The voice recognition software we use has suddenly stopped
working when we moved from 4.1.1 to 4.4
Lookig at the ktraces (it's proprietary code so we can't look
at the code) we see the following differences:
in 4.1.1 we see
5015 hapitest CALL ioctl(0x5,SNDCTL_DSP_GETISPACE,0x85e385c)
5015 hapitest RET ioctl 0
5015 hapitest CALL ioctl(0x5,SNDCTL_DSP_GETISPACE,0x85e385c)
5015 hapitest RET ioctl 0
5015 hapitest CALL ioctl(0x5,SNDCTL_DSP_GETISPACE,0x85e385c)
5015 hapitest RET ioctl 0
5015 hapitest CALL ioctl(0x5,SNDCTL_DSP_GETISPACE,0x85e385c)
5015 hapitest RET ioctl 0
5015 hapitest CALL nanosleep(0xbfbff9c8,0)
5015 hapitest RET nanosleep 0
5015 hapitest CALL ioctl(0x5,SNDCTL_DSP_GETISPACE,0x85e385c)
5015 hapitest RET ioctl 0
5015 hapitest CALL read(0x5,0x84ec826,0x800)
5015 hapitest GIO fd 5 read 2048 bytes
"\b\0
\0\b\0\a\0\a\0\^F\0\^F\0\b\0\a\0\a\0\b\0\^F\0\^F\0\^F\0\^E\0\^F\0\a\0\
\b\0\^F\0\^E\0\a\0\^E\0\^E\0\^F\0\a\0\^E\0\^D\0\^E\0\^E\0\^E\0\^A\0\^C\
\0\^F\0\^C\0\^E\0\^F\0\^D\0\^F\0\^D\0\^A\0\^C\0\^B\0\^A\0\^B\0\^D\0\^E\
\0\^D\0\^D\0\^E\0\^C\0\^C\0\^D\0\^C\0\^D\0\^D\0\^D\0\^C\0\^E\0\^D\0\^E\
\0\^E\0\^B\0\0\0\^C\0\^B\0\^A\0\0\0\^B\0\^C\0\^A\0\^A\0\^B\0\^B\0\^B\0\
\^C\0\^B\0\^D\0\^C\0\0\0\0\0\^A\0\^B\0\0\0\M-}\M^?\M^?\M^?\^A\0\M^?\
\M^?\0\0\^B\0\^A\0\0\0\M^?\M^?\^A\0\^A\0\^A\0\0\0\^A\0\^A\0\^A\0\0\0\
\^A\0\^C\0\0\0\M^?\M^?\^A\0\0\0\M^?\M^?\M-}\M^?\M-~\M^?\0\0\0\0\0\0\^A\
\0\0\0\M-~\M^?\M-~\M^?\M-~\M^?\M-~\M^?\M^?\M^?\M^?\M^?\M-|\M^?\M-}\M^?\
\M-~\M^?\M-~\M^?\M-~\M^?\M-}\M^?\M-~\M^?\M^?\M^?\M-}\M^?\M-}\M^?\M-~\
\M^?\M-|\M^?\M-|\M^?\M-|\M^?\M-{\M^?\M-|\M^?\M-}\M^?\M-|\M^?\M-|\M^?
(and 2048 bytes of sound data read...)
in 4.4 we see:
3121 hapitest RET read 226/0xe2
3121 hapitest CALL ioctl(0x5,SNDCTL_DSP_GETISPACE,0x85e385c)
3121 hapitest RET ioctl 0
3121 hapitest CALL read(0x5,0x8563dea,0x5e)
3121 hapitest GIO fd 5 read 94 bytes
"\^C\0\M-}\M^?\M-}\M^?\^E\0
\0\v\0\a\0\^E\0\^B\0\M-}\M^?\0\0\^D\0\^C\0\^E\0\a\0\^E\0\^E\0\M-{\M^?\
\M-~\M^?\^F\0
\0\b\0\M^?\M^?\M-|\M^?\M-|\M^?\M-}\M^?\0\0\f\0\v\0
\0\^D\0\^B\0\
\^B\0\0\0\^E\0\^E\0\^E\0\^E\0\M^?\M^?\0\0\^F\0\^C\0\^D\0\a\0\^E\0\^E\0\
\^B\0"
3121 hapitest RET read 94/0x5e
3121 hapitest CALL ioctl(0x5,SNDCTL_DSP_GETISPACE,0x85e385c)
3121 hapitest RET ioctl 0
3121 hapitest CALL read(0x5,0x8563e48,0)
3121 hapitest GIO fd 5 read 0 bytes
""
3121 hapitest RET read 0
3121 hapitest CALL ioctl(0x5,SNDCTL_DSP_GETISPACE,0x85e385c)
3121 hapitest RET ioctl 0
3121 hapitest CALL read(0x5,0x8563e48,0x22)
3121 hapitest GIO fd 5 read 34 bytes
"\M-~\M^?\^B\0\^C\0\^B\0\M-~\M^?\M-}\M^?\^E\0\^F\0\^D\0\b\0
\0\^D\
\0\M-}\M^?\M-~\M^?\^C\0\M-|\M^?\M-~\M^?"
3121 hapitest RET read 34/0x22
3121 hapitest CALL ioctl(0x5,SNDCTL_DSP_GETISPACE,0x85e385c)
3121 hapitest RET ioctl 0
3121 hapitest CALL read(0x5,0x8563e6a,0x44)
3121 hapitest GIO fd 5 read 68 bytes
"\^E\0\^D\0\M-~\M^?\^B\0\^E\0\^C\0
\0\^B\0\M-z\M^?\M-z\M^?\^B\0\v\0
\0\^A\0\M^?\M^?\^A\0\^E\0\^A\0\
\0\0\^C\0\M^?\M^?\^C\0\^F\0\0\0\M-z\M^?\M-z\M^?\M-~\M^?\^E\0\M^?\M^?\
\M-~\M^?\^B\0\M-}\M^?\0\0\^C\0"
3121 hapitest RET read 68/0x44
3121 hapitest CALL ioctl(0x5,SNDCTL_DSP_GETISPACE,0x85e385c)
3121 hapitest RET ioctl 0
3121 hapitest CALL read(0x5,0x8563eae,0x88)
3121 hapitest GIO fd 5 read 136 bytes
"\^D\0\M-v\M^?\M-u\M^?\M-z\M^?\M-z\M^?\^D\0\^E\0\M^?\M^?\M^?\M^?\^A\0\0\
\0\M-{\M^?\M^?\M^?\^D\0\^C\0\^F\0\v\0\a\0\M-~\M^?\M-z\M^?\M-|\M^?\0\0\
\M-}\M^?\M-y\M^?\M-x\M^?\M-z\M^?\M-}\M^?\M-~\M^?\M-z\M^?\M-x\M^?\^A\0\
\M-}\M^?\M-{\M^?\M-}\M^?\M-{\M^?\M-~\M^?\M-~\M^?\0\0\M^?\M^?\M-}\M^?\
\M-|\M^?\M^?\M^?\0\0\M-|\M^?\M-}\M^?\M-~\M^?\^C\0\a\0\^E\0\^A\0\M-~\
\M^?\^A\0\^F\0\^E\0\^B\0\M^?\M^?\M^?\M^?\M-|\M^?\M-z\M^?\M-{\M^?\M-|\
\M^?\^C\0\^E\0\0\0\M-}\M^?\M^?\M^?\M-~\M^?\M-|\M^?"
3121 hapitest RET read 136/0x88
3121 hapitest CALL ioctl(0x5,SNDCTL_DSP_GETISPACE,0x85e385c)
3121 hapitest RET ioctl 0
3121 hapitest CALL read(0x5,0x8563f36,0x110)
3121 hapitest GIO fd 5 read 272 bytes
lots of little reads...
now THEORETICALLY the program should read the same data either way and
it
should work the same, but in practice, it no-longer recognises
any speech.
Is there a sound guru who can give as an easy way to make it behave like
before
as we want to see if that is making a difference..
Kernel patches are ok, but since we can't modify the program
a suggestion of "make the program do XXX" doesn't help us.
The hardware is:
[kxt1a] /dev <21> cat sndstat.bak0
FreeBSD Audio Driver (newpcm) Oct 11 2001 16:19:03
Installed devices:
pcm0: <Creative CT5880-A> at io 0x1440 irq 9 (1p/1r/0v channels duplex)
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
