| 1 |
Hi there! |
| 2 |
|
| 3 |
Strange things are going on here. |
| 4 |
|
| 5 |
I've written here in the past about my performance problems. My dual-core |
| 6 |
had trouble playing movies without stuttering when there was I/O. It was |
| 7 |
mainly swapping that caused this, and 8 G were not enough for me running |
| 8 |
KDE4. |
| 9 |
|
| 10 |
Then my hardware broke, and I got new one, except for the system hard |
| 11 |
drive and the PSU. It's an AMD FX-4100 quad-core with 3.6 GHz, 16 G of |
| 12 |
RAM. Running gentoo-sources-3.2.1 as kernel. But it seems playing movies |
| 13 |
got even worse! |
| 14 |
|
| 15 |
The videos do not need to have high quality. When I do this, I get |
| 16 |
interruptions, sometimes for more than a whole second: |
| 17 |
|
| 18 |
# dd if=/dev/zero of=/tmp/argh bs=10M count=1000 |
| 19 |
|
| 20 |
My whole system is encrypted, but the same happens with unencrypted |
| 21 |
partitions. All are on LVM. When I write to another drive, there is no |
| 22 |
effect. Throughput is around 50-60 MB/s. |
| 23 |
|
| 24 |
Any ideas where to look? I think I'll create a completely fresh |
| 25 |
kernel .config with genkernel, maybe my own .config has some weird |
| 26 |
problem. But I tried similar things in the past already, getting a kernel |
| 27 |
from a live cd, to no effect. |
| 28 |
|
| 29 |
I put cache = 10240 into .mplayer/config to get 10 MB of video cached, |
| 30 |
but I see no effect. |
| 31 |
|
| 32 |
Playing music with Amarok is no problem. |
| 33 |
|
| 34 |
My SATA drives are in AHCI mode, here's some dmesg info about that: |
| 35 |
|
| 36 |
ahci 0000:00:11.0: version 3.0 |
| 37 |
ahci 0000:00:11.0: PCI INT A -> GSI 22 (level, low) -> IRQ 22 |
| 38 |
ahci 0000:00:11.0: AHCI 0001.0100 32 slots 4 ports 3 Gbps 0xf impl SATA |
| 39 |
mode |
| 40 |
ahci 0000:00:11.0: flags: 64bit ncq sntf ilck led clo pmp pio slum part |
| 41 |
ccc sxs |
| 42 |
scsi0 : ahci |
| 43 |
scsi1 : ahci |
| 44 |
scsi2 : ahci |
| 45 |
scsi3 : ahci |
| 46 |
ata1: SATA max UDMA/133 abar m1024@0xff70b000 port 0xff70b100 irq 22 |
| 47 |
ata2: SATA max UDMA/133 abar m1024@0xff70b000 port 0xff70b180 irq 22 |
| 48 |
ata3: SATA max UDMA/133 abar m1024@0xff70b000 port 0xff70b200 irq 22 |
| 49 |
ata4: SATA max UDMA/133 abar m1024@0xff70b000 port 0xff70b280 irq 22 |
| 50 |
ahci 0000:02:00.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19 |
| 51 |
ahci 0000:02:00.0: irq 43 for MSI/MSI-X |
| 52 |
ahci: SSS flag set, parallel bus scan disabled |
| 53 |
ahci 0000:02:00.0: AHCI 0001.0200 32 slots 2 ports 6 Gbps 0x3 impl SATA |
| 54 |
mode |
| 55 |
ahci 0000:02:00.0: flags: 64bit ncq sntf stag led clo pmp pio slum part |
| 56 |
ccc sxs |
| 57 |
ahci 0000:02:00.0: setting latency timer to 64 |
| 58 |
scsi4 : ahci |
| 59 |
scsi5 : ahci |
| 60 |
ata5: SATA max UDMA/133 abar m512@0xff600000 port 0xff600100 irq 43 |
| 61 |
ata6: SATA max UDMA/133 abar m512@0xff600000 port 0xff600180 irq 43 |
| 62 |
pata_atiixp 0000:00:14.1: PCI INT A -> GSI 16 (level, low) -> IRQ 16 |
| 63 |
scsi6 : pata_atiixp |
| 64 |
scsi7 : pata_atiixp |
| 65 |
ata7: PATA max UDMA/100 cmd 0x1f0 ctl 0x3f6 bmdma 0xf000 irq 14 |
| 66 |
ata8: PATA max UDMA/100 cmd 0x170 ctl 0x376 bmdma 0xf008 irq 15 |
| 67 |
|
| 68 |
(ata7/8 is the additional PATA controller, seen with the pata_atiixp |
| 69 |
driver. I have one drive there, but it is not being used.) |
| 70 |
|
| 71 |
When my new girl-friend comes over and we want to watch a movie, and it |
| 72 |
stutters... she will ask why I don't simply use Windows to get |
| 73 |
better performance, her five year old PC would do this just fine. Wat do |
| 74 |
I tell her? WHAT DO I TELL HER?? |
| 75 |
|
| 76 |
|
| 77 |
And then there's what happened yesterday. A world update was going on, |
| 78 |
with libreoffice, firefox, wine and thunderbird emerging in parallel, all |
| 79 |
big packages. I have the PORTAGE_TMPDIR on a 5GB tmpfs, only libreoffice |
| 80 |
is being compiled on disk. Suddenly, my system became very unresponsive, |
| 81 |
the mouse had disappeared, the KDE widgets did not update, and xosview |
| 82 |
showed a load of 23. All 4 cores were at 100%, the type of usage was |
| 83 |
io-wait. How can I find out in such a case which processes are waiting |
| 84 |
for I/O? top showed nothing. The Ctrl-Esc task viewer of KDE showed some |
| 85 |
processes being 'inactive on hard drive', does this men those are the |
| 86 |
waiting tasks? They varied, they were mostly Akonadi stuff. I stopped |
| 87 |
akonadi, and after a while the load dropped. But this may be a |
| 88 |
coincidence. |
| 89 |
|
| 90 |
After all had calmed down, I had 2G of swap in use. 16G total RAM, all |
| 91 |
being used of course, but only 8G being needed according to the -/+ |
| 92 |
buffers/cache line in free -m, the other 8G are cache. Does my Linux |
| 93 |
somehow prefer to have this much cache, even if tmpfs stuff gets put into |
| 94 |
swap? I have vm.swappiness = 0 in /etc/sysctl.conf. |
| 95 |
|
| 96 |
Is there a command to show me what processes the memory in swap belongs |
| 97 |
to? |
| 98 |
|
| 99 |
Wonko |
Archives