- 17 Feb, 2007 40 commits
-
-
Carlos O'Donell Jr authored
In copy_siginfo_from_user32: Use compat_uptr_t. Use compat_ptr(). In copy_siginfo_to_user32: Use compat_int_t. Use ptr_to_compat(). The sigevent_t structure has a 64-bit si_ptr field that when copied to a 32-bit si_ptr will copy the wrong word. For the compat copy use the si_int field instead. Signed-off-by:
Carlos O'Donell <carlos@systemhalted.org> Signed-off-by:
Kyle McMartin <kyle@parisc-linux.org>
-
Carlos O'Donell Jr authored
Move migrate_pages into the same position as specified in unistd.h. This fixes migrate_pages, pselect6 and ppoll syscalls. Signed-off-by:
-
Matthew Wilcox authored
This reverts commit d3228a88. DeBunk this code. We need it for compat_sys_rt_sigqueueinfo. Signed-off-by:
-
Kyle McMartin authored
the parisc affecting portion of the patch was inadvertantly reverted a while ago. Signed-off-by: -
Helge Deller authored
fix for two warnings: - Section mismatch: reference to .init.text:snd_harmony_mixer_init from .text.snd_harmony_probe after 'snd_harmony_probe' - Section mismatch: reference to .init.text:snd_harmony_mixer_reset from .text.snd_harmony_mixer_init after 'snd_harmony_mixer_init' Signed-off-by:
Helge Deller <deller@gmx.de> Signed-off-by:
-
Helge Deller authored
- noticed by Randolph Chung (tausq) Signed-off-by:
-
Helge Deller authored
Signed-off-by:
-
Helge Deller authored
Signed-off-by:
-
Helge Deller authored
- additionally update my copyright timestamps Signed-off-by:
-
Helge Deller authored
- this macro unifies the code to add exception table entries - additionally use ENTRY()/ENDPROC() at more places Signed-off-by:
-
Helge Deller authored
Signed-off-by:
-
Helge Deller authored
Signed-off-by:
-
Helge Deller authored
Signed-off-by:
-
Helge Deller authored
Use the macros in entry.S Signed-off-by:
-
Kyle McMartin authored
it's unlikely iCOD will ever happen on parisc-linux now... ;-) Signed-off-by: -
Kyle McMartin authored
Signed-off-by: -
Kyle McMartin authored
looks better this way... ;) Signed-off-by: -
Kyle McMartin authored
Signed-off-by: -
Kyle McMartin authored
Signed-off-by: -
Kyle McMartin authored
Signed-off-by: -
Kyle McMartin authored
Signed-off-by: -
Helge Deller authored
Signed-off-by:
-
Kyle McMartin authored
And unmask the pselect6/ppoll system calls. Signed-off-by: -
Helge Deller authored
Signed-off-by:
-
Helge Deller authored
And remove it's reference in time.c. Allow lcd_print() to take a const char *. Signed-off-by:
-
Helge Deller authored
Signed-off-by:
-
Helge Deller authored
Signed-off-by:
-
Helge Deller authored
And clean up 32/64bit branch assembler statements Signed-off-by:
-
Kyle McMartin authored
Signed-off-by: -
Helge Deller authored
Signed-off-by:
-
Matthew Wilcox authored
Fix resource_size_t warnings in a variety of ways Signed-off-by:
Matthew Wilcox <matthew@wil.cx> Signed-off-by:
-
Helge Deller authored
Signed-off-by:
-
Helge Deller authored
Signed-off-by:
-
Helge Deller authored
Signed-off-by:
-
Helge Deller authored
Signed-off-by:
-
Helge Deller authored
Signed-off-by:
-
Helge Deller authored
Signed-off-by:
-
Matthew Wilcox authored
show_stack() was calling kzalloc() to allocate a struct pt_regs. This meant that *really* early stack dumps would cause a null pointer dereference. x86_64 allocates its pt_regs on the stack, so do the same. Kyle actually committed this exact patch to CVS on Wed Jul 26 14:32:39 2006 UTC, and never moved it across to git. Bad Kyle. Signed-off-by:
-
Matthew Wilcox authored
Al Viro did this for x86-64 and reduced the number of dependencies on sched.h significantly. We had a couple of files which were relying on uaccess.h pulling in sched.h, so they need explicit dependencies added. Signed-off-by:
-
Matthew Wilcox authored
They're called from signal.c, so need to be prototyped Signed-off-by:
-
