Home » Mailing list archives » linux-kernel » 2008 » April » 1

Re: spinlocks -- why are releases inlined and acquires are not?

!MAILaRCHIVE_VOTE_RePLACE
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
[view in full thread]
From: Ingo Molnar <mingo@...>
To: Jiri Kosina <jkosina@...>
Cc: <linux-kernel@...>
Subject: Re: spinlocks -- why are releases inlined and acquires are not?
Date: Tuesday, April 1, 2008 - 4:33 am

* Jiri Kosina <jkosina@suse.cz> wrote:


IIRC the main reason we decided to uninline them was image size. So i'd 
suggest for you to check how this change impacts vmlinux size (on both 
64-bit and 32-bit), a typical distro config (or allyesconfig with lock 
debugging disabled).

If you do the test on x86.git/latest you'll also have the 
CONFIG_OPTIMIZE_INLINING=y and CONFIG_CC_OPTIMIZE_FOR_SIZE=y combination 
as well, which generates the most compact x86 kernel image ever.

	Ingo
--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
Messages in current thread:
spinlocks -- why are releases inlined and acquires are not?, Jiri Kosina, (Mon Mar 31, 8:08 pm)
Re: spinlocks -- why are releases inlined and acquires are n..., Jiri Kosina, (Mon Mar 31, 8:40 pm)
Re: spinlocks -- why are releases inlined and acquires are n..., Ingo Molnar, (Tue Apr 1, 4:33 am)
Re: spinlocks -- why are releases inlined and acquires are n..., Jiri Kosina, (Tue Apr 1, 4:38 am)
Re: spinlocks -- why are releases inlined and acquires are n..., Andi Kleen, (Tue Apr 1, 7:12 am)
Re: spinlocks -- why are releases inlined and acquires are n..., Jiri Kosina, (Tue Apr 1, 8:40 am)
Re: spinlocks -- why are releases inlined and acquires are n..., Andi Kleen, (Tue Apr 1, 9:30 am)
Re: spinlocks -- why are releases inlined and acquires are n..., Ingo Molnar, (Tue Apr 1, 4:43 am)
Re: spinlocks -- why are releases inlined and acquires are n..., Jiri Kosina, (Tue Apr 1, 4:44 am)