| Submitter | Richard Henderson |
|---|---|
| Date | Nov. 1, 2012, 10:20 p.m. |
| Message ID | <1351808446-2436-1-git-send-email-rth@twiddle.net> |
| Download | mbox | patch |
| Permalink | /patch/196393/ |
| State | New |
| Headers | show |
Comments
Thanks, applied. On Thu, Nov 1, 2012 at 10:20 PM, Richard Henderson <rth@twiddle.net> wrote: > After allocating 32MB or more contiguous memory, huge pages > would seem to be ideal. > > Signed-off-by: Richard Henderson <rth@twiddle.net> > --- > exec.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/exec.c b/exec.c > index df67938..ab07f68 100644 > --- a/exec.c > +++ b/exec.c > @@ -602,6 +602,8 @@ static inline void code_gen_alloc(size_t tb_size) > exit(1); > } > > + qemu_madvise(code_gen_buffer, code_gen_buffer_size, QEMU_MADV_HUGEPAGE); > + > /* Steal room for the prologue at the end of the buffer. This ensures > (via the MAX_CODE_GEN_BUFFER_SIZE limits above) that direct branches > from TB's to the prologue are going to be in range. It also means > -- > 1.7.11.7 >
Patch
diff --git a/exec.c b/exec.c index df67938..ab07f68 100644 --- a/exec.c +++ b/exec.c @@ -602,6 +602,8 @@ static inline void code_gen_alloc(size_t tb_size) exit(1); } + qemu_madvise(code_gen_buffer, code_gen_buffer_size, QEMU_MADV_HUGEPAGE); + /* Steal room for the prologue at the end of the buffer. This ensures (via the MAX_CODE_GEN_BUFFER_SIZE limits above) that direct branches from TB's to the prologue are going to be in range. It also means
After allocating 32MB or more contiguous memory, huge pages would seem to be ideal. Signed-off-by: Richard Henderson <rth@twiddle.net> --- exec.c | 2 ++ 1 file changed, 2 insertions(+)