ide retry on 2.4.5-ac7
Hi
Since my first try on 2.4 kernel, I had trouble with DMA when I select activate on boot time because it selects udma4, whereas my HD is only able to do udma2. I correct that with hdparm, but I was quite happy of the patch in ac4 whixh detect ide lost interrupt and retry with a value lower of dma. But it seems that this patch does not work anymore in ac7?? I can not see in the changelog that you come back or made other change in ide?? Should it work in the same way or not?
My hardware: via 686a of course, with Athlon 500 on a k7m MB
Thanks
Mickael
PS: CC to [email protected] please.
Signature · 2 lines
-- Unix IS user-friendly. It just chooses its friends carefully
Quoted 11 lines · PROFETA Mickael
Hi
Since my first try on 2.4 kernel, I had trouble with DMA when I select activate on boot time because it selects udma4, whereas my HD is only able to do udma2. I correct that with hdparm, but I was quite happy of the patch in ac4 whixh detect ide lost interrupt and retry with a value lower of dma. But it seems that this patch does not work anymore in ac7?? I can not see in the changelog that you come back or made other change in ide?? Should it work in the same way or not?
Quoted 1 line
My hardware: via 686a of course, with Athlon 500 on a k7m MB
It worked sucessfully for you in 2.4.5-ac4 but not in -ac7? I can't see any changes to the patch, so more details on the nature of the problem would be helpful.
Signature · 1 line
Jens Axboe
Quoted 4 lines · Jens Axboe
It worked sucessfully for you in 2.4.5-ac4 but not in -ac7? I can't see any changes to the patch, so more details on the nature of the problem would be helpful.
Ok, this is the results of a hdparm -tT on my second hard disk:
hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hdc: dma_intr: error=0x84 { DriveStatusError BadCRC }
hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hdc: dma_intr: error=0x84 { DriveStatusError BadCRC }
ide1: reset: success
64 MB in 7.03 seconds = 9.10 MB/sec
and then the kernel has shifted to udma3 and I have no more error in this session with that hard disk
I tried the same thing on -ac7 and I always have the errors, the kernel does not reset ide... Looking at the different change on ac patches, I can not see why...
Mike
Quoted 3 lines
Signature · 2 lines
-- Jens Axboe
Quoted 15 lines · PROFETA Mickael
Quoted 4 lines · Jens Axboe
It worked sucessfully for you in 2.4.5-ac4 but not in -ac7? I can't see any changes to the patch, so more details on the nature of the problem would be helpful.
Ok, this is the results of a hdparm -tT on my second hard disk:
hdc:
dma_intr: status=0x51 { DriveReady SeekComplete Error }
hdc:dma_intr: error=0x84 { DriveStatusError BadCRC }
hdc:dma_intr: status=0x51 { DriveReady SeekComplete Error }
hdc:dma_intr: error=0x84 { DriveStatusError BadCRC }
ide1: reset: success
64 MB in 7.03 seconds = 9.10 MB/sec
Quoted 6 lines
and then the kernel has shifted to udma3 and I have no more error in this session with that hard disk
I tried the same thing on -ac7 and I always have the errors, the kernel does not reset ide... Looking at the different change on ac patches, I can not see why...
This is not the case that is attempted solve. The above could be a cable error (it looks like it). These are usually genuine and indicate a real hw problem.
Signature · 1 line
Jens Axboe
Quoted 4 lines · Jens Axboe
This is not the case that is attempted solve. The above could be a cable error (it looks like it). These are usually genuine and indicate a real hw problem.
I know about that, but I tried with other cable and the trouble leaves the same. On the other hand, before kernel 2.4.2 I think, (not sure of the number) the kernel does not want to activate UDMA on startup even with the option activated, and this on via mb. At that time, I used to force the DMA with hdparm, and I had no troubles in dma mode 3 with that hard disk, and I can remeber that udma mode 4 was even not indicate when I try hdparm -I /dev/hdc, so I don't know why it detects dma 4, and I did not manage to change that at boot time.
Another example is the hda, which is an ibm DTLA hard drive. In 2.4.2 a 2.4.3 he made the same error in udma mode4, and since 2.4.4 no troubles anymore... I find it difficult to believe that it is hardware trouble! Furthemore, I began this thread because I heard someone else on this mailing list with the same crc error and I thought the patch in ac4 was in a way to solve the dma detect fails.
thanks for your help
Mike
Quoted 3 lines
Signature · 2 lines
-- Jens Axboe