* [PATCH 00/11] Intel HDMI fixes
@ 2009-11-15 11:32 Wu Fengguang
2009-11-15 11:32 ` [PATCH 01/11] intelhdmi - fix audio infoframe fill size Wu Fengguang
` (12 more replies)
0 siblings, 13 replies; 15+ messages in thread
From: Wu Fengguang @ 2009-11-15 11:32 UTC (permalink / raw)
To: Takashi Iwai; +Cc: alsa-devel, Wu Fengguang, Shane W, David Härdeman
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 422 bytes --]
Hi Takashi,
Here is some Intel HDMI fixes, most of them are inspired/written by
David Härdeman (Big thanks!).
- fix audio infoframe size
- sticky infoframe/stream id/channel count to
- not disturb (David's) fragile devices
- avoid the first 0.5s audio samples be losted
- avoid losing presence info during
- module reloading (by querying the pin sense status)
- power off (by disabling it)
Thanks,
Fengguang
[-- Attachment #2: Type: text/plain, Size: 160 bytes --]
_______________________________________________
Alsa-devel mailing list
Alsa-devel@alsa-project.org
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
^ permalink raw reply [flat|nested] 15+ messages in thread* [PATCH 01/11] intelhdmi - fix audio infoframe fill size 2009-11-15 11:32 [PATCH 00/11] Intel HDMI fixes Wu Fengguang @ 2009-11-15 11:32 ` Wu Fengguang 2009-11-15 11:32 ` [PATCH 02/11] intelhdmi - fix channel mapping slot mask Wu Fengguang ` (11 subsequent siblings) 12 siblings, 0 replies; 15+ messages in thread From: Wu Fengguang @ 2009-11-15 11:32 UTC (permalink / raw) To: Takashi Iwai; +Cc: alsa-devel, Wu Fengguang, Shane W, David Härdeman [-- Warning: decoded text below may be mangled, UTF-8 assumed --] [-- Attachment #1: hdmi-fix-sizeof-ai.patch --] [-- Type: text/plain, Size: 829 bytes --] Reported-by: David Härdeman <david@hardeman.nu> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com> --- sound/pci/hda/patch_intelhdmi.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) --- sound-2.6.orig/sound/pci/hda/patch_intelhdmi.c 2009-11-02 15:14:30.000000000 +0800 +++ sound-2.6/sound/pci/hda/patch_intelhdmi.c 2009-11-02 15:14:39.000000000 +0800 @@ -509,12 +509,12 @@ static void hdmi_fill_audio_infoframe(st hdmi_debug_dip_size(codec, pin_nid); hdmi_clear_dip_buffers(codec, pin_nid); /* be paranoid */ - for (i = 0; i < sizeof(ai); i++) + for (i = 0; i < sizeof(*ai); i++) sum += params[i]; ai->checksum = - sum; hdmi_set_dip_index(codec, pin_nid, 0x0, 0x0); - for (i = 0; i < sizeof(ai); i++) + for (i = 0; i < sizeof(*ai); i++) hdmi_write_dip_byte(codec, pin_nid, params[i]); } [-- Attachment #2: Type: text/plain, Size: 160 bytes --] _______________________________________________ Alsa-devel mailing list Alsa-devel@alsa-project.org http://mailman.alsa-project.org/mailman/listinfo/alsa-devel ^ permalink raw reply [flat|nested] 15+ messages in thread
* [PATCH 02/11] intelhdmi - fix channel mapping slot mask 2009-11-15 11:32 [PATCH 00/11] Intel HDMI fixes Wu Fengguang 2009-11-15 11:32 ` [PATCH 01/11] intelhdmi - fix audio infoframe fill size Wu Fengguang @ 2009-11-15 11:32 ` Wu Fengguang 2009-11-15 11:32 ` [PATCH 03/11] intelhdmi - export monitor-presence and ELD-valid status Wu Fengguang ` (10 subsequent siblings) 12 siblings, 0 replies; 15+ messages in thread From: Wu Fengguang @ 2009-11-15 11:32 UTC (permalink / raw) To: Takashi Iwai; +Cc: alsa-devel, Wu Fengguang, Shane W, David Härdeman [-- Attachment #1: hdmi-channel-mapping-fix.patch --] [-- Type: text/plain, Size: 596 bytes --] Signed-off-by: Wu Fengguang <fengguang.wu@intel.com> --- sound/pci/hda/patch_intelhdmi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- sound-2.6.orig/sound/pci/hda/patch_intelhdmi.c 2009-11-02 15:14:39.000000000 +0800 +++ sound-2.6/sound/pci/hda/patch_intelhdmi.c 2009-11-02 15:14:46.000000000 +0800 @@ -433,7 +433,7 @@ static void hdmi_debug_channel_mapping(s slot = snd_hda_codec_read(codec, nid, 0, AC_VERB_GET_HDMI_CHAN_SLOT, i); printk(KERN_DEBUG "HDMI: ASP channel %d => slot %d\n", - slot >> 4, slot & 0x7); + slot >> 4, slot & 0xf); } #endif } ^ permalink raw reply [flat|nested] 15+ messages in thread
* [PATCH 03/11] intelhdmi - export monitor-presence and ELD-valid status 2009-11-15 11:32 [PATCH 00/11] Intel HDMI fixes Wu Fengguang 2009-11-15 11:32 ` [PATCH 01/11] intelhdmi - fix audio infoframe fill size Wu Fengguang 2009-11-15 11:32 ` [PATCH 02/11] intelhdmi - fix channel mapping slot mask Wu Fengguang @ 2009-11-15 11:32 ` Wu Fengguang 2009-11-15 11:32 ` [PATCH 04/11] hda - introduce snd_hda_pin_sense() Wu Fengguang ` (9 subsequent siblings) 12 siblings, 0 replies; 15+ messages in thread From: Wu Fengguang @ 2009-11-15 11:32 UTC (permalink / raw) To: Takashi Iwai; +Cc: alsa-devel, Wu Fengguang, Shane W, David Härdeman [-- Attachment #1: hdmi-export-presence-eld.patch --] [-- Type: text/plain, Size: 2679 bytes --] Signed-off-by: Wu Fengguang <fengguang.wu@intel.com> --- sound/pci/hda/hda_eld.c | 8 +++++++- sound/pci/hda/hda_local.h | 4 +++- sound/pci/hda/patch_intelhdmi.c | 8 +++----- 3 files changed, 13 insertions(+), 7 deletions(-) --- sound-2.6.orig/sound/pci/hda/patch_intelhdmi.c 2009-11-02 15:17:26.000000000 +0800 +++ sound-2.6/sound/pci/hda/patch_intelhdmi.c 2009-11-02 15:18:47.000000000 +0800 @@ -62,8 +62,6 @@ struct intel_hdmi_spec { /* * HDMI sink attached to each pin */ - bool sink_present[INTEL_HDMI_PINS]; - bool sink_eldv[INTEL_HDMI_PINS]; struct hdmi_eld sink_eld[INTEL_HDMI_PINS]; /* @@ -645,7 +643,7 @@ static void hdmi_setup_audio_infoframe(s for (i = 0; i < spec->num_pins; i++) { if (spec->pin_cvt[i] != nid) continue; - if (spec->sink_present[i] != true) + if (!spec->sink_eld[i].monitor_present) continue; pin_nid = spec->pin[i]; @@ -675,8 +673,8 @@ static void hdmi_intrinsic_event(struct if (index < 0) return; - spec->sink_present[index] = pind; - spec->sink_eldv[index] = eldv; + spec->sink_eld[index].monitor_present = pind; + spec->sink_eld[index].eld_valid = eldv; if (pind && eldv) { hdmi_parse_eld(codec, index); --- sound-2.6.orig/sound/pci/hda/hda_eld.c 2009-11-02 15:14:30.000000000 +0800 +++ sound-2.6/sound/pci/hda/hda_eld.c 2009-11-02 15:24:04.000000000 +0800 @@ -477,6 +477,8 @@ static void hdmi_print_eld_info(struct s [4 ... 7] = "reserved" }; + snd_iprintf(buffer, "monitor_present\t\t%d\n", e->monitor_present); + snd_iprintf(buffer, "eld_valid\t\t%d\n", e->eld_valid); snd_iprintf(buffer, "monitor_name\t\t%s\n", e->monitor_name); snd_iprintf(buffer, "connection_type\t\t%s\n", eld_connection_type_names[e->conn_type]); @@ -518,7 +520,11 @@ static void hdmi_write_eld_info(struct s * monitor_name manufacture_id product_id * eld_version edid_version */ - if (!strcmp(name, "connection_type")) + if (!strcmp(name, "monitor_present")) + e->monitor_present = val; + else if (!strcmp(name, "eld_valid")) + e->eld_valid = val; + else if (!strcmp(name, "connection_type")) e->conn_type = val; else if (!strcmp(name, "port_id")) e->port_id = val; --- sound-2.6.orig/sound/pci/hda/hda_local.h 2009-11-02 15:14:30.000000000 +0800 +++ sound-2.6/sound/pci/hda/hda_local.h 2009-11-02 15:21:11.000000000 +0800 @@ -516,9 +516,11 @@ struct cea_sad { * ELD: EDID Like Data */ struct hdmi_eld { + bool monitor_present; + bool eld_valid; int eld_size; int baseline_len; - int eld_ver; /* (eld_ver == 0) indicates invalid ELD */ + int eld_ver; int cea_edid_ver; char monitor_name[ELD_MAX_MNL + 1]; int manufacture_id; ^ permalink raw reply [flat|nested] 15+ messages in thread
* [PATCH 04/11] hda - introduce snd_hda_pin_sense() 2009-11-15 11:32 [PATCH 00/11] Intel HDMI fixes Wu Fengguang ` (2 preceding siblings ...) 2009-11-15 11:32 ` [PATCH 03/11] intelhdmi - export monitor-presence and ELD-valid status Wu Fengguang @ 2009-11-15 11:32 ` Wu Fengguang 2009-11-15 11:32 ` [PATCH 05/11] intelhdmi - probe for monitor/eld presence at module init time Wu Fengguang ` (8 subsequent siblings) 12 siblings, 0 replies; 15+ messages in thread From: Wu Fengguang @ 2009-11-15 11:32 UTC (permalink / raw) To: Takashi Iwai; +Cc: alsa-devel, Wu Fengguang, Shane W, David Härdeman [-- Attachment #1: hda-pin-sense.patch --] [-- Type: text/plain, Size: 17065 bytes --] This helps merge duplicate code. Signed-off-by: Wu Fengguang <fengguang.wu@intel.com> --- sound/pci/hda/hda_codec.c | 12 ++ sound/pci/hda/hda_eld.c | 7 - sound/pci/hda/hda_local.h | 1 sound/pci/hda/patch_cirrus.c | 14 --- sound/pci/hda/patch_realtek.c | 146 ++++++++------------------------ 5 files changed, 58 insertions(+), 122 deletions(-) --- sound-2.6.orig/sound/pci/hda/hda_codec.c 2009-11-02 15:14:30.000000000 +0800 +++ sound-2.6/sound/pci/hda/hda_codec.c 2009-11-02 15:25:17.000000000 +0800 @@ -1229,6 +1229,18 @@ u32 snd_hda_query_pin_caps(struct hda_co } EXPORT_SYMBOL_HDA(snd_hda_query_pin_caps); +u32 snd_hda_pin_sense(struct hda_codec *codec, hda_nid_t nid) +{ + u32 pincap = snd_hda_query_pin_caps(codec, nid); + + if (pincap & AC_PINCAP_TRIG_REQ) /* need trigger? */ + snd_hda_codec_read(codec, nid, 0, AC_VERB_SET_PIN_SENSE, 0); + + return snd_hda_codec_read(codec, nid, 0, + AC_VERB_GET_PIN_SENSE, 0); +} +EXPORT_SYMBOL_HDA(snd_hda_pin_sense); + /* * read the current volume to info * if the cache exists, read the cache value. --- sound-2.6.orig/sound/pci/hda/hda_local.h 2009-11-02 15:21:11.000000000 +0800 +++ sound-2.6/sound/pci/hda/hda_local.h 2009-11-02 15:25:17.000000000 +0800 @@ -424,6 +424,7 @@ u32 query_amp_caps(struct hda_codec *cod int snd_hda_override_amp_caps(struct hda_codec *codec, hda_nid_t nid, int dir, unsigned int caps); u32 snd_hda_query_pin_caps(struct hda_codec *codec, hda_nid_t nid); +u32 snd_hda_pin_sense(struct hda_codec *codec, hda_nid_t nid); int snd_hda_ctl_add(struct hda_codec *codec, struct snd_kcontrol *kctl); void snd_hda_ctls_clear(struct hda_codec *codec); --- sound-2.6.orig/sound/pci/hda/patch_realtek.c 2009-11-02 15:12:14.000000000 +0800 +++ sound-2.6/sound/pci/hda/patch_realtek.c 2009-11-02 15:25:17.000000000 +0800 @@ -961,16 +961,12 @@ static void alc_fix_pll_init(struct hda_ static void alc_automute_pin(struct hda_codec *codec) { struct alc_spec *spec = codec->spec; - unsigned int present, pincap; + unsigned int present; unsigned int nid = spec->autocfg.hp_pins[0]; int i; - pincap = snd_hda_query_pin_caps(codec, nid); - if (pincap & AC_PINCAP_TRIG_REQ) /* need trigger? */ - snd_hda_codec_read(codec, nid, 0, AC_VERB_SET_PIN_SENSE, 0); - present = snd_hda_codec_read(codec, nid, 0, - AC_VERB_GET_PIN_SENSE, 0); - spec->jack_present = (present & AC_PINSENSE_PRESENCE) != 0; + present = snd_hda_pin_sense(codec, nid); + spec->jack_present = !!(present & AC_PINSENSE_PRESENCE); for (i = 0; i < ARRAY_SIZE(spec->autocfg.speaker_pins); i++) { nid = spec->autocfg.speaker_pins[i]; if (!nid) @@ -1511,7 +1507,7 @@ static struct hda_verb alc888_fujitsu_xa static void alc_automute_amp(struct hda_codec *codec) { struct alc_spec *spec = codec->spec; - unsigned int val, mute, pincap; + unsigned int val, mute; hda_nid_t nid; int i; @@ -1520,12 +1516,7 @@ static void alc_automute_amp(struct hda_ nid = spec->autocfg.hp_pins[i]; if (!nid) break; - pincap = snd_hda_query_pin_caps(codec, nid); - if (pincap & AC_PINCAP_TRIG_REQ) /* need trigger? */ - snd_hda_codec_read(codec, nid, 0, - AC_VERB_SET_PIN_SENSE, 0); - val = snd_hda_codec_read(codec, nid, 0, - AC_VERB_GET_PIN_SENSE, 0); + val = snd_hda_pin_sense(codec, nid); if (val & AC_PINSENSE_PRESENCE) { spec->jack_present = 1; break; @@ -2777,8 +2768,7 @@ static void alc880_uniwill_mic_automute( unsigned int present; unsigned char bits; - present = snd_hda_codec_read(codec, 0x18, 0, - AC_VERB_GET_PIN_SENSE, 0) & 0x80000000; + present = snd_hda_pin_sense(codec, 0x18) & AC_PINSENSE_PRESENCE; bits = present ? HDA_AMP_MUTE : 0; snd_hda_codec_amp_stereo(codec, 0x0b, HDA_INPUT, 1, HDA_AMP_MUTE, bits); } @@ -5091,9 +5081,8 @@ static void alc260_hp_automute(struct hd struct alc_spec *spec = codec->spec; unsigned int present; - present = snd_hda_codec_read(codec, 0x10, 0, - AC_VERB_GET_PIN_SENSE, 0); - spec->jack_present = (present & AC_PINSENSE_PRESENCE) != 0; + present = snd_hda_pin_sense(codec, 0x10); + spec->jack_present = !!(present & AC_PINSENSE_PRESENCE); alc260_hp_master_update(codec, 0x0f, 0x10, 0x11); } @@ -5160,9 +5149,8 @@ static void alc260_hp_3013_automute(stru struct alc_spec *spec = codec->spec; unsigned int present; - present = snd_hda_codec_read(codec, 0x15, 0, - AC_VERB_GET_PIN_SENSE, 0); - spec->jack_present = (present & AC_PINSENSE_PRESENCE) != 0; + present = snd_hda_pin_sense(codec, 0x15); + spec->jack_present = !!(present & AC_PINSENSE_PRESENCE); alc260_hp_master_update(codec, 0x15, 0x10, 0x11); } @@ -5177,8 +5165,7 @@ static void alc260_hp_3012_automute(stru { unsigned int present, bits; - present = snd_hda_codec_read(codec, 0x10, 0, - AC_VERB_GET_PIN_SENSE, 0) & AC_PINSENSE_PRESENCE; + present = snd_hda_pin_sense(codec, 0x10) & AC_PINSENSE_PRESENCE; bits = present ? 0 : PIN_OUT; snd_hda_codec_write(codec, 0x0f, 0, AC_VERB_SET_PIN_WIDGET_CONTROL, @@ -5750,8 +5737,7 @@ static void alc260_replacer_672v_automut unsigned int present; /* speaker --> GPIO Data 0, hp or spdif --> GPIO data 1 */ - present = snd_hda_codec_read(codec, 0x0f, 0, - AC_VERB_GET_PIN_SENSE, 0) & 0x80000000; + present = snd_hda_pin_sense(codec, 0x0f) & AC_PINSENSE_PRESENCE; if (present) { snd_hda_codec_write_cache(codec, 0x01, 0, AC_VERB_SET_GPIO_DATA, 1); @@ -8522,8 +8508,7 @@ static void alc883_lenovo_101e_ispeaker_ unsigned int present; unsigned char bits; - present = snd_hda_codec_read(codec, 0x14, 0, AC_VERB_GET_PIN_SENSE, 0) - & AC_PINSENSE_PRESENCE; + present = snd_hda_pin_sense(codec, 0x14) & AC_PINSENSE_PRESENCE; bits = present ? HDA_AMP_MUTE : 0; snd_hda_codec_amp_stereo(codec, 0x15, HDA_OUTPUT, 0, HDA_AMP_MUTE, bits); @@ -8534,8 +8519,7 @@ static void alc883_lenovo_101e_all_autom unsigned int present; unsigned char bits; - present = snd_hda_codec_read(codec, 0x1b, 0, - AC_VERB_GET_PIN_SENSE, 0) & 0x80000000; + present = snd_hda_pin_sense(codec, 0x1b) & AC_PINSENSE_PRESENCE; bits = present ? HDA_AMP_MUTE : 0; snd_hda_codec_amp_stereo(codec, 0x15, HDA_OUTPUT, 0, HDA_AMP_MUTE, bits); @@ -8687,8 +8671,7 @@ static void alc889A_mb31_automute(struct /* Mute only in 2ch or 4ch mode */ if (snd_hda_codec_read(codec, 0x15, 0, AC_VERB_GET_CONNECT_SEL, 0) == 0x00) { - present = snd_hda_codec_read(codec, 0x15, 0, - AC_VERB_GET_PIN_SENSE, 0) & AC_PINSENSE_PRESENCE; + present = snd_hda_pin_sense(codec, 0x15) & AC_PINSENSE_PRESENCE; snd_hda_codec_amp_stereo(codec, 0x14, HDA_OUTPUT, 0, HDA_AMP_MUTE, present ? HDA_AMP_MUTE : 0); snd_hda_codec_amp_stereo(codec, 0x16, HDA_OUTPUT, 0, @@ -10030,9 +10013,7 @@ static void alc262_hp_master_update(stru static void alc262_hp_bpc_automute(struct hda_codec *codec) { struct alc_spec *spec = codec->spec; - unsigned int presence; - presence = snd_hda_codec_read(codec, 0x1b, 0, - AC_VERB_GET_PIN_SENSE, 0); + unsigned int presence = snd_hda_pin_sense(codec, 0x1b); spec->jack_present = !!(presence & AC_PINSENSE_PRESENCE); alc262_hp_master_update(codec); } @@ -10047,9 +10028,7 @@ static void alc262_hp_bpc_unsol_event(st static void alc262_hp_wildwest_automute(struct hda_codec *codec) { struct alc_spec *spec = codec->spec; - unsigned int presence; - presence = snd_hda_codec_read(codec, 0x15, 0, - AC_VERB_GET_PIN_SENSE, 0); + unsigned int presence = snd_hda_pin_sense(codec, 0x15); spec->jack_present = !!(presence & AC_PINSENSE_PRESENCE); alc262_hp_master_update(codec); } @@ -10286,11 +10265,8 @@ static void alc262_hippo_automute(struct hda_nid_t hp_nid = spec->autocfg.hp_pins[0]; unsigned int present; - /* need to execute and sync at first */ - snd_hda_codec_read(codec, hp_nid, 0, AC_VERB_SET_PIN_SENSE, 0); - present = snd_hda_codec_read(codec, hp_nid, 0, - AC_VERB_GET_PIN_SENSE, 0); - spec->jack_present = (present & 0x80000000) != 0; + present = snd_hda_pin_sense(codec, hp_nid) & AC_PINSENSE_PRESENCE; + spec->jack_present = !!present; alc262_hippo_master_update(codec); } @@ -10617,16 +10593,8 @@ static void alc262_fujitsu_automute(stru if (force || !spec->sense_updated) { unsigned int present; - /* need to execute and sync at first */ - snd_hda_codec_read(codec, 0x14, 0, AC_VERB_SET_PIN_SENSE, 0); - /* check laptop HP jack */ - present = snd_hda_codec_read(codec, 0x14, 0, - AC_VERB_GET_PIN_SENSE, 0); - /* need to execute and sync at first */ - snd_hda_codec_read(codec, 0x1b, 0, AC_VERB_SET_PIN_SENSE, 0); - /* check docking HP jack */ - present |= snd_hda_codec_read(codec, 0x1b, 0, - AC_VERB_GET_PIN_SENSE, 0); + present = snd_hda_pin_sense(codec, 0x14); + present |= snd_hda_pin_sense(codec, 0x1b); if (present & AC_PINSENSE_PRESENCE) spec->jack_present = 1; else @@ -10675,12 +10643,8 @@ static void alc262_lenovo_3000_automute( unsigned int mute; if (force || !spec->sense_updated) { - unsigned int present_int_hp; - /* need to execute and sync at first */ - snd_hda_codec_read(codec, 0x1b, 0, AC_VERB_SET_PIN_SENSE, 0); - present_int_hp = snd_hda_codec_read(codec, 0x1b, 0, - AC_VERB_GET_PIN_SENSE, 0); - spec->jack_present = (present_int_hp & 0x80000000) != 0; + unsigned int present_int_hp = snd_hda_pin_sense(codec, 0x1b); + spec->jack_present = !!(present_int_hp & AC_PINSENSE_PRESENCE); spec->sense_updated = 1; } if (spec->jack_present) { @@ -10872,12 +10836,9 @@ static void alc262_ultra_automute(struct mute = 0; /* auto-mute only when HP is used as HP */ if (!spec->cur_mux[0]) { - unsigned int present; - /* need to execute and sync at first */ - snd_hda_codec_read(codec, 0x15, 0, AC_VERB_SET_PIN_SENSE, 0); - present = snd_hda_codec_read(codec, 0x15, 0, - AC_VERB_GET_PIN_SENSE, 0); - spec->jack_present = (present & AC_PINSENSE_PRESENCE) != 0; + unsigned int present = snd_hda_pin_sense(codec, 0x15); + + spec->jack_present = !!(present & AC_PINSENSE_PRESENCE); if (spec->jack_present) mute = HDA_AMP_MUTE; } @@ -11918,7 +11879,7 @@ static void alc268_acer_automute(struct unsigned int present; present = snd_hda_codec_read(codec, 0x14, 0, AC_VERB_GET_PIN_SENSE, 0); - spec->jack_present = (present & 0x80000000) != 0; + spec->jack_present = !!(present & AC_PINSENSE_PRESENCE); spec->sense_updated = 1; } if (spec->jack_present) @@ -13046,12 +13007,10 @@ static void alc269_lifebook_speaker_auto unsigned char bits; /* Check laptop headphone socket */ - present = snd_hda_codec_read(codec, 0x15, 0, - AC_VERB_GET_PIN_SENSE, 0) & 0x80000000; + present = snd_hda_pin_sense(codec, 0x15) & AC_PINSENSE_PRESENCE; /* Check port replicator headphone socket */ - present |= snd_hda_codec_read(codec, 0x1a, 0, - AC_VERB_GET_PIN_SENSE, 0) & 0x80000000; + present = snd_hda_pin_sense(codec, 0x1a) & AC_PINSENSE_PRESENCE; bits = present ? AMP_IN_MUTE(0) : 0; snd_hda_codec_amp_stereo(codec, 0x0c, HDA_INPUT, 0, @@ -13075,11 +13034,8 @@ static void alc269_lifebook_mic_autoswit unsigned int present_laptop; unsigned int present_dock; - present_laptop = snd_hda_codec_read(codec, 0x18, 0, - AC_VERB_GET_PIN_SENSE, 0) & 0x80000000; - - present_dock = snd_hda_codec_read(codec, 0x1b, 0, - AC_VERB_GET_PIN_SENSE, 0) & 0x80000000; + present_laptop = snd_hda_pin_sense(codec, 0x18) & AC_PINSENSE_PRESENCE; + present_dock = snd_hda_pin_sense(codec, 0x1b) & AC_PINSENSE_PRESENCE; /* Laptop mic port overrides dock mic port, design decision */ if (present_dock) @@ -16437,9 +16393,7 @@ static void alc663_m51va_speaker_automut unsigned int present; unsigned char bits; - present = snd_hda_codec_read(codec, 0x21, 0, - AC_VERB_GET_PIN_SENSE, 0) - & AC_PINSENSE_PRESENCE; + present = snd_hda_pin_sense(codec, 0x21) & AC_PINSENSE_PRESENCE; bits = present ? HDA_AMP_MUTE : 0; snd_hda_codec_amp_stereo(codec, 0x0c, HDA_INPUT, 0, AMP_IN_MUTE(0), bits); @@ -16452,9 +16406,7 @@ static void alc663_21jd_two_speaker_auto unsigned int present; unsigned char bits; - present = snd_hda_codec_read(codec, 0x21, 0, - AC_VERB_GET_PIN_SENSE, 0) - & AC_PINSENSE_PRESENCE; + present = snd_hda_pin_sense(codec, 0x21) & AC_PINSENSE_PRESENCE; bits = present ? HDA_AMP_MUTE : 0; snd_hda_codec_amp_stereo(codec, 0x0c, HDA_INPUT, 0, AMP_IN_MUTE(0), bits); @@ -16471,9 +16423,7 @@ static void alc663_15jd_two_speaker_auto unsigned int present; unsigned char bits; - present = snd_hda_codec_read(codec, 0x15, 0, - AC_VERB_GET_PIN_SENSE, 0) - & AC_PINSENSE_PRESENCE; + present = snd_hda_pin_sense(codec, 0x15) & AC_PINSENSE_PRESENCE; bits = present ? HDA_AMP_MUTE : 0; snd_hda_codec_amp_stereo(codec, 0x0c, HDA_INPUT, 0, AMP_IN_MUTE(0), bits); @@ -16490,9 +16440,7 @@ static void alc662_f5z_speaker_automute( unsigned int present; unsigned char bits; - present = snd_hda_codec_read(codec, 0x1b, 0, - AC_VERB_GET_PIN_SENSE, 0) - & AC_PINSENSE_PRESENCE; + present = snd_hda_pin_sense(codec, 0x1b) & AC_PINSENSE_PRESENCE; bits = present ? 0 : PIN_OUT; snd_hda_codec_write(codec, 0x14, 0, AC_VERB_SET_PIN_WIDGET_CONTROL, bits); @@ -16502,12 +16450,8 @@ static void alc663_two_hp_m1_speaker_aut { unsigned int present1, present2; - present1 = snd_hda_codec_read(codec, 0x21, 0, - AC_VERB_GET_PIN_SENSE, 0) - & AC_PINSENSE_PRESENCE; - present2 = snd_hda_codec_read(codec, 0x15, 0, - AC_VERB_GET_PIN_SENSE, 0) - & AC_PINSENSE_PRESENCE; + present1 = snd_hda_pin_sense(codec, 0x21) & AC_PINSENSE_PRESENCE; + present2 = snd_hda_pin_sense(codec, 0x15) & AC_PINSENSE_PRESENCE; if (present1 || present2) { snd_hda_codec_write_cache(codec, 0x14, 0, @@ -16522,12 +16466,8 @@ static void alc663_two_hp_m2_speaker_aut { unsigned int present1, present2; - present1 = snd_hda_codec_read(codec, 0x1b, 0, - AC_VERB_GET_PIN_SENSE, 0) - & AC_PINSENSE_PRESENCE; - present2 = snd_hda_codec_read(codec, 0x15, 0, - AC_VERB_GET_PIN_SENSE, 0) - & AC_PINSENSE_PRESENCE; + present1 = snd_hda_pin_sense(codec, 0x1b) & AC_PINSENSE_PRESENCE; + present2 = snd_hda_pin_sense(codec, 0x15) & AC_PINSENSE_PRESENCE; if (present1 || present2) { snd_hda_codec_amp_stereo(codec, 0x0c, HDA_INPUT, 0, @@ -16687,9 +16627,7 @@ static void alc663_g71v_hp_automute(stru unsigned int present; unsigned char bits; - present = snd_hda_codec_read(codec, 0x21, 0, - AC_VERB_GET_PIN_SENSE, 0) - & AC_PINSENSE_PRESENCE; + present = snd_hda_pin_sense(codec, 0x21) & AC_PINSENSE_PRESENCE; bits = present ? HDA_AMP_MUTE : 0; snd_hda_codec_amp_stereo(codec, 0x15, HDA_OUTPUT, 0, HDA_AMP_MUTE, bits); @@ -16702,9 +16640,7 @@ static void alc663_g71v_front_automute(s unsigned int present; unsigned char bits; - present = snd_hda_codec_read(codec, 0x15, 0, - AC_VERB_GET_PIN_SENSE, 0) - & AC_PINSENSE_PRESENCE; + present = snd_hda_pin_sense(codec, 0x15) & AC_PINSENSE_PRESENCE; bits = present ? HDA_AMP_MUTE : 0; snd_hda_codec_amp_stereo(codec, 0x14, HDA_OUTPUT, 0, HDA_AMP_MUTE, bits); --- sound-2.6.orig/sound/pci/hda/patch_cirrus.c 2009-11-02 15:12:14.000000000 +0800 +++ sound-2.6/sound/pci/hda/patch_cirrus.c 2009-11-02 15:25:17.000000000 +0800 @@ -817,11 +817,7 @@ static void cs_automute(struct hda_codec caps = snd_hda_query_pin_caps(codec, nid); if (!(caps & AC_PINCAP_PRES_DETECT)) continue; - if (caps & AC_PINCAP_TRIG_REQ) - snd_hda_codec_read(codec, nid, 0, - AC_VERB_SET_PIN_SENSE, 0); - present = snd_hda_codec_read(codec, nid, 0, - AC_VERB_GET_PIN_SENSE, 0); + present = snd_hda_pin_sense(codec, nid); hp_present |= (present & AC_PINSENSE_PRESENCE) != 0; if (hp_present) break; @@ -844,14 +840,10 @@ static void cs_automic(struct hda_codec struct cs_spec *spec = codec->spec; struct auto_pin_cfg *cfg = &spec->autocfg; hda_nid_t nid; - unsigned int caps, present; + unsigned int present; nid = cfg->input_pins[spec->automic_idx]; - caps = snd_hda_query_pin_caps(codec, nid); - if (caps & AC_PINCAP_TRIG_REQ) - snd_hda_codec_read(codec, nid, 0, AC_VERB_SET_PIN_SENSE, 0); - present = snd_hda_codec_read(codec, nid, 0, - AC_VERB_GET_PIN_SENSE, 0); + present = snd_hda_pin_sense(codec, nid); if (present & AC_PINSENSE_PRESENCE) change_cur_input(codec, spec->automic_idx, 0); else { --- sound-2.6.orig/sound/pci/hda/hda_eld.c 2009-11-02 15:43:35.000000000 +0800 +++ sound-2.6/sound/pci/hda/hda_eld.c 2009-11-02 15:43:58.000000000 +0800 @@ -309,17 +309,12 @@ out_fail: return -EINVAL; } -static int hdmi_present_sense(struct hda_codec *codec, hda_nid_t nid) -{ - return snd_hda_codec_read(codec, nid, 0, AC_VERB_GET_PIN_SENSE, 0); -} - static int hdmi_eld_valid(struct hda_codec *codec, hda_nid_t nid) { int eldv; int present; - present = hdmi_present_sense(codec, nid); + present = snd_hda_pin_sense(codec, nid); eldv = (present & AC_PINSENSE_ELDV); present = (present & AC_PINSENSE_PRESENCE); ^ permalink raw reply [flat|nested] 15+ messages in thread
* [PATCH 05/11] intelhdmi - probe for monitor/eld presence at module init time 2009-11-15 11:32 [PATCH 00/11] Intel HDMI fixes Wu Fengguang ` (3 preceding siblings ...) 2009-11-15 11:32 ` [PATCH 04/11] hda - introduce snd_hda_pin_sense() Wu Fengguang @ 2009-11-15 11:32 ` Wu Fengguang 2009-11-15 11:32 ` [PATCH 06/11] intelhdmi - separate out infoframe checksum routine Wu Fengguang ` (7 subsequent siblings) 12 siblings, 0 replies; 15+ messages in thread From: Wu Fengguang @ 2009-11-15 11:32 UTC (permalink / raw) To: Takashi Iwai; +Cc: alsa-devel, Wu Fengguang, Shane W, David Härdeman [-- Warning: decoded text below may be mangled, UTF-8 assumed --] [-- Attachment #1: hdmi-pin-sense.patch --] [-- Type: text/plain, Size: 2159 bytes --] This avoids lost of presence info on module reloading. The presence info used to be only updated at the (rare) hotplug events. Proposed by David, thanks! CC: David Härdeman <david@hardeman.nu> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com> --- sound/pci/hda/patch_intelhdmi.c | 33 ++++++++++++++++++++---------- 1 file changed, 23 insertions(+), 10 deletions(-) --- sound-2.6.orig/sound/pci/hda/patch_intelhdmi.c 2009-11-02 15:44:26.000000000 +0800 +++ sound-2.6/sound/pci/hda/patch_intelhdmi.c 2009-11-02 15:47:43.000000000 +0800 @@ -259,6 +259,25 @@ static int intel_hdmi_read_pin_conn(stru return 0; } +static void hdmi_get_show_eld(struct hda_codec *codec, hda_nid_t pin_nid, + struct hdmi_eld *eld) +{ + if (!snd_hdmi_get_eld(eld, codec, pin_nid)) + snd_hdmi_show_eld(eld); +} + +static void hdmi_present_sense(struct hda_codec *codec, hda_nid_t pin_nid, + struct hdmi_eld *eld) +{ + int present = snd_hda_pin_sense(codec, pin_nid); + + eld->monitor_present = !!(present & AC_PINSENSE_PRESENCE); + eld->eld_valid = !!(present & AC_PINSENSE_ELDV); + + if (present & AC_PINSENSE_ELDV) + hdmi_get_show_eld(codec, pin_nid, eld); +} + static int intel_hdmi_add_pin(struct hda_codec *codec, hda_nid_t pin_nid) { struct intel_hdmi_spec *spec = codec->spec; @@ -269,6 +288,8 @@ static int intel_hdmi_add_pin(struct hda return -EINVAL; } + hdmi_present_sense(codec, pin_nid, &spec->sink_eld[spec->num_pins]); + spec->pin[spec->num_pins] = pin_nid; spec->num_pins++; @@ -436,15 +457,6 @@ static void hdmi_debug_channel_mapping(s #endif } -static void hdmi_parse_eld(struct hda_codec *codec, int index) -{ - struct intel_hdmi_spec *spec = codec->spec; - struct hdmi_eld *eld = &spec->sink_eld[index]; - - if (!snd_hdmi_get_eld(eld, codec, spec->pin[index])) - snd_hdmi_show_eld(eld); -} - /* * Audio InfoFrame routines @@ -677,7 +690,7 @@ static void hdmi_intrinsic_event(struct spec->sink_eld[index].eld_valid = eldv; if (pind && eldv) { - hdmi_parse_eld(codec, index); + hdmi_get_show_eld(codec, spec->pin[index], &spec->sink_eld[index]); /* TODO: do real things about ELD */ } } [-- Attachment #2: Type: text/plain, Size: 160 bytes --] _______________________________________________ Alsa-devel mailing list Alsa-devel@alsa-project.org http://mailman.alsa-project.org/mailman/listinfo/alsa-devel ^ permalink raw reply [flat|nested] 15+ messages in thread
* [PATCH 06/11] intelhdmi - separate out infoframe checksum routine 2009-11-15 11:32 [PATCH 00/11] Intel HDMI fixes Wu Fengguang ` (4 preceding siblings ...) 2009-11-15 11:32 ` [PATCH 05/11] intelhdmi - probe for monitor/eld presence at module init time Wu Fengguang @ 2009-11-15 11:32 ` Wu Fengguang 2009-11-15 11:32 ` [PATCH 07/11] intelhdmi - sticky infoframe Wu Fengguang ` (6 subsequent siblings) 12 siblings, 0 replies; 15+ messages in thread From: Wu Fengguang @ 2009-11-15 11:32 UTC (permalink / raw) To: Takashi Iwai; +Cc: alsa-devel, Wu Fengguang, Shane W, David Härdeman [-- Attachment #1: hdmi-checksum.patch --] [-- Type: text/plain, Size: 1340 bytes --] And make it right when called for more than one times. Signed-off-by: Wu Fengguang <fengguang.wu@intel.com> --- sound/pci/hda/patch_intelhdmi.c | 23 +++++++++++++++++------ 1 file changed, 17 insertions(+), 6 deletions(-) --- sound-2.6.orig/sound/pci/hda/patch_intelhdmi.c 2009-11-02 16:11:22.000000000 +0800 +++ sound-2.6/sound/pci/hda/patch_intelhdmi.c 2009-11-02 16:27:42.000000000 +0800 @@ -508,24 +508,35 @@ static void hdmi_clear_dip_buffers(struc #endif } +static void hdmi_checksum_audio_infoframe(struct hdmi_audio_infoframe *ai) +{ + u8 *bytes = (u8 *)ai; + u8 sum = 0; + int i; + + ai->checksum = 0; + + for (i = 0; i < sizeof(*ai); i++) + sum += bytes[i]; + + ai->checksum = - sum; +} + static void hdmi_fill_audio_infoframe(struct hda_codec *codec, hda_nid_t pin_nid, struct hdmi_audio_infoframe *ai) { - u8 *params = (u8 *)ai; - u8 sum = 0; + u8 *bytes = (u8 *)ai; int i; hdmi_debug_dip_size(codec, pin_nid); hdmi_clear_dip_buffers(codec, pin_nid); /* be paranoid */ - for (i = 0; i < sizeof(*ai); i++) - sum += params[i]; - ai->checksum = - sum; + hdmi_checksum_audio_infoframe(ai); hdmi_set_dip_index(codec, pin_nid, 0x0, 0x0); for (i = 0; i < sizeof(*ai); i++) - hdmi_write_dip_byte(codec, pin_nid, params[i]); + hdmi_write_dip_byte(codec, pin_nid, bytes[i]); } /* ^ permalink raw reply [flat|nested] 15+ messages in thread
* [PATCH 07/11] intelhdmi - sticky infoframe 2009-11-15 11:32 [PATCH 00/11] Intel HDMI fixes Wu Fengguang ` (5 preceding siblings ...) 2009-11-15 11:32 ` [PATCH 06/11] intelhdmi - separate out infoframe checksum routine Wu Fengguang @ 2009-11-15 11:32 ` Wu Fengguang 2009-11-15 11:32 ` [PATCH 08/11] intelhdmi - sticky stream id and format Wu Fengguang ` (5 subsequent siblings) 12 siblings, 0 replies; 15+ messages in thread From: Wu Fengguang @ 2009-11-15 11:32 UTC (permalink / raw) To: Takashi Iwai; +Cc: alsa-devel, Wu Fengguang, Shane W, David Härdeman [-- Warning: decoded text below may be mangled, UTF-8 assumed --] [-- Attachment #1: hdmi-sticky-infoframe.patch --] [-- Type: text/plain, Size: 2199 bytes --] Remember the active infoframe, so as to avoid stop/restart infoframe transmission when switching between audio clips of the same format. Proposed by Shang and David. CC: Shane W <shane-alsa@csy.ca> CC: David Härdeman <david@hardeman.nu> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com> --- sound/pci/hda/patch_intelhdmi.c | 38 ++++++++++++++++++++---------- 1 file changed, 26 insertions(+), 12 deletions(-) --- sound-2.6.orig/sound/pci/hda/patch_intelhdmi.c 2009-11-05 13:52:12.000000000 +0800 +++ sound-2.6/sound/pci/hda/patch_intelhdmi.c 2009-11-05 13:56:58.000000000 +0800 @@ -646,6 +646,27 @@ static void hdmi_setup_channel_mapping(s hdmi_debug_channel_mapping(codec, nid); } +static bool hdmi_infoframe_uptodate(struct hda_codec *codec, hda_nid_t pin_nid, + struct hdmi_audio_infoframe *ai) +{ + u8 *bytes = (u8 *)ai; + u8 val; + int i; + + if (snd_hda_codec_read(codec, pin_nid, 0, AC_VERB_GET_HDMI_DIP_XMIT, 0) + != AC_DIPXMIT_BEST) + return false; + + hdmi_set_dip_index(codec, pin_nid, 0x0, 0x0); + for (i = 0; i < sizeof(*ai); i++) { + val = snd_hda_codec_read(codec, pin_nid, 0, + AC_VERB_GET_HDMI_DIP_DATA, 0); + if (val != bytes[i]) + return false; + } + + return true; +} static void hdmi_setup_audio_infoframe(struct hda_codec *codec, hda_nid_t nid, struct snd_pcm_substream *substream) @@ -670,8 +691,11 @@ static void hdmi_setup_audio_infoframe(s continue; pin_nid = spec->pin[i]; - hdmi_fill_audio_infoframe(codec, pin_nid, &ai); - hdmi_start_infoframe_trans(codec, pin_nid); + if (!hdmi_infoframe_uptodate(codec, pin_nid, &ai)) { + hdmi_stop_infoframe_trans(codec, pin_nid); + hdmi_fill_audio_infoframe(codec, pin_nid, &ai); + hdmi_start_infoframe_trans(codec, pin_nid); + } } } @@ -767,16 +791,6 @@ static int intel_hdmi_playback_pcm_clean struct hda_codec *codec, struct snd_pcm_substream *substream) { - struct intel_hdmi_spec *spec = codec->spec; - int i; - - for (i = 0; i < spec->num_pins; i++) { - if (spec->pin_cvt[i] != hinfo->nid) - continue; - - hdmi_stop_infoframe_trans(codec, spec->pin[i]); - } - snd_hda_codec_cleanup_stream(codec, hinfo->nid); return 0; } [-- Attachment #2: Type: text/plain, Size: 160 bytes --] _______________________________________________ Alsa-devel mailing list Alsa-devel@alsa-project.org http://mailman.alsa-project.org/mailman/listinfo/alsa-devel ^ permalink raw reply [flat|nested] 15+ messages in thread
* [PATCH 08/11] intelhdmi - sticky stream id and format 2009-11-15 11:32 [PATCH 00/11] Intel HDMI fixes Wu Fengguang ` (6 preceding siblings ...) 2009-11-15 11:32 ` [PATCH 07/11] intelhdmi - sticky infoframe Wu Fengguang @ 2009-11-15 11:32 ` Wu Fengguang 2009-11-15 11:32 ` [PATCH 09/11] intelhdmi - sticky channel count Wu Fengguang ` (4 subsequent siblings) 12 siblings, 0 replies; 15+ messages in thread From: Wu Fengguang @ 2009-11-15 11:32 UTC (permalink / raw) To: Takashi Iwai; +Cc: alsa-devel, Wu Fengguang, Shane W, David Härdeman [-- Warning: decoded text below may be mangled, UTF-8 assumed --] [-- Attachment #1: hdmi-setup-stream.patch --] [-- Type: text/plain, Size: 2111 bytes --] We tracked down the first-0.5s-hdmi-audio-samples-lost problem to the AC_VERB_SET_CHANNEL_STREAMID command. It is suspected that many HDMI sinks need some time to adapt to the new state. The workaround is to avoid changing stream id/format whenever possible. Proposed by David. Signed-off-by: David Härdeman <david@hardeman.nu> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com> --- sound/pci/hda/patch_intelhdmi.c | 28 ++++++++++++++++++++++++++-- 1 file changed, 26 insertions(+), 2 deletions(-) --- sound-2.6.orig/sound/pci/hda/patch_intelhdmi.c 2009-11-05 13:56:58.000000000 +0800 +++ sound-2.6/sound/pci/hda/patch_intelhdmi.c 2009-11-05 13:58:05.000000000 +0800 @@ -772,6 +772,31 @@ static void intel_hdmi_unsol_event(struc * Callbacks */ +static void hdmi_setup_stream(struct hda_codec *codec, hda_nid_t nid, + u32 stream_tag, int format) +{ + int tag; + int fmt; + + tag = snd_hda_codec_read(codec, nid, 0, AC_VERB_GET_CONV, 0) >> 4; + fmt = snd_hda_codec_read(codec, nid, 0, AC_VERB_GET_STREAM_FORMAT, 0); + + snd_printdd("hdmi_setup_stream: " + "NID=0x%x, %sstream=0x%x, %sformat=0x%x\n", + nid, + tag == stream_tag ? "" : "new-", + stream_tag, + fmt == format ? "" : "new-", + format); + + if (tag != stream_tag) + snd_hda_codec_write(codec, nid, 0, + AC_VERB_SET_CHANNEL_STREAMID, stream_tag << 4); + if (fmt != format) + snd_hda_codec_write(codec, nid, 0, + AC_VERB_SET_STREAM_FORMAT, format); +} + static int intel_hdmi_playback_pcm_prepare(struct hda_pcm_stream *hinfo, struct hda_codec *codec, unsigned int stream_tag, @@ -783,7 +808,7 @@ static int intel_hdmi_playback_pcm_prepa hdmi_setup_audio_infoframe(codec, hinfo->nid, substream); - snd_hda_codec_setup_stream(codec, hinfo->nid, stream_tag, 0, format); + hdmi_setup_stream(codec, hinfo->nid, stream_tag, format); return 0; } @@ -791,7 +816,6 @@ static int intel_hdmi_playback_pcm_clean struct hda_codec *codec, struct snd_pcm_substream *substream) { - snd_hda_codec_cleanup_stream(codec, hinfo->nid); return 0; } [-- Attachment #2: Type: text/plain, Size: 160 bytes --] _______________________________________________ Alsa-devel mailing list Alsa-devel@alsa-project.org http://mailman.alsa-project.org/mailman/listinfo/alsa-devel ^ permalink raw reply [flat|nested] 15+ messages in thread
* [PATCH 09/11] intelhdmi - sticky channel count 2009-11-15 11:32 [PATCH 00/11] Intel HDMI fixes Wu Fengguang ` (7 preceding siblings ...) 2009-11-15 11:32 ` [PATCH 08/11] intelhdmi - sticky stream id and format Wu Fengguang @ 2009-11-15 11:32 ` Wu Fengguang 2009-11-15 11:32 ` [PATCH 10/11] hda - show EPSS capability in proc Wu Fengguang ` (3 subsequent siblings) 12 siblings, 0 replies; 15+ messages in thread From: Wu Fengguang @ 2009-11-15 11:32 UTC (permalink / raw) To: Takashi Iwai; +Cc: alsa-devel, Wu Fengguang, Shane W, David Härdeman [-- Attachment #1: hdmi-sticky-channel-count.patch --] [-- Type: text/plain, Size: 1227 bytes --] Don't change channel count if not necessary. Signed-off-by: Wu Fengguang <fengguang.wu@intel.com> --- sound/pci/hda/patch_intelhdmi.c | 10 ++-------- 1 file changed, 2 insertions(+), 8 deletions(-) --- sound-2.6.orig/sound/pci/hda/patch_intelhdmi.c 2009-11-05 16:48:43.000000000 +0800 +++ sound-2.6/sound/pci/hda/patch_intelhdmi.c 2009-11-05 16:49:22.000000000 +0800 @@ -422,24 +422,18 @@ static void hdmi_stop_infoframe_trans(st AC_DIPXMIT_DISABLE); } -#ifdef CONFIG_SND_DEBUG_VERBOSE static int hdmi_get_channel_count(struct hda_codec *codec, hda_nid_t nid) { return 1 + snd_hda_codec_read(codec, nid, 0, AC_VERB_GET_CVT_CHAN_COUNT, 0); } -#endif static void hdmi_set_channel_count(struct hda_codec *codec, hda_nid_t nid, int chs) { - snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_CVT_CHAN_COUNT, chs - 1); - -#ifdef CONFIG_SND_DEBUG_VERBOSE if (chs != hdmi_get_channel_count(codec, nid)) - snd_printd(KERN_INFO "HDMI channel count: expect %d, get %d\n", - chs, hdmi_get_channel_count(codec, nid)); -#endif + snd_hda_codec_write(codec, nid, 0, + AC_VERB_SET_CVT_CHAN_COUNT, chs - 1); } static void hdmi_debug_channel_mapping(struct hda_codec *codec, hda_nid_t nid) ^ permalink raw reply [flat|nested] 15+ messages in thread
* [PATCH 10/11] hda - show EPSS capability in proc 2009-11-15 11:32 [PATCH 00/11] Intel HDMI fixes Wu Fengguang ` (8 preceding siblings ...) 2009-11-15 11:32 ` [PATCH 09/11] intelhdmi - sticky channel count Wu Fengguang @ 2009-11-15 11:32 ` Wu Fengguang 2009-11-15 11:32 ` [PATCH 11/11] intelhdmi - keep power on to avoid losing hot plug event Wu Fengguang ` (2 subsequent siblings) 12 siblings, 0 replies; 15+ messages in thread From: Wu Fengguang @ 2009-11-15 11:32 UTC (permalink / raw) To: Takashi Iwai; +Cc: alsa-devel, Wu Fengguang, Shane W, David Härdeman [-- Attachment #1: hda-power-epss.patch --] [-- Type: text/plain, Size: 2366 bytes --] Signed-off-by: Wu Fengguang <fengguang.wu@intel.com> --- sound/pci/hda/hda_codec.h | 4 ++++ sound/pci/hda/hda_proc.c | 31 +++++++++++++++++++++++++++++++ 2 files changed, 35 insertions(+) --- sound-2.6.orig/sound/pci/hda/hda_codec.h 2009-11-15 19:28:28.000000000 +0800 +++ sound-2.6/sound/pci/hda/hda_codec.h 2009-11-15 19:28:57.000000000 +0800 @@ -286,6 +286,10 @@ enum { #define AC_PWRST_D1SUP (1<<1) #define AC_PWRST_D2SUP (1<<2) #define AC_PWRST_D3SUP (1<<3) +#define AC_PWRST_D3COLDSUP (1<<4) +#define AC_PWRST_S3D3COLDSUP (1<<29) +#define AC_PWRST_CLKSTOP (1<<30) +#define AC_PWRST_EPSS (1U<<31) /* Power state values */ #define AC_PWRST_SETTING (0xf<<0) --- sound-2.6.orig/sound/pci/hda/hda_proc.c 2009-11-15 19:28:28.000000000 +0800 +++ sound-2.6/sound/pci/hda/hda_proc.c 2009-11-15 19:31:44.000000000 +0800 @@ -26,6 +26,21 @@ #include "hda_codec.h" #include "hda_local.h" +static char *bits_names(unsigned int bits, char *names[], int size) +{ + int i, n; + static char buf[128]; + + for (i = 0, n = 0; i < size; i++) { + if (bits & (1U<<i) && names[i]) + n += snprintf(buf + n, sizeof(buf) - n, " %s", + names[i]); + } + buf[n] = '\0'; + + return buf; +} + static const char *get_wid_type_name(unsigned int wid_value) { static char *names[16] = { @@ -363,8 +378,24 @@ static const char *get_pwr_state(u32 sta static void print_power_state(struct snd_info_buffer *buffer, struct hda_codec *codec, hda_nid_t nid) { + static char *names[] = { + [ilog2(AC_PWRST_D0SUP)] = "D0", + [ilog2(AC_PWRST_D1SUP)] = "D1", + [ilog2(AC_PWRST_D2SUP)] = "D2", + [ilog2(AC_PWRST_D3SUP)] = "D3", + [ilog2(AC_PWRST_D3COLDSUP)] = "D3cold", + [ilog2(AC_PWRST_S3D3COLDSUP)] = "S3D3cold", + [ilog2(AC_PWRST_CLKSTOP)] = "CLKSTOP", + [ilog2(AC_PWRST_EPSS)] = "EPSS", + }; + + int sup = snd_hda_param_read(codec, nid, AC_PAR_POWER_STATE); int pwr = snd_hda_codec_read(codec, nid, 0, AC_VERB_GET_POWER_STATE, 0); + if (sup) + snd_iprintf(buffer, " Power states: %s\n", + bits_names(sup, names, ARRAY_SIZE(names))); + snd_iprintf(buffer, " Power: setting=%s, actual=%s\n", get_pwr_state(pwr & AC_PWRST_SETTING), get_pwr_state((pwr & AC_PWRST_ACTUAL) >> ^ permalink raw reply [flat|nested] 15+ messages in thread
* [PATCH 11/11] intelhdmi - keep power on to avoid losing hot plug event 2009-11-15 11:32 [PATCH 00/11] Intel HDMI fixes Wu Fengguang ` (9 preceding siblings ...) 2009-11-15 11:32 ` [PATCH 10/11] hda - show EPSS capability in proc Wu Fengguang @ 2009-11-15 11:32 ` Wu Fengguang 2009-11-15 15:54 ` [PATCH 00/11] Intel HDMI fixes Takashi Iwai 2009-11-16 17:12 ` Shane W 12 siblings, 0 replies; 15+ messages in thread From: Wu Fengguang @ 2009-11-15 11:32 UTC (permalink / raw) To: Takashi Iwai; +Cc: alsa-devel, Wu Fengguang, Shane W, David Härdeman [-- Attachment #1: hdmi-power-on.patch --] [-- Type: text/plain, Size: 1411 bytes --] The intelhdmi patch relies on reliable pin presence detection to conditionally generate infoframe for the connected pins. However the codec power off risks losing the hot plug event. Interestingly if such hot plug event is missed (it may not be generated by hardware at all), we cannot even get correct pin presence info with the AC_VERB_GET_PIN_SENSE query some time later. So always keep the HDMI codec power on. This also avoids losing 0.5s audio samples if start playing immediately after powering on the codec. Signed-off-by: Wu Fengguang <fengguang.wu@intel.com> --- sound/pci/hda/patch_intelhdmi.c | 9 +++++++++ 1 file changed, 9 insertions(+) --- sound-2.6.orig/sound/pci/hda/patch_intelhdmi.c 2009-11-13 18:21:41.000000000 +0800 +++ sound-2.6/sound/pci/hda/patch_intelhdmi.c 2009-11-13 18:21:47.000000000 +0800 @@ -868,6 +868,15 @@ static int intel_hdmi_init(struct hda_co struct intel_hdmi_spec *spec = codec->spec; int i; + /* + * never power off at idle time: + * - if powered off, the monitor hot plug events will be ignored, + * after that AC_VERB_GET_PIN_SENSE returns inaccurate value + * - we need to keep the HDMI link active to avoid the loss of first + * 0.5s audio samples (due to slow responding HDMI sinks) + */ + codec->power_count++; + for (i = 0; spec->pin[i]; i++) { hdmi_enable_output(codec, spec->pin[i]); snd_hda_codec_write(codec, spec->pin[i], 0, ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH 00/11] Intel HDMI fixes 2009-11-15 11:32 [PATCH 00/11] Intel HDMI fixes Wu Fengguang ` (10 preceding siblings ...) 2009-11-15 11:32 ` [PATCH 11/11] intelhdmi - keep power on to avoid losing hot plug event Wu Fengguang @ 2009-11-15 15:54 ` Takashi Iwai 2009-11-18 4:35 ` Wu Fengguang 2009-11-16 17:12 ` Shane W 12 siblings, 1 reply; 15+ messages in thread From: Takashi Iwai @ 2009-11-15 15:54 UTC (permalink / raw) To: Wu Fengguang; +Cc: alsa-devel, Shane W, David Härdeman At Sun, 15 Nov 2009 19:32:19 +0800, Wu Fengguang wrote: > > Hi Takashi, > > Here is some Intel HDMI fixes, most of them are inspired/written by > David Härdeman (Big thanks!). > > - fix audio infoframe size > - sticky infoframe/stream id/channel count to > - not disturb (David's) fragile devices > - avoid the first 0.5s audio samples be losted > - avoid losing presence info during > - module reloading (by querying the pin sense status) > - power off (by disabling it) Thanks for the patches! Most of them look good; however, I'm not much convinced wrt the disablement of power-saving. In many cases, the power-saving is more important than 0.5sec drop-out (particularly if it's already fixed in the succeeding playbacks). Thus disabling the power-saving unconditionally is no good idea. Maybe some flag, either an option or a mixer control, would be preferred for defining the behavior, as a compromise... Also, it'd be more readable if we define another common jack detection routine that calls the pin sense function, such as int snd_hda_jack_detect(struct hda_codec *codc, hda_nid_t nid) { u32 sense = snd_hda_pin_sense(codec, nid); return !!(sense & AC_PINSENSE_PRESENCE); } since all callers check only that bit. Last but not least, don't forget to add comment to each new common function. Although it's trivial, the documentation is always better to have :) thanks, Takashi _______________________________________________ Alsa-devel mailing list Alsa-devel@alsa-project.org http://mailman.alsa-project.org/mailman/listinfo/alsa-devel ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH 00/11] Intel HDMI fixes 2009-11-15 15:54 ` [PATCH 00/11] Intel HDMI fixes Takashi Iwai @ 2009-11-18 4:35 ` Wu Fengguang 0 siblings, 0 replies; 15+ messages in thread From: Wu Fengguang @ 2009-11-18 4:35 UTC (permalink / raw) To: Takashi Iwai; +Cc: alsa-devel@alsa-project.org, Shane W, David Härdeman On Sun, Nov 15, 2009 at 11:54:57PM +0800, Takashi Iwai wrote: > At Sun, 15 Nov 2009 19:32:19 +0800, > Wu Fengguang wrote: > > > > Hi Takashi, > > > > Here is some Intel HDMI fixes, most of them are inspired/written by > > David Härdeman (Big thanks!). > > > > - fix audio infoframe size > > - sticky infoframe/stream id/channel count to > > - not disturb (David's) fragile devices > > - avoid the first 0.5s audio samples be losted > > - avoid losing presence info during > > - module reloading (by querying the pin sense status) > > - power off (by disabling it) > > Thanks for the patches! > > Most of them look good; however, I'm not much convinced wrt the > disablement of power-saving. In many cases, the power-saving is more > important than 0.5sec drop-out (particularly if it's already fixed in > the succeeding playbacks). Thus disabling the power-saving > unconditionally is no good idea. > > Maybe some flag, either an option or a mixer control, would be > preferred for defining the behavior, as a compromise... OK. I'll do this in a separate patch. > Also, it'd be more readable if we define another common jack > detection routine that calls the pin sense function, such as > > int snd_hda_jack_detect(struct hda_codec *codc, hda_nid_t nid) > { > u32 sense = snd_hda_pin_sense(codec, nid); > return !!(sense & AC_PINSENSE_PRESENCE); > } > > since all callers check only that bit. > > Last but not least, don't forget to add comment to each new common > function. Although it's trivial, the documentation is always better > to have :) Good suggestions, thanks! All have been incorporated into the updated patches. Thanks, Fengguang _______________________________________________ Alsa-devel mailing list Alsa-devel@alsa-project.org http://mailman.alsa-project.org/mailman/listinfo/alsa-devel ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH 00/11] Intel HDMI fixes 2009-11-15 11:32 [PATCH 00/11] Intel HDMI fixes Wu Fengguang ` (11 preceding siblings ...) 2009-11-15 15:54 ` [PATCH 00/11] Intel HDMI fixes Takashi Iwai @ 2009-11-16 17:12 ` Shane W 12 siblings, 0 replies; 15+ messages in thread From: Shane W @ 2009-11-16 17:12 UTC (permalink / raw) To: Wu Fengguang; +Cc: Takashi Iwai, alsa-devel, David H??rdeman [-- Warning: decoded text below may be mangled, UTF-8 assumed --] [-- Attachment #1: Type: text/plain; charset=unknown-8bit, Size: 288 bytes --] On Sun, Nov 15, 2009 at 07:32:19PM +0800, Wu Fengguang wrote: > Here is some Intel HDMI fixes, most of them are inspired/written by > David Härdeman (Big thanks!). Indeed, thanks for these. Just to confirm, this fixes the 0.5s silence on my DG45/RX-V1800 combo. Cheers, Shane [-- Attachment #2: Type: text/plain, Size: 160 bytes --] _______________________________________________ Alsa-devel mailing list Alsa-devel@alsa-project.org http://mailman.alsa-project.org/mailman/listinfo/alsa-devel ^ permalink raw reply [flat|nested] 15+ messages in thread
end of thread, other threads:[~2009-11-18 4:35 UTC | newest] Thread overview: 15+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2009-11-15 11:32 [PATCH 00/11] Intel HDMI fixes Wu Fengguang 2009-11-15 11:32 ` [PATCH 01/11] intelhdmi - fix audio infoframe fill size Wu Fengguang 2009-11-15 11:32 ` [PATCH 02/11] intelhdmi - fix channel mapping slot mask Wu Fengguang 2009-11-15 11:32 ` [PATCH 03/11] intelhdmi - export monitor-presence and ELD-valid status Wu Fengguang 2009-11-15 11:32 ` [PATCH 04/11] hda - introduce snd_hda_pin_sense() Wu Fengguang 2009-11-15 11:32 ` [PATCH 05/11] intelhdmi - probe for monitor/eld presence at module init time Wu Fengguang 2009-11-15 11:32 ` [PATCH 06/11] intelhdmi - separate out infoframe checksum routine Wu Fengguang 2009-11-15 11:32 ` [PATCH 07/11] intelhdmi - sticky infoframe Wu Fengguang 2009-11-15 11:32 ` [PATCH 08/11] intelhdmi - sticky stream id and format Wu Fengguang 2009-11-15 11:32 ` [PATCH 09/11] intelhdmi - sticky channel count Wu Fengguang 2009-11-15 11:32 ` [PATCH 10/11] hda - show EPSS capability in proc Wu Fengguang 2009-11-15 11:32 ` [PATCH 11/11] intelhdmi - keep power on to avoid losing hot plug event Wu Fengguang 2009-11-15 15:54 ` [PATCH 00/11] Intel HDMI fixes Takashi Iwai 2009-11-18 4:35 ` Wu Fengguang 2009-11-16 17:12 ` Shane W
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.