Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1526901
| From | Eric Dumazet <eric.dumazet@gmail.com> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: Linux 4.9-rc6 |
| Date | 2016-11-21 17:50 +0100 |
| Message-ID | <sFZX3-3Fn-11@gated-at.bofh.it> (permalink) |
| References | (3 earlier) <sFLKq-2Qf-1@gated-at.bofh.it> <sFORX-4Q0-3@gated-at.bofh.it> <sFSiS-7gi-17@gated-at.bofh.it> <sFWZc-1K0-11@gated-at.bofh.it> <sFXiC-1QC-19@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
On Mon, 2016-11-21 at 05:51 -0800, Eric Dumazet wrote:
> + while (chunk_order) {
> + if (node == NUMA_NO_NODE)
> + page = alloc_pages(multi_alloc_mask, chunk_order);
> + else
> + page = alloc_pages_node(node, multi_alloc_mask, chunk_order);
> + if (page) {
> + split_page(page, chunk_order);
> + break;
> + }
> + chunk_order--;
> + }
We also could remember the page order with set_page_private() and
speedup show_numa_info()
I wonder if we could avoid the split_page() and speedup vfree().
Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread
Linux 4.9-rc6 Linus Torvalds <torvalds@linux-foundation.org> - 2016-11-20 23:10 +0100
Re: Linux 4.9-rc6 Eric Dumazet <eric.dumazet@gmail.com> - 2016-11-20 23:40 +0100
Re: Linux 4.9-rc6 Linus Torvalds <torvalds@linux-foundation.org> - 2016-11-21 00:30 +0100
Re: Linux 4.9-rc6 Al Viro <viro@ZenIV.linux.org.uk> - 2016-11-21 02:40 +0100
Re: Linux 4.9-rc6 Eric Dumazet <eric.dumazet@gmail.com> - 2016-11-21 06:00 +0100
Re: Linux 4.9-rc6 David Rientjes <rientjes@google.com> - 2016-11-21 09:40 +0100
Re: Linux 4.9-rc6 Eric Dumazet <eric.dumazet@gmail.com> - 2016-11-21 14:40 +0100
Re: Linux 4.9-rc6 Eric Dumazet <eric.dumazet@gmail.com> - 2016-11-21 15:00 +0100
Re: Linux 4.9-rc6 Eric Dumazet <eric.dumazet@gmail.com> - 2016-11-21 17:50 +0100
csiph-web