Re: Lockup in IDE code

Re: Lockup in IDE code

[Zwane Mwaikambo]

Uniform Multi-Platform E-IDE driver Revision: 6.31
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
PIIX4: IDE controller on PCI bus 00 dev 21
PIIX4: chipset revision 1
PIIX4: not 100% native mode: will probe irqs later
    ide0: BM-DMA at 0xd800-0xd807, BIOS settings: hda:pio, hdb:DMA
    ide1: BM-DMA at 0xd808-0xd80f, BIOS settings: hdc:pio, hdd:pio
hdb: CRD-8480B, ATAPI CD/DVD-ROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14

He only has the one device on IDE as you can see that the probe only shows 
hdb, and i have a PIIX4 too and they do UDMA33 well enough, in fact i'm not 
aware of any DMA problems with the PIIX4. Richard, have you tried disabling 
CONFIG_IDEDMA_AUTO and CONFIG_PIIX_TUNING ?

But looking at the code in ide-features.c it doesn't look like we have 
enabled DMA by the time we get to that code.

     if (error) {
                (void) ide_dump_status(drive, "set_drive_speed_status", stat);
                return error;
        }

<----- snip ----- >
#if defined(CONFIG_BLK_DEV_IDEDMA) && !defined(CONFIG_DMA_NONPCI)
        if (speed > XFER_PIO_4) {
                outb(inb(hwif->dma_base+2)|(1<<(5+unit)), hwif->dma_base+2);  
      } else {
                outb(inb(hwif->dma_base+2) & ~(1<<(5+unit)), 
hwif->dma_base+2);
        }
#endif /* (CONFIG_BLK_DEV_IDEDMA) && !(CONFIG_DMA_NONPCI) */

But bear in mind i'm far from being remotely knowledgeable about this 
specific code ;)

Re: Lockup in IDE code

[Ph. Marek]

I've had the same problem.

Lockups with 2.4.14pre1 + ext3-patch, sysrq not possible, HD led is on.

This is a 7IXE4 (AMD756), Duron 650, 384MB. 2 HD's, 1 CDROM.

	
I've noted that on startup I got "spurious 8259A interrupt: IRQ7", and the
machine locked hard on IDE activity (sometimes).
after observing the above message and others like 
	hda: timeout waiting for DMA
	ide_dmaproc: chipset supported ide_dma timeoutfunc only: 14
	hda: read_intr: status=0x51
	hda: read_intr: status=0x04
	hda: read_intr: status=0x51
	hda: read_intr: status=0x04
	hda: read_intr: status=0x51
	hda: read_intr: status=0x04
	ide0: reset: success

The I've tested disabling some options.
So far my best result was disabling APIC on UP (both - IOAPIC and APIC) -
so far I've had no more lockups.

Will test further, but this seemed to be the problem for me.


Regards,

Phil


-
This message is RSA-encrypted: n=33389, e=257

Re: Lockup in IDE code

[Zwane Mwaikambo]

Mark Zealey - 2001-11-09 12:40:41 PST
>Well, obviously your drive or mobo doesn't support DMA. I noticed that all 
>the other devices (hd[acd]) were in PIO mode, why not try setting dma mode 
>on one of them. Most CD's I've come across don't do DMA very well, I believe 
>that the DMA is mostly only useful for DVD's, where you need a high transfer 
>rate to watch a movie.

He probably doesn't have anything on the primary master and seconday 
master/slaves. And UDMA has worked very well for most CDROM drives i've come 
across, but then as always YMMV.

Cheers,
	Zwane

Re: Lockup in IDE code

[Xavier Bestel]

le ven 09-11-2001 à 20:38, Dan Hollis a écrit :
> On Fri, 9 Nov 2001, Richard Gooch wrote:
> > GXavier Bestel writes:
> > > le ven 09-11-2001 à 03:17, Richard Gooch a écrit :
> > > >   Hi, all. I tried to use my IDE CD-ROM today, the first time in a
> > > > long while. When attempting to mount it, the machine locked up,
> > > > hard. Even SysReq didn't work.
> > > Do you have a read error on your CD ?
> > No. I did mention that when I turned off DMA, it worked fine.
> 
> I don't see why read errors should lock up the kernel anyway... would it
> be acceptable for floppy read errors to lock up the kernel too?

Indeed, but they do, on my machine anyway.

	Xav

Re: Lockup in IDE code

[Mark Zealey]

On Fri, Nov 09, 2001 at 09:56:54AM -0700, Richard Gooch wrote:

> GXavier Bestel writes:
> > le ven 09-11-2001 à 03:17, Richard Gooch a écrit :
> > >   Hi, all. I tried to use my IDE CD-ROM today, the first time in a
> > > long while. When attempting to mount it, the machine locked up,
> > > hard. Even SysReq didn't work.
> > 
> > Do you have a read error on your CD ?
> 
> No. I did mention that when I turned off DMA, it worked fine.
> 
> BTW: I've gotten a few "helpful" responses saying "try using hdparm to
> turn off DMA". Well, those people obviously hadn't bothered reading
> all my message, because I stated that when I disabled DMA with hdparm,
> it worked fine.
> </grumble>

Well, obviously your drive or mobo doesn't support DMA. I noticed that all the
other devices (hd[acd]) were in PIO mode, why not try setting dma mode on one of
them. Most CD's I've come across don't do DMA very well, I believe that the DMA
is mostly only useful for DVD's, where you need a high transfer rate to watch a
movie.

-- 

Mark Zealey
mark@zealos.org
mark@itsolve.co.uk

UL++++>$ G!>(GCM/GCS/GS/GM) dpu? s:-@ a16! C++++>$ P++++>+++++$ L+++>+++++$
!E---? W+++>$ N- !o? !w--- O? !M? !V? !PS !PE--@ PGP+? r++ !t---?@ !X---?
!R- b+ !tv b+ DI+ D+? G+++ e>+++++ !h++* r!-- y--

(www.geekcode.com)

Re: Lockup in IDE code

[Dan Hollis]

On Fri, 9 Nov 2001, Richard Gooch wrote:
> GXavier Bestel writes:
> > le ven 09-11-2001 à 03:17, Richard Gooch a écrit :
> > >   Hi, all. I tried to use my IDE CD-ROM today, the first time in a
> > > long while. When attempting to mount it, the machine locked up,
> > > hard. Even SysReq didn't work.
> > Do you have a read error on your CD ?
> No. I did mention that when I turned off DMA, it worked fine.

I don't see why read errors should lock up the kernel anyway... would it
be acceptable for floppy read errors to lock up the kernel too?

-Dan
-- 
[-] Omae no subete no kichi wa ore no mono da. [-]

Re: Lockup in IDE code

[Richard Gooch]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 692 bytes --]

GXavier Bestel writes:
> le ven 09-11-2001 à 03:17, Richard Gooch a écrit :
> >   Hi, all. I tried to use my IDE CD-ROM today, the first time in a
> > long while. When attempting to mount it, the machine locked up,
> > hard. Even SysReq didn't work.
> 
> Do you have a read error on your CD ?

No. I did mention that when I turned off DMA, it worked fine.

BTW: I've gotten a few "helpful" responses saying "try using hdparm to
turn off DMA". Well, those people obviously hadn't bothered reading
all my message, because I stated that when I disabled DMA with hdparm,
it worked fine.
</grumble>

				Regards,

					Richard....
Permanent: rgooch@atnf.csiro.au
Current:   rgooch@ras.ucalgary.ca

Re: Lockup in IDE code

[Xavier Bestel]

le ven 09-11-2001 à 03:17, Richard Gooch a écrit :
>   Hi, all. I tried to use my IDE CD-ROM today, the first time in a
> long while. When attempting to mount it, the machine locked up,
> hard. Even SysReq didn't work.

Do you have a read error on your CD ?

Lockup in IDE code

[Richard Gooch]

  Hi, all. I tried to use my IDE CD-ROM today, the first time in a
long while. When attempting to mount it, the machine locked up,
hard. Even SysReq didn't work. I got the following message:

hdb: timeout waiting for DMA
ide_dmaproc: chipset supported ide_dma_timeout func only: 14

I was running 2.4.7 initially, and then pulled down 2.4.14 and got the
same problem. All my IDE drivers are loadable modules. The kernel
messages I get when I load those modules are appended. If I turn off
DMA using hdparm, I don't get a lockup.

This wasn't always a problem. The last time I used my CD-ROM it worked
fine. My motherboard is an Asus P2B-D (dual PIII).

				Regards,

					Richard....
Permanent: rgooch@atnf.csiro.au
Current:   rgooch@ras.ucalgary.ca
===============================================================================
Uniform Multi-Platform E-IDE driver Revision: 6.31
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
PIIX4: IDE controller on PCI bus 00 dev 21
PIIX4: chipset revision 1
PIIX4: not 100% native mode: will probe irqs later
    ide0: BM-DMA at 0xd800-0xd807, BIOS settings: hda:pio, hdb:DMA
    ide1: BM-DMA at 0xd808-0xd80f, BIOS settings: hdc:pio, hdd:pio
hdb: CRD-8480B, ATAPI CD/DVD-ROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
hdb: set_drive_speed_status: status=0x51 { DriveReady SeekComplete Error }
hdb: set_drive_speed_status: error=0xb4
hdb: ATAPI 48X CD-ROM drive, 128kB Cache, DMA
Uniform CD-ROM driver Revision: 3.12