The patch

ASoC: rsnd: src: Avoid a potential deadlock

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git

All being well this means that it will be integrated into the linux-next tree (usually sometime in the next 24 hours) and sent to Linus during the next merge window (or sooner if it is a bug fix), however if problems are discovered then the patch may be dropped or reverted.

You may get further e-mails resulting from automated or manual testing and review of the tree, please engage with people reporting problems and send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they should be sent as incremental updates against current git, existing patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying to this mail.

Thanks, Mark

From ba164a49f8f7390b036713bf8a70a150a938c670 Mon Sep 17 00:00:00 2001

From: Jiada Wang jiada_wang@mentor.com Date: Thu, 7 Mar 2019 15:15:53 +0900 Subject: [PATCH] ASoC: rsnd: src: Avoid a potential deadlock

lockdep warns us that priv->lock and k->k_lock can cause a deadlock when after acquire of k->k_lock, process is interrupted by src, while in another routine of src .init, k->k_lock is acquired with priv->lock held.

This patch avoids a potential deadlock by not calling soc_device_match() in SRC .init callback, instead it adds new soc fields in priv->flags to differentiate SoCs.

Fixes: linux-next commit 7674bec4fc09 ("ASoC: rsnd: update BSDSR/BSDISR handling") Signed-off-by: Jiada Wang jiada_wang@mentor.com Acked-by: Kuninori Morimoto kuninori.morimoto.gx@renesas.com Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sh/rcar/core.c | 2 ++ sound/soc/sh/rcar/rsnd.h | 5 +++++ sound/soc/sh/rcar/src.c | 9 +-------- 3 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/sound/soc/sh/rcar/core.c b/sound/soc/sh/rcar/core.c index 9834474684b1..8d3758f862f1 100644 --- a/sound/soc/sh/rcar/core.c +++ b/sound/soc/sh/rcar/core.c @@ -110,6 +110,8 @@ static const struct of_device_id rsnd_of_match[] = { { .compatible = "renesas,rcar_sound-gen1", .data = (void *)RSND_GEN1 }, { .compatible = "renesas,rcar_sound-gen2", .data = (void *)RSND_GEN2 }, { .compatible = "renesas,rcar_sound-gen3", .data = (void *)RSND_GEN3 }, + /* Special Handling */ + { .compatible = "renesas,rcar_sound-r8a77990", .data = (void *)(RSND_GEN3 | RSND_SOC_E) }, {}, }; MODULE_DEVICE_TABLE(of, rsnd_of_match); diff --git a/sound/soc/sh/rcar/rsnd.h b/sound/soc/sh/rcar/rsnd.h index 90625c57847b..0e6ef4e18400 100644 --- a/sound/soc/sh/rcar/rsnd.h +++ b/sound/soc/sh/rcar/rsnd.h @@ -607,6 +607,8 @@ struct rsnd_priv { #define RSND_GEN1 (1 << 0) #define RSND_GEN2 (2 << 0) #define RSND_GEN3 (3 << 0) +#define RSND_SOC_MASK (0xFF << 4) +#define RSND_SOC_E (1 << 4) /* E1/E2/E3 */

/* * below value will be filled on rsnd_gen_probe() @@ -679,6 +681,9 @@ struct rsnd_priv { #define rsnd_is_gen1(priv) (((priv)->flags & RSND_GEN_MASK) == RSND_GEN1) #define rsnd_is_gen2(priv) (((priv)->flags & RSND_GEN_MASK) == RSND_GEN2) #define rsnd_is_gen3(priv) (((priv)->flags & RSND_GEN_MASK) == RSND_GEN3) +#define rsnd_is_e3(priv) (((priv)->flags & \ + (RSND_GEN_MASK | RSND_SOC_MASK)) == \ + (RSND_GEN3 | RSND_SOC_E))

#define rsnd_flags_has(p, f) ((p)->flags & (f)) #define rsnd_flags_set(p, f) ((p)->flags |= (f)) diff --git a/sound/soc/sh/rcar/src.c b/sound/soc/sh/rcar/src.c index db81e066b92e..45096a66c074 100644 --- a/sound/soc/sh/rcar/src.c +++ b/sound/soc/sh/rcar/src.c @@ -14,7 +14,6 @@ */

#include "rsnd.h" -#include <linux/sys_soc.h>

#define SRC_NAME "src"

@@ -189,18 +188,12 @@ const static u32 chan222222[] = { 0x00000006, /* 1 to 2 */ };

-static const struct soc_device_attribute ov_soc[] = { - { .soc_id = "r8a77990" }, /* E3 */ - { /* sentinel */ } -}; - static void rsnd_src_set_convert_rate(struct rsnd_dai_stream *io, struct rsnd_mod *mod) { struct rsnd_priv *priv = rsnd_mod_to_priv(mod); struct device *dev = rsnd_priv_to_dev(priv); struct snd_pcm_runtime *runtime = rsnd_io_to_runtime(io); - const struct soc_device_attribute *soc = soc_device_match(ov_soc); int is_play = rsnd_io_is_play(io); int use_src = 0; u32 fin, fout; @@ -307,7 +300,7 @@ static void rsnd_src_set_convert_rate(struct rsnd_dai_stream *io, /* * E3 need to overwrite */ - if (soc) + if (rsnd_is_e3(priv)) switch (rsnd_mod_id(mod)) { case 0: case 4:

Hi Geert

On 2019/03/12 18:15, Geert Uytterhoeven wrote:

Hi Jiada, Fabrizio,

On Thu, Mar 7, 2019 at 7:17 AM Jiada Wang jiada_wang@mentor.com wrote:

...

lockdep warns us that priv->lock and k->k_lock can cause a deadlock when after acquire of k->k_lock, process is interrupted by src, while in another routine of src .init, k->k_lock is acquired with priv->lock held.

This patch avoids a potential deadlock by not calling soc_device_match() in SRC .init callback, instead it adds new soc fields in priv->flags to differentiate SoCs.

Fixes: linux-next commit 7674bec4fc09 ("ASoC: rsnd: update BSDSR/BSDISR handling") Signed-off-by: Jiada Wang jiada_wang@mentor.com

---

sound/soc/sh/rcar/core.c | 2 ++ sound/soc/sh/rcar/rsnd.h | 5 +++++ sound/soc/sh/rcar/src.c | 9 +-------- 3 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/sound/soc/sh/rcar/core.c b/sound/soc/sh/rcar/core.c index 022996d2db13..4fe83e611c01 100644 --- a/sound/soc/sh/rcar/core.c +++ b/sound/soc/sh/rcar/core.c @@ -110,6 +110,8 @@ static const struct of_device_id rsnd_of_match[] = { { .compatible = "renesas,rcar_sound-gen1", .data = (void *)RSND_GEN1 }, { .compatible = "renesas,rcar_sound-gen2", .data = (void *)RSND_GEN2 }, { .compatible = "renesas,rcar_sound-gen3", .data = (void *)RSND_GEN3 },

•   /* Special Handling */
  

•   { .compatible = "renesas,rcar_sound-r8a77990", .data = (void *)(RSND_GEN3 | RSND_SOC_E) },
  

I guess we need an entry for RZ/G2E, too?

Sorry, I don't have user manual for RZ/G2E, if you can let me know where to find RZ/G2E's user manual, I can add entry for these SoCs

Thanks, Jiada

Gr{oetje,eeting}s,

                     Geert

Hi Fabrizio

On 2019/03/14 3:19, Fabrizio Castro wrote:

Hello Jiada,

...

From: Jiada Wang jiada_wang@mentor.com Sent: 13 March 2019 11:56 Subject: Re: [PATCH linux-next v2 1/2] ASoC: rsnd: src: Avoid a potential deadlock

Hi Geert

On 2019/03/12 18:15, Geert Uytterhoeven wrote:

...

Hi Jiada, Fabrizio,

On Thu, Mar 7, 2019 at 7:17 AM Jiada Wang jiada_wang@mentor.com wrote:

...

lockdep warns us that priv->lock and k->k_lock can cause a deadlock when after acquire of k->k_lock, process is interrupted by src, while in another routine of src .init, k->k_lock is acquired with priv->lock held.

This patch avoids a potential deadlock by not calling soc_device_match() in SRC .init callback, instead it adds new soc fields in priv->flags to differentiate SoCs.

Fixes: linux-next commit 7674bec4fc09 ("ASoC: rsnd: update BSDSR/BSDISR handling") Signed-off-by: Jiada Wang jiada_wang@mentor.com

---

sound/soc/sh/rcar/core.c | 2 ++ sound/soc/sh/rcar/rsnd.h | 5 +++++ sound/soc/sh/rcar/src.c | 9 +-------- 3 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/sound/soc/sh/rcar/core.c b/sound/soc/sh/rcar/core.c index 022996d2db13..4fe83e611c01 100644 --- a/sound/soc/sh/rcar/core.c +++ b/sound/soc/sh/rcar/core.c @@ -110,6 +110,8 @@ static const struct of_device_id rsnd_of_match[] = { { .compatible = "renesas,rcar_sound-gen1", .data = (void *)RSND_GEN1 }, { .compatible = "renesas,rcar_sound-gen2", .data = (void *)RSND_GEN2 }, { .compatible = "renesas,rcar_sound-gen3", .data = (void *)RSND_GEN3 },

•   /* Special Handling */
  

•   { .compatible = "renesas,rcar_sound-r8a77990", .data = (void *)(RSND_GEN3 | RSND_SOC_E) },
  

I guess we need an entry for RZ/G2E, too?

Geert has a point, I am looking at the User manual for the RZ/G2E and it looks like the implementation for the RZ/G2E is the same as the one for the r8a77990

...

...

Sorry, I don't have user manual for RZ/G2E, if you can let me know where to find RZ/G2E's user manual, I can add entry for these SoCs

If you prefer I could add that after running some tests? I do have a board with a RZ/G2E.

Yes, please since I have neither RZ/G2E board to test nor its user manual

Thanks, Jiada

Thanks, Fab

...

Thanks, Jiada

...

Gr{oetje,eeting}s,

                      Geert