View previous topic :: View next topic |
Author |
Message |
Demontager n00b
Joined: 31 Oct 2010 Posts: 52 Location: Kherson Ukraine
|
Posted: Sun May 01, 2011 10:15 pm Post subject: SSD drive making system slow |
|
|
I got unknown problem with SSD drive it has Gentoo installed on it. The problem is - when PC not in use( system idle), after back to normal, decreased performance occurred, e.g. gnome interface become running slow, video playing slow, and games which runs from WINE same story.
just for test i have installed Ubuntu 11.04 and left PC idle for night then I started to test interface, movies and Games- everything was normal. But when i mounted partition on SSD drive- all became slow and dmesg has these lines afterwards:
Code: |
[28697.852920] EXT4-fs (sda1): mounted filesystem with ordered data mode. Opts:
(null)
[28730.150690] irq 16: nobody cared (try booting with the "irqpoll" option)
[28730.150692] Pid: 0, comm: swapper Tainted: P 2.6.38-8-generic
#42-Ubuntu
[28730.150693] Call Trace:
[28730.150694] <IRQ> [<ffffffff810d511b>] ?
__report_bad_irq.clone.2+0x2b/0xa0
[28730.150700] [<ffffffff810d551a>] ? note_interrupt+0x19a/0x1e0
[28730.150701] [<ffffffff810d640d>] ? handle_fasteoi_irq+0xdd/0x110
[28730.150703] [<ffffffff8100e9c2>] ? handle_irq+0x22/0x40
[28730.150706] [<ffffffff815caebd>] ? do_IRQ+0x5d/0xe0
[28730.150707] [<ffffffff815c3213>] ? ret_from_intr+0x0/0x15
[28730.150708] <EOI> [<ffffffff814a3af6>] ? poll_idle+0x46/0x80
[28730.150711] [<ffffffff814a3ac3>] ? poll_idle+0x13/0x80
[28730.150713] [<ffffffff814a4e89>] ? menu_select+0x149/0x340
[28730.150714] [<ffffffff814a3bda>] ? cpuidle_idle_call+0xaa/0x1b0
[28730.150717] [<ffffffff8100a266>] ? cpu_idle+0xa6/0xf0
[28730.150718] [<ffffffff815a9205>] ? rest_init+0x75/0x80
[28730.150721] [<ffffffff81acac8b>] ? start_kernel+0x3f5/0x400
[28730.150723] [<ffffffff81aca388>] ? x86_64_start_reservations+0x132/0x136
[28730.150724] [<ffffffff81aca253>] ? zap_identity_mappings+0x3e/0x41
[28730.150726] [<ffffffff81aca458>] ? x86_64_start_kernel+0xcc/0xdb
[28730.150726] handlers:
[28730.150727] [<ffffffffa007d020>] (rtl8169_interrupt+0x0/0x250 [r8169])
[28730.150733] [<ffffffffa05fce90>] (nv_kern_isr+0x0/0x80 [nvidia])
[28730.150818] Disabling IRQ #16
[28877.369428] EXT4-fs (sda3): mounted filesystem with ordered data mode. Opts:
(null)
|
then i found almost same lines in Gentoo
Code: |
[ 6386.385383] irq 16: nobody cared (try booting with the "irqpoll" option)
[ 6386.385385] Pid: 0, comm: swapper Tainted: P 2.6.38-gentoo-r3
#3
[ 6386.385386] Call Trace:
[ 6386.385387] <IRQ> [<ffffffff8108ff66>] ? __report_bad_irq+0x26/0xa0
[ 6386.385391] [<ffffffff8109016c>] ? note_interrupt+0x18c/0x1d0
[ 6386.385393] [<ffffffff81090add>] ? handle_fasteoi_irq+0xad/0xe0
[ 6386.385395] [<ffffffff8100555d>] ? handle_irq+0x1d/0x30
[ 6386.385396] [<ffffffff81004a78>] ? do_IRQ+0x58/0xe0
[ 6386.385398] [<ffffffff8174c793>] ? ret_from_intr+0x0/0xe
[ 6386.385398] <EOI> [<ffffffff81450065>] ?
acpi_hw_read_multiple+0x2a/0x63
[ 6386.385402] [<ffffffff8145fa81>] ? acpi_idle_do_entry+0x36/0x57
[ 6386.385403] [<ffffffff8145faf2>] ? acpi_idle_enter_c1+0x50/0x99
[ 6386.385405] [<ffffffff8157a4a2>] ? menu_select+0x142/0x350
[ 6386.385407] [<ffffffff81579502>] ? cpuidle_idle_call+0x82/0xf0
[ 6386.385408] [<ffffffff81001ee9>] ? cpu_idle+0xa9/0x100
[ 6386.385410] [<ffffffff8173704d>] ? rest_init+0x6d/0x80
[ 6386.385412] [<ffffffff81a3ed7e>] ? start_kernel+0x3a7/0x3b2
[ 6386.385414] [<ffffffff81a3e321>] ?
x86_64_start_reservations+0x131/0x135
[ 6386.385415] [<ffffffff81a3e417>] ? x86_64_start_kernel+0xf2/0xf9
[ 6386.385416] handlers:
[ 6386.385417] [<ffffffff81502c80>] (usb_hcd_irq+0x0/0x70)
[ 6386.385420] [<ffffffffa0084bc0>] (rtl8169_interrupt+0x0/0x3a0 [r8169])
[ 6386.385423] [<ffffffffa05e3d40>] (nv_kern_isr+0x0/0x80 [nvidia])
[ 6386.385481] Disabling IRQ #16
|
And after that - all slow again. To back system to normal state i used to reboot it. And the most biggest problem - upper mentioned lines used to appear sometimes while system boots and then -complete hangs.
I don't know what to do - more likely i need shift to normal HDD, but not so desirable... |
|
Back to top |
|
|
Hu Moderator
Joined: 06 Mar 2007 Posts: 21633
|
Posted: Sun May 01, 2011 11:23 pm Post subject: |
|
|
Can you reproduce the problem with an untainted kernel? Does the problem go away if you use the irqpoll option as the output suggests? |
|
Back to top |
|
|
whiteghost Guru
Joined: 26 Jul 2009 Posts: 374 Location: north dakota
|
Posted: Sun May 01, 2011 11:40 pm Post subject: |
|
|
i can't decipher your dmesg but i can tell you how i set up my ssd
do you have the discard option in fstab? for trim
Code: | /dev/sda2 / ext4 noatime,discard 0 1
/dev/sda3 /home ext4 noatime,discard 0 0 |
also when configuring my kernel i use i/o scheduler = noop
my demsg
Code: | 0.973992] scsi 0:0:0:0: Direct-Access ATA OCZ-AGILITY2 1.32 PQ: 0 ANSI: 5
[ 0.974334] sd 0:0:0:0: [sda] 117231408 512-byte logical blocks: (60.0 GB/55.8 GiB)
[ 0.974355] sd 0:0:0:0: Attached scsi generic sg0 type 0
[ 0.974708] sd 0:0:0:0: [sda] Write Protect is off
[ 0.974821] sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00
[ 0.974837] sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[ 0.975310] sda: sda1 sda2 sda3 sda4
[ 0.975618] sd 0:0:0:0: [sda] Attached SCSI disk
|
demsg Code: | 7.750170] EXT4-fs (sda2): re-mounted. Opts: discard
[ 7.791798] EXT4-fs (sda3): warning: maximal mount count reached, running e2fsck is recommended
[ 7.792318] EXT4-fs (sda3): mounted filesystem with ordered data mode. Opts: |
_________________ www.informationclearinghouse.info
May you re-discover what the poor in 18th century France discovered, that rich people's heads can be mechanically separated from their shoulders if they refuse to listen to reason. |
|
Back to top |
|
|
Demontager n00b
Joined: 31 Oct 2010 Posts: 52 Location: Kherson Ukraine
|
Posted: Mon May 02, 2011 7:06 am Post subject: |
|
|
Tainted kernel means that i have proprietary drivers installed e.g. nvidia, of course, i can uninstall its and try again.
Full dmesg http://pastebin.com/PzBtyj1c You can found that same bug occurred down, but also involves IRQ 17
I didn't put any special settings for SSD, it configured like ordinary drive
Code: |
UUID=6516f268-6c7a-438d-9484-ae7b3577e940 /boot ext4 noatime 1 2
UUID=89e821b4-2db5-43a2-85f2-f21dbdbe7567 / ext4 noatime 0 1
UUID=5e623e00-b433-4cc5-98eb-17e33568a0bf none swap sw 0 0
UUID=2ef70d09-ed88-4351-9999-8ec84733fe5a /home ext4 noatime 0 0
|
all partitions except /home located on SSD
current kernel config http://pastebin.com/z2Ruyg3s
I will do yours suggested settings and report result here afterwards |
|
Back to top |
|
|
whiteghost Guru
Joined: 26 Jul 2009 Posts: 374 Location: north dakota
|
Posted: Mon May 02, 2011 12:27 pm Post subject: |
|
|
i saw some interesting things googling 'disabling irq #16'
i'll let you look that up.
also you may look at kernel-seeds.org to compare against how your kernel is set up. _________________ www.informationclearinghouse.info
May you re-discover what the poor in 18th century France discovered, that rich people's heads can be mechanically separated from their shoulders if they refuse to listen to reason. |
|
Back to top |
|
|
Demontager n00b
Joined: 31 Oct 2010 Posts: 52 Location: Kherson Ukraine
|
Posted: Mon May 02, 2011 6:04 pm Post subject: |
|
|
Well, i done these things-
1.put in fstab
Code: |
UUID=6516f268-6c7a-438d-9484-ae7b3577e940 /boot ext4 noatime,discard 1 2
UUID=89e821b4-2db5-43a2-85f2-f21dbdbe7567 / ext4 noatime,discard 0 0
UUID=5e623e00-b433-4cc5-98eb-17e33568a0bf none swap sw 0 0
UUID=2ef70d09-ed88-4351-9999-8ec84733fe5a /home ext4 noatime 0 0
|
2. then enabled noop as default scheduler
CONFIG_DEFAULT_NOOP=y
CONFIG_DEFAULT_IOSCHED="noop"
full config http://pastebin.com/XfM0RxJf
3. added irqpoll to grub2 kernel options
GRUB_CMDLINE_LINUX_DEFAULT="quiet irqpoll rootfstype=ext4"
And... alas, again happened:
Code: |
[ 2788.745720] irq 16: nobody cared (try booting with the "irqpoll" option)
[ 2788.745722] Pid: 15784, comm: nvidia-settings Tainted: P 2.6.38-gentoo-r3 #4
[ 2788.745723] Call Trace:
[ 2788.745724] <IRQ> [<ffffffff8108ff66>] ? __report_bad_irq+0x26/0xa0
[ 2788.745728] [<ffffffff8109016c>] ? note_interrupt+0x18c/0x1d0
[ 2788.745729] [<ffffffff81090add>] ? handle_fasteoi_irq+0xad/0xe0
[ 2788.745731] [<ffffffff8100555d>] ? handle_irq+0x1d/0x30
[ 2788.745733] [<ffffffff81004a78>] ? do_IRQ+0x58/0xe0
[ 2788.745735] [<ffffffff81747ad3>] ? ret_from_intr+0x0/0xe
[ 2788.745736] <EOI> [<ffffffff81002d7b>] ? system_call_fastpath+0x16/0x1b
[ 2788.745737] handlers:
[ 2788.745738] [<ffffffff814fdfb0>] (usb_hcd_irq+0x0/0x70)
[ 2788.745740] [<ffffffffa0014bc0>] (rtl8169_interrupt+0x0/0x3a0 [r8169])
[ 2788.745743] [<ffffffffa05dbd40>] (nv_kern_isr+0x0/0x80 [nvidia])
[ 2788.745801] Disabling IRQ #16
|
full dmesg output http://pastebin.com/k9krGEJi
Now looking for "disabling irq16"... |
|
Back to top |
|
|
Demontager n00b
Joined: 31 Oct 2010 Posts: 52 Location: Kherson Ukraine
|
Posted: Tue May 03, 2011 12:48 pm Post subject: |
|
|
After searched "disable IRQ16" , didn't find any exact solution how to handle this error.
For now i have disabled all non essential hardware in BIOS, because i found that few devices uses one IRQ16.
This one before disabling:
Code: |
16: 836485 0 0 0 0 0 0 0 IO-APIC-fasteoi xhci_hcd:usb4, eth0, nvidia
|
and after disabling Firewire, first Realtek network card , USB3.0, Bluetooth
Code: |
16: 12375 0 0 0 0 0 0 0 IO-APIC-fasteoi nvidia
|
Till now nothing arise, will keep it more to make sure... |
|
Back to top |
|
|
|