From nobody Tue Dec 2 00:46:24 2025 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by smtp.subspace.kernel.org (Postfix) with ESMTP id 489E9222585; Mon, 24 Nov 2025 13:23:27 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1763990609; cv=none; b=rlkI6kJLEtuLkYgEWQllBpz60VNfS4U2DGGijdWotlDOSO05yAd4CJbtQ74pcct8pPR8DXSz87XBSNn4Thd7dRisWjeWMVnZal6WZpZLH7fJxBU59ygugB+ULaT6X4HZEh9nvDjPQshw78HzSy89oLijQnh5DeXTpKp3YV79zmY= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1763990609; c=relaxed/simple; bh=SpKZ57+V3b0Q5f22/LfL6G+JSBEdRSWZEv7eS39dgZU=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=ZeiW2+Llpd10MsYB2zcrM5sIIkqhaZrO2lALGWQfNPL8BeMOuzKu4F9t53xfWtjp9etY/KoRmig1Jnwu4a8jICRptuwND2lCPEkns7R1z7X0vbn1ZDkZBet54eaklFyvKUe05PMX8vu3xlp3CNeyKWvYSdzQxrHTEJ1cOLQlFjM= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 2C3A11655; Mon, 24 Nov 2025 05:23:19 -0800 (PST) Received: from e123572-lin.arm.com (e123572-lin.cambridge.arm.com [10.1.194.54]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id 7D3443F73B; Mon, 24 Nov 2025 05:23:21 -0800 (PST) From: Kevin Brodsky To: linux-mm@kvack.org Cc: linux-kernel@vger.kernel.org, Kevin Brodsky , Alexander Gordeev , Andreas Larsson , Andrew Morton , Boris Ostrovsky , Borislav Petkov , Catalin Marinas , Christophe Leroy , Dave Hansen , David Hildenbrand , "David S. Miller" , David Woodhouse , "H. Peter Anvin" , Ingo Molnar , Jann Horn , Juergen Gross , "Liam R. Howlett" , Lorenzo Stoakes , Madhavan Srinivasan , Michael Ellerman , Michal Hocko , Mike Rapoport , Nicholas Piggin , Peter Zijlstra , "Ritesh Harjani (IBM)" , Ryan Roberts , Suren Baghdasaryan , Thomas Gleixner , Venkat Rao Bagalkote , Vlastimil Babka , Will Deacon , Yeoreum Yun , linux-arm-kernel@lists.infradead.org, linuxppc-dev@lists.ozlabs.org, sparclinux@vger.kernel.org, xen-devel@lists.xenproject.org, x86@kernel.org Subject: [PATCH v5 08/12] mm: enable lazy_mmu sections to nest Date: Mon, 24 Nov 2025 13:22:24 +0000 Message-ID: <20251124132228.622678-9-kevin.brodsky@arm.com> X-Mailer: git-send-email 2.51.2 In-Reply-To: <20251124132228.622678-1-kevin.brodsky@arm.com> References: <20251124132228.622678-1-kevin.brodsky@arm.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Despite recent efforts to prevent lazy_mmu sections from nesting, it remains difficult to ensure that it never occurs - and in fact it does occur on arm64 in certain situations (CONFIG_DEBUG_PAGEALLOC). Commit 1ef3095b1405 ("arm64/mm: Permit lazy_mmu_mode to be nested") made nesting tolerable on arm64, but without truly supporting it: the inner call to leave() disables the batching optimisation before the outer section ends. This patch actually enables lazy_mmu sections to nest by tracking the nesting level in task_struct, in a similar fashion to e.g. pagefault_{enable,disable}(). This is fully handled by the generic lazy_mmu helpers that were recently introduced. lazy_mmu sections were not initially intended to nest, so we need to clarify the semantics w.r.t. the arch_*_lazy_mmu_mode() callbacks. This patch takes the following approach: * The outermost calls to lazy_mmu_mode_{enable,disable}() trigger calls to arch_{enter,leave}_lazy_mmu_mode() - this is unchanged. * Nested calls to lazy_mmu_mode_{enable,disable}() are not forwarded to the arch via arch_{enter,leave} - lazy MMU remains enabled so the assumption is that these callbacks are not relevant. However, existing code may rely on a call to disable() to flush any batched state, regardless of nesting. arch_flush_lazy_mmu_mode() is therefore called in that situation. A separate interface was recently introduced to temporarily pause the lazy MMU mode: lazy_mmu_mode_{pause,resume}(). pause() fully exits the mode *regardless of the nesting level*, and resume() restores the mode at the same nesting level. pause()/resume() are themselves allowed to nest, so we actually store two nesting levels in task_struct: enable_count and pause_count. A new helper in_lazy_mmu_mode() is introduced to determine whether we are currently in lazy MMU mode; this will be used in subsequent patches to replace the various ways arch's currently track whether the mode is enabled. In summary (enable/pause represent the values *after* the call): lazy_mmu_mode_enable() -> arch_enter() enable=3D1 pause=3D0 lazy_mmu_mode_enable() -> =C3=B8 enable=3D2 pause=3D0 lazy_mmu_mode_pause() -> arch_leave() enable=3D2 pause=3D1 lazy_mmu_mode_resume() -> arch_enter() enable=3D2 pause=3D0 lazy_mmu_mode_disable() -> arch_flush() enable=3D1 pause=3D0 lazy_mmu_mode_disable() -> arch_leave() enable=3D0 pause=3D0 Note: in_lazy_mmu_mode() is added to to allow arch headers included by to use it. Signed-off-by: Kevin Brodsky Acked-by: David Hildenbrand (Red Hat) --- arch/arm64/include/asm/pgtable.h | 12 ---- include/linux/mm_types_task.h | 5 ++ include/linux/pgtable.h | 115 +++++++++++++++++++++++++++++-- include/linux/sched.h | 45 ++++++++++++ 4 files changed, 158 insertions(+), 19 deletions(-) diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgta= ble.h index e596899f4029..a7d99dee3dc4 100644 --- a/arch/arm64/include/asm/pgtable.h +++ b/arch/arm64/include/asm/pgtable.h @@ -82,18 +82,6 @@ static inline void queue_pte_barriers(void) =20 static inline void arch_enter_lazy_mmu_mode(void) { - /* - * lazy_mmu_mode is not supposed to permit nesting. But in practice this - * does happen with CONFIG_DEBUG_PAGEALLOC, where a page allocation - * inside a lazy_mmu_mode section (such as zap_pte_range()) will change - * permissions on the linear map with apply_to_page_range(), which - * re-enters lazy_mmu_mode. So we tolerate nesting in our - * implementation. The first call to arch_leave_lazy_mmu_mode() will - * flush and clear the flag such that the remainder of the work in the - * outer nest behaves as if outside of lazy mmu mode. This is safe and - * keeps tracking simple. - */ - set_thread_flag(TIF_LAZY_MMU); } =20 diff --git a/include/linux/mm_types_task.h b/include/linux/mm_types_task.h index a82aa80c0ba4..11bf319d78ec 100644 --- a/include/linux/mm_types_task.h +++ b/include/linux/mm_types_task.h @@ -88,4 +88,9 @@ struct tlbflush_unmap_batch { #endif }; =20 +struct lazy_mmu_state { + u8 enable_count; + u8 pause_count; +}; + #endif /* _LINUX_MM_TYPES_TASK_H */ diff --git a/include/linux/pgtable.h b/include/linux/pgtable.h index 8ff6fdb4b13d..24fdb6f5c2e1 100644 --- a/include/linux/pgtable.h +++ b/include/linux/pgtable.h @@ -230,39 +230,140 @@ static inline int pmd_dirty(pmd_t pmd) * (In practice, for user PTE updates, the appropriate page table lock(s) = are * held, but for kernel PTE updates, no lock is held). The mode is disable= d in * interrupt context and calls to the lazy_mmu API have no effect. - * Nesting is not permitted. + * + * The lazy MMU mode is enabled for a given block of code using: + * + * lazy_mmu_mode_enable(); + * + * lazy_mmu_mode_disable(); + * + * Nesting is permitted: may itself use an enable()/disable() pair. + * A nested call to enable() has no functional effect; however disable() c= auses + * any batched architectural state to be flushed regardless of nesting. Af= ter a + * call to disable(), the caller can therefore rely on all previous page t= able + * modifications to have taken effect, but the lazy MMU mode may still be + * enabled. + * + * In certain cases, it may be desirable to temporarily pause the lazy MMU= mode. + * This can be done using: + * + * lazy_mmu_mode_pause(); + * + * lazy_mmu_mode_resume(); + * + * pause() ensures that the mode is exited regardless of the nesting level; + * resume() re-enters the mode at the same nesting level. Any call to the + * lazy_mmu_mode_* API between those two calls has no effect. In particula= r, + * this means that pause()/resume() pairs may nest. + * + * in_lazy_mmu_mode() can be used to check whether the lazy MMU mode is + * currently enabled. */ #ifdef CONFIG_ARCH_HAS_LAZY_MMU_MODE +/** + * lazy_mmu_mode_enable() - Enable the lazy MMU mode. + * + * Enters a new lazy MMU mode section; if the mode was not already enabled, + * enables it and calls arch_enter_lazy_mmu_mode(). + * + * Must be paired with a call to lazy_mmu_mode_disable(). + * + * Has no effect if called: + * - While paused - see lazy_mmu_mode_pause() + * - In interrupt context + */ static inline void lazy_mmu_mode_enable(void) { - if (in_interrupt()) + struct lazy_mmu_state *state =3D ¤t->lazy_mmu_state; + + if (in_interrupt() || state->pause_count > 0) return; =20 - arch_enter_lazy_mmu_mode(); + VM_WARN_ON_ONCE(state->enable_count =3D=3D U8_MAX); + + if (state->enable_count++ =3D=3D 0) + arch_enter_lazy_mmu_mode(); } =20 +/** + * lazy_mmu_mode_disable() - Disable the lazy MMU mode. + * + * Exits the current lazy MMU mode section. If it is the outermost section, + * disables the mode and calls arch_leave_lazy_mmu_mode(). Otherwise (nest= ed + * section), calls arch_flush_lazy_mmu_mode(). + * + * Must match a call to lazy_mmu_mode_enable(). + * + * Has no effect if called: + * - While paused - see lazy_mmu_mode_pause() + * - In interrupt context + */ static inline void lazy_mmu_mode_disable(void) { - if (in_interrupt()) + struct lazy_mmu_state *state =3D ¤t->lazy_mmu_state; + + if (in_interrupt() || state->pause_count > 0) return; =20 - arch_leave_lazy_mmu_mode(); + VM_WARN_ON_ONCE(state->enable_count =3D=3D 0); + + if (--state->enable_count =3D=3D 0) + arch_leave_lazy_mmu_mode(); + else /* Exiting a nested section */ + arch_flush_lazy_mmu_mode(); + } =20 +/** + * lazy_mmu_mode_pause() - Pause the lazy MMU mode. + * + * Pauses the lazy MMU mode; if it is currently active, disables it and ca= lls + * arch_leave_lazy_mmu_mode(). + * + * Must be paired with a call to lazy_mmu_mode_resume(). Calls to the + * lazy_mmu_mode_* API have no effect until the matching resume() call. + * + * Has no effect if called: + * - While paused (inside another pause()/resume() pair) + * - In interrupt context + */ static inline void lazy_mmu_mode_pause(void) { + struct lazy_mmu_state *state =3D ¤t->lazy_mmu_state; + if (in_interrupt()) return; =20 - arch_leave_lazy_mmu_mode(); + VM_WARN_ON_ONCE(state->pause_count =3D=3D U8_MAX); + + if (state->pause_count++ =3D=3D 0 && state->enable_count > 0) + arch_leave_lazy_mmu_mode(); } =20 +/** + * lazy_mmu_mode_pause() - Resume the lazy MMU mode. + * + * Resumes the lazy MMU mode; if it was active at the point where the matc= hing + * call to lazy_mmu_mode_pause() was made, re-enables it and calls + * arch_enter_lazy_mmu_mode(). + * + * Must match a call to lazy_mmu_mode_pause(). + * + * Has no effect if called: + * - While paused (inside another pause()/resume() pair) + * - In interrupt context + */ static inline void lazy_mmu_mode_resume(void) { + struct lazy_mmu_state *state =3D ¤t->lazy_mmu_state; + if (in_interrupt()) return; =20 - arch_enter_lazy_mmu_mode(); + VM_WARN_ON_ONCE(state->pause_count =3D=3D 0); + + if (--state->pause_count =3D=3D 0 && state->enable_count > 0) + arch_enter_lazy_mmu_mode(); } #else static inline void lazy_mmu_mode_enable(void) {} diff --git a/include/linux/sched.h b/include/linux/sched.h index b469878de25c..847e242376db 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1441,6 +1441,10 @@ struct task_struct { =20 struct page_frag task_frag; =20 +#ifdef CONFIG_ARCH_HAS_LAZY_MMU_MODE + struct lazy_mmu_state lazy_mmu_state; +#endif + #ifdef CONFIG_TASK_DELAY_ACCT struct task_delay_info *delays; #endif @@ -1724,6 +1728,47 @@ static inline char task_state_to_char(struct task_st= ruct *tsk) return task_index_to_char(task_state_index(tsk)); } =20 +#ifdef CONFIG_ARCH_HAS_LAZY_MMU_MODE +/** + * __task_lazy_mmu_mode_active() - Test the lazy MMU mode state for a task. + * @tsk: The task to check. + * + * Test whether @tsk has its lazy MMU mode state set to active (i.e. enabl= ed + * and not paused). + * + * This function only considers the state saved in task_struct; to test wh= ether + * current actually is in lazy MMU mode, in_lazy_mmu_mode() should be used + * instead. + * + * This function is intended for architectures that implement the lazy MMU + * mode; it must not be called from generic code. + */ +static inline bool __task_lazy_mmu_mode_active(struct task_struct *tsk) +{ + struct lazy_mmu_state *state =3D &tsk->lazy_mmu_state; + + return state->enable_count > 0 && state->pause_count =3D=3D 0; +} + +/** + * in_lazy_mmu_mode() - Test whether we are currently in lazy MMU mode. + * + * Test whether the current context is in lazy MMU mode. This is true if b= oth: + * 1. We are not in interrupt context + * 2. Lazy MMU mode is active for the current task + * + * This function is intended for architectures that implement the lazy MMU + * mode; it must not be called from generic code. + */ +static inline bool in_lazy_mmu_mode(void) +{ + if (in_interrupt()) + return false; + + return __task_lazy_mmu_mode_active(current); +} +#endif + extern struct pid *cad_pid; =20 /* --=20 2.51.2