[alsa-devel] HP dv series mute LED problems?
Hi Takashi,
The commit 26ebe0a28986f4845b2c5bea43ac5cc0b9f27f0a seems to undo the mute LED gpio fixes - at least on my dv6 series laptop.
Maybe, we need to come up with a better solution to accommodate the problem reported in this thread: http://mailman.alsa-project.org/pipermail/alsa-devel/2010-May/027352.html I believe, that's the reason for the commit, right?
Let me do a bit more digging.
Thanks, Kunal
At Wed, 12 May 2010 00:25:07 +0530, Kunal Gangakhedkar wrote:
Hi Takashi,
The commit 26ebe0a28986f4845b2c5bea43ac5cc0b9f27f0a seems to undo the mute LED gpio fixes - at least on my dv6 series laptop.
Maybe, we need to come up with a better solution to accommodate the problem reported in this thread: http://mailman.alsa-project.org/pipermail/alsa-devel/2010-May/027352.html I believe, that's the reason for the commit, right?
Yes. And as I checked the previous code, actually dv4 took GPIO 0. That's why I added it there back.
But, hmm, then it's really depending on the dv model number. Which PCI / codec SSID does your device have?
thanks,
Takashi
At Tue, 11 May 2010 22:23:42 +0200, I wrote:
At Wed, 12 May 2010 00:25:07 +0530, Kunal Gangakhedkar wrote:
Hi Takashi,
The commit 26ebe0a28986f4845b2c5bea43ac5cc0b9f27f0a seems to undo the mute LED gpio fixes - at least on my dv6 series laptop.
Maybe, we need to come up with a better solution to accommodate the problem reported in this thread: http://mailman.alsa-project.org/pipermail/alsa-devel/2010-May/027352.html I believe, that's the reason for the commit, right?
Yes. And as I checked the previous code, actually dv4 took GPIO 0. That's why I added it there back.
But, hmm, then it's really depending on the dv model number. Which PCI / codec SSID does your device have?
As a compromise, how about the additional patch below?
Takashi
--- diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index eb4ea3d..a0e06d8 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -104,6 +104,7 @@ enum { STAC_DELL_M4_2, STAC_DELL_M4_3, STAC_HP_M4, + STAC_HP_DV4, STAC_HP_DV5, STAC_HP_HDX, STAC_HP_DV4_1222NR, @@ -1691,6 +1692,7 @@ static unsigned int *stac92hd71bxx_brd_tbl[STAC_92HD71BXX_MODELS] = { [STAC_DELL_M4_2] = dell_m4_2_pin_configs, [STAC_DELL_M4_3] = dell_m4_3_pin_configs, [STAC_HP_M4] = NULL, + [STAC_HP_DV4] = NULL, [STAC_HP_DV5] = NULL, [STAC_HP_HDX] = NULL, [STAC_HP_DV4_1222NR] = NULL, @@ -1703,6 +1705,7 @@ static const char *stac92hd71bxx_models[STAC_92HD71BXX_MODELS] = { [STAC_DELL_M4_2] = "dell-m4-2", [STAC_DELL_M4_3] = "dell-m4-3", [STAC_HP_M4] = "hp-m4", + [STAC_HP_DV4] = "hp-dv4", [STAC_HP_DV5] = "hp-dv5", [STAC_HP_HDX] = "hp-hdx", [STAC_HP_DV4_1222NR] = "hp-dv4-1222nr", @@ -1721,7 +1724,7 @@ static struct snd_pci_quirk stac92hd71bxx_cfg_tbl[] = { SND_PCI_QUIRK_MASK(PCI_VENDOR_ID_HP, 0xfff0, 0x3080, "HP", STAC_HP_DV5), SND_PCI_QUIRK_MASK(PCI_VENDOR_ID_HP, 0xfff0, 0x30f0, - "HP dv4-7", STAC_HP_DV5), + "HP dv4-7", STAC_HP_DV4), SND_PCI_QUIRK_MASK(PCI_VENDOR_ID_HP, 0xfff0, 0x3600, "HP dv4-7", STAC_HP_DV5), SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x3610, @@ -5678,6 +5681,9 @@ again: spec->num_smuxes = 1; spec->num_dmuxes = 1; /* fallthrough */ + case STAC_HP_DV4: + spec->gpio_led = 0x01; + /* fallthrough */ case STAC_HP_DV5: snd_hda_codec_set_pincfg(codec, 0x0d, 0x90170010); stac92xx_auto_set_pinctl(codec, 0x0d, AC_PINCTL_OUT_EN); @@ -5686,7 +5692,6 @@ again: * detection. */ spec->hp_detect = 1; - spec->gpio_led = 0x01; break; case STAC_HP_HDX: spec->num_dmics = 1; @@ -5749,7 +5754,8 @@ again: }
/* enable bass on HP dv7 */ - if (spec->board_config == STAC_HP_DV5) { + if (spec->board_config == STAC_HP_DV4 || + spec->board_config == STAC_HP_DV5) { unsigned int cap; cap = snd_hda_param_read(codec, 0x1, AC_PAR_GPIO_CAP); cap &= AC_GPIO_IO_COUNT;
On Wednesday 12 May 2010 2:11:13 am Takashi Iwai wrote:
At Tue, 11 May 2010 22:23:42 +0200, I wrote:
At Wed, 12 May 2010 00:25:07 +0530, Kunal Gangakhedkar wrote:
Hi Takashi,
The commit 26ebe0a28986f4845b2c5bea43ac5cc0b9f27f0a seems to undo the mute LED gpio fixes - at least on my dv6 series laptop.
Maybe, we need to come up with a better solution to accommodate the problem reported in this thread: http://mailman.alsa-project.org/pipermail/alsa-devel/2010-May/027352.html I believe, that's the reason for the commit, right?
Yes. And as I checked the previous code, actually dv4 took GPIO 0. That's why I added it there back.
But, hmm, then it's really depending on the dv model number. Which PCI / codec SSID does your device have?
As a compromise, how about the additional patch below?
Yes, this patch fixes the problems on my dv6 machine and looks reasonable to me.
However, we need confirmation from Thoralf - whether it works for him too. I don't know if STAC_HP_DV4 can be reused for dv7 series machines - since I don't have access to verify it.
Tested-by: Kunal Gangakhedkar kunal.gangakhedkar@gmail.com (for dv6-1110ax) Acked-by: Kunal Gangakhedkar kunal.gangakhedkar@gmail.com
Takashi
diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index eb4ea3d..a0e06d8 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -104,6 +104,7 @@ enum { STAC_DELL_M4_2, STAC_DELL_M4_3, STAC_HP_M4,
- STAC_HP_DV4, STAC_HP_DV5, STAC_HP_HDX, STAC_HP_DV4_1222NR,
@@ -1691,6 +1692,7 @@ static unsigned int *stac92hd71bxx_brd_tbl[STAC_92HD71BXX_MODELS] = { [STAC_DELL_M4_2] = dell_m4_2_pin_configs, [STAC_DELL_M4_3] = dell_m4_3_pin_configs, [STAC_HP_M4] = NULL,
- [STAC_HP_DV4] = NULL, [STAC_HP_DV5] = NULL, [STAC_HP_HDX] = NULL, [STAC_HP_DV4_1222NR] = NULL,
@@ -1703,6 +1705,7 @@ static const char *stac92hd71bxx_models[STAC_92HD71BXX_MODELS] = { [STAC_DELL_M4_2] = "dell-m4-2", [STAC_DELL_M4_3] = "dell-m4-3", [STAC_HP_M4] = "hp-m4",
- [STAC_HP_DV4] = "hp-dv4", [STAC_HP_DV5] = "hp-dv5", [STAC_HP_HDX] = "hp-hdx", [STAC_HP_DV4_1222NR] = "hp-dv4-1222nr",
@@ -1721,7 +1724,7 @@ static struct snd_pci_quirk stac92hd71bxx_cfg_tbl[] = { SND_PCI_QUIRK_MASK(PCI_VENDOR_ID_HP, 0xfff0, 0x3080, "HP", STAC_HP_DV5), SND_PCI_QUIRK_MASK(PCI_VENDOR_ID_HP, 0xfff0, 0x30f0,
"HP dv4-7", STAC_HP_DV5),
"HP dv4-7", STAC_HP_DV4),@@ -5678,6 +5681,9 @@ again: spec->num_smuxes = 1; spec->num_dmuxes = 1; /* fallthrough */
- case STAC_HP_DV4:
spec->gpio_led = 0x01;/* fallthrough */@@ -5686,7 +5692,6 @@ again: * detection. */ spec->hp_detect = 1;
spec->gpio_led = 0x01;@@ -5749,7 +5754,8 @@ again: }
/* enable bass on HP dv7 */
- if (spec->board_config == STAC_HP_DV5) {
- if (spec->board_config == STAC_HP_DV4 ||
spec->board_config == STAC_HP_DV5) {
At Wed, 12 May 2010 11:33:06 +0530, Kunal Gangakhedkar wrote:
On Wednesday 12 May 2010 2:11:13 am Takashi Iwai wrote:
At Tue, 11 May 2010 22:23:42 +0200, I wrote:
At Wed, 12 May 2010 00:25:07 +0530, Kunal Gangakhedkar wrote:
Hi Takashi,
The commit 26ebe0a28986f4845b2c5bea43ac5cc0b9f27f0a seems to undo the mute LED gpio fixes - at least on my dv6 series laptop.
Maybe, we need to come up with a better solution to accommodate the problem reported in this thread: http://mailman.alsa-project.org/pipermail/alsa-devel/2010-May/027352.html I believe, that's the reason for the commit, right?
Yes. And as I checked the previous code, actually dv4 took GPIO 0. That's why I added it there back.
But, hmm, then it's really depending on the dv model number. Which PCI / codec SSID does your device have?
As a compromise, how about the additional patch below?
Yes, this patch fixes the problems on my dv6 machine and looks reasonable to me.
However, we need confirmation from Thoralf - whether it works for him too. I don't know if STAC_HP_DV4 can be reused for dv7 series machines - since I don't have access to verify it.
The PCI SSID was already given from his alsa-info.sh output. Actually, dv4 and dv7 seem to have very similar implementations and almost same SSIDs.
Tested-by: Kunal Gangakhedkar kunal.gangakhedkar@gmail.com (for dv6-1110ax) Acked-by: Kunal Gangakhedkar kunal.gangakhedkar@gmail.com
thanks,
Takashi
Takashi Iwai schrieb:
At Wed, 12 May 2010 11:33:06 +0530, Kunal Gangakhedkar wrote:
On Wednesday 12 May 2010 2:11:13 am Takashi Iwai wrote:
At Tue, 11 May 2010 22:23:42 +0200, I wrote:
At Wed, 12 May 2010 00:25:07 +0530, Kunal Gangakhedkar wrote:
Hi Takashi,
The commit 26ebe0a28986f4845b2c5bea43ac5cc0b9f27f0a seems to undo the mute LED gpio fixes - at least on my dv6 series laptop.
Maybe, we need to come up with a better solution to accommodate the problem reported in this thread: http://mailman.alsa-project.org/pipermail/alsa-devel/2010-May/027352.html I believe, that's the reason for the commit, right?
Yes. And as I checked the previous code, actually dv4 took GPIO 0. That's why I added it there back.
But, hmm, then it's really depending on the dv model number. Which PCI / codec SSID does your device have?
As a compromise, how about the additional patch below?
Yes, this patch fixes the problems on my dv6 machine and looks reasonable to me.
However, we need confirmation from Thoralf - whether it works for him too. I don't know if STAC_HP_DV4 can be reused for dv7 series machines - since I don't have access to verify it.
The PCI SSID was already given from his alsa-info.sh output. Actually, dv4 and dv7 seem to have very similar implementations and almost same SSIDs.
Tested-by: Kunal Gangakhedkar kunal.gangakhedkar@gmail.com (for dv6-1110ax) Acked-by: Kunal Gangakhedkar kunal.gangakhedkar@gmail.com
thanks,
Takashi _______________________________________________ Alsa-devel mailing list Alsa-devel@alsa-project.org http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
Hi Takashi,
sorry for my delay. I've a heavy workload with my job.
Try applying the patch, there is one error. IMHO it isn't a problem because my alsa-driver-1.0.23 patch_sigmatel.c file doesn't contain the line
- spec->gpio_led = 0x01;
at this place.
Hunk #2 succeeded at 1690 (offset -2 lines). Hunk #3 succeeded at 1703 (offset -2 lines). Hunk #4 succeeded at 1722 (offset -2 lines). Hunk #5 succeeded at 5676 (offset -5 lines). Hunk #6 FAILED at 5687. Hunk #7 succeeded at 5747 (offset -7 lines).
1 out of 7 hunks FAILED -- saving rejects to file sound/pci/hda/patch_sigmatel.c.rej
Here the content of patch_sigmatel.c.rej
*************** again: *** 5681,5687 **** * detection. */ spec->hp_detect = 1; - spec->gpio_led = 0x01; break; case STAC_HP_HDX: spec->num_dmics = 1; --- 5687,5692 ---- * detection. */ spec->hp_detect = 1; break; case STAC_HP_HDX: spec->num_dmics = 1;
But the patch doesn't solve the problem. Like before, the mute led isn't switched on/off (whithe/orange). Maybe set_hp_led_gpio overwrites the value for gpio_led, which is set static before ?
Thoralf
At Wed, 12 May 2010 21:55:50 +0200, Thoralf Freitag wrote:
Takashi Iwai schrieb:
At Wed, 12 May 2010 11:33:06 +0530, Kunal Gangakhedkar wrote:
On Wednesday 12 May 2010 2:11:13 am Takashi Iwai wrote:
At Tue, 11 May 2010 22:23:42 +0200, I wrote:
At Wed, 12 May 2010 00:25:07 +0530, Kunal Gangakhedkar wrote:
Hi Takashi,
The commit 26ebe0a28986f4845b2c5bea43ac5cc0b9f27f0a seems to undo the mute LED gpio fixes - at least on my dv6 series laptop.
Maybe, we need to come up with a better solution to accommodate the problem reported in this thread: http://mailman.alsa-project.org/pipermail/alsa-devel/2010-May/027352.html I believe, that's the reason for the commit, right?
Yes. And as I checked the previous code, actually dv4 took GPIO 0. That's why I added it there back.
But, hmm, then it's really depending on the dv model number. Which PCI / codec SSID does your device have?
As a compromise, how about the additional patch below?
Yes, this patch fixes the problems on my dv6 machine and looks reasonable to me.
However, we need confirmation from Thoralf - whether it works for him too. I don't know if STAC_HP_DV4 can be reused for dv7 series machines - since I don't have access to verify it.
The PCI SSID was already given from his alsa-info.sh output. Actually, dv4 and dv7 seem to have very similar implementations and almost same SSIDs.
Tested-by: Kunal Gangakhedkar kunal.gangakhedkar@gmail.com (for dv6-1110ax) Acked-by: Kunal Gangakhedkar kunal.gangakhedkar@gmail.com
thanks,
Takashi _______________________________________________ Alsa-devel mailing list Alsa-devel@alsa-project.org http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
Hi Takashi,
sorry for my delay. I've a heavy workload with my job.
Try applying the patch, there is one error. IMHO it isn't a problem because my alsa-driver-1.0.23 patch_sigmatel.c file doesn't contain the line
spec->gpio_led = 0x01;at this place.
Hunk #2 succeeded at 1690 (offset -2 lines). Hunk #3 succeeded at 1703 (offset -2 lines). Hunk #4 succeeded at 1722 (offset -2 lines). Hunk #5 succeeded at 5676 (offset -5 lines). Hunk #6 FAILED at 5687. Hunk #7 succeeded at 5747 (offset -7 lines).
1 out of 7 hunks FAILED -- saving rejects to file sound/pci/hda/patch_sigmatel.c.rej
Here the content of patch_sigmatel.c.rej
*************** again: *** 5681,5687 **** * detection. */ spec->hp_detect = 1;
spec->gpio_led = 0x01; break; case STAC_HP_HDX: spec->num_dmics = 1;--- 5687,5692 ---- * detection. */ spec->hp_detect = 1; break; case STAC_HP_HDX: spec->num_dmics = 1;
But the patch doesn't solve the problem. Like before, the mute led isn't switched on/off (whithe/orange). Maybe set_hp_led_gpio overwrites the value for gpio_led, which is set static before ?
Try alsa-driver snapshot tarball below. ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/snapshot/alsa-driver-snapshot.tar.gz
The fixed patch is already included there.
Takashi
Takashi Iwai schrieb:
At Wed, 12 May 2010 21:55:50 +0200, Thoralf Freitag wrote:
Takashi Iwai schrieb:
At Wed, 12 May 2010 11:33:06 +0530, Kunal Gangakhedkar wrote:
On Wednesday 12 May 2010 2:11:13 am Takashi Iwai wrote:
At Tue, 11 May 2010 22:23:42 +0200, I wrote:
At Wed, 12 May 2010 00:25:07 +0530, Kunal Gangakhedkar wrote:
> Hi Takashi, > > The commit 26ebe0a28986f4845b2c5bea43ac5cc0b9f27f0a seems to undo the mute LED > gpio fixes - at least on my dv6 series laptop. > > Maybe, we need to come up with a better solution to accommodate the problem > reported in this thread: > http://mailman.alsa-project.org/pipermail/alsa-devel/2010-May/027352.html > I believe, that's the reason for the commit, right? > > Yes. And as I checked the previous code, actually dv4 took GPIO 0. That's why I added it there back.
But, hmm, then it's really depending on the dv model number. Which PCI / codec SSID does your device have?
As a compromise, how about the additional patch below?
Yes, this patch fixes the problems on my dv6 machine and looks reasonable to me.
However, we need confirmation from Thoralf - whether it works for him too. I don't know if STAC_HP_DV4 can be reused for dv7 series machines - since I don't have access to verify it.
The PCI SSID was already given from his alsa-info.sh output. Actually, dv4 and dv7 seem to have very similar implementations and almost same SSIDs.
Tested-by: Kunal Gangakhedkar kunal.gangakhedkar@gmail.com (for dv6-1110ax) Acked-by: Kunal Gangakhedkar kunal.gangakhedkar@gmail.com
thanks,
Takashi _______________________________________________ Alsa-devel mailing list Alsa-devel@alsa-project.org http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
Hi Takashi,
sorry for my delay. I've a heavy workload with my job.
Try applying the patch, there is one error. IMHO it isn't a problem because my alsa-driver-1.0.23 patch_sigmatel.c file doesn't contain the line
spec->gpio_led = 0x01;at this place.
Hunk #2 succeeded at 1690 (offset -2 lines). Hunk #3 succeeded at 1703 (offset -2 lines). Hunk #4 succeeded at 1722 (offset -2 lines). Hunk #5 succeeded at 5676 (offset -5 lines). Hunk #6 FAILED at 5687. Hunk #7 succeeded at 5747 (offset -7 lines).
1 out of 7 hunks FAILED -- saving rejects to file sound/pci/hda/patch_sigmatel.c.rej
Here the content of patch_sigmatel.c.rej
*************** again: *** 5681,5687 **** * detection. */ spec->hp_detect = 1;
spec->gpio_led = 0x01; break; case STAC_HP_HDX: spec->num_dmics = 1;--- 5687,5692 ---- * detection. */ spec->hp_detect = 1; break; case STAC_HP_HDX: spec->num_dmics = 1;
But the patch doesn't solve the problem. Like before, the mute led isn't switched on/off (whithe/orange). Maybe set_hp_led_gpio overwrites the value for gpio_led, which is set static before ?
Try alsa-driver snapshot tarball below. ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/snapshot/alsa-driver-snapshot.tar.gz
The fixed patch is already included there.
Takashi _______________________________________________ Alsa-devel mailing list Alsa-devel@alsa-project.org http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
Hi Takashi,
I can confirm you that the snaphot above is working with my HP DV 7-1160eg.
Thank you very much.
TF
At Thu, 13 May 2010 00:35:43 +0200, Thoralf Freitag wrote:
Takashi Iwai schrieb:
At Wed, 12 May 2010 21:55:50 +0200, Thoralf Freitag wrote:
Takashi Iwai schrieb:
At Wed, 12 May 2010 11:33:06 +0530, Kunal Gangakhedkar wrote:
On Wednesday 12 May 2010 2:11:13 am Takashi Iwai wrote:
At Tue, 11 May 2010 22:23:42 +0200, I wrote:
> At Wed, 12 May 2010 00:25:07 +0530, > Kunal Gangakhedkar wrote: > > >> Hi Takashi, >> >> The commit 26ebe0a28986f4845b2c5bea43ac5cc0b9f27f0a seems to undo the mute LED >> gpio fixes - at least on my dv6 series laptop. >> >> Maybe, we need to come up with a better solution to accommodate the problem >> reported in this thread: >> http://mailman.alsa-project.org/pipermail/alsa-devel/2010-May/027352.html >> I believe, that's the reason for the commit, right? >> >> > Yes. And as I checked the previous code, actually dv4 took GPIO 0. > That's why I added it there back. > > But, hmm, then it's really depending on the dv model number. > Which PCI / codec SSID does your device have? > > As a compromise, how about the additional patch below?
Yes, this patch fixes the problems on my dv6 machine and looks reasonable to me.
However, we need confirmation from Thoralf - whether it works for him too. I don't know if STAC_HP_DV4 can be reused for dv7 series machines - since I don't have access to verify it.
The PCI SSID was already given from his alsa-info.sh output. Actually, dv4 and dv7 seem to have very similar implementations and almost same SSIDs.
Tested-by: Kunal Gangakhedkar kunal.gangakhedkar@gmail.com (for dv6-1110ax) Acked-by: Kunal Gangakhedkar kunal.gangakhedkar@gmail.com
thanks,
Takashi _______________________________________________ Alsa-devel mailing list Alsa-devel@alsa-project.org http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
Hi Takashi,
sorry for my delay. I've a heavy workload with my job.
Try applying the patch, there is one error. IMHO it isn't a problem because my alsa-driver-1.0.23 patch_sigmatel.c file doesn't contain the line
spec->gpio_led = 0x01;at this place.
Hunk #2 succeeded at 1690 (offset -2 lines). Hunk #3 succeeded at 1703 (offset -2 lines). Hunk #4 succeeded at 1722 (offset -2 lines). Hunk #5 succeeded at 5676 (offset -5 lines). Hunk #6 FAILED at 5687. Hunk #7 succeeded at 5747 (offset -7 lines).
1 out of 7 hunks FAILED -- saving rejects to file sound/pci/hda/patch_sigmatel.c.rej
Here the content of patch_sigmatel.c.rej
*************** again: *** 5681,5687 **** * detection. */ spec->hp_detect = 1;
spec->gpio_led = 0x01; break; case STAC_HP_HDX: spec->num_dmics = 1;--- 5687,5692 ---- * detection. */ spec->hp_detect = 1; break; case STAC_HP_HDX: spec->num_dmics = 1;
But the patch doesn't solve the problem. Like before, the mute led isn't switched on/off (whithe/orange). Maybe set_hp_led_gpio overwrites the value for gpio_led, which is set static before ?
Try alsa-driver snapshot tarball below. ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/snapshot/alsa-driver-snapshot.tar.gz
The fixed patch is already included there.
Takashi _______________________________________________ Alsa-devel mailing list Alsa-devel@alsa-project.org http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
Hi Takashi,
I can confirm you that the snaphot above is working with my HP DV 7-1160eg.
OK, thanks for confirmation.
Takashi
participants (3)
-
Kunal Gangakhedkar -
Takashi Iwai -
Thoralf Freitag