The code relies on constant folding of cpu_has_feature() based
on possible and always true values as defined per
CPU_FTRS_ALWAYS and CPU_FTRS_POSSIBLE.
Build failure is encountered with for instance
book3e_all_defconfig on kisskb in the AMDGPU driver which uses
cpu_has_feature(CPU_FTR_VSX_COMP) to decide whether calling
kernel_enable_vsx() or not.
The failure is due to cpu_has_feature() not being inlined with
that configuration with gcc 4.9.
In the same way as commit acdad8fb4a15 ("powerpc: Force inlining of
mmu_has_feature to fix build failure"), for inlining of
cpu_has_feature().
Signed-off-by: Christophe Leroy <[email protected]>
---
arch/powerpc/include/asm/cpu_has_feature.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/include/asm/cpu_has_feature.h b/arch/powerpc/include/asm/cpu_has_feature.h
index 7897d16e0990..727d4b321937 100644
--- a/arch/powerpc/include/asm/cpu_has_feature.h
+++ b/arch/powerpc/include/asm/cpu_has_feature.h
@@ -7,7 +7,7 @@
#include <linux/bug.h>
#include <asm/cputable.h>
-static inline bool early_cpu_has_feature(unsigned long feature)
+static __always_inline bool early_cpu_has_feature(unsigned long feature)
{
return !!((CPU_FTRS_ALWAYS & feature) ||
(CPU_FTRS_POSSIBLE & cur_cpu_spec->cpu_features & feature));
@@ -46,7 +46,7 @@ static __always_inline bool cpu_has_feature(unsigned long feature)
return static_branch_likely(&cpu_feature_keys[i]);
}
#else
-static inline bool cpu_has_feature(unsigned long feature)
+static __always_inline bool cpu_has_feature(unsigned long feature)
{
return early_cpu_has_feature(feature);
}
--
2.25.0
Le 10/03/2021 à 13:10, Christophe Leroy a écrit :
> The code relies on constant folding of cpu_has_feature() based
> on possible and always true values as defined per
> CPU_FTRS_ALWAYS and CPU_FTRS_POSSIBLE.
>
> Build failure is encountered with for instance
> book3e_all_defconfig on kisskb in the AMDGPU driver which uses
> cpu_has_feature(CPU_FTR_VSX_COMP) to decide whether calling
> kernel_enable_vsx() or not.
>
> The failure is due to cpu_has_feature() not being inlined with
> that configuration with gcc 4.9.
>
> In the same way as commit acdad8fb4a15 ("powerpc: Force inlining of
> mmu_has_feature to fix build failure"), for inlining of
> cpu_has_feature().
>
> Signed-off-by: Christophe Leroy <[email protected]>
Fixes following build
http://kisskb.ellerman.id.au/kisskb/buildresult/14489391/
> ---
> arch/powerpc/include/asm/cpu_has_feature.h | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/arch/powerpc/include/asm/cpu_has_feature.h b/arch/powerpc/include/asm/cpu_has_feature.h
> index 7897d16e0990..727d4b321937 100644
> --- a/arch/powerpc/include/asm/cpu_has_feature.h
> +++ b/arch/powerpc/include/asm/cpu_has_feature.h
> @@ -7,7 +7,7 @@
> #include <linux/bug.h>
> #include <asm/cputable.h>
>
> -static inline bool early_cpu_has_feature(unsigned long feature)
> +static __always_inline bool early_cpu_has_feature(unsigned long feature)
> {
> return !!((CPU_FTRS_ALWAYS & feature) ||
> (CPU_FTRS_POSSIBLE & cur_cpu_spec->cpu_features & feature));
> @@ -46,7 +46,7 @@ static __always_inline bool cpu_has_feature(unsigned long feature)
> return static_branch_likely(&cpu_feature_keys[i]);
> }
> #else
> -static inline bool cpu_has_feature(unsigned long feature)
> +static __always_inline bool cpu_has_feature(unsigned long feature)
> {
> return early_cpu_has_feature(feature);
> }
>
On Wed, 10 Mar 2021 12:10:34 +0000 (UTC), Christophe Leroy wrote:
> The code relies on constant folding of cpu_has_feature() based
> on possible and always true values as defined per
> CPU_FTRS_ALWAYS and CPU_FTRS_POSSIBLE.
>
> Build failure is encountered with for instance
> book3e_all_defconfig on kisskb in the AMDGPU driver which uses
> cpu_has_feature(CPU_FTR_VSX_COMP) to decide whether calling
> kernel_enable_vsx() or not.
>
> [...]
Applied to powerpc/fixes.
[1/1] powerpc: Force inlining of cpu_has_feature() to avoid build failure
https://git.kernel.org/powerpc/c/eed5fae00593ab9d261a0c1ffc1bdb786a87a55a
cheers