From: Andrew Morton <akpm@linux-foundation.org>
To: Mel Gorman <mgorman@suse.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>,
Andrea Arcangeli <aarcange@redhat.com>,
Ingo Molnar <mingo@kernel.org>,
Simon Jeons <simon.jeons@gmail.com>,
Wanpeng Li <liwanp@linux.vnet.ibm.com>,
Hugh Dickins <hughd@google.com>, Linux-MM <linux-mm@kvack.org>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 5/6] mm: Fold page->_last_nid into page->flags where possible
Date: Tue, 22 Jan 2013 14:46:59 -0800 [thread overview]
Message-ID: <20130122144659.d512e05c.akpm@linux-foundation.org> (raw)
In-Reply-To: <1358874762-19717-6-git-send-email-mgorman@suse.de>
On Tue, 22 Jan 2013 17:12:41 +0000
Mel Gorman <mgorman@suse.de> wrote:
> From: Peter Zijlstra <a.p.zijlstra@chello.nl>
>
> page->_last_nid fits into page->flags on 64-bit. The unlikely 32-bit NUMA
> configuration with NUMA Balancing will still need an extra page field.
> As Peter notes "Completely dropping 32bit support for CONFIG_NUMA_BALANCING
> would simplify things, but it would also remove the warning if we grow
> enough 64bit only page-flags to push the last-cpu out."
How much space remains in the 64-bit page->flags?
Was this the best possible use of the remaining space?
It's good that we can undo this later by flipping
LAST_NID_NOT_IN_PAGE_FLAGS.
> [mgorman@suse.de: Minor modifications]
> Signed-off-by: Mel Gorman <mgorman@suse.de>
Several of these patches are missing signoffs (Peter and Hugh).
>
> ...
>
> +static inline int page_last_nid(struct page *page)
> +{
> + return (page->flags >> LAST_NID_PGSHIFT) & LAST_NID_MASK;
> +}
> +
> +static inline int page_xchg_last_nid(struct page *page, int nid)
> +{
> + unsigned long old_flags, flags;
> + int last_nid;
> +
> + do {
> + old_flags = flags = page->flags;
> + last_nid = page_last_nid(page);
> +
> + flags &= ~(LAST_NID_MASK << LAST_NID_PGSHIFT);
> + flags |= (nid & LAST_NID_MASK) << LAST_NID_PGSHIFT;
> + } while (unlikely(cmpxchg(&page->flags, old_flags, flags) != old_flags));
> +
> + return last_nid;
> +}
> +
> +static inline void reset_page_last_nid(struct page *page)
> +{
> + page_xchg_last_nid(page, (1 << LAST_NID_SHIFT) - 1);
> +}
page_xchg_last_nid() and reset_page_last_nid() are getting nuttily
large. Please investigate uninlining them?
reset_page_last_nid() is poorly named. page_reset_last_nid() would be
better, and consistent.
--
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>
WARNING: multiple messages have this Message-ID (diff)
From: Andrew Morton <akpm@linux-foundation.org>
To: Mel Gorman <mgorman@suse.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>,
Andrea Arcangeli <aarcange@redhat.com>,
Ingo Molnar <mingo@kernel.org>,
Simon Jeons <simon.jeons@gmail.com>,
Wanpeng Li <liwanp@linux.vnet.ibm.com>,
Hugh Dickins <hughd@google.com>, Linux-MM <linux-mm@kvack.org>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 5/6] mm: Fold page->_last_nid into page->flags where possible
Date: Tue, 22 Jan 2013 14:46:59 -0800 [thread overview]
Message-ID: <20130122144659.d512e05c.akpm@linux-foundation.org> (raw)
In-Reply-To: <1358874762-19717-6-git-send-email-mgorman@suse.de>
On Tue, 22 Jan 2013 17:12:41 +0000
Mel Gorman <mgorman@suse.de> wrote:
> From: Peter Zijlstra <a.p.zijlstra@chello.nl>
>
> page->_last_nid fits into page->flags on 64-bit. The unlikely 32-bit NUMA
> configuration with NUMA Balancing will still need an extra page field.
> As Peter notes "Completely dropping 32bit support for CONFIG_NUMA_BALANCING
> would simplify things, but it would also remove the warning if we grow
> enough 64bit only page-flags to push the last-cpu out."
How much space remains in the 64-bit page->flags?
Was this the best possible use of the remaining space?
It's good that we can undo this later by flipping
LAST_NID_NOT_IN_PAGE_FLAGS.
> [mgorman@suse.de: Minor modifications]
> Signed-off-by: Mel Gorman <mgorman@suse.de>
Several of these patches are missing signoffs (Peter and Hugh).
>
> ...
>
> +static inline int page_last_nid(struct page *page)
> +{
> + return (page->flags >> LAST_NID_PGSHIFT) & LAST_NID_MASK;
> +}
> +
> +static inline int page_xchg_last_nid(struct page *page, int nid)
> +{
> + unsigned long old_flags, flags;
> + int last_nid;
> +
> + do {
> + old_flags = flags = page->flags;
> + last_nid = page_last_nid(page);
> +
> + flags &= ~(LAST_NID_MASK << LAST_NID_PGSHIFT);
> + flags |= (nid & LAST_NID_MASK) << LAST_NID_PGSHIFT;
> + } while (unlikely(cmpxchg(&page->flags, old_flags, flags) != old_flags));
> +
> + return last_nid;
> +}
> +
> +static inline void reset_page_last_nid(struct page *page)
> +{
> + page_xchg_last_nid(page, (1 << LAST_NID_SHIFT) - 1);
> +}
page_xchg_last_nid() and reset_page_last_nid() are getting nuttily
large. Please investigate uninlining them?
reset_page_last_nid() is poorly named. page_reset_last_nid() would be
better, and consistent.
next prev parent reply other threads:[~2013-01-22 22:47 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-01-22 17:12 [PATCH 0/6] Follow up work on NUMA Balancing Mel Gorman
2013-01-22 17:12 ` Mel Gorman
2013-01-22 17:12 ` [PATCH 1/6] mm: numa: Fix minor typo in numa_next_scan Mel Gorman
2013-01-22 17:12 ` Mel Gorman
2013-01-22 17:12 ` [PATCH 2/6] mm: numa: Take THP into account when migrating pages for NUMA balancing Mel Gorman
2013-01-22 17:12 ` Mel Gorman
2013-01-22 17:12 ` [PATCH 3/6] mm: numa: Handle side-effects in count_vm_numa_events() for !CONFIG_NUMA_BALANCING Mel Gorman
2013-01-22 17:12 ` Mel Gorman
2013-01-22 22:40 ` Andrew Morton
2013-01-22 22:40 ` Andrew Morton
2013-01-23 9:27 ` Mel Gorman
2013-01-23 9:27 ` Mel Gorman
2013-01-22 17:12 ` [PATCH 4/6] mm: Move page flags layout to separate header Mel Gorman
2013-01-22 17:12 ` Mel Gorman
2013-01-22 17:12 ` [PATCH 5/6] mm: Fold page->_last_nid into page->flags where possible Mel Gorman
2013-01-22 17:12 ` Mel Gorman
2013-01-22 22:46 ` Andrew Morton [this message]
2013-01-22 22:46 ` Andrew Morton
2013-01-23 13:17 ` Mel Gorman
2013-01-23 13:17 ` Mel Gorman
2013-01-23 21:45 ` KOSAKI Motohiro
2013-01-23 21:45 ` KOSAKI Motohiro
2013-01-23 13:18 ` [PATCH] mm: init: Report on last-nid information stored in page->flags Mel Gorman
2013-01-23 13:18 ` Mel Gorman
2013-01-23 14:25 ` [PATCH 5/6] mm: Fold page->_last_nid into page->flags where possible Mel Gorman
2013-01-23 14:25 ` Mel Gorman
2013-01-23 21:56 ` Andrew Morton
2013-01-23 21:56 ` Andrew Morton
2013-01-24 10:55 ` [PATCH] mm: Rename page struct field helpers Mel Gorman
2013-01-24 10:55 ` Mel Gorman
2013-01-29 4:39 ` Hugh Dickins
2013-01-29 4:39 ` Hugh Dickins
2013-01-30 11:58 ` Mel Gorman
2013-01-30 11:58 ` Mel Gorman
2013-01-30 20:32 ` Hugh Dickins
2013-01-30 20:32 ` Hugh Dickins
2013-01-23 15:23 ` [PATCH] mm: uninline page_xchg_last_nid() Mel Gorman
2013-01-23 15:23 ` Mel Gorman
2013-01-22 17:12 ` [PATCH 6/6] mm: numa: Cleanup flow of transhuge page migration Mel Gorman
2013-01-22 17:12 ` Mel Gorman
2013-01-27 21:20 ` Hugh Dickins
2013-01-27 21:20 ` Hugh Dickins
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=20130122144659.d512e05c.akpm@linux-foundation.org \
--to=akpm@linux-foundation.org \
--cc=a.p.zijlstra@chello.nl \
--cc=aarcange@redhat.com \
--cc=hughd@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=liwanp@linux.vnet.ibm.com \
--cc=mgorman@suse.de \
--cc=mingo@kernel.org \
--cc=simon.jeons@gmail.com \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.