* [PATCH v2] dmaengine: tegra: only advertise supported pause operations
@ 2026-11-10 3:02 James Hilliard
2026-11-16 21:38 ` Frank Li
0 siblings, 1 reply; 2+ messages in thread
From: James Hilliard @ 2026-11-10 3:02 UTC (permalink / raw)
To: Laxman Dewangan, Jon Hunter, Vinod Koul, Frank Li,
Thierry Reding, Pavan Kunapuli, Dmitry Osipenko
Cc: Rajesh Gumasta, Thierry Reding, Akhil R, dmaengine, linux-tegra,
linux-kernel, stable, James Hilliard
Tegra186 does not support pausing a channel, but the driver installs
device_pause and device_resume for every variant. Both callbacks return
-ENOSYS on Tegra186.
dma_get_slave_caps() derives cmd_pause and cmd_resume from the presence
of these callbacks. Clients such as generic 8250 therefore accept RX DMA
channels that cannot satisfy their pause requirement. A receive timeout
then reaches an unsupported pause operation instead of being excluded
when the channel capabilities are checked.
Only install the pause and resume callbacks when the matched chip data
advertises hardware support. Keep them available on Tegra194 and later,
and remove their now-redundant checks for unsupported hardware.
Fixes: ee17028009d4 ("dmaengine: tegra: Add tegra gpcdma driver")
Cc: stable@vger.kernel.org
Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
---
Changes in v2:
- Remove the capability checks from the pause and resume callbacks now that
they are registered only on hardware that supports them.
- Link to v1: https://patch.msgid.link/20261108-submit-tegra186-dma-pause-caps-v1-v1-1-9cc9b17a033d@gmail.com
To: Laxman Dewangan <ldewangan@nvidia.com>
To: Jon Hunter <jonathanh@nvidia.com>
To: Vinod Koul <vkoul@kernel.org>
To: Frank Li <Frank.Li@kernel.org>
To: Thierry Reding <thierry.reding@kernel.org>
To: Akhil R <akhilrajeev@nvidia.com>
To: Pavan Kunapuli <pkunapuli@nvidia.com>
To: Dmitry Osipenko <digetx@gmail.com>
Cc: Thierry Reding <treding@nvidia.com>
Cc: dmaengine@vger.kernel.org
Cc: linux-tegra@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
---
drivers/dma/tegra186-gpc-dma.c | 12 ++++--------
1 file changed, 4 insertions(+), 8 deletions(-)
diff --git a/drivers/dma/tegra186-gpc-dma.c b/drivers/dma/tegra186-gpc-dma.c
index e76c58f213d8..bd606b71e714 100644
--- a/drivers/dma/tegra186-gpc-dma.c
+++ b/drivers/dma/tegra186-gpc-dma.c
@@ -419,9 +419,6 @@ static int tegra_dma_device_pause(struct dma_chan *dc)
unsigned long flags;
int ret;
- if (!tdc->tdma->chip_data->hw_support_pause)
- return -ENOSYS;
-
spin_lock_irqsave(&tdc->vc.lock, flags);
ret = tegra_dma_pause(tdc);
spin_unlock_irqrestore(&tdc->vc.lock, flags);
@@ -445,9 +442,6 @@ static int tegra_dma_device_resume(struct dma_chan *dc)
struct tegra_dma_channel *tdc = to_tegra_dma_chan(dc);
unsigned long flags;
- if (!tdc->tdma->chip_data->hw_support_pause)
- return -ENOSYS;
-
spin_lock_irqsave(&tdc->vc.lock, flags);
tegra_dma_resume(tdc);
spin_unlock_irqrestore(&tdc->vc.lock, flags);
@@ -1518,8 +1512,10 @@ static int tegra_dma_probe(struct platform_device *pdev)
tdma->dma_dev.device_terminate_all = tegra_dma_terminate_all;
tdma->dma_dev.device_tx_status = tegra_dma_tx_status;
tdma->dma_dev.device_issue_pending = tegra_dma_issue_pending;
- tdma->dma_dev.device_pause = tegra_dma_device_pause;
- tdma->dma_dev.device_resume = tegra_dma_device_resume;
+ if (cdata->hw_support_pause) {
+ tdma->dma_dev.device_pause = tegra_dma_device_pause;
+ tdma->dma_dev.device_resume = tegra_dma_device_resume;
+ }
tdma->dma_dev.device_synchronize = tegra_dma_chan_synchronize;
tdma->dma_dev.residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
---
base-commit: 6628e7d4c6b25c8d18f1345e3a20d21829aae7e4
change-id: 20261108-submit-tegra186-dma-pause-caps-v1-664d1320d502
Best regards,
--
James Hilliard <james.hilliard1@gmail.com>
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [PATCH v2] dmaengine: tegra: only advertise supported pause operations
2026-11-10 3:02 [PATCH v2] dmaengine: tegra: only advertise supported pause operations James Hilliard
@ 2026-11-16 21:38 ` Frank Li
0 siblings, 0 replies; 2+ messages in thread
From: Frank Li @ 2026-11-16 21:38 UTC (permalink / raw)
To: James Hilliard
Cc: Laxman Dewangan, Jon Hunter, Vinod Koul, Frank Li,
Thierry Reding, Pavan Kunapuli, Dmitry Osipenko, Rajesh Gumasta,
Thierry Reding, Akhil R, dmaengine, linux-tegra, linux-kernel,
stable
On Mon, Nov 09, 2026 at 08:02:14PM -0700, James Hilliard wrote:
> Tegra186 does not support pausing a channel, but the driver installs
> device_pause and device_resume for every variant. Both callbacks return
> -ENOSYS on Tegra186.
>
> dma_get_slave_caps() derives cmd_pause and cmd_resume from the presence
> of these callbacks. Clients such as generic 8250 therefore accept RX DMA
> channels that cannot satisfy their pause requirement. A receive timeout
> then reaches an unsupported pause operation instead of being excluded
> when the channel capabilities are checked.
>
> Only install the pause and resume callbacks when the matched chip data
> advertises hardware support. Keep them available on Tegra194 and later,
> and remove their now-redundant checks for unsupported hardware.
>
> Fixes: ee17028009d4 ("dmaengine: tegra: Add tegra gpcdma driver")
> Cc: stable@vger.kernel.org
> Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
> ---
Reviewed-by: Frank Li <Frank.Li@nxp.com>
> Changes in v2:
> - Remove the capability checks from the pause and resume callbacks now that
> they are registered only on hardware that supports them.
> - Link to v1: https://patch.msgid.link/20261108-submit-tegra186-dma-pause-caps-v1-v1-1-9cc9b17a033d@gmail.com
>
> To: Laxman Dewangan <ldewangan@nvidia.com>
> To: Jon Hunter <jonathanh@nvidia.com>
> To: Vinod Koul <vkoul@kernel.org>
> To: Frank Li <Frank.Li@kernel.org>
> To: Thierry Reding <thierry.reding@kernel.org>
> To: Akhil R <akhilrajeev@nvidia.com>
> To: Pavan Kunapuli <pkunapuli@nvidia.com>
> To: Dmitry Osipenko <digetx@gmail.com>
> Cc: Thierry Reding <treding@nvidia.com>
> Cc: dmaengine@vger.kernel.org
> Cc: linux-tegra@vger.kernel.org
> Cc: linux-kernel@vger.kernel.org
> ---
> drivers/dma/tegra186-gpc-dma.c | 12 ++++--------
> 1 file changed, 4 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/dma/tegra186-gpc-dma.c b/drivers/dma/tegra186-gpc-dma.c
> index e76c58f213d8..bd606b71e714 100644
> --- a/drivers/dma/tegra186-gpc-dma.c
> +++ b/drivers/dma/tegra186-gpc-dma.c
> @@ -419,9 +419,6 @@ static int tegra_dma_device_pause(struct dma_chan *dc)
> unsigned long flags;
> int ret;
>
> - if (!tdc->tdma->chip_data->hw_support_pause)
> - return -ENOSYS;
> -
> spin_lock_irqsave(&tdc->vc.lock, flags);
> ret = tegra_dma_pause(tdc);
> spin_unlock_irqrestore(&tdc->vc.lock, flags);
> @@ -445,9 +442,6 @@ static int tegra_dma_device_resume(struct dma_chan *dc)
> struct tegra_dma_channel *tdc = to_tegra_dma_chan(dc);
> unsigned long flags;
>
> - if (!tdc->tdma->chip_data->hw_support_pause)
> - return -ENOSYS;
> -
> spin_lock_irqsave(&tdc->vc.lock, flags);
> tegra_dma_resume(tdc);
> spin_unlock_irqrestore(&tdc->vc.lock, flags);
> @@ -1518,8 +1512,10 @@ static int tegra_dma_probe(struct platform_device *pdev)
> tdma->dma_dev.device_terminate_all = tegra_dma_terminate_all;
> tdma->dma_dev.device_tx_status = tegra_dma_tx_status;
> tdma->dma_dev.device_issue_pending = tegra_dma_issue_pending;
> - tdma->dma_dev.device_pause = tegra_dma_device_pause;
> - tdma->dma_dev.device_resume = tegra_dma_device_resume;
> + if (cdata->hw_support_pause) {
> + tdma->dma_dev.device_pause = tegra_dma_device_pause;
> + tdma->dma_dev.device_resume = tegra_dma_device_resume;
> + }
> tdma->dma_dev.device_synchronize = tegra_dma_chan_synchronize;
> tdma->dma_dev.residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
>
>
> ---
> base-commit: 6628e7d4c6b25c8d18f1345e3a20d21829aae7e4
> change-id: 20261108-submit-tegra186-dma-pause-caps-v1-664d1320d502
>
> Best regards,
> --
> James Hilliard <james.hilliard1@gmail.com>
>
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-11-16 21:38 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-11-10 3:02 [PATCH v2] dmaengine: tegra: only advertise supported pause operations James Hilliard
2026-11-16 21:38 ` Frank Li
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®