| Message ID | 20240822230635.954557-5-andrew.cooper3@citrix.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | xen/bitops: hweight() cleanup/improvements | expand |
On 23.08.2024 01:06, Andrew Cooper wrote:
> They are no more, and won't be returning in this form.
And what's the plan? Use hweight32((uint8_t)...) in an open-coded manner?
Not overly nice I would say.
Jan
On 26/08/2024 11:39 am, Jan Beulich wrote: > On 23.08.2024 01:06, Andrew Cooper wrote: >> They are no more, and won't be returning in this form. > And what's the plan? Use hweight32((uint8_t)...) in an open-coded manner? > Not overly nice I would say. If we ever regain a genuine need for the 8 or 16 forms, they can go back into bitops.h, in terms of hweightl(), just like hweight32(). But it's been 20 years so far and we haven't actually needed hweight8/16, and I'm expecting this to continue for the forseeable future. ~Andrew
On 27.08.2024 11:49, Andrew Cooper wrote: > On 26/08/2024 11:39 am, Jan Beulich wrote: >> On 23.08.2024 01:06, Andrew Cooper wrote: >>> They are no more, and won't be returning in this form. >> And what's the plan? Use hweight32((uint8_t)...) in an open-coded manner? >> Not overly nice I would say. > > If we ever regain a genuine need for the 8 or 16 forms, they can go back > into bitops.h, in terms of hweightl(), just like hweight32(). > > But it's been 20 years so far and we haven't actually needed > hweight8/16, and I'm expecting this to continue for the forseeable future. Well, I'm not fully convinced. People may (try to) add open-coded forms like in my earlier reply instead. But anyway: Acked-by: Jan Beulich <jbeulich@suse.com> Jan
On 27/08/2024 11:11 am, Jan Beulich wrote: > On 27.08.2024 11:49, Andrew Cooper wrote: >> On 26/08/2024 11:39 am, Jan Beulich wrote: >>> On 23.08.2024 01:06, Andrew Cooper wrote: >>>> They are no more, and won't be returning in this form. >>> And what's the plan? Use hweight32((uint8_t)...) in an open-coded manner? >>> Not overly nice I would say. >> If we ever regain a genuine need for the 8 or 16 forms, they can go back >> into bitops.h, in terms of hweightl(), just like hweight32(). >> >> But it's been 20 years so far and we haven't actually needed >> hweight8/16, and I'm expecting this to continue for the forseeable future. > Well, I'm not fully convinced. People may (try to) add open-coded forms like > in my earlier reply instead. I'd hope we'd spot that during review, and even if not, we can fix it up after the fact. > But anyway: > Acked-by: Jan Beulich <jbeulich@suse.com> Thanks. ~Andrew
diff --git a/xen/arch/arm/include/asm/bitops.h b/xen/arch/arm/include/asm/bitops.h index 3c023103f734..91cd167b6bbb 100644 --- a/xen/arch/arm/include/asm/bitops.h +++ b/xen/arch/arm/include/asm/bitops.h @@ -86,8 +86,6 @@ bool clear_mask16_timeout(uint16_t mask, volatile void *p, */ #define hweight64(x) generic_hweight64(x) #define hweight32(x) generic_hweight32(x) -#define hweight16(x) generic_hweight16(x) -#define hweight8(x) generic_hweight8(x) #endif /* _ARM_BITOPS_H */ /* diff --git a/xen/arch/ppc/include/asm/bitops.h b/xen/arch/ppc/include/asm/bitops.h index eb3355812ea3..a62c4f99c3bb 100644 --- a/xen/arch/ppc/include/asm/bitops.h +++ b/xen/arch/ppc/include/asm/bitops.h @@ -132,7 +132,5 @@ static inline int test_and_set_bit(unsigned int nr, volatile void *addr) */ #define hweight64(x) __builtin_popcountll(x) #define hweight32(x) __builtin_popcount(x) -#define hweight16(x) __builtin_popcount((uint16_t)(x)) -#define hweight8(x) __builtin_popcount((uint8_t)(x)) #endif /* _ASM_PPC_BITOPS_H */ diff --git a/xen/arch/x86/include/asm/bitops.h b/xen/arch/x86/include/asm/bitops.h index 8c0403405aa2..4c5b21907a64 100644 --- a/xen/arch/x86/include/asm/bitops.h +++ b/xen/arch/x86/include/asm/bitops.h @@ -483,7 +483,5 @@ static always_inline unsigned int arch_flsl(unsigned long x) */ #define hweight64(x) generic_hweight64(x) #define hweight32(x) generic_hweight32(x) -#define hweight16(x) generic_hweight16(x) -#define hweight8(x) generic_hweight8(x) #endif /* _X86_BITOPS_H */ diff --git a/xen/include/xen/bitops.h b/xen/include/xen/bitops.h index 74c0d04e6647..64d70a7a1cb5 100644 --- a/xen/include/xen/bitops.h +++ b/xen/include/xen/bitops.h @@ -371,23 +371,6 @@ static inline unsigned int generic_hweight32(unsigned int w) return (w + (w >> 16)) & 0xff; } -static inline unsigned int generic_hweight16(unsigned int w) -{ - w -= ((w >> 1) & 0x5555); - w = (w & 0x3333) + ((w >> 2) & 0x3333); - w = (w + (w >> 4)) & 0x0f0f; - - return (w + (w >> 8)) & 0xff; -} - -static inline unsigned int generic_hweight8(unsigned int w) -{ - w -= ((w >> 1) & 0x55); - w = (w & 0x33) + ((w >> 2) & 0x33); - - return (w + (w >> 4)) & 0x0f; -} - static inline unsigned int generic_hweight64(uint64_t w) { if ( BITS_PER_LONG < 64 )
They are no more, and won't be returning in this form. No functional change. Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com> --- CC: Jan Beulich <JBeulich@suse.com> CC: Roger Pau Monné <roger.pau@citrix.com> CC: Wei Liu <wl@xen.org> CC: Stefano Stabellini <sstabellini@kernel.org> CC: Julien Grall <julien@xen.org> CC: Volodymyr Babchuk <Volodymyr_Babchuk@epam.com> CC: Bertrand Marquis <bertrand.marquis@arm.com> CC: Michal Orzel <michal.orzel@amd.com> CC: Oleksii Kurochko <oleksii.kurochko@gmail.com> CC: Shawn Anastasio <sanastasio@raptorengineering.com> --- xen/arch/arm/include/asm/bitops.h | 2 -- xen/arch/ppc/include/asm/bitops.h | 2 -- xen/arch/x86/include/asm/bitops.h | 2 -- xen/include/xen/bitops.h | 17 ----------------- 4 files changed, 23 deletions(-)