2007-06-27 04:17:35

by Dan

[permalink] [raw]
Subject: PROBLEM: SATA AHCI driver significant throughput loss in 2.6.22-rc5-g75154f40-dirty

Something changed with the AHCI SATA driver in the latest git version.
With 2.6.21 I get ~49MB/sec according to hdparm. Using the latest git
version I only get ~8MB/sec. I've attached my .config and have
included some info about my system below. Please let me know if you
need more information or if I can help test anything.

Lenovo Thinkpad T60p

kernel information:
===================
Linux lappy 2.6.22-rc5-g75154f40-dirty #1 SMP PREEMPT Sat Jun 23
21:14:49 EDT 2007 i686 Genuine Intel(R) CPU T2600 @ 2.16GHz
GenuineIntel GNU/Linux

Gnu C 4.1.2
Gnu make 3.81
binutils 2.17
util-linux 2.12r
mount 2.12r
module-init-tools 3.2.2
e2fsprogs 1.39
Linux C Library > libc.2.5
Dynamic linker (ldd) 2.5
Procps 3.2.7
Net-tools 1.60
Kbd 1.12
Sh-utils 6.9
udev 104
wireless-tools 28
Modules Loaded fglrx agpgart af_packet snd_pcm_oss
snd_mixer_oss ipw3945 ieee80211 ieee80211_crypt snd_hda_intel snd_pcm
snd_timer snd soundcore snd_page_alloc e1000 psmouse evdev unix



hdparm -i /dev/sda
==================
/dev/sda:

Model=HTS721010G9SA00 , FwRev=MCZIC10V,
SerialNo= MPDZN7Y0HKTYZL
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
BuffType=DualPortCache, BuffSize=7538kB, MaxMultSect=16, MultSect=?16?
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=195371568
IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5
AdvancedPM=yes: mode=0x80 (128) WriteCache=enabled
Drive conforms to: ATA/ATAPI-7 T13 1532D revision 1: ATA/ATAPI-2,3,4,5,6,7

cat /proc/modules
=================
fglrx 649632 15 - Live 0xf8a90000 (P)
agpgart 30412 1 fglrx, Live 0xf894e000
af_packet 23816 2 - Live 0xf893c000
snd_pcm_oss 40736 0 - Live 0xf88eb000
snd_mixer_oss 17536 1 snd_pcm_oss, Live 0xf88b4000
ipw3945 201124 1 - Live 0xf8909000
ieee80211 33608 1 ipw3945, Live 0xf88e1000
ieee80211_crypt 8704 1 ieee80211, Live 0xf887d000
snd_hda_intel 243864 0 - Live 0xf8958000
snd_pcm 68996 2 snd_pcm_oss,snd_hda_intel, Live 0xf88f7000
snd_timer 22148 1 snd_pcm, Live 0xf8872000
snd 45412 5 snd_pcm_oss,snd_mixer_oss,snd_hda_intel,snd_pcm,snd_timer,
Live 0xf8881000
soundcore 10208 1 snd, Live 0xf885a000
snd_page_alloc 11400 2 snd_hda_intel,snd_pcm, Live 0xf884b000
e1000 118080 0 - Live 0xf8896000
psmouse 38416 0 - Live 0xf8867000
evdev 11648 0 - Live 0xf8847000
unix 29584 62 - Live 0xf8851000


Attachments:
(No filename) (2.63 kB)
.config (46.45 kB)
lspci-vvv (21.12 kB)
proc-cpuinfo (1.09 kB)
proc-iomem (1.55 kB)
proc-ioports (1.27 kB)
Download all attachments

2007-06-29 13:37:17

by Michal Piotrowski

[permalink] [raw]
Subject: Re: PROBLEM: SATA AHCI driver significant throughput loss in 2.6.22-rc5-g75154f40-dirty

Hi,

[adding linux-ide to cc]

On 27/06/07, Dan <[email protected]> wrote:
> Something changed with the AHCI SATA driver in the latest git version.
> With 2.6.21 I get ~49MB/sec according to hdparm. Using the latest git
> version I only get ~8MB/sec. I've attached my .config and have
> included some info about my system below. Please let me know if you
> need more information or if I can help test anything.
>
> Lenovo Thinkpad T60p
>
> kernel information:
> ===================
> Linux lappy 2.6.22-rc5-g75154f40-dirty #1

Are you using any additional patches?

> SMP PREEMPT Sat Jun 23
> 21:14:49 EDT 2007 i686 Genuine Intel(R) CPU T2600 @ 2.16GHz
> GenuineIntel GNU/Linux
>
> Gnu C 4.1.2
> Gnu make 3.81
> binutils 2.17
> util-linux 2.12r
> mount 2.12r
> module-init-tools 3.2.2
> e2fsprogs 1.39
> Linux C Library > libc.2.5
> Dynamic linker (ldd) 2.5
> Procps 3.2.7
> Net-tools 1.60
> Kbd 1.12
> Sh-utils 6.9
> udev 104
> wireless-tools 28
> Modules Loaded fglrx agpgart af_packet snd_pcm_oss
> snd_mixer_oss ipw3945 ieee80211 ieee80211_crypt snd_hda_intel snd_pcm
> snd_timer snd soundcore snd_page_alloc e1000 psmouse evdev unix
>
>
>
> hdparm -i /dev/sda
> ==================
> /dev/sda:
>
> Model=HTS721010G9SA00 , FwRev=MCZIC10V,
> SerialNo= MPDZN7Y0HKTYZL
> Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
> RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
> BuffType=DualPortCache, BuffSize=7538kB, MaxMultSect=16, MultSect=?16?
> CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=195371568
> IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
> PIO modes: pio0 pio1 pio2 pio3 pio4
> DMA modes: mdma0 mdma1 mdma2
> UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5
> AdvancedPM=yes: mode=0x80 (128) WriteCache=enabled
> Drive conforms to: ATA/ATAPI-7 T13 1532D revision 1: ATA/ATAPI-2,3,4,5,6,7
>
> cat /proc/modules
> =================
> fglrx 649632 15 - Live 0xf8a90000 (P)
> agpgart 30412 1 fglrx, Live 0xf894e000
> af_packet 23816 2 - Live 0xf893c000
> snd_pcm_oss 40736 0 - Live 0xf88eb000
> snd_mixer_oss 17536 1 snd_pcm_oss, Live 0xf88b4000
> ipw3945 201124 1 - Live 0xf8909000
> ieee80211 33608 1 ipw3945, Live 0xf88e1000
> ieee80211_crypt 8704 1 ieee80211, Live 0xf887d000
> snd_hda_intel 243864 0 - Live 0xf8958000
> snd_pcm 68996 2 snd_pcm_oss,snd_hda_intel, Live 0xf88f7000
> snd_timer 22148 1 snd_pcm, Live 0xf8872000
> snd 45412 5 snd_pcm_oss,snd_mixer_oss,snd_hda_intel,snd_pcm,snd_timer,
> Live 0xf8881000
> soundcore 10208 1 snd, Live 0xf885a000
> snd_page_alloc 11400 2 snd_hda_intel,snd_pcm, Live 0xf884b000
> e1000 118080 0 - Live 0xf8896000
> psmouse 38416 0 - Live 0xf8867000
> evdev 11648 0 - Live 0xf8847000
> unix 29584 62 - Live 0xf8851000
>
>

Regards,
Michal

--
LOG
http://www.stardust.webpages.pl/log/

2007-06-30 07:58:08

by Andrew Morton

[permalink] [raw]
Subject: Re: PROBLEM: SATA AHCI driver significant throughput loss in 2.6.22-rc5-g75154f40-dirty

On Wed, 27 Jun 2007 00:17:24 -0400 Dan <[email protected]> wrote:

> Something changed with the AHCI SATA driver in the latest git version.
> With 2.6.21 I get ~49MB/sec according to hdparm. Using the latest git
> version I only get ~8MB/sec. I've attached my .config and have
> included some info about my system below. Please let me know if you
> need more information or if I can help test anything.
>
> Lenovo Thinkpad T60p

Until we know how widespread this damage is, we'd need to consider this
a blocking bug.

Are you able (as a matter of some urgency) to run a git bisection
search to identify the offending commit?

Thanks.