Re: RELENG_7 vchans broken Maestro
On Thu, 13 Dec 2007 20:07:39 +0100 Dominic Fandrey [EMAIL PROTECTED] wrote: Ariff Abdullah wrote: On Wed, 12 Dec 2007 22:33:47 +0100 Dominic Fandrey [EMAIL PROTECTED] wrote: Since the switch to BETA4 (it might actually have happened earlier and I didn't recognize) vchans don't work properly anymore (for me). Instead of hearing a second source there only gets a lot of noise mixed into the first audio stream. Try changing AGG_MAXPLAYCH from 4 to 1 (around line 80, sys/dev/sound/pci/maestro.c). That fixes my problem. Will this be changed or will I always have to do this myself? It will be changed. In the meantime, try playing with the above #def, say like setting it to 2 or 3. -- Ariff Abdullah FreeBSD ... Recording in stereo is obviously too advanced and confusing for us idiot * users :P pgpR3sYRsrIMA.pgp Description: PGP signature
Re: RELENG_7 vchans broken Maestro
On Wed, 12 Dec 2007 22:33:47 +0100 Dominic Fandrey [EMAIL PROTECTED] wrote: Since the switch to BETA4 (it might actually have happened earlier and I didn't recognize) vchans don't work properly anymore (for me). Instead of hearing a second source there only gets a lot of noise mixed into the first audio stream. Try changing AGG_MAXPLAYCH from 4 to 1 (around line 80, sys/dev/sound/pci/maestro.c). # uname -a FreeBSD homeKamikaze.norad 7.0-BETA4 FreeBSD 7.0-BETA4 #0: Wed Dec 12 01:30:18 CET 2007 [EMAIL PROTECTED]:/usr/obj/VECTRA-7/i386/usr/src/sys/VECTRA-7 i386 # cat /dev/sndstat FreeBSD Audio Driver (newpcm: 32bit 2007061600/i386) Installed devices: pcm0: ESS Technology Maestro-2E port 0x5400-0x54ff irq 11 at device 11.0 on pci2 [MPSAFE] (4p:1v/1r:1v channels duplex default) # sysctl -a | grep vchan hw.snd.maxautovchans: 8 dev.pcm.0.play.vchans: 1 dev.pcm.0.play.vchanrate: 48000 dev.pcm.0.play.vchanformat: s16le dev.pcm.0.rec.vchans: 1 dev.pcm.0.rec.vchanrate: 48000 dev.pcm.0.rec.vchanformat: s16le -- Ariff Abdullah FreeBSD ... Recording in stereo is obviously too advanced and confusing for us idiot * users :P pgpaL6CuQXu3X.pgp Description: PGP signature
Re: RELENG_7 jerky mouse and skipping sound (still a problem -BETA3)
On Sat, 1 Dec 2007 12:35:23 +0100 Oliver Peter [EMAIL PROTECTED] wrote: On Mon, Nov 19, 2007 at 10:22:33AM -0800, Kip Macy wrote: On Nov 19, 2007 9:53 AM, Anish Mistry [EMAIL PROTECTED] wrote: On Wednesday 07 November 2007, Anish Mistry wrote: On Monday 05 November 2007, [LoN]Kamikaze wrote: Marc Fonvieille wrote: On Thu, Oct 18, 2007 at 05:53:47PM +0200, [LoN]Kamikaze wrote: Anish Mistry wrote: On Thursday 18 October 2007, Marc Fonvieille wrote: On Wed, Oct 17, 2007 at 12:28:30PM -0400, Anish Mistry wrote: I just updated to RELENG_7 from 6.2 and I'm running into some really annoying issues with jerky mouse movement and skipping sound. This seems to be similar to: Re: SCHED_4BSD in RELENG_7 disturbs workflow This happens both with 4BSD and ULE. I seems to happen when I'm compiling ports and a new cc/bzip2/sh process fires off (I'm just watching top), I'll get the skip/freezeup. [...] Using ULE and UP kernel (i.e. without SMP etc.) helped a bit the things but it's still very annoying to use firefox during ports build. I see this lag/freeze on all boxes I use with 7.0, but it's true that with a fast machine people can ignore the problem, it's less obvious than with a 1GHz box for example. Yeah, I'm still seeing this behavior. Does anyone have suggestions on debugging? Thanks, I did post the solution in this thread. It has nothing to do with the mouse. Does the problem persist for you? It's gone for me, even with moused. Yes, the problem seems to have been fixed. I'm back to kern.hz=1000 and removed FULL_PREEMPTION. No skipping. It looks like I spoke too soon. I've just tried to compile miro and as it was compiling the boost-python dependency I noticed the problem again. Switching kern.hz=100 seems to fix the problem. Can any of the developers in this area reproduce the issue? It's pretty easy to reproduce on my 1.33Ghz Athlon. There is an ithread priority inversion bug that might be causing this. The fix for that should be going in shortly. I have a similar problem under 7.0-BETA3 with SCHED_ULE and snd_ich. Sound is skipping with/without load every few seconds. kern.hz is at 1000 and dev.pcm.0.buffersize: 16384 Try setting hint.pcm.0.highlatency=1 through device.hints [...] Copyright (c) 1992-2007 The FreeBSD Project. Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 The Regents of the University of California. All rights reserved. FreeBSD is a registered trademark of The FreeBSD Foundation. FreeBSD 7.0-BETA3 #0: Fri Nov 30 11:16:35 CET 2007 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/DOGBE RT [...] -- Ariff Abdullah FreeBSD ... Recording in stereo is obviously too advanced and confusing for us idiot * users :P pgp8TUOvQFVcY.pgp Description: PGP signature
Re: Maestro 2E on 6.3-PRERELEASE
On Fri, 23 Nov 2007 20:20:08 -0600 Greg Miller [EMAIL PROTECTED] wrote: I have a Maestro 2E sound card that works perfectly on 6.2-RELEASE-p8, but not on 6.3-PRERELEASE. Here's the dmesg: Could be something else (acpi? bus changes? etc). The sound driver stuffs virtually unchanged. [...] pcm1: ESS Technology Maestro-2E port 0xb000-0xb0ff irq 23 at device 2.0 on pci1 pcm1: agg_rdcodec() RW_DONE timed out. pcm1: will perform cold reset. pcm1: agg_rdcodec() RW_DONE timed out. pcm1: agg_rdcodec() RW_DONE timed out. pcm1: agg_rdcodec() RW_DONE timed out. pcm1: agg_rdcodec() RW_DONE timed out. pcm1: agg_rdcodec() RW_DONE timed out. pcm1: ac97 codec invalid or not present (id == ) pcm1: mixer initialization failed! device_attach: pcm1 attach returned 6 [...] pcm0: HDA Codec: Realtek ALC882 pcm0: HDA Driver Revision: 20071020_0048 Why not using this one? [...] SMP: AP CPU #1 Launched! acd0: FAILURE - INQUIRY ILLEGAL REQUEST asc=0x24 ascq=0x00 afd0: FAILURE - INQUIRY ILLEGAL REQUEST asc=0x24 ascq=0x00 cd0 at ata0 bus 0 target 0 lun 0 cd0: HL-DT-ST RW/DVD GCC-4120B 2.00 Removable CD-ROM SCSI-0 device cd0: 16.000MB/s transfers cd0: cd present [452 x 2048 byte records] da0 at ata0 bus 0 target 1 lun 0 da0: IOMEGA ZIP 100 23.D Removable Direct Access SCSI-0 device da0: 3.300MB/s transfers da0: Attempt to query device size failed: NOT READY, Medium not present Trying to mount root from ufs:/dev/ad4s2a -- Ariff Abdullah FreeBSD ... Recording in stereo is obviously too advanced and confusing for us idiot * users :P pgpehfdDdW4MZ.pgp Description: PGP signature
Re: snd_t4dwave(4) broken in RELENG_7?
On Thu, 8 Nov 2007 14:46:51 +0100 Sascha Klauder [EMAIL PROTECTED] wrote: Hi all, can someone verify whether snd_t4dwave is working in RELENG_7? After upgrading my 6.2-STABLE (from March 2006) to 7.0-BETA2, I get pcm0:play:dsp0.p1: play interrupt timeout, channel dead errors a few seconds after playback started. It's a HP nx9005 laptop, so I'm not sure if it could be ACPI- related, but sound was working fine ever since I installed 5.2 on it. Disabling ACPI makes no difference. Verbose dmesg available here: http://evenstar.shopkeeper.de/~sascha/nx9005/misc/dmesg-7-verbose (actually still from a -CURRENT shortly before the branch) Try changing #define TR_MAXPLAYCH from 4 to 1, in sys/dev/sound/pci/t4dwave.c The driver is virtually unchanged ever since 5.x . I'm at lost (no hardware to test). -- Ariff Abdullah FreeBSD ... Recording in stereo is obviously too advanced and confusing for us idiot * users :P pgpcQsPxPSNeK.pgp Description: PGP signature
Re: snd_t4dwave(4) broken in RELENG_7?
On Thu, 8 Nov 2007 16:04:08 +0100 Sascha Klauder [EMAIL PROTECTED] wrote: On Thu, Nov 08, 2007 at 10:43:06PM +0800, Ariff Abdullah wrote: On Thu, 8 Nov 2007 14:46:51 +0100 Sascha Klauder [EMAIL PROTECTED] wrote: Verbose dmesg available here: http://evenstar.shopkeeper.de/~sascha/nx9005/misc/dmesg-7-verbose (actually still from a -CURRENT shortly before the branch) Try changing #define TR_MAXPLAYCH from 4 to 1, in sys/dev/sound/pci/t4dwave.c Thanks, that did help! Probably a hidden bug. Not all hardwares capable of driving multiple dacs. The driver is virtually unchanged ever since 5.x . I'm at lost (no hardware to test). I could provide ssh access to my system, if that would help. Thanks for the offer, but I'm kind of busy right now. Maybe after a week or two from now, I'll contact you again. -- Ariff Abdullah FreeBSD ... Recording in stereo is obviously too advanced and confusing for us idiot * users :P pgpsvtaNtACRM.pgp Description: PGP signature
Re: snd_t4dwave(4) broken in RELENG_7?
On Fri, 9 Nov 2007 00:13:20 +0800 Ariff Abdullah [EMAIL PROTECTED] wrote: On Thu, 8 Nov 2007 16:04:08 +0100 Sascha Klauder [EMAIL PROTECTED] wrote: On Thu, Nov 08, 2007 at 10:43:06PM +0800, Ariff Abdullah wrote: On Thu, 8 Nov 2007 14:46:51 +0100 Sascha Klauder [EMAIL PROTECTED] wrote: Verbose dmesg available here: http://evenstar.shopkeeper.de/~sascha/nx9005/misc/dmesg-7-verbose (actually still from a -CURRENT shortly before the branch) Try changing #define TR_MAXPLAYCH from 4 to 1, in sys/dev/sound/pci/t4dwave.c Thanks, that did help! Probably a hidden bug. Not all hardwares capable of driving multiple dacs. The driver is virtually unchanged ever since 5.x . I'm at lost (no hardware to test). I could provide ssh access to my system, if that would help. Thanks for the offer, but I'm kind of busy right now. Maybe after a week or two from now, I'll contact you again. Try this one: http://people.freebsd.org/~ariff/test/t4dwave.c -- Ariff Abdullah FreeBSD ... Recording in stereo is obviously too advanced and confusing for us idiot * users :P pgprGWePkHLXG.pgp Description: PGP signature
Re: mini-heads-up: linuxulator fixes (Fw: cvs commit: src/sys/amd64/linux32 linux.h linux32_machdep.c linux32_sysvec.c syscalls.master src/sys/compat/linprocfs linprocfs.c src/sys/c
On Sun, 8 Jul 2007 14:49:47 +0200 Alexander Leidinger [EMAIL PROTECTED] wrote: Hi, I MFCed some fixes to the linuxulator today. One major change is correcting the handling of the linux mmap function. This may break some not so good programmed audio programs (mostly some few bad behaving games). If you get hit by this, don't be afraid. The MFC of a change to the sound system (and an entry to UPDATING) will be done soon by Ariff. Technically, this is not the fault of that particular audio apps. Latest mmap changes in linuxlator apparently force PROT_EXEC mapping on all _typical_ mmap invocations, and this affecting not just sound, but bktr(4) as well. Well, bktr a bit lucky since I haven't encounter any linux-binary + bktr apps. Anyway, I've added sysctl hw.snd.compat_linux_mmap, where it is disabled by default. linux mmap breakage: http://lists.freebsd.org/pipermail/freebsd-current/2007-June/073437.html solution: http://lists.freebsd.org/pipermail/cvs-src/2007-July/080531.html Except for those audio apps, all programs which had problems because of this should work better now. If you stumble upon a problem, please report to [EMAIL PROTECTED] Note: while more MFCs of fixes and missing linux syscalls are planned (before 6.3), the MFC of the linux-2.6 compatibility is not planned (if you need it, you have to update to 7.0 release when it comes out). If someone wants to help test some more patches for the linuxulator in -stable, please contact me and I take you into the loop (the more testers, the faster -stable will see the MFCs). What I need are people which compile and runtest the changes. Most of the changes are straight forward, we only have one part which causes a problem because of an unknown reason (either bad testing hardware or a missing change in my patches). -- Ariff Abdullah FreeBSD ... Recording in stereo is obviously too advanced and confusing for us idiot * users :P pgpnvufXDI2DM.pgp Description: PGP signature
Re: Problems with snd_via8233
[CC: -stable, as this doesn't seems like sound specific issues, or so I thought.] On Thu, 5 Jul 2007 19:51:58 +0200 TooMany Secrets [EMAIL PROTECTED] wrote: Hello. First, please, excuse me my bad english. I've got a laptop with FreeBSD 6.2-STABLE. The upgrade from RELEASE to STABLE was made this monday 2 (aprox.). Before the upgrade, sound work correctly, without any problem. But after compile the system (buildworld...) I don't have sound in the laptop. Here is some information: dmesg: pcm0: VIA VT8235 port 0x1000-0x10ff irq 22 at device 17.5 on pci0 pcm0: primary codec not ready (cnt = 0x7d0) device_attach: pcm0 attach returned 6 This is indeed troublesome, since I couldn't recall any major changes to this driver (snd_via8233) for RELENG_6. What release was that, 6.2 or 6.1? What kind of hardware is this, a laptop? Could it be because of ACPI or APIC (try booting by disabling either both). pciconf -lv: [EMAIL PROTECTED]:17:5: class=0x040100 card=0x2032161f chip=0x30591106 rev=0x50 hdr=0x00 vendor = 'VIA Technologies Inc' device = 'VT8233/33A/8235/8237 AC97 Enhanced Audio Controller' class = multimedia subclass = audio cat /dev/sndstat: FreeBSD Audio Driver (newpcm) Installed devices: sysctl -w hw.snd.verbose=2 hw.snd.verbose: 2 - 2 sysctl -a hw.snd hw.snd.report_soft_formats: 1 hw.snd.targetirqrate: 32 hw.snd.verbose: 2 hw.snd.maxautovchans: 4 hw.snd.unit: 0 uname -a FreeBSD leonsio.toomany.net 6.2-STABLE FreeBSD 6.2-STABLE #1: Mon Jul 2 20:29:31 CEST 2007 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/LEONSIO i386 And attached to this mail here it is the complet kernel config. Please, anybody could help me? -- Ariff Abdullah FreeBSD ... Recording in stereo is obviously too advanced and confusing for us idiot * users :P pgpxzkqw1snVd.pgp Description: PGP signature
Re: HDA merge breaks 6-stable
On Mon, 21 May 2007 12:04:32 +1200 Michael Butler [EMAIL PROTECTED] wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 My apologies for the cross-post .. The merge of HDA support seems to have broken my Toshi A105-S4004 laptop. Any attempt at sound output, e.g. logging onto to KDE, freezes it solid requiring a power-cycle to recover. Per the debugging notes ariff@ posted some time back for the RELENG_6 patch, I've attached two dmesg/sysctl outputs, 'sound-good.txt' is prior to the merge, 'sound-bad.txt' is after. Ok, this probably weird, but please try enabling vchans early enough before login/X. You have either hw.snd.maxautovchans or hw.snd.pcm0.vchans to play with. Virtual channels is disabled by default on RELENG_6 due to issues with other drivers, which I intend to fix and enable it as default sooner. -- Ariff Abdullah FreeBSD ... Recording in stereo is obviously too advanced and confusing for us idiot * users :P pgpGdhOmUlHCC.pgp Description: PGP signature
Re: AMD Turion X2 problems -- Major lag issues
On Fri, 04 May 2007 11:35:10 -0400 Russell Doucette [EMAIL PROTECTED] wrote: Hi, I have a gateway MX3410 laptop, and when I try to install, sysinstall hangs or has unimaginable lag time, and then my laptop just sends out this constant system beep at me. Is there a problem with FreeBSD and AMD Turion CPUs? I hear a few other people are having problems with various AMD chipsets, in particular Turions with the same exact problem. Is there a workaround or patch to installing FreeBSD on my Gateway (AMD Turion 64-x2 technology)? Thanks. Try disabling apic (read: apic, not acpi). set hint.apic.0.disabled=1 through loader prompt. The drawback is that other cpu cores will not be enabled. Update to latest -stable if you want to re-enable apic. -- Ariff Abdullah FreeBSD ... Recording in stereo is obviously too advanced and confusing for us idiot * users :P pgp7LuYOM51eU.pgp Description: PGP signature
Re: PATCH: new acpi embedded controller I/O model
On Mon, 26 Feb 2007 18:20:02 -0800
Nate Lawson [EMAIL PROTECTED] wrote:
If you are having EC timeout problems as in the below PR, please try
the latest EC code. I just committed it in rev 1.69 of acpi_ec.c to
-current. Attached is the patch for 6-stable.
http://www.freebsd.org/cgi/query-pr.cgi?pr=98171
To use it, just recompile your acpi kernel module and load it at
boot: cd /sys/modules/acpi/acpi make cp acpi.ko /
Then at the loader prompt after rebooting:
load /acpi.ko
boot
You should be able to see battery status and thermal settings via
sysctl hw.acpi as normal. Check dmesg for any new errors.
If you notice slower performance or get EC timed out messages on
console, you try increasing these sysctls/tunables:
debug.acpi.ec.timeout
debug.acpi.ec.poll_time
Or turn off this sysctl/tunable, disabling the new burst mode:
debug.acpi.ec.burst=0
To find any performance problems, you'll need to rebuild the kernel
and modules with this added to your kernel config:
options KTR
options KTR_ENTRIES=65536
Then reboot, load this kernel/acpi.ko, use the system for a while to
trigger the problem behavior and generate output:
ktrdump -t | gzip -c ktr.out.gz
This code is pretty well-tested so I expect the only issues we might
see is it not totally fixing some systems that previously didn't
work or needing to add some workaround code for systems that don't
properly support burst mode.
Not quite a success, unless with debug.acpi.ec.burst=0 plus this
additional patch. Even so, the timed out message keep appearing once
in a while though not so frequent as before. In terms of performance,
things are a bit smoother (acpiconf -i 0 works, no longer exhibit
delay or producing timed out messages). Tuning
debug.acpi.ec.timeout/poll_timeout upside-down has no real effect on
eliminating those timed out messages.
Compaq V3000/Turion64 X2.
--
Ariff Abdullah
FreeBSD
... Recording in stereo is obviously too advanced
and confusing for us idiot * users :P
--- sys/dev/acpica/acpi_ec.c Tue Feb 27 19:21:12 2007
+++ sys/dev/acpica/acpi_ec.c Tue Feb 27 19:22:17 2007
@@ -936,6 +936,7 @@
count = ec_poll_time / EC_POLL_DELAY;
if (count = 0)
count = 1;
+slp_ival = max(hz / 1000, 1);
for (i = 0; i count; i++) {
EcStatus = EC_GET_CSR(sc);
if (sc-ec_burstactive (EcStatus EC_FLAG_BURST_MODE) == 0) {
@@ -947,7 +948,15 @@
Status = AE_OK;
break;
}
- AcpiOsStall(EC_POLL_DELAY);
+ if (sc-ec_burstactive)
+ AcpiOsStall(EC_POLL_DELAY);
+ else {
+ if (!cold)
+ msleep(sc-ec_csrvalue, sc-ec_mtx, PZERO, ecpoll,
+ slp_ival);
+ else
+ AcpiOsStall(1000);
+ }
}
/*
pgpqRbRSrOSef.pgp
Description: PGP signature
Re: Recent libxine update seems to cause ICE on 6-STABLE/amd64
On Sun, 18 Feb 2007 14:28:54 +0200 Andy Fawcett [EMAIL PROTECTED] wrote: On Sunday 18 February 2007 11:04, Jack L. wrote: It fails to compile with the gcc version that comes with FreeBSD 6 and 7, but it compiles fine on gcc 4.1 when I tried. It's probably a compiler bug that's been fixed on the 4.x series. Yes, I agree it builds fine on 6.2/amd64 when gcc 4.1 is used. I guess a simple fix for the port would be to update the USE_GCC version to a value known to work (4.1 and 4.2 from ports are ok, no idea about 4.0). Hopefully someone more familiar with the compiler suites would be able to fix the system compilers. How about using default compiler + this patch: http://people.freebsd.org/~ariff/ports/multimedia_libxine/patch-xzz -- Ariff Abdullah FreeBSD ... Recording in stereo is obviously too advanced and confusing for us idiot * users :P pgpehChCyzlgI.pgp Description: PGP signature
Re: Nit in 6-stable
On Sun, 15 Oct 2006 18:03:35 -0200 JoaoBR [EMAIL PROTECTED] wrote: On Sunday 15 October 2006 14:42, Ariff Abdullah wrote: pcm0: Conexant CX20468-21 AC97 Codec * bus_dmamem_alloc failed to align memory properly.bus_dmamem_alloc failed to align memory properly.cd0 at ata1 bus 0 target 0 lun 0 * This DMA misalignment will be fixed before the release. so this is probably the same issue or related? pcm0: Avance Logic ALC655 AC97 Codec bus_dmamem_alloc failed to align memory properly.bus_dmamem_alloc failed to align memory properly. Are you using the same snd_atiixp driver? If true then it is yes. -- Ariff Abdullah FreeBSD ... Recording in stereo is obviously too advanced and confusing for us idiot * users :P pgpri3uP1GL7D.pgp Description: PGP signature
Re: Nit in 6-stable
On Sun, 15 Oct 2006 12:27:32 -0400 Peter Carah [EMAIL PROTECTED] wrote: I saw this come up a couple of weeks ago. Don't know if it is a 6.2 showstopper or not (probably not?). I see the following message come up during boot. Both pcm and cd0 appear to work so I don't know if the message matters or not. Message highlighted by *** below. A truly minor nit: the message is missing its \n. This is a compaq V2310 (AMD64 + ATI) laptop. More nit - the problem with the misidentification of PCI bus 5 subordinate bus is still present. Prevents use of PCcard slot until a pci write workaround is done. End of dmesg: . acpi_acad0: AC Adapter on acpi0 battery0: ACPI Control Method Battery on acpi0 pmtimer0 on isa0 orm0: ISA Option ROMs at iomem 0xc-0xc,0xd-0xd0fff,0xdc000-0xd,0xe-0xe3fff on isa0 sc0: System console at flags 0x100 on isa0 sc0: VGA 16 virtual consoles, flags=0x300 vga0: Generic ISA VGA at port 0x3c0-0x3df iomem 0xa-0xb on isa0 Timecounter TSC frequency 1595371843 Hz quality 800 Timecounters tick every 1.000 msec ad0: 76319MB TOSHIBA MK8025GAS KA024A at ata0-master UDMA100 acd0: DVDR HL-DT-ST DVD-RW GCA-4080N/0C35 at ata1-master PIO4 pcm0: Conexant CX20468-21 AC97 Codec * bus_dmamem_alloc failed to align memory properly.bus_dmamem_alloc failed to align memory properly.cd0 at ata1 bus 0 target 0 lun 0 * This DMA misalignment will be fixed before the release. cd0: HL-DT-ST DVD-RW GCA-4080N 0C35 Removable CD-ROM SCSI-0 device cd0: 16.000MB/s transfers cd0: Attempt to query device size failed: NOT READY, Medium not present Trying to mount root from ufs:/dev/ad0s2a module_register: module pci/radeon already exists! Module pci/radeon failed to register: 17 I've attached the full dmesg in case the server allows attachments. -- Pete -- Ariff Abdullah FreeBSD ... Recording in stereo is obviously too advanced and confusing for us idiot * users :P pgpBpcD5U9TYe.pgp Description: PGP signature
HEADS UP: Last call for snd_hda(4) testers - High Definition Audio driver
[ Please remove current@ and stable@ from your CC: . This mail serves only as an announcement (or a death threat, if you prefer that way). ] Allright folks, I've had enough. It is time to go gold. This driver is proven stable and works (mostly) after several weeks of testing and bug hunting, thanks to those unfortunate unpaid victims at [EMAIL PROTECTED] and few other unsung heroes. If you're using previous driver, please remove it, get a latest/pristine RELENG_6 or -CURRENT, and apply these patches: For RELENG_6: http://people.freebsd.org/~ariff/test/hda_releng6.diff For -CURRENT: http://people.freebsd.org/~ariff/test/hda_current.diff While applying these patches using patch(1), DO NOT forget about -p0 argument. I'm getting tired with people reporting the same old patch failed or compile failed because of this. Stick this into your mind, forever, eternally. The proper way to apply it is like this: patch -d /usr/src -p0 hda_yada.diff You don't have to buildworld, buildkernel or any other sacrificial ritual. Simply cd /usr/src/sys/modules/sound/ make clean cleandir make make install should do the trick. Well, it is up to you, really :) Unfortunately, those who are still stuck with 6.1-RELEASE and earlier had to rely on the binary driver instead. Please grab both sound.ko and snd_hda.ko at http://people.freebsd.org/~ariff/HDA/kmod/ and replace whatever sound.ko you have there. HDA Driver Revision: 20060929_0025 - see ? There are $(( - 25)) more iterations before it reach its equilibrium state. Issues: 1) SPDIF not working - I had to disable it, for now. 2) Multichannel/surround not working - The driver tries to output the sound to all possible path. If you have speakers attach to all of them, chances are it all works, but not in a true sense of multichannel/surround. There are few more works left to do on the upper layer of the sound driver to make it works properly. 3) Recording is broken on few hardwares - As far as I can tell, it should work flawlessly, but not to all. This is a bit tricky to handle, but I'm working on it. 4) Pluging in headset does not mute speakers - This is easy (read #5) 5) Nothing works at all - more like a null driver, isn't it? Please follow the instructions from http://people.freebsd.org/~ariff/HDA/ . The death threat is real :) 6) The driver cause panic, killing my first unborn child - nahh.. I don't believe this. If you're running -CURRENT, the issues are probably elsewhere :) As suggested by netchild@, please report your success or failure like this to [EMAIL PROTECTED]: Success --- Hardware/chipset: Compaq Presario V3000 series http://h10025.www1.hp.com/ewfrf/wc/product?product=3190957lc=encc=usdlc=enlang=encc=us Playback: Works flawlessly Recording: Works flawlessly Specific Issues: None. It works out of the box, including analog CD. Special request: I want to die in peace. Verbose dmesg: pcm0: NVidia MCP51 High Definition Audio Controller mem 0xc000-0xc0003fff irq 10 at device 16.1 on pci0 pcm0: HDA Codec: Conexant Venice pcm0: HDA Driver Revision: 20060929_0025 (optional, you don't have to include those boring and uneventfull kernel noises since the driver already works for you) Failure --- Hardware/chipset: Karipap series http://www.karipap.com/ Playback: NONE! Recording: NONE! Specific Issues: I would rather amaze if this works since it is a food to begin with. Special request: I can donate this to you, but even so, you can buy it at the nearest food stall for an RM 0.50 Verbose dmesg: (please put your verbose dmesg here or I'll send another death threat to your first unborn child) -- Ariff Abdullah FreeBSD ... Recording in stereo is obviously too advanced and confusing for us idiot * users :P pgp1MSEUrHBe6.pgp Description: PGP signature
Re: sound card
On Fri, 01 Sep 2006 21:26:05 +0600 S. M. Ibrahim (Lavlu) [EMAIL PROTECTED] wrote: i am totally new with freeBSD 6.1 I installed it yesterday. Everything goes ok, but problem with sound card, It's a builtin Intl HDA on gigabyte 915G-DU motherboard. I added the line sound_load=YES snd_driver_load=YES snd_cmi_load=YES to /boot/loader.conf but sound is not working. Moreover how to access administrativ app from GUI. Intel HDA isn't supported, yet. A proper driver is in the making. In the meantime, you can try free driver from http://www.opensound.com/ . Stay tune. -- Ariff Abdullah FreeBSD pgpZNmoXNytY5.pgp Description: PGP signature
Re: Sound device reported but no devices created
On Sun, 30 Jul 2006 13:43:52 +1000 Peter Jeremy [EMAIL PROTECTED] wrote: I've been trying to get sound working on a Compaq Armada 1580 with a recent 6-STABLE. With devices 'sound' and 'snd_sbc' built into the kernel and no hints, I get: ESS0004: adding io range 0x250-0x257, size=0x8, align=0 pnpbios: handle 16 device ID ESS0004 (04007316) ESS1878: adding io range 0x220-0x22f, size=0x10, align=0 ESS1878: adding io range 0x388-0x38b, size=0x4, align=0 ESS1878: adding io range 0x330-0x331, size=0x2, align=0 ESS1878: adding irq mask 0x20 ESS1878: adding dma mask 0x2 ESS1878: adding dma mask 0x20 pnpbios: handle 17 device ID ESS1878 (78187316) ... unknown: ESS0004 failed to probe at port 0x250-0x257 on isa0 sbc0: ESS ES1878 at port 0x220-0x22f,0x388-0x38b,0x330-0x331 irq 5 drq 1,5 on isa0 sbc0: [GIANT-LOCKED] unknown: PNP0e03 failed to probe at port 0x3e0-0x3e1 on isa0 Device configuration finished. Note that there's no pcm reported and the only sound-related device is /dev/sndstat. I get the same behaviour if I don't build sound into the kernel and kldload snd_sbc.ko (which autoloads sound.ko). laptop1# sysctl hw.snd.verbose=2 hw.snd.verbose: 1 - 2 laptop1# sysctl hw.snd hw.snd.targetirqrate: 32 hw.snd.report_soft_formats: 1 hw.snd.verbose: 2 hw.snd.unit: 0 hw.snd.maxautovchans: 0 laptop1# cat /dev/sndstat FreeBSD Audio Driver (newpcm) Installed devices: laptop1# I have successfully used sound on a similar laptop with 4.9, where the probes looked like: sbc0: ESS ES1878 at port 0x220-0x22f,0x388-0x38b,0x330-0x331 irq 5 drq 1,5 on isa0 pcm0: ESS 18xx DSP on sbc0 I suspect the problem is that sbc0 is not triggering attachment of pcm0 but I'm uncertain why. Does anyone have any suggestions? Try to disable ACPI, or edit /usr/src/sys/dev/sound/isa/sbc.c and remove acpi module dependency (line 794) -- Ariff Abdullah FreeBSD pgpvAskoCkJWO.pgp Description: PGP signature
acquiring duplicate lock of same type: vnode interlock
I think I've read somewhere about panic during early root mount, fsck etc.. Perhaps this might be related: Full dmesg: http://people.freebsd.org/~ariff/misc/dmesg.boot.amd64 [] acquiring duplicate lock of same type: vnode interlock 1st vnode interlock @ kern/vfs_vnops.c:791 2nd vnode interlock @ kern/vfs_subr.c:2018 KDB: stack backtrace: witness_checkorder() at witness_checkorder+0x4da _mtx_lock_flags() at _mtx_lock_flags+0x4a vrefcnt() at vrefcnt+0x31 null_checkvp() at null_checkvp+0x65 null_lock() at null_lock+0x5b VOP_LOCK_APV() at VOP_LOCK_APV+0x81 vn_lock() at vn_lock+0x6b nullfs_root() at nullfs_root+0x4c vfs_donmount() at vfs_donmount+0x1096 nmount() at nmount+0x82 syscall() at syscall+0x21b Xfast_syscall() at Xfast_syscall+0xa8 --- syscall (378, FreeBSD ELF64, nmount), rip = 0x800679e8c, rsp = 0x7fffe558, rbp = 0x7fffee40 --- -- Ariff Abdullah FreeBSD pgpfKF6YPNouC.pgp Description: PGP signature
Re: Freeze by using sound on -STABLE from today
On Fri, 10 Mar 2006 14:56:21 +0100 Frank Steinborn [EMAIL PROTECTED] wrote: Hi, i just built a 6.1-PRERELEASE and noticed that my system freezes when using mplayer or musicpd. I tried to cat /dev/random /dev/dsp0.0 and it freezes too. The machine has been running 6.0-STABLE before without that problem. Since when it started to break? Last changes against this driver (sys/dev/sound/pci/solo.c) was 2 months ago. Try to reproduce the panic in a clean, single mode state. Things will become much easier to debug if you can provide me with appropriate backtrace after the panic. I wanted to know if someone else had similar issues or can reproduce it, so I could write a PR. Please let me know if you need further informations. Verbose /dev/sndstat: ### ## FreeBSD Audio Driver (newpcm) Installed devices: pcm0: ESS Solo-1E at io 0xd800,0xdc00,0xe000 irq 15 kld snd_solo (1p/1r/4v channels duplex default) -- Ariff Abdullah FreeBSD pgpbLQKhZzsTb.pgp Description: PGP signature
Re: Ferari 4005 64bit Turion
On Sat, 10 Dec 2005 16:56:50 +0200 Maher Mohamed [EMAIL PROTECTED] wrote: I want to know what is the solution to the ATI Driver Since is not ver well suported in freebsd, can anyone share his expirience with Acer Ferrari? If you're talking about SMbus, 3d accelerated graphic, then you're correct. Sound support for ATI just recently added into -current (for RELENG_6, see below), and at least you can have 2d accelerated graphic if you're using xorg-server-snap (and possibily, 3d too). In my experience, it runs quite stable, rock solid and doesn't give much headaches. http://people.freebsd.org/~ariff/acer_ferrari/ -- Ariff Abdullah MyBSD http://www.MyBSD.org.my (IPv6/IPv4) http://staff.MyBSD.org.my (IPv6/IPv4) http://tomoyo.MyBSD.org.my (IPv6/IPv4) ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Sound skipping problems
On Sun, 06 Nov 2005 12:39:02 +0100 Sebastiaan van Erk [EMAIL PROTECTED] wrote: Hi, I have major sound skipping problems on FreeBSD 6.0. I checked the mailing list archives and found a related thread: http://lists.freebsd.org/pipermail/freebsd-current/2005-June/051103.html To quote Jeff Roberson: I have a patch that should greatly improve the sound skipping problems people have under heavy io load. Several people sent me traces that showed the buf daemon running for hundreds of milliseconds with Giant held, which can hold up the pcm code. The patch is available at: http://www.chesapeake.net/~jroberson/flushbuf.diff The problems are definately correlated to io load, however I can't say that I have HEAVY io loads. A simple: # sync;sync;sync; will already cause the sound to skip. I have DMA enabled on all drives, and it seems the above patch is already merged into FreeBSD 6.0-STABLE. This leaves me at a loss, and I don't know what else to try... Does anybody have any ideas of what I could do to solve this problem? Recompile your kernel with options PREEMPTION, and apply this patch: http://people.freebsd.org/~ariff/snd_RELENG_6_0_20051030_058.diff -- Ariff Abdullah MyBSD http://www.MyBSD.org.my (IPv6/IPv4) http://staff.MyBSD.org.my (IPv6/IPv4) http://tomoyo.MyBSD.org.my (IPv6/IPv4) ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Panic: don't do that ... while unloading snd_ich.ko
On Wed, 6 Jul 2005 10:16:59 +0930 Greg 'groggy' Lehey [EMAIL PROTECTED] wrote: On Wednesday, 6 July 2005 at 2:31:37 +0200, Michael Nottebrock wrote: Readily reproducable, 5.4-STABLE as of last week. Kernel messages prior to panic (from memory): pcm0: unregister: mixer busy WARNING: Driver mistake: destroy_dev on 30/3 This is a phkism. It's supposed to tell the developer to fix his code. You'll have to look at the code and work out what's wrong. I don't understand how this could get into STABLE, though. Greg It's a long standing sound driver unloading routine bug, not really phkism. http://staff.mybsd.org.my/skywizard/FreeBSD/sound/RELENG_5/pcm/mixer.c.diff http://staff.mybsd.org.my/skywizard/FreeBSD/sound/RELENG_5/pcm/sound.c.diff http://staff.mybsd.org.my/skywizard/FreeBSD/sound/RELENG_5/pcm/dsp.c.diff will solve the problem. -- Ariff Abdullah MyBSD http://www.MyBSD.org.my (IPv6/IPv4) http://staff.MyBSD.org.my (IPv6/IPv4) http://tomoyo.MyBSD.org.my (IPv6/IPv4) ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Panic: don't do that ... while unloading snd_ich.ko
On Wed, 6 Jul 2005 11:42:41 +0800 Ariff Abdullah [EMAIL PROTECTED] wrote: On Wed, 6 Jul 2005 10:16:59 +0930 Greg 'groggy' Lehey [EMAIL PROTECTED] wrote: On Wednesday, 6 July 2005 at 2:31:37 +0200, Michael Nottebrock wrote: Readily reproducable, 5.4-STABLE as of last week. Kernel messages prior to panic (from memory): pcm0: unregister: mixer busy WARNING: Driver mistake: destroy_dev on 30/3 This is a phkism. It's supposed to tell the developer to fix his code. You'll have to look at the code and work out what's wrong. I don't understand how this could get into STABLE, though. Greg It's a long standing sound driver unloading routine bug, not really phkism. http://staff.mybsd.org.my/skywizard/FreeBSD/sound/RELENG_5/pcm/mixer.c.diff http://staff.mybsd.org.my/skywizard/FreeBSD/sound/RELENG_5/pcm/sound.c.diff http://staff.mybsd.org.my/skywizard/FreeBSD/sound/RELENG_5/pcm/dsp.c.diff will solve the problem. Ah.. sorry. Greg was right. *It's supposed to tell the developer to fix the driver code*. -- Ariff Abdullah MyBSD http://www.MyBSD.org.my (IPv6/IPv4) http://staff.MyBSD.org.my (IPv6/IPv4) http://tomoyo.MyBSD.org.my (IPv6/IPv4) ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: sound module unload panic
On Sat, 28 May 2005 18:18:41 +0200 (CEST) Tomas Pluskal [EMAIL PROTECTED] wrote: Hello, I'm using 5-STABLE and it panics when I try to unload the sound module (for Sound Blaster PCI 128). # kldunload es_137x.ko pcm0: unregister: mixer busy kldunload: can't unload file: Device busy when I do that once more: # kldunload es_137x.ko WARNING: Driver mistake: destroy_dev on 30/3 panic: don't do that My complete dmesg is here: http://plusik.pohoda.cz/dmesg Tomas http://staff.mybsd.org.my/skywizard/FreeBSD/sound/ at least, apply mixer.c.diff and sound.c.diff -- Ariff Abdullah MyBSD http://www.MyBSD.org.my (IPv6/IPv4) http://staff.MyBSD.org.my (IPv6/IPv4) http://tomoyo.MyBSD.org.my (IPv6/IPv4) ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Can't use AC97 sound in 5.4
On Sat, 14 May 2005 21:04:07 +0200 VnPenguin [EMAIL PROTECTED] wrote: Hi list, Just installed 5.4. it works fine except my AC97 on board sound. I added a line in /boot/loader.conf : snd_ich_load=YES But when I run xmms, there is always error like: ** WARNING **: oss_open(): Failed to open audio device (/dev/dsp): No such file or directory Here is info about my sound card: # pciconf -lv | grep -B4 audio [EMAIL PROTECTED]:17:5:class=0x040100 card=0x30051695 chip=0x30591106 rev=0x50 hdr=0x00 vendor = 'VIA Technologies Inc' device = 'VT8233/33A/8235/8237 AC97 Enhanced Audio Controller' class= multimedia subclass = audio use snd_via8233_load=YES instead. -- Ariff Abdullah MyBSD http://www.MyBSD.org.my (IPv6/IPv4) http://staff.MyBSD.org.my (IPv6/IPv4) http://tomoyo.MyBSD.org.my (IPv6/IPv4) ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
