Messages in this thread | | | Date | Thu, 8 Apr 1999 12:28:09 -0700 | From | Garner <> | Subject | Re: DMA disabled and busy errors |
| |
When I do get these messages bad things do seem to happen. ie. compilations die, sometime the machine even locks up completely. These messages are almost always triggered during some combination of heavy load and disk i/o. And once done, DMA remains disabled on all drives on the affected IDE channel. I would get this with the vanilla 2.2 kernels quite often with very little encouragement. The Unified UDMA patch helps quite a bit but does not solve the problem. I end up manually disabling DMA at boot to preserve my sanity.
I used to think that it was my chipset(VIA MVP3 on an FIC PA-2013). But seeing someone with a BX chipset have these problems has caused my renewed interest.
garner
Chris Wedgwood wrote:
> On Wed, Apr 07, 1999 at 08:32:30AM -0400, Mark Lord wrote: > > > Chris Wedgwood wrote: > > > > > ... > > > It works 99% of the time, just every now and then, perhaps once a day under > > > light load, more often under heavier load, I get these: > > > > > > hda: timeout waiting for DMA > > > hda: irq timeout: status=0xd0 { Busy } > > > ide0: reset: success > > > > I wouldn't worry about it then. Likely just a firmware (drive) bug. > > Harmless. > > > > I've seen/had lots of drives with that behaviour. > > MS-Win probably just retries the operation silently; > > with linux, we let you know about spurious problems. > > Windows probably doesn't see these errors -- they only show under > heavy load. Linux 2.2.x produces these error much more often that > 2.0.x, presumably because it drives the hardware harder... > > > > > -cw > > - > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > the body of a message to majordomo@vger.rutgers.edu > Please read the FAQ at http://www.tux.org/lkml/
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |