The current implementation of from64to32() gives a poor result:
0000000000000270 <.from64to32>:
270: 38 00 ff ff li r0,-1
274: 78 69 00 22 rldicl r9,r3,32,32
278: 78 00 00 20 clrldi r0,r0,32
27c: 7c 60 00 38 and r0,r3,r0
280: 7c 09 02 14 add r0,r9,r0
284: 78 09 00 22 rldicl r9,r0,32,32
288: 7c 00 4a 14 add r0,r0,r9
28c: 78 03 00 20 clrldi r3,r0,32
290: 4e 80 00 20 blr
This patch modifies from64to32() to operate in the same
spirit as csum_fold()
It swaps the two 32-bit halves of sum then it adds it with the
unswapped sum. If there is a carry from adding the two 32-bit halves,
it will carry from the lower half into the upper half, giving us the
correct sum in the upper half.
The resulting code is:
0000000000000260 <.from64to32>:
260: 78 60 00 02 rotldi r0,r3,32
264: 7c 60 1a 14 add r3,r0,r3
268: 78 63 00 22 rldicl r3,r3,32,32
26c: 4e 80 00 20 blr
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This reverts commit 6ad966d730.
That commit was pointless, because csum_add() sums two 32 bits
values, so the sum is 0x1fffffffe at the maximum.
And then when adding upper part (1) and lower part (0xfffffffe),
the result is 0xffffffff which doesn't carry.
Any lower value will not carry either.
And behind the fact that this commit is useless, it also kills the
whole purpose of having an arch specific inline csum_add()
because the resulting code gets even worse than what is obtained
with the generic implementation of csum_add()
0000000000000240 <.csum_add>:
240: 38 00 ff ff li r0,-1
244: 7c 84 1a 14 add r4,r4,r3
248: 78 00 00 20 clrldi r0,r0,32
24c: 78 89 00 22 rldicl r9,r4,32,32
250: 7c 80 00 38 and r0,r4,r0
254: 7c 09 02 14 add r0,r9,r0
258: 78 09 00 22 rldicl r9,r0,32,32
25c: 7c 00 4a 14 add r0,r0,r9
260: 78 03 00 20 clrldi r3,r0,32
264: 4e 80 00 20 blr
In comparison, the generic implementation of csum_add() gives:
0000000000000290 <.csum_add>:
290: 7c 63 22 14 add r3,r3,r4
294: 7f 83 20 40 cmplw cr7,r3,r4
298: 7c 10 10 26 mfocrf r0,1
29c: 54 00 ef fe rlwinm r0,r0,29,31,31
2a0: 7c 60 1a 14 add r3,r0,r3
2a4: 78 63 00 20 clrldi r3,r3,32
2a8: 4e 80 00 20 blr
And the reverted implementation for PPC64 gives:
0000000000000240 <.csum_add>:
240: 7c 84 1a 14 add r4,r4,r3
244: 78 80 00 22 rldicl r0,r4,32,32
248: 7c 80 22 14 add r4,r0,r4
24c: 78 83 00 20 clrldi r3,r4,32
250: 4e 80 00 20 blr
Fixes: 6ad966d730 ("powerpc/64: Fix checksum folding in csum_add()")
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Paul's patch to fix checksum folding, commit b492f7e4e0 ("powerpc/64:
Fix checksum folding in csum_tcpudp_nofold and ip_fast_csum_nofold")
missed a case in csum_add(). Fix it.
Signed-off-by: Shile Zhang <shile.zhang@nokia.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Currently we have optimized hand-coded assembly checksum routines for
big-endian 64-bit systems, but for little-endian we use the generic C
routines. This modifies the optimized routines to work for
little-endian. With this, we no longer need to enable
CONFIG_GENERIC_CSUM. This also fixes a couple of comments in
checksum_64.S so they accurately reflect what the associated instruction
does.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
[mpe: Use the more common __BIG_ENDIAN__]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
These functions compute an IP checksum by computing a 64-bit sum and
folding it to 32 bits (the "nofold" in their names refers to folding
down to 16 bits). However, doing (u32) (s + (s >> 32)) is not
sufficient to fold a 64-bit sum to 32 bits correctly. The addition
can produce a carry out from bit 31, which needs to be added in to
the sum to produce the correct result.
To fix this, we copy the from64to32() function from lib/checksum.c
and use that.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Commit 01cfbad "ipv4: Update parameters for csum_tcpudp_magic to their
original types" changed parameters for csum_tcpudp_magic and
csum_tcpudp_nofold for many platforms but not for PowerPC.
Fixes: 01cfbad "ipv4: Update parameters for csum_tcpudp_magic to their original types"
Cc: Alexander Duyck <aduyck@mirantis.com>
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Acked-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
csum_partial is often called for small fixed length packets
for which it is suboptimal to use the generic csum_partial()
function.
For instance, in my configuration, I got:
* One place calling it with constant len 4
* Seven places calling it with constant len 8
* Three places calling it with constant len 14
* One place calling it with constant len 20
* One place calling it with constant len 24
* One place calling it with constant len 32
This patch renames csum_partial() to __csum_partial() and
implements csum_partial() as a wrapper inline function which
* uses csum_add() for small 16bits multiple constant length
* uses ip_fast_csum() for other 32bits multiple constant
* uses __csum_partial() in all other cases
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Scott Wood <oss@buserror.net>
Simplify csum_add(a, b) in case a or b is constant 0
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Scott Wood <oss@buserror.net>
In several architectures, ip_fast_csum() is inlined
There are functions like ip_send_check() which do nothing
much more than calling ip_fast_csum().
Inlining ip_fast_csum() allows the compiler to optimise better
Suggested-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
[scottwood: whitespace and cast fixes]
Signed-off-by: Scott Wood <oss@buserror.net>
The powerpc64 checksum wrapper functions adds csum_and_copy_to_user()
which otherwise is implemented in include/net/checksum.h by using
csum_partial() then copy_to_user()
Those two wrapper fonctions are also applicable to powerpc32 as it is
based on the use of csum_partial_copy_generic() which also
exists on powerpc32
This patch renames arch/powerpc/lib/checksum_wrappers_64.c to
arch/powerpc/lib/checksum_wrappers.c and
makes it non-conditional to CONFIG_WORD_SIZE
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Scott Wood <oss@buserror.net>
addc uses carry so xer is clobbered in csum_add()
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Scott Wood <oss@buserror.net>
The C version of csum_add() as defined in include/net/checksum.h gives
the following assembly in ppc32:
0: 7c 04 1a 14 add r0,r4,r3
4: 7c 64 00 10 subfc r3,r4,r0
8: 7c 63 19 10 subfe r3,r3,r3
c: 7c 63 00 50 subf r3,r3,r0
and the following in ppc64:
0xc000000000001af8 <+0>: add r3,r3,r4
0xc000000000001afc <+4>: cmplw cr7,r3,r4
0xc000000000001b00 <+8>: mfcr r4
0xc000000000001b04 <+12>: rlwinm r4,r4,29,31,31
0xc000000000001b08 <+16>: add r3,r4,r3
0xc000000000001b0c <+20>: clrldi r3,r3,32
0xc000000000001b10 <+24>: blr
include/net/checksum.h also offers the possibility to define an arch
specific function. This patch provides a specific csum_add() inline
function.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Scott Wood <scottwood@freescale.com>
csum_tcpudp_magic() is only a few instructions, and does modify
really few registers. So it is not worth having it as a separate
function and suffer function branching and saving of volatile
registers.
This patch makes it inline by use of the already existing
csum_tcpudp_nofold() function.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Scott Wood <scottwood@freescale.com>
We need to fix some endian issues in our checksum code. For now
just enable the generic checksum routines for little endian builds.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This adds the equivalent of csum_and_copy_from_user for the receive side so we
can copy and checksum in one pass. It is modelled on the generic checksum
routine.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
We use the same core loop as the new csum_partial, adding in the
stores and exception handling code. To keep things simple we do all the
exception fixup in csum_and_copy_from_user. This wrapper function is
modelled on the generic checksum code and is careful to always calculate
a complete checksum even if we only copied part of the data to userspace.
To test this I forced checksumming on over loopback and ran socklib (a
simple TCP benchmark). On a POWER6 575 throughput improved by 19% with
this patch. If I forced both the sender and receiver onto the same cpu
(with the hope of shifting the benchmark from being cache bandwidth limited
to cpu limited), adding this patch improved performance by 55%
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
from include/asm-powerpc. This is the result of a
mkdir arch/powerpc/include/asm
git mv include/asm-powerpc/* arch/powerpc/include/asm
Followed by a few documentation/comment fixups and a couple of places
where <asm-powepc/...> was being used explicitly. Of the latter only
one was outside the arch code and it is a driver only built for powerpc.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Christophe Leroy