On Sat, Mar 31, 2007 at 03:41:32AM +0200, Patrick McHardy wrote: > > > [CRYPTO] api: scatterwalk_copychunks() fails to advance through scatterlist > > This patch seems to cause some problems, I get reproducable freezes > on the receiving system with net-2.6.22 when sending IPsec packets > larger than the mtu (reproduced about 10 times). Reverting this > patch seems to fix it. In a few cases the oops also occured on the > sending system. > > Backtrace from UML (sending system): > > uml:~# ping 10.0.0.1 -s 20000 > PING 10.0.0.1 (10.0.0.1) 20000(20028) bytes of data. > BUG: soft lockup detected on CPU#0! > Call Trace:
Indeed. That patch was buggy. Sorry for not catching this earlier. This should fix the problem. [CRYPTO] api: Use the right value when advancing scatterwalk_copychunks In the scatterwalk_copychunks loop, We should be advancing by len_this_page and not nbytes. The latter is the total length. Signed-off-by: Herbert Xu <[EMAIL PROTECTED]> Cheers, -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} <[EMAIL PROTECTED]> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt -- diff --git a/crypto/scatterwalk.c b/crypto/scatterwalk.c index a664231..0f76175 100644 --- a/crypto/scatterwalk.c +++ b/crypto/scatterwalk.c @@ -91,7 +91,7 @@ void scatterwalk_copychunks(void *buf, struct scatter_walk *walk, memcpy_dir(buf, vaddr, len_this_page, out); scatterwalk_unmap(vaddr, out); - scatterwalk_advance(walk, nbytes); + scatterwalk_advance(walk, len_this_page); if (nbytes == len_this_page) break; - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
