On 6/14/23 00:17, Mike Christie wrote:
If scsi_execute_cmd returns < 0 it will not have set the sshdr, so we can't access it. Signed-off-by: Mike Christie <michael.christie@xxxxxxxxxx> --- drivers/scsi/scsi_transport_spi.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/scsi/scsi_transport_spi.c b/drivers/scsi/scsi_transport_spi.c index 2442d4d2e3f3..2100c3adb456 100644 --- a/drivers/scsi/scsi_transport_spi.c +++ b/drivers/scsi/scsi_transport_spi.c @@ -126,7 +126,7 @@ static int spi_execute(struct scsi_device *sdev, const void *cmd, */ result = scsi_execute_cmd(sdev, cmd, opf, buffer, bufflen, DV_TIMEOUT, 1, &exec_args); - if (result < 0 || !scsi_sense_valid(sshdr) || + if (result <= 0 || !scsi_sense_valid(sshdr) || sshdr->sense_key != UNIT_ATTENTION) break; }
Hmm ... why is this change necessary? If result == 0 and args->sshdr != 0 then scsi_execute() has called scsi_normalize_sense(). The first function call in scsi_normalize_sense() is memset(sshdr, 0, sizeof(struct scsi_sense_hdr)). Does this mean that it is safe to access the contents of sshdr if scsi_execute_cmd() returns 0?
Thanks, Bart.
