fedora 14 kernel performance with ip forwarding workload
by Jesse Brandeburg
The other day I was running the stock fedora kernel on my ip
forwarding setup, to see what the performance was, and the performance
wasn't very good.
system is S5520HC dual socket 2.93GHz Xeon 5570 (Nehalem) with 3 quad
port 82580 adapters (12 ports). Traffic is bidirectional 64 byte
packets being forwarded and received on each port, basically port to
port routing. I am only using 12 flows currently.
The driver is igb, and I am using an affinity script that lines up
each pair of ports that are forwarding traffic into optimal
configurations for cache locality. I am also disabling
remote_node_defrag_ratio to stop cross node traffic.
With the fedora default kernel from F14 it appears that
CONFIG_NETFILTER=y means that I cannot unload all of netfilter even if
I stop iptables service.
perf showed netfilter being prominent, and removing it gives me much
higher throughput. Is there a reason CONFIG_NETFILTER=y ? Isn't it a
good thing to be able to disable netfilter if you want to?
Jesse
8 years, 6 months
[RFC v2 00/19] utrace for 3.0 kernel
by Oleg Nesterov
Another attempt. This version tries to decouple utrace and ptrace.
This way it is much simpler to follow the upstream changes, afaics.
TODO:
- The single-stepping updates in ptrace_resume() can race
with utrace_reset()->user_disable_single_step().
This was fixed by 20/20, but I noticed that this patch
is buggy right before sending.
- Perhaps PTRACE_SYSEMU/TIF_SYSCALL_EMU logic was broken,
I need to recheck.
- Testing.
Oleg.
12 years, 4 months
utrace/uprobes for 3.1 in F16
by Josh Boyer
Hi Oleg,
For now F16 is sitting on 3.0 with the utrace patch applied, but we'll
eventually be bringing 3.1 back to that branch after it gets some
vetting in rawhide. Since I'm rebasing rawhide to the 3.1 pre-rc
snapshots, I'm wondering if you have worked on utrace/uprobes at all for
3.1.
I haven't seen uprobes merged yet, and as of right now the utrace patch
is disabled in the rawhide kernel. Could you let me know if we're
waiting for uprobes to get merged, or what the approach for 3.1 should
be?
josh
12 years, 4 months
strange mcelog mesage
by Genes MailLists
I'm running 3.0-0.rc6.git6.1.fc16.x86_64 on a new sandy bridge laptop on
updated f15 (except for kernel, procps and mdadm from rawhide).
(I'll put rc7 on once the build is done - thanks kyle!
The message I get (temps certainly seem normal):
------------- /var/log/messages -----------------------------
Jul 12 08:39:43 lap3 mcelog[1001]: HARDWARE ERROR. This is *NOT* a
software problem!
Jul 12 08:39:43 lap3 mcelog[1001]: Please contact your hardware vendor
Jul 12 08:39:43 lap3 mcelog[1001]: MCE 11
Jul 12 08:39:43 lap3 mcelog[1001]: CPU 6 THERMAL EVENT TSC 985befbeec
Jul 12 08:39:43 lap3 mcelog[1001]: TIME 1310474383 Tue Jul 12 08:39:43 2011
Jul 12 08:39:43 lap3 mcelog[1001]: Processor 6 below trip temperature.
Throttling disabled
Jul 12 08:39:43 lap3 mcelog[1001]: STATUS c000000088400c08 MCGSTATUS 0
Jul 12 08:39:43 lap3 mcelog[1001]: MCGCAP c09 APICID 6 SOCKETID 0
Jul 12 08:39:43 lap3 mcelog[1001]: CPUID Vendor Intel Family 6 Model 42
Jul 12 08:39:43 lap3 mcelog[1001]: mcelog: Unsupported new Family 6
Model 2a CPU: only decoding architectural err
------------- ver_linux --------------------------------
Linux lap3.prv.sapience.com 3.0-0.rc6.git6.1.fc16.x86_64 #1 SMP Sun Jul
10 16:00:07 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux
Gnu C 4.6.0
Gnu make 3.82
binutils 2.21.51.0.6
util-linux 2.19.1
mount support
module-init-tools 3.16
e2fsprogs 1.41.14
jfsutils 1.1.13
xfsprogs 3.1.4
pcmciautils 017
quota-tools 4.00-pre1.
PPP 2.4.5
isdn4k-utils 3.13
Linux C Library 2.14
Dynamic linker (ldd) 2.14
Procps 3.2.8
Net-tools 1.60
Kbd 1.15.2
oprofile 0.9.6
Sh-utils 8.10
wireless-tools 29
Modules Loaded tcp_lp vboxnetadp vboxnetflt vboxdrv hidp fuse
ebtable_nat ebtables ipt_MASQUERADE iptable_nat nf_nat xt_CHECKSUM
iptable_mangle bridge stp llc ppdev parport_pc lp parport sunrpc capi
kernelcapi rfcomm bnep ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6
ip6table_filter ip6_tables nf_conntrack_ipv4 nf_defrag_ipv4 xt_state
nf_conntrack xts gf128mul dm_crypt arc4 uvcvideo videodev media
v4l2_compat_ioctl32 btusb bluetooth snd_hda_codec_conexant microcode
joydev i2c_i801 snd_hda_intel snd_hda_codec snd_hwdep iwlagn snd_seq
snd_seq_device mac80211 snd_pcm cfg80211 xhci_hcd iTCO_wdt snd_timer
iTCO_vendor_support e1000e snd_page_alloc thinkpad_acpi rfkill snd
soundcore virtio_net kvm_intel kvm firewire_ohci sdhci_pci sdhci
firewire_core mmc_core crc_itu_t wmi i915 drm_kms_helper drm
i2c_algo_bit i2c_core video
12 years, 4 months
f15 2.6.39 rebase
by Dave Jones
I've just pushed a f15-2.6.39 branch which contains a work in progress rebase.
The only thing that I'm really concerned about in this right now is X regressions.
We had a drm-next backport to .38 and moving that to .39 turned up a ton of rejects.
I fixed up a few by hand, but the resulting compile failures made my head hurt, so
I've mostly left them disabled. If the nouveau/intel drm dudes could look over
that branch and fix up whatever is necessary, we can look at getting this out
to people soon.
(looking ahead, after its release pushing 3.0.x as 2.6.40 is probably going to
happen, in a shorter timeframe than what took us to get .39 ready)
Dave
12 years, 4 months
bluetooth problem with kernel-3.0-0.rc7.git3.1.fc16.x86_6
by Genes MailLists
kernel-3.0-0.rc7.git0.1.fc16.x86_6 is fine - but git3 bluetooth is
problematic.
This is on F15 with 3.0 kernel and procps.
symptom is blutooth mouse doesn't work. Restarting bluetooth.service
does not help.
Its a sandy bridge laptop (lenovo w520).
/var/log/messages has a load of these:
Jul 16 22:53:33 lap3 bluetoothd[2843]: Adapter /org/bluez/2843/hci0 has
been enabled
Jul 16 22:53:33 lap3 bluetoothd[2843]: bluetoothd[2843]: Adapter
/org/bluez/2843/hci0 has been enabled
Jul 16 22:53:49 lap3 bluetoothd[2843]: Connection refused (111)
Jul 16 22:53:49 lap3 bluetoothd[2843]: bluetoothd[2843]: Connection
refused (111)
Jul 16 22:53:49 lap3 bluetoothd[2843]: HUP or ERR on socket
Jul 16 22:53:49 lap3 bluetoothd[2843]: bluetoothd[2843]: HUP or ERR on
socket
Jul 16 22:54:04 lap3 bluetoothd[2843]: HUP or ERR on socket
Jul 16 22:54:04 lap3 bluetoothd[2843]: bluetoothd[2843]: HUP or ERR on
socket
Jul 16 22:54:19 lap3 pulseaudio[2332]: ratelimit.c: 24 events suppressed
Jul 16 22:54:23 lap3 bluetoothd[2843]: HUP or ERR on socket
Jul 16 22:54:23 lap3 bluetoothd[2843]: bluetoothd[2843]: HUP or ERR on
socket
Jul 16 22:54:37 lap3 dbus: [system] Activating service
name='net.reactivated.Fprint' (using servicehelper)
Jul 16 22:54:37 lap3 dbus: [system] Successfully activated service
'net.reactivated.Fprint'
Jul 16 22:54:43 lap3 bluetoothd[2843]: HUP or ERR on socket
Jul 16 22:54:43 lap3 bluetoothd[2843]: bluetoothd[2843]: HUP or ERR on
socket
Jul 16 22:54:48 lap3 bluetoothd[2843]: Stopping hci0 event socket
Jul 16 22:54:48 lap3 bluetoothd[2843]: bluetoothd[2843]: Stopping hci0
event socket
Jul 16 22:54:48 lap3 bluetoothd[2843]: Stopping SDP server
Jul 16 22:54:48 lap3 bluetoothd[2843]: bluetoothd[2843]: Stopping SDP server
Jul 16 22:54:48 lap3 bluetoothd[2843]: Exit
Jul 16 22:54:48 lap3 bluetoothd[2843]: bluetoothd[2843]: Exit
Jul 16 22:54:48 lap3 dbus: [system] Activating via systemd: service
name='org.bluez' unit='dbus-org.bluez.service
------------ver_linux
/usr/src/kernels/3.0-0.rc7.git0.1.fc16.x86_64/scripts/ver_linux
If some fields are empty or look unusual you may have an old version.
Compare to the current minimal requirements in Documentation/Changes.
Linux lap3.prv.sapience.com 3.0-0.rc7.git0.1.fc16.x86_64 #1 SMP Tue Jul
12 12:57:40 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux
Gnu C 4.6.0
Gnu make 3.82
binutils 2.21.51.0.6
util-linux 2.19.1
mount support
module-init-tools 3.16
e2fsprogs 1.41.14
jfsutils 1.1.13
xfsprogs 3.1.4
pcmciautils 017
quota-tools 4.00-pre1.
PPP 2.4.5
isdn4k-utils 3.13
Linux C Library 2.14
Dynamic linker (ldd) 2.14
Procps 3.2.8
Net-tools 1.60
Kbd 1.15.2
oprofile 0.9.6
Sh-utils 8.10
wireless-tools 29
Modules Loaded hidp fuse ebtable_nat ebtables ipt_MASQUERADE
iptable_nat nf_nat xt_CHECKSUM iptable_mangle bridge stp llc ppdev
parport_pc lp parport sunrpc vboxnetadp vboxnetflt vboxdrv capi
kernelcapi rfcomm bnep ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6
ip6table_filter ip6_tables nf_conntrack_ipv4 nf_defrag_ipv4 xt_state
nf_conntrack xts gf128mul dm_crypt arc4 uvcvideo videodev media
v4l2_compat_ioctl32 btusb bluetooth snd_hda_codec_conexant microcode
joydev snd_hda_intel snd_hda_codec snd_hwdep iwlagn snd_seq
snd_seq_device mac80211 snd_pcm xhci_hcd i2c_i801 cfg80211 iTCO_wdt
iTCO_vendor_support snd_timer snd_page_alloc e1000e thinkpad_acpi rfkill
snd soundcore binfmt_misc virtio_net kvm_intel kvm sdhci_pci sdhci
firewire_ohci mmc_core firewire_core crc_itu_t wmi i915 drm_kms_helper
drm i2c_algo_bit i2c_core video
---------------- lspci
# lspci
00:00.0 Host bridge: Intel Corporation Sandy Bridge DRAM Controller (rev
09)
00:02.0 VGA compatible controller: Intel Corporation Device 0126 (rev
09)
00:16.0 Communication controller: Intel Corporation Cougar Point HECI
Controller #1 (rev 04)
00:16.3 Serial controller: Intel Corporation Cougar Point KT Controller
(rev 04)
00:19.0 Ethernet controller: Intel Corporation 82579LM Gigabit Network
Connection (rev 04)
00:1a.0 USB Controller: Intel Corporation Cougar Point USB Enhanced Host
Controller #2 (rev 04)
00:1b.0 Audio device: Intel Corporation Cougar Point High Definition
Audio Controller (rev 04)
00:1c.0 PCI bridge: Intel Corporation Cougar Point PCI Express Root Port
1 (rev b4)
00:1c.1 PCI bridge: Intel Corporation Cougar Point PCI Express Root Port
2 (rev b4)
00:1c.3 PCI bridge: Intel Corporation Cougar Point PCI Express Root Port
4 (rev b4)
00:1c.4 PCI bridge: Intel Corporation Cougar Point PCI Express Root Port
5 (rev b4)
00:1c.6 PCI bridge: Intel Corporation Cougar Point PCI Express Root Port
7 (rev b4)
00:1d.0 USB Controller: Intel Corporation Cougar Point USB Enhanced Host
Controller #1 (rev 04)
00:1f.0 ISA bridge: Intel Corporation Cougar Point LPC Controller (rev 04)
00:1f.2 SATA controller: Intel Corporation Cougar Point 6 port SATA AHCI
Controller (rev 04)
00:1f.3 SMBus: Intel Corporation Cougar Point SMBus Controller (rev 04)
03:00.0 Network controller: Intel Corporation Centrino Ultimate-N 6300
(rev 35)
0d:00.0 System peripheral: Ricoh Co Ltd Device e823 (rev 05)
0d:00.3 FireWire (IEEE 1394): Ricoh Co Ltd Device e832 (rev 04)
0e:00.0 USB Controller: NEC Corporation uPD720200 USB 3.0 Host
Controller (rev 04)
12 years, 4 months
kernel 3.0-0.rc7.git0.1
by Genes MailLists
Been using 3.0-0.rc7.git0.1.fc16.x86_64 for a couple of days now on a
Sandy Bridge laptop - it is working generally quit well for me -
the i915 graphics remains very very sluggish (not new to this kernel)
but better than all previous kernels. I am getting occasional window
leftover turds on display background. glxgears has improved from 50 to
270 (not great). Moving windows around in KDE (2-D) is sluggish still.
iwlagn seems to be fine.
auto sched is still wonderful !!
I am not sure where to give karma to this koji build - so mailing here.
gene
12 years, 4 months
Re: [ INFO: possible recursive locking detected ] systemd-logind/651 is trying to acquire lock:
by Michał Piotrowski
Hi,
2011/7/15 Lucas <macachuto(a)gmail.com>:
> Dear All.
>
> Just updated and got the following in dmesg:
It looks like a kernel bug.
>
>
>
> [ 37.653381] =============================================
> [ 37.654015] [ INFO: possible recursive locking detected ]
> [ 37.654015] 3.0-0.rc7.git0.1.fc16.i686 #1
> [ 37.654015] ---------------------------------------------
> [ 37.654015] systemd-logind/651 is trying to acquire lock:
> [ 37.654015] (&ep->mtx){+.+.+.}, at: [<c05285f1>] ep_scan_ready_list+0x32/0x154
> [ 37.654015]
> [ 37.654015] but task is already holding lock:
> [ 37.654015] (&ep->mtx){+.+.+.}, at: [<c0528a90>] sys_epoll_ctl+0x103/0x481
> [ 37.654015]
> [ 37.654015] other info that might help us debug this:
> [ 37.654015] Possible unsafe locking scenario:
> [ 37.654015]
> [ 37.654015] CPU0
> [ 37.654015] ----
> [ 37.654015] lock(&ep->mtx);
> [ 37.654015] lock(&ep->mtx);
> [ 37.654015]
> [ 37.654015] *** DEADLOCK ***
> [ 37.654015]
> [ 37.654015] May be due to missing lock nesting notation
> [ 37.654015]
> [ 37.654015] 2 locks held by systemd-logind/651:
> [ 37.654015] #0: (epmutex){+.+.+.}, at: [<c0528a4b>] sys_epoll_ctl+0xbe/0x481
> [ 37.654015] #1: (&ep->mtx){+.+.+.}, at: [<c0528a90>] sys_epoll_ctl+0x103/0x481
> [ 37.654015]
> [ 37.654015] stack backtrace:
> [ 37.654015] Pid: 651, comm: systemd-logind Not tainted 3.0-0.rc7.git0.1.fc16.i686 #1
> [ 37.654015] Call Trace:
> [ 37.654015] [<c08490fe>] ? printk+0x2d/0x2f
> [ 37.654015] [<c046b2ef>] __lock_acquire+0x811/0xb63
> [ 37.654015] [<c0407c77>] ? sched_clock+0x8/0xb
> [ 37.654015] [<c045d190>] ? sched_clock_local+0x10/0x18b
> [ 37.654015] [<c05285f1>] ? ep_scan_ready_list+0x32/0x154
> [ 37.654015] [<c046ba5e>] lock_acquire+0xad/0xe4
> [ 37.654015] [<c05285f1>] ? ep_scan_ready_list+0x32/0x154
> [ 37.654015] [<c08506bd>] __mutex_lock_common+0x49/0x2ee
> [ 37.654015] [<c05285f1>] ? ep_scan_ready_list+0x32/0x154
> [ 37.654015] [<c04332e6>] ? __might_sleep+0x29/0xfb
> [ 37.654015] [<c046a912>] ? mark_lock+0x26/0x1f2
> [ 37.654015] [<c0850a7c>] mutex_lock_nested+0x43/0x49
> [ 37.654015] [<c05285f1>] ? ep_scan_ready_list+0x32/0x154
> [ 37.654015] [<c05285f1>] ep_scan_ready_list+0x32/0x154
> [ 37.654015] [<c05281cb>] ? ep_remove+0x9b/0x9b
> [ 37.654015] [<c0528727>] ep_poll_readyevents_proc+0x14/0x16
> [ 37.654015] [<c05283d6>] ep_call_nested.constprop.2+0x6d/0x9a
> [ 37.654015] [<c0528713>] ? ep_scan_ready_list+0x154/0x154
> [ 37.654015] [<c05284d2>] ep_eventpoll_poll+0x45/0x55
> [ 37.654015] [<c0528b8c>] sys_epoll_ctl+0x1ff/0x481
> [ 37.654015] [<c05282fb>] ? ep_send_events_proc+0xd5/0xd5
> [ 37.654015] [<c08521ac>] syscall_call+0x7/0xb
> --
> devel mailing list
> devel(a)lists.fedoraproject.org
> https://admin.fedoraproject.org/mailman/listinfo/devel
>
--
Best regards,
Michal
http://eventhorizon.pl/
12 years, 4 months
3.0.rc4
by Genes MailLists
When I install kernel-3.0-0.rc4.git0.2.fc16.x86_64.rpm
from koji on F15 I get these warnings - how to I make them go away?
thanks!
gene
--------------------
W: Possible missing firmware "nouveau/fuc41ad" for kernel module
"nouveau.ko"
W: Possible missing firmware "nouveau/fuc41ac" for kernel module
"nouveau.ko"
W: Possible missing firmware "nouveau/fuc409d" for kernel module
"nouveau.ko"
W: Possible missing firmware "nouveau/fuc409c" for kernel module
"nouveau.ko"
W: Possible missing firmware "nouveau/nvc4_fuc41ad" for kernel module
"nouveau.ko"
W: Possible missing firmware "nouveau/nvc4_fuc41ac" for kernel module
"nouveau.ko"
W: Possible missing firmware "nouveau/nvc4_fuc409d" for kernel module
"nouveau.ko"
W: Possible missing firmware "nouveau/nvc4_fuc409c" for kernel module
"nouveau.ko"
W: Possible missing firmware "nouveau/nvc3_fuc41ad" for kernel module
"nouveau.ko"
W: Possible missing firmware "nouveau/nvc3_fuc41ac" for kernel module
"nouveau.ko"
W: Possible missing firmware "nouveau/nvc3_fuc409d" for kernel module
"nouveau.ko"
W: Possible missing firmware "nouveau/nvc3_fuc409c" for kernel module
"nouveau.ko"
W: Possible missing firmware "nouveau/nvc0_fuc41ad" for kernel module
"nouveau.ko"
W: Possible missing firmware "nouveau/nvc0_fuc41ac" for kernel module
"nouveau.ko"
W: Possible missing firmware "nouveau/nvc0_fuc409d" for kernel module
"nouveau.ko"
W: Possible missing firmware "nouveau/nvc0_fuc409c" for kernel module
"nouveau.ko"
W: Possible missing firmware "cbfw_fc.bin" for kernel module "bfa.ko"
W: Possible missing firmware "ctfw_cna.bin" for kernel module "bfa.ko"
W: Possible missing firmware "ctfw_fc.bin" for kernel module "bfa.ko"
12 years, 5 months