netdev - [PATCH v2] x86/lib: Remove the special case for odd-aligned buffers in csum-partial

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <e2864e9c5d794c79aa7ee7de4abbfc6d@AcuMS.aculab.com>
Date:   Thu, 6 Jan 2022 14:45:41 +0000
From:   David Laight <David.Laight@...LAB.COM>
To:     'Eric Dumazet' <edumazet@...gle.com>,
        Peter Zijlstra <peterz@...radead.org>
CC:     "'tglx@...utronix.de'" <tglx@...utronix.de>,
        "'mingo@...hat.com'" <mingo@...hat.com>,
        'Borislav Petkov' <bp@...en8.de>,
        "'dave.hansen@...ux.intel.com'" <dave.hansen@...ux.intel.com>,
        'X86 ML' <x86@...nel.org>, "'hpa@...or.com'" <hpa@...or.com>,
        "'alexanderduyck@...com'" <alexanderduyck@...com>,
        'open list' <linux-kernel@...r.kernel.org>,
        'netdev' <netdev@...r.kernel.org>,
        "'Noah Goldstein'" <goldstein.w.n@...il.com>
Subject: [PATCH v2] x86/lib: Remove the special case for odd-aligned buffers
 in csum-partial_64.c

There is no need to special case the very unusual odd-aligned buffers.
They are no worse than 4n+2 aligned buffers.

Signed-off-by: David Laight <david.laight@...lab.com>
Acked-by: Eric Dumazet
---

resend - v1 seems to have got lost :-)

v2: Also delete from32to16()
    Add acked-by from Eric (he sent one at some point)
    Fix possible whitespace error in the last hunk.

The penalty for any misaligned access seems to be minimal.
On an i7-7700 misaligned buffers add 2 or 3 clocks (in 115) to a 512 byte
  checksum.
That is less than 1 clock for each cache line!
That is just measuring the main loop with an lfence prior to rdpmc to
read PERF_COUNT_HW_CPU_CYCLES.

 arch/x86/lib/csum-partial_64.c | 28 ++--------------------------
 1 file changed, 2 insertions(+), 26 deletions(-)

diff --git a/arch/x86/lib/csum-partial_64.c b/arch/x86/lib/csum-partial_64.c
index 1f8a8f895173..061b1ed74d6a 100644
--- a/arch/x86/lib/csum-partial_64.c
+++ b/arch/x86/lib/csum-partial_64.c
@@ -11,16 +11,6 @@
 #include <asm/checksum.h>
 #include <asm/word-at-a-time.h>
 
-static inline unsigned short from32to16(unsigned a) 
-{
-	unsigned short b = a >> 16; 
-	asm("addw %w2,%w0\n\t"
-	    "adcw $0,%w0\n" 
-	    : "=r" (b)
-	    : "0" (b), "r" (a));
-	return b;
-}
-
 /*
  * Do a checksum on an arbitrary memory area.
  * Returns a 32bit checksum.
@@ -30,22 +20,12 @@ static inline unsigned short from32to16(unsigned a)
  *
  * Still, with CHECKSUM_COMPLETE this is called to compute
  * checksums on IPv6 headers (40 bytes) and other small parts.
- * it's best to have buff aligned on a 64-bit boundary
+ * The penalty for misaligned buff is negligable.
  */
 __wsum csum_partial(const void *buff, int len, __wsum sum)
 {
 	u64 temp64 = (__force u64)sum;
-	unsigned odd, result;
-
-	odd = 1 & (unsigned long) buff;
-	if (unlikely(odd)) {
-		if (unlikely(len == 0))
-			return sum;
-		temp64 = ror32((__force u32)sum, 8);
-		temp64 += (*(unsigned char *)buff << 8);
-		len--;
-		buff++;
-	}
+	unsigned result;
 
 	while (unlikely(len >= 64)) {
 		asm("addq 0*8(%[src]),%[res]\n\t"
@@ -130,10 +110,6 @@ __wsum csum_partial(const void *buff, int len, __wsum sum)
 #endif
 	}
 	result = add32_with_carry(temp64 >> 32, temp64 & 0xffffffff);
-	if (unlikely(odd)) {
-		result = from32to16(result);
-		result = ((result >> 8) & 0xff) | ((result & 0xff) << 8);
-	}
 	return (__force __wsum)result;
 }
 EXPORT_SYMBOL(csum_partial);
-- 
2.17.1

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)