Date:   Wed, 15 Feb 2023 21:17:49 -0800
In-Reply-To: <20230216051750.3125598-1-surenb@google.com>
Mime-Version: 1.0
References: <20230216051750.3125598-1-surenb@google.com>
Message-ID: <20230216051750.3125598-35-surenb@google.com>
Subject: [PATCH v3 34/35] mm/mmap: free vm_area_struct without call_rcu in exit_mmap
From:   Suren Baghdasaryan <surenb@google.com>
To:     akpm@linux-foundation.org
Cc:     michel@lespinasse.org, jglisse@google.com, mhocko@suse.com,
        vbabka@suse.cz, hannes@cmpxchg.org, mgorman@techsingularity.net,
        dave@stgolabs.net, willy@infradead.org, liam.howlett@oracle.com,
        peterz@infradead.org, ldufour@linux.ibm.com, paulmck@kernel.org,
        mingo@redhat.com, will@kernel.org, luto@kernel.org,
        songliubraving@fb.com, peterx@redhat.com, david@redhat.com,
        dhowells@redhat.com, hughd@google.com, bigeasy@linutronix.de,
        kent.overstreet@linux.dev, punit.agrawal@bytedance.com,
        lstoakes@gmail.com, peterjung1337@gmail.com, rientjes@google.com,
        chriscli@google.com, axelrasmussen@google.com, joelaf@google.com,
        minchan@google.com, rppt@kernel.org, jannh@google.com,
        shakeelb@google.com, tatashin@google.com, edumazet@google.com,
        gthelen@google.com, gurua@google.com, arjunroy@google.com,
        soheil@google.com, leewalsh@google.com, posk@google.com,
        michalechner92@googlemail.com, linux-mm@kvack.org,
        linux-arm-kernel@lists.infradead.org,
        linuxppc-dev@lists.ozlabs.org, x86@kernel.org,
        linux-kernel@vger.kernel.org, kernel-team@android.com,
        Suren Baghdasaryan <surenb@google.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

call_rcu() can take a long time when callback offloading is enabled.
Its use in the vm_area_free can cause regressions in the exit path when
multiple VMAs are being freed.
Because exit_mmap() is called only after the last mm user drops its
refcount, the page fault handlers can't be racing with it. Any other
possible user like oom-reaper or process_mrelease are already synchronized
using mmap_lock. Therefore exit_mmap() can free VMAs directly, without
the use of call_rcu().
Expose __vm_area_free() and use it from exit_mmap() to avoid possible
call_rcu() floods and performance regressions caused by it.

Signed-off-by: Suren Baghdasaryan <surenb@google.com>
---
 include/linux/mm.h |  2 ++
 kernel/fork.c      |  2 +-
 mm/mmap.c          | 11 +++++++----
 3 files changed, 10 insertions(+), 5 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 3c9167529417..cedef02dfd2b 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -256,6 +256,8 @@ void setup_initial_init_mm(void *start_code, void *end_code,
 struct vm_area_struct *vm_area_alloc(struct mm_struct *);
 struct vm_area_struct *vm_area_dup(struct vm_area_struct *);
 void vm_area_free(struct vm_area_struct *);
+/* Use only if VMA has no other users */
+void __vm_area_free(struct vm_area_struct *vma);
 
 #ifndef CONFIG_MMU
 extern struct rb_root nommu_region_tree;
diff --git a/kernel/fork.c b/kernel/fork.c
index a08cc0e2bfde..d0999de82f94 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -480,7 +480,7 @@ struct vm_area_struct *vm_area_dup(struct vm_area_struct *orig)
 	return new;
 }
 
-static void __vm_area_free(struct vm_area_struct *vma)
+void __vm_area_free(struct vm_area_struct *vma)
 {
 	free_anon_vma_name(vma);
 	kmem_cache_free(vm_area_cachep, vma);
diff --git a/mm/mmap.c b/mm/mmap.c
index adf40177e68f..d847be615720 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -133,7 +133,7 @@ void unlink_file_vma(struct vm_area_struct *vma)
 /*
  * Close a vm structure and free it.
  */
-static void remove_vma(struct vm_area_struct *vma)
+static void remove_vma(struct vm_area_struct *vma, bool unreachable)
 {
 	might_sleep();
 	if (vma->vm_ops && vma->vm_ops->close)
@@ -141,7 +141,10 @@ static void remove_vma(struct vm_area_struct *vma)
 	if (vma->vm_file)
 		fput(vma->vm_file);
 	mpol_put(vma_policy(vma));
-	vm_area_free(vma);
+	if (unreachable)
+		__vm_area_free(vma);
+	else
+		vm_area_free(vma);
 }
 
 static inline struct vm_area_struct *vma_prev_limit(struct vma_iterator *vmi,
@@ -2135,7 +2138,7 @@ static inline void remove_mt(struct mm_struct *mm, struct ma_state *mas)
 		if (vma->vm_flags & VM_ACCOUNT)
 			nr_accounted += nrpages;
 		vm_stat_account(mm, vma->vm_flags, -nrpages);
-		remove_vma(vma);
+		remove_vma(vma, false);
 	}
 	vm_unacct_memory(nr_accounted);
 	validate_mm(mm);
@@ -3085,7 +3088,7 @@ void exit_mmap(struct mm_struct *mm)
 	do {
 		if (vma->vm_flags & VM_ACCOUNT)
 			nr_accounted += vma_pages(vma);
-		remove_vma(vma);
+		remove_vma(vma, true);
 		count++;
 		cond_resched();
 	} while ((vma = mas_find(&mas, ULONG_MAX)) != NULL);
-- 
2.39.1