hw/ide/ahci: fix broken SError handling

When encountering an NCQ error, you should not write the NCQ tag to the
SError register. This is completely wrong.

The SError register has a clear definition, where each bit represents a
different error, see PxSERR definition in AHCI 1.3.1.

If we write a random value (like the NCQ tag) in SError, e.g. Linux will
read SError, and will trigger arbitrary error handling depending on the
NCQ tag that happened to be executing.

In case of success, ncq_cb() will call ncq_finish().
In case of error, ncq_cb() will call ncq_err() (which will clear
ncq_tfs->used), and then call ncq_finish(), thus using ncq_tfs->used is
sufficient to tell if finished should get set or not.

Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-id: 20230609140844.202795-9-nks@flawful.org
Signed-off-by: John Snow <jsnow@redhat.com>
(cherry picked from commit 9f89423537)
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
Niklas Cassel 2023-06-09 16:08:44 +02:00 committed by Michael Tokarev
parent 2aa37f5fa5
commit ccac65fbd1
1 changed files with 1 additions and 2 deletions

View File

@ -1011,7 +1011,6 @@ static void ncq_err(NCQTransferState *ncq_tfs)
ide_state->error = ABRT_ERR;
ide_state->status = READY_STAT | ERR_STAT;
ncq_tfs->drive->port_regs.scr_err |= (1 << ncq_tfs->tag);
qemu_sglist_destroy(&ncq_tfs->sglist);
ncq_tfs->used = 0;
}
@ -1021,7 +1020,7 @@ static void ncq_finish(NCQTransferState *ncq_tfs)
/* If we didn't error out, set our finished bit. Errored commands
* do not get a bit set for the SDB FIS ACT register, nor do they
* clear the outstanding bit in scr_act (PxSACT). */
if (!(ncq_tfs->drive->port_regs.scr_err & (1 << ncq_tfs->tag))) {
if (ncq_tfs->used) {
ncq_tfs->drive->finished |= (1 << ncq_tfs->tag);
}