dri-devel.lists.freedesktop.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] i915: fix build error some more
@ 2025-06-20 11:18 Arnd Bergmann
  2025-06-24 21:11 ` Rodrigo Vivi
  0 siblings, 1 reply; 3+ messages in thread
From: Arnd Bergmann @ 2025-06-20 11:18 UTC (permalink / raw)
  To: Jani Nikula, Joonas Lahtinen, Rodrigo Vivi, Tvrtko Ursulin,
	David Airlie, Simona Vetter, Nathan Chancellor, Tzung-Bi Shih,
	Ashutosh Dixit, Umesh Nerlige Ramappa
  Cc: Arnd Bergmann, Nick Desaulniers, Bill Wendling, Justin Stitt,
	Lucas De Marchi, Matt Roper, Ville Syrjälä, Nam Cao,
	intel-gfx, dri-devel, linux-kernel, llvm

From: Arnd Bergmann <arnd@arndb.de>

An earlier patch fixed a build failure with clang, but I still see the
same problem with some configurations using gcc:

drivers/gpu/drm/i915/i915_pmu.c: In function 'config_mask':
include/linux/compiler_types.h:568:38: error: call to '__compiletime_assert_462' declared with attribute error: BUILD_BUG_ON failed: bit > BITS_PER_TYPE(typeof_member(struct i915_pmu, enable)) - 1
drivers/gpu/drm/i915/i915_pmu.c:116:3: note: in expansion of macro 'BUILD_BUG_ON'
  116 |   BUILD_BUG_ON(bit >

As I understand it, the problem is that the function is not always fully
inlined, but the __builtin_constant_p() can still evaluate the argument
as being constant.

Marking it as __always_inline so far works for me in all configurations.

Fixes: a7137b1825b5 ("drm/i915/pmu: Fix build error with GCOV and AutoFDO enabled")
Fixes: a644fde77ff7 ("drm/i915/pmu: Change bitmask of enabled events to u32")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 drivers/gpu/drm/i915/i915_pmu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
index 990bfaba3ce4..5bc696bfbb0f 100644
--- a/drivers/gpu/drm/i915/i915_pmu.c
+++ b/drivers/gpu/drm/i915/i915_pmu.c
@@ -108,7 +108,7 @@ static unsigned int config_bit(const u64 config)
 		return other_bit(config);
 }
 
-static u32 config_mask(const u64 config)
+static __always_inline u32 config_mask(const u64 config)
 {
 	unsigned int bit = config_bit(config);
 
-- 
2.39.5


^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH] i915: fix build error some more
  2025-06-20 11:18 [PATCH] i915: fix build error some more Arnd Bergmann
@ 2025-06-24 21:11 ` Rodrigo Vivi
  2025-06-24 21:24   ` Rodrigo Vivi
  0 siblings, 1 reply; 3+ messages in thread
From: Rodrigo Vivi @ 2025-06-24 21:11 UTC (permalink / raw)
  To: Arnd Bergmann
  Cc: Jani Nikula, Joonas Lahtinen, Tvrtko Ursulin, David Airlie,
	Simona Vetter, Nathan Chancellor, Tzung-Bi Shih, Ashutosh Dixit,
	Umesh Nerlige Ramappa, Arnd Bergmann, Nick Desaulniers,
	Bill Wendling, Justin Stitt, Lucas De Marchi, Matt Roper,
	Ville Syrjälä, Nam Cao, intel-gfx, dri-devel,
	linux-kernel, llvm

On Fri, Jun 20, 2025 at 01:18:18PM +0200, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> An earlier patch fixed a build failure with clang, but I still see the
> same problem with some configurations using gcc:
> 
> drivers/gpu/drm/i915/i915_pmu.c: In function 'config_mask':
> include/linux/compiler_types.h:568:38: error: call to '__compiletime_assert_462' declared with attribute error: BUILD_BUG_ON failed: bit > BITS_PER_TYPE(typeof_member(struct i915_pmu, enable)) - 1
> drivers/gpu/drm/i915/i915_pmu.c:116:3: note: in expansion of macro 'BUILD_BUG_ON'
>   116 |   BUILD_BUG_ON(bit >
> 
> As I understand it, the problem is that the function is not always fully
> inlined, but the __builtin_constant_p() can still evaluate the argument
> as being constant.
> 
> Marking it as __always_inline so far works for me in all configurations.
> 
> Fixes: a7137b1825b5 ("drm/i915/pmu: Fix build error with GCOV and AutoFDO enabled")
> Fixes: a644fde77ff7 ("drm/i915/pmu: Change bitmask of enabled events to u32")
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>  drivers/gpu/drm/i915/i915_pmu.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
> index 990bfaba3ce4..5bc696bfbb0f 100644
> --- a/drivers/gpu/drm/i915/i915_pmu.c
> +++ b/drivers/gpu/drm/i915/i915_pmu.c
> @@ -108,7 +108,7 @@ static unsigned int config_bit(const u64 config)
>  		return other_bit(config);
>  }
>  
> -static u32 config_mask(const u64 config)
> +static __always_inline u32 config_mask(const u64 config)

hmm, it looks like the __builtin_constant_p() with BUILD_BUG_ON is a bit fragile
design down here. But this always_inline will enforce deterministic behavior
indeed.

So,
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

pushing to drm-intel-next right now as well...

>  {
>  	unsigned int bit = config_bit(config);
>  
> -- 
> 2.39.5
> 

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH] i915: fix build error some more
  2025-06-24 21:11 ` Rodrigo Vivi
@ 2025-06-24 21:24   ` Rodrigo Vivi
  0 siblings, 0 replies; 3+ messages in thread
From: Rodrigo Vivi @ 2025-06-24 21:24 UTC (permalink / raw)
  To: Arnd Bergmann
  Cc: Jani Nikula, Joonas Lahtinen, Tvrtko Ursulin, David Airlie,
	Simona Vetter, Nathan Chancellor, Tzung-Bi Shih, Ashutosh Dixit,
	Umesh Nerlige Ramappa, Arnd Bergmann, Nick Desaulniers,
	Bill Wendling, Justin Stitt, Lucas De Marchi, Matt Roper,
	Ville Syrjälä, Nam Cao, intel-gfx, dri-devel,
	linux-kernel, llvm

On Tue, Jun 24, 2025 at 05:11:50PM -0400, Rodrigo Vivi wrote:
> On Fri, Jun 20, 2025 at 01:18:18PM +0200, Arnd Bergmann wrote:

While merging this I adjusted the subject to be drm/i915

> > From: Arnd Bergmann <arnd@arndb.de>
> > 
> > An earlier patch fixed a build failure with clang, but I still see the
> > same problem with some configurations using gcc:
> > 
> > drivers/gpu/drm/i915/i915_pmu.c: In function 'config_mask':
> > include/linux/compiler_types.h:568:38: error: call to '__compiletime_assert_462' declared with attribute error: BUILD_BUG_ON failed: bit > BITS_PER_TYPE(typeof_member(struct i915_pmu, enable)) - 1
> > drivers/gpu/drm/i915/i915_pmu.c:116:3: note: in expansion of macro 'BUILD_BUG_ON'
> >   116 |   BUILD_BUG_ON(bit >
> > 
> > As I understand it, the problem is that the function is not always fully
> > inlined, but the __builtin_constant_p() can still evaluate the argument
> > as being constant.
> > 
> > Marking it as __always_inline so far works for me in all configurations.
> > 
> > Fixes: a7137b1825b5 ("drm/i915/pmu: Fix build error with GCOV and AutoFDO enabled")

and to point to the original right patch instead of the cherry-picked fixes.

> > Fixes: a644fde77ff7 ("drm/i915/pmu: Change bitmask of enabled events to u32")

And then I pushed to drm-intel-gt-next becasue of the mention on this patch

> > Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> > ---
> >  drivers/gpu/drm/i915/i915_pmu.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
> > index 990bfaba3ce4..5bc696bfbb0f 100644
> > --- a/drivers/gpu/drm/i915/i915_pmu.c
> > +++ b/drivers/gpu/drm/i915/i915_pmu.c
> > @@ -108,7 +108,7 @@ static unsigned int config_bit(const u64 config)
> >  		return other_bit(config);
> >  }
> >  
> > -static u32 config_mask(const u64 config)
> > +static __always_inline u32 config_mask(const u64 config)
> 
> hmm, it looks like the __builtin_constant_p() with BUILD_BUG_ON is a bit fragile
> design down here. But this always_inline will enforce deterministic behavior
> indeed.
> 
> So,
> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
> 
> pushing to drm-intel-next right now as well...
> 
> >  {
> >  	unsigned int bit = config_bit(config);
> >  
> > -- 
> > 2.39.5
> > 

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2025-06-24 21:25 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-06-20 11:18 [PATCH] i915: fix build error some more Arnd Bergmann
2025-06-24 21:11 ` Rodrigo Vivi
2025-06-24 21:24   ` Rodrigo Vivi

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).