The patch titled
Subject: mm/vmalloc: eliminated the lock contention from twice to once
has been added to the -mm mm-unstable branch. Its filename is
mm-vmalloc-eliminated-the-lock-contention-from-twice-to-once.patch
This patch will shortly appear at
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-vmalloc-eliminated-the-lock-contention-from-twice-to-once.patch
This patch will later appear in the mm-unstable branch at
git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next via the mm-everything
branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
and is updated there every 2-3 working days
------------------------------------------------------
From: rulinhuang <rulin.huang@xxxxxxxxx>
Subject: mm/vmalloc: eliminated the lock contention from twice to once
Date: Wed, 6 Mar 2024 21:14:40 -0500
When allocating a new memory area where the mapping address range is
known, it is observed that the vmap_node->busy.lock is acquired twice.
The first acquisition occurs in the alloc_vmap_area() function when
inserting the vm area into the vm mapping red-black tree. The second
acquisition occurs in the setup_vmalloc_vm() function when updating the
properties of the vm, such as flags and address, etc.
Combine these two operations together in alloc_vmap_area(), which improves
scalability when the vmap_node->busy.lock is contended. By doing so, the
need to acquire the lock twice can also be eliminated to once.
With the above change, tested on intel sapphire rapids platform(224 vcpu),
a 4% performance improvement is gained on
stress-ng/pthread(https://github.com/ColinIanKing/stress-ng), which is the
stress test of thread creations.
Link: https://lkml.kernel.org/r/20240307021440.64967-1-rulin.huang@xxxxxxxxx
Co-developed-by: "Chen, Tim C" <tim.c.chen@xxxxxxxxx>
Signed-off-by: "Chen, Tim C" <tim.c.chen@xxxxxxxxx>
Co-developed-by: "King, Colin" <colin.king@xxxxxxxxx>
Signed-off-by: "King, Colin" <colin.king@xxxxxxxxx>
Signed-off-by: rulinhuang <rulin.huang@xxxxxxxxx>
Reviewed-by: Baoquan He <bhe@xxxxxxxxxx>
Reviewed-by: Uladzislau Rezki (Sony) <urezki@xxxxxxxxx>
Cc: Tim Chen <tim.c.chen@xxxxxxxxxxxxxxx>
Cc: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Cc: Lorenzo Stoakes <lstoakes@xxxxxxxxx>
Cc: Wangyang Guo <wangyang.guo@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/vmalloc.c | 50 +++++++++++++++++++++----------------------------
1 file changed, 22 insertions(+), 28 deletions(-)
--- a/mm/vmalloc.c~mm-vmalloc-eliminated-the-lock-contention-from-twice-to-once
+++ a/mm/vmalloc.c
@@ -1913,15 +1913,26 @@ node_alloc(unsigned long size, unsigned
return va;
}
+static inline void setup_vmalloc_vm(struct vm_struct *vm,
+ struct vmap_area *va, unsigned long flags, const void *caller)
+{
+ vm->flags = flags;
+ vm->addr = (void *)va->va_start;
+ vm->size = va->va_end - va->va_start;
+ vm->caller = caller;
+ va->vm = vm;
+}
+
/*
* Allocate a region of KVA of the specified size and alignment, within the
- * vstart and vend.
+ * vstart and vend. If vm is passed in, the two will also be bound.
*/
static struct vmap_area *alloc_vmap_area(unsigned long size,
unsigned long align,
unsigned long vstart, unsigned long vend,
int node, gfp_t gfp_mask,
- unsigned long va_flags)
+ unsigned long va_flags, struct vm_struct *vm,
+ unsigned long flags, const void *caller)
{
struct vmap_node *vn;
struct vmap_area *va;
@@ -1984,6 +1995,9 @@ retry:
va->vm = NULL;
va->flags = (va_flags | vn_id);
+ if (vm)
+ setup_vmalloc_vm(vm, va, flags, caller);
+
vn = addr_to_node(va->va_start);
spin_lock(&vn->busy.lock);
@@ -2558,7 +2572,8 @@ static void *new_vmap_block(unsigned int
va = alloc_vmap_area(VMAP_BLOCK_SIZE, VMAP_BLOCK_SIZE,
VMALLOC_START, VMALLOC_END,
node, gfp_mask,
- VMAP_RAM|VMAP_BLOCK);
+ VMAP_RAM|VMAP_BLOCK, NULL,
+ 0, NULL);
if (IS_ERR(va)) {
kfree(vb);
return ERR_CAST(va);
@@ -2915,7 +2930,8 @@ void *vm_map_ram(struct page **pages, un
struct vmap_area *va;
va = alloc_vmap_area(size, PAGE_SIZE,
VMALLOC_START, VMALLOC_END,
- node, GFP_KERNEL, VMAP_RAM);
+ node, GFP_KERNEL, VMAP_RAM,
+ NULL, 0, NULL);
if (IS_ERR(va))
return NULL;
@@ -3018,26 +3034,6 @@ void __init vm_area_register_early(struc
kasan_populate_early_vm_area_shadow(vm->addr, vm->size);
}
-static inline void setup_vmalloc_vm_locked(struct vm_struct *vm,
- struct vmap_area *va, unsigned long flags, const void *caller)
-{
- vm->flags = flags;
- vm->addr = (void *)va->va_start;
- vm->size = va->va_end - va->va_start;
- vm->caller = caller;
- va->vm = vm;
-}
-
-static void setup_vmalloc_vm(struct vm_struct *vm, struct vmap_area *va,
- unsigned long flags, const void *caller)
-{
- struct vmap_node *vn = addr_to_node(va->va_start);
-
- spin_lock(&vn->busy.lock);
- setup_vmalloc_vm_locked(vm, va, flags, caller);
- spin_unlock(&vn->busy.lock);
-}
-
static void clear_vm_uninitialized_flag(struct vm_struct *vm)
{
/*
@@ -3074,14 +3070,12 @@ static struct vm_struct *__get_vm_area_n
if (!(flags & VM_NO_GUARD))
size += PAGE_SIZE;
- va = alloc_vmap_area(size, align, start, end, node, gfp_mask, 0);
+ va = alloc_vmap_area(size, align, start, end, node, gfp_mask, 0, area, flags, caller);
if (IS_ERR(va)) {
kfree(area);
return NULL;
}
- setup_vmalloc_vm(area, va, flags, caller);
-
/*
* Mark pages for non-VM_ALLOC mappings as accessible. Do it now as a
* best-effort approach, as they can be mapped outside of vmalloc code.
@@ -4656,7 +4650,7 @@ retry:
spin_lock(&vn->busy.lock);
insert_vmap_area(vas[area], &vn->busy.root, &vn->busy.head);
- setup_vmalloc_vm_locked(vms[area], vas[area], VM_ALLOC,
+ setup_vmalloc_vm(vms[area], vas[area], VM_ALLOC,
pcpu_get_vm_areas);
spin_unlock(&vn->busy.lock);
}
_
Patches currently in -mm which might be from rulin.huang@xxxxxxxxx are
mm-vmalloc-eliminated-the-lock-contention-from-twice-to-once.patch
