Re: [PATCH 07/15] mm: Add ___GFP_NOTRACE
[Posted March 4, 2015 by corbet]
| From: |
| Steven Rostedt <rostedt-AT-goodmis.org> |
| To: |
| Tom Zanussi <tom.zanussi-AT-linux.intel.com> |
| Subject: |
| Re: [PATCH 07/15] mm: Add ___GFP_NOTRACE |
| Date: |
| Mon, 2 Mar 2015 11:37:01 -0500 |
| Message-ID: |
| <20150302113701.2171bdd5@gandalf.local.home> |
| Cc: |
| masami.hiramatsu.pt-AT-hitachi.com, namhyung-AT-kernel.org, andi-AT-firstfloor.org, ast-AT-plumgrid.com, linux-kernel-AT-vger.kernel.org |
| Archive‑link: | |
Article |
On Mon, 2 Mar 2015 10:01:00 -0600
Tom Zanussi <tom.zanussi@linux.intel.com> wrote:
> Add a gfp flag that allows kmalloc() et al to be used in tracing
> functions.
>
> The problem with using kmalloc for tracing is that the tracing
> subsystem should be able to trace kmalloc itself, which it can't do
> directly because of paths like kmalloc()->trace_kmalloc()->kmalloc()
> or kmalloc()->trace_mm_page_alloc()->kmalloc().
This part I don't like at all. Why can't the memory be preallocated
when the hist is created (the echo 'hist:...')?
kmalloc must never be called from any tracepoint callback.
This change is currently a showstopper.
-- Steve
>
> With this flag, tracing code could use a special version of kmalloc()
> that sets __GFP_NOTRACE on every allocation it does, while leaving the
> normal kmalloc() path untouched.
>
> This would allow any tracepoint in the kmalloc path to be avoided via
> DEFINE_EVENT_CONDITION() redefinitions of those events, which check
> for ___GFP_NOTRACE immediately in their execution and break if set,
> thereby avoiding the recursion.
>
