Re: Tiger oops in ia64_sal_physical_id_info (was [RFC] regression: 113134fcbca83619be4c68d0ca66db6093777b5d)

On Mon, 2008-02-25 at 16:08 -0700, Alex Chiang wrote: > * Bjorn Helgaas <[EMAIL PROTECTED]>: > > On Friday 22 February 2008 12:28:26 am Shaohua Li wrote: > > > My tiger machine hangs since 2.6.23 with commit above. I always saw oops > > > in ia64_sal_

[patch 3/3] remove duplicate code for register access

We have duplicate code to access registers (access_uarea and regset way). They just have different layout, so remove duplicate code. Signed-off-by: Shaohua Li <[EMAIL PROTECTED]> --- arch/ia64/kernel/ptrace.c | 522 +- 1 file changed, 200 inse

[patch 1/3] writeback regset[0] in coredump

ocess_vm to access user RSE. Signed-off-by: Shaohua Li <[EMAIL PROTECTED]> --- fs/binfmt_elf.c |8 1 file changed, 8 insertions(+) Index: linux/fs/binfmt_elf.c === --- linux.orig/fs/binfmt_elf.c 2008-02-25

[patch 2/3] regset IA64 implementation

This is the regset implementation under IA64. Basically register read/write, which is derived from current ptrace register read/write. Signed-off-by: Shaohua Li <[EMAIL PROTECTED]> --- arch/ia64/ia32/sys_ia32.c | 497 +++ arch/ia64/kernel/process.c | 30 -

[RFC] regression: 113134fcbca83619be4c68d0ca66db6093777b5d

My tiger machine hangs since 2.6.23 with commit above. I always saw oops in ia64_sal_physical_id_info(). In 2.6.22, if ia64_pal_logical_to_phys returns UNIMPLENTED, ia64_sal_physical_id_info() isn't called. Below patch fixes the issue. diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/sm

Re: [PATCH]Fix the order of atomic operations in restore_previous_kprobes on ia64

On Sat, 2008-01-19 at 00:20 +0800, Masami Hiramatsu wrote: > From: Masami Hiramatsu <[EMAIL PROTECTED]> > > Fix the order of atomic operations to prevent overwriting > prev_kprobe[0]. > To pop values from stack, we must decrement stack index right AFTER > reading values. > > Signed-off-by: Masam

RE: [PATCH] ptrace RSE bug

On Tue, 2007-11-13 at 12:07 +0100, Petr Tesarik wrote: > On Mon, 2007-11-12 at 16:30 -0800, Roland McGrath wrote: > > [...] > > If you do the artificial test using a long sleep in arch_ptrace_stop, > > then you can probably produce this by hand with gdb. Have the process > > doing raise(SIGCHLD) o

[PATCH] kprobe: make kreturn probe handler stack unwind correct

Restore regs->ccr_iip before kreturn probe handler runs. In this way, if probe handler does unwind, unwind can correctly get the stack trace. http://sourceware.org/bugzilla/show_bug.cgi?id=5051 Signed-off-by: Shaohua Li <[EMAIL PROTECTED]> Index: linux/arch/ia64/kernel/

RE: [PATCH] ptrace RSE bug

On Thu, 2007-10-25 at 07:38 +0800, Luck, Tony wrote: > > Thanks. So Tony, how do think about the IA64 part of the patch? > > Having to use ia64_peek() and ia64_poke() to copy the RBS 8 bytes > at a time looks a bit ugly ... but this isn't in the performance > path (unless you are running your whol

Re: [PATCH] ptrace RSE bug

On Fri, 2007-10-19 at 21:42 +0200, Petr Tesarik wrote: > On Thu, 2007-10-18 at 12:59 +0200, Petr Tesarik wrote: > > Shaohua Li wrote: > > > On Wed, 2007-10-17 at 16:56 +0200, Petr Tesarik wrote: > > >> Shaohua Li wrote: > > >>> On Fri, 2007-09-

Re: [PATCH] ptrace RSE bug

On Thu, 2007-10-18 at 12:59 +0200, Petr Tesarik wrote: > Shaohua Li wrote: > > On Wed, 2007-10-17 at 16:56 +0200, Petr Tesarik wrote: > >> Shaohua Li wrote: > >>> On Fri, 2007-09-07 at 09:11 -0600, David Mosberger-Tang wrote: > >>>> Anything that avoi

Re: [PATCH] ptrace RSE bug

On Wed, 2007-10-17 at 16:56 +0200, Petr Tesarik wrote: > Shaohua Li wrote: > > On Fri, 2007-09-07 at 09:11 -0600, David Mosberger-Tang wrote: > >> Anything that avoids complicating the kernel exit path is worth doing! > >> The exit path is complicated enough as

Re: [PATCH] ptrace RSE bug

-BEGIN PGP SIGNED MESSAGE- > > Hash: SHA1 > > > > Shaohua Li wrote: > > > On Thu, 2007-09-06 at 15:59 +0200, Petr Tesarik wrote: > > >>[...] > > >> So, what happens if upon syscall entry notification the debugger > > >> modifies

Re: [PATCH] ptrace RSE bug

On Thu, 2007-09-06 at 15:59 +0200, Petr Tesarik wrote: > On Thu, 2007-09-06 at 11:16 +0800, Shaohua Li wrote: > > On Wed, 2007-09-05 at 18:25 +0200, Petr Tesarik wrote: > > > Shaohua Li wrote: > > > > This is base kernel patch for ptrace RSE bug. It's basically

Re: [PATCH] ptrace RSE bug

On Wed, 2007-09-05 at 18:25 +0200, Petr Tesarik wrote: > Shaohua Li wrote: > > This is base kernel patch for ptrace RSE bug. It's basically a backport > > from the utrace RSE patch I sent out several weeks ago. please review. > > > > when a thread is stopped

[PATCH] ptrace RSE bug

use the newer RSE to return to user. Signed-off-by: Shaohua Li <[EMAIL PROTECTED]> --- arch/ia64/kernel/perfmon.c | 21 +- arch/ia64/kernel/process.c | 15 +++ arch/ia64/kernel/ptrace.c | 82 + include/asm-ia64/ptrace.h

[PATCH]forbid ptrace changes psr.ri to 3

PSR.ri bits 3 are reserved, change to 3 will break kernel. The patch ignore such change, if you thought access_urea should return an error in the case, please let me know. Signed-off-by: Shaohua Li <[EMAIL PROTECTED]> --- a/arch/ia64/kernel/ptrace.c 2007-08-27 05:48:49.0 +0800

Re: [PATCH]utrace: IA64 RSE bug

scall trace enter, and kernel will use the parameters, so we do an extra synchronization for the case. Thanks, Shaohua > > On Wed, Aug 08, 2007 at 10:30:37AM +0800, Shaohua Li wrote: > > On Tue, 2007-08-07 at 15:13 +0800, Roland McGrath wrote: > > > That looks reasonable to me, not be

Re: [PATCH]utrace: IA64 RSE bug

r space RSE might be newer than kernel RSE. To avoid stale RSE is used when return to userspace, this patch synchronize user space RSE to kernel RSE. Also, as TIF_ALLWORK_MASK bits are limited, TIF_NOTIFY_RESUME is overrided. Signed-off-by: Bibo Mao

[PATCH]utrace: IA64 RSE bug

+ utrace patches) Signed-off-by: Bibo Mao<[EMAIL PROTECTED]> Signed-off-by: Shaohua Li<[EMAIL PROTECTED]> == arch/ia64/kernel/perfmon.c | 21 ++-- arch/ia64/kernel/process.c | 14 +++ arch