parisc/unwind: call callback with toplevel address

parisc/unwind: call callback with toplevel address

[Naresh Kamboju]

Following build errors noticed while building Linux next 20211001
with gcc-11 for parisc architecture.

arch/parisc/kernel/stacktrace.c: In function 'dump_trace':
arch/parisc/kernel/stacktrace.c:20:13: error: 'regs' undeclared (first
use in this function)
   20 |         if (regs)
      |             ^~~~
arch/parisc/kernel/stacktrace.c:20:13: note: each undeclared
identifier is reported only once for each function it appears in
arch/parisc/kernel/stacktrace.c:21:22: error: implicit declaration of
function 'fn' [-Werror=implicit-function-declaration]
   21 |                 if (!fn(cookie, regs->iaoq[0]))
      |                      ^~
arch/parisc/kernel/stacktrace.c:21:25: error: 'cookie' undeclared
(first use in this function)
   21 |                 if (!fn(cookie, regs->iaoq[0]))
      |                         ^~~~~~
cc1: some warnings being treated as errors
make[3]: *** [scripts/Makefile.build:288:
arch/parisc/kernel/stacktrace.o] Error 1

Build config:
https://builds.tuxbuild.com/1ytbtyEg5SDSQgS2Oj9RsCM4ZmS/config

Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>

meta data:
-----------
    git_repo: https://gitlab.com/Linaro/lkft/mirrors/next/linux-next
    git_sha: a25006a77348ba06c7bc96520d331cd9dd370715
    git_short_log: a25006a77348 (\"Add linux-next specific files for 20211001\")
    kconfig:  defconfig
    kernel_version: 5.15.0-rc3
    target_arch: parisc
    toolchain: gcc-11

steps to reproduce:
https://builds.tuxbuild.com/1ytbtyEg5SDSQgS2Oj9RsCM4ZmS/tuxmake_reproducer.sh

--
Linaro LKFT
https://lkft.linaro.org

Re: parisc/unwind: call callback with toplevel address

[Helge Deller]

On 10/1/21 12:30, Naresh Kamboju wrote:
> Following build errors noticed while building Linux next 20211001
> with gcc-11 for parisc architecture.
>
> arch/parisc/kernel/stacktrace.c: In function 'dump_trace':
> arch/parisc/kernel/stacktrace.c:20:13: error: 'regs' undeclared (first
> use in this function)
>    20 |         if (regs)
>       |             ^~~~
> arch/parisc/kernel/stacktrace.c:20:13: note: each undeclared
> identifier is reported only once for each function it appears in
> arch/parisc/kernel/stacktrace.c:21:22: error: implicit declaration of
> function 'fn' [-Werror=implicit-function-declaration]
>    21 |                 if (!fn(cookie, regs->iaoq[0]))
>       |                      ^~
> arch/parisc/kernel/stacktrace.c:21:25: error: 'cookie' undeclared
> (first use in this function)
>    21 |                 if (!fn(cookie, regs->iaoq[0]))
>       |                         ^~~~~~
> cc1: some warnings being treated as errors
> make[3]: *** [scripts/Makefile.build:288:
> arch/parisc/kernel/stacktrace.o] Error 1
>
> Build config:
> https://builds.tuxbuild.com/1ytbtyEg5SDSQgS2Oj9RsCM4ZmS/config
>
> Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>

I've dropped that patch.

Thanks!
Helge



> meta data:
> -----------
>     git_repo: https://gitlab.com/Linaro/lkft/mirrors/next/linux-next
>     git_sha: a25006a77348ba06c7bc96520d331cd9dd370715
>     git_short_log: a25006a77348 (\"Add linux-next specific files for 20211001\")
>     kconfig:  defconfig
>     kernel_version: 5.15.0-rc3
>     target_arch: parisc
>     toolchain: gcc-11
>
> steps to reproduce:
> https://builds.tuxbuild.com/1ytbtyEg5SDSQgS2Oj9RsCM4ZmS/tuxmake_reproducer.sh
>
> --
> Linaro LKFT
> https://lkft.linaro.org
>

ATI Rage XL

[cp]

hi
in the PA wiki (1) the ATI Rage XL (mach64) is reported to have
worked. In combination with the fbdev X-server from XFree86/X.org

I think the news was related to kernel 2.6 or something very old.

Can anyone confirm it's still working with kernel 5.*?

I don't happen to have a card to test myself. Not yet.

C.



(1) https://parisc.wiki.kernel.org/index.php/Graphics_howto

Re: ATI Rage XL

[cp]

Apple PowerBook-G3 use these cards

Name       , Model , GPU             , driver , vram ;
wallstreet , M4753 , ATI Rage LT     , mach64 ,  2MB ;
lombard    , M5343 , ATI Rage Pro LT , mach64 ,  8MB ;
pismo      , M5343 , ATI Rage 128    , ati128 ,  8MB ;

So maybe there is still some interest in supporting their kernel drivers.
Sooner or later I will try to grab some of these PCI cards and try
them with HPPA on modern 5 kernels. *

I know there is also a miniPCI version of the ATI RageXL and it's used
with some IBM servers, so again, maybe there is still some interest
even from those guys ...

On Thu, 7 Oct 2021 at 11:09, cp <carlojpisani@gmail.com> wrote:
>
> hi
> in the PA wiki (1) the ATI Rage XL (mach64) is reported to have
> worked. In combination with the fbdev X-server from XFree86/X.org
>
> I think the news was related to kernel 2.6 or something very old.
>
> Can anyone confirm it's still working with kernel 5.*?
>
> I don't happen to have a card to test myself. Not yet.
>
> C.
>
>
>
> (1) https://parisc.wiki.kernel.org/index.php/Graphics_howto

Re: ATI Rage XL

[cp]

anyone?

any news about the mach64 support with kernel 5.* ?
cheers

> hi
> in the PA wiki (1) the ATI Rage XL (mach64) is reported to have
> worked. In combination with the fbdev X-server from XFree86/X.org
>
> I think the news was related to kernel 2.6 or something very old.
>
> Can anyone confirm it's still working with kernel 5.*?
>
> I don't happen to have a card to test myself. Not yet.
> (1) https://parisc.wiki.kernel.org/index.php/Graphics_howto

ATI Rage XL

[Michael Stickel]

Hi.

I am new to the list.
I have a MIPS32 based platform with miniPCI slots and an ATI Rage XL.
There where rumors that there is code in the 2.6 Kernel that initializes 
the card for non-x86 platforms. I have found a file called xlinit.c
with a function atyfb_xl_init in it, but I did not find a place where it 
is called. I read that it is a patch from SGI to get the Rage XL running
on non X86 platforms.

Is all that only a rumor or does anyone have a pointer where to find
more information about the ATI Rage XL in non x86 platforms.

Michael
michael@cubic.org