libata: fix WARN_ON() condition in *_fill_sg()
PATCH 1/1 libata: fix WARN_ON() condition in *_fill_sg()
For ATAPI commands, padding can reduce qc->n_elem by one and thus to zero making WARN_ON(qc->n_elem == 0) in ata_fill_sg() and qs_fill_sg() trigger for legal commands. This patch fixes the WARN_ON() conditions to take qc->pad_len into account.
Although the condition check seems a bit excessive, as this part of code isn't still stable yet, I think it's worth to keep those.
---
Jeff, this patch is against the current #upstream[1]. Without this patch trying to burn a cd with cdrecord on sii3114 results in a log of Badness messages. One of offending commands is MODE_SENSE(10) with data size of 2 to find out mode page size. Padding leaves no sg entry and the Badness gets trigerred.
[1] f1b318793dcd2d9ff6b5ac06e7762098fa079cee
libata-core.c | 2 +- sata_qstor.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/scsi/libata-core.c +1 -1
@@ -2639,7 +2639,7 @@ static void ata_fill_sg(struct ata_queueunsigned int idx;WARN_ON (qc->__sg ==NULL ); -WARN_ON (qc->n_elem ==0 ); +WARN_ON (qc->n_elem ==0 && qc->pad_len ==0 ); idx =0 ;ata_for_each_sg (sg, qc) {
drivers/scsi/sata_qstor.c +1 -1
@@ -277,7 +277,7 @@ static unsigned int qs_fill_sg(struct atu8 *prd = pp->pkt +QS_CPB_BYTES ;WARN_ON (qc->__sg ==NULL ); -WARN_ON (qc->n_elem ==0 ); +WARN_ON (qc->n_elem ==0 && qc->pad_len ==0 ); nelem =0 ;ata_for_each_sg (sg, qc) {