Skip to contentSkip to conversation

x86: Use xorl %0,%0 in __get_user_asm

PATCH Applied Uros Bizjak 4 participants 5 messages Aug 27, 2020 – Sep 3, 2020 x86

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 x86: Use xorl %0,%0 in __get_user_asm

xorl %0,%0 is equivalent to xorq %0,%0 as both will zero the entire register. Use xorl %0,%0 for all operand sizes to avoid REX prefix byte when legacy registers are used and to avoid size prefix byte when 16bit registers are used.

Zeroing the full register is OK in this use case. xorl %0,%0 also breaks register dependency chains, avoiding potential partial register stalls with 8 and 16bit operands.

The patch lowers the size of .fixup section by 20 bytes.

Changes since v1:
- Rewrite commit message.

Cc Thomas Gleixner <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Borislav Petkov <[email protected]>
Cc "H. Peter Anvin" <[email protected]>
---
 arch/x86/include/asm/uaccess.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/include/asm/uaccess.h +1 -1
index ecefaffd15d4..2bffba2a1b23 100644
@@ -343,7 +343,7 @@ do {									\
 		     "2:\n"						\
 		     ".section .fixup,\"ax\"\n"				\
 		     "3:	mov %[efault],%[errout]\n"		\
-		     "	xor"itype" %[output],%[output]\n"		\
+		     "	xorl %k[output],%k[output]\n"			\
 		     "	jmp 2b\n"					\
 		     ".previous\n"					\
 		     _ASM_EXTABLE_UA(1b, 3b)				\
Signature · 1 line
2.26.2
reviewer Reviewed-by on v2 1/1
#1 lore raw
Quoted 11 lines · Uros Bizjak

xorl %0,%0 is equivalent to xorq %0,%0 as both will zero the entire register. Use xorl %0,%0 for all operand sizes to avoid REX prefix byte when legacy registers are used and to avoid size prefix byte when 16bit registers are used.

Zeroing the full register is OK in this use case. xorl %0,%0 also breaks register dependency chains, avoiding potential partial register stalls with 8 and 16bit operands.

The patch lowers the size of .fixup section by 20 bytes.

Quoted 8 lines

Changes since v1:
- Rewrite commit message.

Cc Thomas Gleixner <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Borislav Petkov <[email protected]>
Cc "H. Peter Anvin" <[email protected]>
#2 lore raw
In reply to Uros Bizjak
Quoted 8 lines · Uros Bizjak

xorl %0,%0 is equivalent to xorq %0,%0 as both will zero the entire register. Use xorl %0,%0 for all operand sizes to avoid REX prefix byte when legacy registers are used and to avoid size prefix byte when 16bit registers are used.

Zeroing the full register is OK in this use case. xorl %0,%0 also breaks register dependency chains, avoiding potential partial register stalls with 8 and 16bit operands.

No objections, but talking about stalls is more than slightly ridiculous - we'd just taken a #PF, failed there, flipped pt_regs %rip to fixup section, returned from fault and are about to fail whatever syscall that had been; a stall here is really not an issue...

#3 lore raw
Quoted 11 lines · Al Viro
Quoted 8 lines · Uros Bizjak

xorl %0,%0 is equivalent to xorq %0,%0 as both will zero the entire register. Use xorl %0,%0 for all operand sizes to avoid REX prefix byte when legacy registers are used and to avoid size prefix byte when 16bit registers are used.

Zeroing the full register is OK in this use case. xorl %0,%0 also breaks register dependency chains, avoiding potential partial register stalls with 8 and 16bit operands.

Quoted 5 lines

No objections, but talking about stalls is more than slightly ridiculous - we'd just taken a #PF, failed there, flipped pt_regs %rip to fixup section, returned from fault and are about to fail whatever syscall that had been; a stall here is really not an issue...

Should I submit a v3 with the offending sentence removed, or could I just ask a committer to remove it on the fly?

Uros.

merge bot Applied
#4 lore raw

[tip: x86/asm] x86/uaccess: Use XORL %0,%0 in __get_user_asm()

In reply to Uros Bizjak
The following commit has been merged into the x86/asm branch of tip:

The following commit has been merged into the x86/asm branch of tip:

Commit-ID:     767ec7289e83721fee205a13b459f12fb2cf922f
Gitweb:        https://git.kernel.org/tip/767ec7289e83721fee205a13b459f12fb2cf922f
Author:        Uros Bizjak <[email protected]>
AuthorDate:    Thu, 27 Aug 2020 20:09:04 +02:00
Committer:     Borislav Petkov <[email protected]>
CommitterDate: Thu, 03 Sep 2020 22:49:03 +02:00

x86/uaccess: Use XORL %0,%0 in __get_user_asm()

XORL %0,%0 is equivalent to XORQ %0,%0 as both will zero the entire register. Use XORL %0,%0 for all operand sizes to avoid REX prefix byte when legacy registers are used and to avoid size prefix byte when 16bit registers are used.

Zeroing the full register is OK in this use case.

As a result, the size of the .fixup section decreases by 20 bytes.

[ bp: Massage commit message. ]

---
 arch/x86/include/asm/uaccess.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/include/asm/uaccess.h +1 -1
index ecefaff..2bffba2 100644
@@ -343,7 +343,7 @@ do {									\
 		     "2:\n"						\
 		     ".section .fixup,\"ax\"\n"				\
 		     "3:	mov %[efault],%[errout]\n"		\
-		     "	xor"itype" %[output],%[output]\n"		\
+		     "	xorl %k[output],%k[output]\n"			\
 		     "	jmp 2b\n"					\
 		     ".previous\n"					\
 		     _ASM_EXTABLE_UA(1b, 3b)				\