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
7 years, 12 months
Supporting Allwinner ARM SoCs in Fedora 21+ kernels
by Hans de Goede
Hi All,
As you probably know in my spare time I'm working on Linux Allwinner SoC support.
We (the linux-sunxi community) are currently making good progress on getting
Allwinner SoCs supported in the upstream kernel. It looks like the majority
of it will land in 3.14 and 3.15. As such I believe the time has come to
start supporting Allwinner SoCs with the official Fedora kernels ootb for
F-21.
I've written a Feature page for this, which you can find here:
https://fedoraproject.org/wiki/Changes/AllwinnerSunxiSupport
This will likely require carrying some kernel patches for 1-2 kernel releases,
until all the necessary bits for headless operation have landed upstream.
As always I'm willing to put my time were my work is, and I'll maintain these
patches for as long as necessary. To give you an idea of where things are
going, here is a break down of the tree I'm currently using for testing
and development, see:
https://github.com/linux-sunxi/linux-sunxi/commits/sunxi-devel
This branch contains 3.13-rc5
+ the following, which should all go upstream for 3.14:
https://github.com/mripard/linux/commits/sunxi-clk-for-3.13
https://github.com/mripard/linux/commits/sunxi-drivers-for-3.14
https://github.com/mripard/linux/commits/sunxi-core-for-3.14
https://github.com/mripard/linux/commits/sunxi-dt-for-3.14
sunxi-bits of:
https://git.linaro.org/people/daniel.lezcano/linux.git/shortlog/refs/head...
Emilio's "[PATCH v3 00/13] clk: sunxi: add PLL5 and PLL6 support" series
+ the following which is not quite ready for 3.14 yet, but people
are working hard to get it ready so hopefully we will see some of
it in 3.14, and the rest should make 3.15:
Emilio's sunxi-clk branch: patches not part of the above set
David's and mine sunxi-mmc work
Chen's gmac work
Oliver's ahci work
Arokux' ehci work
mripard's "ARM: sunxi: Add A31 High Speed Timer Support" series
wens' "v2 of A20 external output clock support patch series"
hansg's resistive touchscreen controller driver (rtp) including temp sensor support
Of the above list, we should at a minimum carry the mmc, gmac (gigabit nic) and
ahci patches if those don't get upstream in time. USB support would also be very nice,
but is not a must have. The last 3 patch-sets we don't need.
Regards,
Hans
9 years, 3 months
mmap_min_addr
by Eric Paris
I notice that on x86_64 we set
CONFIG_DEFAULT_MMAP_MIN_ADDR=4096
CONFIG_LSM_MMAP_MIN_ADDR=65536
I think we should be defaulting the DAC based protection to 64k as well
(or dropping the LSM value to 4k). I guess the Kconfig default is 4k
but testing when we wrote this feature said
ia64, ppc64 and x86 could safely be 64k
arm and maybe others should only be 32k
If it is safe to run with SELinux enforcing 64k it should be safe to run
with root/non-root enforcing 64k...
9 years, 4 months
Rawhide nodebug and 3.12 kernels heads up!
by Justin Forbes
Many of you have been using the Rawhide Nodebug repository to test the
3.12 stable kernels while we got Fedora 20 out the door. We really
appreciate the feedback from those who have tested. This is a heads up
that in one week (December 25th) the rawhide nodebug repository will
return to tracking rawhide and the 3.13 development cycle. Both Fedora
19 and Fedora 20 have the most recent 3.12.5 kernel submitted for
updates-testing. If you wish to remain on rawhide nodebug and test
development kernels, we really appreciate it. If you wish to remain on a
3.12 kernel, now is the time to disable the rawhide-nodebug repository.
To do so, simply flip the value for enabled from 1 to 0
in /etc/yum.repos.d/fedora-rawhide-kernel-nodebug.repo.
Thanks,
Justin
9 years, 5 months
3.12.6-200.fc19.x86_64: stacktrace on HP Elite 8300
by Reindl Harald
Hi
i recently installed the latest kernel-build from today
not i could say anything doe not work but maybe it's worth to look
at the dmesg-stacktrace at the bottom of my message, from my little
understanding it looks soundcard-related, however, MPD is playing
music fine -> no visible regressions
as far as i can say WLAN, network, firewall, services are all fine too
________________________________________________________________________________________
[root@srv-rhsoft:~]$ smbios-sys-info
Libsmbios version: 2.2.28
Product Name: HP Compaq Elite 8300 CMT
Vendor: Hewlett-Packard
BIOS Version: K01 v02.90
________________________________________________________________________________________
00:00.0 Host bridge: Intel Corporation Xeon E3-1200 v2/3rd Gen Core processor DRAM Controller (rev 09)
00:02.0 VGA compatible controller: Intel Corporation Xeon E3-1200 v2/3rd Gen Core processor Graphics Controller
(rev 09)
00:14.0 USB controller: Intel Corporation 7 Series/C210 Series Chipset Family USB xHCI Host Controller (rev 04)
00:19.0 Ethernet controller: Intel Corporation 82579LM Gigabit Network Connection (rev 04)
00:1a.0 USB controller: Intel Corporation 7 Series/C210 Series Chipset Family USB Enhanced Host Controller #2 (rev 04)
00:1b.0 Audio device: Intel Corporation 7 Series/C210 Series Chipset Family High Definition Audio Controller (rev 04)
00:1c.0 PCI bridge: Intel Corporation 7 Series/C210 Series Chipset Family PCI Express Root Port 1 (rev c4)
00:1c.4 PCI bridge: Intel Corporation 7 Series/C210 Series Chipset Family PCI Express Root Port 5 (rev c4)
00:1d.0 USB controller: Intel Corporation 7 Series/C210 Series Chipset Family USB Enhanced Host Controller #1 (rev 04)
00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev a4)
00:1f.0 ISA bridge: Intel Corporation Q77 Express Chipset LPC Controller (rev 04)
00:1f.2 SATA controller: Intel Corporation 7 Series/C210 Series Chipset Family 6-port SATA Controller [AHCI mode]
(rev 04)
00:1f.3 SMBus: Intel Corporation 7 Series/C210 Series Chipset Family SMBus Controller (rev 04)
01:00.0 Network controller: Qualcomm Atheros AR5418 Wireless Network Adapter [AR5008E 802.11(a)bgn] (PCI-Express)
(rev 01)
02:00.0 Ethernet controller: Intel Corporation 82574L Gigabit Network Connection
________________________________________________________________________________________
[ 3.837956] e1000e 0000:00:19.0 eth0: registered PHC clock
[ 3.838025] e1000e 0000:00:19.0 eth0: (PCI Express:2.5GT/s:Width x1) 24:be:05:1a:c0:27
[ 3.838133] e1000e 0000:00:19.0 eth0: Intel(R) PRO/1000 Network Connection
[ 3.838231] e1000e 0000:00:19.0 eth0: MAC: 10, PHY: 11, PBA No: 0100FF-0FF
[ 3.838309] ------------[ cut here ]------------
[ 3.838382] WARNING: CPU: 1 PID: 423 at lib/list_debug.c:59 __list_del_entry+0xa1/0xd0()
[ 3.838425] snd_hda_intel 0000:00:1b.0: irq 43 for MSI/MSI-X
[ 3.838491] list_del corruption. prev->next should be ffff8804060e2ba0, but was ffffffff81c98690
[ 3.838600] Modules linked in: coretemp snd_hda_intel(+) ath snd_hda_codec i2c_algo_bit snd_hwdep
crct10dif_pclmul mac80211 drm_kms_helper crc32_pclmul snd_seq drm cfg80211 usb_storage e1000e(+) crc32c_intel
hp_wmi snd_seq_device iTCO_wdt iTCO_vendor_support i2c_i801 ghash_clmulni_intel sparse_keymap snd_pcm rfkill
snd_page_alloc snd_timer microcode i2c_core ptp snd lpc_ich pps_core wmi serio_raw video mfd_core shpchp soundcore
raid1 raid10
[ 3.839691] CPU: 1 PID: 423 Comm: systemd-udevd Not tainted 3.12.6-200.fc19.x86_64 #1
[ 3.839758] Hardware name: Hewlett-Packard HP Compaq Elite 8300 CMT/3396, BIOS K01 v02.90 07/16/2013
[ 3.839868] 0000000000000009 ffff8804068eda98 ffffffff81662651 ffff8804068edae0
[ 3.840004] ffff8804068edad0 ffffffff810691dd ffff8804060e2ba0 ffff8804060e2bb8
[ 3.840187] ffff8804060e2bb8 0000000000000000 ffffffffa01a8c00 ffff8804068edb30
[ 3.840323] Call Trace:
[ 3.840387] [<ffffffff81662651>] dump_stack+0x45/0x56
[ 3.840453] [<ffffffff810691dd>] warn_slowpath_common+0x7d/0xa0
[ 3.840519] [<ffffffff8106924c>] warn_slowpath_fmt+0x4c/0x50
[ 3.840584] [<ffffffff8131aaa1>] __list_del_entry+0xa1/0xd0
[ 3.840648] [<ffffffff8131aadd>] list_del+0xd/0x30
[ 3.840713] [<ffffffff8132ee8c>] pinctrl_free+0xbc/0xe0
[ 3.840777] [<ffffffff8132f940>] ? pinctrl_put+0x30/0x30
[ 3.840842] [<ffffffff8132eec7>] pinctrl_release+0x17/0x20
[ 3.840906] [<ffffffff8132f969>] devm_pinctrl_release+0x29/0x30
[ 3.840972] [<ffffffff814029ca>] devres_release+0x2a/0x50
[ 3.841037] [<ffffffff8132f6d3>] devm_pinctrl_put+0x23/0x40
[ 3.841150] [<ffffffff81417e5e>] pinctrl_bind_pins+0xae/0x220
[ 3.841237] [<ffffffff813ff4ca>] driver_probe_device+0x6a/0x3a0
[ 3.841325] [<ffffffff813ff8d3>] __driver_attach+0x93/0xa0
[ 3.841396] [<ffffffff813ff840>] ? __device_attach+0x40/0x40
[ 3.841463] [<ffffffff813fd3a3>] bus_for_each_dev+0x63/0xa0
[ 3.841528] [<ffffffff813fef3e>] driver_attach+0x1e/0x20
[ 3.841593] [<ffffffff813feab0>] bus_add_driver+0x200/0x2d0
[ 3.841658] [<ffffffffa01b4000>] ? 0xffffffffa01b3fff
[ 3.841722] [<ffffffff813fff14>] driver_register+0x64/0xf0
[ 3.841786] [<ffffffffa01b4000>] ? 0xffffffffa01b3fff
[ 3.841851] [<ffffffff8133a74c>] __pci_register_driver+0x4c/0x50
[ 3.841921] [<ffffffffa01b4041>] e1000_init_module+0x41/0x1000 [e1000e]
[ 3.841988] [<ffffffff810020f2>] do_one_initcall+0xf2/0x1a0
[ 3.842069] [<ffffffff81054533>] ? set_memory_nx+0x43/0x50
[ 3.842176] [<ffffffff810d687d>] load_module+0x1b9d/0x2640
[ 3.842248] [<ffffffff810d2a30>] ? store_uevent+0x40/0x40
[ 3.842341] [<ffffffff810d7496>] SyS_finit_module+0x86/0xb0
[ 3.842417] [<ffffffff816716a9>] system_call_fastpath+0x16/0x1b
[ 3.842492] ---[ end trace 27592ade283d040b ]---
[ 3.842571] pcieport 0000:00:1c.4: driver skip pci_set_master, fix it!
9 years, 5 months
static dtb location
by Dennis Gilmore
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi all,
Back in August I started a thread[1] on the cross-distro list at linaro
on standardising the location of dtb files. While tehre was not a lot
fo input there was a consensus to use /boot/dtb/ as the location. since
the dtb generation is still tied to kernel building. I would like to
have us have the most recently installed kernel link its dtb directory
to /boot/dtb/ I am working on a change for Fedora 21 to get u-boot that
we provide loading the dtb automatically and there is no way that ive
worked out to have u-boot work out the path as they exist today.
I realise its not perfect and may cause issues if you try to boot an
older kernel. I would like some feedback on how best to implement it so
I can go off and write some patches for review.
Thanks in advance
Dennis
[1]
http://lists.linaro.org/pipermail/cross-distro/2013-August/000463.html
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)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=ke//
-----END PGP SIGNATURE-----
9 years, 5 months
NUMA, migrate/N, and tuned-adm
by David Timothy Strauss
We're running Fedora 19 with the 3.11.10-200.fc19.x86_64 kernel (just
the normal RPM) on large servers (128GB RAM over two NUMA regions,
each with one hex-core processor) with a large number of processes
(more than 700, a couple hundred of which are active fairly
frequently). We encounter situations where there system gets
overwhelmed with migrate/N tasks from the kernel, based on what we've
seen in top.
Here's what we've tried:
* tuned-adm on latency-performance and virtual-host profiles
* kernel.sched_migration_cost_ns=5000000 (which tuned will do for
those profiles in v3.3/Fedora 20)
* numad
Here's what we've used for analysis:
* powertop
* top/htop
* perf record -a -g
* SystemTap with code to print out migrations occurring
* numatop
All we know is that the migration storms correlate with concurrent
Chef runs verifying/configuring containers on the system.
Obviously, Chef invokes many things. But, most of the migrations we
see are for Chef, the Ruby interpreter, the sh interpreter, and Munin.
Responsiveness returns to normal after SystemTap reports a large set
of chef-solo migrations, presumably at their completion.
David Strauss
Pantheon Systems
Fedora Server Working Group
9 years, 5 months
Fedora Kernel Meeting Minutes for 2013-12-13
by Josh Boyer
==============================
#fedora-meeting: Fedora Kernel
==============================
Meeting started by jwb at 19:01:13 UTC. The full logs are available at
http://meetbot.fedoraproject.org/fedora-meeting/2013-12-13/fedora_kernel....
.
Meeting summary
---------------
* init (jwb, 19:01:37)
* release overview (jwb, 19:03:29)
* F18-F20 all on 3.11.10(ish) Rawhide on 3.13-rc3(ish) (jwb,
19:07:24)
* 3.12 rebase (jwb, 19:08:10)
* F18 will not get a 3.12 rebase. critical security fixes only until
EOL (jwb, 19:16:12)
* F19-F20 will get 3.12 rebase in updates-testing very soon (jwb,
19:16:27)
* rawhide-nodebug will go back to tracking rawhide (jwb, 19:17:36)
* ACTION: jforbes to give 1 week warning for nodebug when 3.12 hits
updates-testing (jwb, 19:17:53)
* Removal of kernel-doc (jwb, 19:19:19)
* LINK:
http://kojipkgs.fedoraproject.org//packages/kernel/3.13.0/0.rc3.git4.1.fc...
(jwb, 19:21:37)
* kernel-doc subpackage brings in asciidoc and xmlto, which bloat the
BuildRequires for the kernel significantly (jwb, 19:24:34)
* ACTION: jwb to drop kernel-doc subpackage (jwb, 19:24:44)
* Open Floor (jwb, 19:27:22)
* Trinity (jwb, 19:28:41)
* MMAP rework in trinity is surfacing a number of new bugs. Hopefully
this helps clear up some weird MM bugs we've seen in the past (jwb,
19:31:53)
* probably upstream trinity release next week (jwb, 19:37:15)
* Open Floor (jwb, 19:37:56)
Meeting ended at 19:46:07 UTC.
Action Items
------------
* jforbes to give 1 week warning for nodebug when 3.12 hits
updates-testing
* jwb to drop kernel-doc subpackage
Action Items, by person
-----------------------
* jforbes
* jforbes to give 1 week warning for nodebug when 3.12 hits
updates-testing
* jwb
* jwb to drop kernel-doc subpackage
* **UNASSIGNED**
* (none)
People Present (lines said)
---------------------------
* jwb (95)
* davej (30)
* jforbes (28)
* nirik (19)
* zodbot (3)
Generated by `MeetBot`_ 0.1.4
.. _`MeetBot`: http://wiki.debian.org/MeetBot
9 years, 5 months
Agenda for today's Kernel meeting
by Josh Boyer
Here's a brief agenda for today's meeting. If you have something
you'd like to add, please reply and we'll throw it on.
1) Release overview
2) 3.12 rebase for F20-F18
3) Removal of kernel-doc
4) Open Floor
9 years, 5 months
[PATCH 0/2] MODSIGN: Support MOKx
by Lee, Chun-Yi
It's base on the kernel patches of Fedora 20.
This patch set add the support to MODSIGN mechanism for revoke kernel
module by hash or public key. As MokListRT, EFI bootloader(e.g. shim)
should maintain the MokListXRT container the format the same with dbx.
The patches will check the hash of kernel module before load it.
Lee, Chun-Yi (2):
MODSIGN: check hash of kernel module in blacklist
MODSIGN: load hash blacklist of modules from MOKx
include/linux/efi.h | 12 ++++
kernel/modsign_uefi.c | 150 +++++++++++++++++++++++++++++++++++++++++++++-
kernel/module-internal.h | 14 ++++
kernel/module.c | 9 +++-
kernel/module_signing.c | 79 ++++++++++++++++++++++++
5 files changed, 261 insertions(+), 3 deletions(-)
9 years, 5 months