[PATCH] powerpc: Reduce csum_add() complexity for PPC64

[PATCH] powerpc: Reduce csum_add() complexity for PPC64

[Christophe Leroy]

PPC64 does everything in C, gcc is able to skip calculation
when one of the operands in zero.

Move the constant folding in PPC32 part.

This helps GCC and reduces ppc64_defconfig by 170 bytes.

Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
---
 arch/powerpc/include/asm/checksum.h | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/arch/powerpc/include/asm/checksum.h b/arch/powerpc/include/asm/checksum.h
index 3288a1bf5e8d..e4e25b46ac49 100644
--- a/arch/powerpc/include/asm/checksum.h
+++ b/arch/powerpc/include/asm/checksum.h
@@ -95,16 +95,15 @@ static __always_inline __wsum csum_add(__wsum csum, __wsum addend)
 {
 #ifdef __powerpc64__
 	u64 res = (__force u64)csum;
-#endif
+
+	res += (__force u64)addend;
+	return (__force __wsum)((u32)res + (res >> 32));
+#else
 	if (__builtin_constant_p(csum) && csum == 0)
 		return addend;
 	if (__builtin_constant_p(addend) && addend == 0)
 		return csum;
 
-#ifdef __powerpc64__
-	res += (__force u64)addend;
-	return (__force __wsum)((u32)res + (res >> 32));
-#else
 	asm("addc %0,%0,%1;"
 	    "addze %0,%0;"
 	    : "+r" (csum) : "r" (addend) : "xer");
-- 
2.34.1

Re: [PATCH] powerpc: Reduce csum_add() complexity for PPC64

[Michael Ellerman]

On Sat, 12 Feb 2022 08:36:17 +0100, Christophe Leroy wrote:
> PPC64 does everything in C, gcc is able to skip calculation
> when one of the operands in zero.
> 
> Move the constant folding in PPC32 part.
> 
> This helps GCC and reduces ppc64_defconfig by 170 bytes.
> 
> [...]

Applied to powerpc/next.

[1/1] powerpc: Reduce csum_add() complexity for PPC64
      https://git.kernel.org/powerpc/c/f206fdd9d41bf7deb96219b8ca3499a5abd79b83

cheers