ata: libata-eh: Do not use ATAPI DMA for a device limited to PIO mode

atapi_eh_request_sense() currently uses ATAPI DMA if the SATA controller
has ATA_FLAG_PIO_DMA (PIO cmds via DMA) set.

However, ATA_FLAG_PIO_DMA is a flag that can be set by a low-level driver
on a port at initialization time, before any devices are scanned.

If a controller detects a connected device that only supports PIO, we set
the flag ATA_DFLAG_PIO.

Modify atapi_eh_request_sense() to not use ATAPI DMA if the connected
device only supports PIO.

Reported-by: Philip Pemberton <lists@philpem.me.uk>
Closes: https://lore.kernel.org/linux-ide/c6722ee8-5e21-4169-af59-cbbae9edc02f@philpem.me.uk/
Tested-by: Philip Pemberton <lists@philpem.me.uk>
Reviewed-by: Damien Le Moal <dlemoal@kernel.org>
Link: https://lore.kernel.org/r/20250221015422.20687-2-cassel@kernel.org
Signed-off-by: Niklas Cassel <cassel@kernel.org>

diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 3b303d4ae37a01c55ca2ce5973aa9de256d7d5eb..16cd676eae1f9a27cd1a4714cc3e3be14443e9fe 100644 (file)
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -1542,8 +1542,15 @@ unsigned int atapi_eh_request_sense(struct ata_device *dev,
        tf.flags |= ATA_TFLAG_ISADDR | ATA_TFLAG_DEVICE;
        tf.command = ATA_CMD_PACKET;
 
-       /* is it pointless to prefer PIO for "safety reasons"? */
-       if (ap->flags & ATA_FLAG_PIO_DMA) {
+       /*
+        * Do not use DMA if the connected device only supports PIO, even if the
+        * port prefers PIO commands via DMA.
+        *
+        * Ideally, we should call atapi_check_dma() to check if it is safe for
+        * the LLD to use DMA for REQUEST_SENSE, but we don't have a qc.
+        * Since we can't check the command, perhaps we should only use pio?
+        */
+       if ((ap->flags & ATA_FLAG_PIO_DMA) && !(dev->flags & ATA_DFLAG_PIO)) {
                tf.protocol = ATAPI_PROT_DMA;
                tf.feature |= ATAPI_PKT_DMA;
        } else {