On Mon, 18 Feb 2008 15:30:58 -0800 Tim Pepper <lnxninja@xxxxxxxxxxxxxxxxxx> wrote: > On Mon 18 Feb at 22:32:46 +0900 tomof@xxxxxxx said: > > > > diff --git a/drivers/scsi/ips.c b/drivers/scsi/ips.c > > index 05bb6ea..39cdd68 100644 > > --- a/drivers/scsi/ips.c > > +++ b/drivers/scsi/ips.c > > @@ -6906,7 +6906,7 @@ ips_register_scsi(int index) > > sh->max_channel = ha->nbus - 1; > > sh->can_queue = ha->max_cmds - 1; > > > > - scsi_add_host(sh, NULL); > > + scsi_add_host(sh, &ha->pcidev->dev); > > scsi_scan_host(sh); > > > > return 0; > > Fujita-san, > > This applies and runs well on top of your 0005 patch! The rest of the > patches also then apply in order and run successfully. Great, thanks a lot! > Just to confirm, I applied the above alone to a clean 2.6.24 and things > again build and run successfully. For completeness I also reproduced > the problem against 2.6.23.16 and verified the above patch fixes on that > kernel version as well. Nice. There is another bug on 2.6.24 but we rarely hit this so 2.6.24 works most of the time: http://marc.info/?l=linux-scsi&m=120303487528875&w=2 > Assuming this patch is accepted for 2.6.25, please also queue it for > the 2.6.23/24 stable trees. Yes, I will take care about it. Can you please help me just once more? 2.6.25-rc2 fixed this bug in a bit different way by chance. Please test 2.6.25-rc2 with the attached patch to make sure that ips in 2.6.25 works well. > Thank you very much for your help in tracking this issue down! No problem. I should have fixed it long time ago. Really sorry about it. diff --git a/drivers/scsi/ips.c b/drivers/scsi/ips.c index bb152fb..429592a 100644 --- a/drivers/scsi/ips.c +++ b/drivers/scsi/ips.c @@ -1576,7 +1576,7 @@ ips_make_passthru(ips_ha_t *ha, struct scsi_cmnd *SC, ips_scb_t *scb, int intr) METHOD_TRACE("ips_make_passthru", 1); scsi_for_each_sg(SC, sg, scsi_sg_count(SC), i) - length += sg[i].length; + length += sg->length; if (length < sizeof (ips_passthru_t)) { /* wrong size */ @@ -3510,15 +3510,16 @@ ips_scmd_buf_write(struct scsi_cmnd *scmd, void *data, unsigned int count) struct scatterlist *sg = scsi_sglist(scmd); for (i = 0, xfer_cnt = 0; - (i < scsi_sg_count(scmd)) && (xfer_cnt < count); i++) { - min_cnt = min(count - xfer_cnt, sg[i].length); + (i < scsi_sg_count(scmd)) && (xfer_cnt < count); + i++, sg = sg_next(sg)) { + min_cnt = min(count - xfer_cnt, sg->length); /* kmap_atomic() ensures addressability of the data buffer.*/ /* local_irq_save() protects the KM_IRQ0 address slot. */ local_irq_save(flags); - buffer = kmap_atomic(sg_page(&sg[i]), KM_IRQ0) + sg[i].offset; + buffer = kmap_atomic(sg_page(sg), KM_IRQ0) + sg->offset; memcpy(buffer, &cdata[xfer_cnt], min_cnt); - kunmap_atomic(buffer - sg[i].offset, KM_IRQ0); + kunmap_atomic(buffer - sg->offset, KM_IRQ0); local_irq_restore(flags); xfer_cnt += min_cnt; @@ -3543,15 +3544,16 @@ ips_scmd_buf_read(struct scsi_cmnd *scmd, void *data, unsigned int count) struct scatterlist *sg = scsi_sglist(scmd); for (i = 0, xfer_cnt = 0; - (i < scsi_sg_count(scmd)) && (xfer_cnt < count); i++) { - min_cnt = min(count - xfer_cnt, sg[i].length); + (i < scsi_sg_count(scmd)) && (xfer_cnt < count); + i++, sg = sg_next(sg)) { + min_cnt = min(count - xfer_cnt, sg->length); /* kmap_atomic() ensures addressability of the data buffer.*/ /* local_irq_save() protects the KM_IRQ0 address slot. */ local_irq_save(flags); - buffer = kmap_atomic(sg_page(&sg[i]), KM_IRQ0) + sg[i].offset; + buffer = kmap_atomic(sg_page(sg), KM_IRQ0) + sg->offset; memcpy(&cdata[xfer_cnt], buffer, min_cnt); - kunmap_atomic(buffer - sg[i].offset, KM_IRQ0); + kunmap_atomic(buffer - sg->offset, KM_IRQ0); local_irq_restore(flags); xfer_cnt += min_cnt; - To unsubscribe from this list: 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
