The intention of this patchset is to correct clock enable and disable
order and vote for venus-ebi and cpucfg paths with average bandwidth
instad of peak bandwidth since with current implementation we are seeing
clock related warning during XO-SD and suspend device while video playback
---
Resending as all patches not updated properly because
of some mailing issues
Mansur Alisha Shaik (4):
venus: core: change clk enable and disable order in resume and suspend
venus: core: vote for video-mem path
venus: core: vote with average bandwidth and peak bandwidth as zero
venus: put dummy vote on video-mem path after last session release
drivers/media/platform/qcom/venus/core.c | 29 +++++++++++++++++++-------
drivers/media/platform/qcom/venus/pm_helpers.c | 3 +++
2 files changed, 25 insertions(+), 7 deletions(-)
--
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member
of Code Aurora Forum, hosted by The Linux Foundation
Currently we are voting for venus0-ebi path during buffer processing
with an average bandwidth of all the instances and unvoting during
session release.
While video streaming when we try to do XO-SD using the command
"echo mem > /sys/power/state command" , device is not entering
to suspend state and from interconnect summary seeing votes for venus0-ebi
Corrected this by voting for venus0-ebi path in venus_runtime_resume()
and unvote during venus_runtime_suspend().
Fixes: 7482a983d ("media: venus: redesign clocks and pm domains control")
Signed-off-by: Mansur Alisha Shaik <[email protected]>
---
Resending by adding () for functions to know it as function.
drivers/media/platform/qcom/venus/core.c | 14 +++++++++++++-
1 file changed, 13 insertions(+), 1 deletion(-)
diff --git a/drivers/media/platform/qcom/venus/core.c b/drivers/media/platform/qcom/venus/core.c
index 52a3886..064b6c8 100644
--- a/drivers/media/platform/qcom/venus/core.c
+++ b/drivers/media/platform/qcom/venus/core.c
@@ -363,8 +363,16 @@ static __maybe_unused int venus_runtime_suspend(struct device *dev)
ret = icc_set_bw(core->cpucfg_path, 0, 0);
if (ret)
- return ret;
+ goto err_poweron_core;
+
+ ret = icc_set_bw(core->video_path, 0, 0);
+ if (ret)
+ goto err_poweron_core;
+
+ return ret;
+err_poweron_core:
+ pm_ops->core_power(dev, POWER_ON);
return ret;
}
@@ -374,6 +382,10 @@ static __maybe_unused int venus_runtime_resume(struct device *dev)
const struct venus_pm_ops *pm_ops = core->pm_ops;
int ret;
+ ret = icc_set_bw(core->video_path, 0, kbps_to_icc(1000));
+ if (ret)
+ return ret;
+
ret = icc_set_bw(core->cpucfg_path, 0, kbps_to_icc(1000));
if (ret)
return ret;
--
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member
of Code Aurora Forum, hosted by The Linux Foundation
Currently video driver is voting after clk enable and un voting
before clk disable. Basically we should vote before clk enable
and un vote after clk disable.
Corrected this by changing the order of clk enable and clk disable.
Fixes: 7482a983d ("media: venus: redesign clocks and pm domains control")
Signed-off-by: Mansur Alisha Shaik <[email protected]>
---
drivers/media/platform/qcom/venus/core.c | 17 ++++++++++-------
1 file changed, 10 insertions(+), 7 deletions(-)
diff --git a/drivers/media/platform/qcom/venus/core.c b/drivers/media/platform/qcom/venus/core.c
index 6103aaf..52a3886 100644
--- a/drivers/media/platform/qcom/venus/core.c
+++ b/drivers/media/platform/qcom/venus/core.c
@@ -355,13 +355,16 @@ static __maybe_unused int venus_runtime_suspend(struct device *dev)
if (ret)
return ret;
+ if (pm_ops->core_power) {
+ ret = pm_ops->core_power(dev, POWER_OFF);
+ if (ret)
+ return ret;
+ }
+
ret = icc_set_bw(core->cpucfg_path, 0, 0);
if (ret)
return ret;
- if (pm_ops->core_power)
- ret = pm_ops->core_power(dev, POWER_OFF);
-
return ret;
}
@@ -371,16 +374,16 @@ static __maybe_unused int venus_runtime_resume(struct device *dev)
const struct venus_pm_ops *pm_ops = core->pm_ops;
int ret;
+ ret = icc_set_bw(core->cpucfg_path, 0, kbps_to_icc(1000));
+ if (ret)
+ return ret;
+
if (pm_ops->core_power) {
ret = pm_ops->core_power(dev, POWER_ON);
if (ret)
return ret;
}
- ret = icc_set_bw(core->cpucfg_path, 0, kbps_to_icc(1000));
- if (ret)
- return ret;
-
return hfi_core_resume(core, false);
}
--
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member
of Code Aurora Forum, hosted by The Linux Foundation
As per current implementation, we are unvoting "videom-mem" path
for last video session during vdec_session_release().
While video playback when we try to suspend device, we see video clock
warnings since votes are already removed during vdec_session_release().
corrected this by putting dummy vote on "video-mem" after last video
session release and unvoting it during suspend.
Signed-off-by: Mansur Alisha Shaik <[email protected]>
---
drivers/media/platform/qcom/venus/pm_helpers.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/media/platform/qcom/venus/pm_helpers.c b/drivers/media/platform/qcom/venus/pm_helpers.c
index 57877ea..c0a3524 100644
--- a/drivers/media/platform/qcom/venus/pm_helpers.c
+++ b/drivers/media/platform/qcom/venus/pm_helpers.c
@@ -212,6 +212,9 @@ static int load_scale_bw(struct venus_core *core)
}
mutex_unlock(&core->lock);
+ if (!total_avg && !total_peak)
+ total_avg = kbps_to_icc(1000);
+
dev_dbg(core->dev, VDBGL "total: avg_bw: %u, peak_bw: %u\n",
total_avg, total_peak);
--
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member
of Code Aurora Forum, hosted by The Linux Foundation
As per bandwidth table we are voting with average bandwidth
for "video-mem" and "cpu-cfg" paths as peak bandwidth is zero
in bandwidth table.
Signed-off-by: Mansur Alisha Shaik <[email protected]>
---
drivers/media/platform/qcom/venus/core.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/media/platform/qcom/venus/core.c b/drivers/media/platform/qcom/venus/core.c
index 064b6c8..c9669ad 100644
--- a/drivers/media/platform/qcom/venus/core.c
+++ b/drivers/media/platform/qcom/venus/core.c
@@ -382,11 +382,11 @@ static __maybe_unused int venus_runtime_resume(struct device *dev)
const struct venus_pm_ops *pm_ops = core->pm_ops;
int ret;
- ret = icc_set_bw(core->video_path, 0, kbps_to_icc(1000));
+ ret = icc_set_bw(core->video_path, kbps_to_icc(20000), 0);
if (ret)
return ret;
- ret = icc_set_bw(core->cpucfg_path, 0, kbps_to_icc(1000));
+ ret = icc_set_bw(core->cpucfg_path, kbps_to_icc(1000), 0);
if (ret)
return ret;
--
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member
of Code Aurora Forum, hosted by The Linux Foundation
Quoting Mansur Alisha Shaik (2020-09-16 18:26:01)
> diff --git a/drivers/media/platform/qcom/venus/core.c b/drivers/media/platform/qcom/venus/core.c
> index 52a3886..064b6c8 100644
> --- a/drivers/media/platform/qcom/venus/core.c
> +++ b/drivers/media/platform/qcom/venus/core.c
> @@ -363,8 +363,16 @@ static __maybe_unused int venus_runtime_suspend(struct device *dev)
>
> ret = icc_set_bw(core->cpucfg_path, 0, 0);
> if (ret)
> - return ret;
> + goto err_poweron_core;
> +
> + ret = icc_set_bw(core->video_path, 0, 0);
> + if (ret)
> + goto err_poweron_core;
> +
> + return ret;
>
> +err_poweron_core:
> + pm_ops->core_power(dev, POWER_ON);
Don't we need to put back the bandwidth from before suspend was entered
if the video_path fails?
> return ret;
> }
>
Quoting Mansur Alisha Shaik (2020-09-16 18:26:00)
> Currently video driver is voting after clk enable and un voting
> before clk disable. Basically we should vote before clk enable
> and un vote after clk disable.
>
> Corrected this by changing the order of clk enable and clk disable.
>
> Fixes: 7482a983d ("media: venus: redesign clocks and pm domains control")
> Signed-off-by: Mansur Alisha Shaik <[email protected]>
> ---
Reviewed-by: Stephen Boyd <[email protected]>
Quoting Mansur Alisha Shaik (2020-09-16 18:26:02)
> As per bandwidth table we are voting with average bandwidth
> for "video-mem" and "cpu-cfg" paths as peak bandwidth is zero
> in bandwidth table.
Why? It is in "the bandwidth table" but is there any reason why peak
bandwidth is 0 while average bandwidth is non-zero? Seems odd.
>
> Signed-off-by: Mansur Alisha Shaik <[email protected]>
> ---
Probably needs a Fixes tag?
Quoting Mansur Alisha Shaik (2020-09-16 18:26:03)
> As per current implementation, we are unvoting "videom-mem" path
> for last video session during vdec_session_release().
> While video playback when we try to suspend device, we see video clock
> warnings since votes are already removed during vdec_session_release().
>
> corrected this by putting dummy vote on "video-mem" after last video
> session release and unvoting it during suspend.
>
> Signed-off-by: Mansur Alisha Shaik <[email protected]>
> ---
Fixes tag? Should this be combined with an earlier change?
> drivers/media/platform/qcom/venus/pm_helpers.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/media/platform/qcom/venus/pm_helpers.c b/drivers/media/platform/qcom/venus/pm_helpers.c
> index 57877ea..c0a3524 100644
> --- a/drivers/media/platform/qcom/venus/pm_helpers.c
> +++ b/drivers/media/platform/qcom/venus/pm_helpers.c
> @@ -212,6 +212,9 @@ static int load_scale_bw(struct venus_core *core)
> }
> mutex_unlock(&core->lock);
>
> + if (!total_avg && !total_peak)
> + total_avg = kbps_to_icc(1000);
Can you add a comment in the code here? Just guessing but something
like..
/*
* Keep bandwidth enabled here so that vdec_session_release()
* can disable clks without clks getting stuck. We'll drop the
* bandwidth request when this device is suspended so this is a
* low value to keep things on but not high power until the
* device is powered down.
*/
> +
> dev_dbg(core->dev, VDBGL "total: avg_bw: %u, peak_bw: %u\n",
> total_avg, total_peak);
>