Fixes: 576f1b4bc802 ("soc: mediatek: Add Mediatek CMDQ helper")
There may be possible timeout issue when lots of cmdq packets are
flushed to the same cmdq client. The necessary modifications are as
below.
1.Adjust the timer timeout period as client->timeout_ms * client->pkt_cnt.
2.Optimize the time to start the timer.
Signed-off-by: Houlong Wei <[email protected]>
---
drivers/soc/mediatek/mtk-cmdq-helper.c | 22 ++++++++++++++++++----
1 file changed, 18 insertions(+), 4 deletions(-)
diff --git a/drivers/soc/mediatek/mtk-cmdq-helper.c b/drivers/soc/mediatek/mtk-cmdq-helper.c
index dc644cfb6419..31142c193527 100644
--- a/drivers/soc/mediatek/mtk-cmdq-helper.c
+++ b/drivers/soc/mediatek/mtk-cmdq-helper.c
@@ -350,7 +350,8 @@ static void cmdq_pkt_flush_async_cb(struct cmdq_cb_data data)
del_timer(&client->timer);
else
mod_timer(&client->timer, jiffies +
- msecs_to_jiffies(client->timeout_ms));
+ msecs_to_jiffies(client->timeout_ms *
+ client->pkt_cnt));
spin_unlock_irqrestore(&client->lock, flags);
}
@@ -379,9 +380,7 @@ int cmdq_pkt_flush_async(struct cmdq_pkt *pkt, cmdq_async_flush_cb cb,
if (client->timeout_ms != CMDQ_NO_TIMEOUT) {
spin_lock_irqsave(&client->lock, flags);
- if (client->pkt_cnt++ == 0)
- mod_timer(&client->timer, jiffies +
- msecs_to_jiffies(client->timeout_ms));
+ client->pkt_cnt++;
spin_unlock_irqrestore(&client->lock, flags);
}
@@ -391,6 +390,21 @@ int cmdq_pkt_flush_async(struct cmdq_pkt *pkt, cmdq_async_flush_cb cb,
/* We can send next packet immediately, so just call txdone. */
mbox_client_txdone(client->chan, 0);
+ if (client->timeout_ms != CMDQ_NO_TIMEOUT) {
+ spin_lock_irqsave(&client->lock, flags);
+ /*
+ * GCE HW maybe execute too quickly and the callback function
+ * may be invoked earlier. If this happens, pkt_cnt is reduced
+ * by 1 in cmdq_pkt_flush_async_cb(). The timer is set only if
+ * pkt_cnt is greater than 0.
+ */
+ if (client->pkt_cnt > 0)
+ mod_timer(&client->timer, jiffies +
+ msecs_to_jiffies(client->timeout_ms *
+ client->pkt_cnt));
+ spin_unlock_irqrestore(&client->lock, flags);
+ }
+
return 0;
}
EXPORT_SYMBOL(cmdq_pkt_flush_async);
--
2.18.0
Thanks for the patch.
On Thu, Oct 22, 2020 at 5:44 PM Houlong Wei <[email protected]> wrote:
>
> Fixes: 576f1b4bc802 ("soc: mediatek: Add Mediatek CMDQ helper")
nit: This belongs right next to the Sob line, but I guess/hope
Matthias can help you fix that ,-)
>
> There may be possible timeout issue when lots of cmdq packets are
> flushed to the same cmdq client. The necessary modifications are as
> below.
> 1.Adjust the timer timeout period as client->timeout_ms * client->pkt_cnt.
> 2.Optimize the time to start the timer.
>
Reviewed-by: Nicolas Boichat <[email protected]>
> Signed-off-by: Houlong Wei <[email protected]>
> ---
> drivers/soc/mediatek/mtk-cmdq-helper.c | 22 ++++++++++++++++++----
> 1 file changed, 18 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/soc/mediatek/mtk-cmdq-helper.c b/drivers/soc/mediatek/mtk-cmdq-helper.c
> index dc644cfb6419..31142c193527 100644
> --- a/drivers/soc/mediatek/mtk-cmdq-helper.c
> +++ b/drivers/soc/mediatek/mtk-cmdq-helper.c
> @@ -350,7 +350,8 @@ static void cmdq_pkt_flush_async_cb(struct cmdq_cb_data data)
> del_timer(&client->timer);
> else
> mod_timer(&client->timer, jiffies +
> - msecs_to_jiffies(client->timeout_ms));
> + msecs_to_jiffies(client->timeout_ms *
> + client->pkt_cnt));
> spin_unlock_irqrestore(&client->lock, flags);
> }
>
> @@ -379,9 +380,7 @@ int cmdq_pkt_flush_async(struct cmdq_pkt *pkt, cmdq_async_flush_cb cb,
>
> if (client->timeout_ms != CMDQ_NO_TIMEOUT) {
> spin_lock_irqsave(&client->lock, flags);
> - if (client->pkt_cnt++ == 0)
> - mod_timer(&client->timer, jiffies +
> - msecs_to_jiffies(client->timeout_ms));
> + client->pkt_cnt++;
> spin_unlock_irqrestore(&client->lock, flags);
> }
>
> @@ -391,6 +390,21 @@ int cmdq_pkt_flush_async(struct cmdq_pkt *pkt, cmdq_async_flush_cb cb,
> /* We can send next packet immediately, so just call txdone. */
> mbox_client_txdone(client->chan, 0);
>
> + if (client->timeout_ms != CMDQ_NO_TIMEOUT) {
> + spin_lock_irqsave(&client->lock, flags);
> + /*
> + * GCE HW maybe execute too quickly and the callback function
> + * may be invoked earlier. If this happens, pkt_cnt is reduced
> + * by 1 in cmdq_pkt_flush_async_cb(). The timer is set only if
> + * pkt_cnt is greater than 0.
> + */
> + if (client->pkt_cnt > 0)
> + mod_timer(&client->timer, jiffies +
> + msecs_to_jiffies(client->timeout_ms *
> + client->pkt_cnt));
> + spin_unlock_irqrestore(&client->lock, flags);
> + }
> +
> return 0;
> }
> EXPORT_SYMBOL(cmdq_pkt_flush_async);
> --
> 2.18.0
Hi, Houlong:
Houlong Wei <[email protected]> 於 2020年10月22日 週四 下午5:55寫道：
>
> Fixes: 576f1b4bc802 ("soc: mediatek: Add Mediatek CMDQ helper")
>
> There may be possible timeout issue when lots of cmdq packets are
> flushed to the same cmdq client. The necessary modifications are as
> below.
> 1.Adjust the timer timeout period as client->timeout_ms * client->pkt_cnt.
> 2.Optimize the time to start the timer.
>
> Signed-off-by: Houlong Wei <[email protected]>
> ---
> drivers/soc/mediatek/mtk-cmdq-helper.c | 22 ++++++++++++++++++----
> 1 file changed, 18 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/soc/mediatek/mtk-cmdq-helper.c b/drivers/soc/mediatek/mtk-cmdq-helper.c
> index dc644cfb6419..31142c193527 100644
> --- a/drivers/soc/mediatek/mtk-cmdq-helper.c
> +++ b/drivers/soc/mediatek/mtk-cmdq-helper.c
> @@ -350,7 +350,8 @@ static void cmdq_pkt_flush_async_cb(struct cmdq_cb_data data)
> del_timer(&client->timer);
> else
> mod_timer(&client->timer, jiffies +
> - msecs_to_jiffies(client->timeout_ms));
> + msecs_to_jiffies(client->timeout_ms *
> + client->pkt_cnt));
> spin_unlock_irqrestore(&client->lock, flags);
> }
>
> @@ -379,9 +380,7 @@ int cmdq_pkt_flush_async(struct cmdq_pkt *pkt, cmdq_async_flush_cb cb,
>
> if (client->timeout_ms != CMDQ_NO_TIMEOUT) {
> spin_lock_irqsave(&client->lock, flags);
> - if (client->pkt_cnt++ == 0)
> - mod_timer(&client->timer, jiffies +
> - msecs_to_jiffies(client->timeout_ms));
> + client->pkt_cnt++;
> spin_unlock_irqrestore(&client->lock, flags);
> }
>
> @@ -391,6 +390,21 @@ int cmdq_pkt_flush_async(struct cmdq_pkt *pkt, cmdq_async_flush_cb cb,
> /* We can send next packet immediately, so just call txdone. */
> mbox_client_txdone(client->chan, 0);
>
> + if (client->timeout_ms != CMDQ_NO_TIMEOUT) {
> + spin_lock_irqsave(&client->lock, flags);
> + /*
> + * GCE HW maybe execute too quickly and the callback function
> + * may be invoked earlier. If this happens, pkt_cnt is reduced
> + * by 1 in cmdq_pkt_flush_async_cb(). The timer is set only if
> + * pkt_cnt is greater than 0.
> + */
> + if (client->pkt_cnt > 0)
> + mod_timer(&client->timer, jiffies +
> + msecs_to_jiffies(client->timeout_ms *
> + client->pkt_cnt));
I think for some client, it care about execution time of one packet,
but some client care about total execution time of all packets, so we
should let client driver implement its own timeout handler and remove
handler in cmdq helper [1].
[1] https://patchwork.kernel.org/project/linux-mediatek/patch/[email protected]/
Regards,
Chun-Kuang.
> + spin_unlock_irqrestore(&client->lock, flags);
> + }
> +
> return 0;
> }
> EXPORT_SYMBOL(cmdq_pkt_flush_async);
> --
> 2.18.0
> _______________________________________________
> Linux-mediatek mailing list
> [email protected]
> http://lists.infradead.org/mailman/listinfo/linux-mediatek