Re: [ALSA - driver 0001450]: SPDIF out is limited with 48KHz

[Philip Prindeville <philipp_subx@redfish-solutions.com>]

Takashi Iwai wrote:

>At Tue, 11 Oct 2005 21:52:18 +0400,
>Andrew Gaydenko wrote:
>  
>
>>Takashi,
>>
>>Unfortunately, routing problem still exists for (my host)-(bugtrack.alsa-project.org)
>>path. Is it possible to use the mailing list or my direct email (a@gaydenko.com) to
>>send the patch?
>>    
>>
>
>BTS host seems instable for the time being.
>
>Anyway, the patch is below, and I already committed it to CVS tree now
>:)
>
>
>Takashi
>
>Index: alsa-kernel/pci/ice1712/ice1724.c
>===================================================================
>RCS file: /home/iwai/cvs/alsa/alsa-kernel/pci/ice1712/ice1724.c,v
>retrieving revision 1.67
>diff -u -r1.67 ice1724.c
>--- alsa-kernel/pci/ice1712/ice1724.c	12 Sep 2005 10:27:50 -0000	1.67
>+++ alsa-kernel/pci/ice1712/ice1724.c	11 Oct 2005 15:18:42 -0000
>@@ -675,9 +675,12 @@
> 				 SNDRV_PCM_INFO_MMAP_VALID |
> 				 SNDRV_PCM_INFO_PAUSE | SNDRV_PCM_INFO_SYNC_START),
> 	.formats =		SNDRV_PCM_FMTBIT_S32_LE,
>-	.rates =	        SNDRV_PCM_RATE_32000|SNDRV_PCM_RATE_44100|SNDRV_PCM_RATE_48000,
>+	.rates =	        (SNDRV_PCM_RATE_32000|SNDRV_PCM_RATE_44100|
>+				 SNDRV_PCM_RATE_48000|SNDRV_PCM_RATE_88200|
>+				 SNDRV_PCM_RATE_96000|SNDRV_PCM_RATE_176400|
>+				 SNDRV_PCM_RATE_192000),
> 	.rate_min =		32000,
>-	.rate_max =		48000,
>+	.rate_max =		192000,
> 	.channels_min =		2,
> 	.channels_max =		2,
> 	.buffer_bytes_max =	(1UL << 18),	/* 16bits dword */
>@@ -905,6 +908,10 @@
> 	case 44100: break;
> 	case 48000: nval |= 2 << 12; break;
> 	case 32000: nval |= 3 << 12; break;
>+	case 88200: nval |= 4 << 12; break;
>+	case 96000: nval |= 5 << 12; break;
>+	case 192000: nval |= 6 << 12; break;
>+	case 176400: nval |= 7 << 12; break;
> 	}
> 	if (val != nval)
> 		update_spdif_bits(ice, nval);
>@@ -1292,22 +1299,32 @@
> 
> static unsigned int encode_spdif_bits(snd_aes_iec958_t *diga)
> {
>-	unsigned int val;
>+	unsigned int val, rbits;
> 
> 	val = diga->status[0] & 0x03; /* professional, non-audio */
> 	if (val & 0x01) {
> 		/* professional */
> 		if ((diga->status[0] & IEC958_AES0_PRO_EMPHASIS) == IEC958_AES0_PRO_EMPHASIS_5015)
> 			val |= 1U << 3;
>-		switch (diga->status[0] & IEC958_AES0_PRO_FS) {
>-		case IEC958_AES0_PRO_FS_44100:
>-			break;
>-		case IEC958_AES0_PRO_FS_32000:
>-			val |= 3U << 12;
>-			break;
>-		default:
>-			val |= 2U << 12;
>-			break;
>+		rbits = (diga->status[4] >> 3) & 0x0f;
>+		if (rbits) {
>+			switch (rbits) {
>+			case 2: val |= 5 << 12; break; /* 96k */
>+			case 3: val |= 6 << 12; break; /* 192k */
>+			case 10: val |= 4 << 12; break; /* 88.2k */
>+			case 11: val |= 7 << 12; break; /* 176.4k */
>+			}
>+		} else {
>+			switch (diga->status[0] & IEC958_AES0_PRO_FS) {
>+			case IEC958_AES0_PRO_FS_44100:
>+				break;
>+			case IEC958_AES0_PRO_FS_32000:
>+				val |= 3U << 12;
>+				break;
>+			default:
>+				val |= 2U << 12;
>+				break;
>+			}
> 		}
> 	} else {
> 		/* consumer */
>
>  
>

I don't suppose the patch is anyway applicable to the DRS capable
VIA southbridges, is it?

Also... on a broader note...  SPDIF is an asynchronous transport, right?

The transmission rate is unrelated to the actual playback rate, since you
can carry compressed PCM, DTS, DolbyD, etc. all over the same
transport...

If that's the case, why not send 44.1kHz data as uncompressed "compressed
PCM" (i.e. flag it as being compressed, but then don't really compress it,
or compress it losslessly) over SPDIF if it can only do 48kHz (on the
sender's side)?

-Philip



-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl