ALSA update

ALSA update

[Jaroslav Kysela]

Linus, please do a

	bk pull http://linux-sound.bkbits.net/linux-sound

This set of changes allows compilation without procfs and fixes next 
problems reported on lkml. The major part of code is reducing stack usage 
in several functions using kmalloc(). Also, USB audio/midi code has been 
updated.

					Jaroslav

This will update the following files:

 include/sound/asound.h                |    3 
 include/sound/core.h                  |    1 
 include/sound/info.h                  |    6 
 include/sound/version.h               |    4 
 sound/core/control.c                  |   34 +-
 sound/core/hwdep.c                    |    8 
 sound/core/info_oss.c                 |    2 
 sound/core/init.c                     |    2 
 sound/core/ioctl32/ioctl32.c          |   76 +++---
 sound/core/ioctl32/ioctl32.h          |   38 +++
 sound/core/ioctl32/pcm32.c            |   52 ++--
 sound/core/memory.c                   |    8 
 sound/core/oss/mixer_oss.c            |  174 +++++++++-----
 sound/core/oss/pcm_oss.c              |    4 
 sound/core/pcm_lib.c                  |   66 ++---
 sound/core/pcm_native.c               |  140 ++++++-----
 sound/core/rawmidi.c                  |   16 -
 sound/core/seq/oss/seq_oss_readq.c    |    7 
 sound/core/seq/oss/seq_oss_synth.c    |    4 
 sound/core/seq/oss/seq_oss_writeq.c   |    4 
 sound/core/seq/seq.c                  |    4 
 sound/core/seq/seq_clientmgr.c        |    8 
 sound/core/seq/seq_fifo.c             |    4 
 sound/core/seq/seq_lock.c             |    2 
 sound/core/seq/seq_lock.h             |    9 
 sound/core/seq/seq_memory.c           |   12 -
 sound/core/sound.c                    |   11 
 sound/core/sound_oss.c                |   12 -
 sound/core/timer.c                    |    4 
 sound/isa/es18xx.c                    |   24 +-
 sound/isa/opl3sa2.c                   |    1 
 sound/isa/sb/emu8000.c                |    2 
 sound/isa/wavefront/wavefront_synth.c |    2 
 sound/pci/Config.help                 |    2 
 sound/pci/Config.in                   |   21 -
 sound/pci/emu10k1/emufx.c             |   27 +-
 sound/pci/ens1370.c                   |    6 
 sound/pci/ymfpci/ymfpci_main.c        |    6 
 sound/usb/usbaudio.c                  |   28 +-
 sound/usb/usbmixer.c                  |  405 ++++++++++++++++++++++------------
 sound/usb/usbmixer_maps.c             |  100 ++++++++
 sound/usb/usbquirks.h                 |   66 ++---
 42 files changed, 922 insertions(+), 483 deletions(-)

through these ChangeSets:

<perex@suse.cz> (02/10/14 1.839)
   ALSA update
     - ES18xx - fixed detection and initialization of opl3 and mpu401
     - ENS1370 - make inclusion of *_codec.h selective for each chipset
     - usb audio
       - unified get_min_max() function to retrieve the min and max values
       - added the debug condition to ignore the error at get/put callbacks
       - quirks - use USB_DEVICE without class interface

<perex@suse.cz> (02/10/14 1.838)
   ALSA update
     - reduced stack usage for functions using >1024+ bytes
     - fixed behaviour when OSS emulation is selected
     - fix in kmod support for sequencer

<perex@suse.cz> (02/10/13 1.837)
   ALSA update
     - compilation fixes without CONFIG_PROC_FS

<perex@suse.cz> (02/10/13 1.836)
   ALSA update
     - USB driver update
       - better extdigi support (mixer)
       - cleanups in the audio code

<perex@suse.cz> (02/10/13 1.835)
   ALSA update
     - added SNDRV_PCM_IOCTL_HWSYNC ioctl
     - changed behaviour of read/write/poll functions
       - prefer waiting than return an error code
     - removed snd_seq_sleep_timeout_in_lock and snd_seq_sleep_in_lock helpers

<perex@suse.cz> (02/10/13 1.834)
   ALSA update
     - fixes for compilation without CONFIG_PROC_FS

<perex@suse.cz> (02/10/13 1.833)
   ALSA update
     - reduced stack usage (>1024 bytes) in ioctl32 routines and PCM routines
     - PCM midlevel - fixed drop in release()
     - OPL3SA2 - removed wrong inclusion of <linux/isapnp.h>
     - EMU8000 - fixed compilation when sequencer is not selected
     - Wavefront - fixed compilation for GCC3
     - fixed gameport dependency in pci/Config.in
     - EMU10K1 - fixed icode peek ioctl in emufx()
     - YMFPCI - added FM legacy volume control

-----
Jaroslav Kysela <perex@suse.cz>
Linux Kernel Sound Maintainer
ALSA Project  http://www.alsa-project.org
SuSE Linux    http://www.suse.com

Re: ALSA update

[David S. Miller]

This breaks the build on sparc64:

sound/core/ioctl32/ioctl32.h:95: parse error before `if'
sound/core/ioctl32/ioctl32.h:99: parse error before `sizeof'
sound/core/ioctl32/ioctl32.h:99: `data32' undeclared here (not in a function)
sound/core/ioctl32/ioctl32.h:99: warning: type defaults to `int' in declaration of `CVT_s\ndrv_type'
sound/core/ioctl32/ioctl32.h:99: warning: function declaration isn't a prototype
sound/core/ioctl32/ioctl32.h:99: warning: data definition has no type or storage class
sound/core/ioctl32/ioctl32.h:99: parse error before `}'
sound/core/ioctl32/ioctl32.h:99: warning: type defaults to `int' in declaration of `oldse\g'
sound/core/ioctl32/ioctl32.h:99: incompatible types in initialization
sound/core/ioctl32/ioctl32.h:99: initializer element is not constant
sound/core/ioctl32/ioctl32.h:99: warning: data definition has no type or storage class
sound/core/ioctl32/ioctl32.h:99: parse error before `do'
sound/core/ioctl32/ioctl32.h:99: warning: type defaults to `int' in declaration of `err'
sound/core/ioctl32/ioctl32.h:99: `file' undeclared here (not in a function)
sound/core/ioctl32/ioctl32.h:99: `file' undeclared here (not in a function)
sound/core/ioctl32/ioctl32.h:99: `file' undeclared here (not in a function)
sound/core/ioctl32/ioctl32.h:99: `native_ctl' undeclared here (not in a function)
sound/core/ioctl32/ioctl32.h:99: `data' undeclared here (not in a function)
sound/core/ioctl32/ioctl32.h:99: warning: data definition has no type or storage class
sound/core/ioctl32/ioctl32.h:99: parse error before `if'
sound/core/ioctl32/ioctl32.h:99: warning: type defaults to `int' in declaration of `err'
sound/core/ioctl32/ioctl32.h:99: redefinition of `err'
sound/core/ioctl32/ioctl32.h:99: `err' previously defined here
sound/core/ioctl32/ioctl32.h:99: warning: data definition has no type or storage class
sound/core/ioctl32/ioctl32.h:99: parse error before `if'
sound/core/ioctl32/ioctl32.h:99: `data32' undeclared here (not in a function)
sound/core/ioctl32/ioctl32.h:99: warning: type defaults to `int' in declaration of `CVT_s\ndrv_type'
sound/core/ioctl32/ioctl32.h:99: warning: function declaration isn't a prototype
sound/core/ioctl32/ioctl32.h:99: warning: data definition has no type or storage class
sound/core/ioctl32/ioctl32.h:99: parse error before `}'
sound/core/ioctl32/ioctl32.h:136: warning: This file contains more `}'s than `{'s.
sound/core/ioctl32/ioctl32.c: In function `_snd_ioctl32_ctl_elem_value':
sound/core/ioctl32/ioctl32.c:275: warning: implicit declaration of function `kmalloc_Rsmp\_6f6c1cf7'
sound/core/ioctl32/ioctl32.c:275: `GFP_KERNEL' undeclared (first use in this function)
sound/core/ioctl32/ioctl32.c:275: (Each undeclared identifier is reported only once
sound/core/ioctl32/ioctl32.c:275: for each function it appears in.)
sound/core/ioctl32/ioctl32.c:275: warning: assignment makes pointer from integer without \a cast
sound/core/ioctl32/ioctl32.c:276: warning: assignment makes pointer from integer without \a cast
sound/core/ioctl32/ioctl32.c:290: request for member `value' in something not a structure\ or union
sound/core/ioctl32/ioctl32.c:296: request for member `indirect' in something not a struct\ure or union
sound/core/ioctl32/ioctl32.c:331: request for member `indirect' in something not a struct\ure or union
sound/core/ioctl32/ioctl32.c:362: warning: implicit declaration of function `kfree_Rsmp_0\37a0cba'

Re: ALSA update

[Jaroslav Kysela]

On Mon, 14 Oct 2002, David S. Miller wrote:

> This breaks the build on sparc64:
> 
> sound/core/ioctl32/ioctl32.h:95: parse error before `if'
> sound/core/ioctl32/ioctl32.h:99: parse error before `sizeof'

Oops. Missing two backslashes. It should be corrected with this patch 
(already in linux-sound BK repository):

Index: ioctl32.h
===================================================================
RCS file: /cvsroot/alsa/alsa-kernel/core/ioctl32/ioctl32.h,v
retrieving revision 1.9
retrieving revision 1.10
diff -u -r1.9 -r1.10
--- ioctl32.h	12 Oct 2002 14:18:09 -0000	1.9
+++ ioctl32.h	14 Oct 2002 22:28:27 -0000	1.10
@@ -91,11 +91,11 @@
 	if (data32 == NULL || data == NULL) { \
 		err = -ENOMEM; \
 		goto __end; \
-	}
+	}\
 	if (copy_from_user(data32, (void*)arg, sizeof(*data32))) { \
 		err = -EFAULT; \
 		goto __end; \
-	}
+	}\
 	memset(data, 0, sizeof(*data));\
 	convert_from_32(type, data, data32);\
 	oldseg = get_fs();\

-----
Jaroslav Kysela <perex@suse.cz>
Linux Kernel Sound Maintainer
ALSA Project  http://www.alsa-project.org
SuSE Linux    http://www.suse.com

Re: ALSA update

[David S. Miller]

   From: Jaroslav Kysela <perex@perex.cz>
   Date: Tue, 15 Oct 2002 00:43:53 +0200 (CEST)

   Oops. Missing two backslashes. It should be corrected with this patch 
   (already in linux-sound BK repository):

That's just the tip of the iceberg.

It fails again soon after that, none of the ioctl32.c/pcm32.c
changes were even _compile_ tested.

Try this instead:

--- ./sound/core/ioctl32/pcm32.c.~1~	Mon Oct 14 13:09:55 2002
+++ ./sound/core/ioctl32/pcm32.c	Mon Oct 14 13:12:38 2002
@@ -20,6 +20,7 @@
 
 #include <sound/driver.h>
 #include <linux/time.h>
+#include <linux/slab.h>
 #include <sound/core.h>
 #include <sound/pcm.h>
 #include "ioctl32.h"
@@ -230,7 +231,7 @@ static int _snd_ioctl32_xfern(unsigned i
 	snd_pcm_file_t *pcm_file;
 	snd_pcm_substream_t *substream;
 	struct sndrv_xfern32 data32, *srcptr = (struct sndrv_xfern32*)arg;
-	void *bufs = NULL;
+	void **bufs = NULL;
 	int err = 0, ch, i;
 	u32 *bufptr;
 	mm_segment_t oldseg;
@@ -260,7 +261,7 @@ static int _snd_ioctl32_xfern(unsigned i
 		return -EFAULT;
 	__get_user(data32.bufs, &srcptr->bufs);
 	bufptr = (u32*)TO_PTR(data32.bufs);
-	bufs = kmalloc(sizeof(void *) * 128, GFP_KERNEL)
+	bufs = kmalloc(sizeof(void *) * 128, GFP_KERNEL);
 	if (bufs == NULL)
 		return -ENOMEM;
 	for (i = 0; i < ch; i++) {
@@ -352,8 +353,8 @@ static int _snd_ioctl32_pcm_hw_params_ol
 	mm_segment_t oldseg;
 	int err;
 
-	data32 = kcalloc(sizeof(*data32), GFP_KERNEL);
-	data = kcalloc(sizeof(*data), GFP_KERNEL);
+	data32 = snd_kcalloc(sizeof(*data32), GFP_KERNEL);
+	data = snd_kcalloc(sizeof(*data), GFP_KERNEL);
 	if (data32 == NULL || data == NULL) {
 		err = -ENOMEM;
 		goto __end;
--- ./sound/core/ioctl32/ioctl32.h.~1~	Mon Oct 14 13:01:25 2002
+++ ./sound/core/ioctl32/ioctl32.h	Mon Oct 14 13:12:01 2002
@@ -86,16 +86,16 @@ static int _snd_ioctl32_##type(unsigned 
 	struct sndrv_##type *data;\
 	mm_segment_t oldseg;\
 	int err;\
-	data32 = kcalloc(sizeof(*data32), GFP_KERNEL); \
-	data = kcalloc(sizeof(*data), GFP_KERNEL); \
+	data32 = snd_kcalloc(sizeof(*data32), GFP_KERNEL); \
+	data = snd_kcalloc(sizeof(*data), GFP_KERNEL); \
 	if (data32 == NULL || data == NULL) { \
 		err = -ENOMEM; \
 		goto __end; \
-	}
+	} \
 	if (copy_from_user(data32, (void*)arg, sizeof(*data32))) { \
 		err = -EFAULT; \
 		goto __end; \
-	}
+	} \
 	memset(data, 0, sizeof(*data));\
 	convert_from_32(type, data, data32);\
 	oldseg = get_fs();\
--- ./sound/core/ioctl32/ioctl32.c.~1~	Mon Oct 14 13:02:21 2002
+++ ./sound/core/ioctl32/ioctl32.c	Mon Oct 14 13:03:35 2002
@@ -23,6 +23,7 @@
 #include <linux/smp_lock.h>
 #include <linux/init.h>
 #include <linux/time.h>
+#include <linux/slab.h>
 #include <linux/init.h>
 #include <sound/core.h>
 #include <sound/control.h>
@@ -287,13 +288,13 @@ static int _snd_ioctl32_ctl_elem_value(u
 	data->id = data32->id;
 	data->indirect = data32->indirect;
 	if (data->indirect) /* FIXME: this is not correct for long arrays */
-		data.value.integer.value_ptr = (void*)TO_PTR(data32->value.integer.value_ptr);
+		data->value.integer.value_ptr = (void*)TO_PTR(data32->value.integer.value_ptr);
 	type = get_ctl_type(file, &data->id);
 	if (type < 0) {
 		err = type;
 		goto __end;
 	}
-	if (! data.indirect) {
+	if (! data->indirect) {
 		switch (type) {
 		case SNDRV_CTL_ELEM_TYPE_BOOLEAN:
 		case SNDRV_CTL_ELEM_TYPE_INTEGER:
@@ -328,7 +329,7 @@ static int _snd_ioctl32_ctl_elem_value(u
 	if (err < 0)
 		goto __end;
 	/* restore info to 32bit */
-	if (! data.indirect) {
+	if (! data->indirect) {
 		switch (type) {
 		case SNDRV_CTL_ELEM_TYPE_BOOLEAN:
 		case SNDRV_CTL_ELEM_TYPE_INTEGER:

Re: ALSA update

[Jaroslav Kysela]

On Tue, 15 Oct 2002, David S. Miller wrote:

>    From: Jaroslav Kysela <perex@perex.cz>
>    Date: Tue, 15 Oct 2002 00:43:53 +0200 (CEST)
> 
>    Oops. Missing two backslashes. It should be corrected with this patch 
>    (already in linux-sound BK repository):
> 
> That's just the tip of the iceberg.
> 
> It fails again soon after that, none of the ioctl32.c/pcm32.c
> changes were even _compile_ tested.

Thanks. Applied to linux-sound repository.

						Jaroslav

-----
Jaroslav Kysela <perex@suse.cz>
Linux Kernel Sound Maintainer
ALSA Project  http://www.alsa-project.org
SuSE Linux    http://www.suse.com

Re: ALSA update

[David S. Miller]

   From: Jaroslav Kysela <perex@suse.cz>
   Date: Tue, 15 Oct 2002 21:45:56 +0200 (CEST)

   On Tue, 15 Oct 2002, David S. Miller wrote:
   
   > It fails again soon after that, none of the ioctl32.c/pcm32.c
   > changes were even _compile_ tested.
   
   Thanks. Applied to linux-sound repository.

And what are you going to do in the future to make sure
the sound/ioctl32/ changes you submit in the future even
compile?

Re: ALSA update

[Takashi Iwai]

At Tue, 15 Oct 2002 12:45:40 -0700 (PDT),
David S. Miller <davem@redhat.com> wrote:
> 
>    From: Jaroslav Kysela <perex@suse.cz>
>    Date: Tue, 15 Oct 2002 21:45:56 +0200 (CEST)
> 
>    On Tue, 15 Oct 2002, David S. Miller wrote:
>    
>    > It fails again soon after that, none of the ioctl32.c/pcm32.c
>    > changes were even _compile_ tested.
>    
>    Thanks. Applied to linux-sound repository.
> 
> And what are you going to do in the future to make sure
> the sound/ioctl32/ changes you submit in the future even
> compile?

I'll test before submitting.
this time I had no chance, though...

-- 
Takashi Iwai <tiwai@suse.de>		SuSE Linux AG - www.suse.de
ALSA Developer				ALSA Project - www.alsa-project.org

Re: ALSA update

[Jaroslav Kysela]

On Tue, 15 Oct 2002, Adrian Bunk wrote:

> On Mon, 14 Oct 2002, Jaroslav Kysela wrote:
> 
> > Linus, please do a
> >
> > 	bk pull http://linux-sound.bkbits.net/linux-sound
> >...
> 
> Hi Jaroslav,
> 
> could you make it available as a GNU patch for non-BK users and post the
> URL?

Sure. The latest patch is at
ftp://ftp.alsa-project.org/pub/kernel-patches/alsa-bk-2002-10-15.patch.gz

						Jaroslav

-----
Jaroslav Kysela <perex@suse.cz>
Linux Kernel Sound Maintainer
ALSA Project  http://www.alsa-project.org
SuSE Linux    http://www.suse.com

ALSA update

[Jaroslav Kysela]

Linus, please do a

	bk pull http://linux-sound.bkbits.net/linux-sound

The GNU patch is available at:

	ftp://ftp.alsa-project.org/pub/kernel-patches/alsa-bk-2002-11-10.patch.gz

					Jaroslav

This will update the following files:

 include/sound/core.h                |    6 
 include/sound/cs4231.h              |   45 ++
 include/sound/cs46xx.h              |   83 ---
 include/sound/cs46xx_dsp_spos.h     |   45 +-
 include/sound/info.h                |    3 
 include/sound/pcm.h                 |    4 
 include/sound/pcm_sgbuf.h           |    2 
 include/sound/version.h             |    2 
 sound/core/info.c                   |   19 
 sound/core/init.c                   |  102 +++-
 sound/core/oss/pcm_oss.c            |   10 
 sound/core/pcm_native.c             |   16 
 sound/core/pcm_sgbuf.c              |   24 -
 sound/core/rawmidi.c                |    4 
 sound/core/seq/seq_prioq.c          |    1 
 sound/i2c/l3/uda1341.c              |    4 
 sound/isa/cs423x/cs4231_lib.c       |  247 ++++++++---
 sound/isa/es18xx.c                  |    8 
 sound/isa/sb/emu8000_pcm.c          |   22 -
 sound/pci/ac97/ac97_codec.c         |   31 +
 sound/pci/ac97/ac97_id.h            |    1 
 sound/pci/ac97/ac97_patch.c         |   12 
 sound/pci/ac97/ac97_patch.h         |    1 
 sound/pci/ali5451/ali5451.c         |    9 
 sound/pci/cs46xx/cs46xx.c           |   17 
 sound/pci/cs46xx/cs46xx_lib.c       |  751 +++++++++++++++++++++++++++++-------
 sound/pci/cs46xx/cs46xx_lib.h       |   33 +
 sound/pci/cs46xx/dsp_spos.c         |  125 +++--
 sound/pci/cs46xx/dsp_spos.h         |   25 +
 sound/pci/cs46xx/dsp_spos_scb_lib.c |  326 +++++++++++++--
 sound/pci/ens1370.c                 |  305 ++++++++++----
 sound/pci/ice1712/ak4524.c          |  178 ++++++--
 sound/pci/ice1712/delta.c           |   26 +
 sound/pci/ice1712/delta.h           |    6 
 sound/pci/ice1712/ews.c             |    3 
 sound/pci/ice1712/ice1712.c         |   46 --
 sound/pci/ice1712/ice1712.h         |    6 
 sound/pci/rme9652/hammerfall_mem.c  |    3 
 sound/pci/via82xx.c                 |   56 +-
 sound/sparc/amd7930.c               |    1 
 sound/usb/usbaudio.c                |   76 ++-
 sound/usb/usbaudio.h                |   11 
 sound/usb/usbmidi.c                 |  111 +++--
 43 files changed, 2069 insertions(+), 737 deletions(-)

through these ChangeSets:

<perex@suse.cz> (02/11/10 1.809)
   ALSA update
      - CS4231 - added sparc support to merge sparc/cs4231.c code
      - ICE1712
        - added support for AK4529
        - added support for Midiman M-Audio Delta410
      - USB driver
        - fixed against newer USB API but allow compilation under 2.4

<perex@suse.cz> (02/11/10 1.808)
   ALSA update
      - CS46xx driver
        - DSP is started after initializing AC97 codecs
        - rewrite SPDIF output stuff
        - variable period size support on playback and capture
        - DAC volume mechanism  rewrite
        - IEC958 input volume mechanism rewrite
        - added "AC3 Mode Switch" in mixer
        - code cleanups
      - ENS1371 driver
        - added definitions for the ES1373 chip
        - added code to control IEC958 (S/PDIF) channel status register

<perex@suse.cz> (02/11/10 1.807)
   ALSA update
     - Moved initialization of card->id to card_register() function.
       The new default id is composed from the shortname given by driver.
     - ES18xx - Fixed power management defines
     - VIA82xx - The SG table is build inside hw_params (outside spinlock - memory allocation).

<perex@suse.cz> (02/11/10 1.806)
   ALSA update - small patches
     - added kmalloc_nocheck and vmalloc_nocheck macros
     - PCM
       - the page callback returns 'struct page *'
       - fixed delay function (moved put_user call outside spinlock)
     - OSS PCM emulation
       - fixed read() lock when stream was terminated and no data is available
     - EMU8000
      - added 'can schedule' condition to snd_emu8000_write_wait()
     - AC'97
      - added ALC650 support
     - ALI5451
       - removed double free

-----
Jaroslav Kysela <perex@suse.cz>
Linux Kernel Sound Maintainer
ALSA Project, SuSE Labs

Re: ALSA update

[Greg KH]

> ChangeSet 1.885.1.5, 2002/12/18 10:13:22+01:00, perex@suse.cz

<snip>

> diff -Nru a/sound/usb/usbaudio.c b/sound/usb/usbaudio.c
> --- a/sound/usb/usbaudio.c	Wed Dec 18 10:07:34 2002
> +++ b/sound/usb/usbaudio.c	Wed Dec 18 10:07:34 2002
> @@ -526,7 +526,11 @@
>  /*
>   * complete callback from data urb
>   */
> +#ifndef OLD_USB
>  static void snd_complete_urb(struct urb *urb, struct pt_regs *regs)
> +#else
> +static void snd_complete_urb(struct urb *urb)
> +#endif
>  {
>  	snd_urb_ctx_t *ctx = (snd_urb_ctx_t *)urb->context;
>  	snd_usb_substream_t *subs = ctx->subs;
> @@ -551,7 +555,11 @@
>  /*
>   * complete callback from sync urb
>   */
> +#ifndef OLD_USB
>  static void snd_complete_sync_urb(struct urb *urb, struct pt_regs *regs)
> +#else
> +static void snd_complete_sync_urb(struct urb *urb)
> +#endif
>  {
>  	snd_urb_ctx_t *ctx = (snd_urb_ctx_t *)urb->context;
>  	snd_usb_substream_t *subs = ctx->subs;
> @@ -583,6 +591,9 @@

Ick, you're kidding me, right?  Why do this?  Are you trying to keep a
common code base with 2.4 and 2.5 USB drivers?  If so, I don't recommend
it, as the code will be sprinkled with these ifdef's...

thanks,

greg k-h

Re: ALSA update

[Jaroslav Kysela]

On Wed, 18 Dec 2002, Greg KH wrote:

> > ChangeSet 1.885.1.5, 2002/12/18 10:13:22+01:00, perex@suse.cz
> 
> <snip>
> 
> > diff -Nru a/sound/usb/usbaudio.c b/sound/usb/usbaudio.c
> > --- a/sound/usb/usbaudio.c	Wed Dec 18 10:07:34 2002
> > +++ b/sound/usb/usbaudio.c	Wed Dec 18 10:07:34 2002
> > @@ -526,7 +526,11 @@
> >  /*
> >   * complete callback from data urb
> >   */
> > +#ifndef OLD_USB
> >  static void snd_complete_urb(struct urb *urb, struct pt_regs *regs)
> > +#else
> > +static void snd_complete_urb(struct urb *urb)
> > +#endif
> >  {
> >  	snd_urb_ctx_t *ctx = (snd_urb_ctx_t *)urb->context;
> >  	snd_usb_substream_t *subs = ctx->subs;
> > @@ -551,7 +555,11 @@
> >  /*
> >   * complete callback from sync urb
> >   */
> > +#ifndef OLD_USB
> >  static void snd_complete_sync_urb(struct urb *urb, struct pt_regs *regs)
> > +#else
> > +static void snd_complete_sync_urb(struct urb *urb)
> > +#endif
> >  {
> >  	snd_urb_ctx_t *ctx = (snd_urb_ctx_t *)urb->context;
> >  	snd_usb_substream_t *subs = ctx->subs;
> > @@ -583,6 +591,9 @@
> 
> Ick, you're kidding me, right?  Why do this?  Are you trying to keep a
> common code base with 2.4 and 2.5 USB drivers?  If so, I don't recommend
> it, as the code will be sprinkled with these ifdef's...

Not much. We have 9 #ifdef's and all trying to resolve the conflicts with 
new function prototypes which is difficult to replace with defines or 
inline functions. Perhaps, you'll have an idea to solve this problem.

For us, it's very important to have only one code base for all kernels, 
but on the other hand, we're trying to leave the 2.2/2.4 kernel code 
specific parts separate in our CVS repository if possible.

						Jaroslav

-----
Jaroslav Kysela <perex@suse.cz>
Linux Kernel Sound Maintainer
ALSA Project, SuSE Labs

Re: ALSA update

[Greg KH]

On Wed, Dec 18, 2002 at 08:17:55PM +0100, Jaroslav Kysela wrote:
> 
> Not much. We have 9 #ifdef's and all trying to resolve the conflicts with 
> new function prototypes which is difficult to replace with defines or 
> inline functions. Perhaps, you'll have an idea to solve this problem.

Short of keeping a 2.4 version and a 2.5/2.6 version, no, I do not have
any ideas, sorry.  The USB core has changed a lot between those two
trees, as you know.

> For us, it's very important to have only one code base for all kernels, 
> but on the other hand, we're trying to leave the 2.2/2.4 kernel code 
> specific parts separate in our CVS repository if possible.

Then you might want to leave the OLD_USB stuff out of the 2.5 kernel :)

thanks,

greg k-h

Re: ALSA update

[Takashi Iwai]

At Wed, 18 Dec 2002 11:27:40 -0800,
Greg KH wrote:
> 
> On Wed, Dec 18, 2002 at 08:17:55PM +0100, Jaroslav Kysela wrote:
> > 
> > Not much. We have 9 #ifdef's and all trying to resolve the conflicts with 
> > new function prototypes which is difficult to replace with defines or 
> > inline functions. Perhaps, you'll have an idea to solve this problem.
> 
> Short of keeping a 2.4 version and a 2.5/2.6 version, no, I do not have
> any ideas, sorry.  The USB core has changed a lot between those two
> trees, as you know.

yep, i know :)
but it's also nice to keep the same codebase, because the test
environment of most of users is still on 2.4.
i'll do clean up the codes once when all major bugs are fixed.


Takashi

Re: ALSA update

[Ruslan U. Zakirov]

On Wed, 18 Dec 2002, Jaroslav Kysela wrote:

> On Wed, 18 Dec 2002, Greg KH wrote:
> 
> > > ChangeSet 1.885.1.5, 2002/12/18 10:13:22+01:00, perex@suse.cz
> > 
> > <snip>
> > 
> > > diff -Nru a/sound/usb/usbaudio.c b/sound/usb/usbaudio.c
> > > --- a/sound/usb/usbaudio.c	Wed Dec 18 10:07:34 2002
> > > +++ b/sound/usb/usbaudio.c	Wed Dec 18 10:07:34 2002
> > > @@ -526,7 +526,11 @@
> > >  /*
> > >   * complete callback from data urb
> > >   */
> > > +#ifndef OLD_USB
> > >  static void snd_complete_urb(struct urb *urb, struct pt_regs *regs)
> > > +#else
> > > +static void snd_complete_urb(struct urb *urb)
> > > +#endif
> > >  {
> > >  	snd_urb_ctx_t *ctx = (snd_urb_ctx_t *)urb->context;
> > >  	snd_usb_substream_t *subs = ctx->subs;
> > > @@ -551,7 +555,11 @@
> > >  /*
> > >   * complete callback from sync urb
> > >   */
> > > +#ifndef OLD_USB
> > >  static void snd_complete_sync_urb(struct urb *urb, struct pt_regs *regs)
> > > +#else
> > > +static void snd_complete_sync_urb(struct urb *urb)
> > > +#endif
> > >  {
> > >  	snd_urb_ctx_t *ctx = (snd_urb_ctx_t *)urb->context;
> > >  	snd_usb_substream_t *subs = ctx->subs;
> > > @@ -583,6 +591,9 @@
> > 
> > Ick, you're kidding me, right?  Why do this?  Are you trying to keep a
> > common code base with 2.4 and 2.5 USB drivers?  If so, I don't recommend
> > it, as the code will be sprinkled with these ifdef's...
> 
> Not much. We have 9 #ifdef's and all trying to resolve the conflicts with 
> new function prototypes which is difficult to replace with defines or 
> inline functions. Perhaps, you'll have an idea to solve this problem.
> 
> For us, it's very important to have only one code base for all kernels, 
> but on the other hand, we're trying to leave the 2.2/2.4 kernel code 
> specific parts separate in our CVS repository if possible.
> 
> 						Jaroslav
> 
Hello, Jaroslav and All.
How about other changes in new 2.5 kernel, like new PnP layer (Adam Belay)
or changes with module & boot params (Rusty Russel)? There are now some
changes in 2.5.52 kernel in sound/isa/opl3sa2.c that make this driver not
compatible with other kernels. May be it's better split your tree in
several trees for each version of kernels?
					Ruslan.

Re: ALSA update

[Takashi Iwai]

At Wed, 18 Dec 2002 22:51:27 +0300 (MSK),
Ruslan U. Zakirov <cubic@miee.ru> wrote:
> 
> Hello, Jaroslav and All.
> How about other changes in new 2.5 kernel, like new PnP layer (Adam Belay)
> or changes with module & boot params (Rusty Russel)? There are now some
> changes in 2.5.52 kernel in sound/isa/opl3sa2.c that make this driver not
> compatible with other kernels. May be it's better split your tree in
> several trees for each version of kernels?

if possible, we'll build up some wrapper for 2.4 on alsa-driver (not
the codebase for 2.5) tree.  if not possible, yes, splitting to two
trees would be reasonable for such big changes...

thanks for noticing this issue.  i'll check them now.


-- 
Takashi Iwai <tiwai@suse.de>		SuSE Linux AG - www.suse.de
ALSA Developer				ALSA Project - www.alsa-project.org

Re: ALSA update

[Jaroslav Kysela]

On Thu, 19 Dec 2002, Takashi Iwai wrote:

> At Wed, 18 Dec 2002 22:51:27 +0300 (MSK),
> Ruslan U. Zakirov <cubic@miee.ru> wrote:
> > 
> > Hello, Jaroslav and All.
> > How about other changes in new 2.5 kernel, like new PnP layer (Adam Belay)
> > or changes with module & boot params (Rusty Russel)? There are now some
> > changes in 2.5.52 kernel in sound/isa/opl3sa2.c that make this driver not
> > compatible with other kernels. May be it's better split your tree in
> > several trees for each version of kernels?
> 
> if possible, we'll build up some wrapper for 2.4 on alsa-driver (not
> the codebase for 2.5) tree.  if not possible, yes, splitting to two
> trees would be reasonable for such big changes...
> 
> thanks for noticing this issue.  i'll check them now.

I just removed the complete callback redefinitions from the 2.5 tree, 
but we still have three small OLD_USB sections (mainly commenting 2.5 
code which 2.4 code requires to override). Hopefully, it's quite reasonable.
What do you think, Greg?

						Jaroslav

-----
Jaroslav Kysela <perex@suse.cz>
Linux Kernel Sound Maintainer
ALSA Project, SuSE Labs

Re: ALSA update

[Greg KH]

On Thu, Dec 19, 2002 at 01:18:10PM +0100, Jaroslav Kysela wrote:
> On Thu, 19 Dec 2002, Takashi Iwai wrote:
> 
> > At Wed, 18 Dec 2002 22:51:27 +0300 (MSK),
> > Ruslan U. Zakirov <cubic@miee.ru> wrote:
> > > 
> > > Hello, Jaroslav and All.
> > > How about other changes in new 2.5 kernel, like new PnP layer (Adam Belay)
> > > or changes with module & boot params (Rusty Russel)? There are now some
> > > changes in 2.5.52 kernel in sound/isa/opl3sa2.c that make this driver not
> > > compatible with other kernels. May be it's better split your tree in
> > > several trees for each version of kernels?
> > 
> > if possible, we'll build up some wrapper for 2.4 on alsa-driver (not
> > the codebase for 2.5) tree.  if not possible, yes, splitting to two
> > trees would be reasonable for such big changes...
> > 
> > thanks for noticing this issue.  i'll check them now.
> 
> I just removed the complete callback redefinitions from the 2.5 tree, 
> but we still have three small OLD_USB sections (mainly commenting 2.5 
> code which 2.4 code requires to override). Hopefully, it's quite reasonable.
> What do you think, Greg?

In the end, whatever is easiest for you do to, as I'm not the one who
has to maintain the code :)

That being said, the smaller the number of #ifdefs, the better.
Personally I would just have two different trees, but that's just me...

thanks,

greg k-h

Re: ALSA update

[Bill Davidsen]

On Thu, 19 Dec 2002, Takashi Iwai wrote:

> At Wed, 18 Dec 2002 22:51:27 +0300 (MSK),
> Ruslan U. Zakirov <cubic@miee.ru> wrote:
> > 
> > Hello, Jaroslav and All.
> > How about other changes in new 2.5 kernel, like new PnP layer (Adam Belay)
> > or changes with module & boot params (Rusty Russel)? There are now some
> > changes in 2.5.52 kernel in sound/isa/opl3sa2.c that make this driver not
> > compatible with other kernels. May be it's better split your tree in
> > several trees for each version of kernels?
> 
> if possible, we'll build up some wrapper for 2.4 on alsa-driver (not
> the codebase for 2.5) tree.  if not possible, yes, splitting to two
> trees would be reasonable for such big changes...
> 
> thanks for noticing this issue.  i'll check them now.

If you really want single source you might use something like m4 to split
out both 2.4 and 2.5 versions which are less cluttered.

I admit this doesn't make your job a bit easier, but people reading the
drivers for one series or the other would have an easier job reading the
code. In a perfect world that would means they would find bugs for you and
just send patches. In reality you wouldn't have people telling you your
code was ugly.

-- 
bill davidsen <davidsen@tmr.com>
  CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.