[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20180524200350.GG17342@gate.crashing.org>
Date: Thu, 24 May 2018 15:03:50 -0500
From: Segher Boessenkool <segher@...nel.crashing.org>
To: Christophe Leroy <christophe.leroy@....fr>
Cc: Benjamin Herrenschmidt <benh@...nel.crashing.org>,
Paul Mackerras <paulus@...ba.org>,
Michael Ellerman <mpe@...erman.id.au>,
linux-kernel@...r.kernel.org, linuxppc-dev@...ts.ozlabs.org
Subject: Re: [PATCH v4] powerpc: Implement csum_ipv6_magic in assembly
Looks fine to me (one comment below):
Reviewed-by: Segher Boessenkool <segher@...nel.crashing.org>
On Thu, May 24, 2018 at 11:33:18AM +0000, Christophe Leroy wrote:
> +_GLOBAL(csum_ipv6_magic)
> + lwz r8, 0(r3)
> + lwz r9, 4(r3)
> + addc r0, r7, r8
> + lwz r10, 8(r3)
> + adde r0, r0, r9
> + lwz r11, 12(r3)
> + adde r0, r0, r10
> + lwz r8, 0(r4)
> + adde r0, r0, r11
> + lwz r9, 4(r4)
> + adde r0, r0, r8
> + lwz r10, 8(r4)
> + adde r0, r0, r9
> + lwz r11, 12(r4)
> + adde r0, r0, r10
> + add r5, r5, r6 /* assumption: len + proto doesn't carry */
> + adde r0, r0, r11
> + adde r0, r0, r5
> + addze r0, r0
> + rotlwi r3, r0, 16
> + add r3, r0, r3
> + not r3, r3
> + rlwinm r3, r3, 16, 16, 31
> + blr
> +EXPORT_SYMBOL(csum_ipv6_magic)
> diff --git a/arch/powerpc/lib/checksum_64.S b/arch/powerpc/lib/checksum_64.S
> index d7f1a966136e..bf0546e546fc 100644
> --- a/arch/powerpc/lib/checksum_64.S
> +++ b/arch/powerpc/lib/checksum_64.S
> @@ -429,3 +429,31 @@ dstnr; stb r6,0(r4)
> stw r6,0(r8)
> blr
> EXPORT_SYMBOL(csum_partial_copy_generic)
> +
> +/*
> + * __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
> + * const struct in6_addr *daddr,
> + * __u32 len, __u8 proto, __wsum sum)
> + */
> +
> +_GLOBAL(csum_ipv6_magic)
> + ld r8, 0(r3)
> + ld r9, 8(r3)
> + add r5, r5, r6
> + addc r0, r8, r9
> + ld r10, 0(r4)
> + ld r11, 8(r4)
> + adde r0, r0, r10
> + add r5, r5, r7
> + adde r0, r0, r11
> + adde r0, r0, r5
> + addze r0, r0
> + rotldi r3 ,r0, 32 /* fold two 32 bit halves together */
Typo (s/ ,/, /).
> + add r3, r0, r3
> + srdi r0, r3, 32
> + rotlwi r3, r0, 16 /* fold two 16 bit halves together */
> + add r3, r0, r3
> + not r3, r3
> + rlwinm r3, r3, 16, 16, 31
> + blr
> +EXPORT_SYMBOL(csum_ipv6_magic)
> --
> 2.13.3
Powered by blists - more mailing lists