bus: qcom-ssc-block-bus: Fix some error handling paths
Subsystems (3)
cover bus: qcom-ssc-block-bus: Fix some error handling paths
Patch 1 should be straight forward. It removes some ioumap() from resources allocated with devm_ioremap_resource().
Patch 2 is more speculative. It releases some resources in the error handling path of the probe, as done in the .remove() function.
Finally, the .remove() function also has some
pm_runtime_disable(&pdev->dev); pm_clk_destroy(&pdev->dev);
calls.
Usually pm_runtime_disable() is paired with pm_runtime_enable() and I would expect some pm_clk_add(() calls related to pm_clk_destroy(). Neither exists.
So I wonder if these 2 calls are correct in the .remove() function. Thought?
Christophe JAILLET (2):
bus: qcom-ssc-block-bus: Remove some duplicated iounmap() calls
bus: qcom-ssc-block-bus: Fix the error handling path of
qcom_ssc_block_bus_probe()
drivers/bus/qcom-ssc-block-bus.c | 34 ++++++++++++++++++-------------- 1 file changed, 19 insertions(+), 15 deletions(-)
Signature · 1 line
2.48.1
PATCH 1/2 bus: qcom-ssc-block-bus: Remove some duplicated iounmap() calls
reg_mpm_sscaon_config[01] are allocated with devm_ioremap_resource(). So, they will be unmapped automatically by the manage resource framework.
Remove the incorrect explicit iounmap() calls from the remove function.
--- drivers/bus/qcom-ssc-block-bus.c | 3 --- 1 file changed, 3 deletions(-)
drivers/bus/qcom-ssc-block-bus.c +0 -3
@@ -356,9 +356,6 @@ static void qcom_ssc_block_bus_remove(struct platform_device *pdev)qcom_ssc_block_bus_deinit (&pdev->dev); -iounmap (data->reg_mpm_sscaon_config0); -iounmap (data->reg_mpm_sscaon_config1); -qcom_ssc_block_bus_pds_disable (data->pds, data->num_pds);qcom_ssc_block_bus_pds_detach (&pdev->dev, data->pds, data->num_pds);pm_runtime_disable (&pdev->dev);
Signature · 1 line
2.48.1
Re: [PATCH 1/2] bus: qcom-ssc-block-bus: Remove some duplicated iounmap() calls
Quoted 11 lines · Christophe JAILLET
reg_mpm_sscaon_config[01] are allocated with
devm_ioremap_resource(). So, they will be unmapped automatically by the manage resource framework.Remove the incorrect explicit
iounmap()calls from the remove function.Fixes 97d485edc1d9 ("bus: add driver for initializing the SSC bus on (some) qcom SoCs")Signed-off-by Christophe JAILLET <[email protected]>--- drivers/bus/qcom-ssc-block-bus.c | 3 --- 1 file changed, 3 deletions(-)
Signature · 2 lines
With best wishes Dmitry
PATCH 2/2 bus: qcom-ssc-block-bus: Fix the error handling path of qcom_ssc_block_bus_probe()
In reply to Christophe JAILLETIf qcom_ssc_block_bus_pds_enable() fails, the previous call to qcom_ssc_block_bus_pds_attach() must be undone, as already done in the remove function.
In order to do that, move the code related to the power domains management to the end of the function, in order to avoid many changes in all the error handling path that would need to go through the new error handling path.
---
This patch is compile tested only.
It is also speculative. Power management interaction can be sometimes tricky and I'm not 100% sure that moving this code in fine.
Review with care.
--- drivers/bus/qcom-ssc-block-bus.c | 31 +++++++++++++++++++------------ 1 file changed, 19 insertions(+), 12 deletions(-)
drivers/bus/qcom-ssc-block-bus.c +19 -12
@@ -264,18 +264,6 @@ static int qcom_ssc_block_bus_probe(struct platform_device *pdev)platform_set_drvdata (pdev, data); - data->pd_names = qcom_ssc_block_pd_names; - data->num_pds =ARRAY_SIZE (qcom_ssc_block_pd_names); - -/* power domains */ - ret =qcom_ssc_block_bus_pds_attach (&pdev->dev, data->pds, data->pd_names, data->num_pds); -if (ret <0 ) -return dev_err_probe (&pdev->dev, ret,"error when attaching power domains\n" ); - - ret =qcom_ssc_block_bus_pds_enable (data->pds, data->num_pds); -if (ret <0 ) -return dev_err_probe (&pdev->dev, ret,"error when enabling power domains\n" ); -/* low level overrides for when the HW logic doesn't "just work" */ res =platform_get_resource_byname (pdev,IORESOURCE_MEM ,"mpm_sscaon_config0" ); data->reg_mpm_sscaon_config0 =devm_ioremap_resource (&pdev->dev, res);
@@ -343,11 +331,30 @@ static int qcom_ssc_block_bus_probe(struct platform_device *pdev) data->ssc_axi_halt = halt_args.args[0 ]; +/* power domains */ + data->pd_names = qcom_ssc_block_pd_names; + data->num_pds =ARRAY_SIZE (qcom_ssc_block_pd_names); + + ret =qcom_ssc_block_bus_pds_attach (&pdev->dev, data->pds, data->pd_names, data->num_pds); +if (ret <0 ) +return dev_err_probe (&pdev->dev, ret,"error when attaching power domains\n" ); + + ret =qcom_ssc_block_bus_pds_enable (data->pds, data->num_pds); +if (ret <0 ) { +dev_err_probe (&pdev->dev, ret,"error when enabling power domains\n" ); + goto err_detach_pds_bus; + } +qcom_ssc_block_bus_init (&pdev->dev);of_platform_populate (np,NULL ,NULL , &pdev->dev);return 0 ; + +err_detach_pds_bus: +qcom_ssc_block_bus_pds_detach (&pdev->dev, data->pds, data->num_pds); + +return ret; }static void qcom_ssc_block_bus_remove (struct platform_device *pdev)
Signature · 1 line
2.48.1
Re: [PATCH 2/2] bus: qcom-ssc-block-bus: Fix the error handling path of qcom_ssc_block_bus_probe()
iirc it took me quite a long time to figure out the correct sequence for the bus to come up, so I'd be careful with that indeed. Sadly I can't easily test this on the original device right now, when I have time I want to upstream support for sdm845 which I could test more easily, but the sdm845 case is simpler so idk if testing on that would be sufficient.
Quoted 75 lines · Christophe JAILLET
If
qcom_ssc_block_bus_pds_enable()fails, the previous call toqcom_ssc_block_bus_pds_attach()must be undone, as already done in the remove function.In order to do that, move the code related to the power domains management to the end of the function, in order to avoid many changes in all the error handling path that would need to go through the new error handling path.
Fixes 97d485edc1d9 ("bus: add driver for initializing the SSC bus on (some) qcom SoCs")Signed-off-by Christophe JAILLET <[email protected]>---
This patch is compile tested only.It is also speculative. Power management interaction can be sometimes tricky and I'm not 100% sure that moving this code in fine.
Review with care.
--- drivers/bus/qcom-ssc-block-bus.c | 31 +++++++++++++++++++------------ 1 file changed, 19 insertions(+), 12 deletions(-)drivers/bus/qcom-ssc-block-bus.c +19
-12@@ -264,18 +264,6 @@ static int qcom_ssc_block_bus_probe(struct platform_device *pdev)platform_set_drvdata (pdev, data); - data->pd_names = qcom_ssc_block_pd_names; - data->num_pds =ARRAY_SIZE (qcom_ssc_block_pd_names); - -/* power domains */ - ret =qcom_ssc_block_bus_pds_attach (&pdev->dev, data->pds, data->pd_names, data->num_pds); -if (ret <0 ) -return dev_err_probe (&pdev->dev, ret,"error when attaching power domains\n" ); - - ret =qcom_ssc_block_bus_pds_enable (data->pds, data->num_pds); -if (ret <0 ) -return dev_err_probe (&pdev->dev, ret,"error when enabling power domains\n" ); -/* low level overrides for when the HW logic doesn't "just work" */ res =platform_get_resource_byname (pdev,IORESOURCE_MEM ,"mpm_sscaon_config0" ); data->reg_mpm_sscaon_config0 =devm_ioremap_resource (&pdev->dev, res);@@ -343,11 +331,30 @@ static int qcom_ssc_block_bus_probe(struct platform_device *pdev) data->ssc_axi_halt = halt_args.args[0 ]; +/* power domains */ + data->pd_names = qcom_ssc_block_pd_names; + data->num_pds =ARRAY_SIZE (qcom_ssc_block_pd_names); + + ret =qcom_ssc_block_bus_pds_attach (&pdev->dev, data->pds, data->pd_names, data->num_pds); +if (ret <0 ) +return dev_err_probe (&pdev->dev, ret,"error when attaching power domains\n" ); + + ret =qcom_ssc_block_bus_pds_enable (data->pds, data->num_pds); +if (ret <0 ) { +dev_err_probe (&pdev->dev, ret,"error when enabling power domains\n" ); + goto err_detach_pds_bus; + } +qcom_ssc_block_bus_init (&pdev->dev);of_platform_populate (np,NULL ,NULL , &pdev->dev);return 0 ; + +err_detach_pds_bus: +qcom_ssc_block_bus_pds_detach (&pdev->dev, data->pds, data->num_pds); + +return ret; }static void qcom_ssc_block_bus_remove (struct platform_device *pdev)
Quoted 8 lines · Christophe JAILLET
Patch 1 should be straight forward. It removes some
ioumap()from resources allocated withdevm_ioremap_resource().Patch 2 is more speculative. It releases some resources in the error handling path of the probe, as done in the .remove() function.
[...]
Applied, thanks!
[1/2] bus: qcom-ssc-block-bus: Remove some duplicated iounmap() calls
commit: a9ac4ba7dcace2b3b91e7b87bf0ba97c47edd94f
[2/2] bus: qcom-ssc-block-bus: Fix the error handling path of qcom_ssc_block_bus_probe()
commit: f41658cd081ad7697796b3dacd9a717a57919268Best regards,
Signature · 1 line
Bjorn Andersson <[email protected]>