* [PATCH] dmaengine: Use unique debugfs names
@ 2026-09-18 9:15 Linus Walleij
2026-09-18 14:34 ` Frank Li
` (2 more replies)
0 siblings, 3 replies; 5+ messages in thread
From: Linus Walleij @ 2026-09-18 9:15 UTC (permalink / raw)
To: Vinod Koul, Frank Li; +Cc: dmaengine, Linus Walleij
DMA40 registers multiple dma_device instances for the same platform
device. The DMAengine debugfs code uses dev_name(dma_dev->dev) as the
directory name, so the second and third registrations try to create the
same directory and print duplicate-name warnings.
Keep the existing device-name directory for the first registration, but
use a dev_id-suffixed fallback when that name already exists.
Assisted-by: LLM
Signed-off-by: Linus Walleij <linusw@kernel.org>
---
Fix duplicate direntry.
---
drivers/dma/dmaengine.c | 21 +++++++++++++++++++--
1 file changed, 19 insertions(+), 2 deletions(-)
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 6ffd8bd82154..a1c388def67a 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -33,6 +33,7 @@
#include <linux/acpi.h>
#include <linux/acpi_dma.h>
+#include <linux/dcache.h>
#include <linux/device.h>
#include <linux/dma-mapping.h>
#include <linux/dmaengine.h>
@@ -70,8 +71,24 @@ static struct dentry *rootdir;
static void dmaengine_debug_register(struct dma_device *dma_dev)
{
- dma_dev->dbg_dev_root = debugfs_create_dir(dev_name(dma_dev->dev),
- rootdir);
+ const char *name = dev_name(dma_dev->dev);
+ struct dentry *dentry;
+ char *uniq;
+
+ dentry = debugfs_lookup(name, rootdir);
+ if (dentry) {
+ dput(dentry);
+
+ uniq = kasprintf(GFP_KERNEL, "%s.%d", name, dma_dev->dev_id);
+ if (!uniq)
+ return;
+
+ dma_dev->dbg_dev_root = debugfs_create_dir(uniq, rootdir);
+ kfree(uniq);
+ } else {
+ dma_dev->dbg_dev_root = debugfs_create_dir(name, rootdir);
+ }
+
if (IS_ERR(dma_dev->dbg_dev_root))
dma_dev->dbg_dev_root = NULL;
}
---
base-commit: cee9395acd8043be0644b25c34bfa86623f2b935
change-id: 20260918-dmaengine-fixes-5c893ce71049
Best regards,
--
Linus Walleij <linusw@kernel.org>
^ permalink raw reply related [flat|nested] 5+ messages in thread* Re: [PATCH] dmaengine: Use unique debugfs names
2026-09-18 9:15 [PATCH] dmaengine: Use unique debugfs names Linus Walleij
@ 2026-09-18 14:34 ` Frank Li
2026-09-18 14:55 ` Linus Walleij
2026-09-27 8:58 ` Linus Walleij
2026-10-13 21:31 ` Frank Li
2 siblings, 1 reply; 5+ messages in thread
From: Frank Li @ 2026-09-18 14:34 UTC (permalink / raw)
To: Linus Walleij; +Cc: Vinod Koul, Frank Li, dmaengine
On Fri, Sep 18, 2026 at 11:15:14AM +0200, Linus Walleij wrote:
> DMA40 registers multiple dma_device instances for the same platform
> device.
d40_dmaengine_init() ? I see just register one dmaengine device. I think
it may have other problems.
Frank
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] dmaengine: Use unique debugfs names
2026-09-18 14:34 ` Frank Li
@ 2026-09-18 14:55 ` Linus Walleij
0 siblings, 0 replies; 5+ messages in thread
From: Linus Walleij @ 2026-09-18 14:55 UTC (permalink / raw)
To: Frank Li; +Cc: Vinod Koul, Frank Li, dmaengine
On Fri, Sep 18, 2026 at 4:34 PM Frank Li <Frank.li@oss.nxp.com> wrote:
> On Fri, Sep 18, 2026 at 11:15:14AM +0200, Linus Walleij wrote:
> > DMA40 registers multiple dma_device instances for the same platform
> > device.
>
> d40_dmaengine_init() ? I see just register one dmaengine device. I think
> it may have other problems.
It certainly has other problems because I have a serires of 21 patches
and counting just for the dma40 driver, haha :D
But this patch addresses a real problem.
They are declared together in struct d40_base.
Each gets its own channel list through d40_chan_init(), which points
every channel back to the corresponding dma_device:
d40c->chan.device = dma;
list_add_tail(&d40c->chan.device_node, &dma->channels);
All three deliberately receive the same parent device in
See d40_ops_init():
dev->dev = base->dev;
They are then registered independently:
dmaenginem_async_device_register(&base->dma_slave);
dmaenginem_async_device_register(&base->dma_memcpy);
dmaenginem_async_device_register(&base->dma_both);
Yours,
Linus Walleij
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] dmaengine: Use unique debugfs names
2026-09-18 9:15 [PATCH] dmaengine: Use unique debugfs names Linus Walleij
2026-09-18 14:34 ` Frank Li
@ 2026-09-27 8:58 ` Linus Walleij
2026-10-13 21:31 ` Frank Li
2 siblings, 0 replies; 5+ messages in thread
From: Linus Walleij @ 2026-09-27 8:58 UTC (permalink / raw)
To: Vinod Koul, Frank Li; +Cc: dmaengine
Hi folks,
On Fri, Sep 18, 2026 at 11:15 AM Linus Walleij <linusw@kernel.org> wrote:
> DMA40 registers multiple dma_device instances for the same platform
> device. The DMAengine debugfs code uses dev_name(dma_dev->dev) as the
> directory name, so the second and third registrations try to create the
> same directory and print duplicate-name warnings.
>
> Keep the existing device-name directory for the first registration, but
> use a dev_id-suffixed fallback when that name already exists.
>
> Assisted-by: LLM
> Signed-off-by: Linus Walleij <linusw@kernel.org>
Can we apply this patch?
I think it solves a pretty straight-forward issue with the same
device adding multiple engines.
Yours,
Linus Walleij
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] dmaengine: Use unique debugfs names
2026-09-18 9:15 [PATCH] dmaengine: Use unique debugfs names Linus Walleij
2026-09-18 14:34 ` Frank Li
2026-09-27 8:58 ` Linus Walleij
@ 2026-10-13 21:31 ` Frank Li
2 siblings, 0 replies; 5+ messages in thread
From: Frank Li @ 2026-10-13 21:31 UTC (permalink / raw)
To: Linus Walleij; +Cc: Vinod Koul, Frank Li, dmaengine
On Fri, Sep 18, 2026 at 11:15:14AM +0200, Linus Walleij wrote:
> DMA40 registers multiple dma_device instances for the same platform
> device. The DMAengine debugfs code uses dev_name(dma_dev->dev) as the
> directory name, so the second and third registrations try to create the
> same directory and print duplicate-name warnings.
>
> Keep the existing device-name directory for the first registration, but
> use a dev_id-suffixed fallback when that name already exists.
>
> Assisted-by: LLM
> Signed-off-by: Linus Walleij <linusw@kernel.org>
> ---
Reviewed-by: Frank Li <Frank.Li@nxp.com>
> Fix duplicate direntry.
> ---
> drivers/dma/dmaengine.c | 21 +++++++++++++++++++--
> 1 file changed, 19 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
> index 6ffd8bd82154..a1c388def67a 100644
> --- a/drivers/dma/dmaengine.c
> +++ b/drivers/dma/dmaengine.c
> @@ -33,6 +33,7 @@
>
> #include <linux/acpi.h>
> #include <linux/acpi_dma.h>
> +#include <linux/dcache.h>
> #include <linux/device.h>
> #include <linux/dma-mapping.h>
> #include <linux/dmaengine.h>
> @@ -70,8 +71,24 @@ static struct dentry *rootdir;
>
> static void dmaengine_debug_register(struct dma_device *dma_dev)
> {
> - dma_dev->dbg_dev_root = debugfs_create_dir(dev_name(dma_dev->dev),
> - rootdir);
> + const char *name = dev_name(dma_dev->dev);
> + struct dentry *dentry;
> + char *uniq;
> +
> + dentry = debugfs_lookup(name, rootdir);
> + if (dentry) {
> + dput(dentry);
> +
> + uniq = kasprintf(GFP_KERNEL, "%s.%d", name, dma_dev->dev_id);
> + if (!uniq)
> + return;
> +
> + dma_dev->dbg_dev_root = debugfs_create_dir(uniq, rootdir);
> + kfree(uniq);
> + } else {
> + dma_dev->dbg_dev_root = debugfs_create_dir(name, rootdir);
> + }
> +
> if (IS_ERR(dma_dev->dbg_dev_root))
> dma_dev->dbg_dev_root = NULL;
> }
>
> ---
> base-commit: cee9395acd8043be0644b25c34bfa86623f2b935
> change-id: 20260918-dmaengine-fixes-5c893ce71049
>
> Best regards,
> --
> Linus Walleij <linusw@kernel.org>
>
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2026-10-13 21:31 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-09-18 9:15 [PATCH] dmaengine: Use unique debugfs names Linus Walleij
2026-09-18 14:34 ` Frank Li
2026-09-18 14:55 ` Linus Walleij
2026-09-27 8:58 ` Linus Walleij
2026-10-13 21:31 ` Frank Li
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox