linuxppc-2.5 fixlets

linuxppc-2.5 fixlets

[Michel Dänzer]

[-- Attachment #1: Type: text/plain, Size: 885 bytes --]


First of all, export flush_tlb_page, which is needed by ext3 and SCSI
modules.

Then, thanks to the great new build system, some warnings stood out:

Fix 'comparison is always false due to limited range of data type' in
net/sched/estimator.c by changing char to signed char.

I'm not sure how to fix the same warning in drivers/ide/ppc/pmac.c
though:

if (drive->waiting_for_dma >= DMA_WAIT_TIMEOUT) {

DMA_WAIT_TIMEOUT is 500, but drive->waiting_for_dma is a u8.

Finally, fix

drivers/usb/misc/usbtest.c:438: warning: passing arg 1 of `__swab32s'
from incompatible pointer type

by using le16_to_cpus instead of le32_to_cpus .


PS: I guess Ben's PM fixes haven't made it into linuxppc-2.5 yet? The
display is garbled after sleep.

--
Earthling Michel Dänzer   \  Debian (powerpc), XFree86 and DRI developer
Software libre enthusiast  \     http://svcs.affero.net/rm.php?r=daenzer

[-- Attachment #2: linuxppc-2.5.diff --]
[-- Type: text/plain, Size: 1260 bytes --]

--- linuxppc-2.6.0-test3/arch/ppc/kernel/ppc_ksyms.c	2003-08-20 10:27:35.000000000 +0200
+++ linuxppc-2.6.0-test3-md/arch/ppc/kernel/ppc_ksyms.c	2003-08-21 13:16:55.000000000 +0200
@@ -204,6 +204,7 @@ EXPORT_SYMBOL(flush_dcache_range);
 EXPORT_SYMBOL(flush_icache_user_range);
 EXPORT_SYMBOL(flush_dcache_page);
 EXPORT_SYMBOL(flush_tlb_kernel_range);
+EXPORT_SYMBOL(flush_tlb_page);
 #ifdef CONFIG_ALTIVEC
 EXPORT_SYMBOL(last_task_used_altivec);
 EXPORT_SYMBOL(giveup_altivec);
--- linuxppc-2.6.0-test3/drivers/usb/misc/usbtest.c	2003-08-20 10:27:43.000000000 +0200
+++ linuxppc-2.6.0-test3-md/drivers/usb/misc/usbtest.c	2003-08-21 17:34:01.000000000 +0200
@@ -435,7 +435,7 @@ static int is_good_config (char *buf, in
 		return 0;
 	}

-	le32_to_cpus (&config->wTotalLength);
+	le16_to_cpus (&config->wTotalLength);
 	if (config->wTotalLength == len)		/* read it all */
 		return 1;
 	return config->wTotalLength >= TBUF_SIZE;	/* max partial read */
--- linuxppc-2.6.0-test3/include/linux/pkt_sched.h	2003-08-20 10:26:48.000000000 +0200
+++ linuxppc-2.6.0-test3-md/include/linux/pkt_sched.h	2003-08-21 17:02:55.000000000 +0200
@@ -45,7 +45,7 @@ struct tc_stats

 struct tc_estimator
 {
-	char		interval;
+	signed char	interval;
 	unsigned char	ewma_log;
 };

Re: linuxppc-2.5 fixlets

[Tom Rini]

On Thu, Aug 21, 2003 at 07:36:49PM +0200, Michel D?nzer wrote:

> First of all, export flush_tlb_page, which is needed by ext3 and SCSI
> modules.

Applied.

> Then, thanks to the great new build system, some warnings stood out:
>
> Fix 'comparison is always false due to limited range of data type' in
> net/sched/estimator.c by changing char to signed char.

Send to DaveM (davem at redhat.com).

> Finally, fix
>
> drivers/usb/misc/usbtest.c:438: warning: passing arg 1 of `__swab32s'
> from incompatible pointer type
>
> by using le16_to_cpus instead of le32_to_cpus .

Send to Greg KH (greg at kroah.com)

> PS: I guess Ben's PM fixes haven't made it into linuxppc-2.5 yet? The
> display is garbled after sleep.

linuxppc-2.5-benh tree, I suspect.

--
Tom Rini
http://gate.crashing.org/~trini/

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: linuxppc-2.5 fixlets

[Benjamin Herrenschmidt]

> I'm not sure how to fix the same warning in drivers/ide/ppc/pmac.c
> though:
>
> if (drive->waiting_for_dma >= DMA_WAIT_TIMEOUT) {
>
> DMA_WAIT_TIMEOUT is 500, but drive->waiting_for_dma is a u8.

Fixed in my tree and in the version pending for Linus

> PS: I guess Ben's PM fixes haven't made it into linuxppc-2.5 yet? The
> display is garbled after sleep.

Not yet.

I'm preparing a new set of patches that include all my driver
changes except the video ones. I'll send those ones separately to
James first as they rely on some fbdev infrastructure changes.

ben.


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: linuxppc-2.5 fixlets

[Stefan Berndtsson]

Benjamin Herrenschmidt <benh@kernel.crashing.org> writes:

> I'm preparing a new set of patches that include all my driver
> changes except the video ones. I'll send those ones separately to
> James first as they rely on some fbdev infrastructure changes.

Does this set contain any audio fixes (either alsa or dmasound) for
the tumbler (ibook2 - late 2001)?

Alsa does not seem to produce any sound, whatever I try to do (including
messing around with mute), and dmasound doesn't compile at all.

Used kernel is 2.6.0-test3, via rsync from source.mvista.com::linuxppc-2.5,
as of around 10AM CET (8AM UTC).

/Stefan

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: linuxppc-2.5 fixlets

[Stefan Berndtsson]

Stefan Berndtsson <stefan@nocrew.org> writes:

> Benjamin Herrenschmidt <benh@kernel.crashing.org> writes:
>
>> I'm preparing a new set of patches that include all my driver
>> changes except the video ones. I'll send those ones separately to
>> James first as they rely on some fbdev infrastructure changes.
>
> Does this set contain any audio fixes (either alsa or dmasound) for
> the tumbler (ibook2 - late 2001)?
>
> Alsa does not seem to produce any sound, whatever I try to do (including
> messing around with mute), and dmasound doesn't compile at all.
>
> Used kernel is 2.6.0-test3, via rsync from source.mvista.com::linuxppc-2.5,
> as of around 10AM CET (8AM UTC).

Typically enough, I was just pointed to a fix for this problem (for alsa)
a few minutes after sending this mail.

According to http://www.ussg.iu.edu/hypermail/linux/kernel/0308.2/0288.html
reloading i2c-keywest after loading the alsa-modules initialised things so
that demuting works. There's also a patch included.

/Stefan

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: linuxppc-2.5 fixlets

[Benjamin Herrenschmidt]

On Fri, 2003-08-22 at 13:39, Stefan Berndtsson wrote:
> Benjamin Herrenschmidt <benh@kernel.crashing.org> writes:
>
> > I'm preparing a new set of patches that include all my driver
> > changes except the video ones. I'll send those ones separately to
> > James first as they rely on some fbdev infrastructure changes.
>
> Does this set contain any audio fixes (either alsa or dmasound) for
> the tumbler (ibook2 - late 2001)?
>
> Alsa does not seem to produce any sound, whatever I try to do (including
> messing around with mute), and dmasound doesn't compile at all.
>
> Used kernel is 2.6.0-test3, via rsync from source.mvista.com::linuxppc-2.5,
> as of around 10AM CET (8AM UTC).

No, though I don't do the Alsa stuff (note that I'm also deprecating
dmasound). Regarding Alsa, did you try alsamixer ? I noticed that Alsa
own "Mute" property appear to not be reflected to the /dev/mixer OSS
emulation properly

Ben.


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: linuxppc-2.5 fixlets

[Giuliano Pochini]

> Does this set contain any audio fixes (either alsa or dmasound) for
> the tumbler (ibook2 - late 2001)?
>
> Alsa does not seem to produce any sound, whatever I try to do (including
> messing around with mute), and dmasound doesn't compile at all.

ALSA does not support Mac sound chips very well yet. I also have problems
with a G4mdd (Snapper audio).


Bye.
    Giuliano Pochini ->)|(<- Shiny srl {AS6665} ->)|(<-


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: linuxppc-2.5 fixlets

[Benjamin Herrenschmidt]

On Fri, 2003-08-22 at 14:46, Giuliano Pochini wrote:
> > Does this set contain any audio fixes (either alsa or dmasound) for
> > the tumbler (ibook2 - late 2001)?
> >
> > Alsa does not seem to produce any sound, whatever I try to do (including
> > messing around with mute), and dmasound doesn't compile at all.
>
> ALSA does not support Mac sound chips very well yet. I also have problems
> with a G4mdd (Snapper audio).

The Alsa code base is so much cleaner that I recommend people interested
in looking at it to work with Takashi into fixing his code base, there's
really no point in continuing artificial life support for dmasound on
pmac, that codebase is way to messy and confused (and 2.6 doesn't have
the up-to-date one which is in 2.4 anyway).

Ben.


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: linuxppc-2.5 fixlets

[Giuliano Pochini]

>> ALSA does not support Mac sound chips very well yet. I also have problems
>> with a G4mdd (Snapper audio).
>
> The Alsa code base is so much cleaner that I recommend people interested
> in looking at it to work with Takashi into fixing his code base,

Is there some hw documentation ?

> there's really no point in continuing artificial life support for
> dmasound on pmac, that codebase is way to messy and confused (and
> 2.6 doesn't have the up-to-date one which is in 2.4 anyway).

Well, dmasound doesn't work at all for me, I already moved to
ALSA which is far superior than OSS.


Bye.


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: linuxppc-2.5 fixlets

[Benjamin Herrenschmidt]

On Fri, 2003-08-22 at 18:18, Giuliano Pochini wrote:
> >> ALSA does not support Mac sound chips very well yet. I also have problems
> >> with a G4mdd (Snapper audio).
> >
> > The Alsa code base is so much cleaner that I recommend people interested
> > in looking at it to work with Takashi into fixing his code base,
>
> Is there some hw documentation ?

Some of the chips behind the i2s bus are documented (the Texas ones are)
and there is Apple source code.

Ben


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: linuxppc-2.5 fixlets

[Ernst Persson]

>> Alsa does not seem to produce any sound, whatever I try to do
>> (including messing around with mute), and dmasound doesn't compile at
>> all.
>>
>> Used kernel is 2.6.0-test3, via rsync from
>> source.mvista.com::linuxppc-2.5, as of around 10AM CET (8AM UTC).
>
>Typically enough, I was just pointed to a fix for this problem (for
>alsa) a few minutes after sending this mail.
>
>According to
>http://www.ussg.iu.edu/hypermail/linux/kernel/0308.2/0288.html
>reloading i2c-keywest after loading the alsa-modules initialised things
>so that demuting works. There's also a patch included.

I have something that sounds (no pun intended...) like exactly the same
problem with sound on my iMac, using the burgundy sound driver. Could it
be the same thing? I'll have to try to modify the patch for the burgundy
driver when I get home on friday.

//ernie

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: linuxppc-2.5 fixlets

[Benjamin Herrenschmidt]

On Sat, 2003-08-23 at 16:30, Ernst Persson wrote:
> >> Alsa does not seem to produce any sound, whatever I try to do
> >> (including messing around with mute), and dmasound doesn't compile at
> >> all.
> >>
> >> Used kernel is 2.6.0-test3, via rsync from
> >> source.mvista.com::linuxppc-2.5, as of around 10AM CET (8AM UTC).
> >
> >Typically enough, I was just pointed to a fix for this problem (for
> >alsa) a few minutes after sending this mail.
> >
> >According to
> >http://www.ussg.iu.edu/hypermail/linux/kernel/0308.2/0288.html
> >reloading i2c-keywest after loading the alsa-modules initialised things
> >so that demuting works. There's also a patch included.

Please forward these to the Alsa maintainer Takashi Iwai <tiwai@suse.de>

Ben.


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/