From patchwork Mon Oct 28 01:08:15 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Andrii Nakryiko X-Patchwork-Id: 13852768 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id C700814D43D; Mon, 28 Oct 2024 01:09:09 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1730077749; cv=none; b=YhNUQ92WVsN5GCuKyMlz6xBxLy8a3uquaByT6pDw4qkP+pOcaYyvLoTuuKEO/7Ab2UlD17bXMGIZMeBo4ZRtHLQHOdbL/jmaRgrO3AJyjx3Ns0sMxtgKzMBpP37rRG91TV+Rzpvpp2bMkhEeVVCLZZbZ5pkQRmK/Ty9mOOTQNbo= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1730077749; c=relaxed/simple; bh=Ribok69L0V+IQIfX1LKAa+j7slCTXJ/a5HygjFWCqNY=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=RxsO92yJ4Z8j2ZRVV6mfA0Iy/+6x0Rtvys/L2b21YCewEwIE2aIl6XPewj7oKAgFU+44iDnr8xm3JcDPludSOtzFHRGwpOZpeQmo3S1mgVGBskx9Xa7pA9x1PFuLhj5IMVxghtrXadKd2lE0rYvhEd42ufeJ523+tkZI+MKGWLY= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=XcEsDuxZ; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="XcEsDuxZ" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 65703C4CEE5; Mon, 28 Oct 2024 01:09:09 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1730077749; bh=Ribok69L0V+IQIfX1LKAa+j7slCTXJ/a5HygjFWCqNY=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=XcEsDuxZ44cpUaR4KUbORo/ERRrafxKAF05h7Dd+n1ohShoR33iYzss2xVzkX9bZy NUA2c2/P93p+1NY9BMPbz4ZhkGNAColwIYcLIkLlh2xb7UNFCr/2OVBIwND0TfPClO beYu28enquoY3AqAvW6Rn0Hh1VpYSsaOhhkqDSzC3pNPmq+ARjE5S+TFvAdIRba7vC B251sDrDieSzrwoxJpURaYek9EDwPpQG1pvuqNwhl+W23/1EpA/uFDnIxtKnlI9F/A FmYTcm7H7O71F77iLvdRJxipSDNTwtQR1UO+1v/V0OA1iqT0qzRg8WsMBTP86OkiSe E2A9BGrGM6ZnA== From: Andrii Nakryiko To: linux-trace-kernel@vger.kernel.org, linux-mm@kvack.org, akpm@linux-foundation.org, peterz@infradead.org Cc: oleg@redhat.com, rostedt@goodmis.org, mhiramat@kernel.org, bpf@vger.kernel.org, linux-kernel@vger.kernel.org, jolsa@kernel.org, paulmck@kernel.org, willy@infradead.org, surenb@google.com, mjguzik@gmail.com, brauner@kernel.org, jannh@google.com, mhocko@kernel.org, vbabka@suse.cz, shakeel.butt@linux.dev, hannes@cmpxchg.org, Liam.Howlett@oracle.com, lorenzo.stoakes@oracle.com, david@redhat.com, arnd@arndb.de, richard.weiyang@gmail.com, zhangpeng.00@bytedance.com, linmiaohe@huawei.com, viro@zeniv.linux.org.uk, hca@linux.ibm.com, Andrii Nakryiko Subject: [PATCH v4 tip/perf/core 1/4] mm: Convert mm_lock_seq to a proper seqcount Date: Sun, 27 Oct 2024 18:08:15 -0700 Message-ID: <20241028010818.2487581-2-andrii@kernel.org> X-Mailer: git-send-email 2.43.5 In-Reply-To: <20241028010818.2487581-1-andrii@kernel.org> References: <20241028010818.2487581-1-andrii@kernel.org> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 From: Suren Baghdasaryan Convert mm_lock_seq to be seqcount_t and change all mmap_write_lock variants to increment it, in-line with the usual seqcount usage pattern. This lets us check whether the mmap_lock is write-locked by checking mm_lock_seq.sequence counter (odd=locked, even=unlocked). This will be used when implementing mmap_lock speculation functions. As a result vm_lock_seq is also change to be unsigned to match the type of mm_lock_seq.sequence. Suggested-by: Peter Zijlstra Signed-off-by: Suren Baghdasaryan Signed-off-by: Andrii Nakryiko --- include/linux/mm.h | 12 +++---- include/linux/mm_types.h | 7 ++-- include/linux/mmap_lock.h | 58 +++++++++++++++++++++----------- kernel/fork.c | 5 +-- mm/init-mm.c | 2 +- tools/testing/vma/vma.c | 4 +-- tools/testing/vma/vma_internal.h | 4 +-- 7 files changed, 56 insertions(+), 36 deletions(-) diff --git a/include/linux/mm.h b/include/linux/mm.h index ecf63d2b0582..94b537088142 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -698,7 +698,7 @@ static inline bool vma_start_read(struct vm_area_struct *vma) * we don't rely on for anything - the mm_lock_seq read against which we * need ordering is below. */ - if (READ_ONCE(vma->vm_lock_seq) == READ_ONCE(vma->vm_mm->mm_lock_seq)) + if (READ_ONCE(vma->vm_lock_seq) == READ_ONCE(vma->vm_mm->mm_lock_seq.sequence)) return false; if (unlikely(down_read_trylock(&vma->vm_lock->lock) == 0)) @@ -715,7 +715,7 @@ static inline bool vma_start_read(struct vm_area_struct *vma) * after it has been unlocked. * This pairs with RELEASE semantics in vma_end_write_all(). */ - if (unlikely(vma->vm_lock_seq == smp_load_acquire(&vma->vm_mm->mm_lock_seq))) { + if (unlikely(vma->vm_lock_seq == raw_read_seqcount(&vma->vm_mm->mm_lock_seq))) { up_read(&vma->vm_lock->lock); return false; } @@ -730,7 +730,7 @@ static inline void vma_end_read(struct vm_area_struct *vma) } /* WARNING! Can only be used if mmap_lock is expected to be write-locked */ -static bool __is_vma_write_locked(struct vm_area_struct *vma, int *mm_lock_seq) +static bool __is_vma_write_locked(struct vm_area_struct *vma, unsigned int *mm_lock_seq) { mmap_assert_write_locked(vma->vm_mm); @@ -738,7 +738,7 @@ static bool __is_vma_write_locked(struct vm_area_struct *vma, int *mm_lock_seq) * current task is holding mmap_write_lock, both vma->vm_lock_seq and * mm->mm_lock_seq can't be concurrently modified. */ - *mm_lock_seq = vma->vm_mm->mm_lock_seq; + *mm_lock_seq = vma->vm_mm->mm_lock_seq.sequence; return (vma->vm_lock_seq == *mm_lock_seq); } @@ -749,7 +749,7 @@ static bool __is_vma_write_locked(struct vm_area_struct *vma, int *mm_lock_seq) */ static inline void vma_start_write(struct vm_area_struct *vma) { - int mm_lock_seq; + unsigned int mm_lock_seq; if (__is_vma_write_locked(vma, &mm_lock_seq)) return; @@ -767,7 +767,7 @@ static inline void vma_start_write(struct vm_area_struct *vma) static inline void vma_assert_write_locked(struct vm_area_struct *vma) { - int mm_lock_seq; + unsigned int mm_lock_seq; VM_BUG_ON_VMA(!__is_vma_write_locked(vma, &mm_lock_seq), vma); } diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 6e3bdf8e38bc..76e0cdc0462b 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -715,7 +715,7 @@ struct vm_area_struct { * counter reuse can only lead to occasional unnecessary use of the * slowpath. */ - int vm_lock_seq; + unsigned int vm_lock_seq; /* Unstable RCU readers are allowed to read this. */ struct vma_lock *vm_lock; #endif @@ -887,6 +887,9 @@ struct mm_struct { * Roughly speaking, incrementing the sequence number is * equivalent to releasing locks on VMAs; reading the sequence * number can be part of taking a read lock on a VMA. + * Incremented every time mmap_lock is write-locked/unlocked. + * Initialized to 0, therefore odd values indicate mmap_lock + * is write-locked and even values that it's released. * * Can be modified under write mmap_lock using RELEASE * semantics. @@ -895,7 +898,7 @@ struct mm_struct { * Can be read with ACQUIRE semantics if not holding write * mmap_lock. */ - int mm_lock_seq; + seqcount_t mm_lock_seq; #endif diff --git a/include/linux/mmap_lock.h b/include/linux/mmap_lock.h index de9dc20b01ba..6b3272686860 100644 --- a/include/linux/mmap_lock.h +++ b/include/linux/mmap_lock.h @@ -71,39 +71,38 @@ static inline void mmap_assert_write_locked(const struct mm_struct *mm) } #ifdef CONFIG_PER_VMA_LOCK -/* - * Drop all currently-held per-VMA locks. - * This is called from the mmap_lock implementation directly before releasing - * a write-locked mmap_lock (or downgrading it to read-locked). - * This should normally NOT be called manually from other places. - * If you want to call this manually anyway, keep in mind that this will release - * *all* VMA write locks, including ones from further up the stack. - */ -static inline void vma_end_write_all(struct mm_struct *mm) +static inline void mm_lock_seqcount_init(struct mm_struct *mm) { - mmap_assert_write_locked(mm); - /* - * Nobody can concurrently modify mm->mm_lock_seq due to exclusive - * mmap_lock being held. - * We need RELEASE semantics here to ensure that preceding stores into - * the VMA take effect before we unlock it with this store. - * Pairs with ACQUIRE semantics in vma_start_read(). - */ - smp_store_release(&mm->mm_lock_seq, mm->mm_lock_seq + 1); + seqcount_init(&mm->mm_lock_seq); +} + +static inline void mm_lock_seqcount_begin(struct mm_struct *mm) +{ + do_raw_write_seqcount_begin(&mm->mm_lock_seq); +} + +static inline void mm_lock_seqcount_end(struct mm_struct *mm) +{ + do_raw_write_seqcount_end(&mm->mm_lock_seq); } + #else -static inline void vma_end_write_all(struct mm_struct *mm) {} +static inline void mm_lock_seqcount_init(struct mm_struct *mm) {} +static inline void mm_lock_seqcount_begin(struct mm_struct *mm) {} +static inline void mm_lock_seqcount_end(struct mm_struct *mm) {} #endif static inline void mmap_init_lock(struct mm_struct *mm) { init_rwsem(&mm->mmap_lock); + mm_lock_seqcount_init(mm); } static inline void mmap_write_lock(struct mm_struct *mm) { __mmap_lock_trace_start_locking(mm, true); down_write(&mm->mmap_lock); + mm_lock_seqcount_begin(mm); __mmap_lock_trace_acquire_returned(mm, true, true); } @@ -111,6 +110,7 @@ static inline void mmap_write_lock_nested(struct mm_struct *mm, int subclass) { __mmap_lock_trace_start_locking(mm, true); down_write_nested(&mm->mmap_lock, subclass); + mm_lock_seqcount_begin(mm); __mmap_lock_trace_acquire_returned(mm, true, true); } @@ -120,10 +120,30 @@ static inline int mmap_write_lock_killable(struct mm_struct *mm) __mmap_lock_trace_start_locking(mm, true); ret = down_write_killable(&mm->mmap_lock); + if (!ret) + mm_lock_seqcount_begin(mm); __mmap_lock_trace_acquire_returned(mm, true, ret == 0); return ret; } +/* + * Drop all currently-held per-VMA locks. + * This is called from the mmap_lock implementation directly before releasing + * a write-locked mmap_lock (or downgrading it to read-locked). + * This should normally NOT be called manually from other places. + * If you want to call this manually anyway, keep in mind that this will release + * *all* VMA write locks, including ones from further up the stack. + */ +static inline void vma_end_write_all(struct mm_struct *mm) +{ + mmap_assert_write_locked(mm); + /* + * Nobody can concurrently modify mm->mm_lock_seq due to exclusive + * mmap_lock being held. + */ + mm_lock_seqcount_end(mm); +} + static inline void mmap_write_unlock(struct mm_struct *mm) { __mmap_lock_trace_released(mm, true); diff --git a/kernel/fork.c b/kernel/fork.c index 89ceb4a68af2..55c4088543dc 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -448,7 +448,7 @@ static bool vma_lock_alloc(struct vm_area_struct *vma) return false; init_rwsem(&vma->vm_lock->lock); - vma->vm_lock_seq = -1; + vma->vm_lock_seq = UINT_MAX; return true; } @@ -1261,9 +1261,6 @@ static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p, seqcount_init(&mm->write_protect_seq); mmap_init_lock(mm); INIT_LIST_HEAD(&mm->mmlist); -#ifdef CONFIG_PER_VMA_LOCK - mm->mm_lock_seq = 0; -#endif mm_pgtables_bytes_init(mm); mm->map_count = 0; mm->locked_vm = 0; diff --git a/mm/init-mm.c b/mm/init-mm.c index 24c809379274..6af3ad675930 100644 --- a/mm/init-mm.c +++ b/mm/init-mm.c @@ -40,7 +40,7 @@ struct mm_struct init_mm = { .arg_lock = __SPIN_LOCK_UNLOCKED(init_mm.arg_lock), .mmlist = LIST_HEAD_INIT(init_mm.mmlist), #ifdef CONFIG_PER_VMA_LOCK - .mm_lock_seq = 0, + .mm_lock_seq = SEQCNT_ZERO(init_mm.mm_lock_seq), #endif .user_ns = &init_user_ns, .cpu_bitmap = CPU_BITS_NONE, diff --git a/tools/testing/vma/vma.c b/tools/testing/vma/vma.c index c53f220eb6cc..bcdf831dfe3e 100644 --- a/tools/testing/vma/vma.c +++ b/tools/testing/vma/vma.c @@ -87,7 +87,7 @@ static struct vm_area_struct *alloc_and_link_vma(struct mm_struct *mm, * begun. Linking to the tree will have caused this to be incremented, * which means we will get a false positive otherwise. */ - vma->vm_lock_seq = -1; + vma->vm_lock_seq = UINT_MAX; return vma; } @@ -212,7 +212,7 @@ static bool vma_write_started(struct vm_area_struct *vma) int seq = vma->vm_lock_seq; /* We reset after each check. */ - vma->vm_lock_seq = -1; + vma->vm_lock_seq = UINT_MAX; /* The vma_start_write() stub simply increments this value. */ return seq > -1; diff --git a/tools/testing/vma/vma_internal.h b/tools/testing/vma/vma_internal.h index c5b9da034511..4007ec580f85 100644 --- a/tools/testing/vma/vma_internal.h +++ b/tools/testing/vma/vma_internal.h @@ -231,7 +231,7 @@ struct vm_area_struct { * counter reuse can only lead to occasional unnecessary use of the * slowpath. */ - int vm_lock_seq; + unsigned int vm_lock_seq; struct vma_lock *vm_lock; #endif @@ -406,7 +406,7 @@ static inline bool vma_lock_alloc(struct vm_area_struct *vma) return false; init_rwsem(&vma->vm_lock->lock); - vma->vm_lock_seq = -1; + vma->vm_lock_seq = UINT_MAX; return true; }