> On Oct 18, 2021, at 7:26 AM, Dmitry Bogdanov <d.bogdanov@xxxxxxxxx> wrote:
>
> sgl is freed in the target stack in target_release_cmd_kref before
> calling qlt_free_cmd, but there is an unmap of sgl in qlt_free_cmd that
> causes such a panic if sgl is not yet DMA unmapped:
>
> NIP dma_direct_unmap_sg+0xdc/0x180
> LR dma_direct_unmap_sg+0xc8/0x180
> Call Trace:
> ql_dbg_prefix+0x68/0xc0 [qla2xxx] (unreliable)
> dma_unmap_sg_attrs+0x54/0xf0
> qlt_unmap_sg.part.19+0x54/0x1c0 [qla2xxx]
> qlt_free_cmd+0x124/0x1d0 [qla2xxx]
> tcm_qla2xxx_release_cmd+0x4c/0xa0 [tcm_qla2xxx]
> target_put_sess_cmd+0x198/0x370 [target_core_mod]
> transport_generic_free_cmd+0x6c/0x1b0 [target_core_mod]
> tcm_qla2xxx_complete_free+0x6c/0x90 [tcm_qla2xxx]
>
> SGL may be left unmapped in error cases of response sending.
> For instance, qlt_rdy_to_xfer maps sgl and exits when sesssion is being
> deleted keeping sgl mapped.
>
> This patch removes use-after-free of sgl, and ensures that sgl is
> unmapped for the cmd that was not send to FW.
>
> Signed-off-by: Dmitry Bogdanov <d.bogdanov@xxxxxxxxx>
> ---
> This patchset is intended for scsi-fix.
> ---
> drivers/scsi/qla2xxx/qla_target.c | 14 +++++---------
> 1 file changed, 5 insertions(+), 9 deletions(-)
>
> diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c
> index b3478ed9b12e..7d8242c120fc 100644
> --- a/drivers/scsi/qla2xxx/qla_target.c
> +++ b/drivers/scsi/qla2xxx/qla_target.c
> @@ -3319,8 +3319,7 @@ int qlt_xmit_response(struct qla_tgt_cmd *cmd, int xmit_type,
> "RESET-RSP online/active/old-count/new-count = %d/%d/%d/%d.\n",
> vha->flags.online, qla2x00_reset_active(vha),
> cmd->reset_count, qpair->chip_reset);
> - spin_unlock_irqrestore(qpair->qp_lock_ptr, flags);
> - return 0;
> + goto out_unmap_unlock;
> }
>
> /* Does F/W have an IOCBs for this request */
> @@ -3445,10 +3444,6 @@ int qlt_rdy_to_xfer(struct qla_tgt_cmd *cmd)
> prm.sg = NULL;
> prm.req_cnt = 1;
>
> - /* Calculate number of entries and segments required */
> - if (qlt_pci_map_calc_cnt(&prm) != 0)
> - return -EAGAIN;
> -
> if (!qpair->fw_started || (cmd->reset_count != qpair->chip_reset) ||
> (cmd->sess && cmd->sess->deleted)) {
> /*
> @@ -3466,6 +3461,10 @@ int qlt_rdy_to_xfer(struct qla_tgt_cmd *cmd)
> return 0;
> }
>
> + /* Calculate number of entries and segments required */
> + if (qlt_pci_map_calc_cnt(&prm) != 0)
> + return -EAGAIN;
> +
> spin_lock_irqsave(qpair->qp_lock_ptr, flags);
> /* Does F/W have an IOCBs for this request */
> res = qlt_check_reserve_free_req(qpair, prm.req_cnt);
> @@ -3870,9 +3869,6 @@ void qlt_free_cmd(struct qla_tgt_cmd *cmd)
>
> BUG_ON(cmd->cmd_in_wq);
>
> - if (cmd->sg_mapped)
> - qlt_unmap_sg(cmd->vha, cmd);
> -
> if (!cmd->q_full)
> qlt_decr_num_pend_cmds(cmd->vha);
>
> --
> 2.25.1
>
Looks Good.
Reviewed-by: Himanshu Madhani <himanshu.madhani@xxxxxxxxxx>
--
Himanshu Madhani Oracle Linux Engineering
