[PATCH] ASoC: SOF: build compression interface into snd_sof.ko
From: Arnd Bergmann arnd@arndb.de
With CONFIG_SND_SOC_SOF_COMPRESS=m, the compression code is not built into a the main SOF driver when that is built-in:
x86_64-linux-ld: sound/soc/sof/ipc.o: in function `ipc_stream_message': ipc.c:(.text+0x5a2): undefined reference to `snd_sof_compr_fragment_elapsed' x86_64-linux-ld: sound/soc/sof/topology.o: in function `sof_dai_load': topology.c:(.text+0x32d1): undefined reference to `snd_sof_compr_init_elapsed_work' x86_64-linux-ld: topology.c:(.text+0x32e1): undefined reference to `snd_sof_compr_init_elapsed_work'
Make this a 'bool' symbol so it just decides whether the code gets built at all.
Fixes: 858f7a5c45ca ("ASoC: SOF: Introduce fragment elapsed notification API") Signed-off-by: Arnd Bergmann arnd@arndb.de --- sound/soc/sof/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/sound/soc/sof/Kconfig b/sound/soc/sof/Kconfig index 6bb4db87af03..041c54639c4d 100644 --- a/sound/soc/sof/Kconfig +++ b/sound/soc/sof/Kconfig @@ -47,7 +47,7 @@ config SND_SOC_SOF_OF Say Y if you need this option. If unsure select "N".
config SND_SOC_SOF_COMPRESS - tristate + bool select SND_SOC_COMPRESS
config SND_SOC_SOF_DEBUG_PROBES
On 11/8/21 5:11 AM, Arnd Bergmann wrote:
From: Arnd Bergmann arnd@arndb.de
With CONFIG_SND_SOC_SOF_COMPRESS=m, the compression code is not built into a the main SOF driver when that is built-in:
x86_64-linux-ld: sound/soc/sof/ipc.o: in function `ipc_stream_message': ipc.c:(.text+0x5a2): undefined reference to `snd_sof_compr_fragment_elapsed' x86_64-linux-ld: sound/soc/sof/topology.o: in function `sof_dai_load': topology.c:(.text+0x32d1): undefined reference to `snd_sof_compr_init_elapsed_work' x86_64-linux-ld: topology.c:(.text+0x32e1): undefined reference to `snd_sof_compr_init_elapsed_work'
Make this a 'bool' symbol so it just decides whether the code gets built at all.
Fixes: 858f7a5c45ca ("ASoC: SOF: Introduce fragment elapsed notification API") Signed-off-by: Arnd Bergmann arnd@arndb.de
It's Monday morning and my memory is still foggy but I think we fixed this problem with https://github.com/thesofproject/linux/pull/3180, where we changed the Kconfigs for i.MX. We haven't sent this update upstream for some reason.
Arnd, can you share the configuration that breaks with the existing upstream code, I can check if the problem still exists.
Thanks!
On 11/8/21 3:39 PM, Pierre-Louis Bossart wrote:
On 11/8/21 5:11 AM, Arnd Bergmann wrote:
From: Arnd Bergmann arnd@arndb.de
With CONFIG_SND_SOC_SOF_COMPRESS=m, the compression code is not built into a the main SOF driver when that is built-in:
x86_64-linux-ld: sound/soc/sof/ipc.o: in function `ipc_stream_message': ipc.c:(.text+0x5a2): undefined reference to `snd_sof_compr_fragment_elapsed' x86_64-linux-ld: sound/soc/sof/topology.o: in function `sof_dai_load': topology.c:(.text+0x32d1): undefined reference to `snd_sof_compr_init_elapsed_work' x86_64-linux-ld: topology.c:(.text+0x32e1): undefined reference to `snd_sof_compr_init_elapsed_work'
Make this a 'bool' symbol so it just decides whether the code gets built at all.
Fixes: 858f7a5c45ca ("ASoC: SOF: Introduce fragment elapsed notification API") Signed-off-by: Arnd Bergmann arnd@arndb.de
It's Monday morning and my memory is still foggy but I think we fixed this problem with https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com..., where we changed the Kconfigs for i.MX. We haven't sent this update upstream for some reason.
Arnd, can you share the configuration that breaks with the existing upstream code, I can check if the problem still exists.
Maybe someone forgot :) to send https://github.com/thesofproject/linux/pull/3180/commits/7122edc88d13db8ba83... upstream.
I think that's me.
Arnd can you run your scripts with https://github.com/thesofproject/linux/pull/3180/commits/7122edc88d13db8ba83.... I also attached the patch
if it's easier to apply.
On Mon, Nov 8, 2021 at 3:13 PM Daniel Baluta daniel.baluta@nxp.com wrote:
On 11/8/21 3:39 PM, Pierre-Louis Bossart wrote:
On 11/8/21 5:11 AM, Arnd Bergmann wrote:
From: Arnd Bergmann arnd@arndb.de
Fixes: 858f7a5c45ca ("ASoC: SOF: Introduce fragment elapsed notification API") Signed-off-by: Arnd Bergmann arnd@arndb.de
It's Monday morning and my memory is still foggy but I think we fixed this problem with https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com..., where we changed the Kconfigs for i.MX. We haven't sent this update upstream for some reason.
Arnd, can you share the configuration that breaks with the existing upstream code, I can check if the problem still exists.
Maybe someone forgot :) to send https://github.com/thesofproject/linux/pull/3180/commits/7122edc88d13db8ba83... upstream.
I think that's me.
Arnd can you run your scripts with https://github.com/thesofproject/linux/pull/3180/commits/7122edc88d13db8ba83.... I also attached the patch
I added that to my randconfig tree now. I did a cherry-pick of that commit, I tried a 'git merge' first, but that pulled in a lot of other stuff that conflicts with linux-next.
Your patch does not fix the link error though, I'm fairly sure we need my one-liner on top still. In this case, SND_SOC_SOF=y gets selected by SND_SOC_SOF_INTEL_COMMON=y, while SND_SOC_SOF_COMPRESS=m gets selected by SND_SOC_SOF_IMX_COMMON=m, and then ignored because the Makefile only cares about =y objects built into the common module.
Arnd
On 11/8/21 10:15 AM, Arnd Bergmann wrote:
On Mon, Nov 8, 2021 at 3:13 PM Daniel Baluta daniel.baluta@nxp.com wrote:
On 11/8/21 3:39 PM, Pierre-Louis Bossart wrote:
On 11/8/21 5:11 AM, Arnd Bergmann wrote:
From: Arnd Bergmann arnd@arndb.de
Fixes: 858f7a5c45ca ("ASoC: SOF: Introduce fragment elapsed notification API") Signed-off-by: Arnd Bergmann arnd@arndb.de
It's Monday morning and my memory is still foggy but I think we fixed this problem with https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com..., where we changed the Kconfigs for i.MX. We haven't sent this update upstream for some reason.
Arnd, can you share the configuration that breaks with the existing upstream code, I can check if the problem still exists.
Ack, the problem still exists and is fixed by this patch
Tested-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com
On a related note, with this randconfig we have a separate problem on an AMD patch on the SOF tree
ERROR: modpost: "snd_amd_acp_find_config" [sound/soc/amd/snd-acp-config] is a static EXPORT_SYMBOL
That wasn't detected earlier, any idea what Kconfig option I need to enable to filter this out?
On Mon, Nov 8, 2021 at 6:18 PM Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com wrote:
On 11/8/21 10:15 AM, Arnd Bergmann wrote:
On Mon, Nov 8, 2021 at 3:13 PM Daniel Baluta daniel.baluta@nxp.com wrote:
On a related note, with this randconfig we have a separate problem on an AMD patch on the SOF tree
ERROR: modpost: "snd_amd_acp_find_config" [sound/soc/amd/snd-acp-config] is a static EXPORT_SYMBOL
That wasn't detected earlier, any idea what Kconfig option I need to enable to filter this out?
I don't see that symbol on linux-next at all, so this must be a bug that is specific to the current SOF tree. The message tells you that snd_amd_acp_find_config should not be exported at all since it's static, or it should not be marked static because it is used by other modules. IOW having a symbol that is both static and exported means it's not usable from built-in drivers.
Arnd
On 11/8/21 1:11 PM, Arnd Bergmann wrote:
From: Arnd Bergmann arnd@arndb.de
With CONFIG_SND_SOC_SOF_COMPRESS=m, the compression code is not built into a the main SOF driver when that is built-in:
x86_64-linux-ld: sound/soc/sof/ipc.o: in function `ipc_stream_message': ipc.c:(.text+0x5a2): undefined reference to `snd_sof_compr_fragment_elapsed' x86_64-linux-ld: sound/soc/sof/topology.o: in function `sof_dai_load': topology.c:(.text+0x32d1): undefined reference to `snd_sof_compr_init_elapsed_work' x86_64-linux-ld: topology.c:(.text+0x32e1): undefined reference to `snd_sof_compr_init_elapsed_work'
Make this a 'bool' symbol so it just decides whether the code gets built at all.
Fixes: 858f7a5c45ca ("ASoC: SOF: Introduce fragment elapsed notification API") Signed-off-by: Arnd Bergmann arnd@arndb.de
Indeed this is a bug and must be fixed. Thanks Arnd!
Reviewed-by: Daniel Baluta daniel.baluta@nxp.com
sound/soc/sof/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/sound/soc/sof/Kconfig b/sound/soc/sof/Kconfig index 6bb4db87af03..041c54639c4d 100644 --- a/sound/soc/sof/Kconfig +++ b/sound/soc/sof/Kconfig @@ -47,7 +47,7 @@ config SND_SOC_SOF_OF Say Y if you need this option. If unsure select "N".
config SND_SOC_SOF_COMPRESS
- tristate
bool select SND_SOC_COMPRESS
config SND_SOC_SOF_DEBUG_PROBES
On Mon, 8 Nov 2021 12:11:14 +0100, Arnd Bergmann wrote:
From: Arnd Bergmann arnd@arndb.de
With CONFIG_SND_SOC_SOF_COMPRESS=m, the compression code is not built into a the main SOF driver when that is built-in:
x86_64-linux-ld: sound/soc/sof/ipc.o: in function `ipc_stream_message': ipc.c:(.text+0x5a2): undefined reference to `snd_sof_compr_fragment_elapsed' x86_64-linux-ld: sound/soc/sof/topology.o: in function `sof_dai_load': topology.c:(.text+0x32d1): undefined reference to `snd_sof_compr_init_elapsed_work' x86_64-linux-ld: topology.c:(.text+0x32e1): undefined reference to `snd_sof_compr_init_elapsed_work'
[...]
Applied to
https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-linus
Thanks!
[1/1] ASoC: SOF: build compression interface into snd_sof.ko commit: 1218f06cb3c6e2c51699998bc17c0d9a41ab37a6
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
participants (4)
-
Arnd Bergmann -
Daniel Baluta -
Mark Brown -
Pierre-Louis Bossart