* [PATCH v2] fsldma: Set correct dma_mask based on hw capability
@ 2025-05-05 17:53 Ben Collins
2025-05-14 15:03 ` Vinod Koul
0 siblings, 1 reply; 2+ messages in thread
From: Ben Collins @ 2025-05-05 17:53 UTC (permalink / raw)
To: dmaengine; +Cc: Arnd Bergmann, Vinod Koul, linuxppc-dev, linux-kernel
The driver currently hardcodes DMA_BIT_MASK to 36-bits, which is only
correct on eloplus:
elo3 supports 40-bits
eloplus supports 36-bits
elo supports 32-bits
This is based on 0x08 cdar register documention in the respective
reference manuals. Set the dma mask accordingly.
Feedback from Arnd Bergmann:
- Use match data to set address bit mask
Signed-off-by: Ben Collins <bcollins@kernel.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Vinod Koul <vkoul@kernel.org>
Cc: linuxppc-dev@lists.ozlabs.org
Cc: dmaengine@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
---
drivers/dma/fsldma.c | 20 ++++++++++++++++----
drivers/dma/fsldma.h | 1 +
2 files changed, 17 insertions(+), 4 deletions(-)
diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c
index b5e7d18b97669..566db5a1b0bab 100644
--- a/drivers/dma/fsldma.c
+++ b/drivers/dma/fsldma.c
@@ -1226,6 +1226,8 @@ static int fsldma_of_probe(struct platform_device *op)
fdev->dev = &op->dev;
INIT_LIST_HEAD(&fdev->common.channels);
+ /* The DMA address bits supported for this device. */
+ fdev->addr_bits = (long)device_get_match_data(fdev->dev);
/* ioremap the registers for use */
fdev->regs = of_iomap(op->dev.of_node, 0);
@@ -1254,7 +1256,7 @@ static int fsldma_of_probe(struct platform_device *op)
fdev->common.directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
fdev->common.residue_granularity = DMA_RESIDUE_GRANULARITY_DESCRIPTOR;
- dma_set_mask(&(op->dev), DMA_BIT_MASK(36));
+ dma_set_mask(&(op->dev), DMA_BIT_MASK(fdev->addr_bits));
platform_set_drvdata(op, fdev);
@@ -1387,10 +1389,20 @@ static const struct dev_pm_ops fsldma_pm_ops = {
};
#endif
+/* The .data field is used for dma-bit-mask. */
static const struct of_device_id fsldma_of_ids[] = {
- { .compatible = "fsl,elo3-dma", },
- { .compatible = "fsl,eloplus-dma", },
- { .compatible = "fsl,elo-dma", },
+ {
+ .compatible = "fsl,elo3-dma",
+ .data = (void *)40,
+ },
+ {
+ .compatible = "fsl,eloplus-dma",
+ .data = (void *)36,
+ },
+ {
+ .compatible = "fsl,elo-dma",
+ .data = (void *)32,
+ },
{}
};
MODULE_DEVICE_TABLE(of, fsldma_of_ids);
diff --git a/drivers/dma/fsldma.h b/drivers/dma/fsldma.h
index 308bed0a560ac..2ea57df9b7f7d 100644
--- a/drivers/dma/fsldma.h
+++ b/drivers/dma/fsldma.h
@@ -124,6 +124,7 @@ struct fsldma_device {
struct fsldma_chan *chan[FSL_DMA_MAX_CHANS_PER_DEVICE];
u32 feature; /* The same as DMA channels */
int irq; /* Channel IRQ */
+ int addr_bits; /* DMA addressing bits supported */
};
/* Define macros for fsldma_chan->feature property */
--
2.49.0
--
Ben Collins
https://libjwt.io
https://github.com/benmcollins
--
3EC9 7598 1672 961A 1139 173A 5D5A 57C7 242B 22CF
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH v2] fsldma: Set correct dma_mask based on hw capability
2025-05-05 17:53 [PATCH v2] fsldma: Set correct dma_mask based on hw capability Ben Collins
@ 2025-05-14 15:03 ` Vinod Koul
0 siblings, 0 replies; 2+ messages in thread
From: Vinod Koul @ 2025-05-14 15:03 UTC (permalink / raw)
To: dmaengine, Ben Collins; +Cc: Arnd Bergmann, linuxppc-dev, linux-kernel
On Mon, 05 May 2025 13:53:07 -0400, Ben Collins wrote:
> The driver currently hardcodes DMA_BIT_MASK to 36-bits, which is only
> correct on eloplus:
>
> elo3 supports 40-bits
> eloplus supports 36-bits
> elo supports 32-bits
>
> [...]
Applied, thanks!
[1/1] fsldma: Set correct dma_mask based on hw capability
commit: 00ff4d68a9ae4c9315c166f1fafa47f4c0a65f6f
Best regards,
--
~Vinod
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2025-05-14 15:03 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-05-05 17:53 [PATCH v2] fsldma: Set correct dma_mask based on hw capability Ben Collins
2025-05-14 15:03 ` Vinod Koul
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).