Re: [PATCH v2] ata: libata-scsi: reject data-in commands with a data-out buffer

[Niklas Cassel <cassel@kernel.org>]

Hello Zhao,

When sending a V2 (or V3), please do not use In-Reply-To:
https://docs.kernel.org/process/submitting-patches.html#explicit-in-reply-to-headers

It creates an unmanageable forest of references in email clients.


On Wed, Oct 28, 2026 at 11:53:37AM +0000, sashiko-bot@kernel.org wrote:
> Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
> - [High] ata: libata-scsi: ata_scsi_translate() direction check misses atapi_xlat()

This looks like a valid comment, we probably need something like:

diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -3193,6 +3193,13 @@ static unsigned int atapi_xlat(struct ata_queued_cmd *qc)
 
 	qc->complete_fn = atapi_qc_complete;
 
+	if (scmd->sc_data_direction == DMA_TO_DEVICE &&
+	    scmd->cmnd[0] == INQUIRY) {
+		ata_scsi_set_sense(dev, scmd, ILLEGAL_REQUEST,
+				   INVALID_FIELD_IN_CDB);
+		return 1;
+	}
+
 	qc->tf.flags |= ATA_TFLAG_ISADDR | ATA_TFLAG_DEVICE;
 	if (scmd->sc_data_direction == DMA_TO_DEVICE) {
 		qc->tf.flags |= ATA_TFLAG_WRITE;



> - [Medium] ata: libata-scsi: missing ascq argument in ata_scsi_set_sense() calls

This Sashiko comment is a false positive.

Because Sashiko first tries to apply a patch to libata/for-next before
mainline, Sashiko did not realize that ata_scsi_set_sense() has changed in
mainline by the recent SCSI pull request.

Since Linus has pulled the both the SCSI and the libata pull request, I
updated libata for-next to point to mainline, so Sashiko should not give
this false positive comment should not happen again.


Kind regards,
Niklas