linux.git/include/asm-mips/checksum.h, branch master Linux kernel source tree MIPS: Move headfiles to new location below arch/mips/include 2008-10-11T15:18:52+00:00 Ralf Baechle ralf@linux-mips.org 2008-09-16T17:48:51+00:00 384740dc49ea651ba350704d13ff6be9976e37fe Signed-off-by: Ralf Baechle <ralf@linux-mips.org> 
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
[MIPS] Fix wrong checksum for split TCP packets on 64-bit MIPS 2007-04-20T13:58:37+00:00 Dave Johnson djohnson+linux-mips@sw.starentnetworks.com 2007-04-18T14:39:41+00:00 1d464c26b5625215c4b35fb336c8f3c57d248c2e I've traced down an off-by-one TCP checksum calculation error under the following conditions: 1) The TCP code needs to split a full-sized packet due to a reduced MSS (typically due to the addition of TCP options mid-stream like SACK). _AND_ 2) The checksum of the 2nd fragment is larger than the checksum of the original packet. After subtraction this results in a checksum for the 1st fragment with bits 16..31 set to 1. (this is ok) _AND_ 3) The checksum of the 1st fragment's TCP header plus the previously 32bit checksum of the 1st fragment DOES NOT cause a 32bit overflow when added together. This results in a checksum of the TCP header plus TCP data that still has the upper 16 bits as 1's. _THEN_ 4) The TCP+data checksum is added to the checksum of the pseudo IP header with csum_tcpudp_nofold() incorrectly (the bug). The problem is the checksum of the TCP+data is passed to csum_tcpudp_nofold() as an 32bit unsigned value, however the assembly code acts on it as if it is a 64bit unsigned value. This causes an incorrect 32->64bit extension if the sum has bit 31 set. The resulting checksum is off by one. This problems is data and TCP header dependent due to #2 and #3 above so it doesn't occur on every TCP packet split. Signed-off-by: Dave Johnson <djohnson+linux-mips@sw.starentnetworks.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org> 
I've traced down an off-by-one TCP checksum calculation error under
the following conditions:

1) The TCP code needs to split a full-sized packet due to a reduced
   MSS (typically due to the addition of TCP options mid-stream like
   SACK).
   _AND_
2) The checksum of the 2nd fragment is larger than the checksum of the
   original packet.  After subtraction this results in a checksum for
   the 1st fragment with bits 16..31 set to 1. (this is ok)
   _AND_
3) The checksum of the 1st fragment's TCP header plus the previously
   32bit checksum of the 1st fragment DOES NOT cause a 32bit overflow
   when added together.  This results in a checksum of the TCP header
   plus TCP data that still has the upper 16 bits as 1's.
   _THEN_
4) The TCP+data checksum is added to the checksum of the pseudo IP
   header with csum_tcpudp_nofold() incorrectly (the bug).
    
The problem is the checksum of the TCP+data is passed to
csum_tcpudp_nofold() as an 32bit unsigned value, however the assembly
code acts on it as if it is a 64bit unsigned value.

This causes an incorrect 32->64bit extension if the sum has bit 31
set.  The resulting checksum is off by one.
    
This problems is data and TCP header dependent due to #2 and #3
above so it doesn't occur on every TCP packet split.
    
Signed-off-by: Dave Johnson <djohnson+linux-mips@sw.starentnetworks.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
[MIPS] Fix wrong checksum calculation on 64-bit MIPS 2007-01-24T19:23:22+00:00 Atsushi Nemoto anemo@mba.ocn.ne.jp 2007-01-24T06:43:34+00:00 66218da212bf141532d678a699f5789c78145ab1 The commit 8e3d8433d8c22ca6c42cba4a67d300c39aae7822 ([NET]: MIPS checksum annotations and cleanups) broke 64-bit MIPS. The problem is the commit replaces some unsigned long with __be32. On 64bit MIPS, a __be32 (i.e. unsigned int) value is represented as a sign-extented 32-bit value in a 64-bit argument register. So the address 192.168.0.1 (0xc0a80001) is passed as 0xffffffffc0a80001 to csum_tcpudp_nofold() but the asm code in the function expects 0x00000000c0a80001, therefore it returns a wrong checksum. Explicit cast to unsigned long is needed to drop high 32bit. Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp> Signed-off-by: Ralf Baechle <ralf@linux-mips.org> 
The commit 8e3d8433d8c22ca6c42cba4a67d300c39aae7822 ([NET]: MIPS
checksum annotations and cleanups) broke 64-bit MIPS.

The problem is the commit replaces some unsigned long with __be32.  On
64bit MIPS, a __be32 (i.e. unsigned int) value is represented as a
sign-extented 32-bit value in a 64-bit argument register.  So the
address 192.168.0.1 (0xc0a80001) is passed as 0xffffffffc0a80001 to
csum_tcpudp_nofold() but the asm code in the function expects
0x00000000c0a80001, therefore it returns a wrong checksum.  Explicit
cast to unsigned long is needed to drop high 32bit.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
[MIPS] csum_partial and copy in parallel 2007-01-08T21:41:04+00:00 Atsushi Nemoto anemo@mba.ocn.ne.jp 2006-12-12T16:22:06+00:00 f860c90bd6ce22c6a0a352cc16acc74fba3d628e Implement optimized asm version of csum_partial_copy_nocheck, csum_partial_copy_from_user and csum_and_copy_to_user which can do calculate and copy in parallel, based on memcpy.S. Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp> Signed-off-by: Ralf Baechle <ralf@linux-mips.org> 
Implement optimized asm version of csum_partial_copy_nocheck,
csum_partial_copy_from_user and csum_and_copy_to_user which can do
calculate and copy in parallel, based on memcpy.S.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
[NET]: MIPS checksum annotations and cleanups. 2006-12-03T05:23:09+00:00 Al Viro viro@zeniv.linux.org.uk 2006-11-15T05:18:18+00:00 8e3d8433d8c22ca6c42cba4a67d300c39aae7822 * sanitize prototypes, annotate * kill shift-by-16 in checksum calculations * htons->shift in l-e checksum calculations Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net> 
* sanitize prototypes, annotate
* kill shift-by-16 in checksum calculations
* htons->shift in l-e checksum calculations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Don't include linux/config.h from anywhere else in include/ 2006-04-26T11:56:16+00:00 David Woodhouse dwmw2@infradead.org 2006-04-26T11:56:16+00:00 62c4f0a2d5a188f73a94f2cb8ea0dba3e7cf0a7f Signed-off-by: David Woodhouse <dwmw2@infradead.org> 
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
The type of sum in csum_tcpudp_nofold is "unsigned int", so when we assign 2005-10-29T18:32:25+00:00 Ralf Baechle ralf@linux-mips.org 2005-09-20T10:56:26+00:00 e97288386a316e3960b758a60ae7cd17b3ee10e7 to it in an asm() block, and we're running on a system with 64-bit registers, it is vitally important that we sign extend it correctly before returning to C. Otherwise the stray high bits will be preserved into csum_fold, and on the SB-1 processor, 32-bit arithmetic on a non sign-extended register will yield surprising results. This caused incorrect checksums in some UDP packets for NFS root. The problem was mild when using a 10.0.1.x IP address, but severe when using 192.168.1.x. Signed-off-by: Daniel Jacobowitz <dan@codesourcery.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org> 
to it in an asm() block, and we're running on a system with 64-bit
registers, it is vitally important that we sign extend it correctly before
returning to C.  Otherwise the stray high bits will be preserved into
csum_fold, and on the SB-1 processor, 32-bit arithmetic on a non
sign-extended register will yield surprising results.
    
This caused incorrect checksums in some UDP packets for NFS root.  The
problem was mild when using a 10.0.1.x IP address, but severe when
using 192.168.1.x.
    
Signed-off-by: Daniel Jacobowitz <dan@codesourcery.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Protect noat assembly with .set push/pop and make it somewhat readable. 2005-10-29T18:31:40+00:00 Thiemo Seufer ths@networkno.de 2005-07-08T09:17:05+00:00 04988d6fda5c1cc670bd4c4d563b3443cf7ccce7 Signed-off-by: Ralf Baechle <ralf@linux-mips.org> Signed-off-by: Thiemo Seufer <ths@networkno.de> 
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Thiemo Seufer <ths@networkno.de>
Sparseify MIPS. 2005-10-29T18:30:50+00:00 Ralf Baechle ralf@linux-mips.org 2005-03-01T19:22:29+00:00 fe00f943e0ef98b4057abcc2940d631a975b43cd Signed-off-by: Ralf Baechle <ralf@linux-mips.org> 
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
[PATCH] mips: clean up 32/64-bit configuration 2005-09-05T07:06:06+00:00 Ralf Baechle ralf@linux-mips.org 2005-09-03T22:56:16+00:00 875d43e72b5bf22161a81de7554f88eccf8a51ae Start cleaning 32-bit vs. 64-bit configuration. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org> 
Start cleaning 32-bit vs. 64-bit configuration.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>