From: Hui Zhu <teawater@gmail.com>
To: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: "Américo Wang" <xiyou.wangcong@gmail.com>,
linux-kernel@vger.kernel.org, gdb@sourceware.org,
hellogcc@freelists.org, linux-kbuild@vger.kernel.org
Subject: Re: [PATCH] Built kernel without -O2 option
Date: Wed, 01 Dec 2010 12:18:00 -0000 [thread overview]
Message-ID: <AANLkTikq32EDVfP8SjNJSih2Z5jNafHGRPCDdc+w1GR6@mail.gmail.com> (raw)
In-Reply-To: <4CF37603.5080301@de.ibm.com>
On Mon, Nov 29, 2010 at 17:44, Christian Borntraeger
<borntraeger@de.ibm.com> wrote:
> Am 29.11.2010 09:16, schrieb Américo Wang:
>> On Mon, Nov 29, 2010 at 11:56:15AM +0800, Hui Zhu wrote:
>>> Hi,
>>>
>>> Now, there are a lot of ways to debug the Linux kernel with GDB, like
>>> qemu, kgtp or kgdb and so on.
>>> But the developer more like add a printk. It have a lot of reason, a big one is:
>>> (gdb) p ret
>>> $3 = <value optimized out>
>>> And the code execution order is not right.
>>>
>>> This is becuase the Kernel is bult with gcc -O2. Gcc will not
>>> generate enough debug message with file with -O2.
>>> So GDB cannot work very well with Linux kernel.
>>>
>>> So I make a patch that add a option in "Kernel hacking" called "Close
>>> GCC optimization". It will make kernel be built without -O2.
>>>
>>> I built and use it in i386 and x86_64. I will try to make it OK in other arch.
>>>
>>
>> The problem is that some functions _have to_ be inlined and gcc without -O2
>> doesn't inline them. Have check all the cases? I doubt.
>
> In essence -O2 just tells gcc to activate a list of optimizations
> gcc -Q -O2 --help=optimizers
> tells you what.
>
> So what about making this patch much smaller by explicitely using the optimizations
> that are absolutely necessary?
> e.g:
> -finline-small-functions -finline-functions-called-once
> (what else do we need?)
>
> We might even be able to collapse this with the optimize for size option,
> by providing a Kconfig entry that allows to choose between
>
> -O0 -finline-small-functions -finline-functions-called-once
> -O1 -finline-small-functions -finline-functions-called-once
> -Os
> -O2
> -O3
>
> Christian
>
Hi Christian.
If we can build without any optimization options just set some file to O2.
I think keep it it will better than add some optimization options to
all files, right?
But make for this file that I add O2 to them, I can change them to to
some others options like "-O0 -finline-small-functions
-finline-functions-called-once".
Thanks,
Hui
next prev parent reply other threads:[~2010-12-01 12:18 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-11-29 3:56 Hui Zhu
2010-11-29 8:11 ` Américo Wang
2010-11-29 8:18 ` microcai
2010-11-29 8:25 ` Hui Zhu
2010-11-29 8:48 ` Américo Wang
2010-11-29 9:11 ` Hui Zhu
2010-11-29 18:05 ` Valdis.Kletnieks
2010-11-29 20:51 ` richard -rw- weinberger
2010-11-29 21:00 ` Arnaud Lacombe
2010-11-29 21:07 ` richard -rw- weinberger
[not found] ` <4CF36741.9000808@loongson.cn>
2010-11-29 8:48 ` [hellogcc] " Hui Zhu
2010-11-29 8:55 ` Steven
2010-11-29 8:58 ` Hui Zhu
[not found] ` <4CF3762B.9010102@loongson.cn>
2010-11-29 18:11 ` Valdis.Kletnieks
2010-11-29 18:35 ` Steven J. Magnani
[not found] ` <20101129220350.GA10923@core2.telecom.by>
2010-12-01 13:22 ` Hui Zhu
2010-12-01 12:23 ` bekars
2010-12-01 13:24 ` Hui Zhu
2010-12-01 13:58 ` bekars
2010-12-01 15:02 ` Hui Zhu
2010-12-01 15:42 ` jovi zhang
2010-11-29 9:44 ` Christian Borntraeger
2010-11-29 11:12 ` Segher Boessenkool
2010-11-29 11:17 ` Christian Borntraeger
2010-11-29 11:57 ` Nicholas Mc Guire
2010-12-01 12:18 ` Hui Zhu [this message]
2010-11-29 9:59 ` Andi Kleen
2010-11-29 10:12 ` Mark Wielaard
2010-11-30 9:26 ` Américo Wang
2010-11-30 18:40 ` Michael Snyder
2010-12-01 12:52 ` Hui Zhu
2010-11-29 19:06 ` Jan Kratochvil
2010-11-29 21:07 ` Arnaud Lacombe
2010-11-30 22:40 ` Petr Hluzín
2010-12-01 13:42 ` Hui Zhu
2010-12-01 13:11 ` Hui Zhu
2011-01-24 7:48 ` Hui Zhu
2011-01-24 8:11 ` Kevin Pouget
2011-01-24 12:42 ` Hui Zhu
2011-01-28 3:41 ` Hui Zhu
2011-02-17 2:44 ` Hui Zhu
[not found] ` <AANLkTik+beFdZZHLyei044q8EPHwvmiKoFAvAHsNuhmw@mail.gmail.com>
2011-02-21 12:52 ` Kevin Pouget
2011-02-23 5:20 ` Hui Zhu
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=AANLkTikq32EDVfP8SjNJSih2Z5jNafHGRPCDdc+w1GR6@mail.gmail.com \
--to=teawater@gmail.com \
--cc=borntraeger@de.ibm.com \
--cc=gdb@sourceware.org \
--cc=hellogcc@freelists.org \
--cc=linux-kbuild@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=xiyou.wangcong@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox