linux-kernel - [PATCH -tip] x86/fpu: Rewrite XSTATE_{XSAVE,XRESTORE} macros as static inline functions

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <20250317093408.107282-1-ubizjak@gmail.com>
Date: Mon, 17 Mar 2025 10:28:38 +0100
From: Uros Bizjak <ubizjak@...il.com>
To: x86@...nel.org,
	linux-kernel@...r.kernel.org
Cc: Uros Bizjak <ubizjak@...il.com>,
	Thomas Gleixner <tglx@...utronix.de>,
	Ingo Molnar <mingo@...nel.org>,
	Borislav Petkov <bp@...en8.de>,
	Dave Hansen <dave.hansen@...ux.intel.com>,
	"H. Peter Anvin" <hpa@...or.com>
Subject: [PATCH -tip] x86/fpu: Rewrite XSTATE_{XSAVE,XRESTORE} macros as static inline functions

Rewrite XSTATE_{XSAVE,XRESTORE} macros as static inline functions
in order to use asm goto to optimize exception handling.

The code that used XSTATE_XSAVE improves from:

   3f1fe:	49 0f ae 64 24 40    	xsave64 0x40(%r12)
   3f204:	31 ed                	xor    %ebp,%ebp
   3f206:	85 ed                	test   %ebp,%ebp
   3f208:	75 37                	jne    3f241 <...>

to just:

   3f201:	48 0f ae 65 40       	xsave64 0x40(%rbp)

because there is no need to set and test temporary result
variable anymore.

bloat-o-meter reports a small code size improvement
(x86_64 defconfig, gcc-14.2.1):

	add/remove: 0/0 grow/shrink: 0/2 up/down: 0/-25 (-25)

	Function                             old     new   delta
	--------------------------------------------------------
	save_fpregs_to_fpstate               141     135      -6
	__fpu_restore_sig                   1448    1429     -19

	Total: Before=22809695, After=22809670, chg -0.00%

Signed-off-by: Uros Bizjak <ubizjak@...il.com>
Cc: Thomas Gleixner <tglx@...utronix.de>
Cc: Ingo Molnar <mingo@...nel.org>
Cc: Borislav Petkov <bp@...en8.de>
Cc: Dave Hansen <dave.hansen@...ux.intel.com>
Cc: "H. Peter Anvin" <hpa@...or.com>
---
 arch/x86/kernel/fpu/xstate.h | 54 ++++++++++++++++++++----------------
 1 file changed, 30 insertions(+), 24 deletions(-)

diff --git a/arch/x86/kernel/fpu/xstate.h b/arch/x86/kernel/fpu/xstate.h
index 1418423bc4c9..c8e31005567a 100644
--- a/arch/x86/kernel/fpu/xstate.h
+++ b/arch/x86/kernel/fpu/xstate.h
@@ -131,32 +131,38 @@ static inline int update_pkru_in_sigframe(struct xregs_state __user *buf, u64 ma
  *
  * Use XSAVE as a fallback.
  */
-#define XSTATE_XSAVE(st, lmask, hmask, err)				\
-	asm volatile("1: " ALTERNATIVE_3(XSAVE,				\
-				   XSAVEOPT, X86_FEATURE_XSAVEOPT,	\
-				   XSAVEC,   X86_FEATURE_XSAVEC,	\
-				   XSAVES,   X86_FEATURE_XSAVES)	\
-		     "\n\t"						\
-		     "xor %[err], %[err]\n"				\
-		     "3:\n"						\
-		     _ASM_EXTABLE_TYPE_REG(1b, 3b, EX_TYPE_EFAULT_REG, %[err]) \
-		     : [err] "=r" (err)					\
-		     : [xa] "m" (*(st)), "a" (lmask), "d" (hmask)	\
-		     : "memory")
+static __always_inline int __xstate_xsave(struct xregs_state *st,
+					  u32 lmask, u32 hmask)
+{
+	asm goto("1: " ALTERNATIVE_3(XSAVE,
+				     XSAVEOPT, X86_FEATURE_XSAVEOPT,
+				     XSAVEC,   X86_FEATURE_XSAVEC,
+				     XSAVES,   X86_FEATURE_XSAVES)
+		     _ASM_EXTABLE(1b, %l[fault])
+		     :
+		     : [xa] "m" (*st), "a" (lmask), "d" (hmask)
+		     : "memory"
+		     : fault);
+	return 0;
+fault:
+	return -EFAULT;
+}
 
 /*
  * Use XRSTORS to restore context if it is enabled. XRSTORS supports compact
  * XSAVE area format.
  */
-#define XSTATE_XRESTORE(st, lmask, hmask)				\
-	asm volatile("1: " ALTERNATIVE(XRSTOR,				\
-				 XRSTORS, X86_FEATURE_XSAVES)		\
-		     "\n"						\
-		     "3:\n"						\
-		     _ASM_EXTABLE_TYPE(1b, 3b, EX_TYPE_FPU_RESTORE)	\
-		     :							\
-		     : [xa] "m" (*(st)), "a" (lmask), "d" (hmask)	\
-		     : "memory")
+static __always_inline void __xstate_xrstor(struct xregs_state *st,
+					    u32 lmask, u32 hmask)
+{
+	asm volatile("1: " ALTERNATIVE(XRSTOR,
+				       XRSTORS, X86_FEATURE_XSAVES)
+		     "3:\n"
+		     _ASM_EXTABLE_TYPE(1b, 3b, EX_TYPE_FPU_RESTORE)
+		     :
+		     : [xa] "m" (*st), "a" (lmask), "d" (hmask)
+		     : "memory");
+}
 
 #if defined(CONFIG_X86_64) && defined(CONFIG_X86_DEBUG_FPU)
 extern void xfd_validate_state(struct fpstate *fpstate, u64 mask, bool rstor);
@@ -208,7 +214,7 @@ static inline void os_xsave(struct fpstate *fpstate)
 	WARN_ON_FPU(!alternatives_patched);
 	xfd_validate_state(fpstate, mask, false);
 
-	XSTATE_XSAVE(&fpstate->regs.xsave, lmask, hmask, err);
+	err = __xstate_xsave(&fpstate->regs.xsave, lmask, hmask);
 
 	/* We should never fault when copying to a kernel buffer: */
 	WARN_ON_FPU(err);
@@ -225,7 +231,7 @@ static inline void os_xrstor(struct fpstate *fpstate, u64 mask)
 	u32 hmask = mask >> 32;
 
 	xfd_validate_state(fpstate, mask, true);
-	XSTATE_XRESTORE(&fpstate->regs.xsave, lmask, hmask);
+	__xstate_xrstor(&fpstate->regs.xsave, lmask, hmask);
 }
 
 /* Restore of supervisor state. Does not require XFD */
@@ -235,7 +241,7 @@ static inline void os_xrstor_supervisor(struct fpstate *fpstate)
 	u32 lmask = mask;
 	u32 hmask = mask >> 32;
 
-	XSTATE_XRESTORE(&fpstate->regs.xsave, lmask, hmask);
+	__xstate_xrstor(&fpstate->regs.xsave, lmask, hmask);
 }
 
 /*
-- 
2.48.1