From: Hugh Dickins <hughd@google.com>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Johannes Weiner <hannes@cmpxchg.org>,
domenico cerasuolo <mimmocerasuolo@gmail.com>,
Andi Kleen <ak@linux.intel.com>, Christoph Lameter <cl@linux.co>,
Matthew Wilcox <willy@infradead.org>,
Mike Kravetz <mike.kravetz@oracle.com>,
David Hildenbrand <david@redhat.com>,
Suren Baghdasaryan <surenb@google.com>,
Yang Shi <shy828301@gmail.com>,
Sidhartha Kumar <sidhartha.kumar@oracle.com>,
Vishal Moola <vishal.moola@gmamil.com>,
Kefeng Wang <wangkefeng.wang@huawei.com>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
Tejun Heo <tj@kernel.org>,
Mel Gorman <mgorman@techsingularity.net>,
Michal Hocko <mhocko@suse.com>,
"Huang, Ying" <ying.huang@intel.com>,
Nhat Pham <nphamcs@gmail.com>,
Yosry Ahmed <yosryahmed@google.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: [PATCH] mempolicy: alloc_pages_mpol() for NUMA policy without vma: fix
Date: Mon, 23 Oct 2023 23:44:18 -0700 (PDT) [thread overview]
Message-ID: <00dc4f56-e623-7c85-29ea-4211e93063f6@google.com> (raw)
In-Reply-To: <20231023190555.GA1964810@cmpxchg.org>
mm-unstable commit 48a7bd12d57f ("mempolicy: alloc_pages_mpol() for NUMA
policy without vma") ended read_swap_cache_async() supporting NULL vma -
okay; but missed the NULL mpol being passed to __read_swap_cache_async()
by zswap_writeback_entry() - oops!
Since its other callers all give good mpol, add get_task_policy(current)
there in mm/zswap.c, to produce the same good-enough behaviour as before
(and task policy, acted on in current task, does not require the refcount
to be dup'ed).
But if that policy is (quite reasonably) MPOL_INTERLEAVE, then ilx must
be NO_INTERLEAVE_INDEX rather than 0, to provide the same distribution
as before: move that definition from mempolicy.c to mempolicy.h.
Reported-by: Domenico Cerasuolo <mimmocerasuolo@gmail.com>
Closes: https://lore.kernel.org/linux-mm/74e34633-6060-f5e3-aee-7040d43f2e93@google.com/T/#mf08c877d1884fc7867f9e328cdf02257ff3b3ae9
Suggested-by: Johannes Weiner <hannes@cmpxchg.org>
Fixes: 48a7bd12d57f ("mempolicy: alloc_pages_mpol() for NUMA policy without vma")
Signed-off-by: Hugh Dickins <hughd@google.com>
---
include/linux/mempolicy.h | 2 ++
mm/mempolicy.c | 2 --
mm/zswap.c | 7 +++++--
3 files changed, 7 insertions(+), 4 deletions(-)
diff --git a/include/linux/mempolicy.h b/include/linux/mempolicy.h
index 2801d5b0a4e9..dd9ed2ce7fd5 100644
--- a/include/linux/mempolicy.h
+++ b/include/linux/mempolicy.h
@@ -17,6 +17,8 @@
struct mm_struct;
+#define NO_INTERLEAVE_INDEX (-1UL) /* use task il_prev for interleaving */
+
#ifdef CONFIG_NUMA
/*
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 898ee2e3c85b..989293180eb6 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -114,8 +114,6 @@
#define MPOL_MF_INVERT (MPOL_MF_INTERNAL << 1) /* Invert check for nodemask */
#define MPOL_MF_WRLOCK (MPOL_MF_INTERNAL << 2) /* Write-lock walked vmas */
-#define NO_INTERLEAVE_INDEX (-1UL)
-
static struct kmem_cache *policy_cache;
static struct kmem_cache *sn_cache;
diff --git a/mm/zswap.c b/mm/zswap.c
index 37d2b1cb2ecb..060857adca76 100644
--- a/mm/zswap.c
+++ b/mm/zswap.c
@@ -24,6 +24,7 @@
#include <linux/swap.h>
#include <linux/crypto.h>
#include <linux/scatterlist.h>
+#include <linux/mempolicy.h>
#include <linux/mempool.h>
#include <linux/zpool.h>
#include <crypto/acompress.h>
@@ -1057,6 +1058,7 @@ static int zswap_writeback_entry(struct zswap_entry *entry,
{
swp_entry_t swpentry = entry->swpentry;
struct page *page;
+ struct mempolicy *mpol;
struct scatterlist input, output;
struct crypto_acomp_ctx *acomp_ctx;
struct zpool *pool = zswap_find_zpool(entry);
@@ -1075,8 +1077,9 @@ static int zswap_writeback_entry(struct zswap_entry *entry,
}
/* try to allocate swap cache page */
- page = __read_swap_cache_async(swpentry, GFP_KERNEL, NULL, 0,
- &page_was_allocated);
+ mpol = get_task_policy(current);
+ page = __read_swap_cache_async(swpentry, GFP_KERNEL, mpol,
+ NO_INTERLEAVE_INDEX, &page_was_allocated);
if (!page) {
ret = -ENOMEM;
goto fail;
--
2.35.3
next prev parent reply other threads:[~2023-10-24 6:44 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-03 9:12 [PATCH v2 00/12] mempolicy: cleanups leading to NUMA mpol without vma Hugh Dickins
2023-10-03 9:15 ` [PATCH v2 01/12] hugetlbfs: drop shared NUMA mempolicy pretence Hugh Dickins
2023-10-03 9:16 ` [PATCH v2 02/12] kernfs: drop shared NUMA mempolicy hooks Hugh Dickins
2023-10-03 9:17 ` [PATCH v2 03/12] mempolicy: fix migrate_pages(2) syscall return nr_failed Hugh Dickins
2023-10-07 7:27 ` Huang, Ying
2023-10-03 9:19 ` [PATCH v2 04/12] mempolicy trivia: delete those ancient pr_debug()s Hugh Dickins
2023-10-03 9:20 ` [PATCH v2 05/12] mempolicy trivia: slightly more consistent naming Hugh Dickins
2023-10-03 9:21 ` [PATCH v2 06/12] mempolicy trivia: use pgoff_t in shared mempolicy tree Hugh Dickins
2023-10-03 9:22 ` [PATCH v2 07/12] mempolicy: mpol_shared_policy_init() without pseudo-vma Hugh Dickins
2023-10-03 9:24 ` [PATCH v2 08/12] mempolicy: remove confusing MPOL_MF_LAZY dead code Hugh Dickins
2023-10-03 22:28 ` Yang Shi
2023-10-03 9:25 ` [PATCH v2 09/12] mm: add page_rmappable_folio() wrapper Hugh Dickins
2023-10-03 9:26 ` [PATCH v2 10/12] mempolicy: alloc_pages_mpol() for NUMA policy without vma Hugh Dickins
2023-10-19 20:39 ` [PATCH v3 " Hugh Dickins
2023-10-23 16:53 ` domenico cerasuolo
2023-10-23 17:53 ` Andrew Morton
2023-10-23 18:10 ` domenico cerasuolo
2023-10-23 19:05 ` Johannes Weiner
2023-10-23 19:48 ` Hugh Dickins
2023-10-24 6:44 ` Hugh Dickins [this message]
2023-10-24 8:17 ` [PATCH] mempolicy: alloc_pages_mpol() for NUMA policy without vma: fix kernel test robot
2023-10-24 15:56 ` Hugh Dickins
2023-10-24 16:09 ` [PATCH v2] " Hugh Dickins
2023-10-23 18:34 ` [PATCH v3 10/12] mempolicy: alloc_pages_mpol() for NUMA policy without vma Zi Yan
2023-10-23 21:10 ` Hugh Dickins
2023-10-23 21:13 ` Zi Yan
2023-10-03 9:27 ` [PATCH v2 11/12] mempolicy: mmap_lock is not needed while migrating folios Hugh Dickins
2023-10-03 9:29 ` [PATCH v2 12/12] mempolicy: migration attempt to match interleave nodes Hugh Dickins
2023-10-24 6:50 ` [PATCH] mempolicy: migration attempt to match interleave nodes: fix Hugh Dickins
2023-10-24 15:18 ` Liam R. Howlett
2023-10-24 16:32 ` Hugh Dickins
2023-10-24 16:45 ` Matthew Wilcox
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=00dc4f56-e623-7c85-29ea-4211e93063f6@google.com \
--to=hughd@google.com \
--cc=ak@linux.intel.com \
--cc=akpm@linux-foundation.org \
--cc=cl@linux.co \
--cc=david@redhat.com \
--cc=gregkh@linuxfoundation.org \
--cc=hannes@cmpxchg.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@techsingularity.net \
--cc=mhocko@suse.com \
--cc=mike.kravetz@oracle.com \
--cc=mimmocerasuolo@gmail.com \
--cc=nphamcs@gmail.com \
--cc=shy828301@gmail.com \
--cc=sidhartha.kumar@oracle.com \
--cc=surenb@google.com \
--cc=tj@kernel.org \
--cc=vishal.moola@gmamil.com \
--cc=wangkefeng.wang@huawei.com \
--cc=willy@infradead.org \
--cc=ying.huang@intel.com \
--cc=yosryahmed@google.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 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).