[git patches] parisc for 2.6.32

[git patches] parisc for 2.6.32

[Kyle McMartin]

Mostly fixes, except for the tracehook stuff... then again cleaning up
ptrace is a pretty big fix in and of itself. ;-)

The following changes since commit 17d857be649a21ca90008c6dc425d849fa83db5c:
  Linus Torvalds (1):
        Linux 2.6.32-rc1

are available in the git repository at:

  hera.kernel.org:/pub/scm/linux/kernel/git/kyle/parisc-2.6.git fixes

Christoph Hellwig (1):
      parisc: convert to asm-generic/hardirq.h

Helge Deller (3):
      parisc: rename parisc's vmalloc_start to parisc_vmalloc_start
      parisc: Fix linker script breakage.
      agp: parisc-agp.c - use correct page_mask function

James Bottomley (1):
      parisc: add me to Maintainers

Jaswinder Singh Rajput (1):
      parisc: includecheck fix: signal.c

Julia Lawall (1):
      parisc: correct use of SHF_ALLOC

Kyle McMartin (7):
      parisc: tracehook_report_syscall
      parisc: tracehook_signal_handler
      parisc: add missing TI_TASK macro in syscall.S
      parisc: split syscall_trace into two halves
      parisc: stop using task->ptrace for {single,block}step flags
      parisc: add skeleton syscall.h
      parisc: HAVE_ARCH_TRACEHOOK

Tim Abbott (1):
      parisc: Make THREAD_SIZE available to assembly files and linker scripts.

 MAINTAINERS                           |    1 +
 arch/parisc/Kconfig                   |    1 +
 arch/parisc/include/asm/fixmap.h      |    4 +-
 arch/parisc/include/asm/hardirq.h     |   20 +---------------
 arch/parisc/include/asm/ptrace.h      |    5 +++-
 arch/parisc/include/asm/syscall.h     |   40 +++++++++++++++++++++++++++++++
 arch/parisc/include/asm/thread_info.h |   14 +++++++----
 arch/parisc/kernel/asm-offsets.c      |    4 +-
 arch/parisc/kernel/entry.S            |   21 ++++++++--------
 arch/parisc/kernel/irq.c              |    5 ----
 arch/parisc/kernel/module.c           |    2 +-
 arch/parisc/kernel/ptrace.c           |   42 ++++++++++++++++----------------
 arch/parisc/kernel/signal.c           |    5 +++-
 arch/parisc/kernel/syscall.S          |   22 +++++++++++------
 arch/parisc/kernel/vmlinux.lds.S      |   10 ++++++++
 arch/parisc/mm/init.c                 |   11 ++++----
 drivers/char/agp/parisc-agp.c         |    2 +-
 17 files changed, 128 insertions(+), 81 deletions(-)
 create mode 100644 arch/parisc/include/asm/syscall.h

Re: [git patches] parisc for 2.6.32

[Linus Torvalds]

On Tue, 6 Oct 2009, Kyle McMartin wrote:
<
> Mostly fixes, except for the tracehook stuff...

I don't see how this can be fixes, since I can't recall any parisc pulls 
during the merge window.

Yes, there have been parisc changes, but they've all come in through 
commits that either were global and people also fixed up parisc while at 
it (eg things like "fcntl: add F_[SG]ETOWN_EX"), or they have been build 
cleanups where you were cc'd but apparently never even bothered to ack the 
change.

So now you come in after the merge window has been closed for over a week, 
and want to do a parisc merge?

Tell me why I should take this. Tell me why I shouldn't just wait until 
the next merge window (or rather "_you_ should have waited").

I've now had two pull requests just today that made me go "I don't really 
see why I should pull this after -rc3 is out".

			Linus

Re: [git patches] parisc for 2.6.32

[Kyle McMartin]

On Tue, Oct 06, 2009 at 10:40:10AM -0700, Linus Torvalds wrote:
> 
> 
> On Tue, 6 Oct 2009, Kyle McMartin wrote:
> <
> > Mostly fixes, except for the tracehook stuff...
> 
> I don't see how this can be fixes, since I can't recall any parisc pulls 
> during the merge window.
> 
> Yes, there have been parisc changes, but they've all come in through 
> commits that either were global and people also fixed up parisc while at 
> it (eg things like "fcntl: add F_[SG]ETOWN_EX"), or they have been build 
> cleanups where you were cc'd but apparently never even bothered to ack the 
> change.
> 

Christoph Hellwig (1):
      parisc: convert to asm-generic/hardirq.h

Jaswinder Singh Rajput (1):
      parisc: includecheck fix: signal.c

James Bottomley (1):
      parisc: add me to Maintainers

- These could wait.

Helge Deller (3):
      parisc: rename parisc's vmalloc_start to parisc_vmalloc_start
      parisc: Fix linker script breakage.
      agp: parisc-agp.c - use correct page_mask function

- 3 build fixes (the vmalloc one is a warning because of a name
  collision.)

Julia Lawall (1):
      parisc: correct use of SHF_ALLOC

- Correctness fix (the two symbols have identical values, but still.)

Kyle McMartin (7):
      parisc: tracehook_report_syscall
      parisc: tracehook_signal_handler
      parisc: add missing TI_TASK macro in syscall.S
      parisc: split syscall_trace into two halves
      parisc: stop using task->ptrace for {single,block}step flags
      parisc: add skeleton syscall.h
      parisc: HAVE_ARCH_TRACEHOOK

- Fine, I can drop these until .33

Tim Abbott (1):
      parisc: Make THREAD_SIZE available to assembly files and linker
scripts.

- Build fix.

I mean, take them, or don't. I'm not going to lose any sleep over it. I
do this on a volunteer basis when I have free time between work and
school... 

regards, Kyle

Re: [git patches] parisc for 2.6.32

[Helge Deller]

Hi Linus,

On 10/06/2009 07:40 PM, Linus Torvalds wrote:
> On Tue, 6 Oct 2009, Kyle McMartin wrote:
>> Mostly fixes, except for the tracehook stuff...
>
> I don't see how this can be fixes, since I can't recall any parisc pulls
> during the merge window.

Yes, sadly it was not in time.
And of course it's no excuse, that you didn't received a real pull-request,
even after Kyle asked for feedback for his tree with his mail on LKML and
parisc-list on 2009-09-28:
http://www.pubbs.net/kernel/200909/123965/

> So now you come in after the merge window has been closed for over a week,
> and want to do a parisc merge?
>
> Tell me why I should take this. Tell me why I shouldn't just wait until
> the next merge window (or rather "_you_ should have waited").

This patch really only touches parisc-specific files and it includes
important fixes without which the parisc kernel won't even compile and link.
So, it would help us parisc-porters a lot if you could do an exception and
still pull from Kyle's tree, even after this merge window has been closed.
At least, it would make me personally very happy...

Helge