public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: LKML <linux-kernel@vger.kernel.org>
Cc: x86@kernel.org, Al Viro <viro@zeniv.linux.org.uk>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Tony Luck <tony.luck@intel.com>, Song Liu <songliubraving@fb.com>,
	Alexei Starovoitov <ast@kernel.org>,
	Daniel Borkmann <daniel@iogearbox.net>,
	Peter Ziljstra <peterz@infradead.org>
Subject: [patch V2.1 10/20] x86/fpu/signal: Clarify exception handling in restore_fpregs_from_user()
Date: Tue,  7 Sep 2021 22:24:55 +0200 (CEST)	[thread overview]
Message-ID: <20210907200848.818578802@linutronix.de> (raw)
In-Reply-To: 20210907200722.067068005@linutronix.de

FPU restore from a signal frame can trigger various exceptions. The
exceptions are caught with an exception table entry. The handler of this
entry stores the trap number in EAX. The FPU specific fixup negates that
trap number to convert it into an negative error code.

Any other exception than #PF is fatal and recovery is not possible. This
relies on the fact that the #PF exception number is the same as EFAULT, but
that's not really obvious.

Remove the negation from the exception fixup as it really has no value and
check for X86_TRAP_PF at the call site.

There is still confusion due to the return code conversion for the error
case which will be cleaned up separately.

Suggested-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
V2: Remove the negation (Al)
---
 arch/x86/include/asm/fpu/internal.h |   21 ++++++++-------------
 arch/x86/kernel/fpu/signal.c        |    5 +++--
 2 files changed, 11 insertions(+), 15 deletions(-)
--- a/arch/x86/include/asm/fpu/internal.h
+++ b/arch/x86/include/asm/fpu/internal.h
@@ -88,7 +88,10 @@ static inline void fpstate_init_soft(str
 #endif
 extern void save_fpregs_to_fpstate(struct fpu *fpu);
 
-/* Returns 0 or the negated trap number, which results in -EFAULT for #PF */
+/*
+ * Returns 0 on success or the trap number when the operation raises an
+ * exception.
+ */
 #define user_insn(insn, output, input...)				\
 ({									\
 	int err;							\
@@ -98,11 +101,7 @@ extern void save_fpregs_to_fpstate(struc
 	asm volatile(ASM_STAC "\n"					\
 		     "1: " #insn "\n"					\
 		     "2: " ASM_CLAC "\n"				\
-		     ".section .fixup,\"ax\"\n"				\
-		     "3:  negl %%eax\n"					\
-		     "    jmp  2b\n"					\
-		     ".previous\n"					\
-		     _ASM_EXTABLE_TYPE(1b, 3b, EX_TYPE_FAULT_MCE_SAFE)	\
+		     _ASM_EXTABLE_TYPE(1b, 2b, EX_TYPE_FAULT_MCE_SAFE)	\
 		     : [err] "=a" (err), output				\
 		     : "0"(0), input);					\
 	err;								\
@@ -198,18 +197,14 @@ static inline void fxsave(struct fxregs_
 #define XRSTORS		".byte " REX_PREFIX "0x0f,0xc7,0x1f"
 
 /*
- * After this @err contains 0 on success or the negated trap number when
- * the operation raises an exception. For faults this results in -EFAULT.
+ * After this @err contains 0 on success or the trap number when the
+ * operation raises an exception.
  */
 #define XSTATE_OP(op, st, lmask, hmask, err)				\
 	asm volatile("1:" op "\n\t"					\
 		     "xor %[err], %[err]\n"				\
 		     "2:\n\t"						\
-		     ".pushsection .fixup,\"ax\"\n\t"			\
-		     "3: negl %%eax\n\t"				\
-		     "jmp 2b\n\t"					\
-		     ".popsection\n\t"					\
-		     _ASM_EXTABLE_TYPE(1b, 3b, EX_TYPE_FAULT_MCE_SAFE)	\
+		     _ASM_EXTABLE_TYPE(1b, 2b, EX_TYPE_FAULT_MCE_SAFE)	\
 		     : [err] "=a" (err)					\
 		     : "D" (st), "m" (*st), "a" (lmask), "d" (hmask)	\
 		     : "memory")
--- a/arch/x86/kernel/fpu/signal.c
+++ b/arch/x86/kernel/fpu/signal.c
@@ -13,6 +13,7 @@
 #include <asm/fpu/xstate.h>
 
 #include <asm/sigframe.h>
+#include <asm/trapnr.h>
 #include <asm/trace/fpu.h>
 
 static struct _fpx_sw_bytes fx_sw_reserved __ro_after_init;
@@ -275,7 +276,7 @@ static int restore_fpregs_from_user(void
 		fpregs_unlock();
 
 		/* Try to handle #PF, but anything else is fatal. */
-		if (ret != -EFAULT)
+		if (ret != X86_TRAP_PF)
 			return -EINVAL;
 
 		ret = fault_in_pages_readable(buf, size);
@@ -405,7 +406,7 @@ static int __fpu_restore_sig(void __user
 		u64 mask = user_xfeatures | xfeatures_mask_supervisor();
 
 		fpu->state.xsave.header.xfeatures &= mask;
-		ret = os_xrstor_safe(&fpu->state.xsave, xfeatures_mask_all);
+		ret = os_xrstor_safe(&fpu->state.xsave, xfeatures_mask_all) ? -EINVAL : 0;
 	} else {
 		ret = fxrstor_safe(&fpu->state.fxsave);
 	}


  parent reply	other threads:[~2021-09-07 20:25 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-07 20:24 [patch V2.1 00/20] x86/fpu: Clean up exception fixups and error handling in sigframe related code Thomas Gleixner
2021-09-07 20:24 ` [patch V2.1 01/20] x86/extable: Tidy up redundant handler functions Thomas Gleixner
2021-09-07 20:24 ` [patch V2.1 02/20] x86/extable: Get rid of redundant macros Thomas Gleixner
2021-09-07 20:24 ` [patch V2.1 03/20] x86/mce: Deduplicate exception handling Thomas Gleixner
2021-09-07 20:24 ` [patch V2.1 04/20] x86/mce: Get rid of stray semicolons Thomas Gleixner
2021-09-07 20:24 ` [patch V2.1 05/20] x86/extable: Rework the exception table mechanics Thomas Gleixner
2021-09-07 20:37   ` Linus Torvalds
2021-09-07 21:44     ` Thomas Gleixner
2021-09-07 20:24 ` [patch V2.1 06/20] x86/extable: Provide EX_TYPE_DEFAULT_MCE_SAFE and EX_TYPE_FAULT_MCE_SAFE Thomas Gleixner
2021-09-07 20:24 ` [patch V2.1 07/20] x86/copy_mc: Use EX_TYPE_DEFAULT_MCE_SAFE for exception fixups Thomas Gleixner
2021-09-07 20:24 ` [patch V2.1 08/20] x86/fpu: Use EX_TYPE_FAULT_MCE_SAFE " Thomas Gleixner
2021-09-07 20:24 ` [patch V2.1 09/20] x86/extable: Remove EX_TYPE_FAULT from MCE safe fixups Thomas Gleixner
2021-09-07 20:24 ` Thomas Gleixner [this message]
2021-09-07 20:24 ` [patch V2.1 11/20] x86/fpu: Dont use MCE safe fixups for writing FPU state to user space Thomas Gleixner
2021-09-07 20:24 ` [patch V2.1 12/20] x86/fpu/signal: Move header zeroing out of xsave_to_user_sigframe() Thomas Gleixner
2021-09-07 20:25 ` [patch V2.1 13/20] x86/fpu/signal: Move xstate clearing out of copy_fpregs_to_sigframe() Thomas Gleixner
2021-09-07 20:25 ` [patch V2.1 14/20] x86/fpu/signal: Change return type of copy_fpstate_to_sigframe() to boolean Thomas Gleixner
2021-09-07 20:25 ` [patch V2.1 15/20] x86/fpu/signal: Change return type of copy_fpregs_to_sigframe() helpers " Thomas Gleixner
2021-09-07 20:25 ` [patch V2.1 16/20] x86/signal: Change return type of restore_sigcontext() " Thomas Gleixner
2021-09-07 20:25 ` [patch V2.1 17/20] x86/fpu/signal: Change return type of fpu__restore_sig() " Thomas Gleixner
2021-09-07 20:25 ` [patch V2.1 18/20] x86/fpu/signal: Change return type of __fpu_restore_sig() " Thomas Gleixner
2021-09-07 20:25 ` [patch V2.1 19/20] x86/fpu/signal: Change return code of check_xstate_in_sigframe() " Thomas Gleixner
2021-09-07 20:25 ` [patch V2.1 20/20] x86/fpu/signal: Change return code of restore_fpregs_from_user() " Thomas Gleixner

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=20210907200848.818578802@linutronix.de \
    --to=tglx@linutronix.de \
    --cc=ast@kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=peterz@infradead.org \
    --cc=songliubraving@fb.com \
    --cc=tony.luck@intel.com \
    --cc=torvalds@linux-foundation.org \
    --cc=viro@zeniv.linux.org.uk \
    --cc=x86@kernel.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox