This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
Re: [RFC][PATCH][take2]kretprobe: kretprobe-booster against 2.6.15-rc5-mm3 for i386
- From: Masami Hiramatsu <hiramatu at sdl dot hitachi dot co dot jp>
- To: ananth at in dot ibm dot com
- Cc: maneesh at in dot ibm dot com, anil dot s dot keshavamurthy at intel dot com, systemtap at sources dot redhat dot com, Yumiko Sugita <sugita at sdl dot hitachi dot co dot jp>, Satoshi Oshima <soshima at redhat dot com>, Hideo Aoki <haoki at redhat dot com>
- Date: Tue, 24 Jan 2006 22:23:24 +0900
- Subject: Re: [RFC][PATCH][take2]kretprobe: kretprobe-booster against 2.6.15-rc5-mm3 for i386
- References: <43AAA8FF.1070404@sdl.hitachi.co.jp> <20060124092205.GC3607@in.ibm.com>
Hi, Ananth
Ananth N Mavinakayanahalli wrote:
>>+ " movl %esp, %eax\n"
>>+ " pushl %eax\n"
>>+ " addl $60, %eax\n"
>>+ " movl %eax, 56(%esp)\n"
>>+ " movl arch_trampoline_callee, %eax\n"
> ^^^^
> Why do we need this aliasing of trampoline_handler to
> arch_trampoline_callee? Can't we just invoke trampoline_handler()
> instead?
OK,
I had thought that inline assembler could not access to C function,
and that patch was based on that thought. However, now I know it is
possible as below.
--
" movl $trampoline_handler, %eax\n"
" call *%eax\n"
---
After all, I will resend clean-up patch to this ML.
--
Masami HIRAMATSU
2nd Research Dept.
Hitachi, Ltd., Systems Development Laboratory
E-mail: hiramatu@sdl.hitachi.co.jp