Skip to contentSkip to conversation

libasound segfault

DISCUSSION Jussi Laako 3 participants 4 messages Mar 10, 2002 – Mar 24, 2002

lore
Share & tools
lore
#0 lore raw

(gdb) bt

#0  0x4015e8d7 in snd_pcm_route_convert1_one () from /usr/lib/libasound.so.2
#1  0x40cae5dc in ?? ()
#2  0x012de908 in ?? ()

Error accessing memory address 0xe8c1018b: No such process. (gdb)

When reading plughw after requesting 24-bit format. I expect the output should be 32-bit words with data in upper 24 bits. 16-bit format works and asking for 32-bit format gives shared memory errors.

alsa-lib-0.9.0beta10a

- Jussi Laako

Signature · 8 lines
PGP key fingerprint: 161D 6FED 6A92 39E2 EB5B  39DD A4DE 63EB C216 1E4B
Available at PGP keyservers


_______________________________________________
Alsa-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/alsa-devel
#1 lore raw
Quoted 11 lines

(gdb) bt

#0  0x4015e8d7 in snd_pcm_route_convert1_one () from /usr/lib/libasound.so.2
#1  0x40cae5dc in ?? ()
#2  0x012de908 in ?? ()

Error accessing memory address 0xe8c1018b: No such process. (gdb)

When reading plughw after requesting 24-bit format. I expect the output should be 32-bit words with data in upper 24 bits. 16-bit format works and asking for 32-bit format gives shared memory errors.

(i was waiting for some insightful comment from jaroslav or abramo :)

no, as I've mentioned several times on LAD, S24_[LB]E is real 24 bit data, not 24-in-32. 24-in-32 is handled by S32_[BL]E, and the msbits value is set in the params struct to indicate where the significant bits are. S32_LE has been used on at least 4 types of ALSA-support with the "hw" device type. I don't know about with "plughw".

--p

_______________________________________________
Alsa-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/alsa-devel

#2 lore raw
Quoted 8 lines · Paul Davis
Quoted 8 lines

(gdb) bt

#0  0x4015e8d7 in snd_pcm_route_convert1_one () from /usr/lib/libasound.so.2
#1  0x40cae5dc in ?? ()
#2  0x012de908 in ?? ()

Error accessing memory address 0xe8c1018b: No such process. (gdb)

Quoted 4 lines
Quoted 4 lines

When reading plughw after requesting 24-bit format. I expect the output should be 32-bit words with data in upper 24 bits. 16-bit format works and

asking for 32-bit format gives shared memory errors.

Can we see your code? The 32-bit format should work.

Quoted 7 lines

(i was waiting for some insightful comment from jaroslav or abramo :)

no, as I've mentioned several times on LAD, S24_[LB]E is real 24 bit data, not 24-in-32. 24-in-32 is handled by S32_[BL]E, and the msbits value is set in the params struct to indicate where the significant bits are. S32_LE has been used on at least 4 types of ALSA-support with the "hw" device type. I don't know about with "plughw".

It's not true. The 24-bit format (S24_[LB]E) in ALSA uses low three bites (most significant byte is ignored). The physical data layout is 32-bit, too. I don't think that any hardware supports / will support 24-bit samples in three bytes.

Jaroslav

-----
Jaroslav Kysela <[email protected]>
Linux Kernel Sound Maintainer
ALSA Project http://www.alsa-project.org
SuSE Linux http://www.suse.com

_______________________________________________
Alsa-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/alsa-devel

#3 lore raw
Quoted 9 lines · Jaroslav Kysela
Quoted 7 lines
Quoted 7 lines

(gdb) bt

#0  0x4015e8d7 in snd_pcm_route_convert1_one () from /usr/lib/libasound.so.2
#1  0x40cae5dc in ?? ()
#2  0x012de908 in ?? ()

Error accessing memory address 0xe8c1018b: No such process. (gdb)

Can we see your code? The 32-bit format should work.

Well, I didn't remember I made 32-bit mean float's, but neither 24 nor 32 bit mode of my code works. I get alsa errors with both.

http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/hasas/HASAS/AudioA2.cc?rev=1.2&content-type=text/vnd.viewcvs-markup

- Jussi Laako

Signature · 2 lines
PGP key fingerprint: 161D 6FED 6A92 39E2 EB5B  39DD A4DE 63EB C216 1E4B
Available at PGP keyservers