[alsa-devel] [PATCH 0/2] ALSA: hda - support HDMI HBR passthrough
Hi all!
This patchset adds support for passing through IEC 61937 encapsulated compressed audio at high bitrates (i.e. those over 6.144Mbps). At least TrueHD and DTS-HD are such formats.
I've tested this using an NVIDIA hdmi codec and the following ffmpeg patch which adds support for TrueHD in its IEC 61937 muxer: http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/2010-August/094317.html
One needs to set the AES0 & 0x02 (non-audio) bit during playback so that the data is not transmitted as normal 8 channel PCM audio.
--- Anssi Hannula (2): ALSA: hda - Set Stream Type in Stream Format according to AES0 ALSA: hda - Add support for HDMI HBR passthrough
sound/pci/hda/hda_codec.c | 6 ++++- sound/pci/hda/hda_codec.h | 6 ++++- sound/pci/hda/hda_intel.c | 3 +- sound/pci/hda/patch_hdmi.c | 40 ++++++++++++++++++++++++++++++++++++++- sound/pci/hda/patch_intelhdmi.c | 3 +- sound/pci/hda/patch_nvhdmi.c | 3 +- 6 files changed, 53 insertions(+), 8 deletions(-)
Set bit 15 (Stream Type) of HDA Stream Format to 1 (Non-PCM) when IEC958 channel status bit 1 (AES0 & 0x02) is set to 1 (non-audio).
This is a prequisite for HDMI HBR passthrough.
Signed-off-by: Anssi Hannula anssi.hannula@iki.fi --- sound/pci/hda/hda_codec.c | 6 +++++- sound/pci/hda/hda_codec.h | 3 ++- sound/pci/hda/hda_intel.c | 3 ++- 3 files changed, 9 insertions(+), 3 deletions(-)
diff --git a/sound/pci/hda/hda_codec.c b/sound/pci/hda/hda_codec.c index d9d1c91..bd8d7a6 100644 --- a/sound/pci/hda/hda_codec.c +++ b/sound/pci/hda/hda_codec.c @@ -3051,7 +3051,8 @@ static struct hda_rate_tbl rate_bits[] = { unsigned int snd_hda_calc_stream_format(unsigned int rate, unsigned int channels, unsigned int format, - unsigned int maxbps) + unsigned int maxbps, + unsigned short spdif_ctls) { int i; unsigned int val = 0; @@ -3095,6 +3096,9 @@ unsigned int snd_hda_calc_stream_format(unsigned int rate, return 0; }
+ if (spdif_ctls & AC_DIG1_NONAUDIO) + val |= 0x8000; + return val; } EXPORT_SYMBOL_HDA(snd_hda_calc_stream_format); diff --git a/sound/pci/hda/hda_codec.h b/sound/pci/hda/hda_codec.h index 5991d14..4797416 100644 --- a/sound/pci/hda/hda_codec.h +++ b/sound/pci/hda/hda_codec.h @@ -928,7 +928,8 @@ void snd_hda_codec_cleanup_stream(struct hda_codec *codec, hda_nid_t nid); unsigned int snd_hda_calc_stream_format(unsigned int rate, unsigned int channels, unsigned int format, - unsigned int maxbps); + unsigned int maxbps, + unsigned short spdif_ctls); int snd_hda_is_supported_format(struct hda_codec *codec, hda_nid_t nid, unsigned int format);
diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index 1df25cf..f8a2f5a 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -1653,7 +1653,8 @@ static int azx_pcm_prepare(struct snd_pcm_substream *substream) format_val = snd_hda_calc_stream_format(runtime->rate, runtime->channels, runtime->format, - hinfo->maxbps); + hinfo->maxbps, + apcm->codec->spdif_ctls); if (!format_val) { snd_printk(KERN_ERR SFX "invalid format_val, rate=%d, ch=%d, format=%d\n",
Passing IEC 61937 encapsulated compressed audio at bitrates over 6.144 Mbps (i.e. more than a single 2-channel 16-bit 192kHz IEC 60958 link) over HDMI requires the use of HBR Audio Stream Packets instead of Audio Sample Packets.
Enable HBR mode when the stream has 8 channels and the Non-PCM bit is set.
If the audio converter is not connected to any HBR-capable pins, return -EINVAL in prepare().
Signed-off-by: Anssi Hannula anssi.hannula@iki.fi --- sound/pci/hda/hda_codec.h | 3 ++ sound/pci/hda/patch_hdmi.c | 40 ++++++++++++++++++++++++++++++++++++++- sound/pci/hda/patch_intelhdmi.c | 3 +- sound/pci/hda/patch_nvhdmi.c | 3 +- 4 files changed, 44 insertions(+), 5 deletions(-)
diff --git a/sound/pci/hda/hda_codec.h b/sound/pci/hda/hda_codec.h index 4797416..48b3367 100644 --- a/sound/pci/hda/hda_codec.h +++ b/sound/pci/hda/hda_codec.h @@ -364,6 +364,9 @@ enum { #define AC_DIG2_CC (0x7f<<0)
/* Pin widget control - 8bit */ +#define AC_PINCTL_EPT (0x3<<0) +#define AC_PINCTL_EPT_NATIVE 0 +#define AC_PINCTL_EPT_HBR 3 #define AC_PINCTL_VREFEN (0x7<<0) #define AC_PINCTL_VREF_HIZ 0 /* Hi-Z */ #define AC_PINCTL_VREF_50 1 /* 50% */ diff --git a/sound/pci/hda/patch_hdmi.c b/sound/pci/hda/patch_hdmi.c index 2fc5396..8534792 100644 --- a/sound/pci/hda/patch_hdmi.c +++ b/sound/pci/hda/patch_hdmi.c @@ -698,11 +698,48 @@ static void hdmi_unsol_event(struct hda_codec *codec, unsigned int res) * Callbacks */
-static void hdmi_setup_stream(struct hda_codec *codec, hda_nid_t nid, +static int hdmi_setup_stream(struct hda_codec *codec, hda_nid_t nid, u32 stream_tag, int format) { + struct hdmi_spec *spec = codec->spec; int tag; int fmt; + int pinctl; + int new_pinctl = 0; + int i; + + for (i = 0; i < spec->num_pins; i++) { + if (spec->pin_cvt[i] != nid) + continue; + if (!(snd_hda_query_pin_caps(codec, spec->pin[i]) & AC_PINCAP_HBR)) + continue; + + pinctl = snd_hda_codec_read(codec, spec->pin[i], 0, + AC_VERB_GET_PIN_WIDGET_CONTROL, 0); + + new_pinctl = pinctl & ~AC_PINCTL_EPT; + /* Non-PCM, 8 channels */ + if ((format & 0x8000) && (format & 0x0f) == 7) + new_pinctl |= AC_PINCTL_EPT_HBR; + else + new_pinctl |= AC_PINCTL_EPT_NATIVE; + + snd_printdd("hdmi_setup_stream: " + "NID=0x%x, %spinctl=0x%x\n", + spec->pin[i], + pinctl == new_pinctl ? "" : "new-", + new_pinctl); + + if (pinctl != new_pinctl) + snd_hda_codec_write(codec, spec->pin[i], 0, + AC_VERB_SET_PIN_WIDGET_CONTROL, + new_pinctl); + } + + if ((format & 0x8000) && (format & 0x0f) == 7 && !new_pinctl) { + snd_printdd("hdmi_setup_stream: HBR is not supported\n"); + return -EINVAL; + }
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); @@ -722,6 +759,7 @@ static void hdmi_setup_stream(struct hda_codec *codec, hda_nid_t nid, if (fmt != format) snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_STREAM_FORMAT, format); + return 0; }
/* diff --git a/sound/pci/hda/patch_intelhdmi.c b/sound/pci/hda/patch_intelhdmi.c index b81d23e..5972d5e 100644 --- a/sound/pci/hda/patch_intelhdmi.c +++ b/sound/pci/hda/patch_intelhdmi.c @@ -66,8 +66,7 @@ static int intel_hdmi_playback_pcm_prepare(struct hda_pcm_stream *hinfo,
hdmi_setup_audio_infoframe(codec, hinfo->nid, substream);
- hdmi_setup_stream(codec, hinfo->nid, stream_tag, format); - return 0; + return hdmi_setup_stream(codec, hinfo->nid, stream_tag, format); }
static int intel_hdmi_playback_pcm_cleanup(struct hda_pcm_stream *hinfo, diff --git a/sound/pci/hda/patch_nvhdmi.c b/sound/pci/hda/patch_nvhdmi.c index b0652ac..a281836 100644 --- a/sound/pci/hda/patch_nvhdmi.c +++ b/sound/pci/hda/patch_nvhdmi.c @@ -202,8 +202,7 @@ static int nvhdmi_dig_playback_pcm_prepare_8ch_89(struct hda_pcm_stream *hinfo,
hdmi_setup_audio_infoframe(codec, hinfo->nid, substream);
- hdmi_setup_stream(codec, hinfo->nid, stream_tag, format); - return 0; + return hdmi_setup_stream(codec, hinfo->nid, stream_tag, format); }
static int nvhdmi_dig_playback_pcm_prepare_8ch(struct hda_pcm_stream *hinfo,
At Tue, 3 Aug 2010 13:28:56 +0300, Anssi Hannula wrote:
Hi all!
This patchset adds support for passing through IEC 61937 encapsulated compressed audio at high bitrates (i.e. those over 6.144Mbps). At least TrueHD and DTS-HD are such formats.
I've tested this using an NVIDIA hdmi codec and the following ffmpeg patch which adds support for TrueHD in its IEC 61937 muxer: http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/2010-August/094317.html
One needs to set the AES0 & 0x02 (non-audio) bit during playback so that the data is not transmitted as normal 8 channel PCM audio.
Thanks, I applied both patches.
There are some magic numbers there like 0x8000, and I'm going to define them...
Takashi
Anssi Hannula (2): ALSA: hda - Set Stream Type in Stream Format according to AES0 ALSA: hda - Add support for HDMI HBR passthrough
sound/pci/hda/hda_codec.c | 6 ++++- sound/pci/hda/hda_codec.h | 6 ++++- sound/pci/hda/hda_intel.c | 3 +- sound/pci/hda/patch_hdmi.c | 40 ++++++++++++++++++++++++++++++++++++++- sound/pci/hda/patch_intelhdmi.c | 3 +- sound/pci/hda/patch_nvhdmi.c | 3 +- 6 files changed, 53 insertions(+), 8 deletions(-)
Takashi Iwai kirjoitti tiistai, 3. elokuuta 2010 13:59:09:
At Tue, 3 Aug 2010 13:28:56 +0300,
Anssi Hannula wrote:
Hi all!
This patchset adds support for passing through IEC 61937 encapsulated compressed audio at high bitrates (i.e. those over 6.144Mbps). At least TrueHD and DTS-HD are such formats.
I've tested this using an NVIDIA hdmi codec and the following ffmpeg patch which adds support for TrueHD in its IEC 61937 muxer: http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/2010-August/094317.html
One needs to set the AES0 & 0x02 (non-audio) bit during playback so that the data is not transmitted as normal 8 channel PCM audio.
Thanks, I applied both patches.
There are some magic numbers there like 0x8000, and I'm going to define them...
Yeah, I guess all the Stream Format bits should be defined to get rid of all the existing magic numbers in snd_hda_calc_stream_format() as well.
Takashi
Anssi Hannula (2): ALSA: hda - Set Stream Type in Stream Format according to AES0 ALSA: hda - Add support for HDMI HBR passthrough
sound/pci/hda/hda_codec.c | 6 ++++- sound/pci/hda/hda_codec.h | 6 ++++- sound/pci/hda/hda_intel.c | 3 +- sound/pci/hda/patch_hdmi.c | 40 ++++++++++++++++++++++++++++++++++++++- sound/pci/hda/patch_intelhdmi.c | 3 +- sound/pci/hda/patch_nvhdmi.c | 3 +- 6 files changed, 53 insertions(+), 8 deletions(-)
At Tue, 3 Aug 2010 14:04:01 +0300, Anssi Hannula wrote:
Takashi Iwai kirjoitti tiistai, 3. elokuuta 2010 13:59:09:
At Tue, 3 Aug 2010 13:28:56 +0300,
Anssi Hannula wrote:
Hi all!
This patchset adds support for passing through IEC 61937 encapsulated compressed audio at high bitrates (i.e. those over 6.144Mbps). At least TrueHD and DTS-HD are such formats.
I've tested this using an NVIDIA hdmi codec and the following ffmpeg patch which adds support for TrueHD in its IEC 61937 muxer: http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/2010-August/094317.html
One needs to set the AES0 & 0x02 (non-audio) bit during playback so that the data is not transmitted as normal 8 channel PCM audio.
Thanks, I applied both patches.
There are some magic numbers there like 0x8000, and I'm going to define them...
Yeah, I guess all the Stream Format bits should be defined to get rid of all the existing magic numbers in snd_hda_calc_stream_format() as well.
Exactly.
Takashi
participants (2)
-
Anssi Hannula -
Takashi Iwai