Messages in this thread | | | Date | Fri, 3 Oct 2008 09:27:22 +0200 | From | Ingo Molnar <> | Subject | Re: [PATCH] ring_buffer: map to cpu not page |
| |
* Steven Rostedt <rostedt@goodmis.org> wrote:
> My original patch had a compile bug when NUMA was configured. I > referenced cpu when it should have been cpu_buffer->cpu. > > Ingo quickly fixed this bug by replacing cpu with 'i' because that was > the loop counter. Unfortunately, the 'i' was the counter of pages, not > CPUs. This caused a crash when the number of pages allocated for the > buffers exceeded the number of pages, which would usually be the case. > > Signed-off-by: Steven Rostedt <srostedt@redhat.com>
> for (i = 0; i < nr_pages; i++) { > page = kzalloc_node(ALIGN(sizeof(*page), cache_line_size()), > - GFP_KERNEL, cpu_to_node(i)); > + GFP_KERNEL, cpu_to_node(cpu_buffer->cpu));
oh, stupid typo of the year :-)
applied to tip/tracing/ring-buffer, thanks for tracking it down! I've reactivated the topic branch for tip/master and i'm running a few tests before pushing it out for wider testing.
Ingo
| |