From mboxrd@z Thu Jan 1 00:00:00 1970 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 52FC622F15E for ; Wed, 21 May 2025 17:49:07 +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=1747849748; cv=none; b=YzXMvU2slFXNinp9TW5r18gqdtihzAv5lIspFM+g+ZXfPp/aQ7XJxlnh6gWMBKxHRuT4DYGyHadgbyrlvxrLx9EnILITKvxQwRddsO3aD5ncUl0kzG2dsukV8Eqk/cNoXtuJmvVQQGvbPm3mSn/Aj9nhr5vfSEqErUhf+6G72tM= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1747849748; c=relaxed/simple; bh=sV+fkYDTYSoZQxgyIxAo3ZKvc3REGDYynP4xsS3e5PM=; h=Date:To:From:Subject:Message-Id; b=tj9MkDMYKoMjokzwl1hNGr1uYi31QkynQdCnFcpO+L2lNuKtQ5Y6FK/z8e/jC7h6Z6+jykl4EHUHL+NOF5QSfv6CM11RBBTdfr4lLnXiQT4XC0u7npgP9FkDYIrH5y+OuLCHEpIwkvr0n9Ojk/wh/7NcQ45QyBaN8dL8DrGnJl4= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux-foundation.org header.i=@linux-foundation.org header.b=A8iD7tev; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux-foundation.org header.i=@linux-foundation.org header.b="A8iD7tev" Received: by smtp.kernel.org (Postfix) with ESMTPSA id BE5D4C4CEEB; Wed, 21 May 2025 17:49:07 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linux-foundation.org; s=korg; t=1747849747; bh=sV+fkYDTYSoZQxgyIxAo3ZKvc3REGDYynP4xsS3e5PM=; h=Date:To:From:Subject:From; b=A8iD7tevifiJBA/nbLQBvKCng0oDC4zqyXDwONEvpl8yR3d0famVrfr83lCqwav9V zUtI5XR04AL6H5RHRKM6wrp32unJZuZH5NG5GUSL7pLs9C0H9FgQTso4DCD6p0ITJ6 xpxN+WL3a1WWfHRCXppll2bVL3XzuMEVuwm57IAo= Date: Wed, 21 May 2025 10:49:07 -0700 To: mm-commits@vger.kernel.org,pasha.tatashin@soleen.com,mjguzik@gmail.com,linus.walleij@linaro.org,akpm@linux-foundation.org From: Andrew Morton Subject: [merged mm-nonmm-stable] fork-define-a-local-gfp_vmap_stack.patch removed from -mm tree Message-Id: <20250521174907.BE5D4C4CEEB@smtp.kernel.org> Precedence: bulk X-Mailing-List: mm-commits@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: The quilt patch titled Subject: fork: define a local GFP_VMAP_STACK has been removed from the -mm tree. Its filename was fork-define-a-local-gfp_vmap_stack.patch This patch was dropped because it was merged into the mm-nonmm-stable branch of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm ------------------------------------------------------ From: Linus Walleij Subject: fork: define a local GFP_VMAP_STACK Date: Fri, 09 May 2025 09:25:09 +0200 The current allocation of VMAP stack memory is using (THREADINFO_GFP & ~__GFP_ACCOUNT) which is a complicated way of saying (GFP_KERNEL | __GFP_ZERO): : define THREADINFO_GFP (GFP_KERNEL_ACCOUNT | __GFP_ZERO) : define GFP_KERNEL_ACCOUNT (GFP_KERNEL | __GFP_ACCOUNT) This is an unfortunate side-effect of independent changes blurring the picture: commit 19809c2da28aee5860ad9a2eff760730a0710df0 changed (THREADINFO_GFP | __GFP_HIGHMEM) to just THREADINFO_GFP since highmem became implicit. commit 9b6f7e163cd0f468d1b9696b785659d3c27c8667 then added stack caching and rewrote the allocation to (THREADINFO_GFP & ~__GFP_ACCOUNT) as cached stacks need to be accounted separately. However that code, when it eventually accounts the memory does this: ret = memcg_kmem_charge(vm->pages[i], GFP_KERNEL, 0) so the memory is charged as a GFP_KERNEL allocation. Define a unique GFP_VMAP_STACK to use GFP_KERNEL | __GFP_ZERO and move the comment there. Link: https://lkml.kernel.org/r/20250509-gfp-stack-v1-1-82f6f7efc210@linaro.org Signed-off-by: Linus Walleij Reported-by: Mateusz Guzik Cc: Pasha Tatashin Signed-off-by: Andrew Morton --- kernel/fork.c | 88 ++++++++++++++++++++++++------------------------ 1 file changed, 45 insertions(+), 43 deletions(-) --- a/kernel/fork.c~fork-define-a-local-gfp_vmap_stack +++ a/kernel/fork.c @@ -185,7 +185,13 @@ static inline void free_task_struct(stru kmem_cache_free(task_struct_cachep, tsk); } -#ifdef CONFIG_VMAP_STACK +/* + * Allocate pages if THREAD_SIZE is >= PAGE_SIZE, otherwise use a + * kmemcache based allocator. + */ +# if THREAD_SIZE >= PAGE_SIZE || defined(CONFIG_VMAP_STACK) + +# ifdef CONFIG_VMAP_STACK /* * vmalloc() is a bit slow, and calling vfree() enough times will force a TLB * flush. Try to minimize the number of calls by caching stacks. @@ -198,14 +204,14 @@ struct vm_stack { struct vm_struct *stack_vm_area; }; -static bool try_release_thread_stack_to_cache(struct vm_struct *vm_area) +static bool try_release_thread_stack_to_cache(struct vm_struct *vm) { unsigned int i; for (i = 0; i < NR_CACHED_STACKS; i++) { struct vm_struct *tmp = NULL; - if (this_cpu_try_cmpxchg(cached_stacks[i], &tmp, vm_area)) + if (this_cpu_try_cmpxchg(cached_stacks[i], &tmp, vm)) return true; } return false; @@ -214,12 +220,11 @@ static bool try_release_thread_stack_to_ static void thread_stack_free_rcu(struct rcu_head *rh) { struct vm_stack *vm_stack = container_of(rh, struct vm_stack, rcu); - struct vm_struct *vm_area = vm_stack->stack_vm_area; if (try_release_thread_stack_to_cache(vm_stack->stack_vm_area)) return; - vfree(vm_area->addr); + vfree(vm_stack); } static void thread_stack_delayed_free(struct task_struct *tsk) @@ -232,32 +237,32 @@ static void thread_stack_delayed_free(st static int free_vm_stack_cache(unsigned int cpu) { - struct vm_struct **cached_vm_stack_areas = per_cpu_ptr(cached_stacks, cpu); + struct vm_struct **cached_vm_stacks = per_cpu_ptr(cached_stacks, cpu); int i; for (i = 0; i < NR_CACHED_STACKS; i++) { - struct vm_struct *vm_area = cached_vm_stack_areas[i]; + struct vm_struct *vm_stack = cached_vm_stacks[i]; - if (!vm_area) + if (!vm_stack) continue; - vfree(vm_area->addr); - cached_vm_stack_areas[i] = NULL; + vfree(vm_stack->addr); + cached_vm_stacks[i] = NULL; } return 0; } -static int memcg_charge_kernel_stack(struct vm_struct *vm_area) +static int memcg_charge_kernel_stack(struct vm_struct *vm) { int i; int ret; int nr_charged = 0; - BUG_ON(vm_area->nr_pages != THREAD_SIZE / PAGE_SIZE); + BUG_ON(vm->nr_pages != THREAD_SIZE / PAGE_SIZE); for (i = 0; i < THREAD_SIZE / PAGE_SIZE; i++) { - ret = memcg_kmem_charge_page(vm_area->pages[i], GFP_KERNEL, 0); + ret = memcg_kmem_charge_page(vm->pages[i], GFP_KERNEL, 0); if (ret) goto err; nr_charged++; @@ -265,35 +270,38 @@ static int memcg_charge_kernel_stack(str return 0; err: for (i = 0; i < nr_charged; i++) - memcg_kmem_uncharge_page(vm_area->pages[i], 0); + memcg_kmem_uncharge_page(vm->pages[i], 0); return ret; } static int alloc_thread_stack_node(struct task_struct *tsk, int node) { - struct vm_struct *vm_area; + struct vm_struct *vm; void *stack; int i; for (i = 0; i < NR_CACHED_STACKS; i++) { - vm_area = this_cpu_xchg(cached_stacks[i], NULL); - if (!vm_area) - continue; + struct vm_struct *s; - if (memcg_charge_kernel_stack(vm_area)) { - vfree(vm_area->addr); - return -ENOMEM; - } + s = this_cpu_xchg(cached_stacks[i], NULL); + + if (!s) + continue; /* Reset stack metadata. */ - kasan_unpoison_range(vm_area->addr, THREAD_SIZE); + kasan_unpoison_range(s->addr, THREAD_SIZE); - stack = kasan_reset_tag(vm_area->addr); + stack = kasan_reset_tag(s->addr); /* Clear stale pointers from reused stack. */ memset(stack, 0, THREAD_SIZE); - tsk->stack_vm_area = vm_area; + if (memcg_charge_kernel_stack(s)) { + vfree(s->addr); + return -ENOMEM; + } + + tsk->stack_vm_area = s; tsk->stack = stack; return 0; } @@ -309,8 +317,8 @@ static int alloc_thread_stack_node(struc if (!stack) return -ENOMEM; - vm_area = find_vm_area(stack); - if (memcg_charge_kernel_stack(vm_area)) { + vm = find_vm_area(stack); + if (memcg_charge_kernel_stack(vm)) { vfree(stack); return -ENOMEM; } @@ -319,7 +327,7 @@ static int alloc_thread_stack_node(struc * free_thread_stack() can be called in interrupt context, * so cache the vm_struct. */ - tsk->stack_vm_area = vm_area; + tsk->stack_vm_area = vm; stack = kasan_reset_tag(stack); tsk->stack = stack; return 0; @@ -334,13 +342,7 @@ static void free_thread_stack(struct tas tsk->stack_vm_area = NULL; } -#else /* !CONFIG_VMAP_STACK */ - -/* - * Allocate pages if THREAD_SIZE is >= PAGE_SIZE, otherwise use a - * kmemcache based allocator. - */ -#if THREAD_SIZE >= PAGE_SIZE +# else /* !CONFIG_VMAP_STACK */ static void thread_stack_free_rcu(struct rcu_head *rh) { @@ -372,7 +374,8 @@ static void free_thread_stack(struct tas tsk->stack = NULL; } -#else /* !(THREAD_SIZE >= PAGE_SIZE) */ +# endif /* CONFIG_VMAP_STACK */ +# else /* !(THREAD_SIZE >= PAGE_SIZE || defined(CONFIG_VMAP_STACK)) */ static struct kmem_cache *thread_stack_cache; @@ -411,8 +414,7 @@ void thread_stack_cache_init(void) BUG_ON(thread_stack_cache == NULL); } -#endif /* THREAD_SIZE >= PAGE_SIZE */ -#endif /* CONFIG_VMAP_STACK */ +# endif /* THREAD_SIZE >= PAGE_SIZE || defined(CONFIG_VMAP_STACK) */ /* SLAB cache for signal_struct structures (tsk->signal) */ static struct kmem_cache *signal_cachep; @@ -515,11 +517,11 @@ void vm_area_free(struct vm_area_struct static void account_kernel_stack(struct task_struct *tsk, int account) { if (IS_ENABLED(CONFIG_VMAP_STACK)) { - struct vm_struct *vm_area = task_stack_vm_area(tsk); + struct vm_struct *vm = task_stack_vm_area(tsk); int i; for (i = 0; i < THREAD_SIZE / PAGE_SIZE; i++) - mod_lruvec_page_state(vm_area->pages[i], NR_KERNEL_STACK_KB, + mod_lruvec_page_state(vm->pages[i], NR_KERNEL_STACK_KB, account * (PAGE_SIZE / 1024)); } else { void *stack = task_stack_page(tsk); @@ -535,12 +537,12 @@ void exit_task_stack_account(struct task account_kernel_stack(tsk, -1); if (IS_ENABLED(CONFIG_VMAP_STACK)) { - struct vm_struct *vm_area; + struct vm_struct *vm; int i; - vm_area = task_stack_vm_area(tsk); + vm = task_stack_vm_area(tsk); for (i = 0; i < THREAD_SIZE / PAGE_SIZE; i++) - memcg_kmem_uncharge_page(vm_area->pages[i], 0); + memcg_kmem_uncharge_page(vm->pages[i], 0); } } _ Patches currently in -mm which might be from linus.walleij@linaro.org are