Andre, results of testing with 2.4.0-test6 and ide.2.4.0-t6-5.0804 patch.
Followed by previous data, coz I forgot to copy to the lkml (slapped wrist).
I still can't boot (2.4.0-t6) unless a zip disk has been inserted,
hdd:hdd: lost interrupt
hdd: lost interrupt (repeats)
but the drive works ok after booting. However, I'm no longer running with UDMA,
I didn't find an option for the ALI 15X chipset in make menuconfig ? I'll look
in to this later.
Ken
My previous replies -
[snip]
[root@desktop ken]# cat /proc/ide/hdd/identify
00c0 0200 0000 000c 4000 0200 0020 0000
0000 0000 3638 3238 3131 3231 2020 2020
2020 2020 2020 2020 0000 0020 0004 422e
3239 2020 2020 494f 4d45 4741 2020 5a49
5020 3130 3020 2020 2020 2020 2020 2020
2020 2020 2020 2020 2020 2020 2020 0000
0000 2a00 0000 0000 0000 0000 0200 000c
0020 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0001
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
[root@desktop ken]# hdparm -acdgkmnru /dev/hdd
/dev/hdd:
multcount = 0 (off)
I/O support = 0 (default 16-bit)
unmaskirq = 0 (off)
using_dma = 0 (off)
keepsettings = 0 (off)
nowerr = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 512/12/32, sectors = 196608, start = 0
-- [snip] Andre,results of trying 2.2.16 with your patch. (Twice, first time might have been an early patch, second time was ide.2.2.16.all.20000805.patch. No significant differences).
Summary : nasty.
1. Still loops in the partition check if no disk inserted- hdd:hdd: lost interrupt hdd: lost interrupt (etc.)
2. Messages at boot if a disk is present Aug 21 22:21:37 desktop kernel: Partition check: Aug 21 22:21:37 desktop kernel: hda: hda3 hda4 < hda5 hda6 hda7 hda8 hda9 hda10 hda11 hda12 hda13 > Aug 21 22:21:37 desktop kernel: hdd:hdd: irq timeout: status=0xd0 { Busy } Aug 21 22:21:37 desktop kernel: hdc: DMA disabled Aug 21 22:21:37 desktop kernel: ide1: reset: success Aug 21 22:21:37 desktop kernel: hdd1 Aug 21 22:21:37 desktop kernel: VFS: Mounted root (ext2 filesystem) readonly.
3. After booting with a zip disk in place, I couldn't remove it (I hadn't tried to mount it yet), the LED stayed lit.
4. tried mounting it mount /zip hdd:hdd: irq timeout:status=0xd0{Busy} (a) ide1: reset:success (b) hdd1 (c) repeated as a,b,c; then looping as a,b; a,b; even after I'd killed that shell and whilst closing everything down for a reboot. [snip]
On Mon, 21 Aug 2000, you wrote: > On Mon, 21 Aug 2000, ken moffat wrote: > > > Andre, > > no, the patch makes things worse on the zip - > > I normally use 2.2.13 with the patch, because it gives UDMA access to the hard > > drive. But it gives big problems trying to access the Zip (I think, /dev/hdd is not a > > valid block device). Sometimes I can access the disk if I first use fdisk to > > print the partition table, sometimes fdisk doesn't manage this. So, I keep an > > unpatched version of 2.2.13 as well. > > Can you send me the hdparm output on the device and the contents of > /proc/ide/hdd/identify? > > > In 2.2.16 the patch doesn't let me boot without a zip disk being inserted, and > > Strange because I do it all the time, and I have a known broken device. > > > I get similar problems trying to access /dev/hdd (if anything, they're worse, > > I've never managed to access a zip disk with patched 2.2.16. > > Thanks for the info on why the code had to be changed. > > Okay you have warrented deeper thought as to why. > Now does it fault out in 2.4.X test kernels? > > Andre Hedrick > The Linux ATA/IDE guy -- Amongst our weaponry are such diverse elements as fear, surprise, ruthless efficiency, and an almost fanatical devotion to the Penguin, and nice red uniforms. "Monty Penguin" (well, what did you expect? I don't do Python ;-() - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Aug 23 2000 - 21:00:07 EST