linux-stable/arch
Hirokazu Takata 8e8ff02c0b [PATCH] m32r: Fix pt_regs for !COFNIG_ISA_DSP_LEVEL2 target
This modification is required to fix debugging function for m32r targets
with !CONFIG_ISA_DSP_LEVEL2, by unifying 'struct pt_regs' and 'struct
sigcontext' size for all M32R ISA.

Some m32r processor core with !CONFIG_ISA_DSP_LEVEL2 configuration has only
single accumulator a0 (ex.  VDEC2 core, M32102 core, etc.), the others with
CONFIG_ISA_DSP_LEVEL2 has two accumulators, a0 and a1.

This means there are two variations of thread context.  So far, we reduced
and changed stackframe size at a syscall for their context size.  However,
this causes a problem that a GDB for processors with CONFIG_ISA_DSP_LEVEL2
cannot be used for processors with !CONFIG_ISA_DSP_LEVEL2.

From the viewpoint of GDB support, we should reduce such variation of
stackframe size for simplicity.

In this patch, dummy members are added to 'struct pt_regs' and 'struct
sigcontext' to adjust their size for !CONFIG_ISA_DSP_LEVEL2.

This modification is also a one step for a GDB update in future.
Currently, on the m32r, GDB can access process's context by using ptrace
functions in a simple way of register by register access.  By unifying
stackframe size, we have a possibility to make use of ptrace functions of
not only a single register access but also block register access,
PTRACE_{GETREGS,PUTREGS}.

However, for this purpose, we might have to modify stackframe structure
some more; for example, PSW (processor status word) register should be
pre-processed before pushing to stack at a syscall, and so on.  In this
case, we must update carefully both kernel and GDB at a time...

Signed-off-by: Hayato Fujiwara <fujiwara@linux-m32r.org>
Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
Cc: Kei Sakamoto <ksakamot@linux-m32r.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-04-19 09:13:50 -07:00
..
alpha [PATCH] No arch-specific strpbrk implementations 2006-04-11 06:18:40 -07:00
arm [ARM] 3479/1: Corgi SSP: Fix potential concurrent access problem 2006-04-18 23:18:54 +01:00
arm26 [PATCH] No arch-specific strpbrk implementations 2006-04-11 06:18:40 -07:00
cris [PATCH] No arch-specific strpbrk implementations 2006-04-11 06:18:40 -07:00
frv [PATCH] No arch-specific strpbrk implementations 2006-04-11 06:18:40 -07:00
h8300 [PATCH] No arch-specific strpbrk implementations 2006-04-11 06:18:40 -07:00
i386 [PATCH] x86 cpuid and msr notifier callback section mismatches 2006-04-19 09:13:50 -07:00
ia64 [PATCH] DMI: move dmi_scan.c from arch/i386 to drivers/firmware/ 2006-04-14 11:41:25 -07:00
m32r [PATCH] m32r: Fix pt_regs for !COFNIG_ISA_DSP_LEVEL2 target 2006-04-19 09:13:50 -07:00
m68k [PATCH] No arch-specific strpbrk implementations 2006-04-11 06:18:40 -07:00
m68knommu [PATCH] No arch-specific strpbrk implementations 2006-04-11 06:18:40 -07:00
mips [MIPS] Use __ffs() instead of ffs() for waybit calculation. 2006-04-19 04:14:30 +02:00
parisc [PATCH] No arch-specific strpbrk implementations 2006-04-11 06:18:40 -07:00
powerpc Merge git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc-merge 2006-04-18 10:34:24 -07:00
ppc Merge git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc-merge 2006-04-18 10:34:24 -07:00
s390 [PATCH] arch/s390/Makefile: remove -finline-limit=10000 2006-04-11 06:18:39 -07:00
sh [PATCH] No arch-specific strpbrk implementations 2006-04-11 06:18:40 -07:00
sh64 [PATCH] No arch-specific strpbrk implementations 2006-04-11 06:18:40 -07:00
sparc [SPARC]: Hook up sys_tee() into syscall tables. 2006-04-14 15:29:32 -07:00
sparc64 [SPARC64]: Fix dependencies of HUGETLB_PAGE_SIZE_64K 2006-04-18 16:19:53 -07:00
um [PATCH] uml: MADV_REMOVE fixes 2006-04-19 09:13:49 -07:00
v850 [PATCH] No arch-specific strpbrk implementations 2006-04-11 06:18:40 -07:00
x86_64 [PATCH] x86_64: Add tee and sync_file_range 2006-04-18 10:39:20 -07:00
xtensa [PATCH] No arch-specific strpbrk implementations 2006-04-11 06:18:40 -07:00