problems on alpha

problems on alpha

[Martin Langer]

Hi,

please don't try to build au88x0 on an alpha. You have to read a lot. 
But here is an extract with some quite interesting points about CVS ;-)

martin


gcc -D__KERNEL__ -DMODULE=1 -I/usr/local/src/alsa-cvs/alsa-driver/include
-I/lib/modules/2.2.19/build/include -O2 -mno-fp-regs -ffixed-8 -Wa,-mev6
-DLINUX -Wall -Wstrict-prototypes -fomit-frame-pointer -Wno-trigraphs -O2
-fno-strict-aliasing -fno-common -pipe -DALSA_BUILD
-DKBUILD_BASENAME=memory_wrapper   -c -o memory_wrapper.o memory_wrapper.c
memory_wrapper.c: In function snd_compat_vmalloc_to_page':
memory_wrapper.c:35: warning: implicit declaration of function VMALLOC_VMADDR'
memory_wrapper.c:39: warning: implicit declaration of function pgd_offset'
memory_wrapper.c:39: warning: assignment makes pointer from integer without
a cast
memory_wrapper.c:40: warning: implicit declaration of function pmd_offset'
memory_wrapper.c:40: warning: assignment makes pointer from integer without
a cast
memory_wrapper.c:41: warning: implicit declaration of function pte_offset'
memory_wrapper.c:41: warning: assignment makes pointer from integer without
a cast
memory_wrapper.c:43: warning: implicit declaration of function pte_page'

[...]

gcc -D__KERNEL__ -DMODULE=1 -I/usr/local/src/alsa-cvs/alsa-driver/include
-I/lib/modules/2.2.19/build/include -O2 -mno-fp-regs -ffixed-8 -Wa,-mev6
-DLINUX -Wall -Wstrict-prototypes -fomit-frame-pointer -Wno-trigraphs -O2
-fno-strict-aliasing -fno-common -pipe -DALSA_BUILD
-DKBUILD_BASENAME=au8810   -c -o au8810.o au8810.c
In file included from au88x0.c:23,
                 from au8810.c:8:
/usr/local/src/alsa-cvs/alsa-driver/include/sound/initval.h: In function
get_option_long':
/usr/local/src/alsa-cvs/alsa-driver/include/sound/initval.h:172: warning:
implicit declaration of function get_option'
In file included from au8810.c:8:
au88x0.c: In function snd_vortex_dev_free':
au88x0.c:76: warning: implicit declaration of function pci_release_regions'
au88x0.c:77: warning: implicit declaration of function pci_disable_device'
au88x0.c: In function snd_vortex_create':
au88x0.c:121: warning: implicit declaration of function pci_request_regions'

[...]

au88x0_core.c: In function vortex_mixer_en_sr':
au88x0_core.c:72: warning: passing arg 2 of cia_writel' makes integer from
pointer without a cast
au88x0_core.c: In function vortex_mixer_dis_sr':
au88x0_core.c:76: warning: passing arg 2 of cia_writel' makes integer from
pointer without a cast
au88x0_core.c: In function vortex_mix_muteinputgain':
au88x0_core.c:80: warning: passing arg 2 of cia_writel' makes integer from
pointer without a cast
au88x0_core.c:81: warning: passing arg 2 of cia_writel' makes integer from
pointer without a cast

[...]

ld  -r -o snd-au8810.o au8810.o au8810_core.o au8810_pcm.o au8810_mixer.o
au8810_mpu401.o au8810_game.o au8810_eq.o
au8810_core.o(.modinfo+0x0): multiple definition of __module_kernel_version'
au8810.o(.modinfo+0x0): first defined here
au8810_pcm.o(.modinfo+0x0): multiple definition of __module_kernel_version'
au8810.o(.modinfo+0x0): first defined here
au8810_mixer.o(.modinfo+0x0): multiple definition of __module_kernel_version'
au8810.o(.modinfo+0x0): first defined here
au8810_mpu401.o(.modinfo+0x0): multiple definition of __module_kernel_version'
au8810.o(.modinfo+0x0): first defined here
au8810_game.o(.modinfo+0x0): multiple definition of __module_kernel_version'
au8810.o(.modinfo+0x0): first defined here
au8810_eq.o(.modinfo+0x0): multiple definition of __module_kernel_version'
au8810.o(.modinfo+0x0): first defined here
make[2]: *** [snd-au8810.o] Error 1


-------------------------------------------------------
This SF. Net email is sponsored by: GoToMyPC
GoToMyPC is the fast, easy and secure way to access your computer from
any Web browser or wireless device. Click here to Try it Free!
https://www.gotomypc.com/tr/OSDN/AW/Q4_2003/t/g22lp?Target=mm/g22lp.tmpl

Re: problems on alpha

[Takashi Iwai]

At Sun, 16 Nov 2003 18:51:17 +0100,
Martin Langer wrote:
> 
> 
> Hi,
> 
> please don't try to build au88x0 on an alpha.

most of problems below are due to 2.2 kernel.
the cvs version is not tested well on 2.2 kernels (also i tried only
i386)

>  You have to read a lot. 
> But here is an extract with some quite interesting points about CVS ;-)
> 
> martin
> 
> 
> gcc -D__KERNEL__ -DMODULE=1 -I/usr/local/src/alsa-cvs/alsa-driver/include
> -I/lib/modules/2.2.19/build/include -O2 -mno-fp-regs -ffixed-8 -Wa,-mev6
> -DLINUX -Wall -Wstrict-prototypes -fomit-frame-pointer -Wno-trigraphs -O2
> -fno-strict-aliasing -fno-common -pipe -DALSA_BUILD
> -DKBUILD_BASENAME=memory_wrapper   -c -o memory_wrapper.o memory_wrapper.c
> memory_wrapper.c: In function snd_compat_vmalloc_to_page':
> memory_wrapper.c:35: warning: implicit declaration of function VMALLOC_VMADDR'
> memory_wrapper.c:39: warning: implicit declaration of function pgd_offset'
> memory_wrapper.c:39: warning: assignment makes pointer from integer without
> a cast
> memory_wrapper.c:40: warning: implicit declaration of function pmd_offset'
> memory_wrapper.c:40: warning: assignment makes pointer from integer without
> a cast
> memory_wrapper.c:41: warning: implicit declaration of function pte_offset'
> memory_wrapper.c:41: warning: assignment makes pointer from integer without
> a cast
> memory_wrapper.c:43: warning: implicit declaration of function pte_page'

it seems that configure script didn't detect these checks correctly.
could you check config.log ?

> 
> [...]
> 
> gcc -D__KERNEL__ -DMODULE=1 -I/usr/local/src/alsa-cvs/alsa-driver/include
> -I/lib/modules/2.2.19/build/include -O2 -mno-fp-regs -ffixed-8 -Wa,-mev6
> -DLINUX -Wall -Wstrict-prototypes -fomit-frame-pointer -Wno-trigraphs -O2
> -fno-strict-aliasing -fno-common -pipe -DALSA_BUILD
> -DKBUILD_BASENAME=au8810   -c -o au8810.o au8810.c
> In file included from au88x0.c:23,
>                  from au8810.c:8:
> /usr/local/src/alsa-cvs/alsa-driver/include/sound/initval.h: In function
> get_option_long':
> /usr/local/src/alsa-cvs/alsa-driver/include/sound/initval.h:172: warning:
> implicit declaration of function get_option'

it must be harmless since the function will be never used unless the
driver is built into kernel.  (should be fixed, though)

> In file included from au8810.c:8:
> au88x0.c: In function snd_vortex_dev_free':
> au88x0.c:76: warning: implicit declaration of function pci_release_regions'
> au88x0.c:77: warning: implicit declaration of function pci_disable_device'
> au88x0.c: In function snd_vortex_create':
> au88x0.c:121: warning: implicit declaration of function pci_request_regions'

grrr, they don't exist on 2.2 kernels...
we need a wrapper for them.

> [...]
> 
> au88x0_core.c: In function vortex_mixer_en_sr':
> au88x0_core.c:72: warning: passing arg 2 of cia_writel' makes integer from
> pointer without a cast
> au88x0_core.c: In function vortex_mixer_dis_sr':
> au88x0_core.c:76: warning: passing arg 2 of cia_writel' makes integer from
> pointer without a cast
> au88x0_core.c: In function vortex_mix_muteinputgain':
> au88x0_core.c:80: warning: passing arg 2 of cia_writel' makes integer from
> pointer without a cast
> au88x0_core.c:81: warning: passing arg 2 of cia_writel' makes integer from
> pointer without a cast

they are ok.   (should be fixed, too)

> 
> [...]
> 
> ld  -r -o snd-au8810.o au8810.o au8810_core.o au8810_pcm.o au8810_mixer.o
> au8810_mpu401.o au8810_game.o au8810_eq.o
> au8810_core.o(.modinfo+0x0): multiple definition of __module_kernel_version'
> au8810.o(.modinfo+0x0): first defined here
> au8810_pcm.o(.modinfo+0x0): multiple definition of __module_kernel_version'
> au8810.o(.modinfo+0x0): first defined here
> au8810_mixer.o(.modinfo+0x0): multiple definition of __module_kernel_version'
> au8810.o(.modinfo+0x0): first defined here
> au8810_mpu401.o(.modinfo+0x0): multiple definition of __module_kernel_version'
> au8810.o(.modinfo+0x0): first defined here
> au8810_game.o(.modinfo+0x0): multiple definition of __module_kernel_version'
> au8810.o(.modinfo+0x0): first defined here
> au8810_eq.o(.modinfo+0x0): multiple definition of __module_kernel_version'
> au8810.o(.modinfo+0x0): first defined here
> make[2]: *** [snd-au8810.o] Error 1

defning __NO_VERSION__ seems missing.
put #define __NO_VERSION__ at the beginning of each au88?0.c.


Takashi


-------------------------------------------------------
This SF. Net email is sponsored by: GoToMyPC
GoToMyPC is the fast, easy and secure way to access your computer from
any Web browser or wireless device. Click here to Try it Free!
https://www.gotomypc.com/tr/OSDN/AW/Q4_2003/t/g22lp?Target=mm/g22lp.tmpl

Re: problems on alpha

[Takashi Iwai]

At Mon, 17 Nov 2003 12:08:39 +0100,
I wrote:
> 
> At Sun, 16 Nov 2003 18:51:17 +0100,
> Martin Langer wrote:
> > 
> > 
> > Hi,
> > 
> > please don't try to build au88x0 on an alpha.
> 
> most of problems below are due to 2.2 kernel.
> the cvs version is not tested well on 2.2 kernels (also i tried only
> i386)
> 
> >  You have to read a lot. 
> > But here is an extract with some quite interesting points about CVS ;-)
> > 
> > martin
> > 
> > 
> > gcc -D__KERNEL__ -DMODULE=1 -I/usr/local/src/alsa-cvs/alsa-driver/include
> > -I/lib/modules/2.2.19/build/include -O2 -mno-fp-regs -ffixed-8 -Wa,-mev6
> > -DLINUX -Wall -Wstrict-prototypes -fomit-frame-pointer -Wno-trigraphs -O2
> > -fno-strict-aliasing -fno-common -pipe -DALSA_BUILD
> > -DKBUILD_BASENAME=memory_wrapper   -c -o memory_wrapper.o memory_wrapper.c
> > memory_wrapper.c: In function snd_compat_vmalloc_to_page':
> > memory_wrapper.c:35: warning: implicit declaration of function VMALLOC_VMADDR'
> > memory_wrapper.c:39: warning: implicit declaration of function pgd_offset'
> > memory_wrapper.c:39: warning: assignment makes pointer from integer without
> > a cast
> > memory_wrapper.c:40: warning: implicit declaration of function pmd_offset'
> > memory_wrapper.c:40: warning: assignment makes pointer from integer without
> > a cast
> > memory_wrapper.c:41: warning: implicit declaration of function pte_offset'
> > memory_wrapper.c:41: warning: assignment makes pointer from integer without
> > a cast
> > memory_wrapper.c:43: warning: implicit declaration of function pte_page'
> 
> it seems that configure script didn't detect these checks correctly.
> could you check config.log ?

it seems that 2.2.19 kernel surely has these definitions.
are you sure that you pass the correct kernel source path to configure
script, or /lib/modules/2.2.19/build is symlinked to its kernel source
tree?


Takashi


-------------------------------------------------------
This SF. Net email is sponsored by: GoToMyPC
GoToMyPC is the fast, easy and secure way to access your computer from
any Web browser or wireless device. Click here to Try it Free!
https://www.gotomypc.com/tr/OSDN/AW/Q4_2003/t/g22lp?Target=mm/g22lp.tmpl

Re: problems on alpha

[Martin Langer]

On Mon, Nov 17, 2003 at 06:55:29PM +0100, Takashi Iwai wrote:
> At Mon, 17 Nov 2003 12:08:39 +0100,
> I wrote:
> > 
> > At Sun, 16 Nov 2003 18:51:17 +0100,
> > Martin Langer wrote:
> > > 
> > > gcc -D__KERNEL__ -DMODULE=1 -I/usr/local/src/alsa-cvs/alsa-driver/include
> > > -I/lib/modules/2.2.19/build/include -O2 -mno-fp-regs -ffixed-8 -Wa,-mev6
> > > -DLINUX -Wall -Wstrict-prototypes -fomit-frame-pointer -Wno-trigraphs -O2
> > > -fno-strict-aliasing -fno-common -pipe -DALSA_BUILD
> > > -DKBUILD_BASENAME=memory_wrapper   -c -o memory_wrapper.o memory_wrapper.c
> > > memory_wrapper.c: In function snd_compat_vmalloc_to_page':
> > > memory_wrapper.c:35: warning: implicit declaration of function VMALLOC_VMADDR'
> > > memory_wrapper.c:39: warning: implicit declaration of function pgd_offset'
> > > memory_wrapper.c:39: warning: assignment makes pointer from integer without
> > > a cast
> > > memory_wrapper.c:40: warning: implicit declaration of function pmd_offset'
> > > memory_wrapper.c:40: warning: assignment makes pointer from integer without
> > > a cast
> > > memory_wrapper.c:41: warning: implicit declaration of function pte_offset'
> > > memory_wrapper.c:41: warning: assignment makes pointer from integer without
> > > a cast
> > > memory_wrapper.c:43: warning: implicit declaration of function pte_page'
> > 
> > it seems that configure script didn't detect these checks correctly.
> > could you check config.log ?
> 
> it seems that 2.2.19 kernel surely has these definitions.
> are you sure that you pass the correct kernel source path to configure
> script, or /lib/modules/2.2.19/build is symlinked to its kernel source
> tree?

Yes. I've found the lines in config.log and it's an alpha problem.
Thanks for that config.log hint. 

martin


configure:3194: checking for vmalloc_to_page
configure:3214: gcc -c -g -O2 -I/lib/modules/2.2.19/build/include  conftest.c 1>&5
In file included from /lib/modules/2.2.19/build/include/asm/semaphore.h:11,
                 from /lib/modules/2.2.19/build/include/linux/sched.h:17,
                 from /lib/modules/2.2.19/build/include/linux/mm.h:4,
                 from configure:3204:
/lib/modules/2.2.19/build/include/asm/current.h:4: global register variable follows a function definition
/lib/modules/2.2.19/build/include/asm/current.h:4: warning: call-clobbered register used for global register variable
configure: In function `main':
configure:3209: `vmalloc_to_page' undeclared (first use in this function)
configure:3209: (Each undeclared identifier is reported only once
configure:3209: for each function it appears in.)



-------------------------------------------------------
This SF. Net email is sponsored by: GoToMyPC
GoToMyPC is the fast, easy and secure way to access your computer from
any Web browser or wireless device. Click here to Try it Free!
https://www.gotomypc.com/tr/OSDN/AW/Q4_2003/t/g22lp?Target=mm/g22lp.tmpl

Re: problems on alpha

[Martin Langer]

On Mon, Nov 17, 2003 at 12:08:39PM +0100, Takashi Iwai wrote:
> At Sun, 16 Nov 2003 18:51:17 +0100,
> Martin Langer wrote:
> > 
> > 
> > Hi,
> > 
> > please don't try to build au88x0 on an alpha.
> 
> most of problems below are due to 2.2 kernel.
> the cvs version is not tested well on 2.2 kernels (also i tried only
> i386)
> 
> >  You have to read a lot. 
> > But here is an extract with some quite interesting points about CVS ;-)
> > 
> > martin
> > 
> > 
> > gcc -D__KERNEL__ -DMODULE=1 -I/usr/local/src/alsa-cvs/alsa-driver/include
> > -I/lib/modules/2.2.19/build/include -O2 -mno-fp-regs -ffixed-8 -Wa,-mev6
> > -DLINUX -Wall -Wstrict-prototypes -fomit-frame-pointer -Wno-trigraphs -O2
> > -fno-strict-aliasing -fno-common -pipe -DALSA_BUILD
> > -DKBUILD_BASENAME=memory_wrapper   -c -o memory_wrapper.o memory_wrapper.c
> > memory_wrapper.c: In function snd_compat_vmalloc_to_page':
> > memory_wrapper.c:35: warning: implicit declaration of function VMALLOC_VMADDR'
> > memory_wrapper.c:39: warning: implicit declaration of function pgd_offset'
> > memory_wrapper.c:39: warning: assignment makes pointer from integer without
> > a cast
> > memory_wrapper.c:40: warning: implicit declaration of function pmd_offset'
> > memory_wrapper.c:40: warning: assignment makes pointer from integer without
> > a cast
> > memory_wrapper.c:41: warning: implicit declaration of function pte_offset'
> > memory_wrapper.c:41: warning: assignment makes pointer from integer without
> > a cast
> > memory_wrapper.c:43: warning: implicit declaration of function pte_page'
> 
> it seems that configure script didn't detect these checks correctly.
> could you check config.log ?
> 
> > 
> > [...]
> > 
> > gcc -D__KERNEL__ -DMODULE=1 -I/usr/local/src/alsa-cvs/alsa-driver/include
> > -I/lib/modules/2.2.19/build/include -O2 -mno-fp-regs -ffixed-8 -Wa,-mev6
> > -DLINUX -Wall -Wstrict-prototypes -fomit-frame-pointer -Wno-trigraphs -O2
> > -fno-strict-aliasing -fno-common -pipe -DALSA_BUILD
> > -DKBUILD_BASENAME=au8810   -c -o au8810.o au8810.c
> > In file included from au88x0.c:23,
> >                  from au8810.c:8:
> > /usr/local/src/alsa-cvs/alsa-driver/include/sound/initval.h: In function
> > get_option_long':
> > /usr/local/src/alsa-cvs/alsa-driver/include/sound/initval.h:172: warning:
> > implicit declaration of function get_option'
> 
> it must be harmless since the function will be never used unless the
> driver is built into kernel.  (should be fixed, though)
> 
> > In file included from au8810.c:8:
> > au88x0.c: In function snd_vortex_dev_free':
> > au88x0.c:76: warning: implicit declaration of function pci_release_regions'
> > au88x0.c:77: warning: implicit declaration of function pci_disable_device'
> > au88x0.c: In function snd_vortex_create':
> > au88x0.c:121: warning: implicit declaration of function pci_request_regions'
> 
> grrr, they don't exist on 2.2 kernels...
> we need a wrapper for them.
> 

It's time to say good-bye to 2.2 users...

> > [...]
> > 
> > au88x0_core.c: In function vortex_mixer_en_sr':
> > au88x0_core.c:72: warning: passing arg 2 of cia_writel' makes integer from
> > pointer without a cast
> > au88x0_core.c: In function vortex_mixer_dis_sr':
> > au88x0_core.c:76: warning: passing arg 2 of cia_writel' makes integer from
> > pointer without a cast
> > au88x0_core.c: In function vortex_mix_muteinputgain':
> > au88x0_core.c:80: warning: passing arg 2 of cia_writel' makes integer from
> > pointer without a cast
> > au88x0_core.c:81: warning: passing arg 2 of cia_writel' makes integer from
> > pointer without a cast
> 
> they are ok.   (should be fixed, too)
> 
> > 
> > [...]
> > 
> > ld  -r -o snd-au8810.o au8810.o au8810_core.o au8810_pcm.o au8810_mixer.o
> > au8810_mpu401.o au8810_game.o au8810_eq.o
> > au8810_core.o(.modinfo+0x0): multiple definition of __module_kernel_version'
> > au8810.o(.modinfo+0x0): first defined here
> > au8810_pcm.o(.modinfo+0x0): multiple definition of __module_kernel_version'
> > au8810.o(.modinfo+0x0): first defined here
> > au8810_mixer.o(.modinfo+0x0): multiple definition of __module_kernel_version'
> > au8810.o(.modinfo+0x0): first defined here
> > au8810_mpu401.o(.modinfo+0x0): multiple definition of __module_kernel_version'
> > au8810.o(.modinfo+0x0): first defined here
> > au8810_game.o(.modinfo+0x0): multiple definition of __module_kernel_version'
> > au8810.o(.modinfo+0x0): first defined here
> > au8810_eq.o(.modinfo+0x0): multiple definition of __module_kernel_version'
> > au8810.o(.modinfo+0x0): first defined here
> > make[2]: *** [snd-au8810.o] Error 1
> 
> defning __NO_VERSION__ seems missing.
> put #define __NO_VERSION__ at the beginning of each au88?0.c.
> 

I guess switching to 2.4 could be the best solution for me... 


martin


-------------------------------------------------------
This SF. Net email is sponsored by: GoToMyPC
GoToMyPC is the fast, easy and secure way to access your computer from
any Web browser or wireless device. Click here to Try it Free!
https://www.gotomypc.com/tr/OSDN/AW/Q4_2003/t/g22lp?Target=mm/g22lp.tmpl

Re: problems on alpha

[Takashi Iwai]

At Mon, 17 Nov 2003 22:26:54 +0100,
Martin Langer wrote:
(snip)
> > grrr, they don't exist on 2.2 kernels...
> > we need a wrapper for them.
> > 
> 
> It's time to say good-bye to 2.2 users...

i fixed (part of) these problems on cvs tree.
still i don't figure out why pmd_table() is not defined...


Takashi


-------------------------------------------------------
This SF. Net email is sponsored by: GoToMyPC
GoToMyPC is the fast, easy and secure way to access your computer from
any Web browser or wireless device. Click here to Try it Free!
https://www.gotomypc.com/tr/OSDN/AW/Q4_2003/t/g22lp?Target=mm/g22lp.tmpl

Re: problems on alpha

[Martin Langer]

On Tue, Nov 18, 2003 at 11:09:12AM +0100, Takashi Iwai wrote:
> At Mon, 17 Nov 2003 22:26:54 +0100,
> Martin Langer wrote:
> (snip)
> > > grrr, they don't exist on 2.2 kernels...
> > > we need a wrapper for them.
> > > 
> > 
> > It's time to say good-bye to 2.2 users...
> 
> i fixed (part of) these problems on cvs tree.
> still i don't figure out why pmd_table() is not defined...

Thanks, Takashi.

But snd_pci_compat_pci_request_region in pre2 looks like a typo to me.
(because it doesn't fit to the entry in acore/memalloc.inc1)



--- alsa-driver/include/compat_22.h.ORIGI	Sun Nov 23 14:54:37 2003
+++ alsa-driver/include/compat_22.h	Sun Nov 23 14:55:14 2003
@@ -289,7 +289,7 @@
 int snd_pci_compat_set_dma_mask(struct pci_dev *, unsigned long mask);
 void * snd_pci_compat_get_driver_data (struct pci_dev *dev);
 void snd_pci_compat_set_driver_data (struct pci_dev *dev, void *driver_data);
-int snd_pci_compat_pci_request_region(struct pci_dev *pdev, int bar, char *res_name);
+int snd_pci_compat_request_region(struct pci_dev *pdev, int bar, char *res_name);
 void snd_pci_compat_release_region(struct pci_dev *pci, int bar);
 int snd_pci_compat_request_regions(struct pci_dev *pdev, char *res_name);
 void snd_pci_compat_release_regions(struct pci_dev *pdev);


--- alsa-driver/acore/pci_compat_22.c.ORIGINAL	Sun Nov 23 14:41:21 2003
+++ alsa-driver/acore/pci_compat_22.c	Sun Nov 23 14:41:44 2003
@@ -343,7 +343,7 @@
 
 /*
  */
-int snd_pci_compat_pci_request_region(struct pci_dev *pdev, int bar, char *res_name)
+int snd_pci_compat_request_region(struct pci_dev *pdev, int bar, char *res_name)
 {
 	int flags;
 


-- 
The only nice thing about spam is that it doesn't ring.


-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/