On Thu, 8 Nov 2007 09:16:35 +0200
Denys Fedoryshchenko [EMAIL PROTECTED] wrote:
Does it work as kernel parameter?
I tried libata_dma_mask=0x4 and to set 0xf or 0xff - doesn't help. How to
disable DMA in libata, if it is compiled in kernel?
libata.dma_mask=3
will leave you with CD and disk
On Thu, Nov 08, 2007 at 06:44:31PM +0200, Denys Fedoryshchenko wrote:
Doesn't help
WRAP ~ #cat /proc/cmdline
console=ttyS0,38400n8 libata.dma_mask=3
It's libata.dma if its built into the kernel, or 'dma' module option
if built as a kernel module.
Jeff
-
To unsubscribe from this
On Thu, 8 Nov 2007 21:56:32 +0200
Denys Fedoryshchenko [EMAIL PROTECTED] wrote:
Thanks, it works like that.
Seems in libata there is no fall-back to non-DMA mode, if DMA didn't work.
It should be falling back from UDMA or MWDMA to PIO, if not please file a
bug
Alan
-
To unsubscribe from
Thanks, it works like that.
Seems in libata there is no fall-back to non-DMA mode, if DMA didn't work.
On Thu, 8 Nov 2007 12:31:39 -0500, Jeff Garzik wrote
On Thu, Nov 08, 2007 at 06:44:31PM +0200, Denys Fedoryshchenko wrote:
Doesn't help
WRAP ~ #cat /proc/cmdline
console=ttyS0,38400n8
Denys Fedoryshchenko wrote:
Thanks, it works like that.
Seems in libata there is no fall-back to non-DMA mode, if DMA didn't work.
There is, it's just too conservative about that. With improvements
pending for 2.6.24, it should be quite snappy at falling back to PIO if
configured transfer
Denys Fedoryshchenko wrote:
On Tue, 6 Nov 2007 22:15:21 -0800, Andrew Morton wrote
On Thu, 1 Nov 2007 23:30:13 +0200 Denys [EMAIL PROTECTED] wrote:
Finally i got full DMESG with 1GB card till end. Seems not readable too.
scsi0 : sc1200
scsi1 : sc1200
ata1: PATA max UDMA/33 cmd 0x1f0 ctl
WRAP ~ #./hdparm -I /dev/hda
/dev/hda:
ATAPI Write-once device, with non-removable media
Model Number: SanDisk SDP3B-16
Serial Number: 24313671615
Firmware Revision: vdd 1.00
Standards:
Likely used: 3
Configuration:
DRQ response: 50us.
On Tue, 6 Nov 2007 22:15:21 -0800, Andrew Morton wrote
On Thu, 1 Nov 2007 23:30:13 +0200 Denys [EMAIL PROTECTED] wrote:
Finally i got full DMESG with 1GB card till end. Seems not readable too.
Linux version 2.6.24-rc1-git10-embedded ([EMAIL PROTECTED]) (gcc
version 4.1.2 (Gentoo 4.1.2
On Thu, 08 Nov 2007 00:23:10 +0900, James Andrewartha wrote
Denys Fedoryshchenko wrote:
On Tue, 6 Nov 2007 22:15:21 -0800, Andrew Morton wrote
On Thu, 1 Nov 2007 23:30:13 +0200 Denys [EMAIL PROTECTED] wrote:
Finally i got full DMESG with 1GB card till end. Seems not readable too.
scsi0 :
On Wed, 07 Nov 2007 14:12:55 -0500, Mark Lord wrote
WRAP ~ #./hdparm -I /dev/hda
/dev/hda:
ATAPI Write-once device, with non-removable media
Model Number: SanDisk SDP3B-16
Serial Number: 24313671615
Firmware Revision: vdd 1.00
Standards:
This is another flash (1GB probably original Sandisk), 16MB one is called
SHARP, and it is 2-4 years old. Maybe it is some noname with incorrect
identification data. But it works well in USB card reader.
[ 12.434864] ide: Assuming 33MHz system bus speed for PIO modes; override
with idebus=xx
Denys Fedoryshchenko wrote:
On Wed, 07 Nov 2007 14:12:55 -0500, Mark Lord wrote
WRAP ~ #./hdparm -I /dev/hda
/dev/hda:
ATAPI Write-once device, with non-removable media
Model Number: SanDisk SDP3B-16
Serial Number: 24313671615
Firmware Revision: vdd 1.00
CFA spec requires 848a in word[0], not 844a, which explains
why hdparm thought it might be ATAPI. I've updated my hdparm
source to detect this for hdparm-7.8 (unreleased).
That card (16MB) is definitely buggy.
But your other (1GB) card looks fine.
Thanks. Somebody else may help with the
On Wed, Nov 07, 2007 at 02:12:55PM -0500, Mark Lord wrote:
That cannot be correct (??). Is this with hdparm-7.7 (latest sourceforge)
??
Can you show us the hdparm --Istdout output as well, please.
If this is applicable... FWIW hdparm was only recently (in past 72
hours) updated from 6.9 to
Jeff Garzik wrote:
On Wed, Nov 07, 2007 at 02:12:55PM -0500, Mark Lord wrote:
That cannot be correct (??). Is this with hdparm-7.7 (latest sourceforge)
??
Can you show us the hdparm --Istdout output as well, please.
If this is applicable... FWIW hdparm was only recently (in past 72
hours)
On Thursday 08 November 2007, Denys Fedoryshchenko wrote:
2.6.24-rc2 not working very well
dmesg
[ 12.386395] Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
[ 12.405579] ide: Assuming 33MHz system bus speed for PIO modes; override
with idebus=xx
[ 12.430441] SC1200: IDE
2.6.24-rc2 not working very well
dmesg
[ 12.386395] Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
[ 12.405579] ide: Assuming 33MHz system bus speed for PIO modes; override
with idebus=xx
[ 12.430441] SC1200: IDE controller (0x100b:0x0502 rev 0x01) at PCI slot
:00:12.2
[
I am using Gentoo (and it is custom build of linux, actually only busybox +
kernel + uclibc and few other tools), hdparm is vanilla 7.7
I will try to compile now -rc2 to see if there any changes.
With 16MB 2.6.24-rc1 works fine, 1GB working also with some errors in dmesg.
And IF that all is
You are right, seems no dma lines in adapter. hda=nodma helped, no errors
anymore. I will try now also libata_dma_mask and will mail result. Btw there
is no notes in Documentation/kernel-parameters.txt about it.
In any case it is complete board, WRAP.2C made by PCEngines in 2003. Kind of
Denys wrote:
Finally i got full DMESG with 1GB card till end. Seems not readable too.
..
ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
ata1.00: cmd c8/00:08:00:00:00/00:00:00:00:00/e0 tag 0 cdb 0x0 data 4096 in
res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4
On Thu, 1 Nov 2007 23:30:13 +0200 Denys [EMAIL PROTECTED] wrote:
Finally i got full DMESG with 1GB card till end. Seems not readable too.
Linux version 2.6.24-rc1-git10-embedded ([EMAIL PROTECTED]) (gcc
version 4.1.2 (Gentoo 4.1.2 p1.0.1)) #1 Thu Nov 1 23:12:53 EET 2007
BIOS-provided
21 matches
Mail list logo