Hi all A problem has first been reported in Date: Thu, 6 Jan 2005 05:18:25 -0500 From: Andrew Schulman <andrex@xxxxxxxxxxxxxxxxx> Subject: dc395x: can't write to tape whereby a bug has become apparent in driver's virtual address calculation. I wrote a patch, which propagated into the mainline, before Jens A. (added to CC:) noticed, that it was buggy. Now Pierre actually hit that bug. Yesterday in http://marc.theaimsgroup.com/?l=linux-scsi&m=112336777508938&w=2 I sent him and to the SCSI list a patch, which made his system again usable. Which doesn't mean, however, that the patch is correct. The patch uses the same 2 functions that I used earlier in a proposed in http://marc.theaimsgroup.com/?l=linux-scsi&m=111435313619444&w=2 patch to tmscsim: static void *dc395x_kmap_atomic_sg(struct scatterlist *sg, int sg_count, size_t *offset, size_t *len) and static void dc395x_kunmap_atomic_sg(void *virt) to atomically map and unmap a byte sequence from an sg. This has also been discussed around April this year in a thread with subject: "Re: [PATCH] dc395x: Fix support for highmem." That discussion didn't come to a definite conclusion. As well as my patch to tmscsim didn't produce any reaction, and, I think, wasn't included in any "official" tree. Now, it would be best, I think, if somebody could review the patches to dc395x and tmscsim. In case the 2 functions are correct, they could be included in a central file (scsi_lib?) for all drivers, needing PIO (I think IDE and libata won't be able to use them in this form). The rest of the 2 patches could then be applied to respective drivers. Thanks Guennadi On Sun, 7 Aug 2005, Pierre Ossman wrote: > Guennadi Liakhovetski wrote: > > >So, now that we have somebody to test alternative patches, let's try it... > >Please, first revert the first patch (apply with -R), and then apply the > >second one. There, probably, will be offsets, don't worry about them. > >Warning: untested. Please, try some read-only test first, if you get so > >far at all... I prepared this patch a long time ago, and didn't revisit it > >since then, hope, it's still fresh enough:-) > > > > > > Ok, I've tried the included patches now and everything seems to work > fine. Note though that I only have a CD writer hooked up to the > controller, no disks. > > Anything special I should try to make sure the patch does what it's > supposed to? > > Rgds > Pierre > --- Guennadi Liakhovetski - : send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
