Messages in this thread | | | Date | Wed, 18 Feb 2009 17:05:35 +0100 | From | Ingo Molnar <> | Subject | Re: [PATCH 2/4] generic-smp: remove kmalloc usage |
| |
* Rusty Russell <rusty@rustcorp.com.au> wrote:
> On Tuesday 17 February 2009 20:13:59 Ingo Molnar wrote: > > We should not bend backwards trying to preserve that kmalloc() > > [and prove that it's safe and race-free] - i.e. the burden of > > proof is on the person insisting that it's needed, not on the > > person wanting to remove it. > > Respectfully disagree. The kmalloc has been there for a very long time, > and doing fine AFAICT.
The kmalloc(GFP_ATOMIC) has been in kernel/smp.c for about half a year, since:
| commit 3d4422332711ef48ef0f132f1fcbfcbd56c7f3d1 | Author: Jens Axboe <jens.axboe@oracle.com> | Date: Thu Jun 26 11:21:34 2008 +0200 | Subject: [PATCH] Add generic helpers for arch IPI function calls
Which introduced kernel/smp.c to begin with. It got merged on July 15.
The precedecessor mechanism, smp_call_function*() on x86, used an on-kernel-stack buffer (and a half-wait mechanism), even in the async case, not a kmalloc().
I'm not sure where the "kmalloc has been there for a very long time" statement comes from. Are we talking about the same thing?
Ingo
| |