Linux Documentation
 help / color / mirror / Atom feed
From: Suren Baghdasaryan <surenb@google.com>
To: akpm@linux-foundation.org
Cc: peterz@infradead.org, willy@infradead.org,
	liam.howlett@oracle.com,  lorenzo.stoakes@oracle.com,
	david.laight.linux@gmail.com, mhocko@suse.com,  vbabka@suse.cz,
	hannes@cmpxchg.org, mjguzik@gmail.com, oliver.sang@intel.com,
	 mgorman@techsingularity.net, david@redhat.com,
	peterx@redhat.com,  oleg@redhat.com, dave@stgolabs.net,
	paulmck@kernel.org, brauner@kernel.org,  dhowells@redhat.com,
	hdanton@sina.com, hughd@google.com,  lokeshgidra@google.com,
	minchan@google.com, jannh@google.com,  shakeel.butt@linux.dev,
	souravpanda@google.com, pasha.tatashin@soleen.com,
	 klarasmodin@gmail.com, richard.weiyang@gmail.com,
	corbet@lwn.net,  linux-doc@vger.kernel.org, linux-mm@kvack.org,
	linux-kernel@vger.kernel.org,  kernel-team@android.com,
	surenb@google.com
Subject: [PATCH v10 13/18] mm: move lesser used vma_area_struct members into the last cacheline
Date: Thu, 13 Feb 2025 14:46:50 -0800	[thread overview]
Message-ID: <20250213224655.1680278-14-surenb@google.com> (raw)
In-Reply-To: <20250213224655.1680278-1-surenb@google.com>

Move several vma_area_struct members which are rarely or never used
during page fault handling into the last cacheline to better pack
vm_area_struct. As a result vm_area_struct will fit into 3 as opposed
to 4 cachelines. New typical vm_area_struct layout:

struct vm_area_struct {
    union {
        struct {
            long unsigned int vm_start;              /*     0     8 */
            long unsigned int vm_end;                /*     8     8 */
        };                                           /*     0    16 */
        freeptr_t          vm_freeptr;               /*     0     8 */
    };                                               /*     0    16 */
    struct mm_struct *         vm_mm;                /*    16     8 */
    pgprot_t                   vm_page_prot;         /*    24     8 */
    union {
        const vm_flags_t   vm_flags;                 /*    32     8 */
        vm_flags_t         __vm_flags;               /*    32     8 */
    };                                               /*    32     8 */
    unsigned int               vm_lock_seq;          /*    40     4 */

    /* XXX 4 bytes hole, try to pack */

    struct list_head           anon_vma_chain;       /*    48    16 */
    /* --- cacheline 1 boundary (64 bytes) --- */
    struct anon_vma *          anon_vma;             /*    64     8 */
    const struct vm_operations_struct  * vm_ops;     /*    72     8 */
    long unsigned int          vm_pgoff;             /*    80     8 */
    struct file *              vm_file;              /*    88     8 */
    void *                     vm_private_data;      /*    96     8 */
    atomic_long_t              swap_readahead_info;  /*   104     8 */
    struct mempolicy *         vm_policy;            /*   112     8 */
    struct vma_numab_state *   numab_state;          /*   120     8 */
    /* --- cacheline 2 boundary (128 bytes) --- */
    refcount_t          vm_refcnt (__aligned__(64)); /*   128     4 */

    /* XXX 4 bytes hole, try to pack */

    struct {
        struct rb_node     rb (__aligned__(8));      /*   136    24 */
        long unsigned int  rb_subtree_last;          /*   160     8 */
    } __attribute__((__aligned__(8))) shared;        /*   136    32 */
    struct anon_vma_name *     anon_name;            /*   168     8 */
    struct vm_userfaultfd_ctx  vm_userfaultfd_ctx;   /*   176     8 */

    /* size: 192, cachelines: 3, members: 18 */
    /* sum members: 176, holes: 2, sum holes: 8 */
    /* padding: 8 */
    /* forced alignments: 2, forced holes: 1, sum forced holes: 4 */
} __attribute__((__aligned__(64)));

Memory consumption per 1000 VMAs becomes 48 pages:

    slabinfo after vm_area_struct changes:
     <name>           ... <objsize> <objperslab> <pagesperslab> : ...
     vm_area_struct   ...    192   42    2 : ...

Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
---
Changes since v9 [1]:
- Update vm_area_struct for tests, per Lorenzo Stoakes
- Add Reviewed-by, per Lorenzo Stoakes

[1] https://lore.kernel.org/all/20250111042604.3230628-13-surenb@google.com/

 include/linux/mm_types.h         | 38 +++++++++++++++-----------------
 tools/testing/vma/vma_internal.h | 37 +++++++++++++++----------------
 2 files changed, 36 insertions(+), 39 deletions(-)

diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 48ddfedfff83..63ab51699120 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -735,17 +735,6 @@ struct vm_area_struct {
 	 */
 	unsigned int vm_lock_seq;
 #endif
-
-	/*
-	 * For areas with an address space and backing store,
-	 * linkage into the address_space->i_mmap interval tree.
-	 *
-	 */
-	struct {
-		struct rb_node rb;
-		unsigned long rb_subtree_last;
-	} shared;
-
 	/*
 	 * A file's MAP_PRIVATE vma can be in both i_mmap tree and anon_vma
 	 * list, after a COW of one of the file pages.	A MAP_SHARED vma
@@ -765,14 +754,6 @@ struct vm_area_struct {
 	struct file * vm_file;		/* File we map to (can be NULL). */
 	void * vm_private_data;		/* was vm_pte (shared mem) */
 
-#ifdef CONFIG_ANON_VMA_NAME
-	/*
-	 * For private and shared anonymous mappings, a pointer to a null
-	 * terminated string containing the name given to the vma, or NULL if
-	 * unnamed. Serialized by mmap_lock. Use anon_vma_name to access.
-	 */
-	struct anon_vma_name *anon_name;
-#endif
 #ifdef CONFIG_SWAP
 	atomic_long_t swap_readahead_info;
 #endif
@@ -785,7 +766,6 @@ struct vm_area_struct {
 #ifdef CONFIG_NUMA_BALANCING
 	struct vma_numab_state *numab_state;	/* NUMA Balancing state */
 #endif
-	struct vm_userfaultfd_ctx vm_userfaultfd_ctx;
 #ifdef CONFIG_PER_VMA_LOCK
 	/* Unstable RCU readers are allowed to read this. */
 	refcount_t vm_refcnt ____cacheline_aligned_in_smp;
@@ -793,6 +773,24 @@ struct vm_area_struct {
 	struct lockdep_map vmlock_dep_map;
 #endif
 #endif
+	/*
+	 * For areas with an address space and backing store,
+	 * linkage into the address_space->i_mmap interval tree.
+	 *
+	 */
+	struct {
+		struct rb_node rb;
+		unsigned long rb_subtree_last;
+	} shared;
+#ifdef CONFIG_ANON_VMA_NAME
+	/*
+	 * For private and shared anonymous mappings, a pointer to a null
+	 * terminated string containing the name given to the vma, or NULL if
+	 * unnamed. Serialized by mmap_lock. Use anon_vma_name to access.
+	 */
+	struct anon_vma_name *anon_name;
+#endif
+	struct vm_userfaultfd_ctx vm_userfaultfd_ctx;
 } __randomize_layout;
 
 #ifdef CONFIG_NUMA
diff --git a/tools/testing/vma/vma_internal.h b/tools/testing/vma/vma_internal.h
index ba838097d3f6..b385170fbb8f 100644
--- a/tools/testing/vma/vma_internal.h
+++ b/tools/testing/vma/vma_internal.h
@@ -279,16 +279,6 @@ struct vm_area_struct {
 	unsigned int vm_lock_seq;
 #endif
 
-	/*
-	 * For areas with an address space and backing store,
-	 * linkage into the address_space->i_mmap interval tree.
-	 *
-	 */
-	struct {
-		struct rb_node rb;
-		unsigned long rb_subtree_last;
-	} shared;
-
 	/*
 	 * A file's MAP_PRIVATE vma can be in both i_mmap tree and anon_vma
 	 * list, after a COW of one of the file pages.	A MAP_SHARED vma
@@ -308,14 +298,6 @@ struct vm_area_struct {
 	struct file * vm_file;		/* File we map to (can be NULL). */
 	void * vm_private_data;		/* was vm_pte (shared mem) */
 
-#ifdef CONFIG_ANON_VMA_NAME
-	/*
-	 * For private and shared anonymous mappings, a pointer to a null
-	 * terminated string containing the name given to the vma, or NULL if
-	 * unnamed. Serialized by mmap_lock. Use anon_vma_name to access.
-	 */
-	struct anon_vma_name *anon_name;
-#endif
 #ifdef CONFIG_SWAP
 	atomic_long_t swap_readahead_info;
 #endif
@@ -328,11 +310,28 @@ struct vm_area_struct {
 #ifdef CONFIG_NUMA_BALANCING
 	struct vma_numab_state *numab_state;	/* NUMA Balancing state */
 #endif
-	struct vm_userfaultfd_ctx vm_userfaultfd_ctx;
 #ifdef CONFIG_PER_VMA_LOCK
 	/* Unstable RCU readers are allowed to read this. */
 	refcount_t vm_refcnt;
 #endif
+	/*
+	 * For areas with an address space and backing store,
+	 * linkage into the address_space->i_mmap interval tree.
+	 *
+	 */
+	struct {
+		struct rb_node rb;
+		unsigned long rb_subtree_last;
+	} shared;
+#ifdef CONFIG_ANON_VMA_NAME
+	/*
+	 * For private and shared anonymous mappings, a pointer to a null
+	 * terminated string containing the name given to the vma, or NULL if
+	 * unnamed. Serialized by mmap_lock. Use anon_vma_name to access.
+	 */
+	struct anon_vma_name *anon_name;
+#endif
+	struct vm_userfaultfd_ctx vm_userfaultfd_ctx;
 } __randomize_layout;
 
 struct vm_fault {};
-- 
2.48.1.601.g30ceb7b040-goog


  parent reply	other threads:[~2025-02-13 22:47 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-02-13 22:46 [PATCH v10 00/18] reimplement per-vma lock as a refcount Suren Baghdasaryan
2025-02-13 22:46 ` [PATCH v10 01/18] mm: introduce vma_start_read_locked{_nested} helpers Suren Baghdasaryan
2025-02-13 22:46 ` [PATCH v10 02/18] mm: move per-vma lock into vm_area_struct Suren Baghdasaryan
2025-02-13 22:46 ` [PATCH v10 03/18] mm: mark vma as detached until it's added into vma tree Suren Baghdasaryan
2025-02-13 22:46 ` [PATCH v10 04/18] mm: introduce vma_iter_store_attached() to use with attached vmas Suren Baghdasaryan
2025-02-21 11:07   ` Vlastimil Babka
2025-02-21 16:11   ` Liam R. Howlett
2025-02-13 22:46 ` [PATCH v10 05/18] mm: mark vmas detached upon exit Suren Baghdasaryan
2025-02-21 16:14   ` Liam R. Howlett
2025-02-13 22:46 ` [PATCH v10 06/18] types: move struct rcuwait into types.h Suren Baghdasaryan
2025-02-13 22:46 ` [PATCH v10 07/18] mm: allow vma_start_read_locked/vma_start_read_locked_nested to fail Suren Baghdasaryan
2025-02-21 11:27   ` Vlastimil Babka
2025-02-13 22:46 ` [PATCH v10 08/18] mm: move mmap_init_lock() out of the header file Suren Baghdasaryan
2025-02-13 22:46 ` [PATCH v10 09/18] mm: uninline the main body of vma_start_write() Suren Baghdasaryan
2025-02-13 22:46 ` [PATCH v10 10/18] refcount: provide ops for cases when object's memory can be reused Suren Baghdasaryan
2025-02-13 22:46 ` [PATCH v10 11/18] refcount: introduce __refcount_{add|inc}_not_zero_limited_acquire Suren Baghdasaryan
2025-02-13 22:46 ` [PATCH v10 12/18] mm: replace vm_lock and detached flag with a reference count Suren Baghdasaryan
2025-02-20 18:53   ` Heiko Carstens
2025-02-20 19:03     ` Suren Baghdasaryan
2025-02-20 20:05       ` Suren Baghdasaryan
2025-02-21 15:18   ` Vlastimil Babka
2025-02-13 22:46 ` Suren Baghdasaryan [this message]
2025-02-21 15:20   ` [PATCH v10 13/18] mm: move lesser used vma_area_struct members into the last cacheline Vlastimil Babka
2025-02-13 22:46 ` [PATCH v10 14/18] mm/debug: print vm_refcnt state when dumping the vma Suren Baghdasaryan
2025-02-13 22:46 ` [PATCH v10 15/18] mm: remove extra vma_numab_state_init() call Suren Baghdasaryan
2025-02-13 22:46 ` [PATCH v10 16/18] mm: prepare lock_vma_under_rcu() for vma reuse possibility Suren Baghdasaryan
2025-02-13 22:46 ` [PATCH v10 17/18] mm: make vma cache SLAB_TYPESAFE_BY_RCU Suren Baghdasaryan
2025-02-13 22:46 ` [PATCH v10 18/18] docs/mm: document latest changes to vm_lock Suren Baghdasaryan
2025-02-21 15:44   ` Vlastimil Babka
2025-02-14 19:34 ` [PATCH v10 00/18] reimplement per-vma lock as a refcount Shivank Garg

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20250213224655.1680278-14-surenb@google.com \
    --to=surenb@google.com \
    --cc=akpm@linux-foundation.org \
    --cc=brauner@kernel.org \
    --cc=corbet@lwn.net \
    --cc=dave@stgolabs.net \
    --cc=david.laight.linux@gmail.com \
    --cc=david@redhat.com \
    --cc=dhowells@redhat.com \
    --cc=hannes@cmpxchg.org \
    --cc=hdanton@sina.com \
    --cc=hughd@google.com \
    --cc=jannh@google.com \
    --cc=kernel-team@android.com \
    --cc=klarasmodin@gmail.com \
    --cc=liam.howlett@oracle.com \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=lokeshgidra@google.com \
    --cc=lorenzo.stoakes@oracle.com \
    --cc=mgorman@techsingularity.net \
    --cc=mhocko@suse.com \
    --cc=minchan@google.com \
    --cc=mjguzik@gmail.com \
    --cc=oleg@redhat.com \
    --cc=oliver.sang@intel.com \
    --cc=pasha.tatashin@soleen.com \
    --cc=paulmck@kernel.org \
    --cc=peterx@redhat.com \
    --cc=peterz@infradead.org \
    --cc=richard.weiyang@gmail.com \
    --cc=shakeel.butt@linux.dev \
    --cc=souravpanda@google.com \
    --cc=vbabka@suse.cz \
    --cc=willy@infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox