From: tip-bot for Suresh Siddha <suresh.b.siddha@intel.com>
To: linux-tip-commits@vger.kernel.org
Cc: linux-kernel@vger.kernel.org, hpa@zytor.com, mingo@redhat.com,
roland@redhat.com, suresh.b.siddha@intel.com, tglx@linutronix.de,
hjl.tools@gmail.com
Subject: [tip:x86/ptrace] x86, ptrace: regset extensions to support xstate
Date: Thu, 11 Feb 2010 23:18:46 GMT [thread overview]
Message-ID: <tip-5b3efd500854d45d305b53c54c97db5970959980@git.kernel.org> (raw)
In-Reply-To: <20100211195614.802495327@sbs-t61.sc.intel.com>
Commit-ID: 5b3efd500854d45d305b53c54c97db5970959980
Gitweb: http://git.kernel.org/tip/5b3efd500854d45d305b53c54c97db5970959980
Author: Suresh Siddha <suresh.b.siddha@intel.com>
AuthorDate: Thu, 11 Feb 2010 11:50:59 -0800
Committer: H. Peter Anvin <hpa@zytor.com>
CommitDate: Thu, 11 Feb 2010 15:08:17 -0800
x86, ptrace: regset extensions to support xstate
Add the xstate regset support which helps extend the kernel ptrace and the
core-dump interfaces to support AVX state etc.
This regset interface is designed to support all the future state that gets
supported using xsave/xrstor infrastructure.
Looking at the memory layout saved by "xsave", one can't say which state
is represented in the memory layout. This is because if a particular state is
in init state, in the xsave hdr it can be represented by bit '0'. And hence
we can't really say by the xsave header wether a state is in init state or
the state is not saved in the memory layout.
And hence the xsave memory layout available through this regset
interface uses SW usable bytes [464..511] to convey what state is represented
in the memory layout.
First 8 bytes of the sw_usable_bytes[464..467] will be set to OS enabled xstate
mask(which is same as the 64bit mask returned by the xgetbv's xCR0).
The note NT_X86_XSTATE represents the extended state information in the
core file, using the above mentioned memory layout.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <20100211195614.802495327@sbs-t61.sc.intel.com>
Signed-off-by: Hongjiu Lu <hjl.tools@gmail.com>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
arch/x86/include/asm/i387.h | 12 +++++-
arch/x86/include/asm/user.h | 58 +++++++++++++++++++++++++++++
arch/x86/include/asm/xsave.h | 2 +
arch/x86/kernel/i387.c | 83 ++++++++++++++++++++++++++++++++++++++++++
arch/x86/kernel/ptrace.c | 34 ++++++++++++++++-
arch/x86/kernel/xsave.c | 1 +
include/linux/elf.h | 1 +
7 files changed, 187 insertions(+), 4 deletions(-)
diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h
index ebfb8a9..da29309 100644
--- a/arch/x86/include/asm/i387.h
+++ b/arch/x86/include/asm/i387.h
@@ -33,8 +33,16 @@ extern void init_thread_xstate(void);
extern int dump_fpu(struct pt_regs *, struct user_i387_struct *);
extern user_regset_active_fn fpregs_active, xfpregs_active;
-extern user_regset_get_fn fpregs_get, xfpregs_get, fpregs_soft_get;
-extern user_regset_set_fn fpregs_set, xfpregs_set, fpregs_soft_set;
+extern user_regset_get_fn fpregs_get, xfpregs_get, fpregs_soft_get,
+ xstateregs_get;
+extern user_regset_set_fn fpregs_set, xfpregs_set, fpregs_soft_set,
+ xstateregs_set;
+
+/*
+ * xstateregs_active == fpregs_active. Please refer to the comment
+ * at the definition of fpregs_active.
+ */
+#define xstateregs_active fpregs_active
extern struct _fpx_sw_bytes fx_sw_reserved;
#ifdef CONFIG_IA32_EMULATION
diff --git a/arch/x86/include/asm/user.h b/arch/x86/include/asm/user.h
index 999873b..24532c7 100644
--- a/arch/x86/include/asm/user.h
+++ b/arch/x86/include/asm/user.h
@@ -1,5 +1,63 @@
+#ifndef _ASM_X86_USER_H
+#define _ASM_X86_USER_H
+
#ifdef CONFIG_X86_32
# include "user_32.h"
#else
# include "user_64.h"
#endif
+
+#include <asm/types.h>
+
+struct user_ymmh_regs {
+ /* 16 * 16 bytes for each YMMH-reg */
+ __u32 ymmh_space[64];
+};
+
+struct user_xsave_hdr {
+ __u64 xstate_bv;
+ __u64 reserved1[2];
+ __u64 reserved2[5];
+};
+
+/*
+ * The structure layout of user_xstateregs, used for exporting the
+ * extended register state through ptrace and core-dump (NT_X86_XSTATE note)
+ * interfaces will be same as the memory layout of xsave used by the processor
+ * (except for the bytes 464..511, which can be used by the software) and hence
+ * the size of this structure varies depending on the features supported by the
+ * processor and OS. The size of the structure that users need to use can be
+ * obtained by doing:
+ * cpuid_count(0xd, 0, &eax, &ptrace_xstateregs_struct_size, &ecx, &edx);
+ * i.e., cpuid.(eax=0xd,ecx=0).ebx will be the size that user (debuggers, etc.)
+ * need to use.
+ *
+ * For now, only the first 8 bytes of the software usable bytes[464..471] will
+ * be used and will be set to OS enabled xstate mask (which is same as the
+ * 64bit mask returned by the xgetbv's xCR0). Users (analyzing core dump
+ * remotely, etc.) can use this mask as well as the mask saved in the
+ * xstate_hdr bytes and interpret what states the processor/OS supports
+ * and what states are in modified/initialized conditions for the
+ * particular process/thread.
+ *
+ * Also when the user modifies certain state FP/SSE/etc through the
+ * ptrace interface, they must ensure that the xsave_hdr.xstate_bv
+ * bytes[512..519] of the memory layout are updated correspondingly.
+ * i.e., for example when FP state is modified to a non-init state,
+ * xsave_hdr.xstate_bv's bit 0 must be set to '1', when SSE is modified to
+ * non-init state, xsave_hdr.xstate_bv's bit 1 must to be set to '1', etc.
+ */
+#define USER_XSTATE_FX_SW_WORDS 6
+#define USER_XSTATE_XCR0_WORD 0
+
+struct user_xstateregs {
+ struct {
+ __u64 fpx_space[58];
+ __u64 xstate_fx_sw[USER_XSTATE_FX_SW_WORDS];
+ } i387;
+ struct user_xsave_hdr xsave_hdr;
+ struct user_ymmh_regs ymmh;
+ /* further processor state extensions go here */
+};
+
+#endif /* _ASM_X86_USER_H */
diff --git a/arch/x86/include/asm/xsave.h b/arch/x86/include/asm/xsave.h
index 727acc1..ddc04cc 100644
--- a/arch/x86/include/asm/xsave.h
+++ b/arch/x86/include/asm/xsave.h
@@ -27,9 +27,11 @@
extern unsigned int xstate_size;
extern u64 pcntxt_mask;
extern struct xsave_struct *init_xstate_buf;
+extern u64 xstate_fx_sw_bytes[USER_XSTATE_FX_SW_WORDS];
extern void xsave_cntxt_init(void);
extern void xsave_init(void);
+extern void update_regset_xstate_info(unsigned int size, u64 xstate_mask);
extern int init_fpu(struct task_struct *child);
extern int check_for_xstate(struct i387_fxsave_struct __user *buf,
void __user *fpstate,
diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c
index f2f8540..7a8a193 100644
--- a/arch/x86/kernel/i387.c
+++ b/arch/x86/kernel/i387.c
@@ -164,6 +164,11 @@ int init_fpu(struct task_struct *tsk)
return 0;
}
+/*
+ * The xstateregs_active() routine is the same as the fpregs_active() routine,
+ * as the "regset->n" for the xstate regset will be updated based on the feature
+ * capabilites supported by the xsave.
+ */
int fpregs_active(struct task_struct *target, const struct user_regset *regset)
{
return tsk_used_math(target) ? regset->n : 0;
@@ -224,6 +229,84 @@ int xfpregs_set(struct task_struct *target, const struct user_regset *regset,
return ret;
}
+int xstateregs_get(struct task_struct *target, const struct user_regset *regset,
+ unsigned int pos, unsigned int count,
+ void *kbuf, void __user *ubuf)
+{
+ int ret;
+
+ if (!cpu_has_xsave)
+ return -ENODEV;
+
+ ret = init_fpu(target);
+ if (ret)
+ return ret;
+
+ /*
+ * First copy the fxsave bytes 0..463.
+ */
+ ret = user_regset_copyout(&pos, &count, &kbuf, &ubuf,
+ &target->thread.xstate->xsave, 0,
+ offsetof(struct user_xstateregs,
+ i387.xstate_fx_sw));
+ if (ret)
+ return ret;
+
+ /*
+ * Copy the 48bytes defined by software.
+ */
+ ret = user_regset_copyout(&pos, &count, &kbuf, &ubuf,
+ xstate_fx_sw_bytes,
+ offsetof(struct user_xstateregs,
+ i387.xstate_fx_sw),
+ offsetof(struct user_xstateregs,
+ xsave_hdr));
+ if (ret)
+ return ret;
+
+ /*
+ * Copy the rest of xstate memory layout.
+ */
+ ret = user_regset_copyout(&pos, &count, &kbuf, &ubuf,
+ &target->thread.xstate->xsave.xsave_hdr,
+ offsetof(struct user_xstateregs,
+ xsave_hdr), -1);
+ return ret;
+}
+
+int xstateregs_set(struct task_struct *target, const struct user_regset *regset,
+ unsigned int pos, unsigned int count,
+ const void *kbuf, const void __user *ubuf)
+{
+ int ret;
+ struct xsave_hdr_struct *xsave_hdr;
+
+ if (!cpu_has_xsave)
+ return -ENODEV;
+
+ ret = init_fpu(target);
+ if (ret)
+ return ret;
+
+ ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
+ &target->thread.xstate->xsave, 0, -1);
+
+ /*
+ * mxcsr reserved bits must be masked to zero for security reasons.
+ */
+ target->thread.xstate->fxsave.mxcsr &= mxcsr_feature_mask;
+
+ xsave_hdr = &target->thread.xstate->xsave.xsave_hdr;
+
+ xsave_hdr->xstate_bv &= pcntxt_mask;
+ /*
+ * These bits must be zero.
+ */
+ xsave_hdr->reserved1[0] = xsave_hdr->reserved1[1] = 0;
+
+ return ret;
+}
+
#if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION
/*
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index 017d937..16433a5 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -48,6 +48,7 @@ enum x86_regset {
REGSET_FP,
REGSET_XFP,
REGSET_IOPERM64 = REGSET_XFP,
+ REGSET_XSTATE,
REGSET_TLS,
REGSET_IOPERM32,
};
@@ -1584,7 +1585,7 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request,
#ifdef CONFIG_X86_64
-static const struct user_regset x86_64_regsets[] = {
+static struct user_regset x86_64_regsets[] __read_mostly = {
[REGSET_GENERAL] = {
.core_note_type = NT_PRSTATUS,
.n = sizeof(struct user_regs_struct) / sizeof(long),
@@ -1597,6 +1598,12 @@ static const struct user_regset x86_64_regsets[] = {
.size = sizeof(long), .align = sizeof(long),
.active = xfpregs_active, .get = xfpregs_get, .set = xfpregs_set
},
+ [REGSET_XSTATE] = {
+ .core_note_type = NT_X86_XSTATE,
+ .size = sizeof(u64), .align = sizeof(u64),
+ .active = xstateregs_active, .get = xstateregs_get,
+ .set = xstateregs_set
+ },
[REGSET_IOPERM64] = {
.core_note_type = NT_386_IOPERM,
.n = IO_BITMAP_LONGS,
@@ -1622,7 +1629,7 @@ static const struct user_regset_view user_x86_64_view = {
#endif /* CONFIG_X86_64 */
#if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION
-static const struct user_regset x86_32_regsets[] = {
+static struct user_regset x86_32_regsets[] __read_mostly = {
[REGSET_GENERAL] = {
.core_note_type = NT_PRSTATUS,
.n = sizeof(struct user_regs_struct32) / sizeof(u32),
@@ -1641,6 +1648,12 @@ static const struct user_regset x86_32_regsets[] = {
.size = sizeof(u32), .align = sizeof(u32),
.active = xfpregs_active, .get = xfpregs_get, .set = xfpregs_set
},
+ [REGSET_XSTATE] = {
+ .core_note_type = NT_X86_XSTATE,
+ .size = sizeof(u64), .align = sizeof(u64),
+ .active = xstateregs_active, .get = xstateregs_get,
+ .set = xstateregs_set
+ },
[REGSET_TLS] = {
.core_note_type = NT_386_TLS,
.n = GDT_ENTRY_TLS_ENTRIES, .bias = GDT_ENTRY_TLS_MIN,
@@ -1663,6 +1676,23 @@ static const struct user_regset_view user_x86_32_view = {
};
#endif
+/*
+ * This represents bytes 464..511 in the memory layout exported through
+ * the REGSET_XSTATE interface.
+ */
+u64 xstate_fx_sw_bytes[USER_XSTATE_FX_SW_WORDS];
+
+void update_regset_xstate_info(unsigned int size, u64 xstate_mask)
+{
+#ifdef CONFIG_X86_64
+ x86_64_regsets[REGSET_XSTATE].n = size / sizeof(u64);
+#endif
+#if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION
+ x86_32_regsets[REGSET_XSTATE].n = size / sizeof(u64);
+#endif
+ xstate_fx_sw_bytes[USER_XSTATE_XCR0_WORD] = xstate_mask;
+}
+
const struct user_regset_view *task_user_regset_view(struct task_struct *task)
{
#ifdef CONFIG_IA32_EMULATION
diff --git a/arch/x86/kernel/xsave.c b/arch/x86/kernel/xsave.c
index c5ee17e..782c3a3 100644
--- a/arch/x86/kernel/xsave.c
+++ b/arch/x86/kernel/xsave.c
@@ -337,6 +337,7 @@ void __ref xsave_cntxt_init(void)
cpuid_count(0xd, 0, &eax, &ebx, &ecx, &edx);
xstate_size = ebx;
+ update_regset_xstate_info(xstate_size, pcntxt_mask);
prepare_fx_sw_frame();
setup_xstate_init();
diff --git a/include/linux/elf.h b/include/linux/elf.h
index 0cc4d55..a8c4af0 100644
--- a/include/linux/elf.h
+++ b/include/linux/elf.h
@@ -361,6 +361,7 @@ typedef struct elf64_shdr {
#define NT_PPC_VSX 0x102 /* PowerPC VSX registers */
#define NT_386_TLS 0x200 /* i386 TLS slots (struct user_desc) */
#define NT_386_IOPERM 0x201 /* x86 io permission bitmap (1=deny) */
+#define NT_X86_XSTATE 0x202 /* x86 extended state using xsave */
#define NT_S390_HIGH_GPRS 0x300 /* s390 upper register halves */
next prev parent reply other threads:[~2010-02-11 23:19 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-02-11 19:50 [patch v3 0/2] updated ptrace/core-dump patches for supporting xstate - v3 Suresh Siddha
2010-02-11 19:50 ` [patch v3 1/2] x86, ptrace: regset extensions to support xstate Suresh Siddha
2010-02-11 23:18 ` tip-bot for Suresh Siddha [this message]
2010-02-12 3:45 ` Roland McGrath
2010-02-12 17:31 ` Oleg Nesterov
2010-02-11 19:51 ` [patch v3 2/2] ptrace: Add support for generic PTRACE_GETREGSET/PTRACE_SETREGSET Suresh Siddha
2010-02-11 23:19 ` [tip:x86/ptrace] " tip-bot for Suresh Siddha
2010-02-22 9:07 ` Ingo Molnar
2010-02-22 9:33 ` linux-next requiements (Was: Re: [tip:x86/ptrace] ptrace: Add support for generic PTRACE_GETREGSET/PTRACE_SETREGSET) Stephen Rothwell
2010-02-22 10:27 ` Ingo Molnar
2010-02-22 11:47 ` linux-next requirements " Stephen Rothwell
2010-02-22 22:57 ` H. Peter Anvin
2010-02-22 23:59 ` Stephen Rothwell
2010-02-23 20:20 ` Roland McGrath
2010-02-23 20:49 ` H. Peter Anvin
2010-02-23 22:54 ` Stephen Rothwell
2010-02-23 8:45 ` Ingo Molnar
2010-02-23 19:52 ` Al Viro
2010-02-23 19:57 ` Al Viro
2010-02-24 7:25 ` linux-next requirements Stephen Rothwell
2010-02-27 1:53 ` Grant Likely
2010-02-27 8:53 ` Geert Uytterhoeven
2010-02-27 9:09 ` Jaswinder Singh Rajput
2010-02-27 9:39 ` Ingo Molnar
2010-02-27 12:23 ` Rafael J. Wysocki
2010-02-27 12:47 ` Ingo Molnar
2010-02-27 19:07 ` Rafael J. Wysocki
2010-02-27 21:50 ` Geert Uytterhoeven
2010-02-27 22:31 ` Rafael J. Wysocki
2010-02-28 7:06 ` Ingo Molnar
2010-02-28 12:22 ` Rafael J. Wysocki
2010-02-28 7:14 ` Ingo Molnar
2010-02-28 7:37 ` Stephen Rothwell
2010-02-28 7:51 ` Ingo Molnar
2010-02-28 8:19 ` Al Viro
2010-02-28 8:53 ` Ingo Molnar
2010-02-28 10:26 ` Stephen Rothwell
2010-02-28 7:23 ` Ingo Molnar
2010-03-01 15:13 ` Nick Bowler
2010-03-03 21:53 ` Pavel Machek
2010-03-04 0:35 ` Thomas Gleixner
2010-03-04 0:42 ` Andrew Morton
2010-03-04 1:17 ` Thomas Gleixner
2010-03-04 2:48 ` Ingo Molnar
2010-02-22 18:37 ` [tip:x86/ptrace] ptrace: Add support for generic PTRACE_GETREGSET/PTRACE_SETREGSET Roland McGrath
2010-02-23 18:36 ` [tip:x86/ptrace] parisc: Disable CONFIG_HAVE_ARCH_TRACEHOOK tip-bot for Roland McGrath
2010-02-12 3:56 ` [patch v3 2/2] ptrace: Add support for generic PTRACE_GETREGSET/PTRACE_SETREGSET Roland McGrath
2010-02-12 15:59 ` Oleg Nesterov
-- strict thread matches above, loose matches on Subject: below --
2010-02-09 20:13 [patch v2 2/4] x86, ptrace: regset extensions to support xstate Suresh Siddha
2010-02-09 22:54 ` [tip:x86/ptrace] x86, ptrace: Regset " tip-bot for Suresh Siddha
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=tip-5b3efd500854d45d305b53c54c97db5970959980@git.kernel.org \
--to=suresh.b.siddha@intel.com \
--cc=hjl.tools@gmail.com \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-tip-commits@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=roland@redhat.com \
--cc=tglx@linutronix.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.