[alsa-devel] [PATCH v3] ASoC: Replace snd_soc_acpi_check_hid with acpi_dev_present
Replace snd_soc_acpi_check_hid() with the generic acpi_dev_present() and remove the now unused snd_soc_acpi_check_hid function. This should have no functional change.
Signed-off-by: Jeremy Cline jeremy@jcline.org --- Changes in v3: - Remove the unnecessary "== true" and "!= true"
include/sound/soc-acpi.h | 3 --- sound/soc/soc-acpi.c | 32 ++------------------------------ 2 files changed, 2 insertions(+), 33 deletions(-)
diff --git a/include/sound/soc-acpi.h b/include/sound/soc-acpi.h index a7d8d335b043..057805489af3 100644 --- a/include/sound/soc-acpi.h +++ b/include/sound/soc-acpi.h @@ -49,9 +49,6 @@ snd_soc_acpi_find_package_from_hid(const u8 hid[ACPI_ID_LEN], struct snd_soc_acpi_mach * snd_soc_acpi_find_machine(struct snd_soc_acpi_mach *machines);
-/* acpi check hid */ -bool snd_soc_acpi_check_hid(const u8 hid[ACPI_ID_LEN]); - /** * snd_soc_acpi_mach: ACPI-based machine descriptor. Most of the fields are * related to the hardware, except for the firmware and topology file names. diff --git a/sound/soc/soc-acpi.c b/sound/soc/soc-acpi.c index f21df28bc28e..3543a7749490 100644 --- a/sound/soc/soc-acpi.c +++ b/sound/soc/soc-acpi.c @@ -49,41 +49,13 @@ const char *snd_soc_acpi_find_name_from_hid(const u8 hid[ACPI_ID_LEN]) } EXPORT_SYMBOL_GPL(snd_soc_acpi_find_name_from_hid);
-static acpi_status snd_soc_acpi_mach_match(acpi_handle handle, u32 level, - void *context, void **ret) -{ - unsigned long long sta; - acpi_status status; - - *(bool *)context = true; - status = acpi_evaluate_integer(handle, "_STA", NULL, &sta); - if (ACPI_FAILURE(status) || !(sta & ACPI_STA_DEVICE_PRESENT)) - *(bool *)context = false; - - return AE_OK; -} - -bool snd_soc_acpi_check_hid(const u8 hid[ACPI_ID_LEN]) -{ - acpi_status status; - bool found = false; - - status = acpi_get_devices(hid, snd_soc_acpi_mach_match, &found, NULL); - - if (ACPI_FAILURE(status)) - return false; - - return found; -} -EXPORT_SYMBOL_GPL(snd_soc_acpi_check_hid); - struct snd_soc_acpi_mach * snd_soc_acpi_find_machine(struct snd_soc_acpi_mach *machines) { struct snd_soc_acpi_mach *mach;
for (mach = machines; mach->id[0]; mach++) { - if (snd_soc_acpi_check_hid(mach->id) == true) { + if (acpi_dev_present(mach->id, NULL, -1)) { if (mach->machine_quirk == NULL) return mach;
@@ -163,7 +135,7 @@ struct snd_soc_acpi_mach *snd_soc_acpi_codec_list(void *arg) return mach;
for (i = 0; i < codec_list->num_codecs; i++) { - if (snd_soc_acpi_check_hid(codec_list->codecs[i]) != true) + if (!acpi_dev_present(codec_list->codecs[i], NULL, -1)) return NULL; }
On 11/10/2017 12:36 PM, Jeremy Cline wrote:
Replace snd_soc_acpi_check_hid() with the generic acpi_dev_present() and remove the now unused snd_soc_acpi_check_hid function. This should have no functional change.
The code looks ok, but I'd like to first add the fix for the quirk handling regression which appeared in 4.13 and is still there in 4.14. Current fix is at https://bugzilla.kernel.org/attachment.cgi?id=258581
Can we wait a couple of days on this one to avoid mixing bug fixes and renames? this would also make the stable branch fix easier to handle.
Signed-off-by: Jeremy Cline jeremy@jcline.org
Changes in v3:
- Remove the unnecessary "== true" and "!= true"
include/sound/soc-acpi.h | 3 --- sound/soc/soc-acpi.c | 32 ++------------------------------ 2 files changed, 2 insertions(+), 33 deletions(-)
diff --git a/include/sound/soc-acpi.h b/include/sound/soc-acpi.h index a7d8d335b043..057805489af3 100644 --- a/include/sound/soc-acpi.h +++ b/include/sound/soc-acpi.h @@ -49,9 +49,6 @@ snd_soc_acpi_find_package_from_hid(const u8 hid[ACPI_ID_LEN], struct snd_soc_acpi_mach * snd_soc_acpi_find_machine(struct snd_soc_acpi_mach *machines);
-/* acpi check hid */ -bool snd_soc_acpi_check_hid(const u8 hid[ACPI_ID_LEN]);
- /**
- snd_soc_acpi_mach: ACPI-based machine descriptor. Most of the fields are
- related to the hardware, except for the firmware and topology file names.
diff --git a/sound/soc/soc-acpi.c b/sound/soc/soc-acpi.c index f21df28bc28e..3543a7749490 100644 --- a/sound/soc/soc-acpi.c +++ b/sound/soc/soc-acpi.c @@ -49,41 +49,13 @@ const char *snd_soc_acpi_find_name_from_hid(const u8 hid[ACPI_ID_LEN]) } EXPORT_SYMBOL_GPL(snd_soc_acpi_find_name_from_hid);
-static acpi_status snd_soc_acpi_mach_match(acpi_handle handle, u32 level,
void *context, void **ret)
-{
- unsigned long long sta;
- acpi_status status;
- *(bool *)context = true;
- status = acpi_evaluate_integer(handle, "_STA", NULL, &sta);
- if (ACPI_FAILURE(status) || !(sta & ACPI_STA_DEVICE_PRESENT))
*(bool *)context = false;
- return AE_OK;
-}
-bool snd_soc_acpi_check_hid(const u8 hid[ACPI_ID_LEN]) -{
- acpi_status status;
- bool found = false;
- status = acpi_get_devices(hid, snd_soc_acpi_mach_match, &found, NULL);
- if (ACPI_FAILURE(status))
return false;
- return found;
-} -EXPORT_SYMBOL_GPL(snd_soc_acpi_check_hid);
struct snd_soc_acpi_mach * snd_soc_acpi_find_machine(struct snd_soc_acpi_mach *machines) { struct snd_soc_acpi_mach *mach;
for (mach = machines; mach->id[0]; mach++) {
if (snd_soc_acpi_check_hid(mach->id) == true) {
if (acpi_dev_present(mach->id, NULL, -1)) { if (mach->machine_quirk == NULL) return mach;
@@ -163,7 +135,7 @@ struct snd_soc_acpi_mach *snd_soc_acpi_codec_list(void *arg) return mach;
for (i = 0; i < codec_list->num_codecs; i++) {
if (snd_soc_acpi_check_hid(codec_list->codecs[i]) != true)
}if (!acpi_dev_present(codec_list->codecs[i], NULL, -1)) return NULL;
On Mon, Nov 13, 2017 at 01:43:57PM -0600, Pierre-Louis Bossart wrote:
On 11/10/2017 12:36 PM, Jeremy Cline wrote:
Replace snd_soc_acpi_check_hid() with the generic acpi_dev_present() and remove the now unused snd_soc_acpi_check_hid function. This should have no functional change.
The code looks ok, but I'd like to first add the fix for the quirk handling regression which appeared in 4.13 and is still there in 4.14. Current fix is at https://bugzilla.kernel.org/attachment.cgi?id=258581
Can we wait a couple of days on this one to avoid mixing bug fixes and renames? this would also make the stable branch fix easier to handle.
That makes perfect sense to me. I'm pretty new to all this, should I just keep an eye on things and rebase this once that patch lands?
Thanks, Jeremy
On 11/14/2017 01:12 PM, Jeremy Cline wrote:
On Mon, Nov 13, 2017 at 01:43:57PM -0600, Pierre-Louis Bossart wrote:
On 11/10/2017 12:36 PM, Jeremy Cline wrote:
Replace snd_soc_acpi_check_hid() with the generic acpi_dev_present() and remove the now unused snd_soc_acpi_check_hid function. This should have no functional change.
The code looks ok, but I'd like to first add the fix for the quirk handling regression which appeared in 4.13 and is still there in 4.14. Current fix is at https://bugzilla.kernel.org/attachment.cgi?id=258581
Can we wait a couple of days on this one to avoid mixing bug fixes and renames? this would also make the stable branch fix easier to handle.
That makes perfect sense to me. I'm pretty new to all this, should I just keep an eye on things and rebase this once that patch lands?
if it's easier I can submit both the fix and your change in the same series (keeping you as the author of course)
Thanks, Jeremy _______________________________________________ Alsa-devel mailing list Alsa-devel@alsa-project.org http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
On Tue, Nov 14, 2017 at 01:54:59PM -0600, Pierre-Louis Bossart wrote:
if it's easier I can submit both the fix and your change in the same series (keeping you as the author of course)
That does sound easier, but whatever is easiest for you works for me!
Thanks, Jeremy
participants (2)
-
Jeremy Cline
-
Pierre-Louis Bossart