Quoting the header comments, IRQF_ONESHOT is "Used by threaded interrupts
which need to keep the irq line disabled until the threaded handler has
been run.". When applied to an interrupt that doesn't request a threaded
irq then IRQF_ONESHOT has a lesser known (undocumented?) side effect,
which it to disable the forced threading of irqs (and for "normal" kernels
it is a nop). In this case I can find no evidence that suppressing forced
threading is intentional. Had it been intentional then a driver must adopt
the raw_spinlock API in order to avoid deadlocks on PREEMPT_RT kernels
(and avoid calling any kernel API that uses regular spinlocks).
Fix this by removing the spurious additional flag.
This change is required for my Snapdragon 7cx Gen2 tablet to boot-to-GUI
with PREEMPT_RT enabled.
Signed-off-by: Daniel Thompson <[email protected]>
---
Notes:
Just in case anybody asks, yes! I did use coccinelle to do a quick scan
for similar issues. I didn't find any other instances in drivers/drm/ .
drivers/gpu/drm/msm/dsi/dsi_host.c | 2 +-
drivers/gpu/drm/msm/hdmi/hdmi.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c
index 6b3ced4aaaf5d..3a3f53f0c8ae1 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_host.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_host.c
@@ -1877,7 +1877,7 @@ int msm_dsi_host_init(struct msm_dsi *msm_dsi)
/* do not autoenable, will be enabled later */
ret = devm_request_irq(&pdev->dev, msm_host->irq, dsi_host_irq,
- IRQF_TRIGGER_HIGH | IRQF_ONESHOT | IRQF_NO_AUTOEN,
+ IRQF_TRIGGER_HIGH | IRQF_NO_AUTOEN,
"dsi_isr", msm_host);
if (ret < 0) {
dev_err(&pdev->dev, "failed to request IRQ%u: %d\n",
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c
index 3acdeae25caf0..a1bfbc4c74bf7 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi.c
@@ -301,7 +301,7 @@ int msm_hdmi_modeset_init(struct hdmi *hdmi,
}
ret = devm_request_irq(&pdev->dev, hdmi->irq,
- msm_hdmi_irq, IRQF_TRIGGER_HIGH | IRQF_ONESHOT,
+ msm_hdmi_irq, IRQF_TRIGGER_HIGH,
"hdmi_isr", hdmi);
if (ret < 0) {
DRM_DEV_ERROR(dev->dev, "failed to request IRQ%u: %d\n",
base-commit: e783362eb54cd99b2cac8b3a9aeac942e6f6ac07
--
2.34.1
On Thu, 27 Jan 2022 at 20:04, Daniel Thompson
<[email protected]> wrote:
>
> Quoting the header comments, IRQF_ONESHOT is "Used by threaded interrupts
> which need to keep the irq line disabled until the threaded handler has
> been run.". When applied to an interrupt that doesn't request a threaded
> irq then IRQF_ONESHOT has a lesser known (undocumented?) side effect,
> which it to disable the forced threading of irqs (and for "normal" kernels
> it is a nop). In this case I can find no evidence that suppressing forced
> threading is intentional. Had it been intentional then a driver must adopt
> the raw_spinlock API in order to avoid deadlocks on PREEMPT_RT kernels
> (and avoid calling any kernel API that uses regular spinlocks).
>
> Fix this by removing the spurious additional flag.
>
> This change is required for my Snapdragon 7cx Gen2 tablet to boot-to-GUI
> with PREEMPT_RT enabled.
>
> Signed-off-by: Daniel Thompson <[email protected]>
Could you please split it into two parts: separate DSI and HDMI patches
> ---
>
> Notes:
> Just in case anybody asks, yes! I did use coccinelle to do a quick scan
> for similar issues. I didn't find any other instances in drivers/drm/ .
>
> drivers/gpu/drm/msm/dsi/dsi_host.c | 2 +-
> drivers/gpu/drm/msm/hdmi/hdmi.c | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c
> index 6b3ced4aaaf5d..3a3f53f0c8ae1 100644
> --- a/drivers/gpu/drm/msm/dsi/dsi_host.c
> +++ b/drivers/gpu/drm/msm/dsi/dsi_host.c
> @@ -1877,7 +1877,7 @@ int msm_dsi_host_init(struct msm_dsi *msm_dsi)
>
> /* do not autoenable, will be enabled later */
> ret = devm_request_irq(&pdev->dev, msm_host->irq, dsi_host_irq,
> - IRQF_TRIGGER_HIGH | IRQF_ONESHOT | IRQF_NO_AUTOEN,
> + IRQF_TRIGGER_HIGH | IRQF_NO_AUTOEN,
> "dsi_isr", msm_host);
> if (ret < 0) {
> dev_err(&pdev->dev, "failed to request IRQ%u: %d\n",
> diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c
> index 3acdeae25caf0..a1bfbc4c74bf7 100644
> --- a/drivers/gpu/drm/msm/hdmi/hdmi.c
> +++ b/drivers/gpu/drm/msm/hdmi/hdmi.c
> @@ -301,7 +301,7 @@ int msm_hdmi_modeset_init(struct hdmi *hdmi,
> }
>
> ret = devm_request_irq(&pdev->dev, hdmi->irq,
> - msm_hdmi_irq, IRQF_TRIGGER_HIGH | IRQF_ONESHOT,
> + msm_hdmi_irq, IRQF_TRIGGER_HIGH,
> "hdmi_isr", hdmi);
> if (ret < 0) {
> DRM_DEV_ERROR(dev->dev, "failed to request IRQ%u: %d\n",
>
> base-commit: e783362eb54cd99b2cac8b3a9aeac942e6f6ac07
> --
> 2.34.1
>
--
With best wishes
Dmitry