[email protected] wrote:
>
>
> The patch titled
>
> v4l-720-alsa-support-for-saa7134-that-should-work-fix
>
> has been added to the -mm tree. Its filename is
>
> v4l-720-alsa-support-for-saa7134-that-should-work-fix.patch
>
>
> From: Andrew Morton <[email protected]>
>
> *** Warning: "saa7134_irq_alsa_done" [drivers/media/video/saa7134/saa7134.ko] undefined!
> *** Warning: "alsa_card_saa7134_exit" [drivers/media/video/saa7134/saa7134.ko] undefined!
> *** Warning: "alsa_card_saa7134_create" [drivers/media/video/saa7134/saa7134.ko] undefined!
>
> Cc: Ricardo Cerqueira <[email protected]>
> Cc: Nickolay V. Shmyrev <[email protected]>
> Cc: Mauro Carvalho Chehab <[email protected]>
> Signed-off-by: Andrew Morton <[email protected]>
Well that didn't work. The problem is that
drivers/media/video/saa7134/saa7134-alsa.c doesn't appear to be wired up
into the build system - it simply doesn't get compiled.
Please send a fix against next -mm?
>
> drivers/media/video/saa7134/saa7134-alsa.c | 3 +++
> 1 files changed, 3 insertions(+)
>
> diff -puN drivers/media/video/saa7134/saa7134-alsa.c~v4l-720-alsa-support-for-saa7134-that-should-work-fix drivers/media/video/saa7134/saa7134-alsa.c
> --- devel/drivers/media/video/saa7134/saa7134-alsa.c~v4l-720-alsa-support-for-saa7134-that-should-work-fix 2005-11-05 23:43:39.000000000 -0800
> +++ devel-akpm/drivers/media/video/saa7134/saa7134-alsa.c 2005-11-05 23:43:39.000000000 -0800
> @@ -220,6 +220,7 @@ static int dsp_buffer_init(struct saa713
> return err;
> return 0;
> }
> +EXPORT_SYMBOL(saa7134_irq_alsa_done);
>
>
> static int snd_card_saa7134_pcm_prepare(snd_pcm_substream_t * substream)
> @@ -811,6 +812,7 @@ __nodev:
> kfree(card);
> return err;
> }
> +EXPORT_SYMBOL(alsa_card_saa7134_create);
>
> void alsa_card_saa7134_exit(void)
> {
> @@ -819,3 +821,4 @@ void alsa_card_saa7134_exit(void)
> snd_card_free(snd_saa7134_cards[idx]);
> }
> }
> +EXPORT_SYMBOL(alsa_card_saa7134_exit);
> _
>
> Patches currently in -mm which might be from [email protected] are
>
> fec_8xx-build-fix.patch
> posix-timers-smp-race-condition-tidy.patch
> increase-maximum-kmalloc-size-to-256k-fix.patch
> git-acpi-pciehprm_acpi-fix.patch
> git-acpi-update-8250_acpi.patch
> pnpacpi-handle-address-descriptors-in-_prs-fix-for-git-acpi-change.patch
> git-agpgart.patch
> git-alsa.patch
> git-blktrace-fixup.patch
> git-cpufreq.patch
> git-drm-prep.patch
> pci-gart-fix.patch
> git-audit-audit_inode_context-warning-fix.patch
> git-audit-selinux_inode_xattr_getsuffix-warning-fix.patch
> git-input.patch
> git-netdev-all-ieee80211_get_payload-warning-fix.patch
> s2io-warning-fixes.patch
> gregkh-pci-pci-driver-owner-removal-intel-agp-fix.patch
> gregkh-pci-pci-driver-owner-removal-ali-agp-fix.patch
> gregkh-pci-pci-driver-owner-removal-more-agp-borkage.patch
> gregkh-pci-pci-driver-owner-removal-even-more-agp-borkage.patch
> gregkh-pci-pci-driver-owner-removal-even-more-even-more-agp-borkage.patch
> gregkh-pci-pci-driver-owner-removal-fix-lpfc.patch
> gregkh-pci-pci-driver-owner-removal-fix-aic94xx_init.patch
> gregkh-usb-usb-pm-09-fix.patch
> eagle-and-adi-930-usb-adsl-modem-driver-tidies.patch
> dma32-change-zones_shift-back-to-2-gfp_zonemask-too.patch
> x86_64-vect-share-build-fix.patch
> slab-dont-bug-on-duplicated-cache.patch
> mm-poison-struct-page-for-ptlock-rename-private.patch
> mm-implement-swap-prefetching-default-y.patch
> mm-implement-swap-prefetching-tweaks.patch
> swap-migration-v5-lru-operations-tweaks.patch
> irda-donauboe-locking-fix.patch
> acx1xx-wireless-driver-usb-is-bust.patch
> acx1xx-allow-modular-build.patch
> acx1xx-wireless-driver-spy_offset-went-away.patch
> x86_64-div-by-zero-fix.patch
> serial-console-touch-nmi-watchdog.patch
> clear_buffer_uptodate-in-discard_buffer-check.patch
> write_inode_now-write-inode-if-not-bdi_cap_no_writeback.patch
> smbfs-readdir-vs-signal-fix.patch
> process-events-connector-fixes.patch
> rcu-signal-handling-tidies.patch
> additional-catchup-rcu-signal-fixes-for-mm-warning-fix.patch
> readahead-commentary.patch
> ext3_readdir-use-generic-readahead.patch
> slob-introduce-the-slob-allocator-fixes.patch
> cpuset-memory-pressure-meter-gcc-295-fix.patch
> gregkh-pci-pci-driver-owner-removal-synclink_gt-fix.patch
> changing-config_localversion-rebuilds-too-much-for-no-good-reason-ipw2200-fix.patch
> remove-ext2-xattr-permission-checks-warning-fixes.patch
> remove-xfs-xattr-permission-checks-warning-fixes.patch
> replace-inode_update_time-with-file_update_time-switch-ntfs-to-touch_atime.patch
> implement-kmap_atomic_irqsave.patch
> edac-atomic-scrub-operations-fix.patch
> edac-core-edac-support-code-ifdef-warnings.patch
> edac-core-edac-support-code-fixes.patch
> edac-core-edac-support-code-edac_mc_scrub_block-kunmap_atomic-fix.patch
> edac-core-edac-support-code-edac_mc_scrub_block-kunmap_atomic-fix-2.patch
> parport-constification-fix.patch
> kprobes-use-rcu-for-unregister-synchronization-base-changes-vs-remove-hlist_for_each_rcu-api-convert-existing-use-to-hlist_for_each_entry_rcu.patch
> dlm-use-configfs-fix.patch
> dvb-usb-urb-printk-fix.patch
> v4l-720-alsa-support-for-saa7134-that-should-work-fix.patch
> sched-disable-preempt-in-idle-tasks-2-powerpc-fixes.patch
> sched-resched-and-cpu_idle-rework-warning-fix.patch
> reiser4-only.patch
> reiser4-swsusp-build-fix.patch
> reiser4-printk-warning-fix.patch
> reiser4-mm-remove-pg_highmem-fix.patch
> reiser4-page-private-fixes.patch
> reiser4-big-update-div64-fix.patch
> reiser4-remove-c99isms.patch
> reiser4-big-update-update_atime-fixes.patch
> reiser4_releasepage-gfp_t-fixes.patch
> ide-promise-flushing-hang-fix.patch
> nvidiafb-fix-mode-setting-ppc-support-warning-fixes.patch
> nr_blockdev_pages-in_interrupt-warning.patch
> sysfs-crash-debugging.patch
> device-suspend-debug.patch
> ide-kmalloc-memset-kzalloc-conversion-fix.patch
> tty-layer-buffering-revamp-pmac_zilog-warning-fix.patch
>
> -
> To unsubscribe from this list: send the line "unsubscribe mm-commits" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
On Sun, 2005-11-06 at 00:12 -0800, Andrew Morton wrote:
> Well that didn't work. The problem is that
> drivers/media/video/saa7134/saa7134-alsa.c doesn't appear to be wired
> up into the build system - it simply doesn't get compiled.
>
> Please send a fix against next -mm?
Also please send all ALSA related patches to
[email protected] for review.
Lee
Lee,
Em Dom, 2005-11-06 ?s 13:33 -0500, Lee Revell escreveu:
> On Sun, 2005-11-06 at 00:12 -0800, Andrew Morton wrote:
> > Well that didn't work. The problem is that
> > drivers/media/video/saa7134/saa7134-alsa.c doesn't appear to be wired
> > up into the build system - it simply doesn't get compiled.
> >
> > Please send a fix against next -mm?
>
> Also please send all ALSA related patches to
> [email protected] for review.
I'm sending you enclosed saa7134-alsa patch. To make easier to
understand, I've merged all stuff. This is highly dependent of the other
saa7134 parts, since PCI stuff are common to both video and audio
funcion on this device.
This is meant to replace saa7134-oss (after more tests) that,
currently, is part of saa7134 module.
>
> Lee
>
>
Cheers,
Mauro.
Em Dom, 2005-11-06 ?s 00:12 -0800, Andrew Morton escreveu:
> [email protected] wrote:
> Well that didn't work. The problem is that
> drivers/media/video/saa7134/saa7134-alsa.c doesn't appear to be wired up
> into the build system - it simply doesn't get compiled.
In fact, it was meant to be part of saa7134 module. We've changed it on the newer patchsets I've sent to you today. It shoud compile well now.
> Please send a fix against next -mm?
Only after sending I noticed you've released -mm1 (I'm suffering a long
delay on my current ISP - maybe some greylist policy). I'll apply the
patches I sent you today against -mm and, if something wents wrong, I'll
send you some correcting patches.
Cheers,
Mauro.
On Mon, 2005-11-07 at 04:26 -0200, Mauro Carvalho Chehab wrote:
> Lee,
>
> Em Dom, 2005-11-06 ?s 13:33 -0500, Lee Revell escreveu:
> > On Sun, 2005-11-06 at 00:12 -0800, Andrew Morton wrote:
> > > Well that didn't work. The problem is that
> > > drivers/media/video/saa7134/saa7134-alsa.c doesn't appear to be wired
> > > up into the build system - it simply doesn't get compiled.
> > >
> > > Please send a fix against next -mm?
> >
> > Also please send all ALSA related patches to
> > [email protected] for review.
>
> I'm sending you enclosed saa7134-alsa patch. To make easier to
> understand, I've merged all stuff. This is highly dependent of the other
> saa7134 parts, since PCI stuff are common to both video and audio
> funcion on this device.
> This is meant to replace saa7134-oss (after more tests) that,
> currently, is part of saa7134 module.
OK, a brief review:
- Why couldn't you use ALSA's DMA API?
- The DMA must be stopped and started in the trigger callback, not the
prepare callback.
- If this device lacks a volume control alsa-lib can emulate it in
software, just create a proper /usr/share/alsa/cards/your_card.conf
file.
- By ALSA convention the acceptable formats, sample rates, etc should
be directly defined in the snd_pcm_hardware_t structure.
- dev->oss needs to go.
Lee
At Mon, 07 Nov 2005 10:33:35 -0500,
Lee Revell wrote:
>
> On Mon, 2005-11-07 at 04:26 -0200, Mauro Carvalho Chehab wrote:
> > Lee,
> >
> > Em Dom, 2005-11-06 ?s 13:33 -0500, Lee Revell escreveu:
> > > On Sun, 2005-11-06 at 00:12 -0800, Andrew Morton wrote:
> > > > Well that didn't work. The problem is that
> > > > drivers/media/video/saa7134/saa7134-alsa.c doesn't appear to be wired
> > > > up into the build system - it simply doesn't get compiled.
> > > >
> > > > Please send a fix against next -mm?
> > >
> > > Also please send all ALSA related patches to
> > > [email protected] for review.
> >
> > I'm sending you enclosed saa7134-alsa patch. To make easier to
> > understand, I've merged all stuff. This is highly dependent of the other
> > saa7134 parts, since PCI stuff are common to both video and audio
> > funcion on this device.
> > This is meant to replace saa7134-oss (after more tests) that,
> > currently, is part of saa7134 module.
>
> OK, a brief review:
>
> - Why couldn't you use ALSA's DMA API?
This is OK, IMO. Basically, it's up to the driver.
> - The DMA must be stopped and started in the trigger callback, not the
> prepare callback.
>
> - If this device lacks a volume control alsa-lib can emulate it in
> software, just create a proper /usr/share/alsa/cards/your_card.conf
> file.
>
> - By ALSA convention the acceptable formats, sample rates, etc should
> be directly defined in the snd_pcm_hardware_t structure.
>
> - dev->oss needs to go.
Agreed about the rest.
Takashi
Takashi Iwai <[email protected]> wrote:
>
> > OK, a brief review:
> >
> > - Why couldn't you use ALSA's DMA API?
>
> This is OK, IMO. Basically, it's up to the driver.
>
>
> > - The DMA must be stopped and started in the trigger callback, not the
> > prepare callback.
> >
> > - If this device lacks a volume control alsa-lib can emulate it in
> > software, just create a proper /usr/share/alsa/cards/your_card.conf
> > file.
> >
> > - By ALSA convention the acceptable formats, sample rates, etc should
> > be directly defined in the snd_pcm_hardware_t structure.
> >
> > - dev->oss needs to go.
>
> Agreed about the rest.
Thanks, guys.
Despite all that, I am inclined to merge this patch into 2.6.15. Because
it's in the middle of a 192-patch series and we do need to get the v4l tree
synced up.
Mauro&co, if we do that, do you think the above points can be addressed
inside the next four weeks or so?
On Mon, 2005-11-07 at 11:48 -0800, Andrew Morton wrote:
> Despite all that, I am inclined to merge this patch into 2.6.15.
> Because it's in the middle of a 192-patch series and we do need to get
> the v4l tree synced up.
>
I think starting the DMA in the prepare callbacks is a major problem.
It should be easy to fix though.
Lee
Andrew,
Em Seg, 2005-11-07 ?s 11:48 -0800, Andrew Morton escreveu:
> Takashi Iwai <[email protected]> wrote:
> >
> > > OK, a brief review:
> > >
> > > - Why couldn't you use ALSA's DMA API?
> >
> > This is OK, IMO. Basically, it's up to the driver.
> >
> >
> > > - The DMA must be stopped and started in the trigger callback, not the
> > > prepare callback.
> > >
> > > - If this device lacks a volume control alsa-lib can emulate it in
> > > software, just create a proper /usr/share/alsa/cards/your_card.conf
> > > file.
> > >
> > > - By ALSA convention the acceptable formats, sample rates, etc should
> > > be directly defined in the snd_pcm_hardware_t structure.
> > >
> > > - dev->oss needs to go.
> >
> > Agreed about the rest.
>
> Thanks, guys.
>
> Despite all that, I am inclined to merge this patch into 2.6.15. Because
> it's in the middle of a 192-patch series and we do need to get the v4l tree
> synced up.
>
> Mauro, if we do that, do you think the above points can be addressed
> inside the next four weeks or so?
We are, in fact already working on these. We are about to send you a
newer patch addressing most of the points that Takashi and Lee showed.
You may apply the patches. If they noticed some weird stuff at the
newer patchset, we will have enough time for fix it.
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
Cheers,
Mauro.