Francois Romieu wrote: Hi, could you give the patch below some testing ? Any post-LLTX-revert 2.6.11-rc2-bk should do. Merge of Realtek's code - code dedicated to a new phy (spotted by Richard Dawe)
Richard Dawe <rich@xxxxxxxxxxxxxxxxxxxx> : So it can be considered that the link detection issue is gone. Ok, I'll ask Jeff to submit it for mainline. If this test is (sys) CPU bound, then you should
Francois Romieu wrote: Richard Dawe <rich@xxxxxxxxxxxxxxxxxxxx> : [...] [snip] It works fine setting the speed & duplex using "ethtool -s eth0 autoneg off speed <speed> duplex <duplex>". Nice. Does
Francois Romieu wrote: [snip] If you can do an extra test, I'd like to know if you can safely bring the r8169 interface down on your computer once the patch below if applied: http://www.fr.zoreil.co
Richard Dawe <rich@xxxxxxxxxxxxxxxxxxxx> : @#$*%! Patch against 2.6.11-rc3: http://www.fr.zoreil.com/people/francois/misc/20050211-2.6.11-rc3-r8169-test.patch Better ? Up-to-date patch-script version
Francois Romieu wrote: Richard Dawe <rich@xxxxxxxxxxxxxxxxxxxx> : [...] I find that the "ip link set dev eth0 down" command hangs, consuming most of the CPU time (~99%). Note that I issued the "netw
Francois Romieu wrote: Richard Dawe <rich@xxxxxxxxxxxxxxxxxxxx> : [...] IRQ routing seems to be disabled in 2.6.10. I got a warning about an unhandled interrupt for the VIA 8255 (I think that's the
Richard Dawe wrote: Francois Romieu wrote: [snip] Could you send an updated dmesg, lspci -vvx, /proc/interrupts please ? Please see the attached, which are for running with 2.6.10. I included .confi
ce benchmarking for 2.6.10 IPSec with 2.6 GHz P5s connected back-to-back with Gigabit connection. I'm using iperf for performance test. I'm really puzzled with
it seems that in case of null encryption, scatterwalk related code takes most of the cpu cycles. Tunnel mode ESP with null-encryption/sha1 (throughput 51MBits/