[PATCH -tip v2] x86/idle: Work around LLVM assembler bug with MONITOR and MWAIT insn

[PATCH -tip v2] x86/idle: Work around LLVM assembler bug with MONITOR and MWAIT insn

[Uros Bizjak]

LLVM assembler is not able to assemble correct forms of MONITOR
and MWAIT instructions with explicit operands:

  error: invalid operand for instruction
          monitor %rax,%ecx,%edx
                       ^~~~

Use instruction mnemonics with implicit operands to
work around this issue.

Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Fixes: cd3b85b27542 ("x86/idle: Use MONITOR and MWAIT mnemonics in <asm/mwait.h>")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202504030802.2lEVBSpN-lkp@intel.com/
---
v2: Use instruction mnemonics with implicit operands.
---
 arch/x86/include/asm/mwait.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/x86/include/asm/mwait.h b/arch/x86/include/asm/mwait.h
index 5141d2acc0ef..8f2cfc9cc3d8 100644
--- a/arch/x86/include/asm/mwait.h
+++ b/arch/x86/include/asm/mwait.h
@@ -27,7 +27,7 @@
 
 static __always_inline void __monitor(const void *eax, u32 ecx, u32 edx)
 {
-	asm volatile("monitor %0, %1, %2" :: "a" (eax), "c" (ecx), "d" (edx));
+	asm volatile("monitor" :: "a" (eax), "c" (ecx), "d" (edx));
 }
 
 static __always_inline void __monitorx(const void *eax, u32 ecx, u32 edx)
@@ -41,7 +41,7 @@ static __always_inline void __mwait(u32 eax, u32 ecx)
 {
 	mds_idle_clear_cpu_buffers();
 
-	asm volatile("mwait %0, %1" :: "a" (eax), "c" (ecx));
+	asm volatile("mwait" :: "a" (eax), "c" (ecx));
 }
 
 /*
@@ -92,7 +92,7 @@ static __always_inline void __sti_mwait(u32 eax, u32 ecx)
 {
 	mds_idle_clear_cpu_buffers();
 
-	asm volatile("sti; mwait %0, %1" :: "a" (eax), "c" (ecx));
+	asm volatile("sti; mwait" :: "a" (eax), "c" (ecx));
 }
 
 /*
-- 
2.49.0

Re: [PATCH -tip v2] x86/idle: Work around LLVM assembler bug with MONITOR and MWAIT insn

[Ingo Molnar]

* Uros Bizjak <ubizjak@gmail.com> wrote:

> LLVM assembler is not able to assemble correct forms of MONITOR
> and MWAIT instructions with explicit operands:
> 
>   error: invalid operand for instruction
>           monitor %rax,%ecx,%edx
>                        ^~~~
> 
> Use instruction mnemonics with implicit operands to
> work around this issue.
> 
> Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Ingo Molnar <mingo@kernel.org>
> Cc: Borislav Petkov <bp@alien8.de>
> Cc: Dave Hansen <dave.hansen@linux.intel.com>
> Cc: "H. Peter Anvin" <hpa@zytor.com>
> Fixes: cd3b85b27542 ("x86/idle: Use MONITOR and MWAIT mnemonics in <asm/mwait.h>")
> Reported-by: kernel test robot <lkp@intel.com>

So I've zapped cd3b85b27542 instead - let's re-try it again and
see if there's any code generation tradeoffs vs. the byte encodings?

Thanks,

	Ingo

Re: [PATCH -tip v2] x86/idle: Work around LLVM assembler bug with MONITOR and MWAIT insn

[Borislav Petkov]

On Thu, Apr 03, 2025 at 01:20:33PM +0200, Ingo Molnar wrote:
> So I've zapped cd3b85b27542 instead - let's re-try it again and
> see if there's any code generation tradeoffs vs. the byte encodings?

Please.

And give the test bots some time to chew on them before committing and/or
build-test with clang too.

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette