Re: [RFC/RFT v2 0/4] ALSA: hda - hdmi: ATI/AMD multi-channel and HBR support

[Takashi Iwai]

At Tue,  1 Oct 2013 23:30:52 +0300,
Anssi Hannula wrote:
> 
> Hi all!
> 
> Here is a second revision of the ATI/AMD multichannel patch, now a
> patchset.
> 
> Since the last revision from a bit over the week ago, AMD has released
> more documentation ( http://www.x.org/docs/AMD/AMD_HDA_verbs_v2.pdf ),
> so the patchset contains new additions:
> 
> - HBR bitstreaming support (HD passthrough, for DTS-HD and Dolby TrueHD)
>   o This completes the set, now we have HBR on NVIDIA, Intel and AMD :)
> 
> - More complete ELD emulation on revision 3+ (0x100300) codecs
>   o Untested, though from what I can see the newly added fields may not
>     be filled in yet by the radeon driver.
> 
> - Ramp up/down configuration according to non-pcm bit on rev3+
>   o I guess this means gradual fade-in/out? Untested.
> 
> Other changes since the previous post:
> - ELD emulation moved to hda_eld.c and cleaned up a bit
> - Some fixes to hdmi_chmap_ctl_tlv() changes
> (and maybe something else I forgot - if so, sorry)
> 
> Also, the pairwise channel mapping code has been tested to work with
> custom channel maps. The TLV contents are shown as follows by amixer:
>     | chmap-paired=FL,FR
>     | chmap-paired=FL,FR,NA,LFE
>     | chmap-paired=FL,FR,FC,NA
>     | chmap-paired=FL,FR,RL,RR
>     | chmap-paired=FL,FR,FC,LFE
>     | chmap-paired=FL,FR,NA,LFE,RL,RR
>     | chmap-paired=FL,FR,FC,NA,RL,RR
> [etc]
> 
> RafaÅ, do you know about the missing (AFAICS) support for lipsync (F7B)
> and sink information (F81) in radeon driver?
> 
> This patchset requires at least more testing before it should be applied.
> 
> A few other somewhat open things:
> - I named everything ati/ATI except for is_amdhdmi_rev3() and
>   has_amd_full_remap_support() that are AMD-only, though AMD-only verbs
>   I still named ATI to have similar names to other verbs. Does that
>   seem sensible or how should this be done?
> - Currently setup_audio_infoframe() writes the channel mapping every
>   time for ATI/AMD, should we try to avoid that? If so, any
>   suggestions?
>   (argh, just noticed this is broken for manual channel maps on non-ATI;
>    if just the chmap is changed, channel mapping is not updated;
>    so this requires fixes in the generic side as well)
> 
> I'm especially interested in testers with:
> 
> - Older codecs other than 0x1002aa01. My best guess still is that the
>   new code works on them as well.
>   o On these I'd like to know if multichannel and the new formats
>     work, i.e. e.g.
>     speaker-test -D hdmi:CARD=Generic,DEV=0 -c8 -r192000 -F S32_LE
> 
> - Codec 0x1002aa01 revisions 0x100300 (Rev ID 3) or later (see
>   /proc/asound/cardX/codec#Y). These are HD7750+ I think. Stuff to be
>   tested on these:
>   o The ramp up/down stuff, i.e. patch 4. Is there any difference seen
>     with these, in the beginning/end (i.e. fade-out/in):
>     speaker-test -D hdmi:CARD=Generic,DEV=0,AES0=0x04 -c2 -r48000
>     speaker-test -D hdmi:CARD=Generic,DEV=0,AES0=0x06 -c2 -r48000
>     Also, is there a difference in the beginning of these
>     (maybe garbage sound and/or slightly slower startup?):
>     aplay -Dhdmi:CARD=Generic,DEV=0,AES0=4 -r44100 -f s16_le -c2 testi.dts.cut.spdif
>     aplay -Dhdmi:CARD=Generic,DEV=0,AES0=6 -r44100 -f s16_le -c2 testi.dts.cut.spdif
> 
>   o Contents of /proc/asound/cardX/eld#0. I'd like to see the contents
>     both with radeon and with the proprietary fglrx driver in use
> 
> Olivier, can you test again on your rev3 hw? :)
> 
> The patchset can be found in combined form (for e.g. testing purposes) at:
> http://onse.fi/files/atihdmi5.patch
> 
> The test file referenced above can be found at:
> http://onse.fi/files/testi.dts.cut.spdif.gz (just regular DTS)
> 
> Anssi Hannula (4):
>       ALSA: hda - hdmi: Add ATI/AMD multi-channel audio support
>       ALSA: hda - hdmi: Add ELD emulation for ATI/AMD codecs
>       ALSA: hda - hdmi: Add HBR bitstreaming support for ATI/AMD HDMI codecs
>       ALSA: hda - hdmi: Disable ramp-up/down for non-PCM on AMD codecs


The patches generally look good, but I'm thinking whether it's cleaner
to create a new ops items.  Maybe it can be specific to HDMI codec,
e.g.

struct hda_hdmi_ops {
	int (*get_chan_slot)(struct hda_codec *codec, hda_nid_t pin, int slot);
	int (*set_chan_slot)(struct hda_codec *codec, hda_nid_t pin, int setup);
	int (*set_ca)(struct hda_codec *codec, hda_nid_t pin, int ca);
	....
}

It's not sexy to have a thing like is_atihdmi() in the common header
and see it in everywhere.


thanks,

Takashi

> 
>  sound/pci/hda/hda_eld.c    | 157 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>  sound/pci/hda/hda_local.h  |   5 +++
>  sound/pci/hda/patch_hdmi.c | 424 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++------------------
>  3 files changed, 547 insertions(+), 39 deletions(-)
> 
> -- 
> Anssi Hannula
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/