[alsa-devel] [PATCH] HDA-Intel: present up to 8 HDMI/DP outputs via "hdmi" device
Some new AMD cards have HDA codecs presenting 6 connected HDMI/DP pin nodes (plus 1 unconnected pin node) according to the ALSA card database.
Example: http://www.alsa-project.org/db/?f=de3ced7af41de0ed54d218650e5e2f16c511787b
Bump the maximum number of presented HDMI outputs per card via the "hdmi" PCM from 4 to 8 (so that the last possible device is DEV=7).
Note that HDMI PCM devices DEV=4..7 use shared PCM device numbers, so HDA cards that have over 4 audio PCM devices or multiple S/PDIF or modem devices will have their remaining PCM devices misrepresented as HDMI devices.
Signed-off-by: Anssi Hannula anssi.hannula@iki.fi ---
Another example: http://www.alsa-project.org/db/?f=cd12a9d38f51c47eeedd380e4c5713f0d9a6c9de
I do not have confirmed that all of these pins are actually usable (and therefore this patch is untested on affected hardware), but I guess that is probable. Maybe the large amount of them has something to do with DisplayPort supporting multiple displays via a single connector, but that is just a wild guess on my part...
But do we want some sort of a solution for the possible conflicts between non-HDMI devices in the 10+ physical PCM device numbering range?
src/conf/cards/HDA-Intel.conf | 92 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 92 insertions(+)
diff --git a/src/conf/cards/HDA-Intel.conf b/src/conf/cards/HDA-Intel.conf index 3957c12..7976b6c 100644 --- a/src/conf/cards/HDA-Intel.conf +++ b/src/conf/cards/HDA-Intel.conf @@ -305,6 +305,98 @@ HDA-Intel.pcm.hdmi.3 { } }
+HDA-Intel.pcm.hdmi.4 { + @args [ CARD AES0 AES1 AES2 AES3 ] + @args.CARD { type string } + @args.AES0 { type integer } + @args.AES1 { type integer } + @args.AES2 { type integer } + @args.AES3 { type integer } + @func refer + name { + @func concat + strings [ + "cards.HDA-Intel.pcm.hdmi.common:" + "CARD=" $CARD "," + "DEVICE=10," + "CTLINDEX=4," + "AES0=" $AES0 "," + "AES1=" $AES1 "," + "AES2=" $AES2 "," + "AES3=" $AES3 + ] + } +} + +HDA-Intel.pcm.hdmi.5 { + @args [ CARD AES0 AES1 AES2 AES3 ] + @args.CARD { type string } + @args.AES0 { type integer } + @args.AES1 { type integer } + @args.AES2 { type integer } + @args.AES3 { type integer } + @func refer + name { + @func concat + strings [ + "cards.HDA-Intel.pcm.hdmi.common:" + "CARD=" $CARD "," + "DEVICE=11," + "CTLINDEX=5," + "AES0=" $AES0 "," + "AES1=" $AES1 "," + "AES2=" $AES2 "," + "AES3=" $AES3 + ] + } +} + +HDA-Intel.pcm.hdmi.6 { + @args [ CARD AES0 AES1 AES2 AES3 ] + @args.CARD { type string } + @args.AES0 { type integer } + @args.AES1 { type integer } + @args.AES2 { type integer } + @args.AES3 { type integer } + @func refer + name { + @func concat + strings [ + "cards.HDA-Intel.pcm.hdmi.common:" + "CARD=" $CARD "," + "DEVICE=12," + "CTLINDEX=6," + "AES0=" $AES0 "," + "AES1=" $AES1 "," + "AES2=" $AES2 "," + "AES3=" $AES3 + ] + } +} + +HDA-Intel.pcm.hdmi.7 { + @args [ CARD AES0 AES1 AES2 AES3 ] + @args.CARD { type string } + @args.AES0 { type integer } + @args.AES1 { type integer } + @args.AES2 { type integer } + @args.AES3 { type integer } + @func refer + name { + @func concat + strings [ + "cards.HDA-Intel.pcm.hdmi.common:" + "CARD=" $CARD "," + "DEVICE=13," + "CTLINDEX=7," + "AES0=" $AES0 "," + "AES1=" $AES1 "," + "AES2=" $AES2 "," + "AES3=" $AES3 + ] + } +} + confdir:pcm/modem.conf
HDA-Intel.pcm.modem.0 {
23.09.2013 00:41, Anssi Hannula kirjoitti:
Some new AMD cards have HDA codecs presenting 6 connected HDMI/DP pin nodes (plus 1 unconnected pin node) according to the ALSA card database.
Example: http://www.alsa-project.org/db/?f=de3ced7af41de0ed54d218650e5e2f16c511787b
Bump the maximum number of presented HDMI outputs per card via the "hdmi" PCM from 4 to 8 (so that the last possible device is DEV=7).
Note that HDMI PCM devices DEV=4..7 use shared PCM device numbers, so HDA cards that have over 4 audio PCM devices or multiple S/PDIF or modem devices will have their remaining PCM devices misrepresented as HDMI devices.
Signed-off-by: Anssi Hannula anssi.hannula@iki.fi
Another example: http://www.alsa-project.org/db/?f=cd12a9d38f51c47eeedd380e4c5713f0d9a6c9de
I do not have confirmed that all of these pins are actually usable (and therefore this patch is untested on affected hardware), but I guess that is probable. Maybe the large amount of them has something to do with DisplayPort supporting multiple displays via a single connector, but that is just a wild guess on my part...
From what I can see from looking at the radeon driver and their wiki,
Radeon display cores DCE4+ have 6 pins, and DCE8+ have 7 pins.
http://www.x.org/wiki/RadeonFeature/#index7h2
So seems this is really needed. Now what is left is whether we care about the 10+ non-HDMI conflicts, and if so, what do we do about that?
But do we want some sort of a solution for the possible conflicts between non-HDMI devices in the 10+ physical PCM device numbering range?
src/conf/cards/HDA-Intel.conf | 92 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 92 insertions(+)
diff --git a/src/conf/cards/HDA-Intel.conf b/src/conf/cards/HDA-Intel.conf index 3957c12..7976b6c 100644 --- a/src/conf/cards/HDA-Intel.conf +++ b/src/conf/cards/HDA-Intel.conf @@ -305,6 +305,98 @@ HDA-Intel.pcm.hdmi.3 { } }
+HDA-Intel.pcm.hdmi.4 {
- @args [ CARD AES0 AES1 AES2 AES3 ]
- @args.CARD { type string }
- @args.AES0 { type integer }
- @args.AES1 { type integer }
- @args.AES2 { type integer }
- @args.AES3 { type integer }
- @func refer
- name {
@func concatstrings ["cards.HDA-Intel.pcm.hdmi.common:""CARD=" $CARD ",""DEVICE=10,""CTLINDEX=4,""AES0=" $AES0 ",""AES1=" $AES1 ",""AES2=" $AES2 ",""AES3=" $AES3]- }
+}
+HDA-Intel.pcm.hdmi.5 {
- @args [ CARD AES0 AES1 AES2 AES3 ]
- @args.CARD { type string }
- @args.AES0 { type integer }
- @args.AES1 { type integer }
- @args.AES2 { type integer }
- @args.AES3 { type integer }
- @func refer
- name {
@func concatstrings ["cards.HDA-Intel.pcm.hdmi.common:""CARD=" $CARD ",""DEVICE=11,""CTLINDEX=5,""AES0=" $AES0 ",""AES1=" $AES1 ",""AES2=" $AES2 ",""AES3=" $AES3]- }
+}
+HDA-Intel.pcm.hdmi.6 {
- @args [ CARD AES0 AES1 AES2 AES3 ]
- @args.CARD { type string }
- @args.AES0 { type integer }
- @args.AES1 { type integer }
- @args.AES2 { type integer }
- @args.AES3 { type integer }
- @func refer
- name {
@func concatstrings ["cards.HDA-Intel.pcm.hdmi.common:""CARD=" $CARD ",""DEVICE=12,""CTLINDEX=6,""AES0=" $AES0 ",""AES1=" $AES1 ",""AES2=" $AES2 ",""AES3=" $AES3]- }
+}
+HDA-Intel.pcm.hdmi.7 {
- @args [ CARD AES0 AES1 AES2 AES3 ]
- @args.CARD { type string }
- @args.AES0 { type integer }
- @args.AES1 { type integer }
- @args.AES2 { type integer }
- @args.AES3 { type integer }
- @func refer
- name {
@func concatstrings ["cards.HDA-Intel.pcm.hdmi.common:""CARD=" $CARD ",""DEVICE=13,""CTLINDEX=7,""AES0=" $AES0 ",""AES1=" $AES1 ",""AES2=" $AES2 ",""AES3=" $AES3]- }
+}
HDA-Intel.pcm.modem.0 {
At Mon, 23 Sep 2013 00:41:50 +0300, Anssi Hannula wrote:
Some new AMD cards have HDA codecs presenting 6 connected HDMI/DP pin nodes (plus 1 unconnected pin node) according to the ALSA card database.
Example: http://www.alsa-project.org/db/?f=de3ced7af41de0ed54d218650e5e2f16c511787b
Bump the maximum number of presented HDMI outputs per card via the "hdmi" PCM from 4 to 8 (so that the last possible device is DEV=7).
Note that HDMI PCM devices DEV=4..7 use shared PCM device numbers, so HDA cards that have over 4 audio PCM devices or multiple S/PDIF or modem devices will have their remaining PCM devices misrepresented as HDMI devices.
Signed-off-by: Anssi Hannula anssi.hannula@iki.fi
Applied now. Thanks.
Another example: http://www.alsa-project.org/db/?f=cd12a9d38f51c47eeedd380e4c5713f0d9a6c9de
I do not have confirmed that all of these pins are actually usable (and therefore this patch is untested on affected hardware), but I guess that is probable. Maybe the large amount of them has something to do with DisplayPort supporting multiple displays via a single connector, but that is just a wild guess on my part...
But do we want some sort of a solution for the possible conflicts between non-HDMI devices in the 10+ physical PCM device numbering range?
Yes, we definitely need a proper scheme for assigning more devices. I have no particular idea yet, so let me know if you have something already in your mind.
Takashi
src/conf/cards/HDA-Intel.conf | 92 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 92 insertions(+)
diff --git a/src/conf/cards/HDA-Intel.conf b/src/conf/cards/HDA-Intel.conf index 3957c12..7976b6c 100644 --- a/src/conf/cards/HDA-Intel.conf +++ b/src/conf/cards/HDA-Intel.conf @@ -305,6 +305,98 @@ HDA-Intel.pcm.hdmi.3 { } }
+HDA-Intel.pcm.hdmi.4 {
- @args [ CARD AES0 AES1 AES2 AES3 ]
- @args.CARD { type string }
- @args.AES0 { type integer }
- @args.AES1 { type integer }
- @args.AES2 { type integer }
- @args.AES3 { type integer }
- @func refer
- name {
@func concatstrings ["cards.HDA-Intel.pcm.hdmi.common:""CARD=" $CARD ",""DEVICE=10,""CTLINDEX=4,""AES0=" $AES0 ",""AES1=" $AES1 ",""AES2=" $AES2 ",""AES3=" $AES3]- }
+}
+HDA-Intel.pcm.hdmi.5 {
- @args [ CARD AES0 AES1 AES2 AES3 ]
- @args.CARD { type string }
- @args.AES0 { type integer }
- @args.AES1 { type integer }
- @args.AES2 { type integer }
- @args.AES3 { type integer }
- @func refer
- name {
@func concatstrings ["cards.HDA-Intel.pcm.hdmi.common:""CARD=" $CARD ",""DEVICE=11,""CTLINDEX=5,""AES0=" $AES0 ",""AES1=" $AES1 ",""AES2=" $AES2 ",""AES3=" $AES3]- }
+}
+HDA-Intel.pcm.hdmi.6 {
- @args [ CARD AES0 AES1 AES2 AES3 ]
- @args.CARD { type string }
- @args.AES0 { type integer }
- @args.AES1 { type integer }
- @args.AES2 { type integer }
- @args.AES3 { type integer }
- @func refer
- name {
@func concatstrings ["cards.HDA-Intel.pcm.hdmi.common:""CARD=" $CARD ",""DEVICE=12,""CTLINDEX=6,""AES0=" $AES0 ",""AES1=" $AES1 ",""AES2=" $AES2 ",""AES3=" $AES3]- }
+}
+HDA-Intel.pcm.hdmi.7 {
- @args [ CARD AES0 AES1 AES2 AES3 ]
- @args.CARD { type string }
- @args.AES0 { type integer }
- @args.AES1 { type integer }
- @args.AES2 { type integer }
- @args.AES3 { type integer }
- @func refer
- name {
@func concatstrings ["cards.HDA-Intel.pcm.hdmi.common:""CARD=" $CARD ",""DEVICE=13,""CTLINDEX=7,""AES0=" $AES0 ",""AES1=" $AES1 ",""AES2=" $AES2 ",""AES3=" $AES3]- }
+}
HDA-Intel.pcm.modem.0 {
1.8.1.5
participants (2)
-
Anssi Hannula -
Takashi Iwai