Re: [PATCH v2] arm64/prefetch: fix a -Wtype-limits warning

[Will Deacon]

On Mon, Aug 05, 2019 at 08:03:10AM -0400, Qian Cai wrote:
> 
> 
> > On Aug 5, 2019, at 6:00 AM, Will Deacon <will@xxxxxxxxxx> wrote:
> > 
> > On Fri, Aug 02, 2019 at 08:33:58PM -0400, Qian Cai wrote:
> >> The commit d5370f754875 ("arm64: prefetch: add alternative pattern for
> >> CPUs without a prefetcher") introduced MIDR_IS_CPU_MODEL_RANGE() to be
> >> used in has_no_hw_prefetch() with rv_min=0 which generates a compilation
> >> warning from GCC,
> >> 
> >> In file included from ./arch/arm64/include/asm/cache.h:8,
> >>                from ./include/linux/cache.h:6,
> >>                from ./include/linux/printk.h:9,
> >>                from ./include/linux/kernel.h:15,
> >>                from ./include/linux/cpumask.h:10,
> >>                from arch/arm64/kernel/cpufeature.c:11:
> >> arch/arm64/kernel/cpufeature.c: In function 'has_no_hw_prefetch':
> >> ./arch/arm64/include/asm/cputype.h:59:26: warning: comparison of
> >> unsigned expression >= 0 is always true [-Wtype-limits]
> >> _model == (model) && rv >= (rv_min) && rv <= (rv_max);  \
> >>                         ^~
> >> arch/arm64/kernel/cpufeature.c:889:9: note: in expansion of macro
> >> 'MIDR_IS_CPU_MODEL_RANGE'
> >> return MIDR_IS_CPU_MODEL_RANGE(midr, MIDR_THUNDERX,
> >>        ^~~~~~~~~~~~~~~~~~~~~~~
> >> 
> >> Fix it by making "rv" a "s32".
> >> 
> >> Signed-off-by: Qian Cai <cai@xxxxxx>
> >> ---
> >> 
> >> v2: Use "s32" for "rv", so "variant 0/revision 0" can be covered.
> >> 
> >> arch/arm64/include/asm/cputype.h | 2 +-
> >> 1 file changed, 1 insertion(+), 1 deletion(-)
> >> 
> >> diff --git a/arch/arm64/include/asm/cputype.h b/arch/arm64/include/asm/cputype.h
> >> index e7d46631cc42..d52fe8651c2d 100644
> >> --- a/arch/arm64/include/asm/cputype.h
> >> +++ b/arch/arm64/include/asm/cputype.h
> >> @@ -54,7 +54,7 @@
> >> #define MIDR_IS_CPU_MODEL_RANGE(midr, model, rv_min, rv_max)		\
> >> ({									\
> >> 	u32 _model = (midr) & MIDR_CPU_MODEL_MASK;			\
> >> -	u32 rv = (midr) & (MIDR_REVISION_MASK | MIDR_VARIANT_MASK);	\
> >> +	s32 rv = (midr) & (MIDR_REVISION_MASK | MIDR_VARIANT_MASK);	\
> > 
> > Hmm, but this really isn't a signed quantity: it's two fields extracted
> > from an ID register. I think the code is fine. Are you explicitly enabling
> > -Wtype-limits somehow?
> 
> Yes, it is useful to catch unintended developer mistakes or simply optimize wasted instructions of
> checking like in,
> 
> 919aef44d73d (âx86/efi: fix a -Wtype-limits compilation warningâ)
> 
> 5a82bdb48f04 (âx86/cacheinfo: Fix a -Wtype-limits warningâ)
> 
> It is normal to fix a false positive this way as in other mainline commits,
> 
> ec6335586953 (âx86/apic: Silence -Wtype-limits compiler warningsâ)
> 
> Once those false-positives are under control, the warning flag could be then enabled by default in
> the future.

If there's a way to fix the code without making it more confusing, sure,
but your proposal of making the field signed does not achieve that goal.

Will