefile | 1 +
include/linux/stacktool.h | 29 +
lib/Kconfig.debug | 12 +
scripts/Makefile.build | 37 +-
scripts/mod/Makefile | 2 +
tools/Makefile | 43 +-
tools/stacktool/.gitignore | 1 +
tools/stacktool/Build | 4 +
tools/stacktool/Documentation/stack-validation.txt | 336 +++++++
tools/stacktool/Makefile | 51 ++
tools/stacktool/arch.h | 44 +
tools/stacktool/arch/x86/Build | 12 +
tools/stacktool/arch/x86/decode.c | 163 ++++
.../stacktool/arch/x86/insn/gen-insn-attr-x86.awk | 387 ++++++++
tools/stacktool/arch/x86/insn/inat.c | 97 ++
tools/stacktool/arch/x86/insn/inat.h | 221 +++++
tools/stacktool/arch/x86/insn/inat_types.h | 29 +
tools/stacktool/arch/x86/insn/insn.c | 594 +++++++++++++
tools/stacktool/arch/x86/insn/insn.h | 201 +++++
tools/stacktool/arch/x86/insn/x86-opcode-map.txt | 984 +++++++++++++++++++++
tools/stacktool/elf.c | 403 +++++++++
tools/stacktool/elf.h | 79 ++
tools/stacktool/special.c | 199 +++++
tools/stacktool/special.h | 42 +
tools/stacktool/stacktool.c | 980 ++++++++++++++++++++
tools/stacktool/warn.h | 60 ++
55 files changed, 5196 insertions(+), 71 deletions(-)
create mode 100644 arch/x86/include/asm/stacktool.h
create mode 100644 include/linux/stacktool.h
create mode 100644 tools/stacktool/.gitignore
create mode 100644 tools/stacktool/Build
create mode 100644 tools/stacktool/Documentation/stack-validation.txt
create mode 100644 tools/stacktool/Makefile
create mode 100644 tools/stacktool/arch.h
create mode 100644 tools/stacktool/arch/x86/Build
create mode 100644 tools/stacktool/arch/x86/decode.c
create mode 100644 tools/stacktool/arch/x86/insn/gen-insn-attr-x86.awk
create mode 100644 tools/stacktool/arch/x86/insn/inat.c
create mode 100644 tools/stacktool/arch/x86/insn/inat.h
create mode 100644 tools/stacktool/arch/x86/insn/inat_types.h
create mode 100644 tools/stacktool/arch/x86/insn/insn.c
create mode 100644 tools/stacktool/arch/x86/insn/insn.h
create mode 100644 tools/stacktool/arch/x86/insn/x86-opcode-map.txt
create mode 100644 tools/stacktool/elf.c
create mode 100644 tools/stacktool/elf.h
create mode 100644 tools/stacktool/special.c
create mode 100644 tools/stacktool/special.h
create mode 100644 tools/stacktool/stacktool.c
create mode 100644 tools/stacktool/warn.h
--
2.4.3
^ permalink raw reply [flat|nested] 38+ messages in thread* [PATCH v13 01/23] tools: Fix formatting of the "make -C tools" help message 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf @ 2015-09-22 15:46 ` Josh Poimboeuf 2015-09-22 15:46 ` [PATCH v13 02/23] x86/asm: Frame pointer macro cleanup Josh Poimboeuf ` (22 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:46 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton Align the x86_energy_perf_policy line with the others and restore the original alphabetical sorting. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> --- tools/Makefile | 32 ++++++++++++++++---------------- 1 file changed, 16 insertions(+), 16 deletions(-) diff --git a/tools/Makefile b/tools/Makefile index d6f307d..c7400a6 100644 --- a/tools/Makefile +++ b/tools/Makefile @@ -8,23 +8,23 @@ include scripts/Makefile.include help: @echo 'Possible targets:' @echo '' - @echo ' acpi - ACPI tools' - @echo ' cgroup - cgroup tools' - @echo ' cpupower - a tool for all things x86 CPU power' - @echo ' firewire - the userspace part of nosy, an IEEE-1394 traffic sniffer' - @echo ' hv - tools used when in Hyper-V clients' - @echo ' iio - IIO tools' - @echo ' lguest - a minimal 32-bit x86 hypervisor' - @echo ' perf - Linux performance measurement and analysis tool' - @echo ' selftests - various kernel selftests' - @echo ' turbostat - Intel CPU idle stats and freq reporting tool' - @echo ' usb - USB testing tools' - @echo ' virtio - vhost test module' - @echo ' net - misc networking tools' - @echo ' vm - misc vm tools' + @echo ' acpi - ACPI tools' + @echo ' cgroup - cgroup tools' + @echo ' cpupower - a tool for all things x86 CPU power' + @echo ' firewire - the userspace part of nosy, an IEEE-1394 traffic sniffer' + @echo ' freefall - laptop accelerometer program for disk protection' + @echo ' hv - tools used when in Hyper-V clients' + @echo ' iio - IIO tools' + @echo ' lguest - a minimal 32-bit x86 hypervisor' + @echo ' net - misc networking tools' + @echo ' perf - Linux performance measurement and analysis tool' + @echo ' selftests - various kernel selftests' + @echo ' tmon - thermal monitoring and tuning tool' + @echo ' turbostat - Intel CPU idle stats and freq reporting tool' + @echo ' usb - USB testing tools' + @echo ' virtio - vhost test module' + @echo ' vm - misc vm tools' @echo ' x86_energy_perf_policy - Intel energy policy tool' - @echo ' tmon - thermal monitoring and tuning tool' - @echo ' freefall - laptop accelerometer program for disk protection' @echo '' @echo 'You can do:' @echo ' $$ make -C tools/ <tool>_install' -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 02/23] x86/asm: Frame pointer macro cleanup 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf 2015-09-22 15:46 ` [PATCH v13 01/23] tools: Fix formatting of the "make -C tools" help message Josh Poimboeuf @ 2015-09-22 15:46 ` Josh Poimboeuf 2015-09-22 15:46 ` [PATCH v13 03/23] x86/asm: Add C versions of frame pointer macros Josh Poimboeuf ` (21 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:46 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton The asm macros for setting up and restoring the frame pointer aren't currently being used. However, they will be needed soon to help asm functions to comply with stacktool. Rename FRAME/ENDFRAME to FRAME_BEGIN/FRAME_END for more symmetry. Also make the code more readable and improve the comments. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> --- arch/x86/include/asm/frame.h | 33 ++++++++++++++++++--------------- 1 file changed, 18 insertions(+), 15 deletions(-) diff --git a/arch/x86/include/asm/frame.h b/arch/x86/include/asm/frame.h index 793179c..8a6cd26 100644 --- a/arch/x86/include/asm/frame.h +++ b/arch/x86/include/asm/frame.h @@ -1,23 +1,26 @@ +#ifndef _ASM_X86_FRAME_H +#define _ASM_X86_FRAME_H + #ifdef __ASSEMBLY__ #include <asm/asm.h> -/* The annotation hides the frame from the unwinder and makes it look - like a ordinary ebp save/restore. This avoids some special cases for - frame pointer later */ +/* + * These are stack frame creation macros. They should be used by every + * callable non-leaf asm function to make kernel stack traces more reliable. + */ +.macro FRAME_BEGIN +#ifdef CONFIG_FRAME_POINTER + push %_ASM_BP + _ASM_MOV %_ASM_SP, %_ASM_BP +#endif +.endm + +.macro FRAME_END #ifdef CONFIG_FRAME_POINTER - .macro FRAME - __ASM_SIZE(push,) %__ASM_REG(bp) - __ASM_SIZE(mov) %__ASM_REG(sp), %__ASM_REG(bp) - .endm - .macro ENDFRAME - __ASM_SIZE(pop,) %__ASM_REG(bp) - .endm -#else - .macro FRAME - .endm - .macro ENDFRAME - .endm + pop %_ASM_BP #endif +.endm #endif /* __ASSEMBLY__ */ +#endif /* _ASM_X86_FRAME_H */ -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 03/23] x86/asm: Add C versions of frame pointer macros 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf 2015-09-22 15:46 ` [PATCH v13 01/23] tools: Fix formatting of the "make -C tools" help message Josh Poimboeuf 2015-09-22 15:46 ` [PATCH v13 02/23] x86/asm: Frame pointer macro cleanup Josh Poimboeuf @ 2015-09-22 15:46 ` Josh Poimboeuf 2015-09-22 15:46 ` [PATCH v13 04/23] x86/stacktool: Compile-time stack metadata validation Josh Poimboeuf ` (20 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:46 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton Add C versions of the frame pointer macros which can be used to create a stack frame in inline assembly. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> --- arch/x86/include/asm/frame.h | 20 ++++++++++++++++++-- 1 file changed, 18 insertions(+), 2 deletions(-) diff --git a/arch/x86/include/asm/frame.h b/arch/x86/include/asm/frame.h index 8a6cd26..9a30ec7 100644 --- a/arch/x86/include/asm/frame.h +++ b/arch/x86/include/asm/frame.h @@ -1,10 +1,10 @@ #ifndef _ASM_X86_FRAME_H #define _ASM_X86_FRAME_H -#ifdef __ASSEMBLY__ - #include <asm/asm.h> +#ifdef __ASSEMBLY__ + /* * These are stack frame creation macros. They should be used by every * callable non-leaf asm function to make kernel stack traces more reliable. @@ -22,5 +22,21 @@ #endif .endm +#else /* !__ASSEMBLY__ */ + +#ifdef CONFIG_FRAME_POINTER + +#define FRAME_BEGIN \ + "push %" _ASM_BP "\n" \ + _ASM_MOV "%" _ASM_SP ", %" _ASM_BP "\n" + +#define FRAME_END "pop %" _ASM_BP "\n" + +#else /* !CONFIG_FRAME_POINTER */ + +#define FRAME_BEGIN "" +#define FRAME_END "" + +#endif /* CONFIG_FRAME_POINTER */ #endif /* __ASSEMBLY__ */ #endif /* _ASM_X86_FRAME_H */ -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 04/23] x86/stacktool: Compile-time stack metadata validation 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (2 preceding siblings ...) 2015-09-22 15:46 ` [PATCH v13 03/23] x86/asm: Add C versions of frame pointer macros Josh Poimboeuf @ 2015-09-22 15:46 ` Josh Poimboeuf 2015-09-23 11:38 ` Jiri Slaby 2015-09-22 15:46 ` [PATCH v13 05/23] x86/stacktool: Add file and directory ignores Josh Poimboeuf ` (19 subsequent siblings) 23 siblings, 1 reply; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:46 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton This adds a CONFIG_STACK_VALIDATION option which enables a host tool named stacktool which runs at compile time. It analyzes every .o file and ensures the validity of its stack metadata. It enforces a set of rules on asm code and C inline assembly code so that stack traces can be reliable. For each function, it recursively follows all possible code paths and validates the correct frame pointer state at each instruction. It also follows code paths involving special sections, like .altinstructions, __jump_table, and __ex_table, which can add alternative execution paths to a given instruction (or set of instructions). Similarly, it knows how to follow switch statements, for which gcc sometimes uses jump tables. Here are some of the benefits of validating stack metadata: a) More reliable stack traces for frame pointer enabled kernels Frame pointers are used for debugging purposes. They allow runtime code and debug tools to be able to walk the stack to determine the chain of function call sites that led to the currently executing code. For some architectures, frame pointers are enabled by CONFIG_FRAME_POINTER. For some other architectures they may be required by the ABI (sometimes referred to as "backchain pointers"). For C code, gcc automatically generates instructions for setting up frame pointers when the -fno-omit-frame-pointer option is used. But for asm code, the frame setup instructions have to be written by hand, which most people don't do. So the end result is that CONFIG_FRAME_POINTER is honored for C code but not for most asm code. For stack traces based on frame pointers to be reliable, all functions which call other functions must first create a stack frame and update the frame pointer. If a first function doesn't properly create a stack frame before calling a second function, the *caller* of the first function will be skipped on the stack trace. For example, consider the following example backtrace with frame pointers enabled: [<ffffffff81812584>] dump_stack+0x4b/0x63 [<ffffffff812d6dc2>] cmdline_proc_show+0x12/0x30 [<ffffffff8127f568>] seq_read+0x108/0x3e0 [<ffffffff812cce62>] proc_reg_read+0x42/0x70 [<ffffffff81256197>] __vfs_read+0x37/0x100 [<ffffffff81256b16>] vfs_read+0x86/0x130 [<ffffffff81257898>] SyS_read+0x58/0xd0 [<ffffffff8181c1f2>] entry_SYSCALL_64_fastpath+0x12/0x76 It correctly shows that the caller of cmdline_proc_show() is seq_read(). If we remove the frame pointer logic from cmdline_proc_show() by replacing the frame pointer related instructions with nops, here's what it looks like instead: [<ffffffff81812584>] dump_stack+0x4b/0x63 [<ffffffff812d6dc2>] cmdline_proc_show+0x12/0x30 [<ffffffff812cce62>] proc_reg_read+0x42/0x70 [<ffffffff81256197>] __vfs_read+0x37/0x100 [<ffffffff81256b16>] vfs_read+0x86/0x130 [<ffffffff81257898>] SyS_read+0x58/0xd0 [<ffffffff8181c1f2>] entry_SYSCALL_64_fastpath+0x12/0x76 Notice that cmdline_proc_show()'s caller, seq_read(), has been skipped. Instead the stack trace seems to show that cmdline_proc_show() was called by proc_reg_read(). The benefit of stacktool here is that because it ensures that *all* functions honor CONFIG_FRAME_POINTER, no functions will ever[*] be skipped on a stack trace. [*] unless an interrupt or exception has occurred at the very beginning of a function before the stack frame has been created, or at the very end of the function after the stack frame has been destroyed. This is an inherent limitation of frame pointers. b) 100% reliable stack traces for DWARF enabled kernels This is not yet implemented. For more details about what is planned, see tools/stacktool/Documentation/stack-validation.txt. c) Higher live patching compatibility rate This is not yet implemented. For more details about what is planned, see tools/stacktool/Documentation/stack-validation.txt. To achieve the validation, stacktool enforces the following rules: 1. Each callable function must be annotated as such with the ELF function type. In asm code, this is typically done using the ENTRY/ENDPROC macros. If stacktool finds a return instruction outside of a function, it flags an error since that usually indicates callable code which should be annotated accordingly. This rule is needed so that stacktool can properly identify each callable function in order to analyze its stack metadata. 2. Conversely, each section of code which is *not* callable should *not* be annotated as an ELF function. The ENDPROC macro shouldn't be used in this case. This rule is needed so that stacktool can ignore non-callable code. Such code doesn't have to follow any of the other rules. 3. Each callable function which calls another function must have the correct frame pointer logic, if required by CONFIG_FRAME_POINTER or the architecture's back chain rules. This can by done in asm code with the FRAME_BEGIN/FRAME_END macros. This rule ensures that frame pointer based stack traces will work as designed. If function A doesn't create a stack frame before calling function B, the _caller_ of function A will be skipped on the stack trace. 4. Dynamic jumps and jumps to undefined symbols are only allowed if: a) the jump is part of a switch statement; or b) the jump matches sibling call semantics and the frame pointer has the same value it had on function entry. This rule is needed so that stacktool can reliably analyze all of a function's code paths. If a function jumps to code in another file, and it's not a sibling call, stacktool has no way to follow the jump because it only analyzes a single file at a time. 5. A callable function may not execute kernel entry/exit instructions. The only code which needs such instructions is kernel entry code, which shouldn't be be in callable functions anyway. This rule is just a sanity check to ensure that callable functions return normally. It currently only supports x86_64. I tried to make the code generic so that support for other architectures can hopefully be plugged in relatively easily. As a first step, CONFIG_STACK_VALIDATION is disabled by default, and all reported non-compliances result in warnings. Once we get them all cleaned up, we can change the default to CONFIG_STACK_VALIDATION=y and change the warnings to errors to keep the stack metadata clean. On my Lenovo laptop with a i7-4810MQ 4-core/8-thread CPU, building the kernel with it enabled adds about three seconds of total build time. It hasn't been optimized for performance yet, so there are probably some opportunities for better build performance. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> --- MAINTAINERS | 7 + Makefile | 5 +- arch/Kconfig | 6 + arch/x86/Kconfig | 1 + lib/Kconfig.debug | 12 + scripts/Makefile.build | 37 +- tools/Makefile | 11 +- tools/stacktool/.gitignore | 1 + tools/stacktool/Build | 4 + tools/stacktool/Documentation/stack-validation.txt | 336 +++++++ tools/stacktool/Makefile | 51 ++ tools/stacktool/arch.h | 44 + tools/stacktool/arch/x86/Build | 12 + tools/stacktool/arch/x86/decode.c | 163 ++++ .../stacktool/arch/x86/insn/gen-insn-attr-x86.awk | 387 ++++++++ tools/stacktool/arch/x86/insn/inat.c | 97 ++ tools/stacktool/arch/x86/insn/inat.h | 221 +++++ tools/stacktool/arch/x86/insn/inat_types.h | 29 + tools/stacktool/arch/x86/insn/insn.c | 594 +++++++++++++ tools/stacktool/arch/x86/insn/insn.h | 201 +++++ tools/stacktool/arch/x86/insn/x86-opcode-map.txt | 984 +++++++++++++++++++++ tools/stacktool/elf.c | 403 +++++++++ tools/stacktool/elf.h | 79 ++ tools/stacktool/special.c | 199 +++++ tools/stacktool/special.h | 42 + tools/stacktool/stacktool.c | 980 ++++++++++++++++++++ tools/stacktool/warn.h | 60 ++ 27 files changed, 4956 insertions(+), 10 deletions(-) create mode 100644 tools/stacktool/.gitignore create mode 100644 tools/stacktool/Build create mode 100644 tools/stacktool/Documentation/stack-validation.txt create mode 100644 tools/stacktool/Makefile create mode 100644 tools/stacktool/arch.h create mode 100644 tools/stacktool/arch/x86/Build create mode 100644 tools/stacktool/arch/x86/decode.c create mode 100644 tools/stacktool/arch/x86/insn/gen-insn-attr-x86.awk create mode 100644 tools/stacktool/arch/x86/insn/inat.c create mode 100644 tools/stacktool/arch/x86/insn/inat.h create mode 100644 tools/stacktool/arch/x86/insn/inat_types.h create mode 100644 tools/stacktool/arch/x86/insn/insn.c create mode 100644 tools/stacktool/arch/x86/insn/insn.h create mode 100644 tools/stacktool/arch/x86/insn/x86-opcode-map.txt create mode 100644 tools/stacktool/elf.c create mode 100644 tools/stacktool/elf.h create mode 100644 tools/stacktool/special.c create mode 100644 tools/stacktool/special.h create mode 100644 tools/stacktool/stacktool.c create mode 100644 tools/stacktool/warn.h diff --git a/MAINTAINERS b/MAINTAINERS index ea3e3ae..a128801 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -9879,6 +9879,13 @@ L: stable@vger.kernel.org S: Supported F: Documentation/stable_kernel_rules.txt +STACK METADATA VALIDATION +M: Josh Poimboeuf <jpoimboe@redhat.com> +S: Supported +F: tools/stacktool/ +F: include/linux/stacktool.h +F: arch/x86/include/asm/stacktool.h + STAGING SUBSYSTEM M: Greg Kroah-Hartman <gregkh@linuxfoundation.org> T: git git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging.git diff --git a/Makefile b/Makefile index 84f4b31..0b9a1ca 100644 --- a/Makefile +++ b/Makefile @@ -984,7 +984,10 @@ prepare0: archprepare FORCE $(Q)$(MAKE) $(build)=. # All the preparing.. -prepare: prepare0 +prepare: prepare0 prepare-stacktool + +PHONY += prepare-stacktool +prepare-stacktool: $(if $(CONFIG_STACK_VALIDATION), tools/stacktool FORCE) # Generate some files # --------------------------------------------------------------------------- diff --git a/arch/Kconfig b/arch/Kconfig index 671810c..b20f472 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -527,6 +527,12 @@ config HAVE_COPY_THREAD_TLS normal C parameter passing, rather than extracting the syscall argument from pt_regs. +config HAVE_STACK_VALIDATION + bool + help + Architecture supports the stacktool host tool, which adds + compile-time stack metadata validation. + # # ABI hall of shame # diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index d8546d4..006b4af 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -153,6 +153,7 @@ config X86 select VIRT_TO_BUS select X86_DEV_DMA_OPS if X86_64 select X86_FEATURE_NAMES if PROC_FS + select HAVE_STACK_VALIDATION if X86_64 config INSTRUCTION_DECODER def_bool y diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index ab76b99..24d2e88 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -332,6 +332,18 @@ config FRAME_POINTER larger and slower, but it gives very useful debugging information in case of kernel bugs. (precise oopses/stacktraces/warnings) +config STACK_VALIDATION + bool "Enable compile-time stack metadata validation" + depends on HAVE_STACK_VALIDATION + default n + help + Add compile-time checks to validate stack metadata, including frame + pointers (if CONFIG_FRAME_POINTER is enabled). This helps ensure + that runtime stack traces are more reliable. + + For more information, see + tools/stacktool/Documentation/stack-validation.txt. + config DEBUG_FORCE_WEAK_PER_CPU bool "Force weak per-cpu definitions" depends on DEBUG_KERNEL diff --git a/scripts/Makefile.build b/scripts/Makefile.build index 01df30a..3ffd994 100644 --- a/scripts/Makefile.build +++ b/scripts/Makefile.build @@ -241,10 +241,30 @@ cmd_record_mcount = \ fi; endif +ifdef CONFIG_STACK_VALIDATION + +__stacktool_obj := $(objtree)/tools/stacktool/stacktool + +ifdef CONFIG_FRAME_POINTER +stacktool_args += --check-frame-pointer +endif + +# Set STACKTOOL_foo.o=n to skip stack metadata validation for a file. +# Set STACKTOOL=n to skip stack metadata validation for a directory. +stacktool_obj = $(if $(patsubst n%,, \ + $(STACKTOOL_$(basetarget).o)$(STACKTOOL)y), \ + $(__stacktool_obj)) +cmd_stacktool = $(if $(patsubst n%,, \ + $(STACKTOOL_$(basetarget).o)$(STACKTOOL)y), \ + $(__stacktool_obj) $(stacktool_args) "$(@)";) + +endif # CONFIG_STACK_VALIDATION + define rule_cc_o_c $(call echo-cmd,checksrc) $(cmd_checksrc) \ $(call echo-cmd,cc_o_c) $(cmd_cc_o_c); \ $(cmd_modversions) \ + $(cmd_stacktool) \ $(call echo-cmd,record_mcount) \ $(cmd_record_mcount) \ scripts/basic/fixdep $(depfile) $@ '$(call make-cmd,cc_o_c)' > \ @@ -253,14 +273,23 @@ define rule_cc_o_c mv -f $(dot-target).tmp $(dot-target).cmd endef +define rule_as_o_S + $(call echo-cmd,as_o_S) $(cmd_as_o_S); \ + $(cmd_stacktool) \ + scripts/basic/fixdep $(depfile) $@ '$(call make-cmd,as_o_S)' > \ + $(dot-target).tmp; \ + rm -f $(depfile); \ + mv -f $(dot-target).tmp $(dot-target).cmd +endef + # Built-in and composite module parts -$(obj)/%.o: $(src)/%.c $(recordmcount_source) FORCE +$(obj)/%.o: $(src)/%.c $(recordmcount_source) $(stacktool_obj) FORCE $(call cmd,force_checksrc) $(call if_changed_rule,cc_o_c) # Single-part modules are special since we need to mark them in $(MODVERDIR) -$(single-used-m): $(obj)/%.o: $(src)/%.c $(recordmcount_source) FORCE +$(single-used-m): $(obj)/%.o: $(src)/%.c $(recordmcount_source) $(stacktool_obj) FORCE $(call cmd,force_checksrc) $(call if_changed_rule,cc_o_c) @{ echo $(@:.o=.ko); echo $@; } > $(MODVERDIR)/$(@F:.o=.mod) @@ -290,8 +319,8 @@ $(obj)/%.s: $(src)/%.S FORCE quiet_cmd_as_o_S = AS $(quiet_modtag) $@ cmd_as_o_S = $(CC) $(a_flags) -c -o $@ $< -$(obj)/%.o: $(src)/%.S FORCE - $(call if_changed_dep,as_o_S) +$(obj)/%.o: $(src)/%.S $(stacktool_obj) FORCE + $(call if_changed_rule,as_o_S) targets += $(real-objs-y) $(real-objs-m) $(lib-y) targets += $(extra-y) $(MAKECMDGOALS) $(always) diff --git a/tools/Makefile b/tools/Makefile index c7400a6..39d5d60 100644 --- a/tools/Makefile +++ b/tools/Makefile @@ -19,6 +19,7 @@ help: @echo ' net - misc networking tools' @echo ' perf - Linux performance measurement and analysis tool' @echo ' selftests - various kernel selftests' + @echo ' stacktool - a stack metadata validation tool' @echo ' tmon - thermal monitoring and tuning tool' @echo ' turbostat - Intel CPU idle stats and freq reporting tool' @echo ' usb - USB testing tools' @@ -48,7 +49,7 @@ acpi: FORCE cpupower: FORCE $(call descend,power/$@) -cgroup firewire hv guest usb virtio vm net iio: FORCE +cgroup firewire hv guest usb virtio vm net iio stacktool: FORCE $(call descend,$@) liblockdep: FORCE @@ -83,7 +84,7 @@ acpi_install: cpupower_install: $(call descend,power/$(@:_install=),install) -cgroup_install firewire_install hv_install lguest_install perf_install usb_install virtio_install vm_install net_install: +cgroup_install firewire_install hv_install lguest_install perf_install usb_install virtio_install vm_install net_install stacktool_install: $(call descend,$(@:_install=),install) selftests_install: @@ -101,7 +102,7 @@ freefall_install: install: acpi_install cgroup_install cpupower_install hv_install firewire_install lguest_install \ perf_install selftests_install turbostat_install usb_install \ virtio_install vm_install net_install x86_energy_perf_policy_install \ - tmon freefall_install + tmon freefall_install stacktool_install acpi_clean: $(call descend,power/acpi,clean) @@ -109,7 +110,7 @@ acpi_clean: cpupower_clean: $(call descend,power/cpupower,clean) -cgroup_clean hv_clean firewire_clean lguest_clean usb_clean virtio_clean vm_clean net_clean iio_clean: +cgroup_clean hv_clean firewire_clean lguest_clean usb_clean virtio_clean vm_clean net_clean iio_clean stacktool_clean: $(call descend,$(@:_clean=),clean) liblockdep_clean: @@ -136,6 +137,6 @@ freefall_clean: clean: acpi_clean cgroup_clean cpupower_clean hv_clean firewire_clean lguest_clean \ perf_clean selftests_clean turbostat_clean usb_clean virtio_clean \ vm_clean net_clean iio_clean x86_energy_perf_policy_clean tmon_clean \ - freefall_clean + freefall_clean stacktool_clean .PHONY: FORCE diff --git a/tools/stacktool/.gitignore b/tools/stacktool/.gitignore new file mode 100644 index 0000000..a4014ec --- /dev/null +++ b/tools/stacktool/.gitignore @@ -0,0 +1 @@ +arch/x86/insn/inat-tables.c diff --git a/tools/stacktool/Build b/tools/stacktool/Build new file mode 100644 index 0000000..7688d3d --- /dev/null +++ b/tools/stacktool/Build @@ -0,0 +1,4 @@ +stacktool-y += stacktool.o +stacktool-y += elf.o +stacktool-y += special.o +stacktool-y += arch/$(ARCH)/ diff --git a/tools/stacktool/Documentation/stack-validation.txt b/tools/stacktool/Documentation/stack-validation.txt new file mode 100644 index 0000000..49373c6 --- /dev/null +++ b/tools/stacktool/Documentation/stack-validation.txt @@ -0,0 +1,336 @@ +Compile-time stack metadata validation +====================================== + + +Overview +-------- + +The kernel CONFIG_STACK_VALIDATION option enables a host tool named +stacktool which runs at compile time. It analyzes every .o file and +ensures the validity of its stack metadata. It enforces a set of rules +on asm code and C inline assembly code so that stack traces can be +reliable. + +Currently it only checks frame pointer usage, but there are plans to add +CFI validation for C files and CFI generation for asm files. + +For each function, it recursively follows all possible code paths and +validates the correct frame pointer state at each instruction. + +It also follows code paths involving special sections, like +.altinstructions, __jump_table, and __ex_table, which can add +alternative execution paths to a given instruction (or set of +instructions). Similarly, it knows how to follow switch statements, for +which gcc sometimes uses jump tables. + + +Why do we need stack metadata validation? +----------------------------------------- + +Here are some of the benefits of validating stack metadata: + +a) More reliable stack traces for frame pointer enabled kernels + + Frame pointers are used for debugging purposes. They allow runtime + code and debug tools to be able to walk the stack to determine the + chain of function call sites that led to the currently executing + code. + + For some architectures, frame pointers are enabled by + CONFIG_FRAME_POINTER. For some other architectures they may be + required by the ABI (sometimes referred to as "backchain pointers"). + + For C code, gcc automatically generates instructions for setting up + frame pointers when the -fno-omit-frame-pointer option is used. + + But for asm code, the frame setup instructions have to be written by + hand, which most people don't do. So the end result is that + CONFIG_FRAME_POINTER is honored for C code but not for most asm code. + + For stack traces based on frame pointers to be reliable, all + functions which call other functions must first create a stack frame + and update the frame pointer. If a first function doesn't properly + create a stack frame before calling a second function, the *caller* + of the first function will be skipped on the stack trace. + + For example, consider the following example backtrace with frame + pointers enabled: + + [<ffffffff81812584>] dump_stack+0x4b/0x63 + [<ffffffff812d6dc2>] cmdline_proc_show+0x12/0x30 + [<ffffffff8127f568>] seq_read+0x108/0x3e0 + [<ffffffff812cce62>] proc_reg_read+0x42/0x70 + [<ffffffff81256197>] __vfs_read+0x37/0x100 + [<ffffffff81256b16>] vfs_read+0x86/0x130 + [<ffffffff81257898>] SyS_read+0x58/0xd0 + [<ffffffff8181c1f2>] entry_SYSCALL_64_fastpath+0x12/0x76 + + It correctly shows that the caller of cmdline_proc_show() is + seq_read(). + + If we remove the frame pointer logic from cmdline_proc_show() by + replacing the frame pointer related instructions with nops, here's + what it looks like instead: + + [<ffffffff81812584>] dump_stack+0x4b/0x63 + [<ffffffff812d6dc2>] cmdline_proc_show+0x12/0x30 + [<ffffffff812cce62>] proc_reg_read+0x42/0x70 + [<ffffffff81256197>] __vfs_read+0x37/0x100 + [<ffffffff81256b16>] vfs_read+0x86/0x130 + [<ffffffff81257898>] SyS_read+0x58/0xd0 + [<ffffffff8181c1f2>] entry_SYSCALL_64_fastpath+0x12/0x76 + + Notice that cmdline_proc_show()'s caller, seq_read(), has been + skipped. Instead the stack trace seems to show that + cmdline_proc_show() was called by proc_reg_read(). + + The benefit of stacktool here is that because it ensures that *all* + functions honor CONFIG_FRAME_POINTER, no functions will ever[*] be + skipped on a stack trace. + + [*] unless an interrupt or exception has occurred at the very + beginning of a function before the stack frame has been created, + or at the very end of the function after the stack frame has been + destroyed. This is an inherent limitation of frame pointers. + +b) 100% reliable stack traces for DWARF enabled kernels + + (NOTE: This is not yet implemented) + + As an alternative to frame pointers, DWARF Call Frame Information + (CFI) metadata can be used to walk the stack. Unlike frame pointers, + CFI metadata is out of band. So it doesn't affect runtime + performance and it can be reliable even when interrupts or exceptions + are involved. + + For C code, gcc automatically generates DWARF CFI metadata. But for + asm code, generating CFI is a tedious manual approach which requires + manually placed .cfi assembler macros to be scattered throughout the + code. It's clumsy and very easy to get wrong, and it makes the real + code harder to read. + + Stacktool will improve this situation in several ways. For code + which already has CFI annotations, it will validate them. For code + which doesn't have CFI annotations, it will generate them. So an + architecture can opt to strip out all the manual .cfi annotations + from their asm code and have stacktool generate them instead. + + We might also add a runtime stack validation debug option where we + periodically walk the stack from schedule() and/or an NMI to ensure + that the stack metadata is sane and that we reach the bottom of the + stack. + + So the benefit of stacktool here will be that external tooling should + always show perfect stack traces. And the same will be true for + kernel warning/oops traces if the architecture has a runtime DWARF + unwinder. + +c) Higher live patching compatibility rate + + (NOTE: This is not yet implemented) + + Currently with CONFIG_LIVEPATCH there's a basic live patching + framework which is safe for roughly 85-90% of "security" fixes. But + patches can't have complex features like function dependency or + prototype changes, or data structure changes. + + There's a strong need to support patches which have the more complex + features so that the patch compatibility rate for security fixes can + eventually approach something resembling 100%. To achieve that, a + "consistency model" is needed, which allows tasks to be safely + transitioned from an unpatched state to a patched state. + + One of the key requirements of the currently proposed livepatch + consistency model [*] is that it needs to walk the stack of each + sleeping task to determine if it can be transitioned to the patched + state. If stacktool can ensure that stack traces are reliable, this + consistency model can be used and the live patching compatibility + rate can be improved significantly. + + [*] https://lkml.kernel.org/r/cover.1423499826.git.jpoimboe@redhat.com + + +Rules +----- + +To achieve the validation, stacktool enforces the following rules: + +1. Each callable function must be annotated as such with the ELF + function type. In asm code, this is typically done using the + ENTRY/ENDPROC macros. If stacktool finds a return instruction + outside of a function, it flags an error since that usually indicates + callable code which should be annotated accordingly. + + This rule is needed so that stacktool can properly identify each + callable function in order to analyze its stack metadata. + +2. Conversely, each section of code which is *not* callable should *not* + be annotated as an ELF function. The ENDPROC macro shouldn't be used + in this case. + + This rule is needed so that stacktool can ignore non-callable code. + Such code doesn't have to follow any of the other rules. + +3. Each callable function which calls another function must have the + correct frame pointer logic, if required by CONFIG_FRAME_POINTER or + the architecture's back chain rules. This can by done in asm code + with the FRAME_BEGIN/FRAME_END macros. + + This rule ensures that frame pointer based stack traces will work as + designed. If function A doesn't create a stack frame before calling + function B, the _caller_ of function A will be skipped on the stack + trace. + +4. Dynamic jumps and jumps to undefined symbols are only allowed if: + + a) the jump is part of a switch statement; or + + b) the jump matches sibling call semantics and the frame pointer has + the same value it had on function entry. + + This rule is needed so that stacktool can reliably analyze all of a + function's code paths. If a function jumps to code in another file, + and it's not a sibling call, stacktool has no way to follow the jump + because it only analyzes a single file at a time. + +5. A callable function may not execute kernel entry/exit instructions. + The only code which needs such instructions is kernel entry code, + which shouldn't be be in callable functions anyway. + + This rule is just a sanity check to ensure that callable functions + return normally. + + +Errors in .S files +------------------ + +If you're getting an error in a compiled .S file which you don't +understand, first make sure that the affected code follows the above +rules. + +Here are some examples of common warnings reported by stacktool, what +they mean, and suggestions for how to fix them. + + +1. stacktool: asm_file.o: func()+0x128: call without frame pointer save/setup + + The func() function made a function call without first saving and/or + updating the frame pointer. + + If func() is indeed a callable function, add proper frame pointer + logic using the FRAME_BEGIN and FRAME_END macros. Otherwise, remove + its ELF function annotation by changing ENDPROC to END. + + If you're getting this error in a .c file, see the "Errors in .c + files" section. + + +2. stacktool: asm_file.o: .text+0x53: return instruction outside of a callable function + + A return instruction was detected, but stacktool couldn't find a way + for a callable function to reach the instruction. + + If the return instruction is inside (or reachable from) a callable + function, the function needs to be annotated with the ENTRY/ENDPROC + macros. + + If you _really_ need a return instruction outside of a function, and + are 100% sure that it won't affect stack traces, you can tell + stacktool to ignore it. See the "Adding exceptions" section below. + + +3. stacktool: asm_file.o: func()+0x9: function has unreachable instruction + + The instruction lives inside of a callable function, but there's no + possible control flow path from the beginning of the function to the + instruction. + + If the instruction is actually needed, and it's actually in a + callable function, ensure that its function is properly annotated + with ENTRY/ENDPROC. + + If it's not actually in a callable function (e.g. kernel entry code), + change ENDPROC to END. + + +4. stacktool: asm_file.o: func(): can't find starting instruction + or + stacktool: asm_file.o: func()+0x11dd: can't decode instruction + + Did you put data in a text section? If so, that can confuse + stacktool's instruction decoder. Move the data to a more appropriate + section like .data or .rodata. + + +5. stacktool: asm_file.o: func()+0x6: kernel entry/exit from callable instruction + + This is a kernel entry/exit instruction like sysenter or sysret. + Such instructions aren't allowed in a callable function, and are most + likely part of the kernel entry code. + + If the instruction isn't actually in a callable function, change + ENDPROC to END. + + +6. stacktool: asm_file.o: func()+0x26: sibling call from callable instruction with changed frame pointer + + This is a dynamic jump or a jump to an undefined symbol. Stacktool + assumed it's a sibling call and detected that the frame pointer + wasn't first restored to its original state. + + If it's not really a sibling call, you may need to move the + destination code to the local file. + + If the instruction is not actually in a callable function (e.g. + kernel entry code), change ENDPROC to END. + + +7. stacktool: asm_file: func()+0x5c: frame pointer state mismatch + + The instruction's frame pointer state is inconsistent, depending on + which execution path was taken to reach the instruction. + + Make sure the function pushes and sets up the frame pointer (for + x86_64, this means rbp) at the beginning of the function and pops it + at the end of the function. Also make sure that no other code in the + function touches the frame pointer. + + +Errors in .c files +------------------ + +If you're getting a stacktool error in a compiled .c file, chances are +the file uses an asm() statement which has a "call" instruction. An +asm() statement with a call instruction must declare the use of the +stack pointer in its output operand. For example, on x86_64: + + register void *__sp asm("rsp"); + asm volatile("call func" : "+r" (__sp)); + +Otherwise the stack frame may not get created before the call. + +Another possible cause for errors in C code is if the Makefile removes +-fno-omit-frame-pointer or adds -fomit-frame-pointer to the gcc options. + +Also see the above section for .S file errors for more information what +the individual error messages mean. + + + +Adding exceptions +----------------- + +If you _really_ need stacktool to ignore something, and are 100% sure +that it won't affect kernel stack traces, you can tell stacktool to +ignore it: + +- To skip validation of an instruction, use the STACKTOOL_IGNORE_INSN + macro immediately before the instruction. + +- To skip validation of a function, use the STACKTOOL_IGNORE_FUNC macro. + +- To skip validation of a file, add "STACKTOOL_filename.o := n" to the + Makefile. + +- To skip validation of a directory, add "STACKTOOL := n" to the + Makefile. diff --git a/tools/stacktool/Makefile b/tools/stacktool/Makefile new file mode 100644 index 0000000..9f5d2f9 --- /dev/null +++ b/tools/stacktool/Makefile @@ -0,0 +1,51 @@ +include ../scripts/Makefile.include + +ifndef ($(ARCH)) +ARCH ?= $(shell uname -m) +ifeq ($(ARCH),x86_64) +ARCH := x86 +endif +endif + +ifeq ($(srctree),) +srctree := $(patsubst %/,%,$(dir $(shell pwd))) +srctree := $(patsubst %/,%,$(dir $(srctree))) +endif + +INCLUDES := -I$(srctree)/tools/include +CFLAGS += -Wall -Werror $(EXTRA_WARNINGS) -fomit-frame-pointer -O2 $(INCLUDES) +LDFLAGS += -lelf + +STACKTOOL := $(OUTPUT)stacktool +STACKTOOL_IN := $(STACKTOOL)-in.o + +AWK = awk +export srctree OUTPUT CFLAGS ARCH AWK + +build := -f $(srctree)/tools/build/Makefile.build dir=. obj + +all: $(STACKTOOL) + +$(STACKTOOL_IN): FORCE + @$(MAKE) $(build)=stacktool + +$(STACKTOOL): $(STACKTOOL_IN) + @(test -d ../../kernel -a -d ../../tools -a -d ../stacktool && (( \ + diff -I'^#include' arch/x86/insn/insn.c ../../arch/x86/lib/insn.c >/dev/null && \ + diff -I'^#include' arch/x86/insn/inat.c ../../arch/x86/lib/inat.c >/dev/null && \ + diff arch/x86/insn/x86-opcode-map.txt ../../arch/x86/lib/x86-opcode-map.txt >/dev/null && \ + diff arch/x86/insn/gen-insn-attr-x86.awk ../../arch/x86/tools/gen-insn-attr-x86.awk >/dev/null && \ + diff -I'^#include' arch/x86/insn/insn.h ../../arch/x86/include/asm/insn.h >/dev/null && \ + diff -I'^#include' arch/x86/insn/inat.h ../../arch/x86/include/asm/inat.h >/dev/null && \ + diff -I'^#include' arch/x86/insn/inat_types.h ../../arch/x86/include/asm/inat_types.h >/dev/null) \ + || echo "Warning: stacktool: x86 instruction decoder differs from kernel" >&2 )) || true + $(QUIET_LINK)$(CC) $(STACKTOOL_IN) $(LDFLAGS) -o $@ + +clean: + $(call QUIET_CLEAN, stacktool) $(RM) $(STACKTOOL) + $(Q)find $(OUTPUT) -name '*.o' -delete -o -name '\.*.cmd' -delete -o -name '\.*.d' -delete + $(Q)$(RM) $(OUTPUT)arch/x86/insn/inat-tables.c + +FORCE: + +.PHONY: clean FORCE diff --git a/tools/stacktool/arch.h b/tools/stacktool/arch.h new file mode 100644 index 0000000..f7350fc --- /dev/null +++ b/tools/stacktool/arch.h @@ -0,0 +1,44 @@ +/* + * Copyright (C) 2015 Josh Poimboeuf <jpoimboe@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef _ARCH_H +#define _ARCH_H + +#include <stdbool.h> +#include "elf.h" + +#define INSN_FP_SAVE 1 +#define INSN_FP_SETUP 2 +#define INSN_FP_RESTORE 3 +#define INSN_JUMP_CONDITIONAL 4 +#define INSN_JUMP_UNCONDITIONAL 5 +#define INSN_JUMP_DYNAMIC 6 +#define INSN_CALL 7 +#define INSN_CALL_DYNAMIC 8 +#define INSN_RETURN 9 +#define INSN_CONTEXT_SWITCH 10 +#define INSN_BUG 11 +#define INSN_NOP 12 +#define INSN_OTHER 13 +#define INSN_LAST INSN_OTHER + +int arch_decode_instruction(struct elf *elf, struct section *sec, + unsigned long offset, unsigned int maxlen, + unsigned int *len, unsigned char *type, + unsigned long *displacement); + +#endif /* _ARCH_H */ diff --git a/tools/stacktool/arch/x86/Build b/tools/stacktool/arch/x86/Build new file mode 100644 index 0000000..9b9777d --- /dev/null +++ b/tools/stacktool/arch/x86/Build @@ -0,0 +1,12 @@ +stacktool-y += decode.o + +inat_tables_script = arch/x86/insn/gen-insn-attr-x86.awk +inat_tables_maps = arch/x86/insn/x86-opcode-map.txt + +$(OUTPUT)arch/x86/insn/inat-tables.c: $(inat_tables_script) $(inat_tables_maps) + $(call rule_mkdir) + $(Q)$(call echo-cmd,gen)$(AWK) -f $(inat_tables_script) $(inat_tables_maps) > $@ + +$(OUTPUT)arch/x86/decode.o: $(OUTPUT)arch/x86/insn/inat-tables.c + +CFLAGS_decode.o += -I$(OUTPUT)arch/x86/insn diff --git a/tools/stacktool/arch/x86/decode.c b/tools/stacktool/arch/x86/decode.c new file mode 100644 index 0000000..5aa78b9 --- /dev/null +++ b/tools/stacktool/arch/x86/decode.c @@ -0,0 +1,163 @@ +/* + * Copyright (C) 2015 Josh Poimboeuf <jpoimboe@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#include <stdio.h> +#include <stdlib.h> + +#define unlikely(cond) (cond) +#include "insn/insn.h" +#include "insn/inat.c" +#include "insn/insn.c" + +#include "../../elf.h" +#include "../../arch.h" +#include "../../warn.h" + +static int is_x86_64(struct elf *elf) +{ + switch (elf->ehdr.e_machine) { + case EM_X86_64: + return 1; + case EM_386: + return 0; + default: + WARN("unexpected ELF machine type %d", elf->ehdr.e_machine); + return -1; + } +} + +int arch_decode_instruction(struct elf *elf, struct section *sec, + unsigned long offset, unsigned int maxlen, + unsigned int *len, unsigned char *type, + unsigned long *immediate) +{ + struct insn insn; + int x86_64; + unsigned char op1, op2, ext; + + x86_64 = is_x86_64(elf); + if (x86_64 == -1) + return -1; + + insn_init(&insn, (void *)(sec->data + offset), maxlen, x86_64); + insn_get_length(&insn); + insn_get_opcode(&insn); + insn_get_modrm(&insn); + insn_get_immediate(&insn); + + if (!insn_complete(&insn)) { + WARN_FUNC("can't decode instruction", sec, offset); + return -1; + } + + *len = insn.length; + *type = INSN_OTHER; + + if (insn.vex_prefix.nbytes) + return 0; + + op1 = insn.opcode.bytes[0]; + op2 = insn.opcode.bytes[1]; + + switch (op1) { + case 0x55: + if (!insn.rex_prefix.nbytes) + /* push rbp */ + *type = INSN_FP_SAVE; + break; + + case 0x5d: + if (!insn.rex_prefix.nbytes) + /* pop rbp */ + *type = INSN_FP_RESTORE; + break; + + case 0x70 ... 0x7f: + *type = INSN_JUMP_CONDITIONAL; + break; + + case 0x89: + if (insn.rex_prefix.nbytes == 1 && + insn.rex_prefix.bytes[0] == 0x48 && + insn.modrm.nbytes && insn.modrm.bytes[0] == 0xe5) + /* mov rsp, rbp */ + *type = INSN_FP_SETUP; + break; + + case 0x90: + *type = INSN_NOP; + break; + + case 0x0f: + if (op2 >= 0x80 && op2 <= 0x8f) + *type = INSN_JUMP_CONDITIONAL; + else if (op2 == 0x05 || op2 == 0x07 || op2 == 0x34 || + op2 == 0x35) + /* sysenter, sysret */ + *type = INSN_CONTEXT_SWITCH; + else if (op2 == 0x0b || op2 == 0xb9) + /* ud2 */ + *type = INSN_BUG; + else if (op2 == 0x0d || op2 == 0x1f) + /* nopl/nopw */ + *type = INSN_NOP; + break; + + case 0xc9: /* leave */ + *type = INSN_FP_RESTORE; + break; + + case 0xe3: /* jecxz/jrcxz */ + *type = INSN_JUMP_CONDITIONAL; + break; + + case 0xe9: + case 0xeb: + *type = INSN_JUMP_UNCONDITIONAL; + break; + + case 0xc2: + case 0xc3: + *type = INSN_RETURN; + break; + + case 0xc5: /* iret */ + case 0xca: /* retf */ + case 0xcb: /* retf */ + *type = INSN_CONTEXT_SWITCH; + break; + + case 0xe8: + *type = INSN_CALL; + break; + + case 0xff: + ext = X86_MODRM_REG(insn.modrm.bytes[0]); + if (ext == 2 || ext == 3) + *type = INSN_CALL_DYNAMIC; + else if (ext == 4 || ext == 5) + *type = INSN_JUMP_DYNAMIC; + break; + + default: + break; + } + + *immediate = insn.immediate.nbytes ? insn.immediate.value : 0; + + return 0; +} diff --git a/tools/stacktool/arch/x86/insn/gen-insn-attr-x86.awk b/tools/stacktool/arch/x86/insn/gen-insn-attr-x86.awk new file mode 100644 index 0000000..093a892 --- /dev/null +++ b/tools/stacktool/arch/x86/insn/gen-insn-attr-x86.awk @@ -0,0 +1,387 @@ +#!/bin/awk -f +# gen-insn-attr-x86.awk: Instruction attribute table generator +# Written by Masami Hiramatsu <mhiramat@redhat.com> +# +# Usage: awk -f gen-insn-attr-x86.awk x86-opcode-map.txt > inat-tables.c + +# Awk implementation sanity check +function check_awk_implement() { + if (sprintf("%x", 0) != "0") + return "Your awk has a printf-format problem." + return "" +} + +# Clear working vars +function clear_vars() { + delete table + delete lptable2 + delete lptable1 + delete lptable3 + eid = -1 # escape id + gid = -1 # group id + aid = -1 # AVX id + tname = "" +} + +BEGIN { + # Implementation error checking + awkchecked = check_awk_implement() + if (awkchecked != "") { + print "Error: " awkchecked > "/dev/stderr" + print "Please try to use gawk." > "/dev/stderr" + exit 1 + } + + # Setup generating tables + print "/* x86 opcode map generated from x86-opcode-map.txt */" + print "/* Do not change this code. */\n" + ggid = 1 + geid = 1 + gaid = 0 + delete etable + delete gtable + delete atable + + opnd_expr = "^[A-Za-z/]" + ext_expr = "^\\(" + sep_expr = "^\\|$" + group_expr = "^Grp[0-9A-Za-z]+" + + imm_expr = "^[IJAOL][a-z]" + imm_flag["Ib"] = "INAT_MAKE_IMM(INAT_IMM_BYTE)" + imm_flag["Jb"] = "INAT_MAKE_IMM(INAT_IMM_BYTE)" + imm_flag["Iw"] = "INAT_MAKE_IMM(INAT_IMM_WORD)" + imm_flag["Id"] = "INAT_MAKE_IMM(INAT_IMM_DWORD)" + imm_flag["Iq"] = "INAT_MAKE_IMM(INAT_IMM_QWORD)" + imm_flag["Ap"] = "INAT_MAKE_IMM(INAT_IMM_PTR)" + imm_flag["Iz"] = "INAT_MAKE_IMM(INAT_IMM_VWORD32)" + imm_flag["Jz"] = "INAT_MAKE_IMM(INAT_IMM_VWORD32)" + imm_flag["Iv"] = "INAT_MAKE_IMM(INAT_IMM_VWORD)" + imm_flag["Ob"] = "INAT_MOFFSET" + imm_flag["Ov"] = "INAT_MOFFSET" + imm_flag["Lx"] = "INAT_MAKE_IMM(INAT_IMM_BYTE)" + + modrm_expr = "^([CDEGMNPQRSUVW/][a-z]+|NTA|T[012])" + force64_expr = "\\([df]64\\)" + rex_expr = "^REX(\\.[XRWB]+)*" + fpu_expr = "^ESC" # TODO + + lprefix1_expr = "\\((66|!F3)\\)" + lprefix2_expr = "\\(F3\\)" + lprefix3_expr = "\\((F2|!F3|66\\&F2)\\)" + lprefix_expr = "\\((66|F2|F3)\\)" + max_lprefix = 4 + + # All opcodes starting with lower-case 'v' or with (v1) superscript + # accepts VEX prefix + vexok_opcode_expr = "^v.*" + vexok_expr = "\\(v1\\)" + # All opcodes with (v) superscript supports *only* VEX prefix + vexonly_expr = "\\(v\\)" + + prefix_expr = "\\(Prefix\\)" + prefix_num["Operand-Size"] = "INAT_PFX_OPNDSZ" + prefix_num["REPNE"] = "INAT_PFX_REPNE" + prefix_num["REP/REPE"] = "INAT_PFX_REPE" + prefix_num["XACQUIRE"] = "INAT_PFX_REPNE" + prefix_num["XRELEASE"] = "INAT_PFX_REPE" + prefix_num["LOCK"] = "INAT_PFX_LOCK" + prefix_num["SEG=CS"] = "INAT_PFX_CS" + prefix_num["SEG=DS"] = "INAT_PFX_DS" + prefix_num["SEG=ES"] = "INAT_PFX_ES" + prefix_num["SEG=FS"] = "INAT_PFX_FS" + prefix_num["SEG=GS"] = "INAT_PFX_GS" + prefix_num["SEG=SS"] = "INAT_PFX_SS" + prefix_num["Address-Size"] = "INAT_PFX_ADDRSZ" + prefix_num["VEX+1byte"] = "INAT_PFX_VEX2" + prefix_num["VEX+2byte"] = "INAT_PFX_VEX3" + + clear_vars() +} + +function semantic_error(msg) { + print "Semantic error at " NR ": " msg > "/dev/stderr" + exit 1 +} + +function debug(msg) { + print "DEBUG: " msg +} + +function array_size(arr, i,c) { + c = 0 + for (i in arr) + c++ + return c +} + +/^Table:/ { + print "/* " $0 " */" + if (tname != "") + semantic_error("Hit Table: before EndTable:."); +} + +/^Referrer:/ { + if (NF != 1) { + # escape opcode table + ref = "" + for (i = 2; i <= NF; i++) + ref = ref $i + eid = escape[ref] + tname = sprintf("inat_escape_table_%d", eid) + } +} + +/^AVXcode:/ { + if (NF != 1) { + # AVX/escape opcode table + aid = $2 + if (gaid <= aid) + gaid = aid + 1 + if (tname == "") # AVX only opcode table + tname = sprintf("inat_avx_table_%d", $2) + } + if (aid == -1 && eid == -1) # primary opcode table + tname = "inat_primary_table" +} + +/^GrpTable:/ { + print "/* " $0 " */" + if (!($2 in group)) + semantic_error("No group: " $2 ) + gid = group[$2] + tname = "inat_group_table_" gid +} + +function print_table(tbl,name,fmt,n) +{ + print "const insn_attr_t " name " = {" + for (i = 0; i < n; i++) { + id = sprintf(fmt, i) + if (tbl[id]) + print " [" id "] = " tbl[id] "," + } + print "};" +} + +/^EndTable/ { + if (gid != -1) { + # print group tables + if (array_size(table) != 0) { + print_table(table, tname "[INAT_GROUP_TABLE_SIZE]", + "0x%x", 8) + gtable[gid,0] = tname + } + if (array_size(lptable1) != 0) { + print_table(lptable1, tname "_1[INAT_GROUP_TABLE_SIZE]", + "0x%x", 8) + gtable[gid,1] = tname "_1" + } + if (array_size(lptable2) != 0) { + print_table(lptable2, tname "_2[INAT_GROUP_TABLE_SIZE]", + "0x%x", 8) + gtable[gid,2] = tname "_2" + } + if (array_size(lptable3) != 0) { + print_table(lptable3, tname "_3[INAT_GROUP_TABLE_SIZE]", + "0x%x", 8) + gtable[gid,3] = tname "_3" + } + } else { + # print primary/escaped tables + if (array_size(table) != 0) { + print_table(table, tname "[INAT_OPCODE_TABLE_SIZE]", + "0x%02x", 256) + etable[eid,0] = tname + if (aid >= 0) + atable[aid,0] = tname + } + if (array_size(lptable1) != 0) { + print_table(lptable1,tname "_1[INAT_OPCODE_TABLE_SIZE]", + "0x%02x", 256) + etable[eid,1] = tname "_1" + if (aid >= 0) + atable[aid,1] = tname "_1" + } + if (array_size(lptable2) != 0) { + print_table(lptable2,tname "_2[INAT_OPCODE_TABLE_SIZE]", + "0x%02x", 256) + etable[eid,2] = tname "_2" + if (aid >= 0) + atable[aid,2] = tname "_2" + } + if (array_size(lptable3) != 0) { + print_table(lptable3,tname "_3[INAT_OPCODE_TABLE_SIZE]", + "0x%02x", 256) + etable[eid,3] = tname "_3" + if (aid >= 0) + atable[aid,3] = tname "_3" + } + } + print "" + clear_vars() +} + +function add_flags(old,new) { + if (old && new) + return old " | " new + else if (old) + return old + else + return new +} + +# convert operands to flags. +function convert_operands(count,opnd, i,j,imm,mod) +{ + imm = null + mod = null + for (j = 1; j <= count; j++) { + i = opnd[j] + if (match(i, imm_expr) == 1) { + if (!imm_flag[i]) + semantic_error("Unknown imm opnd: " i) + if (imm) { + if (i != "Ib") + semantic_error("Second IMM error") + imm = add_flags(imm, "INAT_SCNDIMM") + } else + imm = imm_flag[i] + } else if (match(i, modrm_expr)) + mod = "INAT_MODRM" + } + return add_flags(imm, mod) +} + +/^[0-9a-f]+\:/ { + if (NR == 1) + next + # get index + idx = "0x" substr($1, 1, index($1,":") - 1) + if (idx in table) + semantic_error("Redefine " idx " in " tname) + + # check if escaped opcode + if ("escape" == $2) { + if ($3 != "#") + semantic_error("No escaped name") + ref = "" + for (i = 4; i <= NF; i++) + ref = ref $i + if (ref in escape) + semantic_error("Redefine escape (" ref ")") + escape[ref] = geid + geid++ + table[idx] = "INAT_MAKE_ESCAPE(" escape[ref] ")" + next + } + + variant = null + # converts + i = 2 + while (i <= NF) { + opcode = $(i++) + delete opnds + ext = null + flags = null + opnd = null + # parse one opcode + if (match($i, opnd_expr)) { + opnd = $i + count = split($(i++), opnds, ",") + flags = convert_operands(count, opnds) + } + if (match($i, ext_expr)) + ext = $(i++) + if (match($i, sep_expr)) + i++ + else if (i < NF) + semantic_error($i " is not a separator") + + # check if group opcode + if (match(opcode, group_expr)) { + if (!(opcode in group)) { + group[opcode] = ggid + ggid++ + } + flags = add_flags(flags, "INAT_MAKE_GROUP(" group[opcode] ")") + } + # check force(or default) 64bit + if (match(ext, force64_expr)) + flags = add_flags(flags, "INAT_FORCE64") + + # check REX prefix + if (match(opcode, rex_expr)) + flags = add_flags(flags, "INAT_MAKE_PREFIX(INAT_PFX_REX)") + + # check coprocessor escape : TODO + if (match(opcode, fpu_expr)) + flags = add_flags(flags, "INAT_MODRM") + + # check VEX codes + if (match(ext, vexonly_expr)) + flags = add_flags(flags, "INAT_VEXOK | INAT_VEXONLY") + else if (match(ext, vexok_expr) || match(opcode, vexok_opcode_expr)) + flags = add_flags(flags, "INAT_VEXOK") + + # check prefixes + if (match(ext, prefix_expr)) { + if (!prefix_num[opcode]) + semantic_error("Unknown prefix: " opcode) + flags = add_flags(flags, "INAT_MAKE_PREFIX(" prefix_num[opcode] ")") + } + if (length(flags) == 0) + continue + # check if last prefix + if (match(ext, lprefix1_expr)) { + lptable1[idx] = add_flags(lptable1[idx],flags) + variant = "INAT_VARIANT" + } + if (match(ext, lprefix2_expr)) { + lptable2[idx] = add_flags(lptable2[idx],flags) + variant = "INAT_VARIANT" + } + if (match(ext, lprefix3_expr)) { + lptable3[idx] = add_flags(lptable3[idx],flags) + variant = "INAT_VARIANT" + } + if (!match(ext, lprefix_expr)){ + table[idx] = add_flags(table[idx],flags) + } + } + if (variant) + table[idx] = add_flags(table[idx],variant) +} + +END { + if (awkchecked != "") + exit 1 + # print escape opcode map's array + print "/* Escape opcode map array */" + print "const insn_attr_t * const inat_escape_tables[INAT_ESC_MAX + 1]" \ + "[INAT_LSTPFX_MAX + 1] = {" + for (i = 0; i < geid; i++) + for (j = 0; j < max_lprefix; j++) + if (etable[i,j]) + print " ["i"]["j"] = "etable[i,j]"," + print "};\n" + # print group opcode map's array + print "/* Group opcode map array */" + print "const insn_attr_t * const inat_group_tables[INAT_GRP_MAX + 1]"\ + "[INAT_LSTPFX_MAX + 1] = {" + for (i = 0; i < ggid; i++) + for (j = 0; j < max_lprefix; j++) + if (gtable[i,j]) + print " ["i"]["j"] = "gtable[i,j]"," + print "};\n" + # print AVX opcode map's array + print "/* AVX opcode map array */" + print "const insn_attr_t * const inat_avx_tables[X86_VEX_M_MAX + 1]"\ + "[INAT_LSTPFX_MAX + 1] = {" + for (i = 0; i < gaid; i++) + for (j = 0; j < max_lprefix; j++) + if (atable[i,j]) + print " ["i"]["j"] = "atable[i,j]"," + print "};" +} + diff --git a/tools/stacktool/arch/x86/insn/inat.c b/tools/stacktool/arch/x86/insn/inat.c new file mode 100644 index 0000000..e4bf28e --- /dev/null +++ b/tools/stacktool/arch/x86/insn/inat.c @@ -0,0 +1,97 @@ +/* + * x86 instruction attribute tables + * + * Written by Masami Hiramatsu <mhiramat@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + */ +#include "insn.h" + +/* Attribute tables are generated from opcode map */ +#include "inat-tables.c" + +/* Attribute search APIs */ +insn_attr_t inat_get_opcode_attribute(insn_byte_t opcode) +{ + return inat_primary_table[opcode]; +} + +int inat_get_last_prefix_id(insn_byte_t last_pfx) +{ + insn_attr_t lpfx_attr; + + lpfx_attr = inat_get_opcode_attribute(last_pfx); + return inat_last_prefix_id(lpfx_attr); +} + +insn_attr_t inat_get_escape_attribute(insn_byte_t opcode, int lpfx_id, + insn_attr_t esc_attr) +{ + const insn_attr_t *table; + int n; + + n = inat_escape_id(esc_attr); + + table = inat_escape_tables[n][0]; + if (!table) + return 0; + if (inat_has_variant(table[opcode]) && lpfx_id) { + table = inat_escape_tables[n][lpfx_id]; + if (!table) + return 0; + } + return table[opcode]; +} + +insn_attr_t inat_get_group_attribute(insn_byte_t modrm, int lpfx_id, + insn_attr_t grp_attr) +{ + const insn_attr_t *table; + int n; + + n = inat_group_id(grp_attr); + + table = inat_group_tables[n][0]; + if (!table) + return inat_group_common_attribute(grp_attr); + if (inat_has_variant(table[X86_MODRM_REG(modrm)]) && lpfx_id) { + table = inat_group_tables[n][lpfx_id]; + if (!table) + return inat_group_common_attribute(grp_attr); + } + return table[X86_MODRM_REG(modrm)] | + inat_group_common_attribute(grp_attr); +} + +insn_attr_t inat_get_avx_attribute(insn_byte_t opcode, insn_byte_t vex_m, + insn_byte_t vex_p) +{ + const insn_attr_t *table; + if (vex_m > X86_VEX_M_MAX || vex_p > INAT_LSTPFX_MAX) + return 0; + /* At first, this checks the master table */ + table = inat_avx_tables[vex_m][0]; + if (!table) + return 0; + if (!inat_is_group(table[opcode]) && vex_p) { + /* If this is not a group, get attribute directly */ + table = inat_avx_tables[vex_m][vex_p]; + if (!table) + return 0; + } + return table[opcode]; +} + diff --git a/tools/stacktool/arch/x86/insn/inat.h b/tools/stacktool/arch/x86/insn/inat.h new file mode 100644 index 0000000..611645e --- /dev/null +++ b/tools/stacktool/arch/x86/insn/inat.h @@ -0,0 +1,221 @@ +#ifndef _ASM_X86_INAT_H +#define _ASM_X86_INAT_H +/* + * x86 instruction attributes + * + * Written by Masami Hiramatsu <mhiramat@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + */ +#include "inat_types.h" + +/* + * Internal bits. Don't use bitmasks directly, because these bits are + * unstable. You should use checking functions. + */ + +#define INAT_OPCODE_TABLE_SIZE 256 +#define INAT_GROUP_TABLE_SIZE 8 + +/* Legacy last prefixes */ +#define INAT_PFX_OPNDSZ 1 /* 0x66 */ /* LPFX1 */ +#define INAT_PFX_REPE 2 /* 0xF3 */ /* LPFX2 */ +#define INAT_PFX_REPNE 3 /* 0xF2 */ /* LPFX3 */ +/* Other Legacy prefixes */ +#define INAT_PFX_LOCK 4 /* 0xF0 */ +#define INAT_PFX_CS 5 /* 0x2E */ +#define INAT_PFX_DS 6 /* 0x3E */ +#define INAT_PFX_ES 7 /* 0x26 */ +#define INAT_PFX_FS 8 /* 0x64 */ +#define INAT_PFX_GS 9 /* 0x65 */ +#define INAT_PFX_SS 10 /* 0x36 */ +#define INAT_PFX_ADDRSZ 11 /* 0x67 */ +/* x86-64 REX prefix */ +#define INAT_PFX_REX 12 /* 0x4X */ +/* AVX VEX prefixes */ +#define INAT_PFX_VEX2 13 /* 2-bytes VEX prefix */ +#define INAT_PFX_VEX3 14 /* 3-bytes VEX prefix */ + +#define INAT_LSTPFX_MAX 3 +#define INAT_LGCPFX_MAX 11 + +/* Immediate size */ +#define INAT_IMM_BYTE 1 +#define INAT_IMM_WORD 2 +#define INAT_IMM_DWORD 3 +#define INAT_IMM_QWORD 4 +#define INAT_IMM_PTR 5 +#define INAT_IMM_VWORD32 6 +#define INAT_IMM_VWORD 7 + +/* Legacy prefix */ +#define INAT_PFX_OFFS 0 +#define INAT_PFX_BITS 4 +#define INAT_PFX_MAX ((1 << INAT_PFX_BITS) - 1) +#define INAT_PFX_MASK (INAT_PFX_MAX << INAT_PFX_OFFS) +/* Escape opcodes */ +#define INAT_ESC_OFFS (INAT_PFX_OFFS + INAT_PFX_BITS) +#define INAT_ESC_BITS 2 +#define INAT_ESC_MAX ((1 << INAT_ESC_BITS) - 1) +#define INAT_ESC_MASK (INAT_ESC_MAX << INAT_ESC_OFFS) +/* Group opcodes (1-16) */ +#define INAT_GRP_OFFS (INAT_ESC_OFFS + INAT_ESC_BITS) +#define INAT_GRP_BITS 5 +#define INAT_GRP_MAX ((1 << INAT_GRP_BITS) - 1) +#define INAT_GRP_MASK (INAT_GRP_MAX << INAT_GRP_OFFS) +/* Immediates */ +#define INAT_IMM_OFFS (INAT_GRP_OFFS + INAT_GRP_BITS) +#define INAT_IMM_BITS 3 +#define INAT_IMM_MASK (((1 << INAT_IMM_BITS) - 1) << INAT_IMM_OFFS) +/* Flags */ +#define INAT_FLAG_OFFS (INAT_IMM_OFFS + INAT_IMM_BITS) +#define INAT_MODRM (1 << (INAT_FLAG_OFFS)) +#define INAT_FORCE64 (1 << (INAT_FLAG_OFFS + 1)) +#define INAT_SCNDIMM (1 << (INAT_FLAG_OFFS + 2)) +#define INAT_MOFFSET (1 << (INAT_FLAG_OFFS + 3)) +#define INAT_VARIANT (1 << (INAT_FLAG_OFFS + 4)) +#define INAT_VEXOK (1 << (INAT_FLAG_OFFS + 5)) +#define INAT_VEXONLY (1 << (INAT_FLAG_OFFS + 6)) +/* Attribute making macros for attribute tables */ +#define INAT_MAKE_PREFIX(pfx) (pfx << INAT_PFX_OFFS) +#define INAT_MAKE_ESCAPE(esc) (esc << INAT_ESC_OFFS) +#define INAT_MAKE_GROUP(grp) ((grp << INAT_GRP_OFFS) | INAT_MODRM) +#define INAT_MAKE_IMM(imm) (imm << INAT_IMM_OFFS) + +/* Attribute search APIs */ +extern insn_attr_t inat_get_opcode_attribute(insn_byte_t opcode); +extern int inat_get_last_prefix_id(insn_byte_t last_pfx); +extern insn_attr_t inat_get_escape_attribute(insn_byte_t opcode, + int lpfx_id, + insn_attr_t esc_attr); +extern insn_attr_t inat_get_group_attribute(insn_byte_t modrm, + int lpfx_id, + insn_attr_t esc_attr); +extern insn_attr_t inat_get_avx_attribute(insn_byte_t opcode, + insn_byte_t vex_m, + insn_byte_t vex_pp); + +/* Attribute checking functions */ +static inline int inat_is_legacy_prefix(insn_attr_t attr) +{ + attr &= INAT_PFX_MASK; + return attr && attr <= INAT_LGCPFX_MAX; +} + +static inline int inat_is_address_size_prefix(insn_attr_t attr) +{ + return (attr & INAT_PFX_MASK) == INAT_PFX_ADDRSZ; +} + +static inline int inat_is_operand_size_prefix(insn_attr_t attr) +{ + return (attr & INAT_PFX_MASK) == INAT_PFX_OPNDSZ; +} + +static inline int inat_is_rex_prefix(insn_attr_t attr) +{ + return (attr & INAT_PFX_MASK) == INAT_PFX_REX; +} + +static inline int inat_last_prefix_id(insn_attr_t attr) +{ + if ((attr & INAT_PFX_MASK) > INAT_LSTPFX_MAX) + return 0; + else + return attr & INAT_PFX_MASK; +} + +static inline int inat_is_vex_prefix(insn_attr_t attr) +{ + attr &= INAT_PFX_MASK; + return attr == INAT_PFX_VEX2 || attr == INAT_PFX_VEX3; +} + +static inline int inat_is_vex3_prefix(insn_attr_t attr) +{ + return (attr & INAT_PFX_MASK) == INAT_PFX_VEX3; +} + +static inline int inat_is_escape(insn_attr_t attr) +{ + return attr & INAT_ESC_MASK; +} + +static inline int inat_escape_id(insn_attr_t attr) +{ + return (attr & INAT_ESC_MASK) >> INAT_ESC_OFFS; +} + +static inline int inat_is_group(insn_attr_t attr) +{ + return attr & INAT_GRP_MASK; +} + +static inline int inat_group_id(insn_attr_t attr) +{ + return (attr & INAT_GRP_MASK) >> INAT_GRP_OFFS; +} + +static inline int inat_group_common_attribute(insn_attr_t attr) +{ + return attr & ~INAT_GRP_MASK; +} + +static inline int inat_has_immediate(insn_attr_t attr) +{ + return attr & INAT_IMM_MASK; +} + +static inline int inat_immediate_size(insn_attr_t attr) +{ + return (attr & INAT_IMM_MASK) >> INAT_IMM_OFFS; +} + +static inline int inat_has_modrm(insn_attr_t attr) +{ + return attr & INAT_MODRM; +} + +static inline int inat_is_force64(insn_attr_t attr) +{ + return attr & INAT_FORCE64; +} + +static inline int inat_has_second_immediate(insn_attr_t attr) +{ + return attr & INAT_SCNDIMM; +} + +static inline int inat_has_moffset(insn_attr_t attr) +{ + return attr & INAT_MOFFSET; +} + +static inline int inat_has_variant(insn_attr_t attr) +{ + return attr & INAT_VARIANT; +} + +static inline int inat_accept_vex(insn_attr_t attr) +{ + return attr & INAT_VEXOK; +} + +static inline int inat_must_vex(insn_attr_t attr) +{ + return attr & INAT_VEXONLY; +} +#endif diff --git a/tools/stacktool/arch/x86/insn/inat_types.h b/tools/stacktool/arch/x86/insn/inat_types.h new file mode 100644 index 0000000..cb3c20c --- /dev/null +++ b/tools/stacktool/arch/x86/insn/inat_types.h @@ -0,0 +1,29 @@ +#ifndef _ASM_X86_INAT_TYPES_H +#define _ASM_X86_INAT_TYPES_H +/* + * x86 instruction attributes + * + * Written by Masami Hiramatsu <mhiramat@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + */ + +/* Instruction attributes */ +typedef unsigned int insn_attr_t; +typedef unsigned char insn_byte_t; +typedef signed int insn_value_t; + +#endif diff --git a/tools/stacktool/arch/x86/insn/insn.c b/tools/stacktool/arch/x86/insn/insn.c new file mode 100644 index 0000000..47314a6 --- /dev/null +++ b/tools/stacktool/arch/x86/insn/insn.c @@ -0,0 +1,594 @@ +/* + * x86 instruction analysis + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright (C) IBM Corporation, 2002, 2004, 2009 + */ + +#ifdef __KERNEL__ +#include <linux/string.h> +#else +#include <string.h> +#endif +#include "inat.h" +#include "insn.h" + +/* Verify next sizeof(t) bytes can be on the same instruction */ +#define validate_next(t, insn, n) \ + ((insn)->next_byte + sizeof(t) + n <= (insn)->end_kaddr) + +#define __get_next(t, insn) \ + ({ t r = *(t*)insn->next_byte; insn->next_byte += sizeof(t); r; }) + +#define __peek_nbyte_next(t, insn, n) \ + ({ t r = *(t*)((insn)->next_byte + n); r; }) + +#define get_next(t, insn) \ + ({ if (unlikely(!validate_next(t, insn, 0))) goto err_out; __get_next(t, insn); }) + +#define peek_nbyte_next(t, insn, n) \ + ({ if (unlikely(!validate_next(t, insn, n))) goto err_out; __peek_nbyte_next(t, insn, n); }) + +#define peek_next(t, insn) peek_nbyte_next(t, insn, 0) + +/** + * insn_init() - initialize struct insn + * @insn: &struct insn to be initialized + * @kaddr: address (in kernel memory) of instruction (or copy thereof) + * @x86_64: !0 for 64-bit kernel or 64-bit app + */ +void insn_init(struct insn *insn, const void *kaddr, int buf_len, int x86_64) +{ + /* + * Instructions longer than MAX_INSN_SIZE (15 bytes) are invalid + * even if the input buffer is long enough to hold them. + */ + if (buf_len > MAX_INSN_SIZE) + buf_len = MAX_INSN_SIZE; + + memset(insn, 0, sizeof(*insn)); + insn->kaddr = kaddr; + insn->end_kaddr = kaddr + buf_len; + insn->next_byte = kaddr; + insn->x86_64 = x86_64 ? 1 : 0; + insn->opnd_bytes = 4; + if (x86_64) + insn->addr_bytes = 8; + else + insn->addr_bytes = 4; +} + +/** + * insn_get_prefixes - scan x86 instruction prefix bytes + * @insn: &struct insn containing instruction + * + * Populates the @insn->prefixes bitmap, and updates @insn->next_byte + * to point to the (first) opcode. No effect if @insn->prefixes.got + * is already set. + */ +void insn_get_prefixes(struct insn *insn) +{ + struct insn_field *prefixes = &insn->prefixes; + insn_attr_t attr; + insn_byte_t b, lb; + int i, nb; + + if (prefixes->got) + return; + + nb = 0; + lb = 0; + b = peek_next(insn_byte_t, insn); + attr = inat_get_opcode_attribute(b); + while (inat_is_legacy_prefix(attr)) { + /* Skip if same prefix */ + for (i = 0; i < nb; i++) + if (prefixes->bytes[i] == b) + goto found; + if (nb == 4) + /* Invalid instruction */ + break; + prefixes->bytes[nb++] = b; + if (inat_is_address_size_prefix(attr)) { + /* address size switches 2/4 or 4/8 */ + if (insn->x86_64) + insn->addr_bytes ^= 12; + else + insn->addr_bytes ^= 6; + } else if (inat_is_operand_size_prefix(attr)) { + /* oprand size switches 2/4 */ + insn->opnd_bytes ^= 6; + } +found: + prefixes->nbytes++; + insn->next_byte++; + lb = b; + b = peek_next(insn_byte_t, insn); + attr = inat_get_opcode_attribute(b); + } + /* Set the last prefix */ + if (lb && lb != insn->prefixes.bytes[3]) { + if (unlikely(insn->prefixes.bytes[3])) { + /* Swap the last prefix */ + b = insn->prefixes.bytes[3]; + for (i = 0; i < nb; i++) + if (prefixes->bytes[i] == lb) + prefixes->bytes[i] = b; + } + insn->prefixes.bytes[3] = lb; + } + + /* Decode REX prefix */ + if (insn->x86_64) { + b = peek_next(insn_byte_t, insn); + attr = inat_get_opcode_attribute(b); + if (inat_is_rex_prefix(attr)) { + insn->rex_prefix.value = b; + insn->rex_prefix.nbytes = 1; + insn->next_byte++; + if (X86_REX_W(b)) + /* REX.W overrides opnd_size */ + insn->opnd_bytes = 8; + } + } + insn->rex_prefix.got = 1; + + /* Decode VEX prefix */ + b = peek_next(insn_byte_t, insn); + attr = inat_get_opcode_attribute(b); + if (inat_is_vex_prefix(attr)) { + insn_byte_t b2 = peek_nbyte_next(insn_byte_t, insn, 1); + if (!insn->x86_64) { + /* + * In 32-bits mode, if the [7:6] bits (mod bits of + * ModRM) on the second byte are not 11b, it is + * LDS or LES. + */ + if (X86_MODRM_MOD(b2) != 3) + goto vex_end; + } + insn->vex_prefix.bytes[0] = b; + insn->vex_prefix.bytes[1] = b2; + if (inat_is_vex3_prefix(attr)) { + b2 = peek_nbyte_next(insn_byte_t, insn, 2); + insn->vex_prefix.bytes[2] = b2; + insn->vex_prefix.nbytes = 3; + insn->next_byte += 3; + if (insn->x86_64 && X86_VEX_W(b2)) + /* VEX.W overrides opnd_size */ + insn->opnd_bytes = 8; + } else { + /* + * For VEX2, fake VEX3-like byte#2. + * Makes it easier to decode vex.W, vex.vvvv, + * vex.L and vex.pp. Masking with 0x7f sets vex.W == 0. + */ + insn->vex_prefix.bytes[2] = b2 & 0x7f; + insn->vex_prefix.nbytes = 2; + insn->next_byte += 2; + } + } +vex_end: + insn->vex_prefix.got = 1; + + prefixes->got = 1; + +err_out: + return; +} + +/** + * insn_get_opcode - collect opcode(s) + * @insn: &struct insn containing instruction + * + * Populates @insn->opcode, updates @insn->next_byte to point past the + * opcode byte(s), and set @insn->attr (except for groups). + * If necessary, first collects any preceding (prefix) bytes. + * Sets @insn->opcode.value = opcode1. No effect if @insn->opcode.got + * is already 1. + */ +void insn_get_opcode(struct insn *insn) +{ + struct insn_field *opcode = &insn->opcode; + insn_byte_t op; + int pfx_id; + if (opcode->got) + return; + if (!insn->prefixes.got) + insn_get_prefixes(insn); + + /* Get first opcode */ + op = get_next(insn_byte_t, insn); + opcode->bytes[0] = op; + opcode->nbytes = 1; + + /* Check if there is VEX prefix or not */ + if (insn_is_avx(insn)) { + insn_byte_t m, p; + m = insn_vex_m_bits(insn); + p = insn_vex_p_bits(insn); + insn->attr = inat_get_avx_attribute(op, m, p); + if (!inat_accept_vex(insn->attr) && !inat_is_group(insn->attr)) + insn->attr = 0; /* This instruction is bad */ + goto end; /* VEX has only 1 byte for opcode */ + } + + insn->attr = inat_get_opcode_attribute(op); + while (inat_is_escape(insn->attr)) { + /* Get escaped opcode */ + op = get_next(insn_byte_t, insn); + opcode->bytes[opcode->nbytes++] = op; + pfx_id = insn_last_prefix_id(insn); + insn->attr = inat_get_escape_attribute(op, pfx_id, insn->attr); + } + if (inat_must_vex(insn->attr)) + insn->attr = 0; /* This instruction is bad */ +end: + opcode->got = 1; + +err_out: + return; +} + +/** + * insn_get_modrm - collect ModRM byte, if any + * @insn: &struct insn containing instruction + * + * Populates @insn->modrm and updates @insn->next_byte to point past the + * ModRM byte, if any. If necessary, first collects the preceding bytes + * (prefixes and opcode(s)). No effect if @insn->modrm.got is already 1. + */ +void insn_get_modrm(struct insn *insn) +{ + struct insn_field *modrm = &insn->modrm; + insn_byte_t pfx_id, mod; + if (modrm->got) + return; + if (!insn->opcode.got) + insn_get_opcode(insn); + + if (inat_has_modrm(insn->attr)) { + mod = get_next(insn_byte_t, insn); + modrm->value = mod; + modrm->nbytes = 1; + if (inat_is_group(insn->attr)) { + pfx_id = insn_last_prefix_id(insn); + insn->attr = inat_get_group_attribute(mod, pfx_id, + insn->attr); + if (insn_is_avx(insn) && !inat_accept_vex(insn->attr)) + insn->attr = 0; /* This is bad */ + } + } + + if (insn->x86_64 && inat_is_force64(insn->attr)) + insn->opnd_bytes = 8; + modrm->got = 1; + +err_out: + return; +} + + +/** + * insn_rip_relative() - Does instruction use RIP-relative addressing mode? + * @insn: &struct insn containing instruction + * + * If necessary, first collects the instruction up to and including the + * ModRM byte. No effect if @insn->x86_64 is 0. + */ +int insn_rip_relative(struct insn *insn) +{ + struct insn_field *modrm = &insn->modrm; + + if (!insn->x86_64) + return 0; + if (!modrm->got) + insn_get_modrm(insn); + /* + * For rip-relative instructions, the mod field (top 2 bits) + * is zero and the r/m field (bottom 3 bits) is 0x5. + */ + return (modrm->nbytes && (modrm->value & 0xc7) == 0x5); +} + +/** + * insn_get_sib() - Get the SIB byte of instruction + * @insn: &struct insn containing instruction + * + * If necessary, first collects the instruction up to and including the + * ModRM byte. + */ +void insn_get_sib(struct insn *insn) +{ + insn_byte_t modrm; + + if (insn->sib.got) + return; + if (!insn->modrm.got) + insn_get_modrm(insn); + if (insn->modrm.nbytes) { + modrm = (insn_byte_t)insn->modrm.value; + if (insn->addr_bytes != 2 && + X86_MODRM_MOD(modrm) != 3 && X86_MODRM_RM(modrm) == 4) { + insn->sib.value = get_next(insn_byte_t, insn); + insn->sib.nbytes = 1; + } + } + insn->sib.got = 1; + +err_out: + return; +} + + +/** + * insn_get_displacement() - Get the displacement of instruction + * @insn: &struct insn containing instruction + * + * If necessary, first collects the instruction up to and including the + * SIB byte. + * Displacement value is sign-expanded. + */ +void insn_get_displacement(struct insn *insn) +{ + insn_byte_t mod, rm, base; + + if (insn->displacement.got) + return; + if (!insn->sib.got) + insn_get_sib(insn); + if (insn->modrm.nbytes) { + /* + * Interpreting the modrm byte: + * mod = 00 - no displacement fields (exceptions below) + * mod = 01 - 1-byte displacement field + * mod = 10 - displacement field is 4 bytes, or 2 bytes if + * address size = 2 (0x67 prefix in 32-bit mode) + * mod = 11 - no memory operand + * + * If address size = 2... + * mod = 00, r/m = 110 - displacement field is 2 bytes + * + * If address size != 2... + * mod != 11, r/m = 100 - SIB byte exists + * mod = 00, SIB base = 101 - displacement field is 4 bytes + * mod = 00, r/m = 101 - rip-relative addressing, displacement + * field is 4 bytes + */ + mod = X86_MODRM_MOD(insn->modrm.value); + rm = X86_MODRM_RM(insn->modrm.value); + base = X86_SIB_BASE(insn->sib.value); + if (mod == 3) + goto out; + if (mod == 1) { + insn->displacement.value = get_next(char, insn); + insn->displacement.nbytes = 1; + } else if (insn->addr_bytes == 2) { + if ((mod == 0 && rm == 6) || mod == 2) { + insn->displacement.value = + get_next(short, insn); + insn->displacement.nbytes = 2; + } + } else { + if ((mod == 0 && rm == 5) || mod == 2 || + (mod == 0 && base == 5)) { + insn->displacement.value = get_next(int, insn); + insn->displacement.nbytes = 4; + } + } + } +out: + insn->displacement.got = 1; + +err_out: + return; +} + +/* Decode moffset16/32/64. Return 0 if failed */ +static int __get_moffset(struct insn *insn) +{ + switch (insn->addr_bytes) { + case 2: + insn->moffset1.value = get_next(short, insn); + insn->moffset1.nbytes = 2; + break; + case 4: + insn->moffset1.value = get_next(int, insn); + insn->moffset1.nbytes = 4; + break; + case 8: + insn->moffset1.value = get_next(int, insn); + insn->moffset1.nbytes = 4; + insn->moffset2.value = get_next(int, insn); + insn->moffset2.nbytes = 4; + break; + default: /* opnd_bytes must be modified manually */ + goto err_out; + } + insn->moffset1.got = insn->moffset2.got = 1; + + return 1; + +err_out: + return 0; +} + +/* Decode imm v32(Iz). Return 0 if failed */ +static int __get_immv32(struct insn *insn) +{ + switch (insn->opnd_bytes) { + case 2: + insn->immediate.value = get_next(short, insn); + insn->immediate.nbytes = 2; + break; + case 4: + case 8: + insn->immediate.value = get_next(int, insn); + insn->immediate.nbytes = 4; + break; + default: /* opnd_bytes must be modified manually */ + goto err_out; + } + + return 1; + +err_out: + return 0; +} + +/* Decode imm v64(Iv/Ov), Return 0 if failed */ +static int __get_immv(struct insn *insn) +{ + switch (insn->opnd_bytes) { + case 2: + insn->immediate1.value = get_next(short, insn); + insn->immediate1.nbytes = 2; + break; + case 4: + insn->immediate1.value = get_next(int, insn); + insn->immediate1.nbytes = 4; + break; + case 8: + insn->immediate1.value = get_next(int, insn); + insn->immediate1.nbytes = 4; + insn->immediate2.value = get_next(int, insn); + insn->immediate2.nbytes = 4; + break; + default: /* opnd_bytes must be modified manually */ + goto err_out; + } + insn->immediate1.got = insn->immediate2.got = 1; + + return 1; +err_out: + return 0; +} + +/* Decode ptr16:16/32(Ap) */ +static int __get_immptr(struct insn *insn) +{ + switch (insn->opnd_bytes) { + case 2: + insn->immediate1.value = get_next(short, insn); + insn->immediate1.nbytes = 2; + break; + case 4: + insn->immediate1.value = get_next(int, insn); + insn->immediate1.nbytes = 4; + break; + case 8: + /* ptr16:64 is not exist (no segment) */ + return 0; + default: /* opnd_bytes must be modified manually */ + goto err_out; + } + insn->immediate2.value = get_next(unsigned short, insn); + insn->immediate2.nbytes = 2; + insn->immediate1.got = insn->immediate2.got = 1; + + return 1; +err_out: + return 0; +} + +/** + * insn_get_immediate() - Get the immediates of instruction + * @insn: &struct insn containing instruction + * + * If necessary, first collects the instruction up to and including the + * displacement bytes. + * Basically, most of immediates are sign-expanded. Unsigned-value can be + * get by bit masking with ((1 << (nbytes * 8)) - 1) + */ +void insn_get_immediate(struct insn *insn) +{ + if (insn->immediate.got) + return; + if (!insn->displacement.got) + insn_get_displacement(insn); + + if (inat_has_moffset(insn->attr)) { + if (!__get_moffset(insn)) + goto err_out; + goto done; + } + + if (!inat_has_immediate(insn->attr)) + /* no immediates */ + goto done; + + switch (inat_immediate_size(insn->attr)) { + case INAT_IMM_BYTE: + insn->immediate.value = get_next(char, insn); + insn->immediate.nbytes = 1; + break; + case INAT_IMM_WORD: + insn->immediate.value = get_next(short, insn); + insn->immediate.nbytes = 2; + break; + case INAT_IMM_DWORD: + insn->immediate.value = get_next(int, insn); + insn->immediate.nbytes = 4; + break; + case INAT_IMM_QWORD: + insn->immediate1.value = get_next(int, insn); + insn->immediate1.nbytes = 4; + insn->immediate2.value = get_next(int, insn); + insn->immediate2.nbytes = 4; + break; + case INAT_IMM_PTR: + if (!__get_immptr(insn)) + goto err_out; + break; + case INAT_IMM_VWORD32: + if (!__get_immv32(insn)) + goto err_out; + break; + case INAT_IMM_VWORD: + if (!__get_immv(insn)) + goto err_out; + break; + default: + /* Here, insn must have an immediate, but failed */ + goto err_out; + } + if (inat_has_second_immediate(insn->attr)) { + insn->immediate2.value = get_next(char, insn); + insn->immediate2.nbytes = 1; + } +done: + insn->immediate.got = 1; + +err_out: + return; +} + +/** + * insn_get_length() - Get the length of instruction + * @insn: &struct insn containing instruction + * + * If necessary, first collects the instruction up to and including the + * immediates bytes. + */ +void insn_get_length(struct insn *insn) +{ + if (insn->length) + return; + if (!insn->immediate.got) + insn_get_immediate(insn); + insn->length = (unsigned char)((unsigned long)insn->next_byte + - (unsigned long)insn->kaddr); +} diff --git a/tools/stacktool/arch/x86/insn/insn.h b/tools/stacktool/arch/x86/insn/insn.h new file mode 100644 index 0000000..dd12da0 --- /dev/null +++ b/tools/stacktool/arch/x86/insn/insn.h @@ -0,0 +1,201 @@ +#ifndef _ASM_X86_INSN_H +#define _ASM_X86_INSN_H +/* + * x86 instruction analysis + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright (C) IBM Corporation, 2009 + */ + +/* insn_attr_t is defined in inat.h */ +#include "inat.h" + +struct insn_field { + union { + insn_value_t value; + insn_byte_t bytes[4]; + }; + /* !0 if we've run insn_get_xxx() for this field */ + unsigned char got; + unsigned char nbytes; +}; + +struct insn { + struct insn_field prefixes; /* + * Prefixes + * prefixes.bytes[3]: last prefix + */ + struct insn_field rex_prefix; /* REX prefix */ + struct insn_field vex_prefix; /* VEX prefix */ + struct insn_field opcode; /* + * opcode.bytes[0]: opcode1 + * opcode.bytes[1]: opcode2 + * opcode.bytes[2]: opcode3 + */ + struct insn_field modrm; + struct insn_field sib; + struct insn_field displacement; + union { + struct insn_field immediate; + struct insn_field moffset1; /* for 64bit MOV */ + struct insn_field immediate1; /* for 64bit imm or off16/32 */ + }; + union { + struct insn_field moffset2; /* for 64bit MOV */ + struct insn_field immediate2; /* for 64bit imm or seg16 */ + }; + + insn_attr_t attr; + unsigned char opnd_bytes; + unsigned char addr_bytes; + unsigned char length; + unsigned char x86_64; + + const insn_byte_t *kaddr; /* kernel address of insn to analyze */ + const insn_byte_t *end_kaddr; /* kernel address of last insn in buffer */ + const insn_byte_t *next_byte; +}; + +#define MAX_INSN_SIZE 15 + +#define X86_MODRM_MOD(modrm) (((modrm) & 0xc0) >> 6) +#define X86_MODRM_REG(modrm) (((modrm) & 0x38) >> 3) +#define X86_MODRM_RM(modrm) ((modrm) & 0x07) + +#define X86_SIB_SCALE(sib) (((sib) & 0xc0) >> 6) +#define X86_SIB_INDEX(sib) (((sib) & 0x38) >> 3) +#define X86_SIB_BASE(sib) ((sib) & 0x07) + +#define X86_REX_W(rex) ((rex) & 8) +#define X86_REX_R(rex) ((rex) & 4) +#define X86_REX_X(rex) ((rex) & 2) +#define X86_REX_B(rex) ((rex) & 1) + +/* VEX bit flags */ +#define X86_VEX_W(vex) ((vex) & 0x80) /* VEX3 Byte2 */ +#define X86_VEX_R(vex) ((vex) & 0x80) /* VEX2/3 Byte1 */ +#define X86_VEX_X(vex) ((vex) & 0x40) /* VEX3 Byte1 */ +#define X86_VEX_B(vex) ((vex) & 0x20) /* VEX3 Byte1 */ +#define X86_VEX_L(vex) ((vex) & 0x04) /* VEX3 Byte2, VEX2 Byte1 */ +/* VEX bit fields */ +#define X86_VEX3_M(vex) ((vex) & 0x1f) /* VEX3 Byte1 */ +#define X86_VEX2_M 1 /* VEX2.M always 1 */ +#define X86_VEX_V(vex) (((vex) & 0x78) >> 3) /* VEX3 Byte2, VEX2 Byte1 */ +#define X86_VEX_P(vex) ((vex) & 0x03) /* VEX3 Byte2, VEX2 Byte1 */ +#define X86_VEX_M_MAX 0x1f /* VEX3.M Maximum value */ + +extern void insn_init(struct insn *insn, const void *kaddr, int buf_len, int x86_64); +extern void insn_get_prefixes(struct insn *insn); +extern void insn_get_opcode(struct insn *insn); +extern void insn_get_modrm(struct insn *insn); +extern void insn_get_sib(struct insn *insn); +extern void insn_get_displacement(struct insn *insn); +extern void insn_get_immediate(struct insn *insn); +extern void insn_get_length(struct insn *insn); + +/* Attribute will be determined after getting ModRM (for opcode groups) */ +static inline void insn_get_attribute(struct insn *insn) +{ + insn_get_modrm(insn); +} + +/* Instruction uses RIP-relative addressing */ +extern int insn_rip_relative(struct insn *insn); + +/* Init insn for kernel text */ +static inline void kernel_insn_init(struct insn *insn, + const void *kaddr, int buf_len) +{ +#ifdef CONFIG_X86_64 + insn_init(insn, kaddr, buf_len, 1); +#else /* CONFIG_X86_32 */ + insn_init(insn, kaddr, buf_len, 0); +#endif +} + +static inline int insn_is_avx(struct insn *insn) +{ + if (!insn->prefixes.got) + insn_get_prefixes(insn); + return (insn->vex_prefix.value != 0); +} + +/* Ensure this instruction is decoded completely */ +static inline int insn_complete(struct insn *insn) +{ + return insn->opcode.got && insn->modrm.got && insn->sib.got && + insn->displacement.got && insn->immediate.got; +} + +static inline insn_byte_t insn_vex_m_bits(struct insn *insn) +{ + if (insn->vex_prefix.nbytes == 2) /* 2 bytes VEX */ + return X86_VEX2_M; + else + return X86_VEX3_M(insn->vex_prefix.bytes[1]); +} + +static inline insn_byte_t insn_vex_p_bits(struct insn *insn) +{ + if (insn->vex_prefix.nbytes == 2) /* 2 bytes VEX */ + return X86_VEX_P(insn->vex_prefix.bytes[1]); + else + return X86_VEX_P(insn->vex_prefix.bytes[2]); +} + +/* Get the last prefix id from last prefix or VEX prefix */ +static inline int insn_last_prefix_id(struct insn *insn) +{ + if (insn_is_avx(insn)) + return insn_vex_p_bits(insn); /* VEX_p is a SIMD prefix id */ + + if (insn->prefixes.bytes[3]) + return inat_get_last_prefix_id(insn->prefixes.bytes[3]); + + return 0; +} + +/* Offset of each field from kaddr */ +static inline int insn_offset_rex_prefix(struct insn *insn) +{ + return insn->prefixes.nbytes; +} +static inline int insn_offset_vex_prefix(struct insn *insn) +{ + return insn_offset_rex_prefix(insn) + insn->rex_prefix.nbytes; +} +static inline int insn_offset_opcode(struct insn *insn) +{ + return insn_offset_vex_prefix(insn) + insn->vex_prefix.nbytes; +} +static inline int insn_offset_modrm(struct insn *insn) +{ + return insn_offset_opcode(insn) + insn->opcode.nbytes; +} +static inline int insn_offset_sib(struct insn *insn) +{ + return insn_offset_modrm(insn) + insn->modrm.nbytes; +} +static inline int insn_offset_displacement(struct insn *insn) +{ + return insn_offset_sib(insn) + insn->sib.nbytes; +} +static inline int insn_offset_immediate(struct insn *insn) +{ + return insn_offset_displacement(insn) + insn->displacement.nbytes; +} + +#endif /* _ASM_X86_INSN_H */ diff --git a/tools/stacktool/arch/x86/insn/x86-opcode-map.txt b/tools/stacktool/arch/x86/insn/x86-opcode-map.txt new file mode 100644 index 0000000..d388de7 --- /dev/null +++ b/tools/stacktool/arch/x86/insn/x86-opcode-map.txt @@ -0,0 +1,984 @@ +# x86 Opcode Maps +# +# This is (mostly) based on following documentations. +# - Intel(R) 64 and IA-32 Architectures Software Developer's Manual Vol.2C +# (#326018-047US, June 2013) +# +#<Opcode maps> +# Table: table-name +# Referrer: escaped-name +# AVXcode: avx-code +# opcode: mnemonic|GrpXXX [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...] +# (or) +# opcode: escape # escaped-name +# EndTable +# +#<group maps> +# GrpTable: GrpXXX +# reg: mnemonic [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...] +# EndTable +# +# AVX Superscripts +# (v): this opcode requires VEX prefix. +# (v1): this opcode only supports 128bit VEX. +# +# Last Prefix Superscripts +# - (66): the last prefix is 0x66 +# - (F3): the last prefix is 0xF3 +# - (F2): the last prefix is 0xF2 +# - (!F3) : the last prefix is not 0xF3 (including non-last prefix case) +# - (66&F2): Both 0x66 and 0xF2 prefixes are specified. + +Table: one byte opcode +Referrer: +AVXcode: +# 0x00 - 0x0f +00: ADD Eb,Gb +01: ADD Ev,Gv +02: ADD Gb,Eb +03: ADD Gv,Ev +04: ADD AL,Ib +05: ADD rAX,Iz +06: PUSH ES (i64) +07: POP ES (i64) +08: OR Eb,Gb +09: OR Ev,Gv +0a: OR Gb,Eb +0b: OR Gv,Ev +0c: OR AL,Ib +0d: OR rAX,Iz +0e: PUSH CS (i64) +0f: escape # 2-byte escape +# 0x10 - 0x1f +10: ADC Eb,Gb +11: ADC Ev,Gv +12: ADC Gb,Eb +13: ADC Gv,Ev +14: ADC AL,Ib +15: ADC rAX,Iz +16: PUSH SS (i64) +17: POP SS (i64) +18: SBB Eb,Gb +19: SBB Ev,Gv +1a: SBB Gb,Eb +1b: SBB Gv,Ev +1c: SBB AL,Ib +1d: SBB rAX,Iz +1e: PUSH DS (i64) +1f: POP DS (i64) +# 0x20 - 0x2f +20: AND Eb,Gb +21: AND Ev,Gv +22: AND Gb,Eb +23: AND Gv,Ev +24: AND AL,Ib +25: AND rAx,Iz +26: SEG=ES (Prefix) +27: DAA (i64) +28: SUB Eb,Gb +29: SUB Ev,Gv +2a: SUB Gb,Eb +2b: SUB Gv,Ev +2c: SUB AL,Ib +2d: SUB rAX,Iz +2e: SEG=CS (Prefix) +2f: DAS (i64) +# 0x30 - 0x3f +30: XOR Eb,Gb +31: XOR Ev,Gv +32: XOR Gb,Eb +33: XOR Gv,Ev +34: XOR AL,Ib +35: XOR rAX,Iz +36: SEG=SS (Prefix) +37: AAA (i64) +38: CMP Eb,Gb +39: CMP Ev,Gv +3a: CMP Gb,Eb +3b: CMP Gv,Ev +3c: CMP AL,Ib +3d: CMP rAX,Iz +3e: SEG=DS (Prefix) +3f: AAS (i64) +# 0x40 - 0x4f +40: INC eAX (i64) | REX (o64) +41: INC eCX (i64) | REX.B (o64) +42: INC eDX (i64) | REX.X (o64) +43: INC eBX (i64) | REX.XB (o64) +44: INC eSP (i64) | REX.R (o64) +45: INC eBP (i64) | REX.RB (o64) +46: INC eSI (i64) | REX.RX (o64) +47: INC eDI (i64) | REX.RXB (o64) +48: DEC eAX (i64) | REX.W (o64) +49: DEC eCX (i64) | REX.WB (o64) +4a: DEC eDX (i64) | REX.WX (o64) +4b: DEC eBX (i64) | REX.WXB (o64) +4c: DEC eSP (i64) | REX.WR (o64) +4d: DEC eBP (i64) | REX.WRB (o64) +4e: DEC eSI (i64) | REX.WRX (o64) +4f: DEC eDI (i64) | REX.WRXB (o64) +# 0x50 - 0x5f +50: PUSH rAX/r8 (d64) +51: PUSH rCX/r9 (d64) +52: PUSH rDX/r10 (d64) +53: PUSH rBX/r11 (d64) +54: PUSH rSP/r12 (d64) +55: PUSH rBP/r13 (d64) +56: PUSH rSI/r14 (d64) +57: PUSH rDI/r15 (d64) +58: POP rAX/r8 (d64) +59: POP rCX/r9 (d64) +5a: POP rDX/r10 (d64) +5b: POP rBX/r11 (d64) +5c: POP rSP/r12 (d64) +5d: POP rBP/r13 (d64) +5e: POP rSI/r14 (d64) +5f: POP rDI/r15 (d64) +# 0x60 - 0x6f +60: PUSHA/PUSHAD (i64) +61: POPA/POPAD (i64) +62: BOUND Gv,Ma (i64) +63: ARPL Ew,Gw (i64) | MOVSXD Gv,Ev (o64) +64: SEG=FS (Prefix) +65: SEG=GS (Prefix) +66: Operand-Size (Prefix) +67: Address-Size (Prefix) +68: PUSH Iz (d64) +69: IMUL Gv,Ev,Iz +6a: PUSH Ib (d64) +6b: IMUL Gv,Ev,Ib +6c: INS/INSB Yb,DX +6d: INS/INSW/INSD Yz,DX +6e: OUTS/OUTSB DX,Xb +6f: OUTS/OUTSW/OUTSD DX,Xz +# 0x70 - 0x7f +70: JO Jb +71: JNO Jb +72: JB/JNAE/JC Jb +73: JNB/JAE/JNC Jb +74: JZ/JE Jb +75: JNZ/JNE Jb +76: JBE/JNA Jb +77: JNBE/JA Jb +78: JS Jb +79: JNS Jb +7a: JP/JPE Jb +7b: JNP/JPO Jb +7c: JL/JNGE Jb +7d: JNL/JGE Jb +7e: JLE/JNG Jb +7f: JNLE/JG Jb +# 0x80 - 0x8f +80: Grp1 Eb,Ib (1A) +81: Grp1 Ev,Iz (1A) +82: Grp1 Eb,Ib (1A),(i64) +83: Grp1 Ev,Ib (1A) +84: TEST Eb,Gb +85: TEST Ev,Gv +86: XCHG Eb,Gb +87: XCHG Ev,Gv +88: MOV Eb,Gb +89: MOV Ev,Gv +8a: MOV Gb,Eb +8b: MOV Gv,Ev +8c: MOV Ev,Sw +8d: LEA Gv,M +8e: MOV Sw,Ew +8f: Grp1A (1A) | POP Ev (d64) +# 0x90 - 0x9f +90: NOP | PAUSE (F3) | XCHG r8,rAX +91: XCHG rCX/r9,rAX +92: XCHG rDX/r10,rAX +93: XCHG rBX/r11,rAX +94: XCHG rSP/r12,rAX +95: XCHG rBP/r13,rAX +96: XCHG rSI/r14,rAX +97: XCHG rDI/r15,rAX +98: CBW/CWDE/CDQE +99: CWD/CDQ/CQO +9a: CALLF Ap (i64) +9b: FWAIT/WAIT +9c: PUSHF/D/Q Fv (d64) +9d: POPF/D/Q Fv (d64) +9e: SAHF +9f: LAHF +# 0xa0 - 0xaf +a0: MOV AL,Ob +a1: MOV rAX,Ov +a2: MOV Ob,AL +a3: MOV Ov,rAX +a4: MOVS/B Yb,Xb +a5: MOVS/W/D/Q Yv,Xv +a6: CMPS/B Xb,Yb +a7: CMPS/W/D Xv,Yv +a8: TEST AL,Ib +a9: TEST rAX,Iz +aa: STOS/B Yb,AL +ab: STOS/W/D/Q Yv,rAX +ac: LODS/B AL,Xb +ad: LODS/W/D/Q rAX,Xv +ae: SCAS/B AL,Yb +# Note: The May 2011 Intel manual shows Xv for the second parameter of the +# next instruction but Yv is correct +af: SCAS/W/D/Q rAX,Yv +# 0xb0 - 0xbf +b0: MOV AL/R8L,Ib +b1: MOV CL/R9L,Ib +b2: MOV DL/R10L,Ib +b3: MOV BL/R11L,Ib +b4: MOV AH/R12L,Ib +b5: MOV CH/R13L,Ib +b6: MOV DH/R14L,Ib +b7: MOV BH/R15L,Ib +b8: MOV rAX/r8,Iv +b9: MOV rCX/r9,Iv +ba: MOV rDX/r10,Iv +bb: MOV rBX/r11,Iv +bc: MOV rSP/r12,Iv +bd: MOV rBP/r13,Iv +be: MOV rSI/r14,Iv +bf: MOV rDI/r15,Iv +# 0xc0 - 0xcf +c0: Grp2 Eb,Ib (1A) +c1: Grp2 Ev,Ib (1A) +c2: RETN Iw (f64) +c3: RETN +c4: LES Gz,Mp (i64) | VEX+2byte (Prefix) +c5: LDS Gz,Mp (i64) | VEX+1byte (Prefix) +c6: Grp11A Eb,Ib (1A) +c7: Grp11B Ev,Iz (1A) +c8: ENTER Iw,Ib +c9: LEAVE (d64) +ca: RETF Iw +cb: RETF +cc: INT3 +cd: INT Ib +ce: INTO (i64) +cf: IRET/D/Q +# 0xd0 - 0xdf +d0: Grp2 Eb,1 (1A) +d1: Grp2 Ev,1 (1A) +d2: Grp2 Eb,CL (1A) +d3: Grp2 Ev,CL (1A) +d4: AAM Ib (i64) +d5: AAD Ib (i64) +d6: +d7: XLAT/XLATB +d8: ESC +d9: ESC +da: ESC +db: ESC +dc: ESC +dd: ESC +de: ESC +df: ESC +# 0xe0 - 0xef +# Note: "forced64" is Intel CPU behavior: they ignore 0x66 prefix +# in 64-bit mode. AMD CPUs accept 0x66 prefix, it causes RIP truncation +# to 16 bits. In 32-bit mode, 0x66 is accepted by both Intel and AMD. +e0: LOOPNE/LOOPNZ Jb (f64) +e1: LOOPE/LOOPZ Jb (f64) +e2: LOOP Jb (f64) +e3: JrCXZ Jb (f64) +e4: IN AL,Ib +e5: IN eAX,Ib +e6: OUT Ib,AL +e7: OUT Ib,eAX +# With 0x66 prefix in 64-bit mode, for AMD CPUs immediate offset +# in "near" jumps and calls is 16-bit. For CALL, +# push of return address is 16-bit wide, RSP is decremented by 2 +# but is not truncated to 16 bits, unlike RIP. +e8: CALL Jz (f64) +e9: JMP-near Jz (f64) +ea: JMP-far Ap (i64) +eb: JMP-short Jb (f64) +ec: IN AL,DX +ed: IN eAX,DX +ee: OUT DX,AL +ef: OUT DX,eAX +# 0xf0 - 0xff +f0: LOCK (Prefix) +f1: +f2: REPNE (Prefix) | XACQUIRE (Prefix) +f3: REP/REPE (Prefix) | XRELEASE (Prefix) +f4: HLT +f5: CMC +f6: Grp3_1 Eb (1A) +f7: Grp3_2 Ev (1A) +f8: CLC +f9: STC +fa: CLI +fb: STI +fc: CLD +fd: STD +fe: Grp4 (1A) +ff: Grp5 (1A) +EndTable + +Table: 2-byte opcode (0x0f) +Referrer: 2-byte escape +AVXcode: 1 +# 0x0f 0x00-0x0f +00: Grp6 (1A) +01: Grp7 (1A) +02: LAR Gv,Ew +03: LSL Gv,Ew +04: +05: SYSCALL (o64) +06: CLTS +07: SYSRET (o64) +08: INVD +09: WBINVD +0a: +0b: UD2 (1B) +0c: +# AMD's prefetch group. Intel supports prefetchw(/1) only. +0d: GrpP +0e: FEMMS +# 3DNow! uses the last imm byte as opcode extension. +0f: 3DNow! Pq,Qq,Ib +# 0x0f 0x10-0x1f +# NOTE: According to Intel SDM opcode map, vmovups and vmovupd has no operands +# but it actually has operands. And also, vmovss and vmovsd only accept 128bit. +# MOVSS/MOVSD has too many forms(3) on SDM. This map just shows a typical form. +# Many AVX instructions lack v1 superscript, according to Intel AVX-Prgramming +# Reference A.1 +10: vmovups Vps,Wps | vmovupd Vpd,Wpd (66) | vmovss Vx,Hx,Wss (F3),(v1) | vmovsd Vx,Hx,Wsd (F2),(v1) +11: vmovups Wps,Vps | vmovupd Wpd,Vpd (66) | vmovss Wss,Hx,Vss (F3),(v1) | vmovsd Wsd,Hx,Vsd (F2),(v1) +12: vmovlps Vq,Hq,Mq (v1) | vmovhlps Vq,Hq,Uq (v1) | vmovlpd Vq,Hq,Mq (66),(v1) | vmovsldup Vx,Wx (F3) | vmovddup Vx,Wx (F2) +13: vmovlps Mq,Vq (v1) | vmovlpd Mq,Vq (66),(v1) +14: vunpcklps Vx,Hx,Wx | vunpcklpd Vx,Hx,Wx (66) +15: vunpckhps Vx,Hx,Wx | vunpckhpd Vx,Hx,Wx (66) +16: vmovhps Vdq,Hq,Mq (v1) | vmovlhps Vdq,Hq,Uq (v1) | vmovhpd Vdq,Hq,Mq (66),(v1) | vmovshdup Vx,Wx (F3) +17: vmovhps Mq,Vq (v1) | vmovhpd Mq,Vq (66),(v1) +18: Grp16 (1A) +19: +# Intel SDM opcode map does not list MPX instructions. For now using Gv for +# bnd registers and Ev for everything else is OK because the instruction +# decoder does not use the information except as an indication that there is +# a ModR/M byte. +1a: BNDCL Gv,Ev (F3) | BNDCU Gv,Ev (F2) | BNDMOV Gv,Ev (66) | BNDLDX Gv,Ev +1b: BNDCN Gv,Ev (F2) | BNDMOV Ev,Gv (66) | BNDMK Gv,Ev (F3) | BNDSTX Ev,Gv +1c: +1d: +1e: +1f: NOP Ev +# 0x0f 0x20-0x2f +20: MOV Rd,Cd +21: MOV Rd,Dd +22: MOV Cd,Rd +23: MOV Dd,Rd +24: +25: +26: +27: +28: vmovaps Vps,Wps | vmovapd Vpd,Wpd (66) +29: vmovaps Wps,Vps | vmovapd Wpd,Vpd (66) +2a: cvtpi2ps Vps,Qpi | cvtpi2pd Vpd,Qpi (66) | vcvtsi2ss Vss,Hss,Ey (F3),(v1) | vcvtsi2sd Vsd,Hsd,Ey (F2),(v1) +2b: vmovntps Mps,Vps | vmovntpd Mpd,Vpd (66) +2c: cvttps2pi Ppi,Wps | cvttpd2pi Ppi,Wpd (66) | vcvttss2si Gy,Wss (F3),(v1) | vcvttsd2si Gy,Wsd (F2),(v1) +2d: cvtps2pi Ppi,Wps | cvtpd2pi Qpi,Wpd (66) | vcvtss2si Gy,Wss (F3),(v1) | vcvtsd2si Gy,Wsd (F2),(v1) +2e: vucomiss Vss,Wss (v1) | vucomisd Vsd,Wsd (66),(v1) +2f: vcomiss Vss,Wss (v1) | vcomisd Vsd,Wsd (66),(v1) +# 0x0f 0x30-0x3f +30: WRMSR +31: RDTSC +32: RDMSR +33: RDPMC +34: SYSENTER +35: SYSEXIT +36: +37: GETSEC +38: escape # 3-byte escape 1 +39: +3a: escape # 3-byte escape 2 +3b: +3c: +3d: +3e: +3f: +# 0x0f 0x40-0x4f +40: CMOVO Gv,Ev +41: CMOVNO Gv,Ev +42: CMOVB/C/NAE Gv,Ev +43: CMOVAE/NB/NC Gv,Ev +44: CMOVE/Z Gv,Ev +45: CMOVNE/NZ Gv,Ev +46: CMOVBE/NA Gv,Ev +47: CMOVA/NBE Gv,Ev +48: CMOVS Gv,Ev +49: CMOVNS Gv,Ev +4a: CMOVP/PE Gv,Ev +4b: CMOVNP/PO Gv,Ev +4c: CMOVL/NGE Gv,Ev +4d: CMOVNL/GE Gv,Ev +4e: CMOVLE/NG Gv,Ev +4f: CMOVNLE/G Gv,Ev +# 0x0f 0x50-0x5f +50: vmovmskps Gy,Ups | vmovmskpd Gy,Upd (66) +51: vsqrtps Vps,Wps | vsqrtpd Vpd,Wpd (66) | vsqrtss Vss,Hss,Wss (F3),(v1) | vsqrtsd Vsd,Hsd,Wsd (F2),(v1) +52: vrsqrtps Vps,Wps | vrsqrtss Vss,Hss,Wss (F3),(v1) +53: vrcpps Vps,Wps | vrcpss Vss,Hss,Wss (F3),(v1) +54: vandps Vps,Hps,Wps | vandpd Vpd,Hpd,Wpd (66) +55: vandnps Vps,Hps,Wps | vandnpd Vpd,Hpd,Wpd (66) +56: vorps Vps,Hps,Wps | vorpd Vpd,Hpd,Wpd (66) +57: vxorps Vps,Hps,Wps | vxorpd Vpd,Hpd,Wpd (66) +58: vaddps Vps,Hps,Wps | vaddpd Vpd,Hpd,Wpd (66) | vaddss Vss,Hss,Wss (F3),(v1) | vaddsd Vsd,Hsd,Wsd (F2),(v1) +59: vmulps Vps,Hps,Wps | vmulpd Vpd,Hpd,Wpd (66) | vmulss Vss,Hss,Wss (F3),(v1) | vmulsd Vsd,Hsd,Wsd (F2),(v1) +5a: vcvtps2pd Vpd,Wps | vcvtpd2ps Vps,Wpd (66) | vcvtss2sd Vsd,Hx,Wss (F3),(v1) | vcvtsd2ss Vss,Hx,Wsd (F2),(v1) +5b: vcvtdq2ps Vps,Wdq | vcvtps2dq Vdq,Wps (66) | vcvttps2dq Vdq,Wps (F3) +5c: vsubps Vps,Hps,Wps | vsubpd Vpd,Hpd,Wpd (66) | vsubss Vss,Hss,Wss (F3),(v1) | vsubsd Vsd,Hsd,Wsd (F2),(v1) +5d: vminps Vps,Hps,Wps | vminpd Vpd,Hpd,Wpd (66) | vminss Vss,Hss,Wss (F3),(v1) | vminsd Vsd,Hsd,Wsd (F2),(v1) +5e: vdivps Vps,Hps,Wps | vdivpd Vpd,Hpd,Wpd (66) | vdivss Vss,Hss,Wss (F3),(v1) | vdivsd Vsd,Hsd,Wsd (F2),(v1) +5f: vmaxps Vps,Hps,Wps | vmaxpd Vpd,Hpd,Wpd (66) | vmaxss Vss,Hss,Wss (F3),(v1) | vmaxsd Vsd,Hsd,Wsd (F2),(v1) +# 0x0f 0x60-0x6f +60: punpcklbw Pq,Qd | vpunpcklbw Vx,Hx,Wx (66),(v1) +61: punpcklwd Pq,Qd | vpunpcklwd Vx,Hx,Wx (66),(v1) +62: punpckldq Pq,Qd | vpunpckldq Vx,Hx,Wx (66),(v1) +63: packsswb Pq,Qq | vpacksswb Vx,Hx,Wx (66),(v1) +64: pcmpgtb Pq,Qq | vpcmpgtb Vx,Hx,Wx (66),(v1) +65: pcmpgtw Pq,Qq | vpcmpgtw Vx,Hx,Wx (66),(v1) +66: pcmpgtd Pq,Qq | vpcmpgtd Vx,Hx,Wx (66),(v1) +67: packuswb Pq,Qq | vpackuswb Vx,Hx,Wx (66),(v1) +68: punpckhbw Pq,Qd | vpunpckhbw Vx,Hx,Wx (66),(v1) +69: punpckhwd Pq,Qd | vpunpckhwd Vx,Hx,Wx (66),(v1) +6a: punpckhdq Pq,Qd | vpunpckhdq Vx,Hx,Wx (66),(v1) +6b: packssdw Pq,Qd | vpackssdw Vx,Hx,Wx (66),(v1) +6c: vpunpcklqdq Vx,Hx,Wx (66),(v1) +6d: vpunpckhqdq Vx,Hx,Wx (66),(v1) +6e: movd/q Pd,Ey | vmovd/q Vy,Ey (66),(v1) +6f: movq Pq,Qq | vmovdqa Vx,Wx (66) | vmovdqu Vx,Wx (F3) +# 0x0f 0x70-0x7f +70: pshufw Pq,Qq,Ib | vpshufd Vx,Wx,Ib (66),(v1) | vpshufhw Vx,Wx,Ib (F3),(v1) | vpshuflw Vx,Wx,Ib (F2),(v1) +71: Grp12 (1A) +72: Grp13 (1A) +73: Grp14 (1A) +74: pcmpeqb Pq,Qq | vpcmpeqb Vx,Hx,Wx (66),(v1) +75: pcmpeqw Pq,Qq | vpcmpeqw Vx,Hx,Wx (66),(v1) +76: pcmpeqd Pq,Qq | vpcmpeqd Vx,Hx,Wx (66),(v1) +# Note: Remove (v), because vzeroall and vzeroupper becomes emms without VEX. +77: emms | vzeroupper | vzeroall +78: VMREAD Ey,Gy +79: VMWRITE Gy,Ey +7a: +7b: +7c: vhaddpd Vpd,Hpd,Wpd (66) | vhaddps Vps,Hps,Wps (F2) +7d: vhsubpd Vpd,Hpd,Wpd (66) | vhsubps Vps,Hps,Wps (F2) +7e: movd/q Ey,Pd | vmovd/q Ey,Vy (66),(v1) | vmovq Vq,Wq (F3),(v1) +7f: movq Qq,Pq | vmovdqa Wx,Vx (66) | vmovdqu Wx,Vx (F3) +# 0x0f 0x80-0x8f +# Note: "forced64" is Intel CPU behavior (see comment about CALL insn). +80: JO Jz (f64) +81: JNO Jz (f64) +82: JB/JC/JNAE Jz (f64) +83: JAE/JNB/JNC Jz (f64) +84: JE/JZ Jz (f64) +85: JNE/JNZ Jz (f64) +86: JBE/JNA Jz (f64) +87: JA/JNBE Jz (f64) +88: JS Jz (f64) +89: JNS Jz (f64) +8a: JP/JPE Jz (f64) +8b: JNP/JPO Jz (f64) +8c: JL/JNGE Jz (f64) +8d: JNL/JGE Jz (f64) +8e: JLE/JNG Jz (f64) +8f: JNLE/JG Jz (f64) +# 0x0f 0x90-0x9f +90: SETO Eb +91: SETNO Eb +92: SETB/C/NAE Eb +93: SETAE/NB/NC Eb +94: SETE/Z Eb +95: SETNE/NZ Eb +96: SETBE/NA Eb +97: SETA/NBE Eb +98: SETS Eb +99: SETNS Eb +9a: SETP/PE Eb +9b: SETNP/PO Eb +9c: SETL/NGE Eb +9d: SETNL/GE Eb +9e: SETLE/NG Eb +9f: SETNLE/G Eb +# 0x0f 0xa0-0xaf +a0: PUSH FS (d64) +a1: POP FS (d64) +a2: CPUID +a3: BT Ev,Gv +a4: SHLD Ev,Gv,Ib +a5: SHLD Ev,Gv,CL +a6: GrpPDLK +a7: GrpRNG +a8: PUSH GS (d64) +a9: POP GS (d64) +aa: RSM +ab: BTS Ev,Gv +ac: SHRD Ev,Gv,Ib +ad: SHRD Ev,Gv,CL +ae: Grp15 (1A),(1C) +af: IMUL Gv,Ev +# 0x0f 0xb0-0xbf +b0: CMPXCHG Eb,Gb +b1: CMPXCHG Ev,Gv +b2: LSS Gv,Mp +b3: BTR Ev,Gv +b4: LFS Gv,Mp +b5: LGS Gv,Mp +b6: MOVZX Gv,Eb +b7: MOVZX Gv,Ew +b8: JMPE (!F3) | POPCNT Gv,Ev (F3) +b9: Grp10 (1A) +ba: Grp8 Ev,Ib (1A) +bb: BTC Ev,Gv +bc: BSF Gv,Ev (!F3) | TZCNT Gv,Ev (F3) +bd: BSR Gv,Ev (!F3) | LZCNT Gv,Ev (F3) +be: MOVSX Gv,Eb +bf: MOVSX Gv,Ew +# 0x0f 0xc0-0xcf +c0: XADD Eb,Gb +c1: XADD Ev,Gv +c2: vcmpps Vps,Hps,Wps,Ib | vcmppd Vpd,Hpd,Wpd,Ib (66) | vcmpss Vss,Hss,Wss,Ib (F3),(v1) | vcmpsd Vsd,Hsd,Wsd,Ib (F2),(v1) +c3: movnti My,Gy +c4: pinsrw Pq,Ry/Mw,Ib | vpinsrw Vdq,Hdq,Ry/Mw,Ib (66),(v1) +c5: pextrw Gd,Nq,Ib | vpextrw Gd,Udq,Ib (66),(v1) +c6: vshufps Vps,Hps,Wps,Ib | vshufpd Vpd,Hpd,Wpd,Ib (66) +c7: Grp9 (1A) +c8: BSWAP RAX/EAX/R8/R8D +c9: BSWAP RCX/ECX/R9/R9D +ca: BSWAP RDX/EDX/R10/R10D +cb: BSWAP RBX/EBX/R11/R11D +cc: BSWAP RSP/ESP/R12/R12D +cd: BSWAP RBP/EBP/R13/R13D +ce: BSWAP RSI/ESI/R14/R14D +cf: BSWAP RDI/EDI/R15/R15D +# 0x0f 0xd0-0xdf +d0: vaddsubpd Vpd,Hpd,Wpd (66) | vaddsubps Vps,Hps,Wps (F2) +d1: psrlw Pq,Qq | vpsrlw Vx,Hx,Wx (66),(v1) +d2: psrld Pq,Qq | vpsrld Vx,Hx,Wx (66),(v1) +d3: psrlq Pq,Qq | vpsrlq Vx,Hx,Wx (66),(v1) +d4: paddq Pq,Qq | vpaddq Vx,Hx,Wx (66),(v1) +d5: pmullw Pq,Qq | vpmullw Vx,Hx,Wx (66),(v1) +d6: vmovq Wq,Vq (66),(v1) | movq2dq Vdq,Nq (F3) | movdq2q Pq,Uq (F2) +d7: pmovmskb Gd,Nq | vpmovmskb Gd,Ux (66),(v1) +d8: psubusb Pq,Qq | vpsubusb Vx,Hx,Wx (66),(v1) +d9: psubusw Pq,Qq | vpsubusw Vx,Hx,Wx (66),(v1) +da: pminub Pq,Qq | vpminub Vx,Hx,Wx (66),(v1) +db: pand Pq,Qq | vpand Vx,Hx,Wx (66),(v1) +dc: paddusb Pq,Qq | vpaddusb Vx,Hx,Wx (66),(v1) +dd: paddusw Pq,Qq | vpaddusw Vx,Hx,Wx (66),(v1) +de: pmaxub Pq,Qq | vpmaxub Vx,Hx,Wx (66),(v1) +df: pandn Pq,Qq | vpandn Vx,Hx,Wx (66),(v1) +# 0x0f 0xe0-0xef +e0: pavgb Pq,Qq | vpavgb Vx,Hx,Wx (66),(v1) +e1: psraw Pq,Qq | vpsraw Vx,Hx,Wx (66),(v1) +e2: psrad Pq,Qq | vpsrad Vx,Hx,Wx (66),(v1) +e3: pavgw Pq,Qq | vpavgw Vx,Hx,Wx (66),(v1) +e4: pmulhuw Pq,Qq | vpmulhuw Vx,Hx,Wx (66),(v1) +e5: pmulhw Pq,Qq | vpmulhw Vx,Hx,Wx (66),(v1) +e6: vcvttpd2dq Vx,Wpd (66) | vcvtdq2pd Vx,Wdq (F3) | vcvtpd2dq Vx,Wpd (F2) +e7: movntq Mq,Pq | vmovntdq Mx,Vx (66) +e8: psubsb Pq,Qq | vpsubsb Vx,Hx,Wx (66),(v1) +e9: psubsw Pq,Qq | vpsubsw Vx,Hx,Wx (66),(v1) +ea: pminsw Pq,Qq | vpminsw Vx,Hx,Wx (66),(v1) +eb: por Pq,Qq | vpor Vx,Hx,Wx (66),(v1) +ec: paddsb Pq,Qq | vpaddsb Vx,Hx,Wx (66),(v1) +ed: paddsw Pq,Qq | vpaddsw Vx,Hx,Wx (66),(v1) +ee: pmaxsw Pq,Qq | vpmaxsw Vx,Hx,Wx (66),(v1) +ef: pxor Pq,Qq | vpxor Vx,Hx,Wx (66),(v1) +# 0x0f 0xf0-0xff +f0: vlddqu Vx,Mx (F2) +f1: psllw Pq,Qq | vpsllw Vx,Hx,Wx (66),(v1) +f2: pslld Pq,Qq | vpslld Vx,Hx,Wx (66),(v1) +f3: psllq Pq,Qq | vpsllq Vx,Hx,Wx (66),(v1) +f4: pmuludq Pq,Qq | vpmuludq Vx,Hx,Wx (66),(v1) +f5: pmaddwd Pq,Qq | vpmaddwd Vx,Hx,Wx (66),(v1) +f6: psadbw Pq,Qq | vpsadbw Vx,Hx,Wx (66),(v1) +f7: maskmovq Pq,Nq | vmaskmovdqu Vx,Ux (66),(v1) +f8: psubb Pq,Qq | vpsubb Vx,Hx,Wx (66),(v1) +f9: psubw Pq,Qq | vpsubw Vx,Hx,Wx (66),(v1) +fa: psubd Pq,Qq | vpsubd Vx,Hx,Wx (66),(v1) +fb: psubq Pq,Qq | vpsubq Vx,Hx,Wx (66),(v1) +fc: paddb Pq,Qq | vpaddb Vx,Hx,Wx (66),(v1) +fd: paddw Pq,Qq | vpaddw Vx,Hx,Wx (66),(v1) +fe: paddd Pq,Qq | vpaddd Vx,Hx,Wx (66),(v1) +ff: +EndTable + +Table: 3-byte opcode 1 (0x0f 0x38) +Referrer: 3-byte escape 1 +AVXcode: 2 +# 0x0f 0x38 0x00-0x0f +00: pshufb Pq,Qq | vpshufb Vx,Hx,Wx (66),(v1) +01: phaddw Pq,Qq | vphaddw Vx,Hx,Wx (66),(v1) +02: phaddd Pq,Qq | vphaddd Vx,Hx,Wx (66),(v1) +03: phaddsw Pq,Qq | vphaddsw Vx,Hx,Wx (66),(v1) +04: pmaddubsw Pq,Qq | vpmaddubsw Vx,Hx,Wx (66),(v1) +05: phsubw Pq,Qq | vphsubw Vx,Hx,Wx (66),(v1) +06: phsubd Pq,Qq | vphsubd Vx,Hx,Wx (66),(v1) +07: phsubsw Pq,Qq | vphsubsw Vx,Hx,Wx (66),(v1) +08: psignb Pq,Qq | vpsignb Vx,Hx,Wx (66),(v1) +09: psignw Pq,Qq | vpsignw Vx,Hx,Wx (66),(v1) +0a: psignd Pq,Qq | vpsignd Vx,Hx,Wx (66),(v1) +0b: pmulhrsw Pq,Qq | vpmulhrsw Vx,Hx,Wx (66),(v1) +0c: vpermilps Vx,Hx,Wx (66),(v) +0d: vpermilpd Vx,Hx,Wx (66),(v) +0e: vtestps Vx,Wx (66),(v) +0f: vtestpd Vx,Wx (66),(v) +# 0x0f 0x38 0x10-0x1f +10: pblendvb Vdq,Wdq (66) +11: +12: +13: vcvtph2ps Vx,Wx,Ib (66),(v) +14: blendvps Vdq,Wdq (66) +15: blendvpd Vdq,Wdq (66) +16: vpermps Vqq,Hqq,Wqq (66),(v) +17: vptest Vx,Wx (66) +18: vbroadcastss Vx,Wd (66),(v) +19: vbroadcastsd Vqq,Wq (66),(v) +1a: vbroadcastf128 Vqq,Mdq (66),(v) +1b: +1c: pabsb Pq,Qq | vpabsb Vx,Wx (66),(v1) +1d: pabsw Pq,Qq | vpabsw Vx,Wx (66),(v1) +1e: pabsd Pq,Qq | vpabsd Vx,Wx (66),(v1) +1f: +# 0x0f 0x38 0x20-0x2f +20: vpmovsxbw Vx,Ux/Mq (66),(v1) +21: vpmovsxbd Vx,Ux/Md (66),(v1) +22: vpmovsxbq Vx,Ux/Mw (66),(v1) +23: vpmovsxwd Vx,Ux/Mq (66),(v1) +24: vpmovsxwq Vx,Ux/Md (66),(v1) +25: vpmovsxdq Vx,Ux/Mq (66),(v1) +26: +27: +28: vpmuldq Vx,Hx,Wx (66),(v1) +29: vpcmpeqq Vx,Hx,Wx (66),(v1) +2a: vmovntdqa Vx,Mx (66),(v1) +2b: vpackusdw Vx,Hx,Wx (66),(v1) +2c: vmaskmovps Vx,Hx,Mx (66),(v) +2d: vmaskmovpd Vx,Hx,Mx (66),(v) +2e: vmaskmovps Mx,Hx,Vx (66),(v) +2f: vmaskmovpd Mx,Hx,Vx (66),(v) +# 0x0f 0x38 0x30-0x3f +30: vpmovzxbw Vx,Ux/Mq (66),(v1) +31: vpmovzxbd Vx,Ux/Md (66),(v1) +32: vpmovzxbq Vx,Ux/Mw (66),(v1) +33: vpmovzxwd Vx,Ux/Mq (66),(v1) +34: vpmovzxwq Vx,Ux/Md (66),(v1) +35: vpmovzxdq Vx,Ux/Mq (66),(v1) +36: vpermd Vqq,Hqq,Wqq (66),(v) +37: vpcmpgtq Vx,Hx,Wx (66),(v1) +38: vpminsb Vx,Hx,Wx (66),(v1) +39: vpminsd Vx,Hx,Wx (66),(v1) +3a: vpminuw Vx,Hx,Wx (66),(v1) +3b: vpminud Vx,Hx,Wx (66),(v1) +3c: vpmaxsb Vx,Hx,Wx (66),(v1) +3d: vpmaxsd Vx,Hx,Wx (66),(v1) +3e: vpmaxuw Vx,Hx,Wx (66),(v1) +3f: vpmaxud Vx,Hx,Wx (66),(v1) +# 0x0f 0x38 0x40-0x8f +40: vpmulld Vx,Hx,Wx (66),(v1) +41: vphminposuw Vdq,Wdq (66),(v1) +42: +43: +44: +45: vpsrlvd/q Vx,Hx,Wx (66),(v) +46: vpsravd Vx,Hx,Wx (66),(v) +47: vpsllvd/q Vx,Hx,Wx (66),(v) +# Skip 0x48-0x57 +58: vpbroadcastd Vx,Wx (66),(v) +59: vpbroadcastq Vx,Wx (66),(v) +5a: vbroadcasti128 Vqq,Mdq (66),(v) +# Skip 0x5b-0x77 +78: vpbroadcastb Vx,Wx (66),(v) +79: vpbroadcastw Vx,Wx (66),(v) +# Skip 0x7a-0x7f +80: INVEPT Gy,Mdq (66) +81: INVPID Gy,Mdq (66) +82: INVPCID Gy,Mdq (66) +8c: vpmaskmovd/q Vx,Hx,Mx (66),(v) +8e: vpmaskmovd/q Mx,Vx,Hx (66),(v) +# 0x0f 0x38 0x90-0xbf (FMA) +90: vgatherdd/q Vx,Hx,Wx (66),(v) +91: vgatherqd/q Vx,Hx,Wx (66),(v) +92: vgatherdps/d Vx,Hx,Wx (66),(v) +93: vgatherqps/d Vx,Hx,Wx (66),(v) +94: +95: +96: vfmaddsub132ps/d Vx,Hx,Wx (66),(v) +97: vfmsubadd132ps/d Vx,Hx,Wx (66),(v) +98: vfmadd132ps/d Vx,Hx,Wx (66),(v) +99: vfmadd132ss/d Vx,Hx,Wx (66),(v),(v1) +9a: vfmsub132ps/d Vx,Hx,Wx (66),(v) +9b: vfmsub132ss/d Vx,Hx,Wx (66),(v),(v1) +9c: vfnmadd132ps/d Vx,Hx,Wx (66),(v) +9d: vfnmadd132ss/d Vx,Hx,Wx (66),(v),(v1) +9e: vfnmsub132ps/d Vx,Hx,Wx (66),(v) +9f: vfnmsub132ss/d Vx,Hx,Wx (66),(v),(v1) +a6: vfmaddsub213ps/d Vx,Hx,Wx (66),(v) +a7: vfmsubadd213ps/d Vx,Hx,Wx (66),(v) +a8: vfmadd213ps/d Vx,Hx,Wx (66),(v) +a9: vfmadd213ss/d Vx,Hx,Wx (66),(v),(v1) +aa: vfmsub213ps/d Vx,Hx,Wx (66),(v) +ab: vfmsub213ss/d Vx,Hx,Wx (66),(v),(v1) +ac: vfnmadd213ps/d Vx,Hx,Wx (66),(v) +ad: vfnmadd213ss/d Vx,Hx,Wx (66),(v),(v1) +ae: vfnmsub213ps/d Vx,Hx,Wx (66),(v) +af: vfnmsub213ss/d Vx,Hx,Wx (66),(v),(v1) +b6: vfmaddsub231ps/d Vx,Hx,Wx (66),(v) +b7: vfmsubadd231ps/d Vx,Hx,Wx (66),(v) +b8: vfmadd231ps/d Vx,Hx,Wx (66),(v) +b9: vfmadd231ss/d Vx,Hx,Wx (66),(v),(v1) +ba: vfmsub231ps/d Vx,Hx,Wx (66),(v) +bb: vfmsub231ss/d Vx,Hx,Wx (66),(v),(v1) +bc: vfnmadd231ps/d Vx,Hx,Wx (66),(v) +bd: vfnmadd231ss/d Vx,Hx,Wx (66),(v),(v1) +be: vfnmsub231ps/d Vx,Hx,Wx (66),(v) +bf: vfnmsub231ss/d Vx,Hx,Wx (66),(v),(v1) +# 0x0f 0x38 0xc0-0xff +c8: sha1nexte Vdq,Wdq +c9: sha1msg1 Vdq,Wdq +ca: sha1msg2 Vdq,Wdq +cb: sha256rnds2 Vdq,Wdq +cc: sha256msg1 Vdq,Wdq +cd: sha256msg2 Vdq,Wdq +db: VAESIMC Vdq,Wdq (66),(v1) +dc: VAESENC Vdq,Hdq,Wdq (66),(v1) +dd: VAESENCLAST Vdq,Hdq,Wdq (66),(v1) +de: VAESDEC Vdq,Hdq,Wdq (66),(v1) +df: VAESDECLAST Vdq,Hdq,Wdq (66),(v1) +f0: MOVBE Gy,My | MOVBE Gw,Mw (66) | CRC32 Gd,Eb (F2) | CRC32 Gd,Eb (66&F2) +f1: MOVBE My,Gy | MOVBE Mw,Gw (66) | CRC32 Gd,Ey (F2) | CRC32 Gd,Ew (66&F2) +f2: ANDN Gy,By,Ey (v) +f3: Grp17 (1A) +f5: BZHI Gy,Ey,By (v) | PEXT Gy,By,Ey (F3),(v) | PDEP Gy,By,Ey (F2),(v) +f6: ADCX Gy,Ey (66) | ADOX Gy,Ey (F3) | MULX By,Gy,rDX,Ey (F2),(v) +f7: BEXTR Gy,Ey,By (v) | SHLX Gy,Ey,By (66),(v) | SARX Gy,Ey,By (F3),(v) | SHRX Gy,Ey,By (F2),(v) +EndTable + +Table: 3-byte opcode 2 (0x0f 0x3a) +Referrer: 3-byte escape 2 +AVXcode: 3 +# 0x0f 0x3a 0x00-0xff +00: vpermq Vqq,Wqq,Ib (66),(v) +01: vpermpd Vqq,Wqq,Ib (66),(v) +02: vpblendd Vx,Hx,Wx,Ib (66),(v) +03: +04: vpermilps Vx,Wx,Ib (66),(v) +05: vpermilpd Vx,Wx,Ib (66),(v) +06: vperm2f128 Vqq,Hqq,Wqq,Ib (66),(v) +07: +08: vroundps Vx,Wx,Ib (66) +09: vroundpd Vx,Wx,Ib (66) +0a: vroundss Vss,Wss,Ib (66),(v1) +0b: vroundsd Vsd,Wsd,Ib (66),(v1) +0c: vblendps Vx,Hx,Wx,Ib (66) +0d: vblendpd Vx,Hx,Wx,Ib (66) +0e: vpblendw Vx,Hx,Wx,Ib (66),(v1) +0f: palignr Pq,Qq,Ib | vpalignr Vx,Hx,Wx,Ib (66),(v1) +14: vpextrb Rd/Mb,Vdq,Ib (66),(v1) +15: vpextrw Rd/Mw,Vdq,Ib (66),(v1) +16: vpextrd/q Ey,Vdq,Ib (66),(v1) +17: vextractps Ed,Vdq,Ib (66),(v1) +18: vinsertf128 Vqq,Hqq,Wqq,Ib (66),(v) +19: vextractf128 Wdq,Vqq,Ib (66),(v) +1d: vcvtps2ph Wx,Vx,Ib (66),(v) +20: vpinsrb Vdq,Hdq,Ry/Mb,Ib (66),(v1) +21: vinsertps Vdq,Hdq,Udq/Md,Ib (66),(v1) +22: vpinsrd/q Vdq,Hdq,Ey,Ib (66),(v1) +38: vinserti128 Vqq,Hqq,Wqq,Ib (66),(v) +39: vextracti128 Wdq,Vqq,Ib (66),(v) +40: vdpps Vx,Hx,Wx,Ib (66) +41: vdppd Vdq,Hdq,Wdq,Ib (66),(v1) +42: vmpsadbw Vx,Hx,Wx,Ib (66),(v1) +44: vpclmulqdq Vdq,Hdq,Wdq,Ib (66),(v1) +46: vperm2i128 Vqq,Hqq,Wqq,Ib (66),(v) +4a: vblendvps Vx,Hx,Wx,Lx (66),(v) +4b: vblendvpd Vx,Hx,Wx,Lx (66),(v) +4c: vpblendvb Vx,Hx,Wx,Lx (66),(v1) +60: vpcmpestrm Vdq,Wdq,Ib (66),(v1) +61: vpcmpestri Vdq,Wdq,Ib (66),(v1) +62: vpcmpistrm Vdq,Wdq,Ib (66),(v1) +63: vpcmpistri Vdq,Wdq,Ib (66),(v1) +cc: sha1rnds4 Vdq,Wdq,Ib +df: VAESKEYGEN Vdq,Wdq,Ib (66),(v1) +f0: RORX Gy,Ey,Ib (F2),(v) +EndTable + +GrpTable: Grp1 +0: ADD +1: OR +2: ADC +3: SBB +4: AND +5: SUB +6: XOR +7: CMP +EndTable + +GrpTable: Grp1A +0: POP +EndTable + +GrpTable: Grp2 +0: ROL +1: ROR +2: RCL +3: RCR +4: SHL/SAL +5: SHR +6: +7: SAR +EndTable + +GrpTable: Grp3_1 +0: TEST Eb,Ib +1: +2: NOT Eb +3: NEG Eb +4: MUL AL,Eb +5: IMUL AL,Eb +6: DIV AL,Eb +7: IDIV AL,Eb +EndTable + +GrpTable: Grp3_2 +0: TEST Ev,Iz +1: +2: NOT Ev +3: NEG Ev +4: MUL rAX,Ev +5: IMUL rAX,Ev +6: DIV rAX,Ev +7: IDIV rAX,Ev +EndTable + +GrpTable: Grp4 +0: INC Eb +1: DEC Eb +EndTable + +GrpTable: Grp5 +0: INC Ev +1: DEC Ev +# Note: "forced64" is Intel CPU behavior (see comment about CALL insn). +2: CALLN Ev (f64) +3: CALLF Ep +4: JMPN Ev (f64) +5: JMPF Mp +6: PUSH Ev (d64) +7: +EndTable + +GrpTable: Grp6 +0: SLDT Rv/Mw +1: STR Rv/Mw +2: LLDT Ew +3: LTR Ew +4: VERR Ew +5: VERW Ew +EndTable + +GrpTable: Grp7 +0: SGDT Ms | VMCALL (001),(11B) | VMLAUNCH (010),(11B) | VMRESUME (011),(11B) | VMXOFF (100),(11B) +1: SIDT Ms | MONITOR (000),(11B) | MWAIT (001),(11B) | CLAC (010),(11B) | STAC (011),(11B) +2: LGDT Ms | XGETBV (000),(11B) | XSETBV (001),(11B) | VMFUNC (100),(11B) | XEND (101)(11B) | XTEST (110)(11B) +3: LIDT Ms +4: SMSW Mw/Rv +5: rdpkru (110),(11B) | wrpkru (111),(11B) +6: LMSW Ew +7: INVLPG Mb | SWAPGS (o64),(000),(11B) | RDTSCP (001),(11B) +EndTable + +GrpTable: Grp8 +4: BT +5: BTS +6: BTR +7: BTC +EndTable + +GrpTable: Grp9 +1: CMPXCHG8B/16B Mq/Mdq +3: xrstors +4: xsavec +5: xsaves +6: VMPTRLD Mq | VMCLEAR Mq (66) | VMXON Mq (F3) | RDRAND Rv (11B) +7: VMPTRST Mq | VMPTRST Mq (F3) | RDSEED Rv (11B) +EndTable + +GrpTable: Grp10 +EndTable + +# Grp11A and Grp11B are expressed as Grp11 in Intel SDM +GrpTable: Grp11A +0: MOV Eb,Ib +7: XABORT Ib (000),(11B) +EndTable + +GrpTable: Grp11B +0: MOV Eb,Iz +7: XBEGIN Jz (000),(11B) +EndTable + +GrpTable: Grp12 +2: psrlw Nq,Ib (11B) | vpsrlw Hx,Ux,Ib (66),(11B),(v1) +4: psraw Nq,Ib (11B) | vpsraw Hx,Ux,Ib (66),(11B),(v1) +6: psllw Nq,Ib (11B) | vpsllw Hx,Ux,Ib (66),(11B),(v1) +EndTable + +GrpTable: Grp13 +2: psrld Nq,Ib (11B) | vpsrld Hx,Ux,Ib (66),(11B),(v1) +4: psrad Nq,Ib (11B) | vpsrad Hx,Ux,Ib (66),(11B),(v1) +6: pslld Nq,Ib (11B) | vpslld Hx,Ux,Ib (66),(11B),(v1) +EndTable + +GrpTable: Grp14 +2: psrlq Nq,Ib (11B) | vpsrlq Hx,Ux,Ib (66),(11B),(v1) +3: vpsrldq Hx,Ux,Ib (66),(11B),(v1) +6: psllq Nq,Ib (11B) | vpsllq Hx,Ux,Ib (66),(11B),(v1) +7: vpslldq Hx,Ux,Ib (66),(11B),(v1) +EndTable + +GrpTable: Grp15 +0: fxsave | RDFSBASE Ry (F3),(11B) +1: fxstor | RDGSBASE Ry (F3),(11B) +2: vldmxcsr Md (v1) | WRFSBASE Ry (F3),(11B) +3: vstmxcsr Md (v1) | WRGSBASE Ry (F3),(11B) +4: XSAVE +5: XRSTOR | lfence (11B) +6: XSAVEOPT | clwb (66) | mfence (11B) +7: clflush | clflushopt (66) | sfence (11B) | pcommit (66),(11B) +EndTable + +GrpTable: Grp16 +0: prefetch NTA +1: prefetch T0 +2: prefetch T1 +3: prefetch T2 +EndTable + +GrpTable: Grp17 +1: BLSR By,Ey (v) +2: BLSMSK By,Ey (v) +3: BLSI By,Ey (v) +EndTable + +# AMD's Prefetch Group +GrpTable: GrpP +0: PREFETCH +1: PREFETCHW +EndTable + +GrpTable: GrpPDLK +0: MONTMUL +1: XSHA1 +2: XSHA2 +EndTable + +GrpTable: GrpRNG +0: xstore-rng +1: xcrypt-ecb +2: xcrypt-cbc +4: xcrypt-cfb +5: xcrypt-ofb +EndTable diff --git a/tools/stacktool/elf.c b/tools/stacktool/elf.c new file mode 100644 index 0000000..5ff8ac1 --- /dev/null +++ b/tools/stacktool/elf.c @@ -0,0 +1,403 @@ +/* + * elf.c - ELF access library + * + * Adapted from kpatch (https://github.com/dynup/kpatch): + * Copyright (C) 2013-2015 Josh Poimboeuf <jpoimboe@redhat.com> + * Copyright (C) 2014 Seth Jennings <sjenning@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#include <sys/types.h> +#include <sys/stat.h> +#include <fcntl.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <unistd.h> + +#include "elf.h" +#include "warn.h" + +struct section *find_section_by_name(struct elf *elf, const char *name) +{ + struct section *sec; + + list_for_each_entry(sec, &elf->sections, list) + if (!strcmp(sec->name, name)) + return sec; + + return NULL; +} + +static struct section *find_section_by_index(struct elf *elf, + unsigned int index) +{ + struct section *sec; + + list_for_each_entry(sec, &elf->sections, list) + if (sec->index == index) + return sec; + + return NULL; +} + +static struct symbol *find_symbol_by_index(struct elf *elf, unsigned int index) +{ + struct section *sec; + struct symbol *sym; + + list_for_each_entry(sec, &elf->sections, list) + list_for_each_entry(sym, &sec->symbols, list) + if (sym->index == index) + return sym; + + return NULL; +} + +struct symbol *find_symbol_by_offset(struct section *sec, unsigned long offset) +{ + struct symbol *sym; + + list_for_each_entry(sym, &sec->symbols, list) + if (sym->type != STT_SECTION && + sym->offset == offset) + return sym; + + return NULL; +} + +struct rela *find_rela_by_dest_range(struct section *sec, unsigned long offset, + unsigned int len) +{ + struct rela *rela; + + if (!sec->rela) + return NULL; + + list_for_each_entry(rela, &sec->rela->relas, list) + if (rela->offset >= offset && rela->offset < offset + len) + return rela; + + return NULL; +} + +struct rela *find_rela_by_dest(struct section *sec, unsigned long offset) +{ + return find_rela_by_dest_range(sec, offset, 1); +} + +struct symbol *find_containing_func(struct section *sec, unsigned long offset) +{ + struct symbol *func; + + list_for_each_entry(func, &sec->symbols, list) + if (func->type == STT_FUNC && offset >= func->offset && + offset < func->offset + func->len) + return func; + + return NULL; +} + +static int read_sections(struct elf *elf) +{ + Elf_Scn *s = NULL; + struct section *sec; + size_t shstrndx, sections_nr; + int i; + + if (elf_getshdrnum(elf->elf, §ions_nr)) { + perror("elf_getshdrnum"); + return -1; + } + + if (elf_getshdrstrndx(elf->elf, &shstrndx)) { + perror("elf_getshdrstrndx"); + return -1; + } + + for (i = 0; i < sections_nr; i++) { + sec = malloc(sizeof(*sec)); + if (!sec) { + perror("malloc"); + return -1; + } + memset(sec, 0, sizeof(*sec)); + + INIT_LIST_HEAD(&sec->symbols); + INIT_LIST_HEAD(&sec->relas); + + list_add_tail(&sec->list, &elf->sections); + + s = elf_getscn(elf->elf, i); + if (!s) { + perror("elf_getscn"); + return -1; + } + + sec->index = elf_ndxscn(s); + + if (!gelf_getshdr(s, &sec->sh)) { + perror("gelf_getshdr"); + return -1; + } + + sec->name = elf_strptr(elf->elf, shstrndx, sec->sh.sh_name); + if (!sec->name) { + perror("elf_strptr"); + return -1; + } + + sec->elf_data = elf_getdata(s, NULL); + if (!sec->elf_data) { + perror("elf_getdata"); + return -1; + } + + if (sec->elf_data->d_off != 0 || + sec->elf_data->d_size != sec->sh.sh_size) { + WARN("unexpected data attributes for %s", sec->name); + return -1; + } + + sec->data = (unsigned long)sec->elf_data->d_buf; + sec->len = sec->elf_data->d_size; + } + + /* sanity check, one more call to elf_nextscn() should return NULL */ + if (elf_nextscn(elf->elf, s)) { + WARN("section entry mismatch"); + return -1; + } + + return 0; +} + +static int read_symbols(struct elf *elf) +{ + struct section *symtab; + struct symbol *sym; + struct list_head *entry, *tmp; + int symbols_nr, i; + + symtab = find_section_by_name(elf, ".symtab"); + if (!symtab) { + WARN("missing symbol table"); + return -1; + } + + symbols_nr = symtab->sh.sh_size / symtab->sh.sh_entsize; + + for (i = 0; i < symbols_nr; i++) { + sym = malloc(sizeof(*sym)); + if (!sym) { + perror("malloc"); + return -1; + } + memset(sym, 0, sizeof(*sym)); + + sym->index = i; + + if (!gelf_getsym(symtab->elf_data, i, &sym->sym)) { + perror("gelf_getsym"); + goto err; + } + + sym->name = elf_strptr(elf->elf, symtab->sh.sh_link, + sym->sym.st_name); + if (!sym->name) { + perror("elf_strptr"); + goto err; + } + + sym->type = GELF_ST_TYPE(sym->sym.st_info); + sym->bind = GELF_ST_BIND(sym->sym.st_info); + + if (sym->sym.st_shndx > SHN_UNDEF && + sym->sym.st_shndx < SHN_LORESERVE) { + sym->sec = find_section_by_index(elf, + sym->sym.st_shndx); + if (!sym->sec) { + WARN("couldn't find section for symbol %s", + sym->name); + goto err; + } + if (sym->type == STT_SECTION) { + sym->name = sym->sec->name; + sym->sec->sym = sym; + } + } else + sym->sec = find_section_by_index(elf, 0); + + sym->offset = sym->sym.st_value; + sym->len = sym->sym.st_size; + + /* sorted insert into a per-section list */ + entry = &sym->sec->symbols; + list_for_each_prev(tmp, &sym->sec->symbols) { + struct symbol *s; + + s = list_entry(tmp, struct symbol, list); + + if (sym->offset > s->offset) { + entry = tmp; + break; + } + + if (sym->offset == s->offset && sym->len >= s->len) { + entry = tmp; + break; + } + } + list_add(&sym->list, entry); + } + + return 0; + +err: + free(sym); + return -1; +} + +static int read_relas(struct elf *elf) +{ + struct section *sec; + struct rela *rela; + int i; + unsigned int symndx; + + list_for_each_entry(sec, &elf->sections, list) { + if (sec->sh.sh_type != SHT_RELA) + continue; + + sec->base = find_section_by_name(elf, sec->name + 5); + if (!sec->base) { + WARN("can't find base section for rela section %s", + sec->name); + return -1; + } + + sec->base->rela = sec; + + for (i = 0; i < sec->sh.sh_size / sec->sh.sh_entsize; i++) { + rela = malloc(sizeof(*rela)); + if (!rela) { + perror("malloc"); + return -1; + } + memset(rela, 0, sizeof(*rela)); + + list_add_tail(&rela->list, &sec->relas); + + if (!gelf_getrela(sec->elf_data, i, &rela->rela)) { + perror("gelf_getrela"); + return -1; + } + + rela->type = GELF_R_TYPE(rela->rela.r_info); + rela->addend = rela->rela.r_addend; + rela->offset = rela->rela.r_offset; + symndx = GELF_R_SYM(rela->rela.r_info); + rela->sym = find_symbol_by_index(elf, symndx); + if (!rela->sym) { + WARN("can't find rela entry symbol %d for %s", + symndx, sec->name); + return -1; + } + } + } + + return 0; +} + +struct elf *elf_open(const char *name) +{ + struct elf *elf; + + elf_version(EV_CURRENT); + + elf = malloc(sizeof(*elf)); + if (!elf) { + perror("malloc"); + return NULL; + } + memset(elf, 0, sizeof(*elf)); + + INIT_LIST_HEAD(&elf->sections); + + elf->name = strdup(name); + if (!elf->name) { + perror("strdup"); + goto err; + } + + elf->fd = open(name, O_RDONLY); + if (elf->fd == -1) { + perror("open"); + goto err; + } + + elf->elf = elf_begin(elf->fd, ELF_C_READ_MMAP, NULL); + if (!elf->elf) { + perror("elf_begin"); + goto err; + } + + if (!gelf_getehdr(elf->elf, &elf->ehdr)) { + perror("gelf_getehdr"); + goto err; + } + + if (read_sections(elf)) + goto err; + + if (read_symbols(elf)) + goto err; + + if (read_relas(elf)) + goto err; + + return elf; + +err: + elf_close(elf); + return NULL; +} + +void elf_close(struct elf *elf) +{ + struct section *sec, *tmpsec; + struct symbol *sym, *tmpsym; + struct rela *rela, *tmprela; + + list_for_each_entry_safe(sec, tmpsec, &elf->sections, list) { + list_for_each_entry_safe(sym, tmpsym, &sec->symbols, list) { + list_del(&sym->list); + free(sym); + } + list_for_each_entry_safe(rela, tmprela, &sec->relas, list) { + list_del(&rela->list); + free(rela); + } + list_del(&sec->list); + free(sec); + } + if (elf->name) + free(elf->name); + if (elf->fd > 0) + close(elf->fd); + if (elf->elf) + elf_end(elf->elf); + free(elf); +} diff --git a/tools/stacktool/elf.h b/tools/stacktool/elf.h new file mode 100644 index 0000000..db068a3 --- /dev/null +++ b/tools/stacktool/elf.h @@ -0,0 +1,79 @@ +/* + * Copyright (C) 2015 Josh Poimboeuf <jpoimboe@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef _STACKTOOL_ELF_H +#define _STACKTOOL_ELF_H + +#include <stdio.h> +#include <gelf.h> +#include <linux/list.h> + +struct section { + struct list_head list; + GElf_Shdr sh; + struct list_head symbols; + struct list_head relas; + struct section *base, *rela; + struct symbol *sym; + Elf_Data *elf_data; + char *name; + int index; + unsigned long data; + unsigned int len; +}; + +struct symbol { + struct list_head list; + GElf_Sym sym; + struct section *sec; + char *name; + int index; + unsigned char bind, type; + unsigned long offset; + unsigned int len; +}; + +struct rela { + struct list_head list; + GElf_Rela rela; + struct symbol *sym; + unsigned int type; + int offset; + int addend; +}; + +struct elf { + Elf *elf; + GElf_Ehdr ehdr; + int fd; + char *name; + struct list_head sections; +}; + + +struct elf *elf_open(const char *name); +struct section *find_section_by_name(struct elf *elf, const char *name); +struct symbol *find_symbol_by_offset(struct section *sec, unsigned long offset); +struct rela *find_rela_by_dest(struct section *sec, unsigned long offset); +struct rela *find_rela_by_dest_range(struct section *sec, unsigned long offset, + unsigned int len); +struct symbol *find_containing_func(struct section *sec, unsigned long offset); +void elf_close(struct elf *elf); + + + +#endif /* _STACKTOOL_ELF_H */ diff --git a/tools/stacktool/special.c b/tools/stacktool/special.c new file mode 100644 index 0000000..cffc74e --- /dev/null +++ b/tools/stacktool/special.c @@ -0,0 +1,199 @@ +/* + * Copyright (C) 2015 Josh Poimboeuf <jpoimboe@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +/* + * This file reads all the special sections which have alternate instructions + * which can be patched in or redirected to at runtime. + */ + +#include <stdlib.h> +#include <string.h> + +#include "special.h" +#include "warn.h" + +#define EX_ENTRY_SIZE 8 +#define EX_ORIG_OFFSET 0 +#define EX_NEW_OFFSET 4 + +#define JUMP_ENTRY_SIZE 24 +#define JUMP_ORIG_OFFSET 0 +#define JUMP_NEW_OFFSET 8 + +#define ALT_ENTRY_SIZE 13 +#define ALT_ORIG_OFFSET 0 +#define ALT_NEW_OFFSET 4 +#define ALT_FEATURE_OFFSET 8 +#define ALT_ORIG_LEN_OFFSET 10 +#define ALT_NEW_LEN_OFFSET 11 + +#define X86_FEATURE_POPCNT 4*32+23 + +struct special_entry { + const char *sec; + bool group, jump_or_nop; + unsigned char size, orig, new; + unsigned char orig_len, new_len; /* group only */ + unsigned char feature; /* ALTERNATIVE macro CPU feature */ +}; + +struct special_entry entries[] = { + { + .sec = ".altinstructions", + .group = true, + .size = ALT_ENTRY_SIZE, + .orig = ALT_ORIG_OFFSET, + .orig_len = ALT_ORIG_LEN_OFFSET, + .new = ALT_NEW_OFFSET, + .new_len = ALT_NEW_LEN_OFFSET, + .feature = ALT_FEATURE_OFFSET, + }, + { + .sec = "__jump_table", + .jump_or_nop = true, + .size = JUMP_ENTRY_SIZE, + .orig = JUMP_ORIG_OFFSET, + .new = JUMP_NEW_OFFSET, + }, + { + .sec = "__ex_table", + .size = EX_ENTRY_SIZE, + .orig = EX_ORIG_OFFSET, + .new = EX_NEW_OFFSET, + }, + {}, +}; + +static int get_alt_entry(struct elf *elf, struct special_entry *entry, + struct section *sec, int index, + struct special_alt *alt) +{ + struct rela *orig_rela, *new_rela; + unsigned long offset; + + offset = index * entry->size; + + alt->group = entry->group; + alt->jump_or_nop = entry->jump_or_nop; + + if (alt->group) { + alt->orig_len = *(unsigned char *)(sec->data + offset + + entry->orig_len); + alt->new_len = *(unsigned char *)(sec->data + offset + + entry->new_len); + } + + if (entry->feature) { + unsigned short feature; + + feature = *(unsigned short *)(sec->data + offset + + entry->feature); + + /* + * It has been requested that we don't validate the !POPCNT + * feature path which is a "very very small percentage of + * machines". + */ + if (feature == X86_FEATURE_POPCNT) + alt->skip_orig = true; + } + + orig_rela = find_rela_by_dest(sec, offset + entry->orig); + if (!orig_rela) { + WARN_FUNC("can't find orig rela", sec, offset + entry->orig); + return -1; + } + if (orig_rela->sym->type != STT_SECTION) { + WARN_FUNC("don't know how to handle non-section rela symbol %s", + sec, offset + entry->orig, orig_rela->sym->name); + return -1; + } + + alt->orig_sec = orig_rela->sym->sec; + alt->orig_off = orig_rela->addend; + + if (!entry->group || alt->new_len) { + new_rela = find_rela_by_dest(sec, offset + entry->new); + if (!new_rela) { + WARN_FUNC("can't find new rela", + sec, offset + entry->new); + return -1; + } + if (new_rela->sym->type != STT_SECTION) { + WARN_FUNC("don't know how to handle non-section rela symbol %s", + sec, offset + entry->new, + new_rela->sym->name); + return -1; + } + + alt->new_sec = new_rela->sym->sec; + alt->new_off = (unsigned int)new_rela->addend; + + /* _ASM_EXTABLE_EX hack */ + if (alt->new_off >= 0x7ffffff0) + alt->new_off -= 0x7ffffff0; + } + + return 0; +} + +/* + * Read all the special sections and create a list of special_alt structs which + * describe all the alternate instructions which can be patched in or + * redirected to at runtime. + */ +int special_get_alts(struct elf *elf, struct list_head *alts) +{ + struct special_entry *entry; + struct section *sec; + unsigned int nr_entries; + struct special_alt *alt; + int index, ret; + + INIT_LIST_HEAD(alts); + + for (entry = entries; entry->sec; entry++) { + sec = find_section_by_name(elf, entry->sec); + if (!sec) + continue; + + if (sec->len % entry->size != 0) { + WARN("%s size not a multiple of %d", + sec->name, JUMP_ENTRY_SIZE); + return -1; + } + + nr_entries = sec->len / entry->size; + + for (index = 0; index < nr_entries; index++) { + alt = malloc(sizeof(*alt)); + if (!alt) { + WARN("malloc failed"); + return -1; + } + memset(alt, 0, sizeof(*alt)); + + ret = get_alt_entry(elf, entry, sec, index, alt); + if (ret) + return ret; + + list_add_tail(&alt->list, alts); + } + } + + return 0; +} diff --git a/tools/stacktool/special.h b/tools/stacktool/special.h new file mode 100644 index 0000000..fad1d09 --- /dev/null +++ b/tools/stacktool/special.h @@ -0,0 +1,42 @@ +/* + * Copyright (C) 2015 Josh Poimboeuf <jpoimboe@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef _SPECIAL_H +#define _SPECIAL_H + +#include <stdbool.h> +#include "elf.h" + +struct special_alt { + struct list_head list; + + bool group; + bool skip_orig; + bool jump_or_nop; + + struct section *orig_sec; + unsigned long orig_off; + + struct section *new_sec; + unsigned long new_off; + + unsigned int orig_len, new_len; /* group only */ +}; + +int special_get_alts(struct elf *elf, struct list_head *alts); + +#endif /* _SPECIAL_H */ diff --git a/tools/stacktool/stacktool.c b/tools/stacktool/stacktool.c new file mode 100644 index 0000000..0a7f50f --- /dev/null +++ b/tools/stacktool/stacktool.c @@ -0,0 +1,980 @@ +/* + * Copyright (C) 2015 Josh Poimboeuf <jpoimboe@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +/* + * stacktool: + * + * This tool analyzes every .o file and ensures the validity of its stack trace + * metadata. It enforces a set of rules on asm code and C inline assembly code + * so that stack traces can be reliable. + * + * For more information, see tools/stacktool/Documentation/stack-validation.txt. + */ + +#include <argp.h> +#include <stdbool.h> +#include <string.h> +#include <stdlib.h> + +#include "elf.h" +#include "arch.h" +#include "special.h" +#include "warn.h" + +#define STATE_FP_SAVED 0x1 +#define STATE_FP_SETUP 0x2 + +char *objname; + +struct instruction { + struct list_head list; + struct section *sec; + unsigned long offset; + unsigned int len, state; + unsigned char type; + unsigned long immediate; + bool alt_group, visited, ignore; + struct symbol *call_dest; + struct instruction *jump_dest; + struct list_head alts; +}; + +struct list_head insns; + +struct alternative { + struct list_head list; + struct instruction *insn; +}; + +struct args { + char *args[1]; + bool fp; +}; +struct args args; +static const char args_doc[] = "file.o"; +static struct argp_option options[] = { + {"check-frame-pointer", 'f', 0, 0, "Validate frame pointer usage"}, + {0}, +}; +static error_t parse_opt(int key, char *arg, struct argp_state *state) +{ + /* Get the input argument from argp_parse, which we + know is a pointer to our args structure. */ + struct args *a = state->input; + + switch (key) { + case 'f': + a->fp = true; + break; + + case ARGP_KEY_ARG: + if (state->arg_num >= 1) + /* Too many arguments. */ + argp_usage(state); + a->args[state->arg_num] = arg; + break; + + case ARGP_KEY_END: + if (state->arg_num < 1) + /* Not enough arguments. */ + argp_usage(state); + break; + + default: + return ARGP_ERR_UNKNOWN; + } + return 0; +} +static struct argp argp = { options, parse_opt, args_doc, 0 }; + +/* + * Check for the STACKTOOL_IGNORE_INSN macro. + */ +static bool ignore_insn(struct elf *elf, struct instruction *insn) +{ + struct section *macro_sec; + struct rela *rela; + + macro_sec = find_section_by_name(elf, "__stacktool_ignore_insn"); + if (!macro_sec || !macro_sec->rela) + return false; + + list_for_each_entry(rela, ¯o_sec->rela->relas, list) + if (rela->sym->type == STT_SECTION && + rela->sym == insn->sec->sym && + rela->addend == insn->offset) + return true; + + return false; +} + +/* + * Check for the STACKTOOL_IGNORE_FUNC macro. + */ +static bool ignore_func(struct elf *elf, struct symbol *func) +{ + struct section *macro_sec; + struct rela *rela; + + macro_sec = find_section_by_name(elf, "__stacktool_ignore_func"); + if (!macro_sec || !macro_sec->rela) + return false; + + list_for_each_entry(rela, ¯o_sec->rela->relas, list) + if (rela->sym->sec == func->sec && + rela->addend == func->offset) + return true; + + + return false; +} + +static struct instruction *find_instruction(struct section *sec, + unsigned long offset) +{ + struct instruction *insn; + + list_for_each_entry(insn, &insns, list) + if (insn->sec == sec && insn->offset == offset) + return insn; + + return NULL; +} + +/* + * This checks to see if the given function is a "noreturn" function. + * + * For global functions which are outside the scope of this object file, we + * have to keep a manual list of them. + * + * For local functions, we have to detect them manually by simply looking for + * the lack of a return instruction. + */ +static bool dead_end_function(struct symbol *func) +{ + struct instruction *insn; + + if (func->bind == STB_GLOBAL && + (!strcmp(func->name, "__stack_chk_fail") || + !strcmp(func->name, "panic") || + !strcmp(func->name, "do_exit"))) + return true; + + if (!func->sec) + return false; + + insn = find_instruction(func->sec, func->offset); + if (!insn) + return false; + + list_for_each_entry_from(insn, &insns, list) { + if (insn->sec != func->sec || + insn->offset >= func->offset + func->len) + break; + + if (insn->type == INSN_RETURN) + return false; + + if (insn->type == INSN_JUMP_UNCONDITIONAL) { + struct instruction *dest = insn->jump_dest; + struct symbol *dest_func; + + if (!dest) + /* sibling call to another file */ + return false; + + if (dest->sec != func->sec || + dest->offset < func->offset || + dest->offset >= func->offset + func->len) { + /* local sibling call */ + dest_func = find_symbol_by_offset(dest->sec, + dest->offset); + if (!dest_func) + continue; + + return dead_end_function(dest_func); + } + } + + if (insn->type == INSN_JUMP_DYNAMIC) + /* sibling call */ + return false; + } + + return true; +} + +/* + * Call the arch-specific instruction decoder for all the instructions and add + * them to the global insns list. + */ +static int decode_instructions(struct elf *elf) +{ + struct section *sec; + unsigned long offset; + struct instruction *insn; + int ret; + + INIT_LIST_HEAD(&insns); + + list_for_each_entry(sec, &elf->sections, list) { + + if (!(sec->sh.sh_flags & SHF_EXECINSTR)) + continue; + + for (offset = 0; offset < sec->len; offset += insn->len) { + insn = malloc(sizeof(*insn)); + memset(insn, 0, sizeof(*insn)); + + INIT_LIST_HEAD(&insn->alts); + insn->sec = sec; + insn->offset = offset; + + ret = arch_decode_instruction(elf, sec, offset, + sec->len - offset, + &insn->len, &insn->type, + &insn->immediate); + if (ret) + return ret; + + if (!insn->type || insn->type > INSN_LAST) { + WARN_FUNC("invalid instruction type %d", + insn->sec, insn->offset, insn->type); + return -1; + } + + list_add_tail(&insn->list, &insns); + } + } + + return 0; +} + +/* + * Warnings shouldn't be reported for ignored instructions. Set insn->ignore + * for each ignored instruction and each instruction contained in an ignored + * function. + */ +static void get_ignores(struct elf *elf) +{ + struct instruction *insn; + struct section *sec; + struct symbol *func; + + list_for_each_entry(insn, &insns, list) + insn->ignore = ignore_insn(elf, insn); + + list_for_each_entry(sec, &elf->sections, list) { + list_for_each_entry(func, &sec->symbols, list) { + if (func->type != STT_FUNC) + continue; + + if (!ignore_func(elf, func)) + continue; + + insn = find_instruction(sec, func->offset); + if (!insn) + continue; + + list_for_each_entry_from(insn, &insns, list) { + if (insn->sec != func->sec || + insn->offset >= func->offset + func->len) + break; + + insn->ignore = true; + insn->visited = true; + } + } + } +} + +/* + * Find the destination instructions for all jumps. + */ +static int get_jump_destinations(struct elf *elf) +{ + struct instruction *insn; + struct rela *rela; + struct section *dest_sec; + unsigned long dest_off; + + list_for_each_entry(insn, &insns, list) { + if (insn->type != INSN_JUMP_CONDITIONAL && + insn->type != INSN_JUMP_UNCONDITIONAL) + continue; + + rela = find_rela_by_dest_range(insn->sec, insn->offset, + insn->len); + if (!rela) { + dest_sec = insn->sec; + dest_off = insn->offset + insn->len + insn->immediate; + } else if (rela->sym->type == STT_SECTION) { + dest_sec = rela->sym->sec; + dest_off = rela->addend + 4; + } else if (rela->sym->sec->index) { + dest_sec = rela->sym->sec; + dest_off = rela->sym->sym.st_value + rela->addend + 4; + } else { + /* + * This error (jump to another file) will be handled + * later in validate_functions(). + */ + continue; + } + + insn->jump_dest = find_instruction(dest_sec, dest_off); + if (!insn->jump_dest && !insn->ignore) { + + /* + * This is a special case where an alt instruction + * jumps past the end of the section. These are + * handled later. + */ + if (!strcmp(insn->sec->name, ".altinstr_replacement")) + continue; + + WARN_FUNC("can't find jump dest instruction at %s+0x%lx", + insn->sec, insn->offset, dest_sec->name, + dest_off); + return -1; + } + } + + return 0; +} + +/* + * Find the destination instructions for all calls. + */ +static int get_call_destinations(struct elf *elf) +{ + struct instruction *insn; + unsigned long dest_off; + struct rela *rela; + + list_for_each_entry(insn, &insns, list) { + if (insn->type != INSN_CALL) + continue; + + rela = find_rela_by_dest_range(insn->sec, insn->offset, + insn->len); + if (!rela) { + dest_off = insn->offset + insn->len + insn->immediate; + insn->call_dest = find_symbol_by_offset(insn->sec, + dest_off); + if (!insn->call_dest) { + WARN_FUNC("can't find call dest symbol at offset 0x%lx", + insn->sec, insn->offset, dest_off); + return -1; + } + } else if (rela->sym->type == STT_SECTION) { + insn->call_dest = find_symbol_by_offset(rela->sym->sec, + rela->addend+4); + if (!insn->call_dest || + insn->call_dest->type != STT_FUNC) { + WARN_FUNC("can't find call dest symbol at %s+0x%x", + insn->sec, insn->offset, + rela->sym->sec->name, + rela->addend + 4); + return -1; + } + } else + insn->call_dest = rela->sym; + } + + return 0; +} + +/* + * The .alternatives section requires some extra special care, over and above + * what other special sections require: + * + * 1. Because alternatives are patched in-place, we need to insert a fake jump + * instruction at the end so that validate_branch() skips all the original + * replaced instructions when validating the new instruction path. + * + * 2. An added wrinkle is that the new instruction length might be zero. In + * that case the old instructions are replaced with noops. We simulate that + * by creating a fake jump as the only new instruction. + * + * 3. In some cases, the alternative section includes an instruction which + * conditionally jumps to the _end_ of the entry. We have to modify these + * jumps' destinations to point back to .text rather than the end of the + * entry in .altinstr_replacement. + * + * 4. It has been requested that we don't validate the !POPCNT feature path + * which is a "very very small percentage of machines". + */ +static int handle_group_alt(struct elf *elf, struct special_alt *special_alt, + struct instruction *orig_insn, + struct instruction **new_insn) +{ + struct instruction *last_orig_insn, *last_new_insn, *insn, *fake_jump; + unsigned long dest_off; + + last_orig_insn = NULL; + insn = orig_insn; + list_for_each_entry_from(insn, &insns, list) { + if (insn->sec != special_alt->orig_sec || + insn->offset >= special_alt->orig_off + special_alt->orig_len) + break; + + if (special_alt->skip_orig) + insn->ignore = true; + + insn->alt_group = true; + last_orig_insn = insn; + } + + if (list_is_last(&last_orig_insn->list, &insns) || + list_next_entry(last_orig_insn, list)->sec != special_alt->orig_sec) { + WARN("%s: don't know how to handle alternatives at end of section", + special_alt->orig_sec->name); + return -1; + } + + fake_jump = malloc(sizeof(*fake_jump)); + if (!fake_jump) { + WARN("malloc failed"); + return -1; + } + memset(fake_jump, 0, sizeof(*fake_jump)); + INIT_LIST_HEAD(&fake_jump->alts); + fake_jump->sec = special_alt->new_sec; + fake_jump->offset = -1; + fake_jump->type = INSN_JUMP_UNCONDITIONAL; + fake_jump->jump_dest = list_next_entry(last_orig_insn, list); + + if (!special_alt->new_len) { + *new_insn = fake_jump; + return 0; + } + + last_new_insn = NULL; + insn = *new_insn; + list_for_each_entry_from(insn, &insns, list) { + if (insn->sec != special_alt->new_sec || + insn->offset >= special_alt->new_off + special_alt->new_len) + break; + + last_new_insn = insn; + + if (insn->type != INSN_JUMP_CONDITIONAL && + insn->type != INSN_JUMP_UNCONDITIONAL) + continue; + + if (!insn->immediate) + continue; + + dest_off = insn->offset + insn->len + insn->immediate; + if (dest_off == special_alt->new_off + special_alt->new_len) + insn->jump_dest = fake_jump; + + if (!insn->jump_dest) { + WARN_FUNC("can't find alternative jump destination", + insn->sec, insn->offset); + return -1; + } + } + + if (!last_new_insn) { + WARN_FUNC("can't find last new alternative instruction", + special_alt->new_sec, special_alt->new_off); + return -1; + } + + list_add(&fake_jump->list, &last_new_insn->list); + + return 0; +} + +/* + * A jump table entry can either convert a nop to a jump or a jump to a nop. + * If the original instruction is a jump, make the alt entry an effective nop + * by just skipping the original instruction. + */ +static int handle_jump_alt(struct elf *elf, struct special_alt *special_alt, + struct instruction *orig_insn, + struct instruction **new_insn) +{ + if (orig_insn->type == INSN_NOP) + return 0; + + if (orig_insn->type != INSN_JUMP_UNCONDITIONAL) { + WARN_FUNC("unsupported instruction at jump label", + orig_insn->sec, orig_insn->offset); + return -1; + } + + *new_insn = list_next_entry(orig_insn, list); + return 0; +} + +/* + * Read all the special sections which have alternate instructions which can be + * patched in or redirected to at runtime. Each instruction having alternate + * instruction(s) has them added to its insn->alts list, which will be + * traversed in validate_branch(). + */ +static int get_special_section_alts(struct elf *elf) +{ + struct list_head special_alts; + struct instruction *orig_insn, *new_insn; + struct special_alt *special_alt, *tmp; + struct alternative *alt; + int ret; + + ret = special_get_alts(elf, &special_alts); + if (ret) + return ret; + + list_for_each_entry_safe(special_alt, tmp, &special_alts, list) { + alt = malloc(sizeof(*alt)); + if (!alt) { + WARN("malloc failed"); + ret = -1; + goto out; + } + + orig_insn = find_instruction(special_alt->orig_sec, + special_alt->orig_off); + if (!orig_insn) { + WARN_FUNC("special: can't find orig instruction", + special_alt->orig_sec, special_alt->orig_off); + ret = -1; + goto out; + } + + new_insn = NULL; + if (!special_alt->group || special_alt->new_len) { + new_insn = find_instruction(special_alt->new_sec, + special_alt->new_off); + if (!new_insn) { + WARN_FUNC("special: can't find new instruction", + special_alt->new_sec, + special_alt->new_off); + ret = -1; + goto out; + } + } + + if (special_alt->group) { + ret = handle_group_alt(elf, special_alt, orig_insn, + &new_insn); + if (ret) + goto out; + } else if (special_alt->jump_or_nop) { + ret = handle_jump_alt(elf, special_alt, orig_insn, + &new_insn); + if (ret) + goto out; + } + + alt->insn = new_insn; + list_add_tail(&alt->list, &orig_insn->alts); + + list_del(&special_alt->list); + free(special_alt); + } + +out: + return ret; +} + +/* + * For some switch statements, gcc generates a jump table in the .rodata + * section which contains a list of addresses within the function to jump to. + * This finds these jump tables and adds them to the insn->alts lists. + */ +static int get_switch_alts(struct elf *elf) +{ + struct instruction *insn, *alt_insn; + struct rela *rodata_rela, *rela; + struct section *rodata; + struct symbol *func; + struct alternative *alt; + + list_for_each_entry(insn, &insns, list) { + if (insn->type != INSN_JUMP_DYNAMIC) + continue; + + rodata_rela = find_rela_by_dest_range(insn->sec, insn->offset, + insn->len); + if (!rodata_rela || strcmp(rodata_rela->sym->name, ".rodata")) + continue; + + rodata = find_section_by_name(elf, ".rodata"); + if (!rodata || !rodata->rela) + continue; + + rela = find_rela_by_dest(rodata, rodata_rela->addend); + if (!rela) + continue; + + func = find_containing_func(insn->sec, insn->offset); + if (!func) { + WARN_FUNC("can't find containing func", + insn->sec, insn->offset); + return -1; + } + + list_for_each_entry_from(rela, &rodata->rela->relas, list) { + if (rela->sym->sec != insn->sec || + rela->addend <= func->offset || + rela->addend >= func->offset + func->len) + break; + + alt_insn = find_instruction(insn->sec, rela->addend); + if (!alt_insn) { + WARN("%s: can't find instruction at %s+0x%x", + rodata->rela->name, insn->sec->name, + rela->addend); + return -1; + } + + alt = malloc(sizeof(*alt)); + if (!alt) { + WARN("malloc failed"); + return -1; + } + + alt->insn = alt_insn; + list_add_tail(&alt->list, &insn->alts); + } + } + + return 0; +} + +static int decode_sections(struct elf *elf) +{ + int ret; + + ret = decode_instructions(elf); + if (ret) + return ret; + + get_ignores(elf); + + ret = get_jump_destinations(elf); + if (ret) + return ret; + + ret = get_call_destinations(elf); + if (ret) + return ret; + + ret = get_special_section_alts(elf); + if (ret) + return ret; + + ret = get_switch_alts(elf); + if (ret) + return ret; + + return 0; +} + +/* + * Follow the branch starting at the given instruction, and recursively follow + * any other branches (jumps). Meanwhile, track the frame pointer state at + * each instruction and validate all the rules described in + * tools/stacktool/Documentation/stack-validation.txt. + */ +static int validate_branch(struct elf *elf, struct instruction *first, + unsigned char first_state) +{ + struct alternative *alt; + struct instruction *insn; + struct section *sec; + unsigned char state; + int ret, warnings = 0; + + insn = first; + sec = insn->sec; + state = first_state; + + if (insn->alt_group && list_empty(&insn->alts)) { + WARN_FUNC("don't know how to handle branch to middle of alternative instruction group", + sec, insn->offset); + warnings++; + } + + while (1) { + if (insn->visited) { + if (insn->state != state) { + WARN_FUNC("frame pointer state mismatch", + sec, insn->offset); + warnings++; + } + + return warnings; + } + + insn->visited = true; + insn->state = state; + + list_for_each_entry(alt, &insn->alts, list) { + ret = validate_branch(elf, alt->insn, state); + warnings += ret; + } + + switch (insn->type) { + + case INSN_FP_SAVE: + if (args.fp) { + if (insn->state & STATE_FP_SAVED) { + WARN_FUNC("duplicate frame pointer save", + sec, insn->offset); + warnings++; + } + state |= STATE_FP_SAVED; + } + break; + + case INSN_FP_SETUP: + if (args.fp) { + if (insn->state & STATE_FP_SETUP) { + WARN_FUNC("duplicate frame pointer setup", + sec, insn->offset); + warnings++; + } + state |= STATE_FP_SETUP; + } + break; + + case INSN_FP_RESTORE: + if (args.fp) { + if (!insn->state) { + WARN_FUNC("frame pointer restore without save/setup", + sec, insn->offset); + warnings++; + } + state &= ~STATE_FP_SAVED; + state &= ~STATE_FP_SETUP; + } + break; + + case INSN_RETURN: + if (args.fp && insn->state) { + WARN_FUNC("return without frame pointer restore", + sec, insn->offset); + warnings++; + } + return warnings; + + case INSN_CALL: + if (insn->call_dest->type == STT_NOTYPE && + !strcmp(insn->call_dest->name, "__fentry__")) + break; + + if (dead_end_function(insn->call_dest)) + return warnings; + + if (args.fp && !insn->state && !insn->ignore) { + WARN_FUNC("call without frame pointer save/setup", + sec, insn->offset); + warnings++; + } + break; + + case INSN_CALL_DYNAMIC: + if (args.fp && !insn->state && !insn->ignore) { + WARN_FUNC("call without frame pointer save/setup", + sec, insn->offset); + warnings++; + } + break; + + case INSN_JUMP_CONDITIONAL: + case INSN_JUMP_UNCONDITIONAL: + if (insn->jump_dest) { + ret = validate_branch(elf, insn->jump_dest, + state); + warnings += ret; + } else if (insn->state) { + WARN_FUNC("sibling call from callable instruction with changed frame pointer", + sec, insn->offset); + warnings++; + } /* else it's a sibling call */ + + if (insn->type == INSN_JUMP_UNCONDITIONAL) + return warnings; + + break; + + case INSN_JUMP_DYNAMIC: + if (list_empty(&insn->alts) && insn->state && + !insn->ignore) { + WARN_FUNC("sibling call from callable instruction with changed frame pointer", + sec, insn->offset); + warnings++; + } + + return warnings; + + case INSN_CONTEXT_SWITCH: + if (!insn->ignore) { + WARN_FUNC("kernel entry/exit from callable instruction", + sec, insn->offset); + warnings++; + } + + return warnings; + + case INSN_BUG: + return warnings; + + default: + break; + } + + insn = list_next_entry(insn, list); + + if (&insn->list == &insns || insn->sec != sec) { + WARN("%s: unexpected end of section", sec->name); + warnings++; + return warnings; + } + } + + return warnings; +} + +static int validate_functions(struct elf *elf) +{ + struct section *sec; + struct symbol *func; + struct instruction *insn; + int ret, warnings = 0; + + list_for_each_entry(sec, &elf->sections, list) { + list_for_each_entry(func, &sec->symbols, list) { + if (func->type != STT_FUNC) + continue; + + insn = find_instruction(sec, func->offset); + if (!insn) { + WARN("%s(): can't find starting instruction", + func->name); + warnings++; + continue; + } + + ret = validate_branch(elf, insn, 0); + warnings += ret; + } + } + + list_for_each_entry(sec, &elf->sections, list) { + list_for_each_entry(func, &sec->symbols, list) { + if (func->type != STT_FUNC) + continue; + + insn = find_instruction(sec, func->offset); + if (!insn) + continue; + + list_for_each_entry_from(insn, &insns, list) { + if (insn->sec != func->sec || + insn->offset >= func->offset + func->len) + break; + + if (!insn->visited && insn->type != INSN_NOP) { + WARN_FUNC("function has unreachable instruction", + insn->sec, insn->offset); + warnings++; + } + + insn->visited = true; + } + } + } + + return warnings; +} + +static int validate_uncallable_instructions(struct elf *elf) +{ + struct instruction *insn; + int warnings = 0; + + list_for_each_entry(insn, &insns, list) { + if (!insn->visited && insn->type == INSN_RETURN && + !insn->ignore) { + WARN_FUNC("return instruction outside of a callable function", + insn->sec, insn->offset); + warnings++; + } + } + + return warnings; +} + +static void cleanup(struct elf *elf) +{ + struct instruction *insn, *tmpinsn; + struct alternative *alt, *tmpalt; + + list_for_each_entry_safe(insn, tmpinsn, &insns, list) { + list_for_each_entry_safe(alt, tmpalt, &insn->alts, list) { + list_del(&alt->list); + free(alt); + } + list_del(&insn->list); + free(insn); + } + elf_close(elf); +} + +int main(int argc, char *argv[]) +{ + struct elf *elf; + int ret, warnings = 0; + + argp_parse(&argp, argc, argv, 0, 0, &args); + + objname = args.args[0]; + elf = elf_open(objname); + if (!elf) { + fprintf(stderr, "error reading elf file %s\n", args.args[0]); + return 1; + } + + ret = decode_sections(elf); + if (ret < 0) + goto out; + warnings += ret; + + ret = validate_functions(elf); + if (ret < 0) + goto out; + warnings += ret; + + ret = validate_uncallable_instructions(elf); + if (ret < 0) + goto out; + warnings += ret; + +out: + cleanup(elf); + + /* ignore warnings for now until we get all the code cleaned up */ + if (ret || warnings) + return 0; + return 0; +} diff --git a/tools/stacktool/warn.h b/tools/stacktool/warn.h new file mode 100644 index 0000000..9f54d7e --- /dev/null +++ b/tools/stacktool/warn.h @@ -0,0 +1,60 @@ +/* + * Copyright (C) 2015 Josh Poimboeuf <jpoimboe@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef _WARN_H +#define _WARN_H + +extern char *objname; + +static inline char *offstr(struct section *sec, unsigned long offset) +{ + struct symbol *func; + char *name, *str; + unsigned long name_off; + + func = find_containing_func(sec, offset); + if (func) { + name = func->name; + name_off = offset - func->offset; + } else { + name = sec->name; + name_off = offset; + } + + str = malloc(strlen(name) + 20); + + if (func) + sprintf(str, "%s()+0x%lx", name, name_off); + else + sprintf(str, "%s+0x%lx", name, name_off); + + return str; +} + +#define WARN(format, ...) \ + fprintf(stderr, \ + "stacktool: %s: " format "\n", \ + objname, ##__VA_ARGS__) + +#define WARN_FUNC(format, sec, offset, ...) \ +({ \ + char *_str = offstr(sec, offset); \ + WARN("%s: " format, _str, ##__VA_ARGS__); \ + free(_str); \ +}) + +#endif /* _WARN_H */ -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* Re: [PATCH v13 04/23] x86/stacktool: Compile-time stack metadata validation 2015-09-22 15:46 ` [PATCH v13 04/23] x86/stacktool: Compile-time stack metadata validation Josh Poimboeuf @ 2015-09-23 11:38 ` Jiri Slaby 2015-09-23 11:40 ` Jiri Slaby 0 siblings, 1 reply; 38+ messages in thread From: Jiri Slaby @ 2015-09-23 11:38 UTC (permalink / raw) To: Josh Poimboeuf, Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton On 09/22/2015, 05:46 PM, Josh Poimboeuf wrote: ... > --- /dev/null > +++ b/tools/stacktool/elf.c > @@ -0,0 +1,403 @@ ... > +static struct section *find_section_by_index(struct elf *elf, > + unsigned int index) > +{ ... > +} > + > +static struct symbol *find_symbol_by_index(struct elf *elf, unsigned int index) > +{ Hi, on openSUSE, I am getting: elf.c:45: error: declaration of ‘index’ shadows a global declaration /usr/include/string.h:487: error: shadowed declaration is here elf.c: In function ‘find_symbol_by_index’: elf.c:56: error: declaration of ‘index’ shadows a global declaration /usr/include/string.h:487: error: shadowed declaration is here which is index(3): extern char *index (__const char *__s, int __c) __THROW __attribute_pure__ __nonnull ((1)); The same here: > --- /dev/null > +++ b/tools/stacktool/special.c > @@ -0,0 +1,199 @@ ... > +static int get_alt_entry(struct elf *elf, struct special_entry *entry, > + struct section *sec, int index, > + struct special_alt *alt) > +{ ... > +int special_get_alts(struct elf *elf, struct list_head *alts) > +{ I fixed it obviously by s/index/idx/. thanks, -- js suse labs ^ permalink raw reply [flat|nested] 38+ messages in thread
* Re: [PATCH v13 04/23] x86/stacktool: Compile-time stack metadata validation 2015-09-23 11:38 ` Jiri Slaby @ 2015-09-23 11:40 ` Jiri Slaby 2015-09-23 12:42 ` [PATCH v13.1 " Josh Poimboeuf 0 siblings, 1 reply; 38+ messages in thread From: Jiri Slaby @ 2015-09-23 11:40 UTC (permalink / raw) To: Josh Poimboeuf, Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton On 09/23/2015, 01:38 PM, Jiri Slaby wrote: > On 09/22/2015, 05:46 PM, Josh Poimboeuf wrote: > ... >> --- /dev/null >> +++ b/tools/stacktool/elf.c >> @@ -0,0 +1,403 @@ > ... >> +static struct section *find_section_by_index(struct elf *elf, >> + unsigned int index) >> +{ > ... >> +} >> + >> +static struct symbol *find_symbol_by_index(struct elf *elf, unsigned int index) >> +{ > > Hi, on openSUSE, I am getting: > > elf.c:45: error: declaration of ‘index’ shadows a global declaration > /usr/include/string.h:487: error: shadowed declaration is here > elf.c: In function ‘find_symbol_by_index’: > elf.c:56: error: declaration of ‘index’ shadows a global declaration > /usr/include/string.h:487: error: shadowed declaration is here > > which is index(3): > extern char *index (__const char *__s, int __c) > __THROW __attribute_pure__ __nonnull ((1)); > > The same here: > >> --- /dev/null >> +++ b/tools/stacktool/special.c >> @@ -0,0 +1,199 @@ > ... >> +static int get_alt_entry(struct elf *elf, struct special_entry *entry, >> + struct section *sec, int index, >> + struct special_alt *alt) >> +{ > ... >> +int special_get_alts(struct elf *elf, struct list_head *alts) >> +{ I erased too much here. Index is a local var here: int index, ret; > > I fixed it obviously by s/index/idx/. > > thanks, -- js suse labs ^ permalink raw reply [flat|nested] 38+ messages in thread
* [PATCH v13.1 04/23] x86/stacktool: Compile-time stack metadata validation 2015-09-23 11:40 ` Jiri Slaby @ 2015-09-23 12:42 ` Josh Poimboeuf 2015-10-01 14:50 ` Chris J Arges 0 siblings, 1 reply; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-23 12:42 UTC (permalink / raw) To: Jiri Slaby Cc: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86, linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton On Wed, Sep 23, 2015 at 01:40:17PM +0200, Jiri Slaby wrote: > > I fixed it obviously by s/index/idx/. Thanks Jiri! Here's the patch with s/index/idx/. ---8<--- Subject: [PATCH v13.1 04/23] x86/stacktool: Compile-time stack metadata validation This adds a CONFIG_STACK_VALIDATION option which enables a host tool named stacktool which runs at compile time. It analyzes every .o file and ensures the validity of its stack metadata. It enforces a set of rules on asm code and C inline assembly code so that stack traces can be reliable. For each function, it recursively follows all possible code paths and validates the correct frame pointer state at each instruction. It also follows code paths involving special sections, like .altinstructions, __jump_table, and __ex_table, which can add alternative execution paths to a given instruction (or set of instructions). Similarly, it knows how to follow switch statements, for which gcc sometimes uses jump tables. Here are some of the benefits of validating stack metadata: a) More reliable stack traces for frame pointer enabled kernels Frame pointers are used for debugging purposes. They allow runtime code and debug tools to be able to walk the stack to determine the chain of function call sites that led to the currently executing code. For some architectures, frame pointers are enabled by CONFIG_FRAME_POINTER. For some other architectures they may be required by the ABI (sometimes referred to as "backchain pointers"). For C code, gcc automatically generates instructions for setting up frame pointers when the -fno-omit-frame-pointer option is used. But for asm code, the frame setup instructions have to be written by hand, which most people don't do. So the end result is that CONFIG_FRAME_POINTER is honored for C code but not for most asm code. For stack traces based on frame pointers to be reliable, all functions which call other functions must first create a stack frame and update the frame pointer. If a first function doesn't properly create a stack frame before calling a second function, the *caller* of the first function will be skipped on the stack trace. For example, consider the following example backtrace with frame pointers enabled: [<ffffffff81812584>] dump_stack+0x4b/0x63 [<ffffffff812d6dc2>] cmdline_proc_show+0x12/0x30 [<ffffffff8127f568>] seq_read+0x108/0x3e0 [<ffffffff812cce62>] proc_reg_read+0x42/0x70 [<ffffffff81256197>] __vfs_read+0x37/0x100 [<ffffffff81256b16>] vfs_read+0x86/0x130 [<ffffffff81257898>] SyS_read+0x58/0xd0 [<ffffffff8181c1f2>] entry_SYSCALL_64_fastpath+0x12/0x76 It correctly shows that the caller of cmdline_proc_show() is seq_read(). If we remove the frame pointer logic from cmdline_proc_show() by replacing the frame pointer related instructions with nops, here's what it looks like instead: [<ffffffff81812584>] dump_stack+0x4b/0x63 [<ffffffff812d6dc2>] cmdline_proc_show+0x12/0x30 [<ffffffff812cce62>] proc_reg_read+0x42/0x70 [<ffffffff81256197>] __vfs_read+0x37/0x100 [<ffffffff81256b16>] vfs_read+0x86/0x130 [<ffffffff81257898>] SyS_read+0x58/0xd0 [<ffffffff8181c1f2>] entry_SYSCALL_64_fastpath+0x12/0x76 Notice that cmdline_proc_show()'s caller, seq_read(), has been skipped. Instead the stack trace seems to show that cmdline_proc_show() was called by proc_reg_read(). The benefit of stacktool here is that because it ensures that *all* functions honor CONFIG_FRAME_POINTER, no functions will ever[*] be skipped on a stack trace. [*] unless an interrupt or exception has occurred at the very beginning of a function before the stack frame has been created, or at the very end of the function after the stack frame has been destroyed. This is an inherent limitation of frame pointers. b) 100% reliable stack traces for DWARF enabled kernels This is not yet implemented. For more details about what is planned, see tools/stacktool/Documentation/stack-validation.txt. c) Higher live patching compatibility rate This is not yet implemented. For more details about what is planned, see tools/stacktool/Documentation/stack-validation.txt. To achieve the validation, stacktool enforces the following rules: 1. Each callable function must be annotated as such with the ELF function type. In asm code, this is typically done using the ENTRY/ENDPROC macros. If stacktool finds a return instruction outside of a function, it flags an error since that usually indicates callable code which should be annotated accordingly. This rule is needed so that stacktool can properly identify each callable function in order to analyze its stack metadata. 2. Conversely, each section of code which is *not* callable should *not* be annotated as an ELF function. The ENDPROC macro shouldn't be used in this case. This rule is needed so that stacktool can ignore non-callable code. Such code doesn't have to follow any of the other rules. 3. Each callable function which calls another function must have the correct frame pointer logic, if required by CONFIG_FRAME_POINTER or the architecture's back chain rules. This can by done in asm code with the FRAME_BEGIN/FRAME_END macros. This rule ensures that frame pointer based stack traces will work as designed. If function A doesn't create a stack frame before calling function B, the _caller_ of function A will be skipped on the stack trace. 4. Dynamic jumps and jumps to undefined symbols are only allowed if: a) the jump is part of a switch statement; or b) the jump matches sibling call semantics and the frame pointer has the same value it had on function entry. This rule is needed so that stacktool can reliably analyze all of a function's code paths. If a function jumps to code in another file, and it's not a sibling call, stacktool has no way to follow the jump because it only analyzes a single file at a time. 5. A callable function may not execute kernel entry/exit instructions. The only code which needs such instructions is kernel entry code, which shouldn't be be in callable functions anyway. This rule is just a sanity check to ensure that callable functions return normally. It currently only supports x86_64. I tried to make the code generic so that support for other architectures can hopefully be plugged in relatively easily. As a first step, CONFIG_STACK_VALIDATION is disabled by default, and all reported non-compliances result in warnings. Once we get them all cleaned up, we can change the default to CONFIG_STACK_VALIDATION=y and change the warnings to errors to keep the stack metadata clean. On my Lenovo laptop with a i7-4810MQ 4-core/8-thread CPU, building the kernel with it enabled adds about three seconds of total build time. It hasn't been optimized for performance yet, so there are probably some opportunities for better build performance. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> --- MAINTAINERS | 7 + Makefile | 5 +- arch/Kconfig | 6 + arch/x86/Kconfig | 1 + lib/Kconfig.debug | 12 + scripts/Makefile.build | 37 +- tools/Makefile | 11 +- tools/stacktool/.gitignore | 1 + tools/stacktool/Build | 4 + tools/stacktool/Documentation/stack-validation.txt | 336 +++++++ tools/stacktool/Makefile | 51 ++ tools/stacktool/arch.h | 44 + tools/stacktool/arch/x86/Build | 12 + tools/stacktool/arch/x86/decode.c | 163 ++++ .../stacktool/arch/x86/insn/gen-insn-attr-x86.awk | 387 ++++++++ tools/stacktool/arch/x86/insn/inat.c | 97 ++ tools/stacktool/arch/x86/insn/inat.h | 221 +++++ tools/stacktool/arch/x86/insn/inat_types.h | 29 + tools/stacktool/arch/x86/insn/insn.c | 594 +++++++++++++ tools/stacktool/arch/x86/insn/insn.h | 201 +++++ tools/stacktool/arch/x86/insn/x86-opcode-map.txt | 984 +++++++++++++++++++++ tools/stacktool/elf.c | 403 +++++++++ tools/stacktool/elf.h | 79 ++ tools/stacktool/special.c | 199 +++++ tools/stacktool/special.h | 42 + tools/stacktool/stacktool.c | 980 ++++++++++++++++++++ tools/stacktool/warn.h | 60 ++ 27 files changed, 4956 insertions(+), 10 deletions(-) create mode 100644 tools/stacktool/.gitignore create mode 100644 tools/stacktool/Build create mode 100644 tools/stacktool/Documentation/stack-validation.txt create mode 100644 tools/stacktool/Makefile create mode 100644 tools/stacktool/arch.h create mode 100644 tools/stacktool/arch/x86/Build create mode 100644 tools/stacktool/arch/x86/decode.c create mode 100644 tools/stacktool/arch/x86/insn/gen-insn-attr-x86.awk create mode 100644 tools/stacktool/arch/x86/insn/inat.c create mode 100644 tools/stacktool/arch/x86/insn/inat.h create mode 100644 tools/stacktool/arch/x86/insn/inat_types.h create mode 100644 tools/stacktool/arch/x86/insn/insn.c create mode 100644 tools/stacktool/arch/x86/insn/insn.h create mode 100644 tools/stacktool/arch/x86/insn/x86-opcode-map.txt create mode 100644 tools/stacktool/elf.c create mode 100644 tools/stacktool/elf.h create mode 100644 tools/stacktool/special.c create mode 100644 tools/stacktool/special.h create mode 100644 tools/stacktool/stacktool.c create mode 100644 tools/stacktool/warn.h diff --git a/MAINTAINERS b/MAINTAINERS index ea3e3ae..a128801 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -9879,6 +9879,13 @@ L: stable@vger.kernel.org S: Supported F: Documentation/stable_kernel_rules.txt +STACK METADATA VALIDATION +M: Josh Poimboeuf <jpoimboe@redhat.com> +S: Supported +F: tools/stacktool/ +F: include/linux/stacktool.h +F: arch/x86/include/asm/stacktool.h + STAGING SUBSYSTEM M: Greg Kroah-Hartman <gregkh@linuxfoundation.org> T: git git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging.git diff --git a/Makefile b/Makefile index 84f4b31..0b9a1ca 100644 --- a/Makefile +++ b/Makefile @@ -984,7 +984,10 @@ prepare0: archprepare FORCE $(Q)$(MAKE) $(build)=. # All the preparing.. -prepare: prepare0 +prepare: prepare0 prepare-stacktool + +PHONY += prepare-stacktool +prepare-stacktool: $(if $(CONFIG_STACK_VALIDATION), tools/stacktool FORCE) # Generate some files # --------------------------------------------------------------------------- diff --git a/arch/Kconfig b/arch/Kconfig index 671810c..b20f472 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -527,6 +527,12 @@ config HAVE_COPY_THREAD_TLS normal C parameter passing, rather than extracting the syscall argument from pt_regs. +config HAVE_STACK_VALIDATION + bool + help + Architecture supports the stacktool host tool, which adds + compile-time stack metadata validation. + # # ABI hall of shame # diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index d8546d4..006b4af 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -153,6 +153,7 @@ config X86 select VIRT_TO_BUS select X86_DEV_DMA_OPS if X86_64 select X86_FEATURE_NAMES if PROC_FS + select HAVE_STACK_VALIDATION if X86_64 config INSTRUCTION_DECODER def_bool y diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index ab76b99..24d2e88 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -332,6 +332,18 @@ config FRAME_POINTER larger and slower, but it gives very useful debugging information in case of kernel bugs. (precise oopses/stacktraces/warnings) +config STACK_VALIDATION + bool "Enable compile-time stack metadata validation" + depends on HAVE_STACK_VALIDATION + default n + help + Add compile-time checks to validate stack metadata, including frame + pointers (if CONFIG_FRAME_POINTER is enabled). This helps ensure + that runtime stack traces are more reliable. + + For more information, see + tools/stacktool/Documentation/stack-validation.txt. + config DEBUG_FORCE_WEAK_PER_CPU bool "Force weak per-cpu definitions" depends on DEBUG_KERNEL diff --git a/scripts/Makefile.build b/scripts/Makefile.build index 01df30a..3ffd994 100644 --- a/scripts/Makefile.build +++ b/scripts/Makefile.build @@ -241,10 +241,30 @@ cmd_record_mcount = \ fi; endif +ifdef CONFIG_STACK_VALIDATION + +__stacktool_obj := $(objtree)/tools/stacktool/stacktool + +ifdef CONFIG_FRAME_POINTER +stacktool_args += --check-frame-pointer +endif + +# Set STACKTOOL_foo.o=n to skip stack metadata validation for a file. +# Set STACKTOOL=n to skip stack metadata validation for a directory. +stacktool_obj = $(if $(patsubst n%,, \ + $(STACKTOOL_$(basetarget).o)$(STACKTOOL)y), \ + $(__stacktool_obj)) +cmd_stacktool = $(if $(patsubst n%,, \ + $(STACKTOOL_$(basetarget).o)$(STACKTOOL)y), \ + $(__stacktool_obj) $(stacktool_args) "$(@)";) + +endif # CONFIG_STACK_VALIDATION + define rule_cc_o_c $(call echo-cmd,checksrc) $(cmd_checksrc) \ $(call echo-cmd,cc_o_c) $(cmd_cc_o_c); \ $(cmd_modversions) \ + $(cmd_stacktool) \ $(call echo-cmd,record_mcount) \ $(cmd_record_mcount) \ scripts/basic/fixdep $(depfile) $@ '$(call make-cmd,cc_o_c)' > \ @@ -253,14 +273,23 @@ define rule_cc_o_c mv -f $(dot-target).tmp $(dot-target).cmd endef +define rule_as_o_S + $(call echo-cmd,as_o_S) $(cmd_as_o_S); \ + $(cmd_stacktool) \ + scripts/basic/fixdep $(depfile) $@ '$(call make-cmd,as_o_S)' > \ + $(dot-target).tmp; \ + rm -f $(depfile); \ + mv -f $(dot-target).tmp $(dot-target).cmd +endef + # Built-in and composite module parts -$(obj)/%.o: $(src)/%.c $(recordmcount_source) FORCE +$(obj)/%.o: $(src)/%.c $(recordmcount_source) $(stacktool_obj) FORCE $(call cmd,force_checksrc) $(call if_changed_rule,cc_o_c) # Single-part modules are special since we need to mark them in $(MODVERDIR) -$(single-used-m): $(obj)/%.o: $(src)/%.c $(recordmcount_source) FORCE +$(single-used-m): $(obj)/%.o: $(src)/%.c $(recordmcount_source) $(stacktool_obj) FORCE $(call cmd,force_checksrc) $(call if_changed_rule,cc_o_c) @{ echo $(@:.o=.ko); echo $@; } > $(MODVERDIR)/$(@F:.o=.mod) @@ -290,8 +319,8 @@ $(obj)/%.s: $(src)/%.S FORCE quiet_cmd_as_o_S = AS $(quiet_modtag) $@ cmd_as_o_S = $(CC) $(a_flags) -c -o $@ $< -$(obj)/%.o: $(src)/%.S FORCE - $(call if_changed_dep,as_o_S) +$(obj)/%.o: $(src)/%.S $(stacktool_obj) FORCE + $(call if_changed_rule,as_o_S) targets += $(real-objs-y) $(real-objs-m) $(lib-y) targets += $(extra-y) $(MAKECMDGOALS) $(always) diff --git a/tools/Makefile b/tools/Makefile index c7400a6..39d5d60 100644 --- a/tools/Makefile +++ b/tools/Makefile @@ -19,6 +19,7 @@ help: @echo ' net - misc networking tools' @echo ' perf - Linux performance measurement and analysis tool' @echo ' selftests - various kernel selftests' + @echo ' stacktool - a stack metadata validation tool' @echo ' tmon - thermal monitoring and tuning tool' @echo ' turbostat - Intel CPU idle stats and freq reporting tool' @echo ' usb - USB testing tools' @@ -48,7 +49,7 @@ acpi: FORCE cpupower: FORCE $(call descend,power/$@) -cgroup firewire hv guest usb virtio vm net iio: FORCE +cgroup firewire hv guest usb virtio vm net iio stacktool: FORCE $(call descend,$@) liblockdep: FORCE @@ -83,7 +84,7 @@ acpi_install: cpupower_install: $(call descend,power/$(@:_install=),install) -cgroup_install firewire_install hv_install lguest_install perf_install usb_install virtio_install vm_install net_install: +cgroup_install firewire_install hv_install lguest_install perf_install usb_install virtio_install vm_install net_install stacktool_install: $(call descend,$(@:_install=),install) selftests_install: @@ -101,7 +102,7 @@ freefall_install: install: acpi_install cgroup_install cpupower_install hv_install firewire_install lguest_install \ perf_install selftests_install turbostat_install usb_install \ virtio_install vm_install net_install x86_energy_perf_policy_install \ - tmon freefall_install + tmon freefall_install stacktool_install acpi_clean: $(call descend,power/acpi,clean) @@ -109,7 +110,7 @@ acpi_clean: cpupower_clean: $(call descend,power/cpupower,clean) -cgroup_clean hv_clean firewire_clean lguest_clean usb_clean virtio_clean vm_clean net_clean iio_clean: +cgroup_clean hv_clean firewire_clean lguest_clean usb_clean virtio_clean vm_clean net_clean iio_clean stacktool_clean: $(call descend,$(@:_clean=),clean) liblockdep_clean: @@ -136,6 +137,6 @@ freefall_clean: clean: acpi_clean cgroup_clean cpupower_clean hv_clean firewire_clean lguest_clean \ perf_clean selftests_clean turbostat_clean usb_clean virtio_clean \ vm_clean net_clean iio_clean x86_energy_perf_policy_clean tmon_clean \ - freefall_clean + freefall_clean stacktool_clean .PHONY: FORCE diff --git a/tools/stacktool/.gitignore b/tools/stacktool/.gitignore new file mode 100644 index 0000000..a4014ec --- /dev/null +++ b/tools/stacktool/.gitignore @@ -0,0 +1 @@ +arch/x86/insn/inat-tables.c diff --git a/tools/stacktool/Build b/tools/stacktool/Build new file mode 100644 index 0000000..7688d3d --- /dev/null +++ b/tools/stacktool/Build @@ -0,0 +1,4 @@ +stacktool-y += stacktool.o +stacktool-y += elf.o +stacktool-y += special.o +stacktool-y += arch/$(ARCH)/ diff --git a/tools/stacktool/Documentation/stack-validation.txt b/tools/stacktool/Documentation/stack-validation.txt new file mode 100644 index 0000000..49373c6 --- /dev/null +++ b/tools/stacktool/Documentation/stack-validation.txt @@ -0,0 +1,336 @@ +Compile-time stack metadata validation +====================================== + + +Overview +-------- + +The kernel CONFIG_STACK_VALIDATION option enables a host tool named +stacktool which runs at compile time. It analyzes every .o file and +ensures the validity of its stack metadata. It enforces a set of rules +on asm code and C inline assembly code so that stack traces can be +reliable. + +Currently it only checks frame pointer usage, but there are plans to add +CFI validation for C files and CFI generation for asm files. + +For each function, it recursively follows all possible code paths and +validates the correct frame pointer state at each instruction. + +It also follows code paths involving special sections, like +.altinstructions, __jump_table, and __ex_table, which can add +alternative execution paths to a given instruction (or set of +instructions). Similarly, it knows how to follow switch statements, for +which gcc sometimes uses jump tables. + + +Why do we need stack metadata validation? +----------------------------------------- + +Here are some of the benefits of validating stack metadata: + +a) More reliable stack traces for frame pointer enabled kernels + + Frame pointers are used for debugging purposes. They allow runtime + code and debug tools to be able to walk the stack to determine the + chain of function call sites that led to the currently executing + code. + + For some architectures, frame pointers are enabled by + CONFIG_FRAME_POINTER. For some other architectures they may be + required by the ABI (sometimes referred to as "backchain pointers"). + + For C code, gcc automatically generates instructions for setting up + frame pointers when the -fno-omit-frame-pointer option is used. + + But for asm code, the frame setup instructions have to be written by + hand, which most people don't do. So the end result is that + CONFIG_FRAME_POINTER is honored for C code but not for most asm code. + + For stack traces based on frame pointers to be reliable, all + functions which call other functions must first create a stack frame + and update the frame pointer. If a first function doesn't properly + create a stack frame before calling a second function, the *caller* + of the first function will be skipped on the stack trace. + + For example, consider the following example backtrace with frame + pointers enabled: + + [<ffffffff81812584>] dump_stack+0x4b/0x63 + [<ffffffff812d6dc2>] cmdline_proc_show+0x12/0x30 + [<ffffffff8127f568>] seq_read+0x108/0x3e0 + [<ffffffff812cce62>] proc_reg_read+0x42/0x70 + [<ffffffff81256197>] __vfs_read+0x37/0x100 + [<ffffffff81256b16>] vfs_read+0x86/0x130 + [<ffffffff81257898>] SyS_read+0x58/0xd0 + [<ffffffff8181c1f2>] entry_SYSCALL_64_fastpath+0x12/0x76 + + It correctly shows that the caller of cmdline_proc_show() is + seq_read(). + + If we remove the frame pointer logic from cmdline_proc_show() by + replacing the frame pointer related instructions with nops, here's + what it looks like instead: + + [<ffffffff81812584>] dump_stack+0x4b/0x63 + [<ffffffff812d6dc2>] cmdline_proc_show+0x12/0x30 + [<ffffffff812cce62>] proc_reg_read+0x42/0x70 + [<ffffffff81256197>] __vfs_read+0x37/0x100 + [<ffffffff81256b16>] vfs_read+0x86/0x130 + [<ffffffff81257898>] SyS_read+0x58/0xd0 + [<ffffffff8181c1f2>] entry_SYSCALL_64_fastpath+0x12/0x76 + + Notice that cmdline_proc_show()'s caller, seq_read(), has been + skipped. Instead the stack trace seems to show that + cmdline_proc_show() was called by proc_reg_read(). + + The benefit of stacktool here is that because it ensures that *all* + functions honor CONFIG_FRAME_POINTER, no functions will ever[*] be + skipped on a stack trace. + + [*] unless an interrupt or exception has occurred at the very + beginning of a function before the stack frame has been created, + or at the very end of the function after the stack frame has been + destroyed. This is an inherent limitation of frame pointers. + +b) 100% reliable stack traces for DWARF enabled kernels + + (NOTE: This is not yet implemented) + + As an alternative to frame pointers, DWARF Call Frame Information + (CFI) metadata can be used to walk the stack. Unlike frame pointers, + CFI metadata is out of band. So it doesn't affect runtime + performance and it can be reliable even when interrupts or exceptions + are involved. + + For C code, gcc automatically generates DWARF CFI metadata. But for + asm code, generating CFI is a tedious manual approach which requires + manually placed .cfi assembler macros to be scattered throughout the + code. It's clumsy and very easy to get wrong, and it makes the real + code harder to read. + + Stacktool will improve this situation in several ways. For code + which already has CFI annotations, it will validate them. For code + which doesn't have CFI annotations, it will generate them. So an + architecture can opt to strip out all the manual .cfi annotations + from their asm code and have stacktool generate them instead. + + We might also add a runtime stack validation debug option where we + periodically walk the stack from schedule() and/or an NMI to ensure + that the stack metadata is sane and that we reach the bottom of the + stack. + + So the benefit of stacktool here will be that external tooling should + always show perfect stack traces. And the same will be true for + kernel warning/oops traces if the architecture has a runtime DWARF + unwinder. + +c) Higher live patching compatibility rate + + (NOTE: This is not yet implemented) + + Currently with CONFIG_LIVEPATCH there's a basic live patching + framework which is safe for roughly 85-90% of "security" fixes. But + patches can't have complex features like function dependency or + prototype changes, or data structure changes. + + There's a strong need to support patches which have the more complex + features so that the patch compatibility rate for security fixes can + eventually approach something resembling 100%. To achieve that, a + "consistency model" is needed, which allows tasks to be safely + transitioned from an unpatched state to a patched state. + + One of the key requirements of the currently proposed livepatch + consistency model [*] is that it needs to walk the stack of each + sleeping task to determine if it can be transitioned to the patched + state. If stacktool can ensure that stack traces are reliable, this + consistency model can be used and the live patching compatibility + rate can be improved significantly. + + [*] https://lkml.kernel.org/r/cover.1423499826.git.jpoimboe@redhat.com + + +Rules +----- + +To achieve the validation, stacktool enforces the following rules: + +1. Each callable function must be annotated as such with the ELF + function type. In asm code, this is typically done using the + ENTRY/ENDPROC macros. If stacktool finds a return instruction + outside of a function, it flags an error since that usually indicates + callable code which should be annotated accordingly. + + This rule is needed so that stacktool can properly identify each + callable function in order to analyze its stack metadata. + +2. Conversely, each section of code which is *not* callable should *not* + be annotated as an ELF function. The ENDPROC macro shouldn't be used + in this case. + + This rule is needed so that stacktool can ignore non-callable code. + Such code doesn't have to follow any of the other rules. + +3. Each callable function which calls another function must have the + correct frame pointer logic, if required by CONFIG_FRAME_POINTER or + the architecture's back chain rules. This can by done in asm code + with the FRAME_BEGIN/FRAME_END macros. + + This rule ensures that frame pointer based stack traces will work as + designed. If function A doesn't create a stack frame before calling + function B, the _caller_ of function A will be skipped on the stack + trace. + +4. Dynamic jumps and jumps to undefined symbols are only allowed if: + + a) the jump is part of a switch statement; or + + b) the jump matches sibling call semantics and the frame pointer has + the same value it had on function entry. + + This rule is needed so that stacktool can reliably analyze all of a + function's code paths. If a function jumps to code in another file, + and it's not a sibling call, stacktool has no way to follow the jump + because it only analyzes a single file at a time. + +5. A callable function may not execute kernel entry/exit instructions. + The only code which needs such instructions is kernel entry code, + which shouldn't be be in callable functions anyway. + + This rule is just a sanity check to ensure that callable functions + return normally. + + +Errors in .S files +------------------ + +If you're getting an error in a compiled .S file which you don't +understand, first make sure that the affected code follows the above +rules. + +Here are some examples of common warnings reported by stacktool, what +they mean, and suggestions for how to fix them. + + +1. stacktool: asm_file.o: func()+0x128: call without frame pointer save/setup + + The func() function made a function call without first saving and/or + updating the frame pointer. + + If func() is indeed a callable function, add proper frame pointer + logic using the FRAME_BEGIN and FRAME_END macros. Otherwise, remove + its ELF function annotation by changing ENDPROC to END. + + If you're getting this error in a .c file, see the "Errors in .c + files" section. + + +2. stacktool: asm_file.o: .text+0x53: return instruction outside of a callable function + + A return instruction was detected, but stacktool couldn't find a way + for a callable function to reach the instruction. + + If the return instruction is inside (or reachable from) a callable + function, the function needs to be annotated with the ENTRY/ENDPROC + macros. + + If you _really_ need a return instruction outside of a function, and + are 100% sure that it won't affect stack traces, you can tell + stacktool to ignore it. See the "Adding exceptions" section below. + + +3. stacktool: asm_file.o: func()+0x9: function has unreachable instruction + + The instruction lives inside of a callable function, but there's no + possible control flow path from the beginning of the function to the + instruction. + + If the instruction is actually needed, and it's actually in a + callable function, ensure that its function is properly annotated + with ENTRY/ENDPROC. + + If it's not actually in a callable function (e.g. kernel entry code), + change ENDPROC to END. + + +4. stacktool: asm_file.o: func(): can't find starting instruction + or + stacktool: asm_file.o: func()+0x11dd: can't decode instruction + + Did you put data in a text section? If so, that can confuse + stacktool's instruction decoder. Move the data to a more appropriate + section like .data or .rodata. + + +5. stacktool: asm_file.o: func()+0x6: kernel entry/exit from callable instruction + + This is a kernel entry/exit instruction like sysenter or sysret. + Such instructions aren't allowed in a callable function, and are most + likely part of the kernel entry code. + + If the instruction isn't actually in a callable function, change + ENDPROC to END. + + +6. stacktool: asm_file.o: func()+0x26: sibling call from callable instruction with changed frame pointer + + This is a dynamic jump or a jump to an undefined symbol. Stacktool + assumed it's a sibling call and detected that the frame pointer + wasn't first restored to its original state. + + If it's not really a sibling call, you may need to move the + destination code to the local file. + + If the instruction is not actually in a callable function (e.g. + kernel entry code), change ENDPROC to END. + + +7. stacktool: asm_file: func()+0x5c: frame pointer state mismatch + + The instruction's frame pointer state is inconsistent, depending on + which execution path was taken to reach the instruction. + + Make sure the function pushes and sets up the frame pointer (for + x86_64, this means rbp) at the beginning of the function and pops it + at the end of the function. Also make sure that no other code in the + function touches the frame pointer. + + +Errors in .c files +------------------ + +If you're getting a stacktool error in a compiled .c file, chances are +the file uses an asm() statement which has a "call" instruction. An +asm() statement with a call instruction must declare the use of the +stack pointer in its output operand. For example, on x86_64: + + register void *__sp asm("rsp"); + asm volatile("call func" : "+r" (__sp)); + +Otherwise the stack frame may not get created before the call. + +Another possible cause for errors in C code is if the Makefile removes +-fno-omit-frame-pointer or adds -fomit-frame-pointer to the gcc options. + +Also see the above section for .S file errors for more information what +the individual error messages mean. + + + +Adding exceptions +----------------- + +If you _really_ need stacktool to ignore something, and are 100% sure +that it won't affect kernel stack traces, you can tell stacktool to +ignore it: + +- To skip validation of an instruction, use the STACKTOOL_IGNORE_INSN + macro immediately before the instruction. + +- To skip validation of a function, use the STACKTOOL_IGNORE_FUNC macro. + +- To skip validation of a file, add "STACKTOOL_filename.o := n" to the + Makefile. + +- To skip validation of a directory, add "STACKTOOL := n" to the + Makefile. diff --git a/tools/stacktool/Makefile b/tools/stacktool/Makefile new file mode 100644 index 0000000..9f5d2f9 --- /dev/null +++ b/tools/stacktool/Makefile @@ -0,0 +1,51 @@ +include ../scripts/Makefile.include + +ifndef ($(ARCH)) +ARCH ?= $(shell uname -m) +ifeq ($(ARCH),x86_64) +ARCH := x86 +endif +endif + +ifeq ($(srctree),) +srctree := $(patsubst %/,%,$(dir $(shell pwd))) +srctree := $(patsubst %/,%,$(dir $(srctree))) +endif + +INCLUDES := -I$(srctree)/tools/include +CFLAGS += -Wall -Werror $(EXTRA_WARNINGS) -fomit-frame-pointer -O2 $(INCLUDES) +LDFLAGS += -lelf + +STACKTOOL := $(OUTPUT)stacktool +STACKTOOL_IN := $(STACKTOOL)-in.o + +AWK = awk +export srctree OUTPUT CFLAGS ARCH AWK + +build := -f $(srctree)/tools/build/Makefile.build dir=. obj + +all: $(STACKTOOL) + +$(STACKTOOL_IN): FORCE + @$(MAKE) $(build)=stacktool + +$(STACKTOOL): $(STACKTOOL_IN) + @(test -d ../../kernel -a -d ../../tools -a -d ../stacktool && (( \ + diff -I'^#include' arch/x86/insn/insn.c ../../arch/x86/lib/insn.c >/dev/null && \ + diff -I'^#include' arch/x86/insn/inat.c ../../arch/x86/lib/inat.c >/dev/null && \ + diff arch/x86/insn/x86-opcode-map.txt ../../arch/x86/lib/x86-opcode-map.txt >/dev/null && \ + diff arch/x86/insn/gen-insn-attr-x86.awk ../../arch/x86/tools/gen-insn-attr-x86.awk >/dev/null && \ + diff -I'^#include' arch/x86/insn/insn.h ../../arch/x86/include/asm/insn.h >/dev/null && \ + diff -I'^#include' arch/x86/insn/inat.h ../../arch/x86/include/asm/inat.h >/dev/null && \ + diff -I'^#include' arch/x86/insn/inat_types.h ../../arch/x86/include/asm/inat_types.h >/dev/null) \ + || echo "Warning: stacktool: x86 instruction decoder differs from kernel" >&2 )) || true + $(QUIET_LINK)$(CC) $(STACKTOOL_IN) $(LDFLAGS) -o $@ + +clean: + $(call QUIET_CLEAN, stacktool) $(RM) $(STACKTOOL) + $(Q)find $(OUTPUT) -name '*.o' -delete -o -name '\.*.cmd' -delete -o -name '\.*.d' -delete + $(Q)$(RM) $(OUTPUT)arch/x86/insn/inat-tables.c + +FORCE: + +.PHONY: clean FORCE diff --git a/tools/stacktool/arch.h b/tools/stacktool/arch.h new file mode 100644 index 0000000..f7350fc --- /dev/null +++ b/tools/stacktool/arch.h @@ -0,0 +1,44 @@ +/* + * Copyright (C) 2015 Josh Poimboeuf <jpoimboe@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef _ARCH_H +#define _ARCH_H + +#include <stdbool.h> +#include "elf.h" + +#define INSN_FP_SAVE 1 +#define INSN_FP_SETUP 2 +#define INSN_FP_RESTORE 3 +#define INSN_JUMP_CONDITIONAL 4 +#define INSN_JUMP_UNCONDITIONAL 5 +#define INSN_JUMP_DYNAMIC 6 +#define INSN_CALL 7 +#define INSN_CALL_DYNAMIC 8 +#define INSN_RETURN 9 +#define INSN_CONTEXT_SWITCH 10 +#define INSN_BUG 11 +#define INSN_NOP 12 +#define INSN_OTHER 13 +#define INSN_LAST INSN_OTHER + +int arch_decode_instruction(struct elf *elf, struct section *sec, + unsigned long offset, unsigned int maxlen, + unsigned int *len, unsigned char *type, + unsigned long *displacement); + +#endif /* _ARCH_H */ diff --git a/tools/stacktool/arch/x86/Build b/tools/stacktool/arch/x86/Build new file mode 100644 index 0000000..9b9777d --- /dev/null +++ b/tools/stacktool/arch/x86/Build @@ -0,0 +1,12 @@ +stacktool-y += decode.o + +inat_tables_script = arch/x86/insn/gen-insn-attr-x86.awk +inat_tables_maps = arch/x86/insn/x86-opcode-map.txt + +$(OUTPUT)arch/x86/insn/inat-tables.c: $(inat_tables_script) $(inat_tables_maps) + $(call rule_mkdir) + $(Q)$(call echo-cmd,gen)$(AWK) -f $(inat_tables_script) $(inat_tables_maps) > $@ + +$(OUTPUT)arch/x86/decode.o: $(OUTPUT)arch/x86/insn/inat-tables.c + +CFLAGS_decode.o += -I$(OUTPUT)arch/x86/insn diff --git a/tools/stacktool/arch/x86/decode.c b/tools/stacktool/arch/x86/decode.c new file mode 100644 index 0000000..5aa78b9 --- /dev/null +++ b/tools/stacktool/arch/x86/decode.c @@ -0,0 +1,163 @@ +/* + * Copyright (C) 2015 Josh Poimboeuf <jpoimboe@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#include <stdio.h> +#include <stdlib.h> + +#define unlikely(cond) (cond) +#include "insn/insn.h" +#include "insn/inat.c" +#include "insn/insn.c" + +#include "../../elf.h" +#include "../../arch.h" +#include "../../warn.h" + +static int is_x86_64(struct elf *elf) +{ + switch (elf->ehdr.e_machine) { + case EM_X86_64: + return 1; + case EM_386: + return 0; + default: + WARN("unexpected ELF machine type %d", elf->ehdr.e_machine); + return -1; + } +} + +int arch_decode_instruction(struct elf *elf, struct section *sec, + unsigned long offset, unsigned int maxlen, + unsigned int *len, unsigned char *type, + unsigned long *immediate) +{ + struct insn insn; + int x86_64; + unsigned char op1, op2, ext; + + x86_64 = is_x86_64(elf); + if (x86_64 == -1) + return -1; + + insn_init(&insn, (void *)(sec->data + offset), maxlen, x86_64); + insn_get_length(&insn); + insn_get_opcode(&insn); + insn_get_modrm(&insn); + insn_get_immediate(&insn); + + if (!insn_complete(&insn)) { + WARN_FUNC("can't decode instruction", sec, offset); + return -1; + } + + *len = insn.length; + *type = INSN_OTHER; + + if (insn.vex_prefix.nbytes) + return 0; + + op1 = insn.opcode.bytes[0]; + op2 = insn.opcode.bytes[1]; + + switch (op1) { + case 0x55: + if (!insn.rex_prefix.nbytes) + /* push rbp */ + *type = INSN_FP_SAVE; + break; + + case 0x5d: + if (!insn.rex_prefix.nbytes) + /* pop rbp */ + *type = INSN_FP_RESTORE; + break; + + case 0x70 ... 0x7f: + *type = INSN_JUMP_CONDITIONAL; + break; + + case 0x89: + if (insn.rex_prefix.nbytes == 1 && + insn.rex_prefix.bytes[0] == 0x48 && + insn.modrm.nbytes && insn.modrm.bytes[0] == 0xe5) + /* mov rsp, rbp */ + *type = INSN_FP_SETUP; + break; + + case 0x90: + *type = INSN_NOP; + break; + + case 0x0f: + if (op2 >= 0x80 && op2 <= 0x8f) + *type = INSN_JUMP_CONDITIONAL; + else if (op2 == 0x05 || op2 == 0x07 || op2 == 0x34 || + op2 == 0x35) + /* sysenter, sysret */ + *type = INSN_CONTEXT_SWITCH; + else if (op2 == 0x0b || op2 == 0xb9) + /* ud2 */ + *type = INSN_BUG; + else if (op2 == 0x0d || op2 == 0x1f) + /* nopl/nopw */ + *type = INSN_NOP; + break; + + case 0xc9: /* leave */ + *type = INSN_FP_RESTORE; + break; + + case 0xe3: /* jecxz/jrcxz */ + *type = INSN_JUMP_CONDITIONAL; + break; + + case 0xe9: + case 0xeb: + *type = INSN_JUMP_UNCONDITIONAL; + break; + + case 0xc2: + case 0xc3: + *type = INSN_RETURN; + break; + + case 0xc5: /* iret */ + case 0xca: /* retf */ + case 0xcb: /* retf */ + *type = INSN_CONTEXT_SWITCH; + break; + + case 0xe8: + *type = INSN_CALL; + break; + + case 0xff: + ext = X86_MODRM_REG(insn.modrm.bytes[0]); + if (ext == 2 || ext == 3) + *type = INSN_CALL_DYNAMIC; + else if (ext == 4 || ext == 5) + *type = INSN_JUMP_DYNAMIC; + break; + + default: + break; + } + + *immediate = insn.immediate.nbytes ? insn.immediate.value : 0; + + return 0; +} diff --git a/tools/stacktool/arch/x86/insn/gen-insn-attr-x86.awk b/tools/stacktool/arch/x86/insn/gen-insn-attr-x86.awk new file mode 100644 index 0000000..093a892 --- /dev/null +++ b/tools/stacktool/arch/x86/insn/gen-insn-attr-x86.awk @@ -0,0 +1,387 @@ +#!/bin/awk -f +# gen-insn-attr-x86.awk: Instruction attribute table generator +# Written by Masami Hiramatsu <mhiramat@redhat.com> +# +# Usage: awk -f gen-insn-attr-x86.awk x86-opcode-map.txt > inat-tables.c + +# Awk implementation sanity check +function check_awk_implement() { + if (sprintf("%x", 0) != "0") + return "Your awk has a printf-format problem." + return "" +} + +# Clear working vars +function clear_vars() { + delete table + delete lptable2 + delete lptable1 + delete lptable3 + eid = -1 # escape id + gid = -1 # group id + aid = -1 # AVX id + tname = "" +} + +BEGIN { + # Implementation error checking + awkchecked = check_awk_implement() + if (awkchecked != "") { + print "Error: " awkchecked > "/dev/stderr" + print "Please try to use gawk." > "/dev/stderr" + exit 1 + } + + # Setup generating tables + print "/* x86 opcode map generated from x86-opcode-map.txt */" + print "/* Do not change this code. */\n" + ggid = 1 + geid = 1 + gaid = 0 + delete etable + delete gtable + delete atable + + opnd_expr = "^[A-Za-z/]" + ext_expr = "^\\(" + sep_expr = "^\\|$" + group_expr = "^Grp[0-9A-Za-z]+" + + imm_expr = "^[IJAOL][a-z]" + imm_flag["Ib"] = "INAT_MAKE_IMM(INAT_IMM_BYTE)" + imm_flag["Jb"] = "INAT_MAKE_IMM(INAT_IMM_BYTE)" + imm_flag["Iw"] = "INAT_MAKE_IMM(INAT_IMM_WORD)" + imm_flag["Id"] = "INAT_MAKE_IMM(INAT_IMM_DWORD)" + imm_flag["Iq"] = "INAT_MAKE_IMM(INAT_IMM_QWORD)" + imm_flag["Ap"] = "INAT_MAKE_IMM(INAT_IMM_PTR)" + imm_flag["Iz"] = "INAT_MAKE_IMM(INAT_IMM_VWORD32)" + imm_flag["Jz"] = "INAT_MAKE_IMM(INAT_IMM_VWORD32)" + imm_flag["Iv"] = "INAT_MAKE_IMM(INAT_IMM_VWORD)" + imm_flag["Ob"] = "INAT_MOFFSET" + imm_flag["Ov"] = "INAT_MOFFSET" + imm_flag["Lx"] = "INAT_MAKE_IMM(INAT_IMM_BYTE)" + + modrm_expr = "^([CDEGMNPQRSUVW/][a-z]+|NTA|T[012])" + force64_expr = "\\([df]64\\)" + rex_expr = "^REX(\\.[XRWB]+)*" + fpu_expr = "^ESC" # TODO + + lprefix1_expr = "\\((66|!F3)\\)" + lprefix2_expr = "\\(F3\\)" + lprefix3_expr = "\\((F2|!F3|66\\&F2)\\)" + lprefix_expr = "\\((66|F2|F3)\\)" + max_lprefix = 4 + + # All opcodes starting with lower-case 'v' or with (v1) superscript + # accepts VEX prefix + vexok_opcode_expr = "^v.*" + vexok_expr = "\\(v1\\)" + # All opcodes with (v) superscript supports *only* VEX prefix + vexonly_expr = "\\(v\\)" + + prefix_expr = "\\(Prefix\\)" + prefix_num["Operand-Size"] = "INAT_PFX_OPNDSZ" + prefix_num["REPNE"] = "INAT_PFX_REPNE" + prefix_num["REP/REPE"] = "INAT_PFX_REPE" + prefix_num["XACQUIRE"] = "INAT_PFX_REPNE" + prefix_num["XRELEASE"] = "INAT_PFX_REPE" + prefix_num["LOCK"] = "INAT_PFX_LOCK" + prefix_num["SEG=CS"] = "INAT_PFX_CS" + prefix_num["SEG=DS"] = "INAT_PFX_DS" + prefix_num["SEG=ES"] = "INAT_PFX_ES" + prefix_num["SEG=FS"] = "INAT_PFX_FS" + prefix_num["SEG=GS"] = "INAT_PFX_GS" + prefix_num["SEG=SS"] = "INAT_PFX_SS" + prefix_num["Address-Size"] = "INAT_PFX_ADDRSZ" + prefix_num["VEX+1byte"] = "INAT_PFX_VEX2" + prefix_num["VEX+2byte"] = "INAT_PFX_VEX3" + + clear_vars() +} + +function semantic_error(msg) { + print "Semantic error at " NR ": " msg > "/dev/stderr" + exit 1 +} + +function debug(msg) { + print "DEBUG: " msg +} + +function array_size(arr, i,c) { + c = 0 + for (i in arr) + c++ + return c +} + +/^Table:/ { + print "/* " $0 " */" + if (tname != "") + semantic_error("Hit Table: before EndTable:."); +} + +/^Referrer:/ { + if (NF != 1) { + # escape opcode table + ref = "" + for (i = 2; i <= NF; i++) + ref = ref $i + eid = escape[ref] + tname = sprintf("inat_escape_table_%d", eid) + } +} + +/^AVXcode:/ { + if (NF != 1) { + # AVX/escape opcode table + aid = $2 + if (gaid <= aid) + gaid = aid + 1 + if (tname == "") # AVX only opcode table + tname = sprintf("inat_avx_table_%d", $2) + } + if (aid == -1 && eid == -1) # primary opcode table + tname = "inat_primary_table" +} + +/^GrpTable:/ { + print "/* " $0 " */" + if (!($2 in group)) + semantic_error("No group: " $2 ) + gid = group[$2] + tname = "inat_group_table_" gid +} + +function print_table(tbl,name,fmt,n) +{ + print "const insn_attr_t " name " = {" + for (i = 0; i < n; i++) { + id = sprintf(fmt, i) + if (tbl[id]) + print " [" id "] = " tbl[id] "," + } + print "};" +} + +/^EndTable/ { + if (gid != -1) { + # print group tables + if (array_size(table) != 0) { + print_table(table, tname "[INAT_GROUP_TABLE_SIZE]", + "0x%x", 8) + gtable[gid,0] = tname + } + if (array_size(lptable1) != 0) { + print_table(lptable1, tname "_1[INAT_GROUP_TABLE_SIZE]", + "0x%x", 8) + gtable[gid,1] = tname "_1" + } + if (array_size(lptable2) != 0) { + print_table(lptable2, tname "_2[INAT_GROUP_TABLE_SIZE]", + "0x%x", 8) + gtable[gid,2] = tname "_2" + } + if (array_size(lptable3) != 0) { + print_table(lptable3, tname "_3[INAT_GROUP_TABLE_SIZE]", + "0x%x", 8) + gtable[gid,3] = tname "_3" + } + } else { + # print primary/escaped tables + if (array_size(table) != 0) { + print_table(table, tname "[INAT_OPCODE_TABLE_SIZE]", + "0x%02x", 256) + etable[eid,0] = tname + if (aid >= 0) + atable[aid,0] = tname + } + if (array_size(lptable1) != 0) { + print_table(lptable1,tname "_1[INAT_OPCODE_TABLE_SIZE]", + "0x%02x", 256) + etable[eid,1] = tname "_1" + if (aid >= 0) + atable[aid,1] = tname "_1" + } + if (array_size(lptable2) != 0) { + print_table(lptable2,tname "_2[INAT_OPCODE_TABLE_SIZE]", + "0x%02x", 256) + etable[eid,2] = tname "_2" + if (aid >= 0) + atable[aid,2] = tname "_2" + } + if (array_size(lptable3) != 0) { + print_table(lptable3,tname "_3[INAT_OPCODE_TABLE_SIZE]", + "0x%02x", 256) + etable[eid,3] = tname "_3" + if (aid >= 0) + atable[aid,3] = tname "_3" + } + } + print "" + clear_vars() +} + +function add_flags(old,new) { + if (old && new) + return old " | " new + else if (old) + return old + else + return new +} + +# convert operands to flags. +function convert_operands(count,opnd, i,j,imm,mod) +{ + imm = null + mod = null + for (j = 1; j <= count; j++) { + i = opnd[j] + if (match(i, imm_expr) == 1) { + if (!imm_flag[i]) + semantic_error("Unknown imm opnd: " i) + if (imm) { + if (i != "Ib") + semantic_error("Second IMM error") + imm = add_flags(imm, "INAT_SCNDIMM") + } else + imm = imm_flag[i] + } else if (match(i, modrm_expr)) + mod = "INAT_MODRM" + } + return add_flags(imm, mod) +} + +/^[0-9a-f]+\:/ { + if (NR == 1) + next + # get index + idx = "0x" substr($1, 1, index($1,":") - 1) + if (idx in table) + semantic_error("Redefine " idx " in " tname) + + # check if escaped opcode + if ("escape" == $2) { + if ($3 != "#") + semantic_error("No escaped name") + ref = "" + for (i = 4; i <= NF; i++) + ref = ref $i + if (ref in escape) + semantic_error("Redefine escape (" ref ")") + escape[ref] = geid + geid++ + table[idx] = "INAT_MAKE_ESCAPE(" escape[ref] ")" + next + } + + variant = null + # converts + i = 2 + while (i <= NF) { + opcode = $(i++) + delete opnds + ext = null + flags = null + opnd = null + # parse one opcode + if (match($i, opnd_expr)) { + opnd = $i + count = split($(i++), opnds, ",") + flags = convert_operands(count, opnds) + } + if (match($i, ext_expr)) + ext = $(i++) + if (match($i, sep_expr)) + i++ + else if (i < NF) + semantic_error($i " is not a separator") + + # check if group opcode + if (match(opcode, group_expr)) { + if (!(opcode in group)) { + group[opcode] = ggid + ggid++ + } + flags = add_flags(flags, "INAT_MAKE_GROUP(" group[opcode] ")") + } + # check force(or default) 64bit + if (match(ext, force64_expr)) + flags = add_flags(flags, "INAT_FORCE64") + + # check REX prefix + if (match(opcode, rex_expr)) + flags = add_flags(flags, "INAT_MAKE_PREFIX(INAT_PFX_REX)") + + # check coprocessor escape : TODO + if (match(opcode, fpu_expr)) + flags = add_flags(flags, "INAT_MODRM") + + # check VEX codes + if (match(ext, vexonly_expr)) + flags = add_flags(flags, "INAT_VEXOK | INAT_VEXONLY") + else if (match(ext, vexok_expr) || match(opcode, vexok_opcode_expr)) + flags = add_flags(flags, "INAT_VEXOK") + + # check prefixes + if (match(ext, prefix_expr)) { + if (!prefix_num[opcode]) + semantic_error("Unknown prefix: " opcode) + flags = add_flags(flags, "INAT_MAKE_PREFIX(" prefix_num[opcode] ")") + } + if (length(flags) == 0) + continue + # check if last prefix + if (match(ext, lprefix1_expr)) { + lptable1[idx] = add_flags(lptable1[idx],flags) + variant = "INAT_VARIANT" + } + if (match(ext, lprefix2_expr)) { + lptable2[idx] = add_flags(lptable2[idx],flags) + variant = "INAT_VARIANT" + } + if (match(ext, lprefix3_expr)) { + lptable3[idx] = add_flags(lptable3[idx],flags) + variant = "INAT_VARIANT" + } + if (!match(ext, lprefix_expr)){ + table[idx] = add_flags(table[idx],flags) + } + } + if (variant) + table[idx] = add_flags(table[idx],variant) +} + +END { + if (awkchecked != "") + exit 1 + # print escape opcode map's array + print "/* Escape opcode map array */" + print "const insn_attr_t * const inat_escape_tables[INAT_ESC_MAX + 1]" \ + "[INAT_LSTPFX_MAX + 1] = {" + for (i = 0; i < geid; i++) + for (j = 0; j < max_lprefix; j++) + if (etable[i,j]) + print " ["i"]["j"] = "etable[i,j]"," + print "};\n" + # print group opcode map's array + print "/* Group opcode map array */" + print "const insn_attr_t * const inat_group_tables[INAT_GRP_MAX + 1]"\ + "[INAT_LSTPFX_MAX + 1] = {" + for (i = 0; i < ggid; i++) + for (j = 0; j < max_lprefix; j++) + if (gtable[i,j]) + print " ["i"]["j"] = "gtable[i,j]"," + print "};\n" + # print AVX opcode map's array + print "/* AVX opcode map array */" + print "const insn_attr_t * const inat_avx_tables[X86_VEX_M_MAX + 1]"\ + "[INAT_LSTPFX_MAX + 1] = {" + for (i = 0; i < gaid; i++) + for (j = 0; j < max_lprefix; j++) + if (atable[i,j]) + print " ["i"]["j"] = "atable[i,j]"," + print "};" +} + diff --git a/tools/stacktool/arch/x86/insn/inat.c b/tools/stacktool/arch/x86/insn/inat.c new file mode 100644 index 0000000..e4bf28e --- /dev/null +++ b/tools/stacktool/arch/x86/insn/inat.c @@ -0,0 +1,97 @@ +/* + * x86 instruction attribute tables + * + * Written by Masami Hiramatsu <mhiramat@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + */ +#include "insn.h" + +/* Attribute tables are generated from opcode map */ +#include "inat-tables.c" + +/* Attribute search APIs */ +insn_attr_t inat_get_opcode_attribute(insn_byte_t opcode) +{ + return inat_primary_table[opcode]; +} + +int inat_get_last_prefix_id(insn_byte_t last_pfx) +{ + insn_attr_t lpfx_attr; + + lpfx_attr = inat_get_opcode_attribute(last_pfx); + return inat_last_prefix_id(lpfx_attr); +} + +insn_attr_t inat_get_escape_attribute(insn_byte_t opcode, int lpfx_id, + insn_attr_t esc_attr) +{ + const insn_attr_t *table; + int n; + + n = inat_escape_id(esc_attr); + + table = inat_escape_tables[n][0]; + if (!table) + return 0; + if (inat_has_variant(table[opcode]) && lpfx_id) { + table = inat_escape_tables[n][lpfx_id]; + if (!table) + return 0; + } + return table[opcode]; +} + +insn_attr_t inat_get_group_attribute(insn_byte_t modrm, int lpfx_id, + insn_attr_t grp_attr) +{ + const insn_attr_t *table; + int n; + + n = inat_group_id(grp_attr); + + table = inat_group_tables[n][0]; + if (!table) + return inat_group_common_attribute(grp_attr); + if (inat_has_variant(table[X86_MODRM_REG(modrm)]) && lpfx_id) { + table = inat_group_tables[n][lpfx_id]; + if (!table) + return inat_group_common_attribute(grp_attr); + } + return table[X86_MODRM_REG(modrm)] | + inat_group_common_attribute(grp_attr); +} + +insn_attr_t inat_get_avx_attribute(insn_byte_t opcode, insn_byte_t vex_m, + insn_byte_t vex_p) +{ + const insn_attr_t *table; + if (vex_m > X86_VEX_M_MAX || vex_p > INAT_LSTPFX_MAX) + return 0; + /* At first, this checks the master table */ + table = inat_avx_tables[vex_m][0]; + if (!table) + return 0; + if (!inat_is_group(table[opcode]) && vex_p) { + /* If this is not a group, get attribute directly */ + table = inat_avx_tables[vex_m][vex_p]; + if (!table) + return 0; + } + return table[opcode]; +} + diff --git a/tools/stacktool/arch/x86/insn/inat.h b/tools/stacktool/arch/x86/insn/inat.h new file mode 100644 index 0000000..611645e --- /dev/null +++ b/tools/stacktool/arch/x86/insn/inat.h @@ -0,0 +1,221 @@ +#ifndef _ASM_X86_INAT_H +#define _ASM_X86_INAT_H +/* + * x86 instruction attributes + * + * Written by Masami Hiramatsu <mhiramat@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + */ +#include "inat_types.h" + +/* + * Internal bits. Don't use bitmasks directly, because these bits are + * unstable. You should use checking functions. + */ + +#define INAT_OPCODE_TABLE_SIZE 256 +#define INAT_GROUP_TABLE_SIZE 8 + +/* Legacy last prefixes */ +#define INAT_PFX_OPNDSZ 1 /* 0x66 */ /* LPFX1 */ +#define INAT_PFX_REPE 2 /* 0xF3 */ /* LPFX2 */ +#define INAT_PFX_REPNE 3 /* 0xF2 */ /* LPFX3 */ +/* Other Legacy prefixes */ +#define INAT_PFX_LOCK 4 /* 0xF0 */ +#define INAT_PFX_CS 5 /* 0x2E */ +#define INAT_PFX_DS 6 /* 0x3E */ +#define INAT_PFX_ES 7 /* 0x26 */ +#define INAT_PFX_FS 8 /* 0x64 */ +#define INAT_PFX_GS 9 /* 0x65 */ +#define INAT_PFX_SS 10 /* 0x36 */ +#define INAT_PFX_ADDRSZ 11 /* 0x67 */ +/* x86-64 REX prefix */ +#define INAT_PFX_REX 12 /* 0x4X */ +/* AVX VEX prefixes */ +#define INAT_PFX_VEX2 13 /* 2-bytes VEX prefix */ +#define INAT_PFX_VEX3 14 /* 3-bytes VEX prefix */ + +#define INAT_LSTPFX_MAX 3 +#define INAT_LGCPFX_MAX 11 + +/* Immediate size */ +#define INAT_IMM_BYTE 1 +#define INAT_IMM_WORD 2 +#define INAT_IMM_DWORD 3 +#define INAT_IMM_QWORD 4 +#define INAT_IMM_PTR 5 +#define INAT_IMM_VWORD32 6 +#define INAT_IMM_VWORD 7 + +/* Legacy prefix */ +#define INAT_PFX_OFFS 0 +#define INAT_PFX_BITS 4 +#define INAT_PFX_MAX ((1 << INAT_PFX_BITS) - 1) +#define INAT_PFX_MASK (INAT_PFX_MAX << INAT_PFX_OFFS) +/* Escape opcodes */ +#define INAT_ESC_OFFS (INAT_PFX_OFFS + INAT_PFX_BITS) +#define INAT_ESC_BITS 2 +#define INAT_ESC_MAX ((1 << INAT_ESC_BITS) - 1) +#define INAT_ESC_MASK (INAT_ESC_MAX << INAT_ESC_OFFS) +/* Group opcodes (1-16) */ +#define INAT_GRP_OFFS (INAT_ESC_OFFS + INAT_ESC_BITS) +#define INAT_GRP_BITS 5 +#define INAT_GRP_MAX ((1 << INAT_GRP_BITS) - 1) +#define INAT_GRP_MASK (INAT_GRP_MAX << INAT_GRP_OFFS) +/* Immediates */ +#define INAT_IMM_OFFS (INAT_GRP_OFFS + INAT_GRP_BITS) +#define INAT_IMM_BITS 3 +#define INAT_IMM_MASK (((1 << INAT_IMM_BITS) - 1) << INAT_IMM_OFFS) +/* Flags */ +#define INAT_FLAG_OFFS (INAT_IMM_OFFS + INAT_IMM_BITS) +#define INAT_MODRM (1 << (INAT_FLAG_OFFS)) +#define INAT_FORCE64 (1 << (INAT_FLAG_OFFS + 1)) +#define INAT_SCNDIMM (1 << (INAT_FLAG_OFFS + 2)) +#define INAT_MOFFSET (1 << (INAT_FLAG_OFFS + 3)) +#define INAT_VARIANT (1 << (INAT_FLAG_OFFS + 4)) +#define INAT_VEXOK (1 << (INAT_FLAG_OFFS + 5)) +#define INAT_VEXONLY (1 << (INAT_FLAG_OFFS + 6)) +/* Attribute making macros for attribute tables */ +#define INAT_MAKE_PREFIX(pfx) (pfx << INAT_PFX_OFFS) +#define INAT_MAKE_ESCAPE(esc) (esc << INAT_ESC_OFFS) +#define INAT_MAKE_GROUP(grp) ((grp << INAT_GRP_OFFS) | INAT_MODRM) +#define INAT_MAKE_IMM(imm) (imm << INAT_IMM_OFFS) + +/* Attribute search APIs */ +extern insn_attr_t inat_get_opcode_attribute(insn_byte_t opcode); +extern int inat_get_last_prefix_id(insn_byte_t last_pfx); +extern insn_attr_t inat_get_escape_attribute(insn_byte_t opcode, + int lpfx_id, + insn_attr_t esc_attr); +extern insn_attr_t inat_get_group_attribute(insn_byte_t modrm, + int lpfx_id, + insn_attr_t esc_attr); +extern insn_attr_t inat_get_avx_attribute(insn_byte_t opcode, + insn_byte_t vex_m, + insn_byte_t vex_pp); + +/* Attribute checking functions */ +static inline int inat_is_legacy_prefix(insn_attr_t attr) +{ + attr &= INAT_PFX_MASK; + return attr && attr <= INAT_LGCPFX_MAX; +} + +static inline int inat_is_address_size_prefix(insn_attr_t attr) +{ + return (attr & INAT_PFX_MASK) == INAT_PFX_ADDRSZ; +} + +static inline int inat_is_operand_size_prefix(insn_attr_t attr) +{ + return (attr & INAT_PFX_MASK) == INAT_PFX_OPNDSZ; +} + +static inline int inat_is_rex_prefix(insn_attr_t attr) +{ + return (attr & INAT_PFX_MASK) == INAT_PFX_REX; +} + +static inline int inat_last_prefix_id(insn_attr_t attr) +{ + if ((attr & INAT_PFX_MASK) > INAT_LSTPFX_MAX) + return 0; + else + return attr & INAT_PFX_MASK; +} + +static inline int inat_is_vex_prefix(insn_attr_t attr) +{ + attr &= INAT_PFX_MASK; + return attr == INAT_PFX_VEX2 || attr == INAT_PFX_VEX3; +} + +static inline int inat_is_vex3_prefix(insn_attr_t attr) +{ + return (attr & INAT_PFX_MASK) == INAT_PFX_VEX3; +} + +static inline int inat_is_escape(insn_attr_t attr) +{ + return attr & INAT_ESC_MASK; +} + +static inline int inat_escape_id(insn_attr_t attr) +{ + return (attr & INAT_ESC_MASK) >> INAT_ESC_OFFS; +} + +static inline int inat_is_group(insn_attr_t attr) +{ + return attr & INAT_GRP_MASK; +} + +static inline int inat_group_id(insn_attr_t attr) +{ + return (attr & INAT_GRP_MASK) >> INAT_GRP_OFFS; +} + +static inline int inat_group_common_attribute(insn_attr_t attr) +{ + return attr & ~INAT_GRP_MASK; +} + +static inline int inat_has_immediate(insn_attr_t attr) +{ + return attr & INAT_IMM_MASK; +} + +static inline int inat_immediate_size(insn_attr_t attr) +{ + return (attr & INAT_IMM_MASK) >> INAT_IMM_OFFS; +} + +static inline int inat_has_modrm(insn_attr_t attr) +{ + return attr & INAT_MODRM; +} + +static inline int inat_is_force64(insn_attr_t attr) +{ + return attr & INAT_FORCE64; +} + +static inline int inat_has_second_immediate(insn_attr_t attr) +{ + return attr & INAT_SCNDIMM; +} + +static inline int inat_has_moffset(insn_attr_t attr) +{ + return attr & INAT_MOFFSET; +} + +static inline int inat_has_variant(insn_attr_t attr) +{ + return attr & INAT_VARIANT; +} + +static inline int inat_accept_vex(insn_attr_t attr) +{ + return attr & INAT_VEXOK; +} + +static inline int inat_must_vex(insn_attr_t attr) +{ + return attr & INAT_VEXONLY; +} +#endif diff --git a/tools/stacktool/arch/x86/insn/inat_types.h b/tools/stacktool/arch/x86/insn/inat_types.h new file mode 100644 index 0000000..cb3c20c --- /dev/null +++ b/tools/stacktool/arch/x86/insn/inat_types.h @@ -0,0 +1,29 @@ +#ifndef _ASM_X86_INAT_TYPES_H +#define _ASM_X86_INAT_TYPES_H +/* + * x86 instruction attributes + * + * Written by Masami Hiramatsu <mhiramat@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + */ + +/* Instruction attributes */ +typedef unsigned int insn_attr_t; +typedef unsigned char insn_byte_t; +typedef signed int insn_value_t; + +#endif diff --git a/tools/stacktool/arch/x86/insn/insn.c b/tools/stacktool/arch/x86/insn/insn.c new file mode 100644 index 0000000..47314a6 --- /dev/null +++ b/tools/stacktool/arch/x86/insn/insn.c @@ -0,0 +1,594 @@ +/* + * x86 instruction analysis + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright (C) IBM Corporation, 2002, 2004, 2009 + */ + +#ifdef __KERNEL__ +#include <linux/string.h> +#else +#include <string.h> +#endif +#include "inat.h" +#include "insn.h" + +/* Verify next sizeof(t) bytes can be on the same instruction */ +#define validate_next(t, insn, n) \ + ((insn)->next_byte + sizeof(t) + n <= (insn)->end_kaddr) + +#define __get_next(t, insn) \ + ({ t r = *(t*)insn->next_byte; insn->next_byte += sizeof(t); r; }) + +#define __peek_nbyte_next(t, insn, n) \ + ({ t r = *(t*)((insn)->next_byte + n); r; }) + +#define get_next(t, insn) \ + ({ if (unlikely(!validate_next(t, insn, 0))) goto err_out; __get_next(t, insn); }) + +#define peek_nbyte_next(t, insn, n) \ + ({ if (unlikely(!validate_next(t, insn, n))) goto err_out; __peek_nbyte_next(t, insn, n); }) + +#define peek_next(t, insn) peek_nbyte_next(t, insn, 0) + +/** + * insn_init() - initialize struct insn + * @insn: &struct insn to be initialized + * @kaddr: address (in kernel memory) of instruction (or copy thereof) + * @x86_64: !0 for 64-bit kernel or 64-bit app + */ +void insn_init(struct insn *insn, const void *kaddr, int buf_len, int x86_64) +{ + /* + * Instructions longer than MAX_INSN_SIZE (15 bytes) are invalid + * even if the input buffer is long enough to hold them. + */ + if (buf_len > MAX_INSN_SIZE) + buf_len = MAX_INSN_SIZE; + + memset(insn, 0, sizeof(*insn)); + insn->kaddr = kaddr; + insn->end_kaddr = kaddr + buf_len; + insn->next_byte = kaddr; + insn->x86_64 = x86_64 ? 1 : 0; + insn->opnd_bytes = 4; + if (x86_64) + insn->addr_bytes = 8; + else + insn->addr_bytes = 4; +} + +/** + * insn_get_prefixes - scan x86 instruction prefix bytes + * @insn: &struct insn containing instruction + * + * Populates the @insn->prefixes bitmap, and updates @insn->next_byte + * to point to the (first) opcode. No effect if @insn->prefixes.got + * is already set. + */ +void insn_get_prefixes(struct insn *insn) +{ + struct insn_field *prefixes = &insn->prefixes; + insn_attr_t attr; + insn_byte_t b, lb; + int i, nb; + + if (prefixes->got) + return; + + nb = 0; + lb = 0; + b = peek_next(insn_byte_t, insn); + attr = inat_get_opcode_attribute(b); + while (inat_is_legacy_prefix(attr)) { + /* Skip if same prefix */ + for (i = 0; i < nb; i++) + if (prefixes->bytes[i] == b) + goto found; + if (nb == 4) + /* Invalid instruction */ + break; + prefixes->bytes[nb++] = b; + if (inat_is_address_size_prefix(attr)) { + /* address size switches 2/4 or 4/8 */ + if (insn->x86_64) + insn->addr_bytes ^= 12; + else + insn->addr_bytes ^= 6; + } else if (inat_is_operand_size_prefix(attr)) { + /* oprand size switches 2/4 */ + insn->opnd_bytes ^= 6; + } +found: + prefixes->nbytes++; + insn->next_byte++; + lb = b; + b = peek_next(insn_byte_t, insn); + attr = inat_get_opcode_attribute(b); + } + /* Set the last prefix */ + if (lb && lb != insn->prefixes.bytes[3]) { + if (unlikely(insn->prefixes.bytes[3])) { + /* Swap the last prefix */ + b = insn->prefixes.bytes[3]; + for (i = 0; i < nb; i++) + if (prefixes->bytes[i] == lb) + prefixes->bytes[i] = b; + } + insn->prefixes.bytes[3] = lb; + } + + /* Decode REX prefix */ + if (insn->x86_64) { + b = peek_next(insn_byte_t, insn); + attr = inat_get_opcode_attribute(b); + if (inat_is_rex_prefix(attr)) { + insn->rex_prefix.value = b; + insn->rex_prefix.nbytes = 1; + insn->next_byte++; + if (X86_REX_W(b)) + /* REX.W overrides opnd_size */ + insn->opnd_bytes = 8; + } + } + insn->rex_prefix.got = 1; + + /* Decode VEX prefix */ + b = peek_next(insn_byte_t, insn); + attr = inat_get_opcode_attribute(b); + if (inat_is_vex_prefix(attr)) { + insn_byte_t b2 = peek_nbyte_next(insn_byte_t, insn, 1); + if (!insn->x86_64) { + /* + * In 32-bits mode, if the [7:6] bits (mod bits of + * ModRM) on the second byte are not 11b, it is + * LDS or LES. + */ + if (X86_MODRM_MOD(b2) != 3) + goto vex_end; + } + insn->vex_prefix.bytes[0] = b; + insn->vex_prefix.bytes[1] = b2; + if (inat_is_vex3_prefix(attr)) { + b2 = peek_nbyte_next(insn_byte_t, insn, 2); + insn->vex_prefix.bytes[2] = b2; + insn->vex_prefix.nbytes = 3; + insn->next_byte += 3; + if (insn->x86_64 && X86_VEX_W(b2)) + /* VEX.W overrides opnd_size */ + insn->opnd_bytes = 8; + } else { + /* + * For VEX2, fake VEX3-like byte#2. + * Makes it easier to decode vex.W, vex.vvvv, + * vex.L and vex.pp. Masking with 0x7f sets vex.W == 0. + */ + insn->vex_prefix.bytes[2] = b2 & 0x7f; + insn->vex_prefix.nbytes = 2; + insn->next_byte += 2; + } + } +vex_end: + insn->vex_prefix.got = 1; + + prefixes->got = 1; + +err_out: + return; +} + +/** + * insn_get_opcode - collect opcode(s) + * @insn: &struct insn containing instruction + * + * Populates @insn->opcode, updates @insn->next_byte to point past the + * opcode byte(s), and set @insn->attr (except for groups). + * If necessary, first collects any preceding (prefix) bytes. + * Sets @insn->opcode.value = opcode1. No effect if @insn->opcode.got + * is already 1. + */ +void insn_get_opcode(struct insn *insn) +{ + struct insn_field *opcode = &insn->opcode; + insn_byte_t op; + int pfx_id; + if (opcode->got) + return; + if (!insn->prefixes.got) + insn_get_prefixes(insn); + + /* Get first opcode */ + op = get_next(insn_byte_t, insn); + opcode->bytes[0] = op; + opcode->nbytes = 1; + + /* Check if there is VEX prefix or not */ + if (insn_is_avx(insn)) { + insn_byte_t m, p; + m = insn_vex_m_bits(insn); + p = insn_vex_p_bits(insn); + insn->attr = inat_get_avx_attribute(op, m, p); + if (!inat_accept_vex(insn->attr) && !inat_is_group(insn->attr)) + insn->attr = 0; /* This instruction is bad */ + goto end; /* VEX has only 1 byte for opcode */ + } + + insn->attr = inat_get_opcode_attribute(op); + while (inat_is_escape(insn->attr)) { + /* Get escaped opcode */ + op = get_next(insn_byte_t, insn); + opcode->bytes[opcode->nbytes++] = op; + pfx_id = insn_last_prefix_id(insn); + insn->attr = inat_get_escape_attribute(op, pfx_id, insn->attr); + } + if (inat_must_vex(insn->attr)) + insn->attr = 0; /* This instruction is bad */ +end: + opcode->got = 1; + +err_out: + return; +} + +/** + * insn_get_modrm - collect ModRM byte, if any + * @insn: &struct insn containing instruction + * + * Populates @insn->modrm and updates @insn->next_byte to point past the + * ModRM byte, if any. If necessary, first collects the preceding bytes + * (prefixes and opcode(s)). No effect if @insn->modrm.got is already 1. + */ +void insn_get_modrm(struct insn *insn) +{ + struct insn_field *modrm = &insn->modrm; + insn_byte_t pfx_id, mod; + if (modrm->got) + return; + if (!insn->opcode.got) + insn_get_opcode(insn); + + if (inat_has_modrm(insn->attr)) { + mod = get_next(insn_byte_t, insn); + modrm->value = mod; + modrm->nbytes = 1; + if (inat_is_group(insn->attr)) { + pfx_id = insn_last_prefix_id(insn); + insn->attr = inat_get_group_attribute(mod, pfx_id, + insn->attr); + if (insn_is_avx(insn) && !inat_accept_vex(insn->attr)) + insn->attr = 0; /* This is bad */ + } + } + + if (insn->x86_64 && inat_is_force64(insn->attr)) + insn->opnd_bytes = 8; + modrm->got = 1; + +err_out: + return; +} + + +/** + * insn_rip_relative() - Does instruction use RIP-relative addressing mode? + * @insn: &struct insn containing instruction + * + * If necessary, first collects the instruction up to and including the + * ModRM byte. No effect if @insn->x86_64 is 0. + */ +int insn_rip_relative(struct insn *insn) +{ + struct insn_field *modrm = &insn->modrm; + + if (!insn->x86_64) + return 0; + if (!modrm->got) + insn_get_modrm(insn); + /* + * For rip-relative instructions, the mod field (top 2 bits) + * is zero and the r/m field (bottom 3 bits) is 0x5. + */ + return (modrm->nbytes && (modrm->value & 0xc7) == 0x5); +} + +/** + * insn_get_sib() - Get the SIB byte of instruction + * @insn: &struct insn containing instruction + * + * If necessary, first collects the instruction up to and including the + * ModRM byte. + */ +void insn_get_sib(struct insn *insn) +{ + insn_byte_t modrm; + + if (insn->sib.got) + return; + if (!insn->modrm.got) + insn_get_modrm(insn); + if (insn->modrm.nbytes) { + modrm = (insn_byte_t)insn->modrm.value; + if (insn->addr_bytes != 2 && + X86_MODRM_MOD(modrm) != 3 && X86_MODRM_RM(modrm) == 4) { + insn->sib.value = get_next(insn_byte_t, insn); + insn->sib.nbytes = 1; + } + } + insn->sib.got = 1; + +err_out: + return; +} + + +/** + * insn_get_displacement() - Get the displacement of instruction + * @insn: &struct insn containing instruction + * + * If necessary, first collects the instruction up to and including the + * SIB byte. + * Displacement value is sign-expanded. + */ +void insn_get_displacement(struct insn *insn) +{ + insn_byte_t mod, rm, base; + + if (insn->displacement.got) + return; + if (!insn->sib.got) + insn_get_sib(insn); + if (insn->modrm.nbytes) { + /* + * Interpreting the modrm byte: + * mod = 00 - no displacement fields (exceptions below) + * mod = 01 - 1-byte displacement field + * mod = 10 - displacement field is 4 bytes, or 2 bytes if + * address size = 2 (0x67 prefix in 32-bit mode) + * mod = 11 - no memory operand + * + * If address size = 2... + * mod = 00, r/m = 110 - displacement field is 2 bytes + * + * If address size != 2... + * mod != 11, r/m = 100 - SIB byte exists + * mod = 00, SIB base = 101 - displacement field is 4 bytes + * mod = 00, r/m = 101 - rip-relative addressing, displacement + * field is 4 bytes + */ + mod = X86_MODRM_MOD(insn->modrm.value); + rm = X86_MODRM_RM(insn->modrm.value); + base = X86_SIB_BASE(insn->sib.value); + if (mod == 3) + goto out; + if (mod == 1) { + insn->displacement.value = get_next(char, insn); + insn->displacement.nbytes = 1; + } else if (insn->addr_bytes == 2) { + if ((mod == 0 && rm == 6) || mod == 2) { + insn->displacement.value = + get_next(short, insn); + insn->displacement.nbytes = 2; + } + } else { + if ((mod == 0 && rm == 5) || mod == 2 || + (mod == 0 && base == 5)) { + insn->displacement.value = get_next(int, insn); + insn->displacement.nbytes = 4; + } + } + } +out: + insn->displacement.got = 1; + +err_out: + return; +} + +/* Decode moffset16/32/64. Return 0 if failed */ +static int __get_moffset(struct insn *insn) +{ + switch (insn->addr_bytes) { + case 2: + insn->moffset1.value = get_next(short, insn); + insn->moffset1.nbytes = 2; + break; + case 4: + insn->moffset1.value = get_next(int, insn); + insn->moffset1.nbytes = 4; + break; + case 8: + insn->moffset1.value = get_next(int, insn); + insn->moffset1.nbytes = 4; + insn->moffset2.value = get_next(int, insn); + insn->moffset2.nbytes = 4; + break; + default: /* opnd_bytes must be modified manually */ + goto err_out; + } + insn->moffset1.got = insn->moffset2.got = 1; + + return 1; + +err_out: + return 0; +} + +/* Decode imm v32(Iz). Return 0 if failed */ +static int __get_immv32(struct insn *insn) +{ + switch (insn->opnd_bytes) { + case 2: + insn->immediate.value = get_next(short, insn); + insn->immediate.nbytes = 2; + break; + case 4: + case 8: + insn->immediate.value = get_next(int, insn); + insn->immediate.nbytes = 4; + break; + default: /* opnd_bytes must be modified manually */ + goto err_out; + } + + return 1; + +err_out: + return 0; +} + +/* Decode imm v64(Iv/Ov), Return 0 if failed */ +static int __get_immv(struct insn *insn) +{ + switch (insn->opnd_bytes) { + case 2: + insn->immediate1.value = get_next(short, insn); + insn->immediate1.nbytes = 2; + break; + case 4: + insn->immediate1.value = get_next(int, insn); + insn->immediate1.nbytes = 4; + break; + case 8: + insn->immediate1.value = get_next(int, insn); + insn->immediate1.nbytes = 4; + insn->immediate2.value = get_next(int, insn); + insn->immediate2.nbytes = 4; + break; + default: /* opnd_bytes must be modified manually */ + goto err_out; + } + insn->immediate1.got = insn->immediate2.got = 1; + + return 1; +err_out: + return 0; +} + +/* Decode ptr16:16/32(Ap) */ +static int __get_immptr(struct insn *insn) +{ + switch (insn->opnd_bytes) { + case 2: + insn->immediate1.value = get_next(short, insn); + insn->immediate1.nbytes = 2; + break; + case 4: + insn->immediate1.value = get_next(int, insn); + insn->immediate1.nbytes = 4; + break; + case 8: + /* ptr16:64 is not exist (no segment) */ + return 0; + default: /* opnd_bytes must be modified manually */ + goto err_out; + } + insn->immediate2.value = get_next(unsigned short, insn); + insn->immediate2.nbytes = 2; + insn->immediate1.got = insn->immediate2.got = 1; + + return 1; +err_out: + return 0; +} + +/** + * insn_get_immediate() - Get the immediates of instruction + * @insn: &struct insn containing instruction + * + * If necessary, first collects the instruction up to and including the + * displacement bytes. + * Basically, most of immediates are sign-expanded. Unsigned-value can be + * get by bit masking with ((1 << (nbytes * 8)) - 1) + */ +void insn_get_immediate(struct insn *insn) +{ + if (insn->immediate.got) + return; + if (!insn->displacement.got) + insn_get_displacement(insn); + + if (inat_has_moffset(insn->attr)) { + if (!__get_moffset(insn)) + goto err_out; + goto done; + } + + if (!inat_has_immediate(insn->attr)) + /* no immediates */ + goto done; + + switch (inat_immediate_size(insn->attr)) { + case INAT_IMM_BYTE: + insn->immediate.value = get_next(char, insn); + insn->immediate.nbytes = 1; + break; + case INAT_IMM_WORD: + insn->immediate.value = get_next(short, insn); + insn->immediate.nbytes = 2; + break; + case INAT_IMM_DWORD: + insn->immediate.value = get_next(int, insn); + insn->immediate.nbytes = 4; + break; + case INAT_IMM_QWORD: + insn->immediate1.value = get_next(int, insn); + insn->immediate1.nbytes = 4; + insn->immediate2.value = get_next(int, insn); + insn->immediate2.nbytes = 4; + break; + case INAT_IMM_PTR: + if (!__get_immptr(insn)) + goto err_out; + break; + case INAT_IMM_VWORD32: + if (!__get_immv32(insn)) + goto err_out; + break; + case INAT_IMM_VWORD: + if (!__get_immv(insn)) + goto err_out; + break; + default: + /* Here, insn must have an immediate, but failed */ + goto err_out; + } + if (inat_has_second_immediate(insn->attr)) { + insn->immediate2.value = get_next(char, insn); + insn->immediate2.nbytes = 1; + } +done: + insn->immediate.got = 1; + +err_out: + return; +} + +/** + * insn_get_length() - Get the length of instruction + * @insn: &struct insn containing instruction + * + * If necessary, first collects the instruction up to and including the + * immediates bytes. + */ +void insn_get_length(struct insn *insn) +{ + if (insn->length) + return; + if (!insn->immediate.got) + insn_get_immediate(insn); + insn->length = (unsigned char)((unsigned long)insn->next_byte + - (unsigned long)insn->kaddr); +} diff --git a/tools/stacktool/arch/x86/insn/insn.h b/tools/stacktool/arch/x86/insn/insn.h new file mode 100644 index 0000000..dd12da0 --- /dev/null +++ b/tools/stacktool/arch/x86/insn/insn.h @@ -0,0 +1,201 @@ +#ifndef _ASM_X86_INSN_H +#define _ASM_X86_INSN_H +/* + * x86 instruction analysis + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright (C) IBM Corporation, 2009 + */ + +/* insn_attr_t is defined in inat.h */ +#include "inat.h" + +struct insn_field { + union { + insn_value_t value; + insn_byte_t bytes[4]; + }; + /* !0 if we've run insn_get_xxx() for this field */ + unsigned char got; + unsigned char nbytes; +}; + +struct insn { + struct insn_field prefixes; /* + * Prefixes + * prefixes.bytes[3]: last prefix + */ + struct insn_field rex_prefix; /* REX prefix */ + struct insn_field vex_prefix; /* VEX prefix */ + struct insn_field opcode; /* + * opcode.bytes[0]: opcode1 + * opcode.bytes[1]: opcode2 + * opcode.bytes[2]: opcode3 + */ + struct insn_field modrm; + struct insn_field sib; + struct insn_field displacement; + union { + struct insn_field immediate; + struct insn_field moffset1; /* for 64bit MOV */ + struct insn_field immediate1; /* for 64bit imm or off16/32 */ + }; + union { + struct insn_field moffset2; /* for 64bit MOV */ + struct insn_field immediate2; /* for 64bit imm or seg16 */ + }; + + insn_attr_t attr; + unsigned char opnd_bytes; + unsigned char addr_bytes; + unsigned char length; + unsigned char x86_64; + + const insn_byte_t *kaddr; /* kernel address of insn to analyze */ + const insn_byte_t *end_kaddr; /* kernel address of last insn in buffer */ + const insn_byte_t *next_byte; +}; + +#define MAX_INSN_SIZE 15 + +#define X86_MODRM_MOD(modrm) (((modrm) & 0xc0) >> 6) +#define X86_MODRM_REG(modrm) (((modrm) & 0x38) >> 3) +#define X86_MODRM_RM(modrm) ((modrm) & 0x07) + +#define X86_SIB_SCALE(sib) (((sib) & 0xc0) >> 6) +#define X86_SIB_INDEX(sib) (((sib) & 0x38) >> 3) +#define X86_SIB_BASE(sib) ((sib) & 0x07) + +#define X86_REX_W(rex) ((rex) & 8) +#define X86_REX_R(rex) ((rex) & 4) +#define X86_REX_X(rex) ((rex) & 2) +#define X86_REX_B(rex) ((rex) & 1) + +/* VEX bit flags */ +#define X86_VEX_W(vex) ((vex) & 0x80) /* VEX3 Byte2 */ +#define X86_VEX_R(vex) ((vex) & 0x80) /* VEX2/3 Byte1 */ +#define X86_VEX_X(vex) ((vex) & 0x40) /* VEX3 Byte1 */ +#define X86_VEX_B(vex) ((vex) & 0x20) /* VEX3 Byte1 */ +#define X86_VEX_L(vex) ((vex) & 0x04) /* VEX3 Byte2, VEX2 Byte1 */ +/* VEX bit fields */ +#define X86_VEX3_M(vex) ((vex) & 0x1f) /* VEX3 Byte1 */ +#define X86_VEX2_M 1 /* VEX2.M always 1 */ +#define X86_VEX_V(vex) (((vex) & 0x78) >> 3) /* VEX3 Byte2, VEX2 Byte1 */ +#define X86_VEX_P(vex) ((vex) & 0x03) /* VEX3 Byte2, VEX2 Byte1 */ +#define X86_VEX_M_MAX 0x1f /* VEX3.M Maximum value */ + +extern void insn_init(struct insn *insn, const void *kaddr, int buf_len, int x86_64); +extern void insn_get_prefixes(struct insn *insn); +extern void insn_get_opcode(struct insn *insn); +extern void insn_get_modrm(struct insn *insn); +extern void insn_get_sib(struct insn *insn); +extern void insn_get_displacement(struct insn *insn); +extern void insn_get_immediate(struct insn *insn); +extern void insn_get_length(struct insn *insn); + +/* Attribute will be determined after getting ModRM (for opcode groups) */ +static inline void insn_get_attribute(struct insn *insn) +{ + insn_get_modrm(insn); +} + +/* Instruction uses RIP-relative addressing */ +extern int insn_rip_relative(struct insn *insn); + +/* Init insn for kernel text */ +static inline void kernel_insn_init(struct insn *insn, + const void *kaddr, int buf_len) +{ +#ifdef CONFIG_X86_64 + insn_init(insn, kaddr, buf_len, 1); +#else /* CONFIG_X86_32 */ + insn_init(insn, kaddr, buf_len, 0); +#endif +} + +static inline int insn_is_avx(struct insn *insn) +{ + if (!insn->prefixes.got) + insn_get_prefixes(insn); + return (insn->vex_prefix.value != 0); +} + +/* Ensure this instruction is decoded completely */ +static inline int insn_complete(struct insn *insn) +{ + return insn->opcode.got && insn->modrm.got && insn->sib.got && + insn->displacement.got && insn->immediate.got; +} + +static inline insn_byte_t insn_vex_m_bits(struct insn *insn) +{ + if (insn->vex_prefix.nbytes == 2) /* 2 bytes VEX */ + return X86_VEX2_M; + else + return X86_VEX3_M(insn->vex_prefix.bytes[1]); +} + +static inline insn_byte_t insn_vex_p_bits(struct insn *insn) +{ + if (insn->vex_prefix.nbytes == 2) /* 2 bytes VEX */ + return X86_VEX_P(insn->vex_prefix.bytes[1]); + else + return X86_VEX_P(insn->vex_prefix.bytes[2]); +} + +/* Get the last prefix id from last prefix or VEX prefix */ +static inline int insn_last_prefix_id(struct insn *insn) +{ + if (insn_is_avx(insn)) + return insn_vex_p_bits(insn); /* VEX_p is a SIMD prefix id */ + + if (insn->prefixes.bytes[3]) + return inat_get_last_prefix_id(insn->prefixes.bytes[3]); + + return 0; +} + +/* Offset of each field from kaddr */ +static inline int insn_offset_rex_prefix(struct insn *insn) +{ + return insn->prefixes.nbytes; +} +static inline int insn_offset_vex_prefix(struct insn *insn) +{ + return insn_offset_rex_prefix(insn) + insn->rex_prefix.nbytes; +} +static inline int insn_offset_opcode(struct insn *insn) +{ + return insn_offset_vex_prefix(insn) + insn->vex_prefix.nbytes; +} +static inline int insn_offset_modrm(struct insn *insn) +{ + return insn_offset_opcode(insn) + insn->opcode.nbytes; +} +static inline int insn_offset_sib(struct insn *insn) +{ + return insn_offset_modrm(insn) + insn->modrm.nbytes; +} +static inline int insn_offset_displacement(struct insn *insn) +{ + return insn_offset_sib(insn) + insn->sib.nbytes; +} +static inline int insn_offset_immediate(struct insn *insn) +{ + return insn_offset_displacement(insn) + insn->displacement.nbytes; +} + +#endif /* _ASM_X86_INSN_H */ diff --git a/tools/stacktool/arch/x86/insn/x86-opcode-map.txt b/tools/stacktool/arch/x86/insn/x86-opcode-map.txt new file mode 100644 index 0000000..d388de7 --- /dev/null +++ b/tools/stacktool/arch/x86/insn/x86-opcode-map.txt @@ -0,0 +1,984 @@ +# x86 Opcode Maps +# +# This is (mostly) based on following documentations. +# - Intel(R) 64 and IA-32 Architectures Software Developer's Manual Vol.2C +# (#326018-047US, June 2013) +# +#<Opcode maps> +# Table: table-name +# Referrer: escaped-name +# AVXcode: avx-code +# opcode: mnemonic|GrpXXX [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...] +# (or) +# opcode: escape # escaped-name +# EndTable +# +#<group maps> +# GrpTable: GrpXXX +# reg: mnemonic [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...] +# EndTable +# +# AVX Superscripts +# (v): this opcode requires VEX prefix. +# (v1): this opcode only supports 128bit VEX. +# +# Last Prefix Superscripts +# - (66): the last prefix is 0x66 +# - (F3): the last prefix is 0xF3 +# - (F2): the last prefix is 0xF2 +# - (!F3) : the last prefix is not 0xF3 (including non-last prefix case) +# - (66&F2): Both 0x66 and 0xF2 prefixes are specified. + +Table: one byte opcode +Referrer: +AVXcode: +# 0x00 - 0x0f +00: ADD Eb,Gb +01: ADD Ev,Gv +02: ADD Gb,Eb +03: ADD Gv,Ev +04: ADD AL,Ib +05: ADD rAX,Iz +06: PUSH ES (i64) +07: POP ES (i64) +08: OR Eb,Gb +09: OR Ev,Gv +0a: OR Gb,Eb +0b: OR Gv,Ev +0c: OR AL,Ib +0d: OR rAX,Iz +0e: PUSH CS (i64) +0f: escape # 2-byte escape +# 0x10 - 0x1f +10: ADC Eb,Gb +11: ADC Ev,Gv +12: ADC Gb,Eb +13: ADC Gv,Ev +14: ADC AL,Ib +15: ADC rAX,Iz +16: PUSH SS (i64) +17: POP SS (i64) +18: SBB Eb,Gb +19: SBB Ev,Gv +1a: SBB Gb,Eb +1b: SBB Gv,Ev +1c: SBB AL,Ib +1d: SBB rAX,Iz +1e: PUSH DS (i64) +1f: POP DS (i64) +# 0x20 - 0x2f +20: AND Eb,Gb +21: AND Ev,Gv +22: AND Gb,Eb +23: AND Gv,Ev +24: AND AL,Ib +25: AND rAx,Iz +26: SEG=ES (Prefix) +27: DAA (i64) +28: SUB Eb,Gb +29: SUB Ev,Gv +2a: SUB Gb,Eb +2b: SUB Gv,Ev +2c: SUB AL,Ib +2d: SUB rAX,Iz +2e: SEG=CS (Prefix) +2f: DAS (i64) +# 0x30 - 0x3f +30: XOR Eb,Gb +31: XOR Ev,Gv +32: XOR Gb,Eb +33: XOR Gv,Ev +34: XOR AL,Ib +35: XOR rAX,Iz +36: SEG=SS (Prefix) +37: AAA (i64) +38: CMP Eb,Gb +39: CMP Ev,Gv +3a: CMP Gb,Eb +3b: CMP Gv,Ev +3c: CMP AL,Ib +3d: CMP rAX,Iz +3e: SEG=DS (Prefix) +3f: AAS (i64) +# 0x40 - 0x4f +40: INC eAX (i64) | REX (o64) +41: INC eCX (i64) | REX.B (o64) +42: INC eDX (i64) | REX.X (o64) +43: INC eBX (i64) | REX.XB (o64) +44: INC eSP (i64) | REX.R (o64) +45: INC eBP (i64) | REX.RB (o64) +46: INC eSI (i64) | REX.RX (o64) +47: INC eDI (i64) | REX.RXB (o64) +48: DEC eAX (i64) | REX.W (o64) +49: DEC eCX (i64) | REX.WB (o64) +4a: DEC eDX (i64) | REX.WX (o64) +4b: DEC eBX (i64) | REX.WXB (o64) +4c: DEC eSP (i64) | REX.WR (o64) +4d: DEC eBP (i64) | REX.WRB (o64) +4e: DEC eSI (i64) | REX.WRX (o64) +4f: DEC eDI (i64) | REX.WRXB (o64) +# 0x50 - 0x5f +50: PUSH rAX/r8 (d64) +51: PUSH rCX/r9 (d64) +52: PUSH rDX/r10 (d64) +53: PUSH rBX/r11 (d64) +54: PUSH rSP/r12 (d64) +55: PUSH rBP/r13 (d64) +56: PUSH rSI/r14 (d64) +57: PUSH rDI/r15 (d64) +58: POP rAX/r8 (d64) +59: POP rCX/r9 (d64) +5a: POP rDX/r10 (d64) +5b: POP rBX/r11 (d64) +5c: POP rSP/r12 (d64) +5d: POP rBP/r13 (d64) +5e: POP rSI/r14 (d64) +5f: POP rDI/r15 (d64) +# 0x60 - 0x6f +60: PUSHA/PUSHAD (i64) +61: POPA/POPAD (i64) +62: BOUND Gv,Ma (i64) +63: ARPL Ew,Gw (i64) | MOVSXD Gv,Ev (o64) +64: SEG=FS (Prefix) +65: SEG=GS (Prefix) +66: Operand-Size (Prefix) +67: Address-Size (Prefix) +68: PUSH Iz (d64) +69: IMUL Gv,Ev,Iz +6a: PUSH Ib (d64) +6b: IMUL Gv,Ev,Ib +6c: INS/INSB Yb,DX +6d: INS/INSW/INSD Yz,DX +6e: OUTS/OUTSB DX,Xb +6f: OUTS/OUTSW/OUTSD DX,Xz +# 0x70 - 0x7f +70: JO Jb +71: JNO Jb +72: JB/JNAE/JC Jb +73: JNB/JAE/JNC Jb +74: JZ/JE Jb +75: JNZ/JNE Jb +76: JBE/JNA Jb +77: JNBE/JA Jb +78: JS Jb +79: JNS Jb +7a: JP/JPE Jb +7b: JNP/JPO Jb +7c: JL/JNGE Jb +7d: JNL/JGE Jb +7e: JLE/JNG Jb +7f: JNLE/JG Jb +# 0x80 - 0x8f +80: Grp1 Eb,Ib (1A) +81: Grp1 Ev,Iz (1A) +82: Grp1 Eb,Ib (1A),(i64) +83: Grp1 Ev,Ib (1A) +84: TEST Eb,Gb +85: TEST Ev,Gv +86: XCHG Eb,Gb +87: XCHG Ev,Gv +88: MOV Eb,Gb +89: MOV Ev,Gv +8a: MOV Gb,Eb +8b: MOV Gv,Ev +8c: MOV Ev,Sw +8d: LEA Gv,M +8e: MOV Sw,Ew +8f: Grp1A (1A) | POP Ev (d64) +# 0x90 - 0x9f +90: NOP | PAUSE (F3) | XCHG r8,rAX +91: XCHG rCX/r9,rAX +92: XCHG rDX/r10,rAX +93: XCHG rBX/r11,rAX +94: XCHG rSP/r12,rAX +95: XCHG rBP/r13,rAX +96: XCHG rSI/r14,rAX +97: XCHG rDI/r15,rAX +98: CBW/CWDE/CDQE +99: CWD/CDQ/CQO +9a: CALLF Ap (i64) +9b: FWAIT/WAIT +9c: PUSHF/D/Q Fv (d64) +9d: POPF/D/Q Fv (d64) +9e: SAHF +9f: LAHF +# 0xa0 - 0xaf +a0: MOV AL,Ob +a1: MOV rAX,Ov +a2: MOV Ob,AL +a3: MOV Ov,rAX +a4: MOVS/B Yb,Xb +a5: MOVS/W/D/Q Yv,Xv +a6: CMPS/B Xb,Yb +a7: CMPS/W/D Xv,Yv +a8: TEST AL,Ib +a9: TEST rAX,Iz +aa: STOS/B Yb,AL +ab: STOS/W/D/Q Yv,rAX +ac: LODS/B AL,Xb +ad: LODS/W/D/Q rAX,Xv +ae: SCAS/B AL,Yb +# Note: The May 2011 Intel manual shows Xv for the second parameter of the +# next instruction but Yv is correct +af: SCAS/W/D/Q rAX,Yv +# 0xb0 - 0xbf +b0: MOV AL/R8L,Ib +b1: MOV CL/R9L,Ib +b2: MOV DL/R10L,Ib +b3: MOV BL/R11L,Ib +b4: MOV AH/R12L,Ib +b5: MOV CH/R13L,Ib +b6: MOV DH/R14L,Ib +b7: MOV BH/R15L,Ib +b8: MOV rAX/r8,Iv +b9: MOV rCX/r9,Iv +ba: MOV rDX/r10,Iv +bb: MOV rBX/r11,Iv +bc: MOV rSP/r12,Iv +bd: MOV rBP/r13,Iv +be: MOV rSI/r14,Iv +bf: MOV rDI/r15,Iv +# 0xc0 - 0xcf +c0: Grp2 Eb,Ib (1A) +c1: Grp2 Ev,Ib (1A) +c2: RETN Iw (f64) +c3: RETN +c4: LES Gz,Mp (i64) | VEX+2byte (Prefix) +c5: LDS Gz,Mp (i64) | VEX+1byte (Prefix) +c6: Grp11A Eb,Ib (1A) +c7: Grp11B Ev,Iz (1A) +c8: ENTER Iw,Ib +c9: LEAVE (d64) +ca: RETF Iw +cb: RETF +cc: INT3 +cd: INT Ib +ce: INTO (i64) +cf: IRET/D/Q +# 0xd0 - 0xdf +d0: Grp2 Eb,1 (1A) +d1: Grp2 Ev,1 (1A) +d2: Grp2 Eb,CL (1A) +d3: Grp2 Ev,CL (1A) +d4: AAM Ib (i64) +d5: AAD Ib (i64) +d6: +d7: XLAT/XLATB +d8: ESC +d9: ESC +da: ESC +db: ESC +dc: ESC +dd: ESC +de: ESC +df: ESC +# 0xe0 - 0xef +# Note: "forced64" is Intel CPU behavior: they ignore 0x66 prefix +# in 64-bit mode. AMD CPUs accept 0x66 prefix, it causes RIP truncation +# to 16 bits. In 32-bit mode, 0x66 is accepted by both Intel and AMD. +e0: LOOPNE/LOOPNZ Jb (f64) +e1: LOOPE/LOOPZ Jb (f64) +e2: LOOP Jb (f64) +e3: JrCXZ Jb (f64) +e4: IN AL,Ib +e5: IN eAX,Ib +e6: OUT Ib,AL +e7: OUT Ib,eAX +# With 0x66 prefix in 64-bit mode, for AMD CPUs immediate offset +# in "near" jumps and calls is 16-bit. For CALL, +# push of return address is 16-bit wide, RSP is decremented by 2 +# but is not truncated to 16 bits, unlike RIP. +e8: CALL Jz (f64) +e9: JMP-near Jz (f64) +ea: JMP-far Ap (i64) +eb: JMP-short Jb (f64) +ec: IN AL,DX +ed: IN eAX,DX +ee: OUT DX,AL +ef: OUT DX,eAX +# 0xf0 - 0xff +f0: LOCK (Prefix) +f1: +f2: REPNE (Prefix) | XACQUIRE (Prefix) +f3: REP/REPE (Prefix) | XRELEASE (Prefix) +f4: HLT +f5: CMC +f6: Grp3_1 Eb (1A) +f7: Grp3_2 Ev (1A) +f8: CLC +f9: STC +fa: CLI +fb: STI +fc: CLD +fd: STD +fe: Grp4 (1A) +ff: Grp5 (1A) +EndTable + +Table: 2-byte opcode (0x0f) +Referrer: 2-byte escape +AVXcode: 1 +# 0x0f 0x00-0x0f +00: Grp6 (1A) +01: Grp7 (1A) +02: LAR Gv,Ew +03: LSL Gv,Ew +04: +05: SYSCALL (o64) +06: CLTS +07: SYSRET (o64) +08: INVD +09: WBINVD +0a: +0b: UD2 (1B) +0c: +# AMD's prefetch group. Intel supports prefetchw(/1) only. +0d: GrpP +0e: FEMMS +# 3DNow! uses the last imm byte as opcode extension. +0f: 3DNow! Pq,Qq,Ib +# 0x0f 0x10-0x1f +# NOTE: According to Intel SDM opcode map, vmovups and vmovupd has no operands +# but it actually has operands. And also, vmovss and vmovsd only accept 128bit. +# MOVSS/MOVSD has too many forms(3) on SDM. This map just shows a typical form. +# Many AVX instructions lack v1 superscript, according to Intel AVX-Prgramming +# Reference A.1 +10: vmovups Vps,Wps | vmovupd Vpd,Wpd (66) | vmovss Vx,Hx,Wss (F3),(v1) | vmovsd Vx,Hx,Wsd (F2),(v1) +11: vmovups Wps,Vps | vmovupd Wpd,Vpd (66) | vmovss Wss,Hx,Vss (F3),(v1) | vmovsd Wsd,Hx,Vsd (F2),(v1) +12: vmovlps Vq,Hq,Mq (v1) | vmovhlps Vq,Hq,Uq (v1) | vmovlpd Vq,Hq,Mq (66),(v1) | vmovsldup Vx,Wx (F3) | vmovddup Vx,Wx (F2) +13: vmovlps Mq,Vq (v1) | vmovlpd Mq,Vq (66),(v1) +14: vunpcklps Vx,Hx,Wx | vunpcklpd Vx,Hx,Wx (66) +15: vunpckhps Vx,Hx,Wx | vunpckhpd Vx,Hx,Wx (66) +16: vmovhps Vdq,Hq,Mq (v1) | vmovlhps Vdq,Hq,Uq (v1) | vmovhpd Vdq,Hq,Mq (66),(v1) | vmovshdup Vx,Wx (F3) +17: vmovhps Mq,Vq (v1) | vmovhpd Mq,Vq (66),(v1) +18: Grp16 (1A) +19: +# Intel SDM opcode map does not list MPX instructions. For now using Gv for +# bnd registers and Ev for everything else is OK because the instruction +# decoder does not use the information except as an indication that there is +# a ModR/M byte. +1a: BNDCL Gv,Ev (F3) | BNDCU Gv,Ev (F2) | BNDMOV Gv,Ev (66) | BNDLDX Gv,Ev +1b: BNDCN Gv,Ev (F2) | BNDMOV Ev,Gv (66) | BNDMK Gv,Ev (F3) | BNDSTX Ev,Gv +1c: +1d: +1e: +1f: NOP Ev +# 0x0f 0x20-0x2f +20: MOV Rd,Cd +21: MOV Rd,Dd +22: MOV Cd,Rd +23: MOV Dd,Rd +24: +25: +26: +27: +28: vmovaps Vps,Wps | vmovapd Vpd,Wpd (66) +29: vmovaps Wps,Vps | vmovapd Wpd,Vpd (66) +2a: cvtpi2ps Vps,Qpi | cvtpi2pd Vpd,Qpi (66) | vcvtsi2ss Vss,Hss,Ey (F3),(v1) | vcvtsi2sd Vsd,Hsd,Ey (F2),(v1) +2b: vmovntps Mps,Vps | vmovntpd Mpd,Vpd (66) +2c: cvttps2pi Ppi,Wps | cvttpd2pi Ppi,Wpd (66) | vcvttss2si Gy,Wss (F3),(v1) | vcvttsd2si Gy,Wsd (F2),(v1) +2d: cvtps2pi Ppi,Wps | cvtpd2pi Qpi,Wpd (66) | vcvtss2si Gy,Wss (F3),(v1) | vcvtsd2si Gy,Wsd (F2),(v1) +2e: vucomiss Vss,Wss (v1) | vucomisd Vsd,Wsd (66),(v1) +2f: vcomiss Vss,Wss (v1) | vcomisd Vsd,Wsd (66),(v1) +# 0x0f 0x30-0x3f +30: WRMSR +31: RDTSC +32: RDMSR +33: RDPMC +34: SYSENTER +35: SYSEXIT +36: +37: GETSEC +38: escape # 3-byte escape 1 +39: +3a: escape # 3-byte escape 2 +3b: +3c: +3d: +3e: +3f: +# 0x0f 0x40-0x4f +40: CMOVO Gv,Ev +41: CMOVNO Gv,Ev +42: CMOVB/C/NAE Gv,Ev +43: CMOVAE/NB/NC Gv,Ev +44: CMOVE/Z Gv,Ev +45: CMOVNE/NZ Gv,Ev +46: CMOVBE/NA Gv,Ev +47: CMOVA/NBE Gv,Ev +48: CMOVS Gv,Ev +49: CMOVNS Gv,Ev +4a: CMOVP/PE Gv,Ev +4b: CMOVNP/PO Gv,Ev +4c: CMOVL/NGE Gv,Ev +4d: CMOVNL/GE Gv,Ev +4e: CMOVLE/NG Gv,Ev +4f: CMOVNLE/G Gv,Ev +# 0x0f 0x50-0x5f +50: vmovmskps Gy,Ups | vmovmskpd Gy,Upd (66) +51: vsqrtps Vps,Wps | vsqrtpd Vpd,Wpd (66) | vsqrtss Vss,Hss,Wss (F3),(v1) | vsqrtsd Vsd,Hsd,Wsd (F2),(v1) +52: vrsqrtps Vps,Wps | vrsqrtss Vss,Hss,Wss (F3),(v1) +53: vrcpps Vps,Wps | vrcpss Vss,Hss,Wss (F3),(v1) +54: vandps Vps,Hps,Wps | vandpd Vpd,Hpd,Wpd (66) +55: vandnps Vps,Hps,Wps | vandnpd Vpd,Hpd,Wpd (66) +56: vorps Vps,Hps,Wps | vorpd Vpd,Hpd,Wpd (66) +57: vxorps Vps,Hps,Wps | vxorpd Vpd,Hpd,Wpd (66) +58: vaddps Vps,Hps,Wps | vaddpd Vpd,Hpd,Wpd (66) | vaddss Vss,Hss,Wss (F3),(v1) | vaddsd Vsd,Hsd,Wsd (F2),(v1) +59: vmulps Vps,Hps,Wps | vmulpd Vpd,Hpd,Wpd (66) | vmulss Vss,Hss,Wss (F3),(v1) | vmulsd Vsd,Hsd,Wsd (F2),(v1) +5a: vcvtps2pd Vpd,Wps | vcvtpd2ps Vps,Wpd (66) | vcvtss2sd Vsd,Hx,Wss (F3),(v1) | vcvtsd2ss Vss,Hx,Wsd (F2),(v1) +5b: vcvtdq2ps Vps,Wdq | vcvtps2dq Vdq,Wps (66) | vcvttps2dq Vdq,Wps (F3) +5c: vsubps Vps,Hps,Wps | vsubpd Vpd,Hpd,Wpd (66) | vsubss Vss,Hss,Wss (F3),(v1) | vsubsd Vsd,Hsd,Wsd (F2),(v1) +5d: vminps Vps,Hps,Wps | vminpd Vpd,Hpd,Wpd (66) | vminss Vss,Hss,Wss (F3),(v1) | vminsd Vsd,Hsd,Wsd (F2),(v1) +5e: vdivps Vps,Hps,Wps | vdivpd Vpd,Hpd,Wpd (66) | vdivss Vss,Hss,Wss (F3),(v1) | vdivsd Vsd,Hsd,Wsd (F2),(v1) +5f: vmaxps Vps,Hps,Wps | vmaxpd Vpd,Hpd,Wpd (66) | vmaxss Vss,Hss,Wss (F3),(v1) | vmaxsd Vsd,Hsd,Wsd (F2),(v1) +# 0x0f 0x60-0x6f +60: punpcklbw Pq,Qd | vpunpcklbw Vx,Hx,Wx (66),(v1) +61: punpcklwd Pq,Qd | vpunpcklwd Vx,Hx,Wx (66),(v1) +62: punpckldq Pq,Qd | vpunpckldq Vx,Hx,Wx (66),(v1) +63: packsswb Pq,Qq | vpacksswb Vx,Hx,Wx (66),(v1) +64: pcmpgtb Pq,Qq | vpcmpgtb Vx,Hx,Wx (66),(v1) +65: pcmpgtw Pq,Qq | vpcmpgtw Vx,Hx,Wx (66),(v1) +66: pcmpgtd Pq,Qq | vpcmpgtd Vx,Hx,Wx (66),(v1) +67: packuswb Pq,Qq | vpackuswb Vx,Hx,Wx (66),(v1) +68: punpckhbw Pq,Qd | vpunpckhbw Vx,Hx,Wx (66),(v1) +69: punpckhwd Pq,Qd | vpunpckhwd Vx,Hx,Wx (66),(v1) +6a: punpckhdq Pq,Qd | vpunpckhdq Vx,Hx,Wx (66),(v1) +6b: packssdw Pq,Qd | vpackssdw Vx,Hx,Wx (66),(v1) +6c: vpunpcklqdq Vx,Hx,Wx (66),(v1) +6d: vpunpckhqdq Vx,Hx,Wx (66),(v1) +6e: movd/q Pd,Ey | vmovd/q Vy,Ey (66),(v1) +6f: movq Pq,Qq | vmovdqa Vx,Wx (66) | vmovdqu Vx,Wx (F3) +# 0x0f 0x70-0x7f +70: pshufw Pq,Qq,Ib | vpshufd Vx,Wx,Ib (66),(v1) | vpshufhw Vx,Wx,Ib (F3),(v1) | vpshuflw Vx,Wx,Ib (F2),(v1) +71: Grp12 (1A) +72: Grp13 (1A) +73: Grp14 (1A) +74: pcmpeqb Pq,Qq | vpcmpeqb Vx,Hx,Wx (66),(v1) +75: pcmpeqw Pq,Qq | vpcmpeqw Vx,Hx,Wx (66),(v1) +76: pcmpeqd Pq,Qq | vpcmpeqd Vx,Hx,Wx (66),(v1) +# Note: Remove (v), because vzeroall and vzeroupper becomes emms without VEX. +77: emms | vzeroupper | vzeroall +78: VMREAD Ey,Gy +79: VMWRITE Gy,Ey +7a: +7b: +7c: vhaddpd Vpd,Hpd,Wpd (66) | vhaddps Vps,Hps,Wps (F2) +7d: vhsubpd Vpd,Hpd,Wpd (66) | vhsubps Vps,Hps,Wps (F2) +7e: movd/q Ey,Pd | vmovd/q Ey,Vy (66),(v1) | vmovq Vq,Wq (F3),(v1) +7f: movq Qq,Pq | vmovdqa Wx,Vx (66) | vmovdqu Wx,Vx (F3) +# 0x0f 0x80-0x8f +# Note: "forced64" is Intel CPU behavior (see comment about CALL insn). +80: JO Jz (f64) +81: JNO Jz (f64) +82: JB/JC/JNAE Jz (f64) +83: JAE/JNB/JNC Jz (f64) +84: JE/JZ Jz (f64) +85: JNE/JNZ Jz (f64) +86: JBE/JNA Jz (f64) +87: JA/JNBE Jz (f64) +88: JS Jz (f64) +89: JNS Jz (f64) +8a: JP/JPE Jz (f64) +8b: JNP/JPO Jz (f64) +8c: JL/JNGE Jz (f64) +8d: JNL/JGE Jz (f64) +8e: JLE/JNG Jz (f64) +8f: JNLE/JG Jz (f64) +# 0x0f 0x90-0x9f +90: SETO Eb +91: SETNO Eb +92: SETB/C/NAE Eb +93: SETAE/NB/NC Eb +94: SETE/Z Eb +95: SETNE/NZ Eb +96: SETBE/NA Eb +97: SETA/NBE Eb +98: SETS Eb +99: SETNS Eb +9a: SETP/PE Eb +9b: SETNP/PO Eb +9c: SETL/NGE Eb +9d: SETNL/GE Eb +9e: SETLE/NG Eb +9f: SETNLE/G Eb +# 0x0f 0xa0-0xaf +a0: PUSH FS (d64) +a1: POP FS (d64) +a2: CPUID +a3: BT Ev,Gv +a4: SHLD Ev,Gv,Ib +a5: SHLD Ev,Gv,CL +a6: GrpPDLK +a7: GrpRNG +a8: PUSH GS (d64) +a9: POP GS (d64) +aa: RSM +ab: BTS Ev,Gv +ac: SHRD Ev,Gv,Ib +ad: SHRD Ev,Gv,CL +ae: Grp15 (1A),(1C) +af: IMUL Gv,Ev +# 0x0f 0xb0-0xbf +b0: CMPXCHG Eb,Gb +b1: CMPXCHG Ev,Gv +b2: LSS Gv,Mp +b3: BTR Ev,Gv +b4: LFS Gv,Mp +b5: LGS Gv,Mp +b6: MOVZX Gv,Eb +b7: MOVZX Gv,Ew +b8: JMPE (!F3) | POPCNT Gv,Ev (F3) +b9: Grp10 (1A) +ba: Grp8 Ev,Ib (1A) +bb: BTC Ev,Gv +bc: BSF Gv,Ev (!F3) | TZCNT Gv,Ev (F3) +bd: BSR Gv,Ev (!F3) | LZCNT Gv,Ev (F3) +be: MOVSX Gv,Eb +bf: MOVSX Gv,Ew +# 0x0f 0xc0-0xcf +c0: XADD Eb,Gb +c1: XADD Ev,Gv +c2: vcmpps Vps,Hps,Wps,Ib | vcmppd Vpd,Hpd,Wpd,Ib (66) | vcmpss Vss,Hss,Wss,Ib (F3),(v1) | vcmpsd Vsd,Hsd,Wsd,Ib (F2),(v1) +c3: movnti My,Gy +c4: pinsrw Pq,Ry/Mw,Ib | vpinsrw Vdq,Hdq,Ry/Mw,Ib (66),(v1) +c5: pextrw Gd,Nq,Ib | vpextrw Gd,Udq,Ib (66),(v1) +c6: vshufps Vps,Hps,Wps,Ib | vshufpd Vpd,Hpd,Wpd,Ib (66) +c7: Grp9 (1A) +c8: BSWAP RAX/EAX/R8/R8D +c9: BSWAP RCX/ECX/R9/R9D +ca: BSWAP RDX/EDX/R10/R10D +cb: BSWAP RBX/EBX/R11/R11D +cc: BSWAP RSP/ESP/R12/R12D +cd: BSWAP RBP/EBP/R13/R13D +ce: BSWAP RSI/ESI/R14/R14D +cf: BSWAP RDI/EDI/R15/R15D +# 0x0f 0xd0-0xdf +d0: vaddsubpd Vpd,Hpd,Wpd (66) | vaddsubps Vps,Hps,Wps (F2) +d1: psrlw Pq,Qq | vpsrlw Vx,Hx,Wx (66),(v1) +d2: psrld Pq,Qq | vpsrld Vx,Hx,Wx (66),(v1) +d3: psrlq Pq,Qq | vpsrlq Vx,Hx,Wx (66),(v1) +d4: paddq Pq,Qq | vpaddq Vx,Hx,Wx (66),(v1) +d5: pmullw Pq,Qq | vpmullw Vx,Hx,Wx (66),(v1) +d6: vmovq Wq,Vq (66),(v1) | movq2dq Vdq,Nq (F3) | movdq2q Pq,Uq (F2) +d7: pmovmskb Gd,Nq | vpmovmskb Gd,Ux (66),(v1) +d8: psubusb Pq,Qq | vpsubusb Vx,Hx,Wx (66),(v1) +d9: psubusw Pq,Qq | vpsubusw Vx,Hx,Wx (66),(v1) +da: pminub Pq,Qq | vpminub Vx,Hx,Wx (66),(v1) +db: pand Pq,Qq | vpand Vx,Hx,Wx (66),(v1) +dc: paddusb Pq,Qq | vpaddusb Vx,Hx,Wx (66),(v1) +dd: paddusw Pq,Qq | vpaddusw Vx,Hx,Wx (66),(v1) +de: pmaxub Pq,Qq | vpmaxub Vx,Hx,Wx (66),(v1) +df: pandn Pq,Qq | vpandn Vx,Hx,Wx (66),(v1) +# 0x0f 0xe0-0xef +e0: pavgb Pq,Qq | vpavgb Vx,Hx,Wx (66),(v1) +e1: psraw Pq,Qq | vpsraw Vx,Hx,Wx (66),(v1) +e2: psrad Pq,Qq | vpsrad Vx,Hx,Wx (66),(v1) +e3: pavgw Pq,Qq | vpavgw Vx,Hx,Wx (66),(v1) +e4: pmulhuw Pq,Qq | vpmulhuw Vx,Hx,Wx (66),(v1) +e5: pmulhw Pq,Qq | vpmulhw Vx,Hx,Wx (66),(v1) +e6: vcvttpd2dq Vx,Wpd (66) | vcvtdq2pd Vx,Wdq (F3) | vcvtpd2dq Vx,Wpd (F2) +e7: movntq Mq,Pq | vmovntdq Mx,Vx (66) +e8: psubsb Pq,Qq | vpsubsb Vx,Hx,Wx (66),(v1) +e9: psubsw Pq,Qq | vpsubsw Vx,Hx,Wx (66),(v1) +ea: pminsw Pq,Qq | vpminsw Vx,Hx,Wx (66),(v1) +eb: por Pq,Qq | vpor Vx,Hx,Wx (66),(v1) +ec: paddsb Pq,Qq | vpaddsb Vx,Hx,Wx (66),(v1) +ed: paddsw Pq,Qq | vpaddsw Vx,Hx,Wx (66),(v1) +ee: pmaxsw Pq,Qq | vpmaxsw Vx,Hx,Wx (66),(v1) +ef: pxor Pq,Qq | vpxor Vx,Hx,Wx (66),(v1) +# 0x0f 0xf0-0xff +f0: vlddqu Vx,Mx (F2) +f1: psllw Pq,Qq | vpsllw Vx,Hx,Wx (66),(v1) +f2: pslld Pq,Qq | vpslld Vx,Hx,Wx (66),(v1) +f3: psllq Pq,Qq | vpsllq Vx,Hx,Wx (66),(v1) +f4: pmuludq Pq,Qq | vpmuludq Vx,Hx,Wx (66),(v1) +f5: pmaddwd Pq,Qq | vpmaddwd Vx,Hx,Wx (66),(v1) +f6: psadbw Pq,Qq | vpsadbw Vx,Hx,Wx (66),(v1) +f7: maskmovq Pq,Nq | vmaskmovdqu Vx,Ux (66),(v1) +f8: psubb Pq,Qq | vpsubb Vx,Hx,Wx (66),(v1) +f9: psubw Pq,Qq | vpsubw Vx,Hx,Wx (66),(v1) +fa: psubd Pq,Qq | vpsubd Vx,Hx,Wx (66),(v1) +fb: psubq Pq,Qq | vpsubq Vx,Hx,Wx (66),(v1) +fc: paddb Pq,Qq | vpaddb Vx,Hx,Wx (66),(v1) +fd: paddw Pq,Qq | vpaddw Vx,Hx,Wx (66),(v1) +fe: paddd Pq,Qq | vpaddd Vx,Hx,Wx (66),(v1) +ff: +EndTable + +Table: 3-byte opcode 1 (0x0f 0x38) +Referrer: 3-byte escape 1 +AVXcode: 2 +# 0x0f 0x38 0x00-0x0f +00: pshufb Pq,Qq | vpshufb Vx,Hx,Wx (66),(v1) +01: phaddw Pq,Qq | vphaddw Vx,Hx,Wx (66),(v1) +02: phaddd Pq,Qq | vphaddd Vx,Hx,Wx (66),(v1) +03: phaddsw Pq,Qq | vphaddsw Vx,Hx,Wx (66),(v1) +04: pmaddubsw Pq,Qq | vpmaddubsw Vx,Hx,Wx (66),(v1) +05: phsubw Pq,Qq | vphsubw Vx,Hx,Wx (66),(v1) +06: phsubd Pq,Qq | vphsubd Vx,Hx,Wx (66),(v1) +07: phsubsw Pq,Qq | vphsubsw Vx,Hx,Wx (66),(v1) +08: psignb Pq,Qq | vpsignb Vx,Hx,Wx (66),(v1) +09: psignw Pq,Qq | vpsignw Vx,Hx,Wx (66),(v1) +0a: psignd Pq,Qq | vpsignd Vx,Hx,Wx (66),(v1) +0b: pmulhrsw Pq,Qq | vpmulhrsw Vx,Hx,Wx (66),(v1) +0c: vpermilps Vx,Hx,Wx (66),(v) +0d: vpermilpd Vx,Hx,Wx (66),(v) +0e: vtestps Vx,Wx (66),(v) +0f: vtestpd Vx,Wx (66),(v) +# 0x0f 0x38 0x10-0x1f +10: pblendvb Vdq,Wdq (66) +11: +12: +13: vcvtph2ps Vx,Wx,Ib (66),(v) +14: blendvps Vdq,Wdq (66) +15: blendvpd Vdq,Wdq (66) +16: vpermps Vqq,Hqq,Wqq (66),(v) +17: vptest Vx,Wx (66) +18: vbroadcastss Vx,Wd (66),(v) +19: vbroadcastsd Vqq,Wq (66),(v) +1a: vbroadcastf128 Vqq,Mdq (66),(v) +1b: +1c: pabsb Pq,Qq | vpabsb Vx,Wx (66),(v1) +1d: pabsw Pq,Qq | vpabsw Vx,Wx (66),(v1) +1e: pabsd Pq,Qq | vpabsd Vx,Wx (66),(v1) +1f: +# 0x0f 0x38 0x20-0x2f +20: vpmovsxbw Vx,Ux/Mq (66),(v1) +21: vpmovsxbd Vx,Ux/Md (66),(v1) +22: vpmovsxbq Vx,Ux/Mw (66),(v1) +23: vpmovsxwd Vx,Ux/Mq (66),(v1) +24: vpmovsxwq Vx,Ux/Md (66),(v1) +25: vpmovsxdq Vx,Ux/Mq (66),(v1) +26: +27: +28: vpmuldq Vx,Hx,Wx (66),(v1) +29: vpcmpeqq Vx,Hx,Wx (66),(v1) +2a: vmovntdqa Vx,Mx (66),(v1) +2b: vpackusdw Vx,Hx,Wx (66),(v1) +2c: vmaskmovps Vx,Hx,Mx (66),(v) +2d: vmaskmovpd Vx,Hx,Mx (66),(v) +2e: vmaskmovps Mx,Hx,Vx (66),(v) +2f: vmaskmovpd Mx,Hx,Vx (66),(v) +# 0x0f 0x38 0x30-0x3f +30: vpmovzxbw Vx,Ux/Mq (66),(v1) +31: vpmovzxbd Vx,Ux/Md (66),(v1) +32: vpmovzxbq Vx,Ux/Mw (66),(v1) +33: vpmovzxwd Vx,Ux/Mq (66),(v1) +34: vpmovzxwq Vx,Ux/Md (66),(v1) +35: vpmovzxdq Vx,Ux/Mq (66),(v1) +36: vpermd Vqq,Hqq,Wqq (66),(v) +37: vpcmpgtq Vx,Hx,Wx (66),(v1) +38: vpminsb Vx,Hx,Wx (66),(v1) +39: vpminsd Vx,Hx,Wx (66),(v1) +3a: vpminuw Vx,Hx,Wx (66),(v1) +3b: vpminud Vx,Hx,Wx (66),(v1) +3c: vpmaxsb Vx,Hx,Wx (66),(v1) +3d: vpmaxsd Vx,Hx,Wx (66),(v1) +3e: vpmaxuw Vx,Hx,Wx (66),(v1) +3f: vpmaxud Vx,Hx,Wx (66),(v1) +# 0x0f 0x38 0x40-0x8f +40: vpmulld Vx,Hx,Wx (66),(v1) +41: vphminposuw Vdq,Wdq (66),(v1) +42: +43: +44: +45: vpsrlvd/q Vx,Hx,Wx (66),(v) +46: vpsravd Vx,Hx,Wx (66),(v) +47: vpsllvd/q Vx,Hx,Wx (66),(v) +# Skip 0x48-0x57 +58: vpbroadcastd Vx,Wx (66),(v) +59: vpbroadcastq Vx,Wx (66),(v) +5a: vbroadcasti128 Vqq,Mdq (66),(v) +# Skip 0x5b-0x77 +78: vpbroadcastb Vx,Wx (66),(v) +79: vpbroadcastw Vx,Wx (66),(v) +# Skip 0x7a-0x7f +80: INVEPT Gy,Mdq (66) +81: INVPID Gy,Mdq (66) +82: INVPCID Gy,Mdq (66) +8c: vpmaskmovd/q Vx,Hx,Mx (66),(v) +8e: vpmaskmovd/q Mx,Vx,Hx (66),(v) +# 0x0f 0x38 0x90-0xbf (FMA) +90: vgatherdd/q Vx,Hx,Wx (66),(v) +91: vgatherqd/q Vx,Hx,Wx (66),(v) +92: vgatherdps/d Vx,Hx,Wx (66),(v) +93: vgatherqps/d Vx,Hx,Wx (66),(v) +94: +95: +96: vfmaddsub132ps/d Vx,Hx,Wx (66),(v) +97: vfmsubadd132ps/d Vx,Hx,Wx (66),(v) +98: vfmadd132ps/d Vx,Hx,Wx (66),(v) +99: vfmadd132ss/d Vx,Hx,Wx (66),(v),(v1) +9a: vfmsub132ps/d Vx,Hx,Wx (66),(v) +9b: vfmsub132ss/d Vx,Hx,Wx (66),(v),(v1) +9c: vfnmadd132ps/d Vx,Hx,Wx (66),(v) +9d: vfnmadd132ss/d Vx,Hx,Wx (66),(v),(v1) +9e: vfnmsub132ps/d Vx,Hx,Wx (66),(v) +9f: vfnmsub132ss/d Vx,Hx,Wx (66),(v),(v1) +a6: vfmaddsub213ps/d Vx,Hx,Wx (66),(v) +a7: vfmsubadd213ps/d Vx,Hx,Wx (66),(v) +a8: vfmadd213ps/d Vx,Hx,Wx (66),(v) +a9: vfmadd213ss/d Vx,Hx,Wx (66),(v),(v1) +aa: vfmsub213ps/d Vx,Hx,Wx (66),(v) +ab: vfmsub213ss/d Vx,Hx,Wx (66),(v),(v1) +ac: vfnmadd213ps/d Vx,Hx,Wx (66),(v) +ad: vfnmadd213ss/d Vx,Hx,Wx (66),(v),(v1) +ae: vfnmsub213ps/d Vx,Hx,Wx (66),(v) +af: vfnmsub213ss/d Vx,Hx,Wx (66),(v),(v1) +b6: vfmaddsub231ps/d Vx,Hx,Wx (66),(v) +b7: vfmsubadd231ps/d Vx,Hx,Wx (66),(v) +b8: vfmadd231ps/d Vx,Hx,Wx (66),(v) +b9: vfmadd231ss/d Vx,Hx,Wx (66),(v),(v1) +ba: vfmsub231ps/d Vx,Hx,Wx (66),(v) +bb: vfmsub231ss/d Vx,Hx,Wx (66),(v),(v1) +bc: vfnmadd231ps/d Vx,Hx,Wx (66),(v) +bd: vfnmadd231ss/d Vx,Hx,Wx (66),(v),(v1) +be: vfnmsub231ps/d Vx,Hx,Wx (66),(v) +bf: vfnmsub231ss/d Vx,Hx,Wx (66),(v),(v1) +# 0x0f 0x38 0xc0-0xff +c8: sha1nexte Vdq,Wdq +c9: sha1msg1 Vdq,Wdq +ca: sha1msg2 Vdq,Wdq +cb: sha256rnds2 Vdq,Wdq +cc: sha256msg1 Vdq,Wdq +cd: sha256msg2 Vdq,Wdq +db: VAESIMC Vdq,Wdq (66),(v1) +dc: VAESENC Vdq,Hdq,Wdq (66),(v1) +dd: VAESENCLAST Vdq,Hdq,Wdq (66),(v1) +de: VAESDEC Vdq,Hdq,Wdq (66),(v1) +df: VAESDECLAST Vdq,Hdq,Wdq (66),(v1) +f0: MOVBE Gy,My | MOVBE Gw,Mw (66) | CRC32 Gd,Eb (F2) | CRC32 Gd,Eb (66&F2) +f1: MOVBE My,Gy | MOVBE Mw,Gw (66) | CRC32 Gd,Ey (F2) | CRC32 Gd,Ew (66&F2) +f2: ANDN Gy,By,Ey (v) +f3: Grp17 (1A) +f5: BZHI Gy,Ey,By (v) | PEXT Gy,By,Ey (F3),(v) | PDEP Gy,By,Ey (F2),(v) +f6: ADCX Gy,Ey (66) | ADOX Gy,Ey (F3) | MULX By,Gy,rDX,Ey (F2),(v) +f7: BEXTR Gy,Ey,By (v) | SHLX Gy,Ey,By (66),(v) | SARX Gy,Ey,By (F3),(v) | SHRX Gy,Ey,By (F2),(v) +EndTable + +Table: 3-byte opcode 2 (0x0f 0x3a) +Referrer: 3-byte escape 2 +AVXcode: 3 +# 0x0f 0x3a 0x00-0xff +00: vpermq Vqq,Wqq,Ib (66),(v) +01: vpermpd Vqq,Wqq,Ib (66),(v) +02: vpblendd Vx,Hx,Wx,Ib (66),(v) +03: +04: vpermilps Vx,Wx,Ib (66),(v) +05: vpermilpd Vx,Wx,Ib (66),(v) +06: vperm2f128 Vqq,Hqq,Wqq,Ib (66),(v) +07: +08: vroundps Vx,Wx,Ib (66) +09: vroundpd Vx,Wx,Ib (66) +0a: vroundss Vss,Wss,Ib (66),(v1) +0b: vroundsd Vsd,Wsd,Ib (66),(v1) +0c: vblendps Vx,Hx,Wx,Ib (66) +0d: vblendpd Vx,Hx,Wx,Ib (66) +0e: vpblendw Vx,Hx,Wx,Ib (66),(v1) +0f: palignr Pq,Qq,Ib | vpalignr Vx,Hx,Wx,Ib (66),(v1) +14: vpextrb Rd/Mb,Vdq,Ib (66),(v1) +15: vpextrw Rd/Mw,Vdq,Ib (66),(v1) +16: vpextrd/q Ey,Vdq,Ib (66),(v1) +17: vextractps Ed,Vdq,Ib (66),(v1) +18: vinsertf128 Vqq,Hqq,Wqq,Ib (66),(v) +19: vextractf128 Wdq,Vqq,Ib (66),(v) +1d: vcvtps2ph Wx,Vx,Ib (66),(v) +20: vpinsrb Vdq,Hdq,Ry/Mb,Ib (66),(v1) +21: vinsertps Vdq,Hdq,Udq/Md,Ib (66),(v1) +22: vpinsrd/q Vdq,Hdq,Ey,Ib (66),(v1) +38: vinserti128 Vqq,Hqq,Wqq,Ib (66),(v) +39: vextracti128 Wdq,Vqq,Ib (66),(v) +40: vdpps Vx,Hx,Wx,Ib (66) +41: vdppd Vdq,Hdq,Wdq,Ib (66),(v1) +42: vmpsadbw Vx,Hx,Wx,Ib (66),(v1) +44: vpclmulqdq Vdq,Hdq,Wdq,Ib (66),(v1) +46: vperm2i128 Vqq,Hqq,Wqq,Ib (66),(v) +4a: vblendvps Vx,Hx,Wx,Lx (66),(v) +4b: vblendvpd Vx,Hx,Wx,Lx (66),(v) +4c: vpblendvb Vx,Hx,Wx,Lx (66),(v1) +60: vpcmpestrm Vdq,Wdq,Ib (66),(v1) +61: vpcmpestri Vdq,Wdq,Ib (66),(v1) +62: vpcmpistrm Vdq,Wdq,Ib (66),(v1) +63: vpcmpistri Vdq,Wdq,Ib (66),(v1) +cc: sha1rnds4 Vdq,Wdq,Ib +df: VAESKEYGEN Vdq,Wdq,Ib (66),(v1) +f0: RORX Gy,Ey,Ib (F2),(v) +EndTable + +GrpTable: Grp1 +0: ADD +1: OR +2: ADC +3: SBB +4: AND +5: SUB +6: XOR +7: CMP +EndTable + +GrpTable: Grp1A +0: POP +EndTable + +GrpTable: Grp2 +0: ROL +1: ROR +2: RCL +3: RCR +4: SHL/SAL +5: SHR +6: +7: SAR +EndTable + +GrpTable: Grp3_1 +0: TEST Eb,Ib +1: +2: NOT Eb +3: NEG Eb +4: MUL AL,Eb +5: IMUL AL,Eb +6: DIV AL,Eb +7: IDIV AL,Eb +EndTable + +GrpTable: Grp3_2 +0: TEST Ev,Iz +1: +2: NOT Ev +3: NEG Ev +4: MUL rAX,Ev +5: IMUL rAX,Ev +6: DIV rAX,Ev +7: IDIV rAX,Ev +EndTable + +GrpTable: Grp4 +0: INC Eb +1: DEC Eb +EndTable + +GrpTable: Grp5 +0: INC Ev +1: DEC Ev +# Note: "forced64" is Intel CPU behavior (see comment about CALL insn). +2: CALLN Ev (f64) +3: CALLF Ep +4: JMPN Ev (f64) +5: JMPF Mp +6: PUSH Ev (d64) +7: +EndTable + +GrpTable: Grp6 +0: SLDT Rv/Mw +1: STR Rv/Mw +2: LLDT Ew +3: LTR Ew +4: VERR Ew +5: VERW Ew +EndTable + +GrpTable: Grp7 +0: SGDT Ms | VMCALL (001),(11B) | VMLAUNCH (010),(11B) | VMRESUME (011),(11B) | VMXOFF (100),(11B) +1: SIDT Ms | MONITOR (000),(11B) | MWAIT (001),(11B) | CLAC (010),(11B) | STAC (011),(11B) +2: LGDT Ms | XGETBV (000),(11B) | XSETBV (001),(11B) | VMFUNC (100),(11B) | XEND (101)(11B) | XTEST (110)(11B) +3: LIDT Ms +4: SMSW Mw/Rv +5: rdpkru (110),(11B) | wrpkru (111),(11B) +6: LMSW Ew +7: INVLPG Mb | SWAPGS (o64),(000),(11B) | RDTSCP (001),(11B) +EndTable + +GrpTable: Grp8 +4: BT +5: BTS +6: BTR +7: BTC +EndTable + +GrpTable: Grp9 +1: CMPXCHG8B/16B Mq/Mdq +3: xrstors +4: xsavec +5: xsaves +6: VMPTRLD Mq | VMCLEAR Mq (66) | VMXON Mq (F3) | RDRAND Rv (11B) +7: VMPTRST Mq | VMPTRST Mq (F3) | RDSEED Rv (11B) +EndTable + +GrpTable: Grp10 +EndTable + +# Grp11A and Grp11B are expressed as Grp11 in Intel SDM +GrpTable: Grp11A +0: MOV Eb,Ib +7: XABORT Ib (000),(11B) +EndTable + +GrpTable: Grp11B +0: MOV Eb,Iz +7: XBEGIN Jz (000),(11B) +EndTable + +GrpTable: Grp12 +2: psrlw Nq,Ib (11B) | vpsrlw Hx,Ux,Ib (66),(11B),(v1) +4: psraw Nq,Ib (11B) | vpsraw Hx,Ux,Ib (66),(11B),(v1) +6: psllw Nq,Ib (11B) | vpsllw Hx,Ux,Ib (66),(11B),(v1) +EndTable + +GrpTable: Grp13 +2: psrld Nq,Ib (11B) | vpsrld Hx,Ux,Ib (66),(11B),(v1) +4: psrad Nq,Ib (11B) | vpsrad Hx,Ux,Ib (66),(11B),(v1) +6: pslld Nq,Ib (11B) | vpslld Hx,Ux,Ib (66),(11B),(v1) +EndTable + +GrpTable: Grp14 +2: psrlq Nq,Ib (11B) | vpsrlq Hx,Ux,Ib (66),(11B),(v1) +3: vpsrldq Hx,Ux,Ib (66),(11B),(v1) +6: psllq Nq,Ib (11B) | vpsllq Hx,Ux,Ib (66),(11B),(v1) +7: vpslldq Hx,Ux,Ib (66),(11B),(v1) +EndTable + +GrpTable: Grp15 +0: fxsave | RDFSBASE Ry (F3),(11B) +1: fxstor | RDGSBASE Ry (F3),(11B) +2: vldmxcsr Md (v1) | WRFSBASE Ry (F3),(11B) +3: vstmxcsr Md (v1) | WRGSBASE Ry (F3),(11B) +4: XSAVE +5: XRSTOR | lfence (11B) +6: XSAVEOPT | clwb (66) | mfence (11B) +7: clflush | clflushopt (66) | sfence (11B) | pcommit (66),(11B) +EndTable + +GrpTable: Grp16 +0: prefetch NTA +1: prefetch T0 +2: prefetch T1 +3: prefetch T2 +EndTable + +GrpTable: Grp17 +1: BLSR By,Ey (v) +2: BLSMSK By,Ey (v) +3: BLSI By,Ey (v) +EndTable + +# AMD's Prefetch Group +GrpTable: GrpP +0: PREFETCH +1: PREFETCHW +EndTable + +GrpTable: GrpPDLK +0: MONTMUL +1: XSHA1 +2: XSHA2 +EndTable + +GrpTable: GrpRNG +0: xstore-rng +1: xcrypt-ecb +2: xcrypt-cbc +4: xcrypt-cfb +5: xcrypt-ofb +EndTable diff --git a/tools/stacktool/elf.c b/tools/stacktool/elf.c new file mode 100644 index 0000000..d547e3f --- /dev/null +++ b/tools/stacktool/elf.c @@ -0,0 +1,403 @@ +/* + * elf.c - ELF access library + * + * Adapted from kpatch (https://github.com/dynup/kpatch): + * Copyright (C) 2013-2015 Josh Poimboeuf <jpoimboe@redhat.com> + * Copyright (C) 2014 Seth Jennings <sjenning@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#include <sys/types.h> +#include <sys/stat.h> +#include <fcntl.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <unistd.h> + +#include "elf.h" +#include "warn.h" + +struct section *find_section_by_name(struct elf *elf, const char *name) +{ + struct section *sec; + + list_for_each_entry(sec, &elf->sections, list) + if (!strcmp(sec->name, name)) + return sec; + + return NULL; +} + +static struct section *find_section_by_index(struct elf *elf, + unsigned int idx) +{ + struct section *sec; + + list_for_each_entry(sec, &elf->sections, list) + if (sec->idx == idx) + return sec; + + return NULL; +} + +static struct symbol *find_symbol_by_index(struct elf *elf, unsigned int idx) +{ + struct section *sec; + struct symbol *sym; + + list_for_each_entry(sec, &elf->sections, list) + list_for_each_entry(sym, &sec->symbols, list) + if (sym->idx == idx) + return sym; + + return NULL; +} + +struct symbol *find_symbol_by_offset(struct section *sec, unsigned long offset) +{ + struct symbol *sym; + + list_for_each_entry(sym, &sec->symbols, list) + if (sym->type != STT_SECTION && + sym->offset == offset) + return sym; + + return NULL; +} + +struct rela *find_rela_by_dest_range(struct section *sec, unsigned long offset, + unsigned int len) +{ + struct rela *rela; + + if (!sec->rela) + return NULL; + + list_for_each_entry(rela, &sec->rela->relas, list) + if (rela->offset >= offset && rela->offset < offset + len) + return rela; + + return NULL; +} + +struct rela *find_rela_by_dest(struct section *sec, unsigned long offset) +{ + return find_rela_by_dest_range(sec, offset, 1); +} + +struct symbol *find_containing_func(struct section *sec, unsigned long offset) +{ + struct symbol *func; + + list_for_each_entry(func, &sec->symbols, list) + if (func->type == STT_FUNC && offset >= func->offset && + offset < func->offset + func->len) + return func; + + return NULL; +} + +static int read_sections(struct elf *elf) +{ + Elf_Scn *s = NULL; + struct section *sec; + size_t shstrndx, sections_nr; + int i; + + if (elf_getshdrnum(elf->elf, §ions_nr)) { + perror("elf_getshdrnum"); + return -1; + } + + if (elf_getshdrstrndx(elf->elf, &shstrndx)) { + perror("elf_getshdrstrndx"); + return -1; + } + + for (i = 0; i < sections_nr; i++) { + sec = malloc(sizeof(*sec)); + if (!sec) { + perror("malloc"); + return -1; + } + memset(sec, 0, sizeof(*sec)); + + INIT_LIST_HEAD(&sec->symbols); + INIT_LIST_HEAD(&sec->relas); + + list_add_tail(&sec->list, &elf->sections); + + s = elf_getscn(elf->elf, i); + if (!s) { + perror("elf_getscn"); + return -1; + } + + sec->idx = elf_ndxscn(s); + + if (!gelf_getshdr(s, &sec->sh)) { + perror("gelf_getshdr"); + return -1; + } + + sec->name = elf_strptr(elf->elf, shstrndx, sec->sh.sh_name); + if (!sec->name) { + perror("elf_strptr"); + return -1; + } + + sec->elf_data = elf_getdata(s, NULL); + if (!sec->elf_data) { + perror("elf_getdata"); + return -1; + } + + if (sec->elf_data->d_off != 0 || + sec->elf_data->d_size != sec->sh.sh_size) { + WARN("unexpected data attributes for %s", sec->name); + return -1; + } + + sec->data = (unsigned long)sec->elf_data->d_buf; + sec->len = sec->elf_data->d_size; + } + + /* sanity check, one more call to elf_nextscn() should return NULL */ + if (elf_nextscn(elf->elf, s)) { + WARN("section entry mismatch"); + return -1; + } + + return 0; +} + +static int read_symbols(struct elf *elf) +{ + struct section *symtab; + struct symbol *sym; + struct list_head *entry, *tmp; + int symbols_nr, i; + + symtab = find_section_by_name(elf, ".symtab"); + if (!symtab) { + WARN("missing symbol table"); + return -1; + } + + symbols_nr = symtab->sh.sh_size / symtab->sh.sh_entsize; + + for (i = 0; i < symbols_nr; i++) { + sym = malloc(sizeof(*sym)); + if (!sym) { + perror("malloc"); + return -1; + } + memset(sym, 0, sizeof(*sym)); + + sym->idx = i; + + if (!gelf_getsym(symtab->elf_data, i, &sym->sym)) { + perror("gelf_getsym"); + goto err; + } + + sym->name = elf_strptr(elf->elf, symtab->sh.sh_link, + sym->sym.st_name); + if (!sym->name) { + perror("elf_strptr"); + goto err; + } + + sym->type = GELF_ST_TYPE(sym->sym.st_info); + sym->bind = GELF_ST_BIND(sym->sym.st_info); + + if (sym->sym.st_shndx > SHN_UNDEF && + sym->sym.st_shndx < SHN_LORESERVE) { + sym->sec = find_section_by_index(elf, + sym->sym.st_shndx); + if (!sym->sec) { + WARN("couldn't find section for symbol %s", + sym->name); + goto err; + } + if (sym->type == STT_SECTION) { + sym->name = sym->sec->name; + sym->sec->sym = sym; + } + } else + sym->sec = find_section_by_index(elf, 0); + + sym->offset = sym->sym.st_value; + sym->len = sym->sym.st_size; + + /* sorted insert into a per-section list */ + entry = &sym->sec->symbols; + list_for_each_prev(tmp, &sym->sec->symbols) { + struct symbol *s; + + s = list_entry(tmp, struct symbol, list); + + if (sym->offset > s->offset) { + entry = tmp; + break; + } + + if (sym->offset == s->offset && sym->len >= s->len) { + entry = tmp; + break; + } + } + list_add(&sym->list, entry); + } + + return 0; + +err: + free(sym); + return -1; +} + +static int read_relas(struct elf *elf) +{ + struct section *sec; + struct rela *rela; + int i; + unsigned int symndx; + + list_for_each_entry(sec, &elf->sections, list) { + if (sec->sh.sh_type != SHT_RELA) + continue; + + sec->base = find_section_by_name(elf, sec->name + 5); + if (!sec->base) { + WARN("can't find base section for rela section %s", + sec->name); + return -1; + } + + sec->base->rela = sec; + + for (i = 0; i < sec->sh.sh_size / sec->sh.sh_entsize; i++) { + rela = malloc(sizeof(*rela)); + if (!rela) { + perror("malloc"); + return -1; + } + memset(rela, 0, sizeof(*rela)); + + list_add_tail(&rela->list, &sec->relas); + + if (!gelf_getrela(sec->elf_data, i, &rela->rela)) { + perror("gelf_getrela"); + return -1; + } + + rela->type = GELF_R_TYPE(rela->rela.r_info); + rela->addend = rela->rela.r_addend; + rela->offset = rela->rela.r_offset; + symndx = GELF_R_SYM(rela->rela.r_info); + rela->sym = find_symbol_by_index(elf, symndx); + if (!rela->sym) { + WARN("can't find rela entry symbol %d for %s", + symndx, sec->name); + return -1; + } + } + } + + return 0; +} + +struct elf *elf_open(const char *name) +{ + struct elf *elf; + + elf_version(EV_CURRENT); + + elf = malloc(sizeof(*elf)); + if (!elf) { + perror("malloc"); + return NULL; + } + memset(elf, 0, sizeof(*elf)); + + INIT_LIST_HEAD(&elf->sections); + + elf->name = strdup(name); + if (!elf->name) { + perror("strdup"); + goto err; + } + + elf->fd = open(name, O_RDONLY); + if (elf->fd == -1) { + perror("open"); + goto err; + } + + elf->elf = elf_begin(elf->fd, ELF_C_READ_MMAP, NULL); + if (!elf->elf) { + perror("elf_begin"); + goto err; + } + + if (!gelf_getehdr(elf->elf, &elf->ehdr)) { + perror("gelf_getehdr"); + goto err; + } + + if (read_sections(elf)) + goto err; + + if (read_symbols(elf)) + goto err; + + if (read_relas(elf)) + goto err; + + return elf; + +err: + elf_close(elf); + return NULL; +} + +void elf_close(struct elf *elf) +{ + struct section *sec, *tmpsec; + struct symbol *sym, *tmpsym; + struct rela *rela, *tmprela; + + list_for_each_entry_safe(sec, tmpsec, &elf->sections, list) { + list_for_each_entry_safe(sym, tmpsym, &sec->symbols, list) { + list_del(&sym->list); + free(sym); + } + list_for_each_entry_safe(rela, tmprela, &sec->relas, list) { + list_del(&rela->list); + free(rela); + } + list_del(&sec->list); + free(sec); + } + if (elf->name) + free(elf->name); + if (elf->fd > 0) + close(elf->fd); + if (elf->elf) + elf_end(elf->elf); + free(elf); +} diff --git a/tools/stacktool/elf.h b/tools/stacktool/elf.h new file mode 100644 index 0000000..76b86b3 --- /dev/null +++ b/tools/stacktool/elf.h @@ -0,0 +1,79 @@ +/* + * Copyright (C) 2015 Josh Poimboeuf <jpoimboe@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef _STACKTOOL_ELF_H +#define _STACKTOOL_ELF_H + +#include <stdio.h> +#include <gelf.h> +#include <linux/list.h> + +struct section { + struct list_head list; + GElf_Shdr sh; + struct list_head symbols; + struct list_head relas; + struct section *base, *rela; + struct symbol *sym; + Elf_Data *elf_data; + char *name; + int idx; + unsigned long data; + unsigned int len; +}; + +struct symbol { + struct list_head list; + GElf_Sym sym; + struct section *sec; + char *name; + int idx; + unsigned char bind, type; + unsigned long offset; + unsigned int len; +}; + +struct rela { + struct list_head list; + GElf_Rela rela; + struct symbol *sym; + unsigned int type; + int offset; + int addend; +}; + +struct elf { + Elf *elf; + GElf_Ehdr ehdr; + int fd; + char *name; + struct list_head sections; +}; + + +struct elf *elf_open(const char *name); +struct section *find_section_by_name(struct elf *elf, const char *name); +struct symbol *find_symbol_by_offset(struct section *sec, unsigned long offset); +struct rela *find_rela_by_dest(struct section *sec, unsigned long offset); +struct rela *find_rela_by_dest_range(struct section *sec, unsigned long offset, + unsigned int len); +struct symbol *find_containing_func(struct section *sec, unsigned long offset); +void elf_close(struct elf *elf); + + + +#endif /* _STACKTOOL_ELF_H */ diff --git a/tools/stacktool/special.c b/tools/stacktool/special.c new file mode 100644 index 0000000..3c7027e --- /dev/null +++ b/tools/stacktool/special.c @@ -0,0 +1,199 @@ +/* + * Copyright (C) 2015 Josh Poimboeuf <jpoimboe@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +/* + * This file reads all the special sections which have alternate instructions + * which can be patched in or redirected to at runtime. + */ + +#include <stdlib.h> +#include <string.h> + +#include "special.h" +#include "warn.h" + +#define EX_ENTRY_SIZE 8 +#define EX_ORIG_OFFSET 0 +#define EX_NEW_OFFSET 4 + +#define JUMP_ENTRY_SIZE 24 +#define JUMP_ORIG_OFFSET 0 +#define JUMP_NEW_OFFSET 8 + +#define ALT_ENTRY_SIZE 13 +#define ALT_ORIG_OFFSET 0 +#define ALT_NEW_OFFSET 4 +#define ALT_FEATURE_OFFSET 8 +#define ALT_ORIG_LEN_OFFSET 10 +#define ALT_NEW_LEN_OFFSET 11 + +#define X86_FEATURE_POPCNT 4*32+23 + +struct special_entry { + const char *sec; + bool group, jump_or_nop; + unsigned char size, orig, new; + unsigned char orig_len, new_len; /* group only */ + unsigned char feature; /* ALTERNATIVE macro CPU feature */ +}; + +struct special_entry entries[] = { + { + .sec = ".altinstructions", + .group = true, + .size = ALT_ENTRY_SIZE, + .orig = ALT_ORIG_OFFSET, + .orig_len = ALT_ORIG_LEN_OFFSET, + .new = ALT_NEW_OFFSET, + .new_len = ALT_NEW_LEN_OFFSET, + .feature = ALT_FEATURE_OFFSET, + }, + { + .sec = "__jump_table", + .jump_or_nop = true, + .size = JUMP_ENTRY_SIZE, + .orig = JUMP_ORIG_OFFSET, + .new = JUMP_NEW_OFFSET, + }, + { + .sec = "__ex_table", + .size = EX_ENTRY_SIZE, + .orig = EX_ORIG_OFFSET, + .new = EX_NEW_OFFSET, + }, + {}, +}; + +static int get_alt_entry(struct elf *elf, struct special_entry *entry, + struct section *sec, int idx, + struct special_alt *alt) +{ + struct rela *orig_rela, *new_rela; + unsigned long offset; + + offset = idx * entry->size; + + alt->group = entry->group; + alt->jump_or_nop = entry->jump_or_nop; + + if (alt->group) { + alt->orig_len = *(unsigned char *)(sec->data + offset + + entry->orig_len); + alt->new_len = *(unsigned char *)(sec->data + offset + + entry->new_len); + } + + if (entry->feature) { + unsigned short feature; + + feature = *(unsigned short *)(sec->data + offset + + entry->feature); + + /* + * It has been requested that we don't validate the !POPCNT + * feature path which is a "very very small percentage of + * machines". + */ + if (feature == X86_FEATURE_POPCNT) + alt->skip_orig = true; + } + + orig_rela = find_rela_by_dest(sec, offset + entry->orig); + if (!orig_rela) { + WARN_FUNC("can't find orig rela", sec, offset + entry->orig); + return -1; + } + if (orig_rela->sym->type != STT_SECTION) { + WARN_FUNC("don't know how to handle non-section rela symbol %s", + sec, offset + entry->orig, orig_rela->sym->name); + return -1; + } + + alt->orig_sec = orig_rela->sym->sec; + alt->orig_off = orig_rela->addend; + + if (!entry->group || alt->new_len) { + new_rela = find_rela_by_dest(sec, offset + entry->new); + if (!new_rela) { + WARN_FUNC("can't find new rela", + sec, offset + entry->new); + return -1; + } + if (new_rela->sym->type != STT_SECTION) { + WARN_FUNC("don't know how to handle non-section rela symbol %s", + sec, offset + entry->new, + new_rela->sym->name); + return -1; + } + + alt->new_sec = new_rela->sym->sec; + alt->new_off = (unsigned int)new_rela->addend; + + /* _ASM_EXTABLE_EX hack */ + if (alt->new_off >= 0x7ffffff0) + alt->new_off -= 0x7ffffff0; + } + + return 0; +} + +/* + * Read all the special sections and create a list of special_alt structs which + * describe all the alternate instructions which can be patched in or + * redirected to at runtime. + */ +int special_get_alts(struct elf *elf, struct list_head *alts) +{ + struct special_entry *entry; + struct section *sec; + unsigned int nr_entries; + struct special_alt *alt; + int idx, ret; + + INIT_LIST_HEAD(alts); + + for (entry = entries; entry->sec; entry++) { + sec = find_section_by_name(elf, entry->sec); + if (!sec) + continue; + + if (sec->len % entry->size != 0) { + WARN("%s size not a multiple of %d", + sec->name, JUMP_ENTRY_SIZE); + return -1; + } + + nr_entries = sec->len / entry->size; + + for (idx = 0; idx < nr_entries; idx++) { + alt = malloc(sizeof(*alt)); + if (!alt) { + WARN("malloc failed"); + return -1; + } + memset(alt, 0, sizeof(*alt)); + + ret = get_alt_entry(elf, entry, sec, idx, alt); + if (ret) + return ret; + + list_add_tail(&alt->list, alts); + } + } + + return 0; +} diff --git a/tools/stacktool/special.h b/tools/stacktool/special.h new file mode 100644 index 0000000..fad1d09 --- /dev/null +++ b/tools/stacktool/special.h @@ -0,0 +1,42 @@ +/* + * Copyright (C) 2015 Josh Poimboeuf <jpoimboe@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef _SPECIAL_H +#define _SPECIAL_H + +#include <stdbool.h> +#include "elf.h" + +struct special_alt { + struct list_head list; + + bool group; + bool skip_orig; + bool jump_or_nop; + + struct section *orig_sec; + unsigned long orig_off; + + struct section *new_sec; + unsigned long new_off; + + unsigned int orig_len, new_len; /* group only */ +}; + +int special_get_alts(struct elf *elf, struct list_head *alts); + +#endif /* _SPECIAL_H */ diff --git a/tools/stacktool/stacktool.c b/tools/stacktool/stacktool.c new file mode 100644 index 0000000..cbd5ca6 --- /dev/null +++ b/tools/stacktool/stacktool.c @@ -0,0 +1,980 @@ +/* + * Copyright (C) 2015 Josh Poimboeuf <jpoimboe@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +/* + * stacktool: + * + * This tool analyzes every .o file and ensures the validity of its stack trace + * metadata. It enforces a set of rules on asm code and C inline assembly code + * so that stack traces can be reliable. + * + * For more information, see tools/stacktool/Documentation/stack-validation.txt. + */ + +#include <argp.h> +#include <stdbool.h> +#include <string.h> +#include <stdlib.h> + +#include "elf.h" +#include "arch.h" +#include "special.h" +#include "warn.h" + +#define STATE_FP_SAVED 0x1 +#define STATE_FP_SETUP 0x2 + +char *objname; + +struct instruction { + struct list_head list; + struct section *sec; + unsigned long offset; + unsigned int len, state; + unsigned char type; + unsigned long immediate; + bool alt_group, visited, ignore; + struct symbol *call_dest; + struct instruction *jump_dest; + struct list_head alts; +}; + +struct list_head insns; + +struct alternative { + struct list_head list; + struct instruction *insn; +}; + +struct args { + char *args[1]; + bool fp; +}; +struct args args; +static const char args_doc[] = "file.o"; +static struct argp_option options[] = { + {"check-frame-pointer", 'f', 0, 0, "Validate frame pointer usage"}, + {0}, +}; +static error_t parse_opt(int key, char *arg, struct argp_state *state) +{ + /* Get the input argument from argp_parse, which we + know is a pointer to our args structure. */ + struct args *a = state->input; + + switch (key) { + case 'f': + a->fp = true; + break; + + case ARGP_KEY_ARG: + if (state->arg_num >= 1) + /* Too many arguments. */ + argp_usage(state); + a->args[state->arg_num] = arg; + break; + + case ARGP_KEY_END: + if (state->arg_num < 1) + /* Not enough arguments. */ + argp_usage(state); + break; + + default: + return ARGP_ERR_UNKNOWN; + } + return 0; +} +static struct argp argp = { options, parse_opt, args_doc, 0 }; + +/* + * Check for the STACKTOOL_IGNORE_INSN macro. + */ +static bool ignore_insn(struct elf *elf, struct instruction *insn) +{ + struct section *macro_sec; + struct rela *rela; + + macro_sec = find_section_by_name(elf, "__stacktool_ignore_insn"); + if (!macro_sec || !macro_sec->rela) + return false; + + list_for_each_entry(rela, ¯o_sec->rela->relas, list) + if (rela->sym->type == STT_SECTION && + rela->sym == insn->sec->sym && + rela->addend == insn->offset) + return true; + + return false; +} + +/* + * Check for the STACKTOOL_IGNORE_FUNC macro. + */ +static bool ignore_func(struct elf *elf, struct symbol *func) +{ + struct section *macro_sec; + struct rela *rela; + + macro_sec = find_section_by_name(elf, "__stacktool_ignore_func"); + if (!macro_sec || !macro_sec->rela) + return false; + + list_for_each_entry(rela, ¯o_sec->rela->relas, list) + if (rela->sym->sec == func->sec && + rela->addend == func->offset) + return true; + + + return false; +} + +static struct instruction *find_instruction(struct section *sec, + unsigned long offset) +{ + struct instruction *insn; + + list_for_each_entry(insn, &insns, list) + if (insn->sec == sec && insn->offset == offset) + return insn; + + return NULL; +} + +/* + * This checks to see if the given function is a "noreturn" function. + * + * For global functions which are outside the scope of this object file, we + * have to keep a manual list of them. + * + * For local functions, we have to detect them manually by simply looking for + * the lack of a return instruction. + */ +static bool dead_end_function(struct symbol *func) +{ + struct instruction *insn; + + if (func->bind == STB_GLOBAL && + (!strcmp(func->name, "__stack_chk_fail") || + !strcmp(func->name, "panic") || + !strcmp(func->name, "do_exit"))) + return true; + + if (!func->sec) + return false; + + insn = find_instruction(func->sec, func->offset); + if (!insn) + return false; + + list_for_each_entry_from(insn, &insns, list) { + if (insn->sec != func->sec || + insn->offset >= func->offset + func->len) + break; + + if (insn->type == INSN_RETURN) + return false; + + if (insn->type == INSN_JUMP_UNCONDITIONAL) { + struct instruction *dest = insn->jump_dest; + struct symbol *dest_func; + + if (!dest) + /* sibling call to another file */ + return false; + + if (dest->sec != func->sec || + dest->offset < func->offset || + dest->offset >= func->offset + func->len) { + /* local sibling call */ + dest_func = find_symbol_by_offset(dest->sec, + dest->offset); + if (!dest_func) + continue; + + return dead_end_function(dest_func); + } + } + + if (insn->type == INSN_JUMP_DYNAMIC) + /* sibling call */ + return false; + } + + return true; +} + +/* + * Call the arch-specific instruction decoder for all the instructions and add + * them to the global insns list. + */ +static int decode_instructions(struct elf *elf) +{ + struct section *sec; + unsigned long offset; + struct instruction *insn; + int ret; + + INIT_LIST_HEAD(&insns); + + list_for_each_entry(sec, &elf->sections, list) { + + if (!(sec->sh.sh_flags & SHF_EXECINSTR)) + continue; + + for (offset = 0; offset < sec->len; offset += insn->len) { + insn = malloc(sizeof(*insn)); + memset(insn, 0, sizeof(*insn)); + + INIT_LIST_HEAD(&insn->alts); + insn->sec = sec; + insn->offset = offset; + + ret = arch_decode_instruction(elf, sec, offset, + sec->len - offset, + &insn->len, &insn->type, + &insn->immediate); + if (ret) + return ret; + + if (!insn->type || insn->type > INSN_LAST) { + WARN_FUNC("invalid instruction type %d", + insn->sec, insn->offset, insn->type); + return -1; + } + + list_add_tail(&insn->list, &insns); + } + } + + return 0; +} + +/* + * Warnings shouldn't be reported for ignored instructions. Set insn->ignore + * for each ignored instruction and each instruction contained in an ignored + * function. + */ +static void get_ignores(struct elf *elf) +{ + struct instruction *insn; + struct section *sec; + struct symbol *func; + + list_for_each_entry(insn, &insns, list) + insn->ignore = ignore_insn(elf, insn); + + list_for_each_entry(sec, &elf->sections, list) { + list_for_each_entry(func, &sec->symbols, list) { + if (func->type != STT_FUNC) + continue; + + if (!ignore_func(elf, func)) + continue; + + insn = find_instruction(sec, func->offset); + if (!insn) + continue; + + list_for_each_entry_from(insn, &insns, list) { + if (insn->sec != func->sec || + insn->offset >= func->offset + func->len) + break; + + insn->ignore = true; + insn->visited = true; + } + } + } +} + +/* + * Find the destination instructions for all jumps. + */ +static int get_jump_destinations(struct elf *elf) +{ + struct instruction *insn; + struct rela *rela; + struct section *dest_sec; + unsigned long dest_off; + + list_for_each_entry(insn, &insns, list) { + if (insn->type != INSN_JUMP_CONDITIONAL && + insn->type != INSN_JUMP_UNCONDITIONAL) + continue; + + rela = find_rela_by_dest_range(insn->sec, insn->offset, + insn->len); + if (!rela) { + dest_sec = insn->sec; + dest_off = insn->offset + insn->len + insn->immediate; + } else if (rela->sym->type == STT_SECTION) { + dest_sec = rela->sym->sec; + dest_off = rela->addend + 4; + } else if (rela->sym->sec->idx) { + dest_sec = rela->sym->sec; + dest_off = rela->sym->sym.st_value + rela->addend + 4; + } else { + /* + * This error (jump to another file) will be handled + * later in validate_functions(). + */ + continue; + } + + insn->jump_dest = find_instruction(dest_sec, dest_off); + if (!insn->jump_dest && !insn->ignore) { + + /* + * This is a special case where an alt instruction + * jumps past the end of the section. These are + * handled later. + */ + if (!strcmp(insn->sec->name, ".altinstr_replacement")) + continue; + + WARN_FUNC("can't find jump dest instruction at %s+0x%lx", + insn->sec, insn->offset, dest_sec->name, + dest_off); + return -1; + } + } + + return 0; +} + +/* + * Find the destination instructions for all calls. + */ +static int get_call_destinations(struct elf *elf) +{ + struct instruction *insn; + unsigned long dest_off; + struct rela *rela; + + list_for_each_entry(insn, &insns, list) { + if (insn->type != INSN_CALL) + continue; + + rela = find_rela_by_dest_range(insn->sec, insn->offset, + insn->len); + if (!rela) { + dest_off = insn->offset + insn->len + insn->immediate; + insn->call_dest = find_symbol_by_offset(insn->sec, + dest_off); + if (!insn->call_dest) { + WARN_FUNC("can't find call dest symbol at offset 0x%lx", + insn->sec, insn->offset, dest_off); + return -1; + } + } else if (rela->sym->type == STT_SECTION) { + insn->call_dest = find_symbol_by_offset(rela->sym->sec, + rela->addend+4); + if (!insn->call_dest || + insn->call_dest->type != STT_FUNC) { + WARN_FUNC("can't find call dest symbol at %s+0x%x", + insn->sec, insn->offset, + rela->sym->sec->name, + rela->addend + 4); + return -1; + } + } else + insn->call_dest = rela->sym; + } + + return 0; +} + +/* + * The .alternatives section requires some extra special care, over and above + * what other special sections require: + * + * 1. Because alternatives are patched in-place, we need to insert a fake jump + * instruction at the end so that validate_branch() skips all the original + * replaced instructions when validating the new instruction path. + * + * 2. An added wrinkle is that the new instruction length might be zero. In + * that case the old instructions are replaced with noops. We simulate that + * by creating a fake jump as the only new instruction. + * + * 3. In some cases, the alternative section includes an instruction which + * conditionally jumps to the _end_ of the entry. We have to modify these + * jumps' destinations to point back to .text rather than the end of the + * entry in .altinstr_replacement. + * + * 4. It has been requested that we don't validate the !POPCNT feature path + * which is a "very very small percentage of machines". + */ +static int handle_group_alt(struct elf *elf, struct special_alt *special_alt, + struct instruction *orig_insn, + struct instruction **new_insn) +{ + struct instruction *last_orig_insn, *last_new_insn, *insn, *fake_jump; + unsigned long dest_off; + + last_orig_insn = NULL; + insn = orig_insn; + list_for_each_entry_from(insn, &insns, list) { + if (insn->sec != special_alt->orig_sec || + insn->offset >= special_alt->orig_off + special_alt->orig_len) + break; + + if (special_alt->skip_orig) + insn->ignore = true; + + insn->alt_group = true; + last_orig_insn = insn; + } + + if (list_is_last(&last_orig_insn->list, &insns) || + list_next_entry(last_orig_insn, list)->sec != special_alt->orig_sec) { + WARN("%s: don't know how to handle alternatives at end of section", + special_alt->orig_sec->name); + return -1; + } + + fake_jump = malloc(sizeof(*fake_jump)); + if (!fake_jump) { + WARN("malloc failed"); + return -1; + } + memset(fake_jump, 0, sizeof(*fake_jump)); + INIT_LIST_HEAD(&fake_jump->alts); + fake_jump->sec = special_alt->new_sec; + fake_jump->offset = -1; + fake_jump->type = INSN_JUMP_UNCONDITIONAL; + fake_jump->jump_dest = list_next_entry(last_orig_insn, list); + + if (!special_alt->new_len) { + *new_insn = fake_jump; + return 0; + } + + last_new_insn = NULL; + insn = *new_insn; + list_for_each_entry_from(insn, &insns, list) { + if (insn->sec != special_alt->new_sec || + insn->offset >= special_alt->new_off + special_alt->new_len) + break; + + last_new_insn = insn; + + if (insn->type != INSN_JUMP_CONDITIONAL && + insn->type != INSN_JUMP_UNCONDITIONAL) + continue; + + if (!insn->immediate) + continue; + + dest_off = insn->offset + insn->len + insn->immediate; + if (dest_off == special_alt->new_off + special_alt->new_len) + insn->jump_dest = fake_jump; + + if (!insn->jump_dest) { + WARN_FUNC("can't find alternative jump destination", + insn->sec, insn->offset); + return -1; + } + } + + if (!last_new_insn) { + WARN_FUNC("can't find last new alternative instruction", + special_alt->new_sec, special_alt->new_off); + return -1; + } + + list_add(&fake_jump->list, &last_new_insn->list); + + return 0; +} + +/* + * A jump table entry can either convert a nop to a jump or a jump to a nop. + * If the original instruction is a jump, make the alt entry an effective nop + * by just skipping the original instruction. + */ +static int handle_jump_alt(struct elf *elf, struct special_alt *special_alt, + struct instruction *orig_insn, + struct instruction **new_insn) +{ + if (orig_insn->type == INSN_NOP) + return 0; + + if (orig_insn->type != INSN_JUMP_UNCONDITIONAL) { + WARN_FUNC("unsupported instruction at jump label", + orig_insn->sec, orig_insn->offset); + return -1; + } + + *new_insn = list_next_entry(orig_insn, list); + return 0; +} + +/* + * Read all the special sections which have alternate instructions which can be + * patched in or redirected to at runtime. Each instruction having alternate + * instruction(s) has them added to its insn->alts list, which will be + * traversed in validate_branch(). + */ +static int get_special_section_alts(struct elf *elf) +{ + struct list_head special_alts; + struct instruction *orig_insn, *new_insn; + struct special_alt *special_alt, *tmp; + struct alternative *alt; + int ret; + + ret = special_get_alts(elf, &special_alts); + if (ret) + return ret; + + list_for_each_entry_safe(special_alt, tmp, &special_alts, list) { + alt = malloc(sizeof(*alt)); + if (!alt) { + WARN("malloc failed"); + ret = -1; + goto out; + } + + orig_insn = find_instruction(special_alt->orig_sec, + special_alt->orig_off); + if (!orig_insn) { + WARN_FUNC("special: can't find orig instruction", + special_alt->orig_sec, special_alt->orig_off); + ret = -1; + goto out; + } + + new_insn = NULL; + if (!special_alt->group || special_alt->new_len) { + new_insn = find_instruction(special_alt->new_sec, + special_alt->new_off); + if (!new_insn) { + WARN_FUNC("special: can't find new instruction", + special_alt->new_sec, + special_alt->new_off); + ret = -1; + goto out; + } + } + + if (special_alt->group) { + ret = handle_group_alt(elf, special_alt, orig_insn, + &new_insn); + if (ret) + goto out; + } else if (special_alt->jump_or_nop) { + ret = handle_jump_alt(elf, special_alt, orig_insn, + &new_insn); + if (ret) + goto out; + } + + alt->insn = new_insn; + list_add_tail(&alt->list, &orig_insn->alts); + + list_del(&special_alt->list); + free(special_alt); + } + +out: + return ret; +} + +/* + * For some switch statements, gcc generates a jump table in the .rodata + * section which contains a list of addresses within the function to jump to. + * This finds these jump tables and adds them to the insn->alts lists. + */ +static int get_switch_alts(struct elf *elf) +{ + struct instruction *insn, *alt_insn; + struct rela *rodata_rela, *rela; + struct section *rodata; + struct symbol *func; + struct alternative *alt; + + list_for_each_entry(insn, &insns, list) { + if (insn->type != INSN_JUMP_DYNAMIC) + continue; + + rodata_rela = find_rela_by_dest_range(insn->sec, insn->offset, + insn->len); + if (!rodata_rela || strcmp(rodata_rela->sym->name, ".rodata")) + continue; + + rodata = find_section_by_name(elf, ".rodata"); + if (!rodata || !rodata->rela) + continue; + + rela = find_rela_by_dest(rodata, rodata_rela->addend); + if (!rela) + continue; + + func = find_containing_func(insn->sec, insn->offset); + if (!func) { + WARN_FUNC("can't find containing func", + insn->sec, insn->offset); + return -1; + } + + list_for_each_entry_from(rela, &rodata->rela->relas, list) { + if (rela->sym->sec != insn->sec || + rela->addend <= func->offset || + rela->addend >= func->offset + func->len) + break; + + alt_insn = find_instruction(insn->sec, rela->addend); + if (!alt_insn) { + WARN("%s: can't find instruction at %s+0x%x", + rodata->rela->name, insn->sec->name, + rela->addend); + return -1; + } + + alt = malloc(sizeof(*alt)); + if (!alt) { + WARN("malloc failed"); + return -1; + } + + alt->insn = alt_insn; + list_add_tail(&alt->list, &insn->alts); + } + } + + return 0; +} + +static int decode_sections(struct elf *elf) +{ + int ret; + + ret = decode_instructions(elf); + if (ret) + return ret; + + get_ignores(elf); + + ret = get_jump_destinations(elf); + if (ret) + return ret; + + ret = get_call_destinations(elf); + if (ret) + return ret; + + ret = get_special_section_alts(elf); + if (ret) + return ret; + + ret = get_switch_alts(elf); + if (ret) + return ret; + + return 0; +} + +/* + * Follow the branch starting at the given instruction, and recursively follow + * any other branches (jumps). Meanwhile, track the frame pointer state at + * each instruction and validate all the rules described in + * tools/stacktool/Documentation/stack-validation.txt. + */ +static int validate_branch(struct elf *elf, struct instruction *first, + unsigned char first_state) +{ + struct alternative *alt; + struct instruction *insn; + struct section *sec; + unsigned char state; + int ret, warnings = 0; + + insn = first; + sec = insn->sec; + state = first_state; + + if (insn->alt_group && list_empty(&insn->alts)) { + WARN_FUNC("don't know how to handle branch to middle of alternative instruction group", + sec, insn->offset); + warnings++; + } + + while (1) { + if (insn->visited) { + if (insn->state != state) { + WARN_FUNC("frame pointer state mismatch", + sec, insn->offset); + warnings++; + } + + return warnings; + } + + insn->visited = true; + insn->state = state; + + list_for_each_entry(alt, &insn->alts, list) { + ret = validate_branch(elf, alt->insn, state); + warnings += ret; + } + + switch (insn->type) { + + case INSN_FP_SAVE: + if (args.fp) { + if (insn->state & STATE_FP_SAVED) { + WARN_FUNC("duplicate frame pointer save", + sec, insn->offset); + warnings++; + } + state |= STATE_FP_SAVED; + } + break; + + case INSN_FP_SETUP: + if (args.fp) { + if (insn->state & STATE_FP_SETUP) { + WARN_FUNC("duplicate frame pointer setup", + sec, insn->offset); + warnings++; + } + state |= STATE_FP_SETUP; + } + break; + + case INSN_FP_RESTORE: + if (args.fp) { + if (!insn->state) { + WARN_FUNC("frame pointer restore without save/setup", + sec, insn->offset); + warnings++; + } + state &= ~STATE_FP_SAVED; + state &= ~STATE_FP_SETUP; + } + break; + + case INSN_RETURN: + if (args.fp && insn->state) { + WARN_FUNC("return without frame pointer restore", + sec, insn->offset); + warnings++; + } + return warnings; + + case INSN_CALL: + if (insn->call_dest->type == STT_NOTYPE && + !strcmp(insn->call_dest->name, "__fentry__")) + break; + + if (dead_end_function(insn->call_dest)) + return warnings; + + if (args.fp && !insn->state && !insn->ignore) { + WARN_FUNC("call without frame pointer save/setup", + sec, insn->offset); + warnings++; + } + break; + + case INSN_CALL_DYNAMIC: + if (args.fp && !insn->state && !insn->ignore) { + WARN_FUNC("call without frame pointer save/setup", + sec, insn->offset); + warnings++; + } + break; + + case INSN_JUMP_CONDITIONAL: + case INSN_JUMP_UNCONDITIONAL: + if (insn->jump_dest) { + ret = validate_branch(elf, insn->jump_dest, + state); + warnings += ret; + } else if (insn->state) { + WARN_FUNC("sibling call from callable instruction with changed frame pointer", + sec, insn->offset); + warnings++; + } /* else it's a sibling call */ + + if (insn->type == INSN_JUMP_UNCONDITIONAL) + return warnings; + + break; + + case INSN_JUMP_DYNAMIC: + if (list_empty(&insn->alts) && insn->state && + !insn->ignore) { + WARN_FUNC("sibling call from callable instruction with changed frame pointer", + sec, insn->offset); + warnings++; + } + + return warnings; + + case INSN_CONTEXT_SWITCH: + if (!insn->ignore) { + WARN_FUNC("kernel entry/exit from callable instruction", + sec, insn->offset); + warnings++; + } + + return warnings; + + case INSN_BUG: + return warnings; + + default: + break; + } + + insn = list_next_entry(insn, list); + + if (&insn->list == &insns || insn->sec != sec) { + WARN("%s: unexpected end of section", sec->name); + warnings++; + return warnings; + } + } + + return warnings; +} + +static int validate_functions(struct elf *elf) +{ + struct section *sec; + struct symbol *func; + struct instruction *insn; + int ret, warnings = 0; + + list_for_each_entry(sec, &elf->sections, list) { + list_for_each_entry(func, &sec->symbols, list) { + if (func->type != STT_FUNC) + continue; + + insn = find_instruction(sec, func->offset); + if (!insn) { + WARN("%s(): can't find starting instruction", + func->name); + warnings++; + continue; + } + + ret = validate_branch(elf, insn, 0); + warnings += ret; + } + } + + list_for_each_entry(sec, &elf->sections, list) { + list_for_each_entry(func, &sec->symbols, list) { + if (func->type != STT_FUNC) + continue; + + insn = find_instruction(sec, func->offset); + if (!insn) + continue; + + list_for_each_entry_from(insn, &insns, list) { + if (insn->sec != func->sec || + insn->offset >= func->offset + func->len) + break; + + if (!insn->visited && insn->type != INSN_NOP) { + WARN_FUNC("function has unreachable instruction", + insn->sec, insn->offset); + warnings++; + } + + insn->visited = true; + } + } + } + + return warnings; +} + +static int validate_uncallable_instructions(struct elf *elf) +{ + struct instruction *insn; + int warnings = 0; + + list_for_each_entry(insn, &insns, list) { + if (!insn->visited && insn->type == INSN_RETURN && + !insn->ignore) { + WARN_FUNC("return instruction outside of a callable function", + insn->sec, insn->offset); + warnings++; + } + } + + return warnings; +} + +static void cleanup(struct elf *elf) +{ + struct instruction *insn, *tmpinsn; + struct alternative *alt, *tmpalt; + + list_for_each_entry_safe(insn, tmpinsn, &insns, list) { + list_for_each_entry_safe(alt, tmpalt, &insn->alts, list) { + list_del(&alt->list); + free(alt); + } + list_del(&insn->list); + free(insn); + } + elf_close(elf); +} + +int main(int argc, char *argv[]) +{ + struct elf *elf; + int ret, warnings = 0; + + argp_parse(&argp, argc, argv, 0, 0, &args); + + objname = args.args[0]; + elf = elf_open(objname); + if (!elf) { + fprintf(stderr, "error reading elf file %s\n", args.args[0]); + return 1; + } + + ret = decode_sections(elf); + if (ret < 0) + goto out; + warnings += ret; + + ret = validate_functions(elf); + if (ret < 0) + goto out; + warnings += ret; + + ret = validate_uncallable_instructions(elf); + if (ret < 0) + goto out; + warnings += ret; + +out: + cleanup(elf); + + /* ignore warnings for now until we get all the code cleaned up */ + if (ret || warnings) + return 0; + return 0; +} diff --git a/tools/stacktool/warn.h b/tools/stacktool/warn.h new file mode 100644 index 0000000..9f54d7e --- /dev/null +++ b/tools/stacktool/warn.h @@ -0,0 +1,60 @@ +/* + * Copyright (C) 2015 Josh Poimboeuf <jpoimboe@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef _WARN_H +#define _WARN_H + +extern char *objname; + +static inline char *offstr(struct section *sec, unsigned long offset) +{ + struct symbol *func; + char *name, *str; + unsigned long name_off; + + func = find_containing_func(sec, offset); + if (func) { + name = func->name; + name_off = offset - func->offset; + } else { + name = sec->name; + name_off = offset; + } + + str = malloc(strlen(name) + 20); + + if (func) + sprintf(str, "%s()+0x%lx", name, name_off); + else + sprintf(str, "%s+0x%lx", name, name_off); + + return str; +} + +#define WARN(format, ...) \ + fprintf(stderr, \ + "stacktool: %s: " format "\n", \ + objname, ##__VA_ARGS__) + +#define WARN_FUNC(format, sec, offset, ...) \ +({ \ + char *_str = offstr(sec, offset); \ + WARN("%s: " format, _str, ##__VA_ARGS__); \ + free(_str); \ +}) + +#endif /* _WARN_H */ -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* Re: [PATCH v13.1 04/23] x86/stacktool: Compile-time stack metadata validation 2015-09-23 12:42 ` [PATCH v13.1 " Josh Poimboeuf @ 2015-10-01 14:50 ` Chris J Arges 2015-10-01 15:25 ` Josh Poimboeuf 0 siblings, 1 reply; 38+ messages in thread From: Chris J Arges @ 2015-10-01 14:50 UTC (permalink / raw) To: Josh Poimboeuf Cc: Jiri Slaby, Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86, linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Andrew Morton <snip> > + > +$(STACKTOOL_IN): FORCE > + @$(MAKE) $(build)=stacktool > + > +$(STACKTOOL): $(STACKTOOL_IN) > + @(test -d ../../kernel -a -d ../../tools -a -d ../stacktool && (( \ > + diff -I'^#include' arch/x86/insn/insn.c ../../arch/x86/lib/insn.c >/dev/null && \ > + diff -I'^#include' arch/x86/insn/inat.c ../../arch/x86/lib/inat.c >/dev/null && \ > + diff arch/x86/insn/x86-opcode-map.txt ../../arch/x86/lib/x86-opcode-map.txt >/dev/null && \ > + diff arch/x86/insn/gen-insn-attr-x86.awk ../../arch/x86/tools/gen-insn-attr-x86.awk >/dev/null && \ > + diff -I'^#include' arch/x86/insn/insn.h ../../arch/x86/include/asm/insn.h >/dev/null && \ > + diff -I'^#include' arch/x86/insn/inat.h ../../arch/x86/include/asm/inat.h >/dev/null && \ > + diff -I'^#include' arch/x86/insn/inat_types.h ../../arch/x86/include/asm/inat_types.h >/dev/null) \ > + || echo "Warning: stacktool: x86 instruction decoder differs from kernel" >&2 )) || true > + $(QUIET_LINK)$(CC) $(STACKTOOL_IN) $(LDFLAGS) -o $@ > + > +clean: > + $(call QUIET_CLEAN, stacktool) $(RM) $(STACKTOOL) > + $(Q)find $(OUTPUT) -name '*.o' -delete -o -name '\.*.cmd' -delete -o -name '\.*.d' -delete > + $(Q)$(RM) $(OUTPUT)arch/x86/insn/inat-tables.c > + > +FORCE: > + > +.PHONY: clean FORCE Josh, I'm getting the following error when the above makefile is called: Warning: stacktool: x86 instruction decoder differs from kernel This was seen with your patches rebased on v4.3-rc2. It is failing on this diff: diff arch/x86/insn/x86-opcode-map.txt ../../arch/x86/lib/x86-opcode-map.txt And the output is as follows: 902,904d890 < 3: xrstors < 4: xsavec < 5: xsaves 949,950c935,936 < 6: XSAVEOPT | clwb (66) | mfence (11B) < 7: clflush | clflushopt (66) | sfence (11B) | pcommit (66),(11B) --- > 6: XSAVEOPT | mfence (11B) > 7: clflush | sfence (11B) Why do these files essentially need to be duplicated? Would it be possible to point Make at the files that are already in ../../arch/x86/*? --chris j arges ^ permalink raw reply [flat|nested] 38+ messages in thread
* Re: [PATCH v13.1 04/23] x86/stacktool: Compile-time stack metadata validation 2015-10-01 14:50 ` Chris J Arges @ 2015-10-01 15:25 ` Josh Poimboeuf 2015-10-01 15:38 ` Chris J Arges 0 siblings, 1 reply; 38+ messages in thread From: Josh Poimboeuf @ 2015-10-01 15:25 UTC (permalink / raw) To: Chris J Arges Cc: Jiri Slaby, Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86, linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Andrew Morton On Thu, Oct 01, 2015 at 09:50:51AM -0500, Chris J Arges wrote: > <snip> > > + > > +$(STACKTOOL_IN): FORCE > > + @$(MAKE) $(build)=stacktool > > + > > +$(STACKTOOL): $(STACKTOOL_IN) > > + @(test -d ../../kernel -a -d ../../tools -a -d ../stacktool && (( \ > > + diff -I'^#include' arch/x86/insn/insn.c ../../arch/x86/lib/insn.c >/dev/null && \ > > + diff -I'^#include' arch/x86/insn/inat.c ../../arch/x86/lib/inat.c >/dev/null && \ > > + diff arch/x86/insn/x86-opcode-map.txt ../../arch/x86/lib/x86-opcode-map.txt >/dev/null && \ > > + diff arch/x86/insn/gen-insn-attr-x86.awk ../../arch/x86/tools/gen-insn-attr-x86.awk >/dev/null && \ > > + diff -I'^#include' arch/x86/insn/insn.h ../../arch/x86/include/asm/insn.h >/dev/null && \ > > + diff -I'^#include' arch/x86/insn/inat.h ../../arch/x86/include/asm/inat.h >/dev/null && \ > > + diff -I'^#include' arch/x86/insn/inat_types.h ../../arch/x86/include/asm/inat_types.h >/dev/null) \ > > + || echo "Warning: stacktool: x86 instruction decoder differs from kernel" >&2 )) || true > > + $(QUIET_LINK)$(CC) $(STACKTOOL_IN) $(LDFLAGS) -o $@ > > + > > +clean: > > + $(call QUIET_CLEAN, stacktool) $(RM) $(STACKTOOL) > > + $(Q)find $(OUTPUT) -name '*.o' -delete -o -name '\.*.cmd' -delete -o -name '\.*.d' -delete > > + $(Q)$(RM) $(OUTPUT)arch/x86/insn/inat-tables.c > > + > > +FORCE: > > + > > +.PHONY: clean FORCE > > Josh, > > I'm getting the following error when the above makefile is called: > Warning: stacktool: x86 instruction decoder differs from kernel It's just a warning, not an error, right? Just want to make sure it didn't fail the build. The files are in sync on tip/master. > This was seen with your patches rebased on v4.3-rc2. > > It is failing on this diff: > diff arch/x86/insn/x86-opcode-map.txt ../../arch/x86/lib/x86-opcode-map.txt > > And the output is as follows: > 902,904d890 > < 3: xrstors > < 4: xsavec > < 5: xsaves > 949,950c935,936 > < 6: XSAVEOPT | clwb (66) | mfence (11B) > < 7: clflush | clflushopt (66) | sfence (11B) | pcommit (66),(11B) > --- > > 6: XSAVEOPT | mfence (11B) > > 7: clflush | sfence (11B) > > Why do these files essentially need to be duplicated? Would it be possible to > point Make at the files that are already in ../../arch/x86/*? They're duplicated because the "tools" dir is supposed to be source independent from the rest of the kernel tree. -- Josh ^ permalink raw reply [flat|nested] 38+ messages in thread
* Re: [PATCH v13.1 04/23] x86/stacktool: Compile-time stack metadata validation 2015-10-01 15:25 ` Josh Poimboeuf @ 2015-10-01 15:38 ` Chris J Arges 0 siblings, 0 replies; 38+ messages in thread From: Chris J Arges @ 2015-10-01 15:38 UTC (permalink / raw) To: Josh Poimboeuf Cc: Jiri Slaby, Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86, linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Andrew Morton On 10/01/2015 10:25 AM, Josh Poimboeuf wrote: > On Thu, Oct 01, 2015 at 09:50:51AM -0500, Chris J Arges wrote: >> <snip> >>> + >>> +$(STACKTOOL_IN): FORCE >>> + @$(MAKE) $(build)=stacktool >>> + >>> +$(STACKTOOL): $(STACKTOOL_IN) >>> + @(test -d ../../kernel -a -d ../../tools -a -d ../stacktool && (( \ >>> + diff -I'^#include' arch/x86/insn/insn.c ../../arch/x86/lib/insn.c >/dev/null && \ >>> + diff -I'^#include' arch/x86/insn/inat.c ../../arch/x86/lib/inat.c >/dev/null && \ >>> + diff arch/x86/insn/x86-opcode-map.txt ../../arch/x86/lib/x86-opcode-map.txt >/dev/null && \ >>> + diff arch/x86/insn/gen-insn-attr-x86.awk ../../arch/x86/tools/gen-insn-attr-x86.awk >/dev/null && \ >>> + diff -I'^#include' arch/x86/insn/insn.h ../../arch/x86/include/asm/insn.h >/dev/null && \ >>> + diff -I'^#include' arch/x86/insn/inat.h ../../arch/x86/include/asm/inat.h >/dev/null && \ >>> + diff -I'^#include' arch/x86/insn/inat_types.h ../../arch/x86/include/asm/inat_types.h >/dev/null) \ >>> + || echo "Warning: stacktool: x86 instruction decoder differs from kernel" >&2 )) || true >>> + $(QUIET_LINK)$(CC) $(STACKTOOL_IN) $(LDFLAGS) -o $@ >>> + >>> +clean: >>> + $(call QUIET_CLEAN, stacktool) $(RM) $(STACKTOOL) >>> + $(Q)find $(OUTPUT) -name '*.o' -delete -o -name '\.*.cmd' -delete -o -name '\.*.d' -delete >>> + $(Q)$(RM) $(OUTPUT)arch/x86/insn/inat-tables.c >>> + >>> +FORCE: >>> + >>> +.PHONY: clean FORCE >> >> Josh, >> >> I'm getting the following error when the above makefile is called: >> Warning: stacktool: x86 instruction decoder differs from kernel > > It's just a warning, not an error, right? Just want to make sure it > didn't fail the build. The files are in sync on tip/master. > Correct just a warning. And once I rebased correctly things are in sync. Thanks and sorry for the noise. --chris >> This was seen with your patches rebased on v4.3-rc2. >> >> It is failing on this diff: >> diff arch/x86/insn/x86-opcode-map.txt ../../arch/x86/lib/x86-opcode-map.txt >> >> And the output is as follows: >> 902,904d890 >> < 3: xrstors >> < 4: xsavec >> < 5: xsaves >> 949,950c935,936 >> < 6: XSAVEOPT | clwb (66) | mfence (11B) >> < 7: clflush | clflushopt (66) | sfence (11B) | pcommit (66),(11B) >> --- >>> 6: XSAVEOPT | mfence (11B) >>> 7: clflush | sfence (11B) >> >> Why do these files essentially need to be duplicated? Would it be possible to >> point Make at the files that are already in ../../arch/x86/*? > > They're duplicated because the "tools" dir is supposed to be source > independent from the rest of the kernel tree. > ^ permalink raw reply [flat|nested] 38+ messages in thread
* [PATCH v13 05/23] x86/stacktool: Add file and directory ignores 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (3 preceding siblings ...) 2015-09-22 15:46 ` [PATCH v13 04/23] x86/stacktool: Compile-time stack metadata validation Josh Poimboeuf @ 2015-09-22 15:46 ` Josh Poimboeuf 2015-09-22 15:46 ` [PATCH v13 06/23] x86/stacktool: Add ignore macros Josh Poimboeuf ` (18 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:46 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton Tell stacktool to skip validation of the following code: - boot image - vdso image - kexec purgatory - realmode - efi libstub - scripts/mod They all run outside the kernel's normal mode of operation and they don't affect runtime kernel stack traces, so they're free to skirt the stacktool rules. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> --- arch/x86/boot/Makefile | 3 ++- arch/x86/boot/compressed/Makefile | 3 ++- arch/x86/entry/vdso/Makefile | 5 ++++- arch/x86/purgatory/Makefile | 2 ++ arch/x86/realmode/Makefile | 4 +++- arch/x86/realmode/rm/Makefile | 3 ++- drivers/firmware/efi/libstub/Makefile | 1 + scripts/mod/Makefile | 2 ++ 8 files changed, 18 insertions(+), 5 deletions(-) diff --git a/arch/x86/boot/Makefile b/arch/x86/boot/Makefile index 0d553e5..fe96b2f 100644 --- a/arch/x86/boot/Makefile +++ b/arch/x86/boot/Makefile @@ -14,7 +14,8 @@ # Set it to -DSVGA_MODE=NORMAL_VGA if you just want the EGA/VGA mode. # The number is the same as you would ordinarily press at bootup. -KASAN_SANITIZE := n +KASAN_SANITIZE := n +STACKTOOL := n SVGA_MODE := -DSVGA_MODE=NORMAL_VGA diff --git a/arch/x86/boot/compressed/Makefile b/arch/x86/boot/compressed/Makefile index 0a291cd..8cea814 100644 --- a/arch/x86/boot/compressed/Makefile +++ b/arch/x86/boot/compressed/Makefile @@ -16,7 +16,8 @@ # (see scripts/Makefile.lib size_append) # compressed vmlinux.bin.all + u32 size of vmlinux.bin.all -KASAN_SANITIZE := n +KASAN_SANITIZE := n +STACKTOOL := n targets := vmlinux vmlinux.bin vmlinux.bin.gz vmlinux.bin.bz2 vmlinux.bin.lzma \ vmlinux.bin.xz vmlinux.bin.lzo vmlinux.bin.lz4 diff --git a/arch/x86/entry/vdso/Makefile b/arch/x86/entry/vdso/Makefile index a3d0767..8095694 100644 --- a/arch/x86/entry/vdso/Makefile +++ b/arch/x86/entry/vdso/Makefile @@ -3,7 +3,9 @@ # KBUILD_CFLAGS += $(DISABLE_LTO) -KASAN_SANITIZE := n + +KASAN_SANITIZE := n +STACKTOOL := n VDSO64-$(CONFIG_X86_64) := y VDSOX32-$(CONFIG_X86_X32_ABI) := y @@ -15,6 +17,7 @@ vobjs-y := vdso-note.o vclock_gettime.o vgetcpu.o # files to link into kernel obj-y += vma.o +STACKTOOL_vma.o := y # vDSO images to build vdso_img-$(VDSO64-y) += 64 diff --git a/arch/x86/purgatory/Makefile b/arch/x86/purgatory/Makefile index 2c835e3..30d7d58 100644 --- a/arch/x86/purgatory/Makefile +++ b/arch/x86/purgatory/Makefile @@ -1,3 +1,5 @@ +STACKTOOL := n + purgatory-y := purgatory.o stack.o setup-x86_$(BITS).o sha256.o entry64.o string.o targets += $(purgatory-y) diff --git a/arch/x86/realmode/Makefile b/arch/x86/realmode/Makefile index e02c2c6..0c24689 100644 --- a/arch/x86/realmode/Makefile +++ b/arch/x86/realmode/Makefile @@ -6,7 +6,9 @@ # for more details. # # -KASAN_SANITIZE := n +KASAN_SANITIZE := n +STACKTOOL := n + subdir- := rm obj-y += init.o diff --git a/arch/x86/realmode/rm/Makefile b/arch/x86/realmode/rm/Makefile index 2730d77..1da2e5b 100644 --- a/arch/x86/realmode/rm/Makefile +++ b/arch/x86/realmode/rm/Makefile @@ -6,7 +6,8 @@ # for more details. # # -KASAN_SANITIZE := n +KASAN_SANITIZE := n +STACKTOOL := n always := realmode.bin realmode.relocs diff --git a/drivers/firmware/efi/libstub/Makefile b/drivers/firmware/efi/libstub/Makefile index 816dbe9..a612cdb 100644 --- a/drivers/firmware/efi/libstub/Makefile +++ b/drivers/firmware/efi/libstub/Makefile @@ -20,6 +20,7 @@ KBUILD_CFLAGS := $(cflags-y) \ GCOV_PROFILE := n KASAN_SANITIZE := n +STACKTOOL := n lib-y := efi-stub-helper.o lib-$(CONFIG_EFI_ARMSTUB) += arm-stub.o fdt.o diff --git a/scripts/mod/Makefile b/scripts/mod/Makefile index c11212f..496184d 100644 --- a/scripts/mod/Makefile +++ b/scripts/mod/Makefile @@ -1,3 +1,5 @@ +STACKTOOL := n + hostprogs-y := modpost mk_elfconfig always := $(hostprogs-y) empty.o -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 06/23] x86/stacktool: Add ignore macros 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (4 preceding siblings ...) 2015-09-22 15:46 ` [PATCH v13 05/23] x86/stacktool: Add file and directory ignores Josh Poimboeuf @ 2015-09-22 15:46 ` Josh Poimboeuf 2015-09-22 15:46 ` [PATCH v13 07/23] x86/xen: Add stack frame dependency to hypercall inline asm calls Josh Poimboeuf ` (17 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:46 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton Add new stacktool ignore macros: STACKTOOL_IGNORE_INSN and STACKTOOL_IGNORE_FUNC. These can be used to tell stacktool to skip validation of an instruction or a function, respectively. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> --- arch/x86/include/asm/stacktool.h | 45 ++++++++++++++++++++++++++++++++++++++++ arch/x86/kernel/vmlinux.lds.S | 5 ++++- include/linux/stacktool.h | 29 ++++++++++++++++++++++++++ 3 files changed, 78 insertions(+), 1 deletion(-) create mode 100644 arch/x86/include/asm/stacktool.h create mode 100644 include/linux/stacktool.h diff --git a/arch/x86/include/asm/stacktool.h b/arch/x86/include/asm/stacktool.h new file mode 100644 index 0000000..250a37e --- /dev/null +++ b/arch/x86/include/asm/stacktool.h @@ -0,0 +1,45 @@ +#ifndef _ASM_X86_STACKTOOL_H +#define _ASM_X86_STACKTOOL_H + +#include <asm/asm.h> + +#ifdef __ASSEMBLY__ + +/* + * This asm macro tells stacktool to ignore the instruction immediately after + * the macro when doing stack metadata validation. It should only be used in + * special cases where you're 100% sure it won't affect the reliability of + * frame pointers and kernel stack traces. + * + * For more information, see tools/stacktool/Documentation/stack-validation.txt. + */ +.macro STACKTOOL_IGNORE_INSN +#ifdef CONFIG_STACK_VALIDATION + .Lstacktool_ignore_\@: + .pushsection __stacktool_ignore_insn, "a" + _ASM_ALIGN + .long .Lstacktool_ignore_\@ - . + .popsection +#endif +.endm + +#else /* !__ASSEMBLY__ */ + +#ifdef CONFIG_STACK_VALIDATION + +#define STACKTOOL_IGNORE_INSN \ + "1:\n" \ + ".pushsection __stacktool_ignore_insn, \"a\"\n" \ + _ASM_ALIGN "\n" \ + ".long 1b - .\n" \ + ".popsection\n" + +#else /* !CONFIG_STACK_VALIDATION */ + +#define STACKTOOL_IGNORE_INSN "" + +#endif /* CONFIG_STACK_VALIDATION */ + +#endif /* __ASSEMBLY__ */ + +#endif /* _ASM_X86_STACKTOOL_H */ diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S index 74e4bf1..8512b2d 100644 --- a/arch/x86/kernel/vmlinux.lds.S +++ b/arch/x86/kernel/vmlinux.lds.S @@ -332,7 +332,10 @@ SECTIONS /* Sections to be discarded */ DISCARDS - /DISCARD/ : { *(.eh_frame) } + /DISCARD/ : { + *(.eh_frame) + *(__stacktool_ignore_*) + } } diff --git a/include/linux/stacktool.h b/include/linux/stacktool.h new file mode 100644 index 0000000..c1e151b --- /dev/null +++ b/include/linux/stacktool.h @@ -0,0 +1,29 @@ +#ifndef _LINUX_STACKTOOL_H +#define _LINUX_STACKTOOL_H + +#include <asm/stacktool.h> + +#ifndef __ASSEMBLY__ + +#ifdef CONFIG_STACK_VALIDATION +/* + * This C macro tells stacktool to ignore the function when doing stack + * metadata validation. It should only be used in special cases where you're + * 100% sure it won't affect the reliability of frame pointers and kernel stack + * traces. + * + * For more information, see tools/stacktool/Documentation/stack-validation.txt. + */ +#define STACKTOOL_IGNORE_FUNC(_func) \ + static void __used __section(__stacktool_ignore_func) \ + *__stacktool_ignore_func_##_func = _func + +#else /* !CONFIG_STACK_VALIDATION */ + +#define STACKTOOL_IGNORE_FUNC(_func) + +#endif /* CONFIG_STACK_VALIDATION */ + +#endif /* !__ASSEMBLY__ */ + +#endif /* _LINUX_STACKTOOL_H */ -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 07/23] x86/xen: Add stack frame dependency to hypercall inline asm calls 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (5 preceding siblings ...) 2015-09-22 15:46 ` [PATCH v13 06/23] x86/stacktool: Add ignore macros Josh Poimboeuf @ 2015-09-22 15:46 ` Josh Poimboeuf 2015-09-22 15:46 ` [PATCH v13 08/23] x86/paravirt: Add stack frame dependency to PVOP " Josh Poimboeuf ` (16 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:46 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, David Vrabel, Konrad Rzeszutek Wilk, Boris Ostrovsky If a hypercall is inlined at the beginning of a function, gcc can insert the call instruction before setting up a stack frame, which breaks frame pointer convention if CONFIG_FRAME_POINTER is enabled and can result in a bad stack trace. Force a stack frame to be created if CONFIG_FRAME_POINTER is enabled by listing the stack pointer as an output operand for the hypercall inline asm statements. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> Reviewed-by: David Vrabel <david.vrabel@citrix.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com> --- arch/x86/include/asm/xen/hypercall.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/x86/include/asm/xen/hypercall.h b/arch/x86/include/asm/xen/hypercall.h index 83aea80..9b7dc62 100644 --- a/arch/x86/include/asm/xen/hypercall.h +++ b/arch/x86/include/asm/xen/hypercall.h @@ -110,9 +110,10 @@ extern struct { char _entry[32]; } hypercall_page[]; register unsigned long __arg2 asm(__HYPERCALL_ARG2REG) = __arg2; \ register unsigned long __arg3 asm(__HYPERCALL_ARG3REG) = __arg3; \ register unsigned long __arg4 asm(__HYPERCALL_ARG4REG) = __arg4; \ - register unsigned long __arg5 asm(__HYPERCALL_ARG5REG) = __arg5; + register unsigned long __arg5 asm(__HYPERCALL_ARG5REG) = __arg5; \ + register void *__sp asm(_ASM_SP); -#define __HYPERCALL_0PARAM "=r" (__res) +#define __HYPERCALL_0PARAM "=r" (__res), "+r" (__sp) #define __HYPERCALL_1PARAM __HYPERCALL_0PARAM, "+r" (__arg1) #define __HYPERCALL_2PARAM __HYPERCALL_1PARAM, "+r" (__arg2) #define __HYPERCALL_3PARAM __HYPERCALL_2PARAM, "+r" (__arg3) -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 08/23] x86/paravirt: Add stack frame dependency to PVOP inline asm calls 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (6 preceding siblings ...) 2015-09-22 15:46 ` [PATCH v13 07/23] x86/xen: Add stack frame dependency to hypercall inline asm calls Josh Poimboeuf @ 2015-09-22 15:46 ` Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 09/23] x86/paravirt: Create a stack frame in PV_CALLEE_SAVE_REGS_THUNK Josh Poimboeuf ` (15 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:46 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, Jeremy Fitzhardinge, Chris Wright, Alok Kataria, Rusty Russell If a PVOP call macro is inlined at the beginning of a function, gcc can insert the call instruction before setting up a stack frame, which breaks frame pointer convention if CONFIG_FRAME_POINTER is enabled and can result in a bad stack trace. Force a stack frame to be created if CONFIG_FRAME_POINTER is enabled by listing the stack pointer as an output operand for the PVOP inline asm statements. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Jeremy Fitzhardinge <jeremy@goop.org> Cc: Chris Wright <chrisw@sous-sol.org> Cc: Alok Kataria <akataria@vmware.com> Cc: Rusty Russell <rusty@rustcorp.com.au> --- arch/x86/include/asm/paravirt_types.h | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) diff --git a/arch/x86/include/asm/paravirt_types.h b/arch/x86/include/asm/paravirt_types.h index 31247b5..1885961 100644 --- a/arch/x86/include/asm/paravirt_types.h +++ b/arch/x86/include/asm/paravirt_types.h @@ -496,8 +496,9 @@ int paravirt_disable_iospace(void); * makes sure the incoming and outgoing types are always correct. */ #ifdef CONFIG_X86_32 -#define PVOP_VCALL_ARGS \ - unsigned long __eax = __eax, __edx = __edx, __ecx = __ecx +#define PVOP_VCALL_ARGS \ + unsigned long __eax = __eax, __edx = __edx, __ecx = __ecx; \ + register void *__sp asm("esp") #define PVOP_CALL_ARGS PVOP_VCALL_ARGS #define PVOP_CALL_ARG1(x) "a" ((unsigned long)(x)) @@ -515,9 +516,10 @@ int paravirt_disable_iospace(void); #define VEXTRA_CLOBBERS #else /* CONFIG_X86_64 */ /* [re]ax isn't an arg, but the return val */ -#define PVOP_VCALL_ARGS \ - unsigned long __edi = __edi, __esi = __esi, \ - __edx = __edx, __ecx = __ecx, __eax = __eax +#define PVOP_VCALL_ARGS \ + unsigned long __edi = __edi, __esi = __esi, \ + __edx = __edx, __ecx = __ecx, __eax = __eax; \ + register void *__sp asm("rsp") #define PVOP_CALL_ARGS PVOP_VCALL_ARGS #define PVOP_CALL_ARG1(x) "D" ((unsigned long)(x)) @@ -556,7 +558,7 @@ int paravirt_disable_iospace(void); asm volatile(pre \ paravirt_alt(PARAVIRT_CALL) \ post \ - : call_clbr \ + : "+r" (__sp), call_clbr \ : paravirt_type(op), \ paravirt_clobber(clbr), \ ##__VA_ARGS__ \ @@ -566,7 +568,7 @@ int paravirt_disable_iospace(void); asm volatile(pre \ paravirt_alt(PARAVIRT_CALL) \ post \ - : call_clbr \ + : "+r" (__sp), call_clbr \ : paravirt_type(op), \ paravirt_clobber(clbr), \ ##__VA_ARGS__ \ @@ -593,7 +595,7 @@ int paravirt_disable_iospace(void); asm volatile(pre \ paravirt_alt(PARAVIRT_CALL) \ post \ - : call_clbr \ + : "+r" (__sp), call_clbr \ : paravirt_type(op), \ paravirt_clobber(clbr), \ ##__VA_ARGS__ \ -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 09/23] x86/paravirt: Create a stack frame in PV_CALLEE_SAVE_REGS_THUNK 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (7 preceding siblings ...) 2015-09-22 15:46 ` [PATCH v13 08/23] x86/paravirt: Add stack frame dependency to PVOP " Josh Poimboeuf @ 2015-09-22 15:47 ` Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 10/23] x86/amd: Set ELF function type for vide() Josh Poimboeuf ` (14 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:47 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, Jeremy Fitzhardinge, Chris Wright, Alok Kataria, Rusty Russell A function created with the PV_CALLEE_SAVE_REGS_THUNK macro doesn't set up a new stack frame before the call instruction, which breaks frame pointer convention if CONFIG_FRAME_POINTER is enabled and can result in a bad stack trace. Also, the thunk functions aren't annotated as ELF callable functions. Create a stack frame when CONFIG_FRAME_POINTER is enabled and add the ELF function type. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Jeremy Fitzhardinge <jeremy@goop.org> Cc: Chris Wright <chrisw@sous-sol.org> Cc: Alok Kataria <akataria@vmware.com> Cc: Rusty Russell <rusty@rustcorp.com.au> --- arch/x86/include/asm/paravirt.h | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index 10d0596..23a382b 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -13,6 +13,7 @@ #include <linux/bug.h> #include <linux/types.h> #include <linux/cpumask.h> +#include <asm/frame.h> static inline int paravirt_enabled(void) { @@ -776,15 +777,19 @@ static __always_inline void __ticket_unlock_kick(struct arch_spinlock *lock, * call. The return value in rax/eax will not be saved, even for void * functions. */ +#define PV_THUNK_NAME(func) "__raw_callee_save_" #func #define PV_CALLEE_SAVE_REGS_THUNK(func) \ extern typeof(func) __raw_callee_save_##func; \ \ asm(".pushsection .text;" \ - ".globl __raw_callee_save_" #func " ; " \ - "__raw_callee_save_" #func ": " \ + ".globl " PV_THUNK_NAME(func) ";" \ + ".type " PV_THUNK_NAME(func) ", @function;" \ + PV_THUNK_NAME(func) ":" \ + FRAME_BEGIN \ PV_SAVE_ALL_CALLER_REGS \ "call " #func ";" \ PV_RESTORE_ALL_CALLER_REGS \ + FRAME_END \ "ret;" \ ".popsection") -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 10/23] x86/amd: Set ELF function type for vide() 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (8 preceding siblings ...) 2015-09-22 15:47 ` [PATCH v13 09/23] x86/paravirt: Create a stack frame in PV_CALLEE_SAVE_REGS_THUNK Josh Poimboeuf @ 2015-09-22 15:47 ` Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 11/23] x86/reboot: Add ljmp instructions to stacktool whitelist Josh Poimboeuf ` (13 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:47 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton vide() is a callable function, but is missing the ELF function type, which confuses tools like stacktool. Properly annotate it to be a callable function. The generated code is unchanged. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> --- arch/x86/kernel/cpu/amd.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 4a70fc6..c010de5 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -75,7 +75,10 @@ static inline int wrmsrl_amd_safe(unsigned msr, unsigned long long val) */ extern __visible void vide(void); -__asm__(".globl vide\n\t.align 4\nvide: ret"); +__asm__(".globl vide\n" + ".type vide, @function\n" + ".align 4\n" + "vide: ret\n"); static void init_amd_k5(struct cpuinfo_x86 *c) { -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 11/23] x86/reboot: Add ljmp instructions to stacktool whitelist 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (9 preceding siblings ...) 2015-09-22 15:47 ` [PATCH v13 10/23] x86/amd: Set ELF function type for vide() Josh Poimboeuf @ 2015-09-22 15:47 ` Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 12/23] x86/xen: Add xen_cpuid() and xen_setup_gdt() to stacktool whitelists Josh Poimboeuf ` (12 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:47 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton stacktool reports a false positive warning for the ljmp instruction in machine_real_restart(). Normally, ljmp isn't allowed in a function, but this is a special case where it's jumping into real mode. Add the jumps to a whitelist which tells stacktool to ignore them. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> --- arch/x86/kernel/reboot.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 02693dd..1ea1c5e 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -9,6 +9,7 @@ #include <linux/sched.h> #include <linux/tboot.h> #include <linux/delay.h> +#include <linux/stacktool.h> #include <acpi/reboot.h> #include <asm/io.h> #include <asm/apic.h> @@ -97,11 +98,13 @@ void __noreturn machine_real_restart(unsigned int type) /* Jump to the identity-mapped low memory code */ #ifdef CONFIG_X86_32 - asm volatile("jmpl *%0" : : + asm volatile(STACKTOOL_IGNORE_INSN + "jmpl *%0;" : : "rm" (real_mode_header->machine_real_restart_asm), "a" (type)); #else - asm volatile("ljmpl *%0" : : + asm volatile(STACKTOOL_IGNORE_INSN + "ljmpl *%0" : : "m" (real_mode_header->machine_real_restart_asm), "D" (type)); #endif -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 12/23] x86/xen: Add xen_cpuid() and xen_setup_gdt() to stacktool whitelists 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (10 preceding siblings ...) 2015-09-22 15:47 ` [PATCH v13 11/23] x86/reboot: Add ljmp instructions to stacktool whitelist Josh Poimboeuf @ 2015-09-22 15:47 ` Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 13/23] x86/asm/crypto: Create stack frames in aesni-intel_asm.S Josh Poimboeuf ` (11 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:47 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, David Vrabel, Konrad Rzeszutek Wilk, Boris Ostrovsky stacktool reports the following false positive warnings: stacktool: arch/x86/xen/enlighten.o: xen_cpuid()+0x41: can't find jump dest instruction at .text+0x108 stacktool: arch/x86/xen/enlighten.o: xen_setup_gdt.constprop.23()+0x2e: kernel entry/exit from callable instruction The first warning is due to xen_cpuid()'s use of XEN_EMULATE_PREFIX to insert some fake instructions which stacktool doesn't know how to decode. The second warning is due to xen_setup_gdt()'s use of an lret instruction, which stacktool normally doesn't allow in callable functions. But this seems to be a valid use of the instruction. Add both functions to the stacktool whitelist. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> Reviewed-by: David Vrabel <david.vrabel@citrix.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com> --- arch/x86/xen/enlighten.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 30d12af..1e0fe3a 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -32,6 +32,7 @@ #include <linux/gfp.h> #include <linux/memblock.h> #include <linux/edd.h> +#include <linux/stacktool.h> #include <xen/xen.h> #include <xen/events.h> @@ -346,8 +347,8 @@ static void xen_cpuid(unsigned int *ax, unsigned int *bx, *cx &= maskecx; *cx |= setecx; *dx &= maskedx; - } +STACKTOOL_IGNORE_FUNC(xen_cpuid); static bool __init xen_check_mwait(void) { @@ -1451,6 +1452,7 @@ static void __ref xen_setup_gdt(int cpu) pv_cpu_ops.write_gdt_entry = xen_write_gdt_entry; pv_cpu_ops.load_gdt = xen_load_gdt; } +STACKTOOL_IGNORE_FUNC(xen_setup_gdt); #ifdef CONFIG_XEN_PVH /* -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 13/23] x86/asm/crypto: Create stack frames in aesni-intel_asm.S 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (11 preceding siblings ...) 2015-09-22 15:47 ` [PATCH v13 12/23] x86/xen: Add xen_cpuid() and xen_setup_gdt() to stacktool whitelists Josh Poimboeuf @ 2015-09-22 15:47 ` Josh Poimboeuf 2015-10-01 6:10 ` minipli 2015-09-22 15:47 ` [PATCH v13 14/23] x86/asm/crypto: Move .Lbswap_mask data to .rodata section Josh Poimboeuf ` (10 subsequent siblings) 23 siblings, 1 reply; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:47 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, Herbert Xu, David S. Miller aesni-intel_asm.S has several callable non-leaf functions which don't honor CONFIG_FRAME_POINTER, which can result in bad stack traces. Create stack frames for them when CONFIG_FRAME_POINTER is enabled. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Herbert Xu <herbert@gondor.apana.org.au> Cc: David S. Miller <davem@davemloft.net> --- arch/x86/crypto/aesni-intel_asm.S | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/arch/x86/crypto/aesni-intel_asm.S b/arch/x86/crypto/aesni-intel_asm.S index 6bd2c6c..289ef12 100644 --- a/arch/x86/crypto/aesni-intel_asm.S +++ b/arch/x86/crypto/aesni-intel_asm.S @@ -31,6 +31,7 @@ #include <linux/linkage.h> #include <asm/inst.h> +#include <asm/frame.h> /* * The following macros are used to move an (un)aligned 16 byte value to/from @@ -1800,6 +1801,7 @@ ENDPROC(_key_expansion_256b) * unsigned int key_len) */ ENTRY(aesni_set_key) + FRAME_BEGIN #ifndef __x86_64__ pushl KEYP movl 8(%esp), KEYP # ctx @@ -1905,6 +1907,7 @@ ENTRY(aesni_set_key) #ifndef __x86_64__ popl KEYP #endif + FRAME_END ret ENDPROC(aesni_set_key) @@ -1912,6 +1915,7 @@ ENDPROC(aesni_set_key) * void aesni_enc(struct crypto_aes_ctx *ctx, u8 *dst, const u8 *src) */ ENTRY(aesni_enc) + FRAME_BEGIN #ifndef __x86_64__ pushl KEYP pushl KLEN @@ -1927,6 +1931,7 @@ ENTRY(aesni_enc) popl KLEN popl KEYP #endif + FRAME_END ret ENDPROC(aesni_enc) @@ -2101,6 +2106,7 @@ ENDPROC(_aesni_enc4) * void aesni_dec (struct crypto_aes_ctx *ctx, u8 *dst, const u8 *src) */ ENTRY(aesni_dec) + FRAME_BEGIN #ifndef __x86_64__ pushl KEYP pushl KLEN @@ -2117,6 +2123,7 @@ ENTRY(aesni_dec) popl KLEN popl KEYP #endif + FRAME_END ret ENDPROC(aesni_dec) @@ -2292,6 +2299,7 @@ ENDPROC(_aesni_dec4) * size_t len) */ ENTRY(aesni_ecb_enc) + FRAME_BEGIN #ifndef __x86_64__ pushl LEN pushl KEYP @@ -2342,6 +2350,7 @@ ENTRY(aesni_ecb_enc) popl KEYP popl LEN #endif + FRAME_END ret ENDPROC(aesni_ecb_enc) @@ -2350,6 +2359,7 @@ ENDPROC(aesni_ecb_enc) * size_t len); */ ENTRY(aesni_ecb_dec) + FRAME_BEGIN #ifndef __x86_64__ pushl LEN pushl KEYP @@ -2401,6 +2411,7 @@ ENTRY(aesni_ecb_dec) popl KEYP popl LEN #endif + FRAME_END ret ENDPROC(aesni_ecb_dec) @@ -2409,6 +2420,7 @@ ENDPROC(aesni_ecb_dec) * size_t len, u8 *iv) */ ENTRY(aesni_cbc_enc) + FRAME_BEGIN #ifndef __x86_64__ pushl IVP pushl LEN @@ -2443,6 +2455,7 @@ ENTRY(aesni_cbc_enc) popl LEN popl IVP #endif + FRAME_END ret ENDPROC(aesni_cbc_enc) @@ -2451,6 +2464,7 @@ ENDPROC(aesni_cbc_enc) * size_t len, u8 *iv) */ ENTRY(aesni_cbc_dec) + FRAME_BEGIN #ifndef __x86_64__ pushl IVP pushl LEN @@ -2534,6 +2548,7 @@ ENTRY(aesni_cbc_dec) popl LEN popl IVP #endif + FRAME_END ret ENDPROC(aesni_cbc_dec) @@ -2598,6 +2613,7 @@ ENDPROC(_aesni_inc) * size_t len, u8 *iv) */ ENTRY(aesni_ctr_enc) + FRAME_BEGIN cmp $16, LEN jb .Lctr_enc_just_ret mov 480(KEYP), KLEN @@ -2651,6 +2667,7 @@ ENTRY(aesni_ctr_enc) .Lctr_enc_ret: movups IV, (IVP) .Lctr_enc_just_ret: + FRAME_END ret ENDPROC(aesni_ctr_enc) @@ -2677,6 +2694,7 @@ ENDPROC(aesni_ctr_enc) * bool enc, u8 *iv) */ ENTRY(aesni_xts_crypt8) + FRAME_BEGIN cmpb $0, %cl movl $0, %ecx movl $240, %r10d @@ -2777,6 +2795,7 @@ ENTRY(aesni_xts_crypt8) pxor INC, STATE4 movdqu STATE4, 0x70(OUTP) + FRAME_END ret ENDPROC(aesni_xts_crypt8) -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* Re: [PATCH v13 13/23] x86/asm/crypto: Create stack frames in aesni-intel_asm.S 2015-09-22 15:47 ` [PATCH v13 13/23] x86/asm/crypto: Create stack frames in aesni-intel_asm.S Josh Poimboeuf @ 2015-10-01 6:10 ` minipli 2015-10-01 13:29 ` Josh Poimboeuf 0 siblings, 1 reply; 38+ messages in thread From: minipli @ 2015-10-01 6:10 UTC (permalink / raw) To: Josh Poimboeuf Cc: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86, linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, Herbert Xu, David S. Miller On Tue, Sep 22, 2015 at 10:47:04AM -0500, Josh Poimboeuf wrote: > aesni-intel_asm.S has several callable non-leaf functions which don't > honor CONFIG_FRAME_POINTER, which can result in bad stack traces. > > Create stack frames for them when CONFIG_FRAME_POINTER is enabled. > > Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> > Cc: Herbert Xu <herbert@gondor.apana.org.au> > Cc: David S. Miller <davem@davemloft.net> > --- > arch/x86/crypto/aesni-intel_asm.S | 19 +++++++++++++++++++ > 1 file changed, 19 insertions(+) > > diff --git a/arch/x86/crypto/aesni-intel_asm.S b/arch/x86/crypto/aesni-intel_asm.S > index 6bd2c6c..289ef12 100644 > --- a/arch/x86/crypto/aesni-intel_asm.S > +++ b/arch/x86/crypto/aesni-intel_asm.S > @@ -31,6 +31,7 @@ > > #include <linux/linkage.h> > #include <asm/inst.h> > +#include <asm/frame.h> > > /* > * The following macros are used to move an (un)aligned 16 byte value to/from > @@ -1800,6 +1801,7 @@ ENDPROC(_key_expansion_256b) > * unsigned int key_len) > */ > ENTRY(aesni_set_key) > + FRAME_BEGIN > #ifndef __x86_64__ > pushl KEYP > movl 8(%esp), KEYP # ctx This will break 32 bit builds using the aesni-intel.ko module. You need to adjust the esp-based offsets for the non-x86_64 case, as FRAME_BEGIN may do another push. How about adding a FRAME_OFFSET() macro to <asm/frame.h> to wrap the offsets?: #ifdef CONFIG_FRAME_POINTER # define FRAME_OFFSET(x) ((x) + (BITS_PER_LONG / 8)) #else # define FRAME_OFFSET(x) (x) #endif And using them like this: movl FRAME_OFFSET(8)(%esp), KEYP # ctx > @@ -1905,6 +1907,7 @@ ENTRY(aesni_set_key) > #ifndef __x86_64__ > popl KEYP > #endif > + FRAME_END > ret > ENDPROC(aesni_set_key) > > @@ -1912,6 +1915,7 @@ ENDPROC(aesni_set_key) > * void aesni_enc(struct crypto_aes_ctx *ctx, u8 *dst, const u8 *src) > */ > ENTRY(aesni_enc) > + FRAME_BEGIN > #ifndef __x86_64__ > pushl KEYP > pushl KLEN > @@ -1927,6 +1931,7 @@ ENTRY(aesni_enc) > popl KLEN > popl KEYP > #endif > + FRAME_END > ret > ENDPROC(aesni_enc) Here, too.. > > @@ -2101,6 +2106,7 @@ ENDPROC(_aesni_enc4) > * void aesni_dec (struct crypto_aes_ctx *ctx, u8 *dst, const u8 *src) > */ > ENTRY(aesni_dec) > + FRAME_BEGIN > #ifndef __x86_64__ > pushl KEYP > pushl KLEN > @@ -2117,6 +2123,7 @@ ENTRY(aesni_dec) > popl KLEN > popl KEYP > #endif > + FRAME_END > ret > ENDPROC(aesni_dec) Dito. > > @@ -2292,6 +2299,7 @@ ENDPROC(_aesni_dec4) > * size_t len) > */ > ENTRY(aesni_ecb_enc) > + FRAME_BEGIN > #ifndef __x86_64__ > pushl LEN > pushl KEYP > @@ -2342,6 +2350,7 @@ ENTRY(aesni_ecb_enc) > popl KEYP > popl LEN > #endif > + FRAME_END > ret > ENDPROC(aesni_ecb_enc) Dito. > > @@ -2350,6 +2359,7 @@ ENDPROC(aesni_ecb_enc) > * size_t len); > */ > ENTRY(aesni_ecb_dec) > + FRAME_BEGIN > #ifndef __x86_64__ > pushl LEN > pushl KEYP > @@ -2401,6 +2411,7 @@ ENTRY(aesni_ecb_dec) > popl KEYP > popl LEN > #endif > + FRAME_END > ret > ENDPROC(aesni_ecb_dec) Dito. > > @@ -2409,6 +2420,7 @@ ENDPROC(aesni_ecb_dec) > * size_t len, u8 *iv) > */ > ENTRY(aesni_cbc_enc) > + FRAME_BEGIN > #ifndef __x86_64__ > pushl IVP > pushl LEN > @@ -2443,6 +2455,7 @@ ENTRY(aesni_cbc_enc) > popl LEN > popl IVP > #endif > + FRAME_END > ret > ENDPROC(aesni_cbc_enc) Dito. > > @@ -2451,6 +2464,7 @@ ENDPROC(aesni_cbc_enc) > * size_t len, u8 *iv) > */ > ENTRY(aesni_cbc_dec) > + FRAME_BEGIN > #ifndef __x86_64__ > pushl IVP > pushl LEN > @@ -2534,6 +2548,7 @@ ENTRY(aesni_cbc_dec) > popl LEN > popl IVP > #endif > + FRAME_END > ret > ENDPROC(aesni_cbc_dec) Dito. > > @@ -2598,6 +2613,7 @@ ENDPROC(_aesni_inc) > * size_t len, u8 *iv) > */ > ENTRY(aesni_ctr_enc) > + FRAME_BEGIN > cmp $16, LEN > jb .Lctr_enc_just_ret > mov 480(KEYP), KLEN > @@ -2651,6 +2667,7 @@ ENTRY(aesni_ctr_enc) > .Lctr_enc_ret: > movups IV, (IVP) > .Lctr_enc_just_ret: > + FRAME_END > ret > ENDPROC(aesni_ctr_enc) > > @@ -2677,6 +2694,7 @@ ENDPROC(aesni_ctr_enc) > * bool enc, u8 *iv) > */ > ENTRY(aesni_xts_crypt8) > + FRAME_BEGIN > cmpb $0, %cl > movl $0, %ecx > movl $240, %r10d > @@ -2777,6 +2795,7 @@ ENTRY(aesni_xts_crypt8) > pxor INC, STATE4 > movdqu STATE4, 0x70(OUTP) > > + FRAME_END > ret > ENDPROC(aesni_xts_crypt8) > Regards, Mathias ^ permalink raw reply [flat|nested] 38+ messages in thread
* Re: [PATCH v13 13/23] x86/asm/crypto: Create stack frames in aesni-intel_asm.S 2015-10-01 6:10 ` minipli @ 2015-10-01 13:29 ` Josh Poimboeuf 2015-10-01 17:36 ` Mathias Krause 0 siblings, 1 reply; 38+ messages in thread From: Josh Poimboeuf @ 2015-10-01 13:29 UTC (permalink / raw) To: minipli Cc: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86, linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, Herbert Xu, David S. Miller On Thu, Oct 01, 2015 at 08:10:26AM +0200, minipli@ld-linux.so wrote: > On Tue, Sep 22, 2015 at 10:47:04AM -0500, Josh Poimboeuf wrote: > > aesni-intel_asm.S has several callable non-leaf functions which don't > > honor CONFIG_FRAME_POINTER, which can result in bad stack traces. > > > > Create stack frames for them when CONFIG_FRAME_POINTER is enabled. > > > > Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> > > Cc: Herbert Xu <herbert@gondor.apana.org.au> > > Cc: David S. Miller <davem@davemloft.net> > > --- > > arch/x86/crypto/aesni-intel_asm.S | 19 +++++++++++++++++++ > > 1 file changed, 19 insertions(+) > > > > diff --git a/arch/x86/crypto/aesni-intel_asm.S b/arch/x86/crypto/aesni-intel_asm.S > > index 6bd2c6c..289ef12 100644 > > --- a/arch/x86/crypto/aesni-intel_asm.S > > +++ b/arch/x86/crypto/aesni-intel_asm.S > > @@ -31,6 +31,7 @@ > > > > #include <linux/linkage.h> > > #include <asm/inst.h> > > +#include <asm/frame.h> > > > > /* > > * The following macros are used to move an (un)aligned 16 byte value to/from > > @@ -1800,6 +1801,7 @@ ENDPROC(_key_expansion_256b) > > * unsigned int key_len) > > */ > > > ENTRY(aesni_set_key) > > + FRAME_BEGIN > > #ifndef __x86_64__ > > pushl KEYP > > movl 8(%esp), KEYP # ctx > > This will break 32 bit builds using the aesni-intel.ko module. You need > to adjust the esp-based offsets for the non-x86_64 case, as FRAME_BEGIN > may do another push. > > How about adding a FRAME_OFFSET() macro to <asm/frame.h> to wrap the > offsets?: > > #ifdef CONFIG_FRAME_POINTER > # define FRAME_OFFSET(x) ((x) + (BITS_PER_LONG / 8)) > #else > # define FRAME_OFFSET(x) (x) > #endif > > And using them like this: > > movl FRAME_OFFSET(8)(%esp), KEYP # ctx Ah, right. The 32-bit ABI passes arguments on the stack instead of via registers. For now, I'm inclined to just make FRAME_BEGIN and FRAME_END do nothing on 32-bit. We're only doing stack validation on x86_64 and I don't know if anybody cares about frame pointers on 32-bit at this point. -- Josh ^ permalink raw reply [flat|nested] 38+ messages in thread
* Re: [PATCH v13 13/23] x86/asm/crypto: Create stack frames in aesni-intel_asm.S 2015-10-01 13:29 ` Josh Poimboeuf @ 2015-10-01 17:36 ` Mathias Krause 2015-10-01 17:51 ` Josh Poimboeuf 0 siblings, 1 reply; 38+ messages in thread From: Mathias Krause @ 2015-10-01 17:36 UTC (permalink / raw) To: Josh Poimboeuf Cc: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86, linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, Herbert Xu, David S. Miller On Do, Okt 01, 2015 at 08:29:50 -0500, Josh Poimboeuf wrote: > On Thu, Oct 01, 2015 at 08:10:26AM +0200, minipli@ld-linux.so wrote: > > On Tue, Sep 22, 2015 at 10:47:04AM -0500, Josh Poimboeuf wrote: > > > ENTRY(aesni_set_key) > > > + FRAME_BEGIN > > > #ifndef __x86_64__ > > > pushl KEYP > > > movl 8(%esp), KEYP # ctx > > > > This will break 32 bit builds using the aesni-intel.ko module. You need > > to adjust the esp-based offsets for the non-x86_64 case, as FRAME_BEGIN > > may do another push. > > > > How about adding a FRAME_OFFSET() macro to <asm/frame.h> to wrap the > > offsets?: > > > > #ifdef CONFIG_FRAME_POINTER > > # define FRAME_OFFSET(x) ((x) + (BITS_PER_LONG / 8)) > > #else > > # define FRAME_OFFSET(x) (x) > > #endif > > > > And using them like this: > > > > movl FRAME_OFFSET(8)(%esp), KEYP # ctx > > Ah, right. The 32-bit ABI passes arguments on the stack instead of via > registers. > > For now, I'm inclined to just make FRAME_BEGIN and FRAME_END do nothing > on 32-bit. We're only doing stack validation on x86_64 and I don't know > if anybody cares about frame pointers on 32-bit at this point. Well, we had issues in the past, especially in that very module, but only on 32 bit systems. So it would be nice to get frame pointers right for 32 bit, too. Regards, Mathias ^ permalink raw reply [flat|nested] 38+ messages in thread
* Re: [PATCH v13 13/23] x86/asm/crypto: Create stack frames in aesni-intel_asm.S 2015-10-01 17:36 ` Mathias Krause @ 2015-10-01 17:51 ` Josh Poimboeuf 0 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-10-01 17:51 UTC (permalink / raw) To: Mathias Krause Cc: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86, linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, Herbert Xu, David S. Miller On Thu, Oct 01, 2015 at 07:36:54PM +0200, Mathias Krause wrote: > On Do, Okt 01, 2015 at 08:29:50 -0500, Josh Poimboeuf wrote: > > On Thu, Oct 01, 2015 at 08:10:26AM +0200, minipli@ld-linux.so wrote: > > > On Tue, Sep 22, 2015 at 10:47:04AM -0500, Josh Poimboeuf wrote: > > > > ENTRY(aesni_set_key) > > > > + FRAME_BEGIN > > > > #ifndef __x86_64__ > > > > pushl KEYP > > > > movl 8(%esp), KEYP # ctx > > > > > > This will break 32 bit builds using the aesni-intel.ko module. You need > > > to adjust the esp-based offsets for the non-x86_64 case, as FRAME_BEGIN > > > may do another push. > > > > > > How about adding a FRAME_OFFSET() macro to <asm/frame.h> to wrap the > > > offsets?: > > > > > > #ifdef CONFIG_FRAME_POINTER > > > # define FRAME_OFFSET(x) ((x) + (BITS_PER_LONG / 8)) > > > #else > > > # define FRAME_OFFSET(x) (x) > > > #endif > > > > > > And using them like this: > > > > > > movl FRAME_OFFSET(8)(%esp), KEYP # ctx > > > > Ah, right. The 32-bit ABI passes arguments on the stack instead of via > > registers. > > > > For now, I'm inclined to just make FRAME_BEGIN and FRAME_END do nothing > > on 32-bit. We're only doing stack validation on x86_64 and I don't know > > if anybody cares about frame pointers on 32-bit at this point. > > Well, we had issues in the past, especially in that very module, but > only on 32 bit systems. So it would be nice to get frame pointers right > for 32 bit, too. Ok, I'll make your suggested changes for this file with FRAME_OFFSET. (But note that stacktool doesn't currently support 32 bit so it won't detect any frame pointer issues in 32 bit code.) -- Josh ^ permalink raw reply [flat|nested] 38+ messages in thread
* [PATCH v13 14/23] x86/asm/crypto: Move .Lbswap_mask data to .rodata section 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (12 preceding siblings ...) 2015-09-22 15:47 ` [PATCH v13 13/23] x86/asm/crypto: Create stack frames in aesni-intel_asm.S Josh Poimboeuf @ 2015-09-22 15:47 ` Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 15/23] x86/asm/crypto: Move jump_table " Josh Poimboeuf ` (9 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:47 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, Herbert Xu, David S. Miller stacktool reports the following warning: stacktool: arch/x86/crypto/aesni-intel_asm.o: _aesni_inc_init(): can't find starting instruction stacktool gets confused when it tries to disassemble the following data in the .text section: .Lbswap_mask: .byte 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0 Move it to .rodata which is a more appropriate section for read-only data. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Herbert Xu <herbert@gondor.apana.org.au> Cc: David S. Miller <davem@davemloft.net> --- arch/x86/crypto/aesni-intel_asm.S | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/x86/crypto/aesni-intel_asm.S b/arch/x86/crypto/aesni-intel_asm.S index 289ef12..18f0ced 100644 --- a/arch/x86/crypto/aesni-intel_asm.S +++ b/arch/x86/crypto/aesni-intel_asm.S @@ -2553,9 +2553,11 @@ ENTRY(aesni_cbc_dec) ENDPROC(aesni_cbc_dec) #ifdef __x86_64__ +.pushsection .rodata .align 16 .Lbswap_mask: .byte 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0 +.popsection /* * _aesni_inc_init: internal ABI -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 15/23] x86/asm/crypto: Move jump_table to .rodata section 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (13 preceding siblings ...) 2015-09-22 15:47 ` [PATCH v13 14/23] x86/asm/crypto: Move .Lbswap_mask data to .rodata section Josh Poimboeuf @ 2015-09-22 15:47 ` Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 16/23] x86/asm/crypto: Create stack frames in clmul_ghash_mul/update() Josh Poimboeuf ` (8 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:47 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, Herbert Xu, David S. Miller stacktool reports the following warning: stacktool: arch/x86/crypto/crc32c-pcl-intel-asm_64.o: crc_pcl()+0x11dd: can't decode instruction It gets confused when trying to decode jump_table data. Move jump_table to the .rodata section which is a more appropriate home for read-only data. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Herbert Xu <herbert@gondor.apana.org.au> Cc: David S. Miller <davem@davemloft.net> --- arch/x86/crypto/crc32c-pcl-intel-asm_64.S | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/x86/crypto/crc32c-pcl-intel-asm_64.S b/arch/x86/crypto/crc32c-pcl-intel-asm_64.S index 225be06..dc05f01 100644 --- a/arch/x86/crypto/crc32c-pcl-intel-asm_64.S +++ b/arch/x86/crypto/crc32c-pcl-intel-asm_64.S @@ -170,8 +170,8 @@ continue_block: ## branch into array lea jump_table(%rip), bufp movzxw (bufp, %rax, 2), len - offset=crc_array-jump_table - lea offset(bufp, len, 1), bufp + lea crc_array(%rip), bufp + lea (bufp, len, 1), bufp jmp *bufp ################################################################ @@ -310,7 +310,9 @@ do_return: popq %rdi popq %rbx ret +ENDPROC(crc_pcl) +.section .rodata, "a", %progbits ################################################################ ## jump table Table is 129 entries x 2 bytes each ################################################################ @@ -324,13 +326,11 @@ JMPTBL_ENTRY %i i=i+1 .endr -ENDPROC(crc_pcl) ################################################################ ## PCLMULQDQ tables ## Table is 128 entries x 2 words (8 bytes) each ################################################################ -.section .rotata, "a", %progbits .align 8 K_table: .long 0x493c7d27, 0x00000001 -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 16/23] x86/asm/crypto: Create stack frames in clmul_ghash_mul/update() 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (14 preceding siblings ...) 2015-09-22 15:47 ` [PATCH v13 15/23] x86/asm/crypto: Move jump_table " Josh Poimboeuf @ 2015-09-22 15:47 ` Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 17/23] x86/asm/entry: Create stack frames in thunk functions Josh Poimboeuf ` (7 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:47 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, Herbert Xu, David S. Miller clmul_ghash_mul() and clmul_ghash_update() are callable non-leaf functions which don't honor CONFIG_FRAME_POINTER, which can result in bad stack traces. Create stack frames for them when CONFIG_FRAME_POINTER is enabled. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Herbert Xu <herbert@gondor.apana.org.au> Cc: David S. Miller <davem@davemloft.net> --- arch/x86/crypto/ghash-clmulni-intel_asm.S | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/arch/x86/crypto/ghash-clmulni-intel_asm.S b/arch/x86/crypto/ghash-clmulni-intel_asm.S index 5d1e007..eed55c8 100644 --- a/arch/x86/crypto/ghash-clmulni-intel_asm.S +++ b/arch/x86/crypto/ghash-clmulni-intel_asm.S @@ -18,6 +18,7 @@ #include <linux/linkage.h> #include <asm/inst.h> +#include <asm/frame.h> .data @@ -94,6 +95,7 @@ ENDPROC(__clmul_gf128mul_ble) /* void clmul_ghash_mul(char *dst, const u128 *shash) */ ENTRY(clmul_ghash_mul) + FRAME_BEGIN movups (%rdi), DATA movups (%rsi), SHASH movaps .Lbswap_mask, BSWAP @@ -101,6 +103,7 @@ ENTRY(clmul_ghash_mul) call __clmul_gf128mul_ble PSHUFB_XMM BSWAP DATA movups DATA, (%rdi) + FRAME_END ret ENDPROC(clmul_ghash_mul) @@ -109,6 +112,7 @@ ENDPROC(clmul_ghash_mul) * const u128 *shash); */ ENTRY(clmul_ghash_update) + FRAME_BEGIN cmp $16, %rdx jb .Lupdate_just_ret # check length movaps .Lbswap_mask, BSWAP @@ -128,5 +132,6 @@ ENTRY(clmul_ghash_update) PSHUFB_XMM BSWAP DATA movups DATA, (%rdi) .Lupdate_just_ret: + FRAME_END ret ENDPROC(clmul_ghash_update) -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 17/23] x86/asm/entry: Create stack frames in thunk functions 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (15 preceding siblings ...) 2015-09-22 15:47 ` [PATCH v13 16/23] x86/asm/crypto: Create stack frames in clmul_ghash_mul/update() Josh Poimboeuf @ 2015-09-22 15:47 ` Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 18/23] x86/asm/acpi: Create a stack frame in do_suspend_lowlevel() Josh Poimboeuf ` (6 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:47 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton Thunk functions are callable non-leaf functions that don't honor CONFIG_FRAME_POINTER, which can result in bad stack traces. Also they aren't annotated as ELF callable functions which can confuse tooling. Create stack frames for them when CONFIG_FRAME_POINTER is enabled and add the ELF function type. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> --- arch/x86/entry/thunk_64.S | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/x86/entry/thunk_64.S b/arch/x86/entry/thunk_64.S index efb2b93..98df1fa 100644 --- a/arch/x86/entry/thunk_64.S +++ b/arch/x86/entry/thunk_64.S @@ -8,11 +8,14 @@ #include <linux/linkage.h> #include "calling.h" #include <asm/asm.h> +#include <asm/frame.h> /* rdi: arg1 ... normal C conventions. rax is saved/restored. */ .macro THUNK name, func, put_ret_addr_in_rdi=0 .globl \name + .type \name, @function \name: + FRAME_BEGIN /* this one pushes 9 elems, the next one would be %rIP */ pushq %rdi @@ -62,6 +65,7 @@ restore: popq %rdx popq %rsi popq %rdi + FRAME_END ret _ASM_NOKPROBE(restore) #endif -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 18/23] x86/asm/acpi: Create a stack frame in do_suspend_lowlevel() 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (16 preceding siblings ...) 2015-09-22 15:47 ` [PATCH v13 17/23] x86/asm/entry: Create stack frames in thunk functions Josh Poimboeuf @ 2015-09-22 15:47 ` Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 19/23] x86/asm: Create stack frames in rwsem functions Josh Poimboeuf ` (5 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:47 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, Pavel Machek, Rafael J. Wysocki, Len Brown do_suspend_lowlevel() is a callable non-leaf function which doesn't honor CONFIG_FRAME_POINTER, which can result in bad stack traces. Create a stack frame for it when CONFIG_FRAME_POINTER is enabled. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> Acked-by: Pavel Machek <pavel@ucw.cz> Cc: Rafael J. Wysocki <rjw@rjwysocki.net> Cc: Len Brown <len.brown@intel.com> --- arch/x86/kernel/acpi/wakeup_64.S | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/x86/kernel/acpi/wakeup_64.S b/arch/x86/kernel/acpi/wakeup_64.S index 8c35df4..169963f 100644 --- a/arch/x86/kernel/acpi/wakeup_64.S +++ b/arch/x86/kernel/acpi/wakeup_64.S @@ -5,6 +5,7 @@ #include <asm/page_types.h> #include <asm/msr.h> #include <asm/asm-offsets.h> +#include <asm/frame.h> # Copyright 2003 Pavel Machek <pavel@suse.cz>, distribute under GPLv2 @@ -39,6 +40,7 @@ bogus_64_magic: jmp bogus_64_magic ENTRY(do_suspend_lowlevel) + FRAME_BEGIN subq $8, %rsp xorl %eax, %eax call save_processor_state @@ -109,6 +111,7 @@ ENTRY(do_suspend_lowlevel) xorl %eax, %eax addq $8, %rsp + FRAME_END jmp restore_processor_state ENDPROC(do_suspend_lowlevel) -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 19/23] x86/asm: Create stack frames in rwsem functions 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (17 preceding siblings ...) 2015-09-22 15:47 ` [PATCH v13 18/23] x86/asm/acpi: Create a stack frame in do_suspend_lowlevel() Josh Poimboeuf @ 2015-09-22 15:47 ` Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 20/23] x86/asm/efi: Create a stack frame in efi_call() Josh Poimboeuf ` (4 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:47 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton rwsem.S has several callable non-leaf functions which don't honor CONFIG_FRAME_POINTER, which can result in bad stack traces. Create stack frames for them when CONFIG_FRAME_POINTER is enabled. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> --- arch/x86/lib/rwsem.S | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/arch/x86/lib/rwsem.S b/arch/x86/lib/rwsem.S index 40027db..be110ef 100644 --- a/arch/x86/lib/rwsem.S +++ b/arch/x86/lib/rwsem.S @@ -15,6 +15,7 @@ #include <linux/linkage.h> #include <asm/alternative-asm.h> +#include <asm/frame.h> #define __ASM_HALF_REG(reg) __ASM_SEL(reg, e##reg) #define __ASM_HALF_SIZE(inst) __ASM_SEL(inst##w, inst##l) @@ -84,24 +85,29 @@ /* Fix up special calling conventions */ ENTRY(call_rwsem_down_read_failed) + FRAME_BEGIN save_common_regs __ASM_SIZE(push,) %__ASM_REG(dx) movq %rax,%rdi call rwsem_down_read_failed __ASM_SIZE(pop,) %__ASM_REG(dx) restore_common_regs + FRAME_END ret ENDPROC(call_rwsem_down_read_failed) ENTRY(call_rwsem_down_write_failed) + FRAME_BEGIN save_common_regs movq %rax,%rdi call rwsem_down_write_failed restore_common_regs + FRAME_END ret ENDPROC(call_rwsem_down_write_failed) ENTRY(call_rwsem_wake) + FRAME_BEGIN /* do nothing if still outstanding active readers */ __ASM_HALF_SIZE(dec) %__ASM_HALF_REG(dx) jnz 1f @@ -109,15 +115,18 @@ ENTRY(call_rwsem_wake) movq %rax,%rdi call rwsem_wake restore_common_regs -1: ret +1: FRAME_END + ret ENDPROC(call_rwsem_wake) ENTRY(call_rwsem_downgrade_wake) + FRAME_BEGIN save_common_regs __ASM_SIZE(push,) %__ASM_REG(dx) movq %rax,%rdi call rwsem_downgrade_wake __ASM_SIZE(pop,) %__ASM_REG(dx) restore_common_regs + FRAME_END ret ENDPROC(call_rwsem_downgrade_wake) -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 20/23] x86/asm/efi: Create a stack frame in efi_call() 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (18 preceding siblings ...) 2015-09-22 15:47 ` [PATCH v13 19/23] x86/asm: Create stack frames in rwsem functions Josh Poimboeuf @ 2015-09-22 15:47 ` Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 21/23] x86/asm/power: Create stack frames in hibernate_asm_64.S Josh Poimboeuf ` (3 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:47 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, Matt Fleming efi_call() is a callable non-leaf function which doesn't honor CONFIG_FRAME_POINTER, which can result in bad stack traces. Create a stack frame for it when CONFIG_FRAME_POINTER is enabled. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> Reviewed-by: Matt Fleming <matt.fleming@intel.com> --- arch/x86/platform/efi/efi_stub_64.S | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/x86/platform/efi/efi_stub_64.S b/arch/x86/platform/efi/efi_stub_64.S index 86d0f9e..0df2dcc 100644 --- a/arch/x86/platform/efi/efi_stub_64.S +++ b/arch/x86/platform/efi/efi_stub_64.S @@ -11,6 +11,7 @@ #include <asm/msr.h> #include <asm/processor-flags.h> #include <asm/page_types.h> +#include <asm/frame.h> #define SAVE_XMM \ mov %rsp, %rax; \ @@ -74,6 +75,7 @@ .endm ENTRY(efi_call) + FRAME_BEGIN SAVE_XMM mov (%rsp), %rax mov 8(%rax), %rax @@ -88,6 +90,7 @@ ENTRY(efi_call) RESTORE_PGT addq $48, %rsp RESTORE_XMM + FRAME_END ret ENDPROC(efi_call) -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 21/23] x86/asm/power: Create stack frames in hibernate_asm_64.S 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (19 preceding siblings ...) 2015-09-22 15:47 ` [PATCH v13 20/23] x86/asm/efi: Create a stack frame in efi_call() Josh Poimboeuf @ 2015-09-22 15:47 ` Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 22/23] x86/uaccess: Add stack frame output operand in get_user inline asm Josh Poimboeuf ` (2 subsequent siblings) 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:47 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, Pavel Machek, Rafael J. Wysocki swsusp_arch_suspend() and restore_registers() are callable non-leaf functions which don't honor CONFIG_FRAME_POINTER, which can result in bad stack traces. Also they aren't annotated as ELF callable functions which can confuse tooling. Create a stack frame for them when CONFIG_FRAME_POINTER is enabled and give them proper ELF function annotations. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> Acked-by: Pavel Machek <pavel@ucw.cz> Cc: Rafael J. Wysocki <rjw@rjwysocki.net> --- arch/x86/power/hibernate_asm_64.S | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/arch/x86/power/hibernate_asm_64.S b/arch/x86/power/hibernate_asm_64.S index e2386cb..4400a43 100644 --- a/arch/x86/power/hibernate_asm_64.S +++ b/arch/x86/power/hibernate_asm_64.S @@ -21,8 +21,10 @@ #include <asm/page_types.h> #include <asm/asm-offsets.h> #include <asm/processor-flags.h> +#include <asm/frame.h> ENTRY(swsusp_arch_suspend) + FRAME_BEGIN movq $saved_context, %rax movq %rsp, pt_regs_sp(%rax) movq %rbp, pt_regs_bp(%rax) @@ -50,7 +52,9 @@ ENTRY(swsusp_arch_suspend) movq %rax, restore_cr3(%rip) call swsusp_save + FRAME_END ret +ENDPROC(swsusp_arch_suspend) ENTRY(restore_image) /* switch to temporary page tables */ @@ -107,6 +111,7 @@ ENTRY(core_restore_code) */ ENTRY(restore_registers) + FRAME_BEGIN /* go back to the original page tables */ movq %rbx, %cr3 @@ -147,4 +152,6 @@ ENTRY(restore_registers) /* tell the hibernation core that we've just restored the memory */ movq %rax, in_suspend(%rip) + FRAME_END ret +ENDPROC(restore_registers) -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 22/23] x86/uaccess: Add stack frame output operand in get_user inline asm 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (20 preceding siblings ...) 2015-09-22 15:47 ` [PATCH v13 21/23] x86/asm/power: Create stack frames in hibernate_asm_64.S Josh Poimboeuf @ 2015-09-22 15:47 ` Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 23/23] x86/stacktool: Ignore head_$(BITS) files Josh Poimboeuf 2015-09-28 13:45 ` [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:47 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: Chris J Arges, linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Andrew Morton From: Chris J Arges <chris.j.arges@canonical.com> Numerous 'call without frame pointer save/setup' warnings are introduced by stacktool because of functions using the get_user macro. Bad stack traces could occur due to lack of or misplacement of stack frame setup code. This patch forces a stack frame to be created before the inline asm code if CONFIG_FRAME_POINTER is enabled by listing the stack pointer as an output operand for the get_user inline assembly statement. Signed-off-by: Chris J Arges <chris.j.arges@canonical.com> Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> --- arch/x86/include/asm/uaccess.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h index a8df874..55b8db5 100644 --- a/arch/x86/include/asm/uaccess.h +++ b/arch/x86/include/asm/uaccess.h @@ -176,10 +176,11 @@ __typeof__(__builtin_choose_expr(sizeof(x) > sizeof(0UL), 0ULL, 0UL)) ({ \ int __ret_gu; \ register __inttype(*(ptr)) __val_gu asm("%"_ASM_DX); \ + register void *__sp asm(_ASM_SP); \ __chk_user_ptr(ptr); \ might_fault(); \ - asm volatile("call __get_user_%P3" \ - : "=a" (__ret_gu), "=r" (__val_gu) \ + asm volatile("call __get_user_%P4" \ + : "=a" (__ret_gu), "=r" (__val_gu), "+r" (__sp) \ : "0" (ptr), "i" (sizeof(*(ptr)))); \ (x) = (__force __typeof__(*(ptr))) __val_gu; \ __ret_gu; \ -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* [PATCH v13 23/23] x86/stacktool: Ignore head_$(BITS) files. 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (21 preceding siblings ...) 2015-09-22 15:47 ` [PATCH v13 22/23] x86/uaccess: Add stack frame output operand in get_user inline asm Josh Poimboeuf @ 2015-09-22 15:47 ` Josh Poimboeuf 2015-09-28 13:45 ` [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf 23 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-22 15:47 UTC (permalink / raw) To: Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86 Cc: Chris J Arges, linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Andrew Morton From: Chris J Arges <chris.j.arges@canonical.com> Tell stacktool to skip validation of head_64, head_32 as they don't affect runtime kernel stack traces. Signed-off-by: Chris J Arges <chris.j.arges@canonical.com> Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> --- arch/x86/kernel/Makefile | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index b1b78ff..f331fef 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -19,6 +19,7 @@ endif KASAN_SANITIZE_head$(BITS).o := n KASAN_SANITIZE_dumpstack.o := n KASAN_SANITIZE_dumpstack_$(BITS).o := n +STACKTOOL_head_$(BITS).o := n CFLAGS_irq.o := -I$(src)/../include/asm/trace -- 2.4.3 ^ permalink raw reply related [flat|nested] 38+ messages in thread
* Re: [PATCH v13 00/23] Compile-time stack metadata validation 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf ` (22 preceding siblings ...) 2015-09-22 15:47 ` [PATCH v13 23/23] x86/stacktool: Ignore head_$(BITS) files Josh Poimboeuf @ 2015-09-28 13:45 ` Josh Poimboeuf 2015-10-12 7:41 ` Ingo Molnar 23 siblings, 1 reply; 38+ messages in thread From: Josh Poimboeuf @ 2015-09-28 13:45 UTC (permalink / raw) To: Ingo Molnar Cc: linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, David Vrabel, Konrad Rzeszutek Wilk, Boris Ostrovsky, Jeremy Fitzhardinge, Chris Wright, Alok Kataria, Rusty Russell, Herbert Xu, David S. Miller, Pavel Machek, Rafael J. Wysocki, Len Brown, Matt Fleming, Thomas Gleixner, H. Peter Anvin, x86 Hi Ingo, Do you have any more objections to these patches? Would you be willing to apply them? Thanks! Josh On Tue, Sep 22, 2015 at 10:46:51AM -0500, Josh Poimboeuf wrote: > This is v13 of the compile-time stack metadata validation patch set, > along with proposed fixes for many of the warnings it found. It's based > on the tip/master branch. > > v12 can be found here: > > https://lkml.kernel.org/r/cover.1441749989.git.jpoimboe@redhat.com > > For more information about the motivation behind this patch set, and > more details about what it does, see the patch 4 changelog and > tools/stacktool/Documentation/stack-validation.txt. > > Patch 1 fixes the formatting for the "make -C tools" help message. > > Patches 2-6 add stacktool and some related macros. > > Patches 7-21 are some proposed fixes for several of the warnings > reported by stacktool. They've been compile-tested and boot-tested in a > VM, but I haven't attempted any meaningful testing for many of them. > > Patches 22-23 are some additional warning fixes from Chris J Arges. > > v13: > - LDFLAGS order fix from Chris J Arges > - new warning fix patches from Chris J Arges > - "--frame-pointer" -> "--check-frame-pointer" > > v12: > - rename "stackvalidate" -> "stacktool" > - move from scripts/ to tools/: > - makefile rework > - make a copy of the x86 insn code (and warn if the code diverges) > - use tools/include/linux/list.h > - move warning macros to a new warn.h file > - change wording: "stack validation" -> "stack metadata validation" > > v11: > - attempt to answer the "why" question better in the documentation and > commit message > - s/FP_SAVE/FRAME_BEGIN/ in documentation > > v10: > - add scripts/mod to directory ignores > - remove circular dependencies for ignored objects which are built > before stackvalidate > - fix CONFIG_MODVERSIONS incompatibility > > v9: > - rename FRAME/ENDFRAME -> FRAME_BEGIN/FRAME_END > - fix jump table issue for when the original instruction is a jump > - drop paravirt thunk alignment patch > - add maintainers to CC for proposed warning fixes > > v8: > - add proposed fixes for warnings > - fix all memory leaks > - process ignores earlier and add more ignore checks > - always assume POPCNT alternative is enabled > - drop hweight inline asm fix > - drop __schedule() ignore patch > - change .Ltemp_\@ to .Lstackvalidate_ignore_\@ in asm macro > - fix CONFIG_* checks in asm macros > - add C versions of ignore macros and frame macros > - change ";" to "\n" in C macros > - add ifdef CONFIG_STACK_VALIDATION checks in C ignore macros > - use numbered label in C ignore macro > - add missing break in switch case statement in arch-x86.c > > v7: > - sibling call support > - document proposed solution for inline asm() frame pointer issues > - say "kernel entry/exit" instead of "context switch" > - clarify the checking of switch statement jump tables > - discard __stackvalidate_ignore_* sections in linker script > - use .Ltemp_\@ to get a unique label instead of static 3-digit number > - change STACKVALIDATE_IGNORE_FUNC variable to a static > - move STACKVALIDATE_IGNORE_INSN to arch-specific .h file > > v6: > - rename asmvalidate -> stackvalidate (again) > - gcc-generated object file support > - recursive branch state analysis > - external jump support > - fixup/exception table support > - jump label support > - switch statement jump table support > - added documentation > - detection of "noreturn" dead end functions > - added a Kbuild mechanism for skipping files and dirs > - moved frame pointer macros to arch/x86/include/asm/frame.h > - moved ignore macros to include/linux/stackvalidate.h > > v5: > - stackvalidate -> asmvalidate > - frame pointers only required for non-leaf functions > - check for the use of the FP_SAVE/RESTORE macros instead of manually > analyzing code to detect frame pointer usage > - additional checks to ensure each function doesn't leave its boundaries > - make the macros simpler and more flexible > - support for analyzing ALTERNATIVE macros > - simplified the arch interfaces in scripts/asmvalidate/arch.h > - fixed some asmvalidate warnings > - rebased onto latest tip asm cleanups > - many more small changes > > v4: > - Changed the default to CONFIG_STACK_VALIDATION=n, until all the asm > code can get cleaned up. > - Fixed a stackvalidate error path exit code issue found by Michal > Marek. > > v3: > - Added a patch to make the push/pop CFI macros arch-independent, as > suggested by H. Peter Anvin > > v2: > - Fixed memory leaks reported by Petr Mladek > > Cc: linux-kernel@vger.kernel.org > Cc: live-patching@vger.kernel.org > Cc: Michal Marek <mmarek@suse.cz> > Cc: Peter Zijlstra <peterz@infradead.org> > Cc: Andy Lutomirski <luto@kernel.org> > Cc: Borislav Petkov <bp@alien8.de> > Cc: Linus Torvalds <torvalds@linux-foundation.org> > Cc: Andi Kleen <andi@firstfloor.org> > Cc: Pedro Alves <palves@redhat.com> > Cc: Namhyung Kim <namhyung@gmail.com> > Cc: Bernd Petrovitsch <bernd@petrovitsch.priv.at> > Cc: Chris J Arges <chris.j.arges@canonical.com> > Cc: Andrew Morton <akpm@linux-foundation.org> > > > Chris J Arges (2): > x86/uaccess: Add stack frame output operand in get_user inline asm > x86/stacktool: Ignore head_$(BITS) files. > > Josh Poimboeuf (21): > tools: Fix formatting of the "make -C tools" help message > x86/asm: Frame pointer macro cleanup > x86/asm: Add C versions of frame pointer macros > x86/stacktool: Compile-time stack metadata validation > x86/stacktool: Add file and directory ignores > x86/stacktool: Add ignore macros > x86/xen: Add stack frame dependency to hypercall inline asm calls > x86/paravirt: Add stack frame dependency to PVOP inline asm calls > x86/paravirt: Create a stack frame in PV_CALLEE_SAVE_REGS_THUNK > x86/amd: Set ELF function type for vide() > x86/reboot: Add ljmp instructions to stacktool whitelist > x86/xen: Add xen_cpuid() and xen_setup_gdt() to stacktool whitelists > x86/asm/crypto: Create stack frames in aesni-intel_asm.S > x86/asm/crypto: Move .Lbswap_mask data to .rodata section > x86/asm/crypto: Move jump_table to .rodata section > x86/asm/crypto: Create stack frames in clmul_ghash_mul/update() > x86/asm/entry: Create stack frames in thunk functions > x86/asm/acpi: Create a stack frame in do_suspend_lowlevel() > x86/asm: Create stack frames in rwsem functions > x86/asm/efi: Create a stack frame in efi_call() > x86/asm/power: Create stack frames in hibernate_asm_64.S > > MAINTAINERS | 7 + > Makefile | 5 +- > arch/Kconfig | 6 + > arch/x86/Kconfig | 1 + > arch/x86/boot/Makefile | 3 +- > arch/x86/boot/compressed/Makefile | 3 +- > arch/x86/crypto/aesni-intel_asm.S | 21 + > arch/x86/crypto/crc32c-pcl-intel-asm_64.S | 8 +- > arch/x86/crypto/ghash-clmulni-intel_asm.S | 5 + > arch/x86/entry/thunk_64.S | 4 + > arch/x86/entry/vdso/Makefile | 5 +- > arch/x86/include/asm/frame.h | 51 +- > arch/x86/include/asm/paravirt.h | 9 +- > arch/x86/include/asm/paravirt_types.h | 18 +- > arch/x86/include/asm/stacktool.h | 45 + > arch/x86/include/asm/uaccess.h | 5 +- > arch/x86/include/asm/xen/hypercall.h | 5 +- > arch/x86/kernel/Makefile | 1 + > arch/x86/kernel/acpi/wakeup_64.S | 3 + > arch/x86/kernel/cpu/amd.c | 5 +- > arch/x86/kernel/reboot.c | 7 +- > arch/x86/kernel/vmlinux.lds.S | 5 +- > arch/x86/lib/rwsem.S | 11 +- > arch/x86/platform/efi/efi_stub_64.S | 3 + > arch/x86/power/hibernate_asm_64.S | 7 + > arch/x86/purgatory/Makefile | 2 + > arch/x86/realmode/Makefile | 4 +- > arch/x86/realmode/rm/Makefile | 3 +- > arch/x86/xen/enlighten.c | 4 +- > drivers/firmware/efi/libstub/Makefile | 1 + > include/linux/stacktool.h | 29 + > lib/Kconfig.debug | 12 + > scripts/Makefile.build | 37 +- > scripts/mod/Makefile | 2 + > tools/Makefile | 43 +- > tools/stacktool/.gitignore | 1 + > tools/stacktool/Build | 4 + > tools/stacktool/Documentation/stack-validation.txt | 336 +++++++ > tools/stacktool/Makefile | 51 ++ > tools/stacktool/arch.h | 44 + > tools/stacktool/arch/x86/Build | 12 + > tools/stacktool/arch/x86/decode.c | 163 ++++ > .../stacktool/arch/x86/insn/gen-insn-attr-x86.awk | 387 ++++++++ > tools/stacktool/arch/x86/insn/inat.c | 97 ++ > tools/stacktool/arch/x86/insn/inat.h | 221 +++++ > tools/stacktool/arch/x86/insn/inat_types.h | 29 + > tools/stacktool/arch/x86/insn/insn.c | 594 +++++++++++++ > tools/stacktool/arch/x86/insn/insn.h | 201 +++++ > tools/stacktool/arch/x86/insn/x86-opcode-map.txt | 984 +++++++++++++++++++++ > tools/stacktool/elf.c | 403 +++++++++ > tools/stacktool/elf.h | 79 ++ > tools/stacktool/special.c | 199 +++++ > tools/stacktool/special.h | 42 + > tools/stacktool/stacktool.c | 980 ++++++++++++++++++++ > tools/stacktool/warn.h | 60 ++ > 55 files changed, 5196 insertions(+), 71 deletions(-) > create mode 100644 arch/x86/include/asm/stacktool.h > create mode 100644 include/linux/stacktool.h > create mode 100644 tools/stacktool/.gitignore > create mode 100644 tools/stacktool/Build > create mode 100644 tools/stacktool/Documentation/stack-validation.txt > create mode 100644 tools/stacktool/Makefile > create mode 100644 tools/stacktool/arch.h > create mode 100644 tools/stacktool/arch/x86/Build > create mode 100644 tools/stacktool/arch/x86/decode.c > create mode 100644 tools/stacktool/arch/x86/insn/gen-insn-attr-x86.awk > create mode 100644 tools/stacktool/arch/x86/insn/inat.c > create mode 100644 tools/stacktool/arch/x86/insn/inat.h > create mode 100644 tools/stacktool/arch/x86/insn/inat_types.h > create mode 100644 tools/stacktool/arch/x86/insn/insn.c > create mode 100644 tools/stacktool/arch/x86/insn/insn.h > create mode 100644 tools/stacktool/arch/x86/insn/x86-opcode-map.txt > create mode 100644 tools/stacktool/elf.c > create mode 100644 tools/stacktool/elf.h > create mode 100644 tools/stacktool/special.c > create mode 100644 tools/stacktool/special.h > create mode 100644 tools/stacktool/stacktool.c > create mode 100644 tools/stacktool/warn.h > > -- > 2.4.3 ^ permalink raw reply [flat|nested] 38+ messages in thread
* Re: [PATCH v13 00/23] Compile-time stack metadata validation 2015-09-28 13:45 ` [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf @ 2015-10-12 7:41 ` Ingo Molnar 2015-10-12 14:23 ` Josh Poimboeuf 0 siblings, 1 reply; 38+ messages in thread From: Ingo Molnar @ 2015-10-12 7:41 UTC (permalink / raw) To: Josh Poimboeuf Cc: Ingo Molnar, linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, David Vrabel, Konrad Rzeszutek Wilk, Boris Ostrovsky, Jeremy Fitzhardinge, Chris Wright, Alok Kataria, Rusty Russell, Herbert Xu, David S. Miller, Pavel Machek, Rafael J. Wysocki, Len Brown, Matt Fleming, Thomas Gleixner, H. Peter Anvin, x86 * Josh Poimboeuf <jpoimboe@redhat.com> wrote: > Hi Ingo, > > Do you have any more objections to these patches? Would you be willing > to apply them? So I still don't like the tool namespace you picked: Git-alike generic naming plus subcommands work so much better that I'm not sure why we are even having that discussion: if you name your tool 'stacktool' and alias everything you have today to under 'stacktool run ...' and add 'stacktool help' as a second, obvious subcommand then you'll have your current syntax and a lot more future flexibility and ability to branch off various functionality a'la Git, perf or kvmtool ... It's an 100% equivalent superset of your current stacktool's parameter structure, so none of your prior objections apply. Thanks, Ingo ^ permalink raw reply [flat|nested] 38+ messages in thread
* Re: [PATCH v13 00/23] Compile-time stack metadata validation 2015-10-12 7:41 ` Ingo Molnar @ 2015-10-12 14:23 ` Josh Poimboeuf 2015-11-11 18:14 ` Josh Poimboeuf 0 siblings, 1 reply; 38+ messages in thread From: Josh Poimboeuf @ 2015-10-12 14:23 UTC (permalink / raw) To: Ingo Molnar Cc: Ingo Molnar, linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, David Vrabel, Konrad Rzeszutek Wilk, Boris Ostrovsky, Jeremy Fitzhardinge, Chris Wright, Alok Kataria, Rusty Russell, Herbert Xu, David S. Miller, Pavel Machek, Rafael J. Wysocki, Len Brown, Matt Fleming, Thomas Gleixner, H. Peter Anvin, x86 On Mon, Oct 12, 2015 at 09:41:11AM +0200, Ingo Molnar wrote: > > * Josh Poimboeuf <jpoimboe@redhat.com> wrote: > > > Hi Ingo, > > > > Do you have any more objections to these patches? Would you be willing > > to apply them? > > So I still don't like the tool namespace you picked: Git-alike generic > naming plus subcommands work so much better that I'm not sure why we > are even having that discussion: Because subcommands are useful in _some_ cases, but they aren't a panacea that should be blindly applied everywhere. > if you name your tool 'stacktool' and > alias everything you have today to under 'stacktool run ...' and add > 'stacktool help' as a second, obvious subcommand then you'll have your > current syntax and a lot more future flexibility and ability to branch > off various functionality a'la Git, perf or kvmtool ... Sure, subcommands work great for monolithic framework tools like git, perf, yum, docker, etc. But stacktool is not (and never will be) a monolithic framework type of tool. The suggestion to put 100% of the functionality under 'stacktool run [options]' is certainly possible. But 'run' is so broad. What else could the tool ever do but 'run'? -- Josh ^ permalink raw reply [flat|nested] 38+ messages in thread
* Re: [PATCH v13 00/23] Compile-time stack metadata validation 2015-10-12 14:23 ` Josh Poimboeuf @ 2015-11-11 18:14 ` Josh Poimboeuf 0 siblings, 0 replies; 38+ messages in thread From: Josh Poimboeuf @ 2015-11-11 18:14 UTC (permalink / raw) To: Ingo Molnar Cc: Ingo Molnar, linux-kernel, live-patching, Michal Marek, Peter Zijlstra, Andy Lutomirski, Borislav Petkov, Linus Torvalds, Andi Kleen, Pedro Alves, Namhyung Kim, Bernd Petrovitsch, Chris J Arges, Andrew Morton, David Vrabel, Konrad Rzeszutek Wilk, Boris Ostrovsky, Jeremy Fitzhardinge, Chris Wright, Alok Kataria, Rusty Russell, Herbert Xu, David S. Miller, Pavel Machek, Rafael J. Wysocki, Len Brown, Matt Fleming, Thomas Gleixner, H. Peter Anvin, x86 Hi Ingo, Do you still have objections to my proposed command-line interface for stacktool? As I described below, I still don't think subcommands are a good fit for stacktool. However, if you strongly disagree, I can change it. The patch set is now quite mature at v14. Further, it's holding up work on several other items: - more frame pointer fixes - CFI generation and validation - livepatch consistency model - DWARF unwinder So I'm wondering if it would be possible for you to merge stacktool in time for the next merge window (4.5). Thanks! Josh On Mon, Oct 12, 2015 at 09:23:14AM -0500, Josh Poimboeuf wrote: > On Mon, Oct 12, 2015 at 09:41:11AM +0200, Ingo Molnar wrote: > > > > * Josh Poimboeuf <jpoimboe@redhat.com> wrote: > > > > > Hi Ingo, > > > > > > Do you have any more objections to these patches? Would you be willing > > > to apply them? > > > > So I still don't like the tool namespace you picked: Git-alike generic > > naming plus subcommands work so much better that I'm not sure why we > > are even having that discussion: > > Because subcommands are useful in _some_ cases, but they aren't a > panacea that should be blindly applied everywhere. > > > if you name your tool 'stacktool' and > > alias everything you have today to under 'stacktool run ...' and add > > 'stacktool help' as a second, obvious subcommand then you'll have your > > current syntax and a lot more future flexibility and ability to branch > > off various functionality a'la Git, perf or kvmtool ... > > Sure, subcommands work great for monolithic framework tools like git, > perf, yum, docker, etc. But stacktool is not (and never will be) a > monolithic framework type of tool. > > The suggestion to put 100% of the functionality under 'stacktool run > [options]' is certainly possible. But 'run' is so broad. What else > could the tool ever do but 'run'? ^ permalink raw reply [flat|nested] 38+ messages in thread
end of thread, other threads:[~2015-11-11 18:15 UTC | newest] Thread overview: 38+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2015-09-22 15:46 [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf 2015-09-22 15:46 ` [PATCH v13 01/23] tools: Fix formatting of the "make -C tools" help message Josh Poimboeuf 2015-09-22 15:46 ` [PATCH v13 02/23] x86/asm: Frame pointer macro cleanup Josh Poimboeuf 2015-09-22 15:46 ` [PATCH v13 03/23] x86/asm: Add C versions of frame pointer macros Josh Poimboeuf 2015-09-22 15:46 ` [PATCH v13 04/23] x86/stacktool: Compile-time stack metadata validation Josh Poimboeuf 2015-09-23 11:38 ` Jiri Slaby 2015-09-23 11:40 ` Jiri Slaby 2015-09-23 12:42 ` [PATCH v13.1 " Josh Poimboeuf 2015-10-01 14:50 ` Chris J Arges 2015-10-01 15:25 ` Josh Poimboeuf 2015-10-01 15:38 ` Chris J Arges 2015-09-22 15:46 ` [PATCH v13 05/23] x86/stacktool: Add file and directory ignores Josh Poimboeuf 2015-09-22 15:46 ` [PATCH v13 06/23] x86/stacktool: Add ignore macros Josh Poimboeuf 2015-09-22 15:46 ` [PATCH v13 07/23] x86/xen: Add stack frame dependency to hypercall inline asm calls Josh Poimboeuf 2015-09-22 15:46 ` [PATCH v13 08/23] x86/paravirt: Add stack frame dependency to PVOP " Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 09/23] x86/paravirt: Create a stack frame in PV_CALLEE_SAVE_REGS_THUNK Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 10/23] x86/amd: Set ELF function type for vide() Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 11/23] x86/reboot: Add ljmp instructions to stacktool whitelist Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 12/23] x86/xen: Add xen_cpuid() and xen_setup_gdt() to stacktool whitelists Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 13/23] x86/asm/crypto: Create stack frames in aesni-intel_asm.S Josh Poimboeuf 2015-10-01 6:10 ` minipli 2015-10-01 13:29 ` Josh Poimboeuf 2015-10-01 17:36 ` Mathias Krause 2015-10-01 17:51 ` Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 14/23] x86/asm/crypto: Move .Lbswap_mask data to .rodata section Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 15/23] x86/asm/crypto: Move jump_table " Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 16/23] x86/asm/crypto: Create stack frames in clmul_ghash_mul/update() Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 17/23] x86/asm/entry: Create stack frames in thunk functions Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 18/23] x86/asm/acpi: Create a stack frame in do_suspend_lowlevel() Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 19/23] x86/asm: Create stack frames in rwsem functions Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 20/23] x86/asm/efi: Create a stack frame in efi_call() Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 21/23] x86/asm/power: Create stack frames in hibernate_asm_64.S Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 22/23] x86/uaccess: Add stack frame output operand in get_user inline asm Josh Poimboeuf 2015-09-22 15:47 ` [PATCH v13 23/23] x86/stacktool: Ignore head_$(BITS) files Josh Poimboeuf 2015-09-28 13:45 ` [PATCH v13 00/23] Compile-time stack metadata validation Josh Poimboeuf 2015-10-12 7:41 ` Ingo Molnar 2015-10-12 14:23 ` Josh Poimboeuf 2015-11-11 18:14 ` Josh Poimboeuf
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).