From: Shay Agroskin <shayagr@amazon.com>
To: Mel Gorman <mgorman@techsingularity.net>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Chuck Lever <chuck.lever@oracle.com>,
Jesper Dangaard Brouer <brouer@redhat.com>,
LKML <linux-kernel@vger.kernel.org>,
Linux-Net <netdev@vger.kernel.org>, Linux-MM <linux-mm@kvack.org>,
Linux-NFS <linux-nfs@vger.kernel.org>
Subject: Re: [PATCH 2/5] mm/page_alloc: Add a bulk page allocator
Date: Wed, 10 Mar 2021 13:04:17 +0200 [thread overview]
Message-ID: <pj41zl7dmfnuby.fsf@u570694869fb251.ant.amazon.com> (raw)
In-Reply-To: <20210301161200.18852-3-mgorman@techsingularity.net>
Mel Gorman <mgorman@techsingularity.net> writes:
>
> diff --git a/include/linux/gfp.h b/include/linux/gfp.h
> index 8572a1474e16..4903d1cc48dc 100644
> --- a/include/linux/gfp.h
> +++ b/include/linux/gfp.h
> @@ -515,6 +515,10 @@ static inline int
> arch_make_page_accessible(struct page *page)
> }
> #endif
>
> +int __alloc_pages_bulk_nodemask(gfp_t gfp_mask, int
> preferred_nid,
> + nodemask_t *nodemask, int
> nr_pages,
> + struct list_head *list);
> +
> struct page *
> __alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order, int
> preferred_nid,
> nodemask_t
> *nodemask);
> @@ -525,6 +529,14 @@ __alloc_pages(gfp_t gfp_mask, unsigned int
> order, int preferred_nid)
> return __alloc_pages_nodemask(gfp_mask, order,
> preferred_nid, NULL);
> }
>
> +/* Bulk allocate order-0 pages */
> +static inline unsigned long
> +alloc_pages_bulk(gfp_t gfp_mask, unsigned long nr_pages, struct
> list_head *list)
> +{
> + return __alloc_pages_bulk_nodemask(gfp_mask,
> numa_mem_id(), NULL,
> + nr_pages,
> list);
Is the second line indentation intentional ? Why not align it to
the first argument (gfp_mask) ?
> +}
> +
> /*
> * Allocate pages, preferring the node given as nid. The node
> must be valid and
> * online. For more general interface, see alloc_pages_node().
> @@ -594,6 +606,7 @@ void * __meminit alloc_pages_exact_nid(int
> nid, size_t size, gfp_t gfp_mask);
>
> extern void __free_pages(struct page *page, unsigned int
> order);
> extern void free_pages(unsigned long addr, unsigned int order);
> +extern void free_pages_bulk(struct list_head *list);
>
> struct page_frag_cache;
> extern void __page_frag_cache_drain(struct page *page, unsigned
> int count);
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index 3e4b29ee2b1e..ff1e55793786 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -4436,6 +4436,21 @@ static void wake_all_kswapds(unsigned int
> order, gfp_t gfp_mask,
> }
> }
> ...
>
> +/*
> + * This is a batched version of the page allocator that
> attempts to
> + * allocate nr_pages quickly from the preferred zone and add
> them to list.
> + */
> +int __alloc_pages_bulk_nodemask(gfp_t gfp_mask, int
> preferred_nid,
> + nodemask_t *nodemask, int nr_pages,
> + struct list_head *alloc_list)
> +{
> + struct page *page;
> + unsigned long flags;
> + struct zone *zone;
> + struct zoneref *z;
> + struct per_cpu_pages *pcp;
> + struct list_head *pcp_list;
> + struct alloc_context ac;
> + gfp_t alloc_mask;
> + unsigned int alloc_flags;
> + int alloced = 0;
Does alloced count the number of allocated pages ? Do you mind
renaming it to 'allocated' ?
> +
> + if (nr_pages == 1)
> + goto failed;
> +
> + /* May set ALLOC_NOFRAGMENT, fragmentation will return 1
> page. */
> + if (!prepare_alloc_pages(gfp_mask, 0, preferred_nid,
> nodemask, &ac, &alloc_mask, &alloc_flags))
> + return 0;
> + gfp_mask = alloc_mask;
> +
> + /* Find an allowed local zone that meets the high
> watermark. */
> + for_each_zone_zonelist_nodemask(zone, z, ac.zonelist,
> ac.highest_zoneidx, ac.nodemask) {
> + unsigned long mark;
> +
> + if (cpusets_enabled() && (alloc_flags &
> ALLOC_CPUSET) &&
> + !__cpuset_zone_allowed(zone, gfp_mask)) {
> + continue;
> + }
> +
> + if (nr_online_nodes > 1 && zone !=
> ac.preferred_zoneref->zone &&
> + zone_to_nid(zone) !=
> zone_to_nid(ac.preferred_zoneref->zone)) {
> + goto failed;
> + }
> +
> + mark = wmark_pages(zone, alloc_flags &
> ALLOC_WMARK_MASK) + nr_pages;
> + if (zone_watermark_fast(zone, 0, mark,
> +
> zonelist_zone_idx(ac.preferred_zoneref),
> + alloc_flags, gfp_mask)) {
> + break;
> + }
> + }
> + if (!zone)
> + return 0;
> +
> + /* Attempt the batch allocation */
> + local_irq_save(flags);
> + pcp = &this_cpu_ptr(zone->pageset)->pcp;
> + pcp_list = &pcp->lists[ac.migratetype];
> +
> + while (alloced < nr_pages) {
> + page = __rmqueue_pcplist(zone, ac.migratetype,
> alloc_flags,
> +
> pcp, pcp_list);
Same indentation comment as before
> + if (!page)
> + break;
> +
> + prep_new_page(page, 0, gfp_mask, 0);
> + list_add(&page->lru, alloc_list);
> + alloced++;
> + }
> +
> + if (!alloced)
> + goto failed_irq;
> +
> + if (alloced) {
> + __count_zid_vm_events(PGALLOC, zone_idx(zone),
> alloced);
> + zone_statistics(zone, zone);
> + }
> +
> + local_irq_restore(flags);
> +
> + return alloced;
> +
> +failed_irq:
> + local_irq_restore(flags);
> +
> +failed:
> + page = __alloc_pages_nodemask(gfp_mask, 0, preferred_nid,
> nodemask);
> + if (page) {
> + alloced++;
> + list_add(&page->lru, alloc_list);
> + }
> +
> + return alloced;
> +}
> +EXPORT_SYMBOL_GPL(__alloc_pages_bulk_nodemask);
> +
> /*
> * This is the 'heart' of the zoned buddy allocator.
> */
> @@ -4981,8 +5092,6 @@ __alloc_pages_nodemask(gfp_t gfp_mask,
> unsigned int order, int preferred_nid,
> return NULL;
> }
>
> - gfp_mask &= gfp_allowed_mask;
> - alloc_mask = gfp_mask;
> if (!prepare_alloc_pages(gfp_mask, order, preferred_nid,
> nodemask, &ac, &alloc_mask, &alloc_flags))
> return NULL;
next prev parent reply other threads:[~2021-03-10 11:12 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-01 16:11 [PATCH 0/5] Introduce a bulk order-0 page allocator with two in-tree users Mel Gorman
2021-03-01 16:11 ` [PATCH 1/5] SUNRPC: Set rq_page_end differently Mel Gorman
2021-03-01 16:11 ` [PATCH 2/5] mm/page_alloc: Add a bulk page allocator Mel Gorman
2021-03-09 17:12 ` Christoph Hellwig
2021-03-09 18:10 ` Mel Gorman
2021-03-10 11:04 ` Shay Agroskin [this message]
2021-03-10 11:38 ` Mel Gorman
2021-03-12 12:01 ` Jesper Dangaard Brouer
2021-03-01 16:11 ` [PATCH 3/5] SUNRPC: Refresh rq_pages using " Mel Gorman
2021-03-01 16:11 ` [PATCH 4/5] net: page_pool: refactor dma_map into own function page_pool_dma_map Mel Gorman
2021-03-02 18:49 ` Ilias Apalodimas
2021-03-03 9:18 ` Mel Gorman
2021-03-03 10:19 ` Jesper Dangaard Brouer
2021-03-01 16:12 ` [PATCH 5/5] net: page_pool: use alloc_pages_bulk in refill code path Mel Gorman
2021-03-02 19:05 ` Ilias Apalodimas
2021-03-10 10:46 [PATCH 0/5] Introduce a bulk order-0 page allocator with two in-tree users Mel Gorman
2021-03-10 10:46 ` [PATCH 2/5] mm/page_alloc: Add a bulk page allocator Mel Gorman
2021-03-10 23:46 ` Andrew Morton
2021-03-11 8:42 ` Mel Gorman
2021-03-12 11:46 ` Jesper Dangaard Brouer
2021-03-12 13:44 ` Mel Gorman
2021-03-12 14:58 ` Matthew Wilcox
2021-03-12 16:03 ` Mel Gorman
2021-03-12 21:08 ` Matthew Wilcox
2021-03-13 13:16 ` Mel Gorman
2021-03-13 16:39 ` Matthew Wilcox
2021-03-13 16:56 ` Chuck Lever III
2021-03-13 19:33 ` Matthew Wilcox
2021-03-14 12:52 ` Mel Gorman
2021-03-14 15:22 ` Chuck Lever III
2021-03-15 10:42 ` Mel Gorman
2021-03-15 16:42 ` Jesper Dangaard Brouer
2021-03-19 17:10 ` Jesper Dangaard Brouer
2021-03-12 12:43 ` Matthew Wilcox
2021-03-12 14:15 ` Mel Gorman
2021-03-11 11:49 [PATCH 0/5 v3] Introduce a bulk order-0 page allocator with two in-tree users Mel Gorman
2021-03-11 11:49 ` [PATCH 2/5] mm/page_alloc: Add a bulk page allocator Mel Gorman
2021-03-11 16:42 ` Alexander Duyck
2021-03-12 7:32 ` Mel Gorman
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=pj41zl7dmfnuby.fsf@u570694869fb251.ant.amazon.com \
--to=shayagr@amazon.com \
--cc=akpm@linux-foundation.org \
--cc=brouer@redhat.com \
--cc=chuck.lever@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-nfs@vger.kernel.org \
--cc=mgorman@techsingularity.net \
--cc=netdev@vger.kernel.org \
/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).