Messages in this thread | | | Subject | Re: [patch] voluntary-preempt-2.6.9-rc1-bk4-R1 | From | Thomas Charbonnel <> | Date | Thu, 09 Sep 2004 18:44:46 +0200 |
| |
Alan Cox wrote : > On Mer, 2004-09-08 at 21:33, Mark_H_Johnson@raytheon.com wrote: > > >.... Please disable IDE DMA and see > > >what happens (after hiding the PIO IDE codepath via > > >touch_preempt_timing()). > > > > Not quite sure where to add touch_preempt_timing() calls - somewhere in the > > loop in ide_outsl and ide_insl? [so we keep resetting the start /end > > times?] > > If you haven't done hdparm -u1 that may be a reason you want to touch > these. To defend against some very bad old h/w where a stall in the I/O > stream to the disk causes corruption we disable IRQ's across the > transfer in PIO mode by default. >
I had the exact same problem showing in the output of latencytest, and enabling unmaskirq on the drive being stressed solved it, thanks !
See this for the problem : http://www.undata.org/~thomas/unmaskirq_0/index.html and this for the (impressive) results : http://www.undata.org/~thomas/unmaskirq_1/index.html
Thomas
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |