[tip:x86/urgent 1/1] arch/x86/coco/sev/core.c:2170:30: warning: variable 'dummy' set but not used

[tip:x86/urgent 1/1] arch/x86/coco/sev/core.c:2170:30: warning: variable 'dummy' set but not used

[kernel test robot]

tree:   https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git x86/urgent
head:   4a35d2b5254af89595fd90dae9ee0c8f990a148d
commit: 4a35d2b5254af89595fd90dae9ee0c8f990a148d [1/1] x86/sev: Use TSC_FACTOR for Secure TSC frequency calculation
config: x86_64-allyesconfig (https://download.01.org/0day-ci/archive/20250701/202507010218.3O5Ge0Xt-lkp@intel.com/config)
compiler: clang version 20.1.7 (https://github.com/llvm/llvm-project 6146a88f60492b520a36f8f8f3231e15f3cc6082)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20250701/202507010218.3O5Ge0Xt-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202507010218.3O5Ge0Xt-lkp@intel.com/

All warnings (new ones prefixed by >>):

>> arch/x86/coco/sev/core.c:2170:30: warning: variable 'dummy' set but not used [-Wunused-but-set-variable]
    2170 |         unsigned long tsc_freq_mhz, dummy;
         |                                     ^
   1 warning generated.


vim +/dummy +2170 arch/x86/coco/sev/core.c

  2167	
  2168	void __init snp_secure_tsc_init(void)
  2169	{
> 2170		unsigned long tsc_freq_mhz, dummy;

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

Re: [tip:x86/urgent 1/1] arch/x86/coco/sev/core.c:2170:30: warning: variable 'dummy' set but not used

[Borislav Petkov]

On Tue, Jul 01, 2025 at 02:43:31AM +0800, kernel test robot wrote:
> tree:   https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git x86/urgent
> head:   4a35d2b5254af89595fd90dae9ee0c8f990a148d
> commit: 4a35d2b5254af89595fd90dae9ee0c8f990a148d [1/1] x86/sev: Use TSC_FACTOR for Secure TSC frequency calculation
> config: x86_64-allyesconfig (https://download.01.org/0day-ci/archive/20250701/202507010218.3O5Ge0Xt-lkp@intel.com/config)
> compiler: clang version 20.1.7 (https://github.com/llvm/llvm-project 6146a88f60492b520a36f8f8f3231e15f3cc6082)

Pff, doesn't fire with clang-19.

That damn compiler.

gcc used to complain about those but they fixed their interprocedural analysis
or whatnot.

clang is simply complaining more.

/facepalm 1

And the 0day bot - because it doesn't have anything better to do - is doing
W=1 builds. I did complain about that waste of resources in the past but
nothing happened.

/facepalm 2

> reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20250701/202507010218.3O5Ge0Xt-lkp@intel.com/reproduce)
> 
> If you fix the issue in a separate patch/commit (i.e. not just a new version of
> the same patch/commit), kindly add following tags
> | Reported-by: kernel test robot <lkp@intel.com>
> | Closes: https://lore.kernel.org/oe-kbuild-all/202507010218.3O5Ge0Xt-lkp@intel.com/
> 
> All warnings (new ones prefixed by >>):
> 
> >> arch/x86/coco/sev/core.c:2170:30: warning: variable 'dummy' set but not used [-Wunused-but-set-variable]
>     2170 |         unsigned long tsc_freq_mhz, dummy;
>          |                                     ^
>    1 warning generated.
> 
> 
> vim +/dummy +2170 arch/x86/coco/sev/core.c
> 
>   2167	
>   2168	void __init snp_secure_tsc_init(void)
>   2169	{
> > 2170		unsigned long tsc_freq_mhz, dummy;

And it is actually my damn fault because even if that dummy crap variable was
bugging me, I thought we want to "harmonize".

/facepalm 3

Which is a total nonsense because that code doesn't build in 32-bit.

So this should've stayed like this:

diff --git a/arch/x86/coco/sev/core.c b/arch/x86/coco/sev/core.c
index 47d10d9a28e9..36c167333d04 100644
--- a/arch/x86/coco/sev/core.c
+++ b/arch/x86/coco/sev/core.c
@@ -2167,8 +2167,8 @@ static unsigned long securetsc_get_tsc_khz(void)
 
 void __init snp_secure_tsc_init(void)
 {
-	unsigned long tsc_freq_mhz, dummy;
 	struct snp_secrets_page *secrets;
+	unsigned long tsc_freq_mhz;
 	void *mem;
 
 	if (!cc_platform_has(CC_ATTR_GUEST_SNP_SECURE_TSC))
@@ -2183,7 +2183,7 @@ void __init snp_secure_tsc_init(void)
 	secrets = (__force struct snp_secrets_page *)mem;
 
 	setup_force_cpu_cap(X86_FEATURE_TSC_KNOWN_FREQ);
-	rdmsr(MSR_AMD64_GUEST_TSC_FREQ, tsc_freq_mhz, dummy);
+	rdmsrq(MSR_AMD64_GUEST_TSC_FREQ, tsc_freq_mhz);
 	/* Extract the GUEST TSC MHZ from BIT[17:0], rest is reserved space */
 	tsc_freq_mhz = tsc_freq_mhz & GENMASK_ULL(17, 0);
 	snp_tsc_freq_khz = SNP_SCALE_TSC_FREQ(tsc_freq_mhz * 1000, secrets->tsc_factor);

What a f*cking mess. I don't have enough room on my face anymore from
facepalms.

Oh boy.

-- 
Regards/Gruss,
    Boris.

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

Re: [tip:x86/urgent 1/1] arch/x86/coco/sev/core.c:2170:30: warning: variable 'dummy' set but not used

[Nathan Chancellor]

On Mon, Jun 30, 2025 at 09:27:26PM +0200, Borislav Petkov wrote:
> On Tue, Jul 01, 2025 at 02:43:31AM +0800, kernel test robot wrote:
> > tree:   https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git x86/urgent
> > head:   4a35d2b5254af89595fd90dae9ee0c8f990a148d
> > commit: 4a35d2b5254af89595fd90dae9ee0c8f990a148d [1/1] x86/sev: Use TSC_FACTOR for Secure TSC frequency calculation
> > config: x86_64-allyesconfig (https://download.01.org/0day-ci/archive/20250701/202507010218.3O5Ge0Xt-lkp@intel.com/config)
> > compiler: clang version 20.1.7 (https://github.com/llvm/llvm-project 6146a88f60492b520a36f8f8f3231e15f3cc6082)
> 
> Pff, doesn't fire with clang-19.
> 
> That damn compiler.
> 
> gcc used to complain about those but they fixed their interprocedural analysis
> or whatnot.
> 
> clang is simply complaining more.

Citation needed. This reproduces with Clang 15 (earliest supported on
x86 now), Clang 19, and GCC 15 for me. Were you missing W=1?

  $ x86_64-linux-gcc --version | head -1
  x86_64-linux-gcc (GCC) 15.1.0

  $ curl -LSso .config https://download.01.org/0day-ci/archive/20250701/202507010218.3O5Ge0Xt-lkp@intel.com/config

  $ make -skj"$(nproc)" ARCH=x86_64 CROSS_COMPILE=x86_64-linux- W=1 olddefconfig arch/x86/coco/sev/core.o
  arch/x86/coco/sev/core.c: In function 'snp_secure_tsc_init':
  arch/x86/coco/sev/core.c:2170:37: warning: variable 'dummy' set but not used [-Wunused-but-set-variable]
   2170 |         unsigned long tsc_freq_mhz, dummy;
        |                                     ^~~~~

Cheers,
Nathan

Re: [tip:x86/urgent 1/1] arch/x86/coco/sev/core.c:2170:30: warning: variable 'dummy' set but not used

[Borislav Petkov]

On Tue, Jul 01, 2025 at 08:06:44AM -0700, Nathan Chancellor wrote:
> Citation needed. This reproduces with Clang 15 (earliest supported on
> x86 now), Clang 19, and GCC 15 for me. Were you missing W=1?

Nope.

$ git checkout -b test 4a35d2b5254af89595fd90dae9ee0c8f990a148d
Switched to a new branch 'test'
$ make CC=clang-19 HOSTCC=clang-19 W=1 arch/x86/coco/sev/core.o
...
  LINK    /mnt/kernel/kernel/linux/tools/objtool/objtool
  CC      arch/x86/coco/sev/core.o
$ clang-19 --version
Debian clang version 19.1.7 (1+b1)
Target: x86_64-pc-linux-gnu
Thread model: posix
InstalledDir: /usr/lib/llvm-19/bin

-- 
Regards/Gruss,
    Boris.

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

Re: [tip:x86/urgent 1/1] arch/x86/coco/sev/core.c:2170:30: warning: variable 'dummy' set but not used

[Nathan Chancellor]

On Tue, Jul 01, 2025 at 05:29:53PM +0200, Borislav Petkov wrote:
> On Tue, Jul 01, 2025 at 08:06:44AM -0700, Nathan Chancellor wrote:
> > Citation needed. This reproduces with Clang 15 (earliest supported on
> > x86 now), Clang 19, and GCC 15 for me. Were you missing W=1?
> 
> Nope.
> 
> $ git checkout -b test 4a35d2b5254af89595fd90dae9ee0c8f990a148d
> Switched to a new branch 'test'
> $ make CC=clang-19 HOSTCC=clang-19 W=1 arch/x86/coco/sev/core.o
> ...
>   LINK    /mnt/kernel/kernel/linux/tools/objtool/objtool
>   CC      arch/x86/coco/sev/core.o
> $ clang-19 --version
> Debian clang version 19.1.7 (1+b1)
> Target: x86_64-pc-linux-gnu
> Thread model: posix
> InstalledDir: /usr/lib/llvm-19/bin

Spooky, same SHA, build command, and configuration from this thread with

  Debian clang version 19.1.7 (3+b1)
  Target: x86_64-pc-linux-gnu
  Thread model: posix
  InstalledDir: /usr/lib/llvm-19/bin

from unstable Debian (since it is what I base on for my Debian
development container) shows it for me... Can't imagine the two Debian
patch levels would have caused anything here but it is probably not
worth thinking too much more about since it is already resolved.

Cheers,
Nathan