Messages in this thread Patch in this message | | | Subject | Re: [PATCH] perf_counter: allow and require one-page mmap on counting counters | From | Peter Zijlstra <> | Date | Wed, 25 Mar 2009 12:48:31 +0100 |
| |
On Wed, 2009-03-25 at 19:15 +1100, Paul Mackerras wrote: > Impact: bug fix > > Currently the mmap code requires that the length of the mmap be at least > two pages. That is fine for sampling counters, but for counting > counters the second and subsequent pages are just wasted, since counting > counters don't generate events. > > This changes the code to require that the mmap be one page in length > for counting counters, and applies the existing check to sampling > counters.
Does the below work for you Paul?
--- Subject: perf_counter: allow one-page mmap on counters From: Peter Zijlstra <a.p.zijlstra@chello.nl> Date: Wed Mar 25 12:44:16 CET 2009
A brainfart stopped single page mmap()s working. The rest of the code should be perfectly fine with not having any data pages.
Reported-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> --- kernel/perf_counter.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
Index: linux-2.6/kernel/perf_counter.c =================================================================== --- linux-2.6.orig/kernel/perf_counter.c +++ linux-2.6/kernel/perf_counter.c @@ -1369,7 +1369,11 @@ static int perf_mmap(struct file *file, vma_size = vma->vm_end - vma->vm_start; nr_pages = (vma_size / PAGE_SIZE) - 1; - if (nr_pages == 0 || !is_power_of_2(nr_pages)) + /* + * If we have data pages ensure they're a power-of-two number, so we + * can do bitmasks instead of modulo. + */ + if (nr_pages != 0 && !is_power_of_2(nr_pages)) return -EINVAL; if (vma_size != PAGE_SIZE * (1 + nr_pages))
| |