From nobody Sun Oct 5 10:47:09 2025 Received: from mail-pg1-f202.google.com (mail-pg1-f202.google.com [209.85.215.202]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id EE23A29A9D3 for ; Wed, 6 Aug 2025 15:59:14 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.215.202 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1754495957; cv=none; b=UBN0Hz+BHu/82UJzU0eaSZQw40GdSW8yBYRSZ6BXBdzgq/0OESZ9gDFkcgQMHU4eqDe0iGBFS2cqNv5OHDJwO9QDpU7/SZ9bRdfSEEKNObRbnmTDLq4LqrXlF8koWEDZbnDjKaoweUd6QoEsyYq23XiISBhNoSy71FzCsLWsyuw= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1754495957; c=relaxed/simple; bh=JRfgX1mdsRBhfL5iUYnonjqsaZrBRF+pfu+2IoIqfMs=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Cc:Content-Type; b=Q9f492Pb+x8JZldO+behoOKXCLGsPiSmn8JSOUzHTmpBPoWpaZUama2FTLYS8nCBuJT7DK39HvrxI3yZuzjiRfHPI5occVUk0Z3kdTn7bcI4zjuZiWNf8AO2XEPLx4VVqwvpsVVZT/CGhpRx3jwWl+GDaOL+p0HlOFm29kWCc6s= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com; spf=pass smtp.mailfrom=flex--surenb.bounces.google.com; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b=Budp1pRU; arc=none smtp.client-ip=209.85.215.202 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=flex--surenb.bounces.google.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="Budp1pRU" Received: by mail-pg1-f202.google.com with SMTP id 41be03b00d2f7-b4227538a47so4828777a12.1 for ; Wed, 06 Aug 2025 08:59:14 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1754495954; x=1755100754; darn=vger.kernel.org; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=79cST+DpT8JLjhSc853c7Vp44fp4YLe4km92lxkzOhQ=; b=Budp1pRUx6mYa0epyo7zbp+kkDKr3RN/i8el8R7Gka6OmrHe2dFDFDdc+iDTLUKKiC iq40KT4Z5+N9+8HmW49DP/HxNSsTFIzoP30ciMT9uw8AeFodCJqI6ZEdtg9hnks+JGHl xZoBMksg8rh0B18GhOcwGnnPUsoKD54LGiyKNA3erREfPj6wpwDRuMrPdDoFPXgCrh0q Lo0mJa3l3ezZkJrQ9XJPpiSLn0TwfCXe/2ffk5VJZXxr8LYJF29c9J5g6v9woFjbRH+8 Ryz0j8obquz+9LqjwKt+gymvn94TrB4q8xSYgGRCXLW7KO3UMTw2Aas+6mWmJKGs5X1s 9U7A== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1754495954; x=1755100754; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=79cST+DpT8JLjhSc853c7Vp44fp4YLe4km92lxkzOhQ=; b=wUwQmJxv1wAEvVtVZ+HIg/1GoZV1eI2mfkTiV2YhEHds0whrZVOr7uA7EuX/yP2AcD 8ZNpTqWA1g41yqqegbra+z3rlq6h+/C90lF+lkB3g9P/hrHH0oVaUXGbG+joqxo1yE7K FDPCUqCcZqlw2uTWRm+9ecp5s2DodHiBHt9o1V0xWYIYjXaLT4gVN0VMYJ2S8XuD0V6d T2EOhWn7u4uPcQ5WindBZ8DG5abj/ROm8WL5ukRxIdF/Z/69/SSTwtI+Nqmyb1xAm5BC 455WDpEOgQ33vN1eAfKDROB9Za61GLbQvUk/XRMBpr17I+3eFoSMw40gsMnwscBZDTYi gffQ== X-Forwarded-Encrypted: i=1; AJvYcCVTCyQF3aujYCICsJZNdbpgZOcRzDgaZRlKm64a39JoIlSIhtEv2sURn5tf320h/X2AU/GpuQe/FV0WGrk=@vger.kernel.org X-Gm-Message-State: AOJu0YzzjocDXsbpwFjAOR04AJ27FXPhUa+3FFGfO8HPqVOfg2Veobtr sHNQFp5KimC9KlOtirFJCOJJyMro4pnO/LNnCYFladO/c+zaYirFXFjDmjakW4LI7FnAcJ4YXHR w2rTqUQ== X-Google-Smtp-Source: AGHT+IFlS+JvIsWnRvs7bTfyfgC/5w+RmEgonVOp9GJDKi3AshY2Fhxqa2lT8Ee9l8rZG2GebQuiHk0ZMyQ= X-Received: from pjboe1.prod.google.com ([2002:a17:90b:3941:b0:31e:c61e:663b]) (user=surenb job=prod-delivery.src-stubby-dispatcher) by 2002:a17:902:dacc:b0:240:92cc:8fcf with SMTP id d9443c01a7336-2429f496c6bmr47463745ad.49.1754495954146; Wed, 06 Aug 2025 08:59:14 -0700 (PDT) Date: Wed, 6 Aug 2025 08:59:04 -0700 In-Reply-To: <20250806155905.824388-1-surenb@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20250806155905.824388-1-surenb@google.com> X-Mailer: git-send-email 2.50.1.565.gc32cd1483b-goog Message-ID: <20250806155905.824388-4-surenb@google.com> Subject: [PATCH v3 3/3] fs/proc/task_mmu: execute PROCMAP_QUERY ioctl under per-vma locks From: Suren Baghdasaryan To: akpm@linux-foundation.org Cc: Liam.Howlett@oracle.com, lorenzo.stoakes@oracle.com, david@redhat.com, vbabka@suse.cz, peterx@redhat.com, jannh@google.com, hannes@cmpxchg.org, mhocko@kernel.org, paulmck@kernel.org, shuah@kernel.org, adobriyan@gmail.com, brauner@kernel.org, josef@toxicpanda.com, yebin10@huawei.com, linux@weissschuh.net, willy@infradead.org, osalvador@suse.de, andrii@kernel.org, ryan.roberts@arm.com, christophe.leroy@csgroup.eu, tjmercier@google.com, kaleshsingh@google.com, aha310510@gmail.com, linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, linux-mm@kvack.org, linux-kselftest@vger.kernel.org, surenb@google.com Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Utilize per-vma locks to stabilize vma after lookup without taking mmap_lock during PROCMAP_QUERY ioctl execution. If vma lock is contended, we fall back to mmap_lock but take it only momentarily to lock the vma and release the mmap_lock. In a very unlikely case of vm_refcnt overflow, this fall back path will fail and ioctl is done under mmap_lock protection. This change is designed to reduce mmap_lock contention and prevent PROCMAP_QUERY ioctl calls from blocking address space updates. Signed-off-by: Suren Baghdasaryan Acked-by: SeongJae Park --- fs/proc/task_mmu.c | 84 +++++++++++++++++++++++++++++++++++++--------- 1 file changed, 68 insertions(+), 16 deletions(-) diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 45134335e086..0396315dfaee 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -517,28 +517,81 @@ static int pid_maps_open(struct inode *inode, struct = file *file) PROCMAP_QUERY_VMA_FLAGS \ ) =20 -static int query_vma_setup(struct mm_struct *mm) +#ifdef CONFIG_PER_VMA_LOCK + +static int query_vma_setup(struct proc_maps_locking_ctx *lock_ctx) { - return mmap_read_lock_killable(mm); + lock_ctx->locked_vma =3D NULL; + lock_ctx->mmap_locked =3D false; + + return 0; } =20 -static void query_vma_teardown(struct mm_struct *mm, struct vm_area_struct= *vma) +static void query_vma_teardown(struct proc_maps_locking_ctx *lock_ctx) { - mmap_read_unlock(mm); + if (lock_ctx->mmap_locked) + mmap_read_unlock(lock_ctx->mm); + else + unlock_vma(lock_ctx); +} + +static struct vm_area_struct *query_vma_find_by_addr(struct proc_maps_lock= ing_ctx *lock_ctx, + unsigned long addr) +{ + struct vm_area_struct *vma; + struct vma_iterator vmi; + + if (lock_ctx->mmap_locked) + return find_vma(lock_ctx->mm, addr); + + unlock_vma(lock_ctx); + rcu_read_lock(); + vma_iter_init(&vmi, lock_ctx->mm, addr); + vma =3D lock_next_vma(lock_ctx->mm, &vmi, addr); + rcu_read_unlock(); + + if (!IS_ERR_OR_NULL(vma)) { + lock_ctx->locked_vma =3D vma; + } else if (PTR_ERR(vma) =3D=3D -EAGAIN) { + /* Fallback to mmap_lock on vma->vm_refcnt overflow */ + mmap_read_lock(lock_ctx->mm); + vma =3D find_vma(lock_ctx->mm, addr); + lock_ctx->mmap_locked =3D true; + } + + return vma; +} + +#else /* CONFIG_PER_VMA_LOCK */ + +static int query_vma_setup(struct proc_maps_locking_ctx *lock_ctx) +{ + return mmap_read_lock_killable(lock_ctx->mm); } =20 -static struct vm_area_struct *query_vma_find_by_addr(struct mm_struct *mm,= unsigned long addr) +static void query_vma_teardown(struct proc_maps_locking_ctx *lock_ctx) { - return find_vma(mm, addr); + mmap_read_unlock(lock_ctx->mm); } =20 -static struct vm_area_struct *query_matching_vma(struct mm_struct *mm, +static struct vm_area_struct *query_vma_find_by_addr(struct proc_maps_lock= ing_ctx *lock_ctx, + unsigned long addr) +{ + return find_vma(lock_ctx->mm, addr); +} + +#endif /* CONFIG_PER_VMA_LOCK */ + +static struct vm_area_struct *query_matching_vma(struct proc_maps_locking_= ctx *lock_ctx, unsigned long addr, u32 flags) { struct vm_area_struct *vma; =20 next_vma: - vma =3D query_vma_find_by_addr(mm, addr); + vma =3D query_vma_find_by_addr(lock_ctx, addr); + if (IS_ERR(vma)) + return vma; + if (!vma) goto no_vma; =20 @@ -579,11 +632,11 @@ static struct vm_area_struct *query_matching_vma(stru= ct mm_struct *mm, return ERR_PTR(-ENOENT); } =20 -static int do_procmap_query(struct proc_maps_private *priv, void __user *u= arg) +static int do_procmap_query(struct mm_struct *mm, void __user *uarg) { + struct proc_maps_locking_ctx lock_ctx =3D { .mm =3D mm }; struct procmap_query karg; struct vm_area_struct *vma; - struct mm_struct *mm; const char *name =3D NULL; char build_id_buf[BUILD_ID_SIZE_MAX], *name_buf =3D NULL; __u64 usize; @@ -610,17 +663,16 @@ static int do_procmap_query(struct proc_maps_private = *priv, void __user *uarg) if (!!karg.build_id_size !=3D !!karg.build_id_addr) return -EINVAL; =20 - mm =3D priv->lock_ctx.mm; if (!mm || !mmget_not_zero(mm)) return -ESRCH; =20 - err =3D query_vma_setup(mm); + err =3D query_vma_setup(&lock_ctx); if (err) { mmput(mm); return err; } =20 - vma =3D query_matching_vma(mm, karg.query_addr, karg.query_flags); + vma =3D query_matching_vma(&lock_ctx, karg.query_addr, karg.query_flags); if (IS_ERR(vma)) { err =3D PTR_ERR(vma); vma =3D NULL; @@ -705,7 +757,7 @@ static int do_procmap_query(struct proc_maps_private *p= riv, void __user *uarg) } =20 /* unlock vma or mmap_lock, and put mm_struct before copying data to user= */ - query_vma_teardown(mm, vma); + query_vma_teardown(&lock_ctx); mmput(mm); =20 if (karg.vma_name_size && copy_to_user(u64_to_user_ptr(karg.vma_name_addr= ), @@ -725,7 +777,7 @@ static int do_procmap_query(struct proc_maps_private *p= riv, void __user *uarg) return 0; =20 out: - query_vma_teardown(mm, vma); + query_vma_teardown(&lock_ctx); mmput(mm); kfree(name_buf); return err; @@ -738,7 +790,7 @@ static long procfs_procmap_ioctl(struct file *file, uns= igned int cmd, unsigned l =20 switch (cmd) { case PROCMAP_QUERY: - return do_procmap_query(priv, (void __user *)arg); + return do_procmap_query(priv->lock_ctx.mm, (void __user *)arg); default: return -ENOIOCTLCMD; } --=20 2.50.1.565.gc32cd1483b-goog