From: Adam Litke <agl@us.ibm.com>
To: npiggin@suse.de
Cc: linux-mm@kvack.org, kniht@us.ibm.com, andi-suse@firstfloor.org,
nacc@us.ibm.com, abh@cray.com, joachim.deguara@amd.com
Subject: Re: [patch 11/23] hugetlb: support larger than MAX_ORDER
Date: Tue, 27 May 2008 16:23:38 -0500 [thread overview]
Message-ID: <1211923418.12036.38.camel@localhost.localdomain> (raw)
In-Reply-To: <20080525143453.269965000@nick.local0.net>
On Mon, 2008-05-26 at 00:23 +1000, npiggin@suse.de wrote:
> @@ -549,6 +560,51 @@ static struct page *alloc_huge_page(stru
> return page;
> }
>
> +static __initdata LIST_HEAD(huge_boot_pages);
> +
> +struct huge_bootmem_page {
> + struct list_head list;
> + struct hstate *hstate;
> +};
> +
> +static int __init alloc_bootmem_huge_page(struct hstate *h)
> +{
> + struct huge_bootmem_page *m;
> + int nr_nodes = nodes_weight(node_online_map);
> +
> + while (nr_nodes) {
> + m = __alloc_bootmem_node_nopanic(NODE_DATA(h->hugetlb_next_nid),
> + huge_page_size(h), huge_page_size(h),
> + 0);
> + if (m)
> + goto found;
> + hstate_next_node(h);
> + nr_nodes--;
> + }
> + return 0;
> +
> +found:
> + BUG_ON((unsigned long)virt_to_phys(m) & (huge_page_size(h) - 1));
> + /* Put them into a private list first because mem_map is not up yet */
> + list_add(&m->list, &huge_boot_pages);
> + m->hstate = h;
> + return 1;
> +}
At first I was pretty confused by how you are directly using the
newly-allocated bootmem page to create a temporary list until the mem
map comes up. Clever. I bet I would have understood right away if it
were written like the following:
void *vaddr;
struct huge_bootmem_page *m;
vaddr = __alloc_bootmem_node_nopanic(...);
if (vaddr) {
/*
* Use the beginning of this block to store some temporary
* meta-data until the mem_map comes up.
*/
m = (huge_bootmem_page *) vaddr;
goto found;
}
If you don't like that level of verbosity, could we add a comment just
to make it immediately clear to the reader?
> +/* Put bootmem huge pages into the standard lists after mem_map is up */
> +static void __init gather_bootmem_prealloc(void)
> +{
> + struct huge_bootmem_page *m;
> + list_for_each_entry (m, &huge_boot_pages, list) {
> + struct page *page = virt_to_page(m);
> + struct hstate *h = m->hstate;
> + __ClearPageReserved(page);
> + WARN_ON(page_count(page) != 1);
> + prep_compound_page(page, h->order);
> + prep_new_huge_page(h, page, page_to_nid(page));
> + }
> +}
> +
> static void __init hugetlb_init_one_hstate(struct hstate *h)
> {
> unsigned long i;
--
Adam Litke - (agl at us.ibm.com)
IBM Linux Technology Center
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2008-05-27 21:23 UTC|newest]
Thread overview: 88+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-05-25 14:23 [patch 00/23] multi size, giant hugetlb support, 1GB for x86, 16GB for powerpc npiggin
2008-05-25 14:23 ` [patch 01/23] hugetlb: fix lockdep error npiggin
2008-05-27 16:30 ` Nishanth Aravamudan
2008-05-27 19:55 ` Adam Litke
2008-05-25 14:23 ` [patch 02/23] hugetlb: factor out huge_new_page npiggin
2008-05-27 16:31 ` Nishanth Aravamudan
2008-05-27 20:03 ` Adam Litke
2008-05-25 14:23 ` [patch 03/23] hugetlb: modular state npiggin
2008-05-27 16:44 ` Nishanth Aravamudan
2008-05-28 8:40 ` Nick Piggin
2008-05-27 20:38 ` Adam Litke
2008-05-28 9:13 ` Nick Piggin
2008-05-25 14:23 ` [patch 04/23] hugetlb: multiple hstates npiggin
2008-05-27 16:52 ` Nishanth Aravamudan
2008-05-27 20:43 ` Adam Litke
2008-05-25 14:23 ` [patch 05/23] hugetlb: multi hstate proc files npiggin
2008-05-29 5:07 ` Nishanth Aravamudan
2008-05-29 5:44 ` Nishanth Aravamudan
2008-05-29 6:30 ` Nishanth Aravamudan
2008-05-29 9:04 ` Nick Piggin
2008-05-25 14:23 ` [patch 06/23] hugetlbfs: per mount hstates npiggin
2008-05-27 16:58 ` Nishanth Aravamudan
2008-05-27 20:50 ` Adam Litke
2008-05-25 14:23 ` [patch 07/23] hugetlb: multi hstate sysctls npiggin
2008-05-27 21:00 ` Adam Litke
2008-05-28 9:59 ` Nick Piggin
2008-05-29 4:59 ` Nishanth Aravamudan
2008-05-29 5:36 ` Nishanth Aravamudan
2008-05-29 8:59 ` Nick Piggin
2008-05-29 6:39 ` [RFC][PATCH 1/2] hugetlb: present information in sysfs Nishanth Aravamudan
2008-05-29 6:42 ` [RFC][PATCH 2/2] hugetlb: remove multi-valued proc files Nishanth Aravamudan
2008-05-30 3:51 ` Nick Piggin
2008-05-30 7:43 ` Nishanth Aravamudan
2008-05-30 2:58 ` [RFC][PATCH 1/2] hugetlb: present information in sysfs Greg KH
2008-05-30 3:37 ` Nick Piggin
2008-05-30 4:21 ` Greg KH
2008-05-30 4:28 ` Nick Piggin
2008-05-30 7:44 ` Nishanth Aravamudan
2008-05-30 7:41 ` Nishanth Aravamudan
2008-05-30 13:40 ` Adam Litke
2008-05-30 7:39 ` Nishanth Aravamudan
2008-05-25 14:23 ` [patch 08/23] hugetlb: abstract numa round robin selection npiggin
2008-05-27 17:01 ` Nishanth Aravamudan
2008-05-27 21:02 ` Adam Litke
2008-05-25 14:23 ` [patch 09/23] mm: introduce non panic alloc_bootmem npiggin
2008-05-25 14:23 ` [patch 10/23] mm: export prep_compound_page to mm npiggin
2008-05-27 21:05 ` Adam Litke
2008-05-25 14:23 ` [patch 11/23] hugetlb: support larger than MAX_ORDER npiggin
2008-05-27 21:23 ` Adam Litke [this message]
2008-05-28 10:22 ` Nick Piggin
2008-05-25 14:23 ` [patch 12/23] hugetlb: support boot allocate different sizes npiggin
2008-05-27 17:04 ` Nishanth Aravamudan
2008-05-27 21:28 ` Adam Litke
2008-05-28 10:57 ` Nick Piggin
2008-05-28 14:01 ` Nick Piggin
2008-05-28 14:35 ` Adam Litke
2008-05-25 14:23 ` [patch 13/23] hugetlb: printk cleanup npiggin
2008-05-27 17:05 ` Nishanth Aravamudan
2008-05-27 21:30 ` Adam Litke
2008-05-25 14:23 ` [patch 14/23] hugetlb: introduce huge_pud npiggin
2008-05-26 11:09 ` Hugh Dickins
2008-05-27 2:24 ` Nick Piggin
2008-05-25 14:23 ` [patch 15/23] x86: support GB hugepages on 64-bit npiggin
2008-05-27 21:35 ` Adam Litke
2008-05-25 14:23 ` [patch 16/23] x86: add hugepagesz option " npiggin
2008-05-25 14:23 ` [patch 17/23] hugetlb: do not always register default HPAGE_SIZE huge page size npiggin
2008-05-27 21:39 ` Adam Litke
2008-05-25 14:23 ` [patch 18/23] hugetlb: allow arch overried hugepage allocation npiggin
2008-05-27 21:41 ` Adam Litke
2008-05-25 14:23 ` [patch 19/23] powerpc: function to allocate gigantic hugepages npiggin
2008-05-27 21:44 ` Adam Litke
2008-05-25 14:23 ` [patch 20/23] powerpc: scan device tree for gigantic pages npiggin
2008-05-27 21:47 ` Adam Litke
2008-05-25 14:23 ` [patch 21/23] powerpc: define support for 16G hugepages npiggin
2008-05-25 14:23 ` [patch 22/23] fs: check for statfs overflow npiggin
2008-05-27 17:14 ` Nishanth Aravamudan
2008-05-27 17:19 ` Jon Tollefson
2008-05-28 9:02 ` Nick Piggin
2008-05-29 23:56 ` Andreas Dilger
2008-05-30 0:12 ` Nishanth Aravamudan
2008-05-30 1:14 ` Nick Piggin
2008-06-02 3:16 ` Andreas Dilger
2008-06-03 3:27 ` Nick Piggin
2008-06-03 17:17 ` Andreas Dilger
2008-05-25 14:23 ` [patch 23/23] powerpc: support multiple hugepage sizes npiggin
2008-05-27 17:14 ` Nishanth Aravamudan
2008-05-28 8:49 ` Nick Piggin
2008-05-25 14:42 ` [patch 00/23] multi size, giant hugetlb support, 1GB for x86, 16GB for powerpc Nick Piggin
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1211923418.12036.38.camel@localhost.localdomain \
--to=agl@us.ibm.com \
--cc=abh@cray.com \
--cc=andi-suse@firstfloor.org \
--cc=joachim.deguara@amd.com \
--cc=kniht@us.ibm.com \
--cc=linux-mm@kvack.org \
--cc=nacc@us.ibm.com \
--cc=npiggin@suse.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).