DMA not enabled for IDE disks

Bug #55929 reported by Gustavo Carneiro
14
Affects Status Importance Assigned to Milestone
linux-source-2.6.15 (Baltix)
Invalid
Undecided
Unassigned
linux-source-2.6.15 (Ubuntu)
Fix Released
Undecided
Ben Collins
linux-source-2.6.17 (Ubuntu)
Fix Released
Undecided
Ben Collins

Bug Description

I just bought a new motherboard[1] and CPU[2], and now DMA doesn't work for my IDE drives:

root@emperor:/home/gjc # hdparm -d 1 /dev/hda

/dev/hda:
 setting using_dma to 1 (on)
 HDIO_SET_DMA failed: Operation not permitted
 using_dma = 0 (off)

root@emperor:/home/gjc # dmesg |grep DMA
[17179569.184000] DMA zone: 4096 pages, LIFO batch:0
[17179576.032000] VP_IDE: Unknown VIA SouthBridge, disabling DMA.
[17179576.104000] ata1: SATA max UDMA/133 cmd 0xE000 ctl 0xDC02 bmdma 0xD480 irq 193
[17179576.104000] ata2: SATA max UDMA/133 cmd 0xD880 ctl 0xD802 bmdma 0xD488 irq 193
[17179592.952000] parport0: PC-style at 0x378 (0x778), irq 7, dma 3 [PCSPP,TRISTATE,COMPAT,EPP,ECP,DMA]

(notice the "VP_IDE: Unknown VIA SouthBridge, disabling DMA." line)

[1] ASROCK Chip.VIA PT880Pro SK775 AGP / PCI-E- 775DUAL-VSTA
[2] Intel CORE 2 DUO E6300

Revision history for this message
Gustavo Carneiro (gjc) wrote :
Revision history for this message
Matthew Garrett (mjg59) wrote :

Just to make sure I don't get things confused - you have a SATA motherboard, but your hard drive is still connected via traidtional IDE, right?

Revision history for this message
Matthew Garrett (mjg59) wrote :

And can you attach the output of the lspci command, please?

Revision history for this message
Matthew Garrett (mjg59) wrote :

Ben - at a guess it's this patch: http://bugzilla.kernel.org/attachment.cgi?id=8646&action=view . Mind including it? Looks like it should be safe.

Revision history for this message
Ben Collins (ben-collins) wrote :

Patch added for next dapper update. Will be included in edgy too.

Changed in linux-source-2.6.17:
assignee: nobody → ben-collins
status: Unconfirmed → Fix Committed
Revision history for this message
Gustavo Carneiro (gjc) wrote :

Probably no longer relevant, but here's the requested info.

Yes, my MB supports both SATA and PATA but I'm using only PATA disks (what I had before).

Revision history for this message
Ben Collins (ben-collins) wrote :

Just to make sure we get this right, can you add the output of "lspci -n" as well?

Thanks

Revision history for this message
Gustavo Carneiro (gjc) wrote :
Revision history for this message
Gustavo Carneiro (gjc) wrote :

I just received a new kernel update, 2.6.17-6-686, but the problem remains:

[17179572.928000] VP_IDE: Unknown VIA SouthBridge, disabling DMA.

Changed in linux-source-2.6.17:
status: Fix Committed → Unconfirmed
Revision history for this message
Gustavo Carneiro (gjc) wrote :

I compiled my own kernel, with apt-get install linux-source-2.6.17 and some other commands, while applying the patch. Looks like the -6 kernel in ubuntu edgy didn't include the patch, and looks like the patch fixes my problem completely (hdparm -tT goes from 3 to 25 MB/s).

Are you sure the patch was included in the ubuntu kernel tree?

Revision history for this message
Gustavo Carneiro (gjc) wrote :

With the patched kernel, i'm getting a couple of error messages during boot, though everything appears to be working fine:

[17179609.028000] hda: dma_timer_expiry: dma status == 0x60
[17179609.028000] hda: DMA timeout retry
[17179609.028000] hda: timeout waiting for DMA
[17179609.240000] hda: status error: status=0x58 { DriveReady SeekComplete DataR
equest }
[17179609.240000] ide: failed opcode was: unknown
[17179609.240000] hda: drive not ready for command
--
[17179609.528000] hda: set_drive_speed_status: status=0x58 { DriveReady SeekComp
lete DataRequest }
[17179609.528000] ide: failed opcode was: unknown
[17179629.512000] hda: dma_timer_expiry: dma status == 0x60
[17179629.512000] hda: DMA timeout retry
[17179629.512000] hda: timeout waiting for DMA

Revision history for this message
Ben Collins (ben-collins) wrote :

It's fix committed, so it's in the repo, just not uploaded yet.

Changed in linux-source-2.6.15:
status: Unconfirmed → Rejected
assignee: nobody → ben-collins
status: Unconfirmed → Fix Committed
Changed in linux-source-2.6.17:
status: Unconfirmed → Fix Committed
Revision history for this message
Gustavo Carneiro (gjc) wrote :

Just a heads up: the DMA problem resurfaced in the 2.6.19 kernels in feisty.

Revision history for this message
Gustavo Carneiro (gjc) wrote :

It is strange because the patch indicated in comment 4 is already applied to the the 2.6.19 source tree (I checked). Also there is no "disabling DMA" message like before. However, the fact is that DMA is turned off, and hdparm doesn't allow me to turn it on.

Revision history for this message
Gustavo Carneiro (gjc) wrote :

OK, maybe I was confused, but it seems 2.6.19-7 is not detecting my hard drive as IDE, it is mapping it as a scsi drive, /dev/sda. And although hdparm -d doesn't work with scsi devices I can notice that DMA is indeed enabled.

So nevermind my last comment.

Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

This report was marked 'Fix Committed' a while ago and there hasn't been any activity in it recently. I'm going to go ahead and mark the status as 'Fix Released'. If this is not the case, please retest against the latest Hardy Alpha release and report back your results. Thanks in advance.

Changed in linux-source-2.6.17:
status: Fix Committed → Fix Released
Changed in linux-source-2.6.15:
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.