| Message ID | 20201123102118.3866195-1-zhangqilong3@huawei.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | memory: mtk-smi: Fix PM usage counter unbalance in mtk_smi ops | expand |
On Mon, Nov 23, 2020 at 06:21:18PM +0800, Zhang Qilong wrote: > pm_runtime_get_sync will increment pm usage counter > even it failed. Forgetting to putting operation will > result in reference leak here. We fix it by replacing > it with pm_runtime_resume_and_get to keep usage counter > balanced. > > Fixes: 4f0a1a1ae3519 ("memory: mtk-smi: Invoke pm runtime_callback to enable clocks") > Signed-off-by: Zhang Qilong <zhangqilong3@huawei.com> > --- > drivers/memory/mtk-smi.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/drivers/memory/mtk-smi.c b/drivers/memory/mtk-smi.c > index ac350f8d1e20..82d09b88240e 100644 > --- a/drivers/memory/mtk-smi.c > +++ b/drivers/memory/mtk-smi.c > @@ -130,7 +130,7 @@ static void mtk_smi_clk_disable(const struct mtk_smi *smi) > > int mtk_smi_larb_get(struct device *larbdev) > { > - int ret = pm_runtime_get_sync(larbdev); > + int ret = pm_runtime_resume_and_get(larbdev); Thanks for the patch, but it cannot be applied. There is no pm_runtime_resume_and_get() in v5.10-rc1. I see you authored it around middle of November, so don't expect that it gets magically to all other trees. In such case - when you know that your patch cannot be applied or you suspect any dependencies, you should describe requirements for the patch. Best regards, Krzysztof
> > On Mon, Nov 23, 2020 at 06:21:18PM +0800, Zhang Qilong wrote: > > pm_runtime_get_sync will increment pm usage counter even it failed. > > Forgetting to putting operation will result in reference leak here. We > > fix it by replacing it with pm_runtime_resume_and_get to keep usage > > counter balanced. > > > > Fixes: 4f0a1a1ae3519 ("memory: mtk-smi: Invoke pm runtime_callback to > > enable clocks") > > Signed-off-by: Zhang Qilong <zhangqilong3@huawei.com> > > --- > > drivers/memory/mtk-smi.c | 4 ++-- > > 1 file changed, 2 insertions(+), 2 deletions(-) > > > > diff --git a/drivers/memory/mtk-smi.c b/drivers/memory/mtk-smi.c index > > ac350f8d1e20..82d09b88240e 100644 > > --- a/drivers/memory/mtk-smi.c > > +++ b/drivers/memory/mtk-smi.c > > @@ -130,7 +130,7 @@ static void mtk_smi_clk_disable(const struct > > mtk_smi *smi) > > > > int mtk_smi_larb_get(struct device *larbdev) { > > - int ret = pm_runtime_get_sync(larbdev); > > + int ret = pm_runtime_resume_and_get(larbdev); > > Thanks for the patch, but it cannot be applied. > > There is no pm_runtime_resume_and_get() in v5.10-rc1. I see you authored it > around middle of November, so don't expect that it gets magically to all other > trees. In such case - when you know that your patch cannot be applied or you > suspect any dependencies, you should describe requirements for the patch. OK, I get it. Best regards, Zhang > > Best regards, > Krzysztof
On Mon, Nov 23, 2020 at 06:21:18PM +0800, Zhang Qilong wrote: > pm_runtime_get_sync will increment pm usage counter > even it failed. Forgetting to putting operation will > result in reference leak here. We fix it by replacing > it with pm_runtime_resume_and_get to keep usage counter > balanced. > > Fixes: 4f0a1a1ae3519 ("memory: mtk-smi: Invoke pm runtime_callback to enable clocks") > Signed-off-by: Zhang Qilong <zhangqilong3@huawei.com> > --- > drivers/memory/mtk-smi.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) Thanks, applied. Best regards, Krzysztof
diff --git a/drivers/memory/mtk-smi.c b/drivers/memory/mtk-smi.c index ac350f8d1e20..82d09b88240e 100644 --- a/drivers/memory/mtk-smi.c +++ b/drivers/memory/mtk-smi.c @@ -130,7 +130,7 @@ static void mtk_smi_clk_disable(const struct mtk_smi *smi) int mtk_smi_larb_get(struct device *larbdev) { - int ret = pm_runtime_get_sync(larbdev); + int ret = pm_runtime_resume_and_get(larbdev); return (ret < 0) ? ret : 0; } @@ -374,7 +374,7 @@ static int __maybe_unused mtk_smi_larb_resume(struct device *dev) int ret; /* Power on smi-common. */ - ret = pm_runtime_get_sync(larb->smi_common_dev); + ret = pm_runtime_resume_and_get(larb->smi_common_dev); if (ret < 0) { dev_err(dev, "Failed to pm get for smi-common(%d).\n", ret); return ret;
pm_runtime_get_sync will increment pm usage counter even it failed. Forgetting to putting operation will result in reference leak here. We fix it by replacing it with pm_runtime_resume_and_get to keep usage counter balanced. Fixes: 4f0a1a1ae3519 ("memory: mtk-smi: Invoke pm runtime_callback to enable clocks") Signed-off-by: Zhang Qilong <zhangqilong3@huawei.com> --- drivers/memory/mtk-smi.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)