From: John Hubbard <jhubbard@nvidia.com>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: "Al Viro" <viro@zeniv.linux.org.uk>,
"Christoph Hellwig" <hch@infradead.org>,
"Dan Williams" <dan.j.williams@intel.com>,
"Dave Chinner" <david@fromorbit.com>,
"Ira Weiny" <ira.weiny@intel.com>, "Jan Kara" <jack@suse.cz>,
"Jason Gunthorpe" <jgg@ziepe.ca>,
"Jonathan Corbet" <corbet@lwn.net>,
"Jérôme Glisse" <jglisse@redhat.com>,
"Kirill A . Shutemov" <kirill@shutemov.name>,
"Michal Hocko" <mhocko@suse.com>,
"Mike Kravetz" <mike.kravetz@oracle.com>,
"Shuah Khan" <shuah@kernel.org>,
"Vlastimil Babka" <vbabka@suse.cz>,
linux-doc@vger.kernel.org, linux-fsdevel@vger.kernel.org,
linux-kselftest@vger.kernel.org, linux-rdma@vger.kernel.org,
linux-mm@kvack.org, LKML <linux-kernel@vger.kernel.org>,
"John Hubbard" <jhubbard@nvidia.com>
Subject: [PATCH v2 6/8] mm/gup: /proc/vmstat: pin_user_pages (FOLL_PIN) reporting
Date: Tue, 28 Jan 2020 19:24:15 -0800 [thread overview]
Message-ID: <20200129032417.3085670-7-jhubbard@nvidia.com> (raw)
In-Reply-To: <20200129032417.3085670-1-jhubbard@nvidia.com>
Now that pages are "DMA-pinned" via pin_user_page*(), and unpinned via
unpin_user_pages*(), we need some visibility into whether all of this is
working correctly.
Add two new fields to /proc/vmstat:
nr_foll_pin_requested
nr_foll_pin_returned
These are documented in Documentation/core-api/pin_user_pages.rst.
They represent the number of pages (since boot time) that have been
pinned ("nr_foll_pin_requested") and unpinned ("nr_foll_pin_returned"),
via pin_user_pages*() and unpin_user_pages*().
In the absence of long-running DMA or RDMA operations that hold pages
pinned, the above two fields will normally be equal to each other.
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
---
include/linux/mmzone.h | 2 ++
mm/gup.c | 21 +++++++++++++++++++++
mm/vmstat.c | 2 ++
3 files changed, 25 insertions(+)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 462f6873905a..392868bc4763 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -243,6 +243,8 @@ enum node_stat_item {
NR_DIRTIED, /* page dirtyings since bootup */
NR_WRITTEN, /* page writings since bootup */
NR_KERNEL_MISC_RECLAIMABLE, /* reclaimable non-slab kernel pages */
+ NR_FOLL_PIN_REQUESTED, /* via: pin_user_page(), gup flag: FOLL_PIN */
+ NR_FOLL_PIN_RETURNED, /* pages returned via unpin_user_page() */
NR_VM_NODE_STAT_ITEMS
};
diff --git a/mm/gup.c b/mm/gup.c
index 03e7a5cfa6a9..d536bda383c4 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -29,6 +29,19 @@ struct follow_page_context {
unsigned int page_mask;
};
+#ifdef CONFIG_DEBUG_VM
+static inline void __update_proc_vmstat(struct page *page,
+ enum node_stat_item item, int count)
+{
+ mod_node_page_state(page_pgdat(page), item, count);
+}
+#else
+static inline void __update_proc_vmstat(struct page *page,
+ enum node_stat_item item, int count)
+{
+}
+#endif
+
static void hpage_pincount_add(struct page *page, int refs)
{
VM_BUG_ON_PAGE(!hpage_pincount_available(page), page);
@@ -86,6 +99,8 @@ static __maybe_unused struct page *try_grab_compound_head(struct page *page,
if (flags & FOLL_GET)
return try_get_compound_head(page, refs);
else if (flags & FOLL_PIN) {
+ int orig_refs = refs;
+
/*
* When pinning a compound page of order > 1 (which is what
* hpage_pincount_available() checks for), use an exact count to
@@ -104,6 +119,7 @@ static __maybe_unused struct page *try_grab_compound_head(struct page *page,
if (hpage_pincount_available(page))
hpage_pincount_add(page, refs);
+ __update_proc_vmstat(page, NR_FOLL_PIN_REQUESTED, orig_refs);
return page;
}
@@ -159,6 +175,8 @@ bool __must_check try_grab_page(struct page *page, unsigned int flags)
* once, so that the page really is pinned.
*/
page_ref_add(page, refs);
+
+ __update_proc_vmstat(page, NR_FOLL_PIN_REQUESTED, 1);
}
return true;
@@ -179,6 +197,7 @@ static bool __unpin_devmap_managed_user_page(struct page *page)
count = page_ref_sub_return(page, refs);
+ __update_proc_vmstat(page, NR_FOLL_PIN_RETURNED, 1);
/*
* devmap page refcounts are 1-based, rather than 0-based: if
* refcount is 1, then the page is free and the refcount is
@@ -229,6 +248,8 @@ void unpin_user_page(struct page *page)
if (page_ref_sub_and_test(page, refs))
__put_page(page);
+
+ __update_proc_vmstat(page, NR_FOLL_PIN_RETURNED, 1);
}
EXPORT_SYMBOL(unpin_user_page);
diff --git a/mm/vmstat.c b/mm/vmstat.c
index 78d53378db99..b56808bae1b4 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -1168,6 +1168,8 @@ const char * const vmstat_text[] = {
"nr_dirtied",
"nr_written",
"nr_kernel_misc_reclaimable",
+ "nr_foll_pin_requested",
+ "nr_foll_pin_returned",
/* enum writeback_stat_item counters */
"nr_dirty_threshold",
--
2.25.0
next prev parent reply other threads:[~2020-01-29 3:24 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-01-29 3:24 [PATCH v2 0/8] mm/gup: track FOLL_PIN pages (follow on from v12) John Hubbard
2020-01-29 3:24 ` [PATCH v2 1/8] mm: dump_page: print head page's refcount, for compound pages John Hubbard
2020-01-29 11:25 ` Kirill A. Shutemov
2020-01-29 22:26 ` John Hubbard
2020-01-29 22:59 ` Matthew Wilcox
2020-01-30 6:23 ` John Hubbard
2020-01-30 6:30 ` John Hubbard
2020-01-29 3:24 ` [PATCH v2 2/8] mm/gup: split get_user_pages_remote() into two routines John Hubbard
2020-01-29 3:24 ` [PATCH v2 3/8] mm/gup: pass a flags arg to __gup_device_* functions John Hubbard
2020-01-29 3:24 ` [PATCH v2 4/8] mm/gup: track FOLL_PIN pages John Hubbard
2020-01-29 13:51 ` Kirill A. Shutemov
2020-01-30 6:44 ` John Hubbard
2020-01-30 11:31 ` Kirill A. Shutemov
2020-01-31 3:19 ` John Hubbard
2020-01-29 3:24 ` [PATCH v2 5/8] mm/gup: page->hpage_pinned_refcount: exact pin counts for huge pages John Hubbard
2020-01-29 3:24 ` John Hubbard [this message]
2020-01-29 3:24 ` [PATCH v2 7/8] mm/gup_benchmark: support pin_user_pages() and related calls John Hubbard
2020-01-29 3:24 ` [PATCH v2 8/8] selftests/vm: run_vmtests: invoke gup_benchmark with basic FOLL_PIN coverage John Hubbard
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=20200129032417.3085670-7-jhubbard@nvidia.com \
--to=jhubbard@nvidia.com \
--cc=akpm@linux-foundation.org \
--cc=corbet@lwn.net \
--cc=dan.j.williams@intel.com \
--cc=david@fromorbit.com \
--cc=hch@infradead.org \
--cc=ira.weiny@intel.com \
--cc=jack@suse.cz \
--cc=jgg@ziepe.ca \
--cc=jglisse@redhat.com \
--cc=kirill@shutemov.name \
--cc=linux-doc@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-rdma@vger.kernel.org \
--cc=mhocko@suse.com \
--cc=mike.kravetz@oracle.com \
--cc=shuah@kernel.org \
--cc=vbabka@suse.cz \
--cc=viro@zeniv.linux.org.uk \
/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).