The patch titled
Subject: mm: rework virtual memory accounting
has been added to the -mm tree. Its filename is
mm-rework-virtual-memory-accounting.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/mm-rework-virtual-memory-accounting.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/mm-rework-virtual-memory-accounting.patch
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/SubmitChecklist when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Konstantin Khlebnikov <koct9i@xxxxxxxxx>
Subject: mm: rework virtual memory accounting
When inspecting a vague code inside prctl(PR_SET_MM_MEM) call (which
testing the RLIMIT_DATA value to figure out if we're allowed to assign new
@start_brk, @brk, @start_data, @end_data from mm_struct) it's been
commited that RLIMIT_DATA in a form it's implemented now doesn't do
anything useful because most of user-space libraries use mmap() syscall
for dynamic memory allocations.
Linus suggested to convert RLIMIT_DATA rlimit into something suitable for
anonymous memory accounting. But in this patch we go further, and the
changes are bundled together as:
* keep vma counting if CONFIG_PROC_FS=n, will be used for limits
* replace mm->shared_vm with better defined mm->data_vm
* account anonymous executable areas as executable
* account file-backed growsdown/up areas as stack
* drop struct file* argument from vm_stat_account
* enforce RLIMIT_DATA for size of data areas
This way code looks cleaner: now code/stack/data classification depends
only on vm_flags state:
VM_EXEC & ~VM_WRITE -> code (VmExe + VmLib in proc)
VM_GROWSUP | VM_GROWSDOWN -> stack (VmStk)
VM_WRITE & ~VM_SHARED & !stack -> data (VmData)
The rest (VmSize - VmData - VmStk - VmExe - VmLib) could be called
"shared", but that might be strange beast like readonly-private or VM_IO
area.
- RLIMIT_AS limits whole address space "VmSize"
- RLIMIT_STACK limits stack "VmStk" (but each vma individually)
- RLIMIT_DATA now limits "VmData"
Signed-off-by: Konstantin Khlebnikov <koct9i@xxxxxxxxx>
Signed-off-by: Cyrill Gorcunov <gorcunov@xxxxxxxxxx>
Cc: Quentin Casasnovas <quentin.casasnovas@xxxxxxxxxx>
Cc: Vegard Nossum <vegard.nossum@xxxxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: Willy Tarreau <w@xxxxxx>
Cc: Andy Lutomirski <luto@xxxxxxxxxxxxxx>
Cc: Kees Cook <keescook@xxxxxxxxxx>
Cc: Vladimir Davydov <vdavydov@xxxxxxxxxxxxx>
Cc: Pavel Emelyanov <xemul@xxxxxxxxxxxxx>
Cc: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
arch/ia64/kernel/perfmon.c | 3 -
fs/proc/task_mmu.c | 7 +--
include/linux/mm.h | 13 +------
include/linux/mm_types.h | 2 -
kernel/fork.c | 5 +-
mm/debug.c | 4 +-
mm/mmap.c | 63 +++++++++++++++++------------------
mm/mprotect.c | 8 +++-
mm/mremap.c | 7 ++-
9 files changed, 54 insertions(+), 58 deletions(-)
diff -puN arch/ia64/kernel/perfmon.c~mm-rework-virtual-memory-accounting arch/ia64/kernel/perfmon.c
--- a/arch/ia64/kernel/perfmon.c~mm-rework-virtual-memory-accounting
+++ a/arch/ia64/kernel/perfmon.c
@@ -2332,8 +2332,7 @@ pfm_smpl_buffer_alloc(struct task_struct
*/
insert_vm_struct(mm, vma);
- vm_stat_account(vma->vm_mm, vma->vm_flags, vma->vm_file,
- vma_pages(vma));
+ vm_stat_account(vma->vm_mm, vma->vm_flags, vma_pages(vma));
up_write(&task->mm->mmap_sem);
/*
diff -puN fs/proc/task_mmu.c~mm-rework-virtual-memory-accounting fs/proc/task_mmu.c
--- a/fs/proc/task_mmu.c~mm-rework-virtual-memory-accounting
+++ a/fs/proc/task_mmu.c
@@ -23,7 +23,7 @@
void task_mem(struct seq_file *m, struct mm_struct *mm)
{
- unsigned long data, text, lib, swap, ptes, pmds, anon, file, shmem;
+ unsigned long text, lib, swap, ptes, pmds, anon, file, shmem;
unsigned long hiwater_vm, total_vm, hiwater_rss, total_rss;
anon = get_mm_counter(mm, MM_ANONPAGES);
@@ -44,7 +44,6 @@ void task_mem(struct seq_file *m, struct
if (hiwater_rss < mm->hiwater_rss)
hiwater_rss = mm->hiwater_rss;
- data = mm->total_vm - mm->shared_vm - mm->stack_vm;
text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK)) >> 10;
lib = (mm->exec_vm << (PAGE_SHIFT-10)) - text;
swap = get_mm_counter(mm, MM_SWAPENTS);
@@ -76,7 +75,7 @@ void task_mem(struct seq_file *m, struct
anon << (PAGE_SHIFT-10),
file << (PAGE_SHIFT-10),
shmem << (PAGE_SHIFT-10),
- data << (PAGE_SHIFT-10),
+ mm->data_vm << (PAGE_SHIFT-10),
mm->stack_vm << (PAGE_SHIFT-10), text, lib,
ptes >> 10,
pmds >> 10,
@@ -97,7 +96,7 @@ unsigned long task_statm(struct mm_struc
get_mm_counter(mm, MM_SHMEMPAGES);
*text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK))
>> PAGE_SHIFT;
- *data = mm->total_vm - mm->shared_vm;
+ *data = mm->data_vm + mm->stack_vm;
*resident = *shared + get_mm_counter(mm, MM_ANONPAGES);
return mm->total_vm;
}
diff -puN include/linux/mm.h~mm-rework-virtual-memory-accounting include/linux/mm.h
--- a/include/linux/mm.h~mm-rework-virtual-memory-accounting
+++ a/include/linux/mm.h
@@ -1929,7 +1929,9 @@ extern void mm_drop_all_locks(struct mm_
extern void set_mm_exe_file(struct mm_struct *mm, struct file *new_exe_file);
extern struct file *get_mm_exe_file(struct mm_struct *mm);
-extern int may_expand_vm(struct mm_struct *mm, unsigned long npages);
+extern bool may_expand_vm(struct mm_struct *, vm_flags_t, unsigned long npages);
+extern void vm_stat_account(struct mm_struct *, vm_flags_t, long npages);
+
extern struct vm_area_struct *_install_special_mapping(struct mm_struct *mm,
unsigned long addr, unsigned long len,
unsigned long flags,
@@ -2147,15 +2149,6 @@ typedef int (*pte_fn_t)(pte_t *pte, pgta
extern int apply_to_page_range(struct mm_struct *mm, unsigned long address,
unsigned long size, pte_fn_t fn, void *data);
-#ifdef CONFIG_PROC_FS
-void vm_stat_account(struct mm_struct *, unsigned long, struct file *, long);
-#else
-static inline void vm_stat_account(struct mm_struct *mm,
- unsigned long flags, struct file *file, long pages)
-{
- mm->total_vm += pages;
-}
-#endif /* CONFIG_PROC_FS */
#ifdef CONFIG_DEBUG_PAGEALLOC
extern bool _debug_pagealloc_enabled;
diff -puN include/linux/mm_types.h~mm-rework-virtual-memory-accounting include/linux/mm_types.h
--- a/include/linux/mm_types.h~mm-rework-virtual-memory-accounting
+++ a/include/linux/mm_types.h
@@ -427,7 +427,7 @@ struct mm_struct {
unsigned long total_vm; /* Total pages mapped */
unsigned long locked_vm; /* Pages that have PG_mlocked set */
unsigned long pinned_vm; /* Refcount permanently increased */
- unsigned long shared_vm; /* Shared pages (files) */
+ unsigned long data_vm; /* VM_WRITE & ~VM_SHARED/GROWSDOWN */
unsigned long exec_vm; /* VM_EXEC & ~VM_WRITE */
unsigned long stack_vm; /* VM_GROWSUP/DOWN */
unsigned long def_flags;
diff -puN kernel/fork.c~mm-rework-virtual-memory-accounting kernel/fork.c
--- a/kernel/fork.c~mm-rework-virtual-memory-accounting
+++ a/kernel/fork.c
@@ -413,7 +413,7 @@ static int dup_mmap(struct mm_struct *mm
RCU_INIT_POINTER(mm->exe_file, get_mm_exe_file(oldmm));
mm->total_vm = oldmm->total_vm;
- mm->shared_vm = oldmm->shared_vm;
+ mm->data_vm = oldmm->data_vm;
mm->exec_vm = oldmm->exec_vm;
mm->stack_vm = oldmm->stack_vm;
@@ -432,8 +432,7 @@ static int dup_mmap(struct mm_struct *mm
struct file *file;
if (mpnt->vm_flags & VM_DONTCOPY) {
- vm_stat_account(mm, mpnt->vm_flags, mpnt->vm_file,
- -vma_pages(mpnt));
+ vm_stat_account(mm, mpnt->vm_flags, -vma_pages(mpnt));
continue;
}
charge = 0;
diff -puN mm/debug.c~mm-rework-virtual-memory-accounting mm/debug.c
--- a/mm/debug.c~mm-rework-virtual-memory-accounting
+++ a/mm/debug.c
@@ -175,7 +175,7 @@ void dump_mm(const struct mm_struct *mm)
"mmap_base %lu mmap_legacy_base %lu highest_vm_end %lu\n"
"pgd %p mm_users %d mm_count %d nr_ptes %lu nr_pmds %lu map_count %d\n"
"hiwater_rss %lx hiwater_vm %lx total_vm %lx locked_vm %lx\n"
- "pinned_vm %lx shared_vm %lx exec_vm %lx stack_vm %lx\n"
+ "pinned_vm %lx data_vm %lx exec_vm %lx stack_vm %lx\n"
"start_code %lx end_code %lx start_data %lx end_data %lx\n"
"start_brk %lx brk %lx start_stack %lx\n"
"arg_start %lx arg_end %lx env_start %lx env_end %lx\n"
@@ -209,7 +209,7 @@ void dump_mm(const struct mm_struct *mm)
mm_nr_pmds((struct mm_struct *)mm),
mm->map_count,
mm->hiwater_rss, mm->hiwater_vm, mm->total_vm, mm->locked_vm,
- mm->pinned_vm, mm->shared_vm, mm->exec_vm, mm->stack_vm,
+ mm->pinned_vm, mm->data_vm, mm->exec_vm, mm->stack_vm,
mm->start_code, mm->end_code, mm->start_data, mm->end_data,
mm->start_brk, mm->brk, mm->start_stack,
mm->arg_start, mm->arg_end, mm->env_start, mm->env_end,
diff -puN mm/mmap.c~mm-rework-virtual-memory-accounting mm/mmap.c
--- a/mm/mmap.c~mm-rework-virtual-memory-accounting
+++ a/mm/mmap.c
@@ -1220,24 +1220,6 @@ none:
return NULL;
}
-#ifdef CONFIG_PROC_FS
-void vm_stat_account(struct mm_struct *mm, unsigned long flags,
- struct file *file, long pages)
-{
- const unsigned long stack_flags
- = VM_STACK_FLAGS & (VM_GROWSUP|VM_GROWSDOWN);
-
- mm->total_vm += pages;
-
- if (file) {
- mm->shared_vm += pages;
- if ((flags & (VM_EXEC|VM_WRITE)) == VM_EXEC)
- mm->exec_vm += pages;
- } else if (flags & stack_flags)
- mm->stack_vm += pages;
-}
-#endif /* CONFIG_PROC_FS */
-
/*
* If a hint addr is less than mmap_min_addr change hint to be as
* low as possible but still greater than mmap_min_addr
@@ -1556,7 +1538,7 @@ unsigned long mmap_region(struct file *f
unsigned long charged = 0;
/* Check against address space limit. */
- if (!may_expand_vm(mm, len >> PAGE_SHIFT)) {
+ if (!may_expand_vm(mm, vm_flags, len >> PAGE_SHIFT)) {
unsigned long nr_pages;
/*
@@ -1565,7 +1547,8 @@ unsigned long mmap_region(struct file *f
*/
nr_pages = count_vma_pages_range(mm, addr, addr + len);
- if (!may_expand_vm(mm, (len >> PAGE_SHIFT) - nr_pages))
+ if (!may_expand_vm(mm, vm_flags,
+ (len >> PAGE_SHIFT) - nr_pages))
return -ENOMEM;
}
@@ -1664,7 +1647,7 @@ unsigned long mmap_region(struct file *f
out:
perf_event_mmap(vma);
- vm_stat_account(mm, vm_flags, file, len >> PAGE_SHIFT);
+ vm_stat_account(mm, vm_flags, len >> PAGE_SHIFT);
if (vm_flags & VM_LOCKED) {
if (!((vm_flags & VM_SPECIAL) || is_vm_hugetlb_page(vma) ||
vma == get_gate_vma(current->mm)))
@@ -2111,7 +2094,7 @@ static int acct_stack_growth(struct vm_a
unsigned long new_start, actual_size;
/* address space limit tests */
- if (!may_expand_vm(mm, grow))
+ if (!may_expand_vm(mm, vma->vm_flags, grow))
return -ENOMEM;
/* Stack limit test */
@@ -2208,8 +2191,7 @@ int expand_upwards(struct vm_area_struct
spin_lock(&mm->page_table_lock);
if (vma->vm_flags & VM_LOCKED)
mm->locked_vm += grow;
- vm_stat_account(mm, vma->vm_flags,
- vma->vm_file, grow);
+ vm_stat_account(mm, vma->vm_flags, grow);
anon_vma_interval_tree_pre_update_vma(vma);
vma->vm_end = address;
anon_vma_interval_tree_post_update_vma(vma);
@@ -2284,8 +2266,7 @@ int expand_downwards(struct vm_area_stru
spin_lock(&mm->page_table_lock);
if (vma->vm_flags & VM_LOCKED)
mm->locked_vm += grow;
- vm_stat_account(mm, vma->vm_flags,
- vma->vm_file, grow);
+ vm_stat_account(mm, vma->vm_flags, grow);
anon_vma_interval_tree_pre_update_vma(vma);
vma->vm_start = address;
vma->vm_pgoff -= grow;
@@ -2399,7 +2380,7 @@ static void remove_vma_list(struct mm_st
if (vma->vm_flags & VM_ACCOUNT)
nr_accounted += nrpages;
- vm_stat_account(mm, vma->vm_flags, vma->vm_file, -nrpages);
+ vm_stat_account(mm, vma->vm_flags, -nrpages);
vma = remove_vma(vma);
} while (vma);
vm_unacct_memory(nr_accounted);
@@ -2769,7 +2750,7 @@ static unsigned long do_brk(unsigned lon
}
/* Check against address space limits *after* clearing old maps... */
- if (!may_expand_vm(mm, len >> PAGE_SHIFT))
+ if (!may_expand_vm(mm, flags, len >> PAGE_SHIFT))
return -ENOMEM;
if (mm->map_count > sysctl_max_map_count)
@@ -2804,6 +2785,7 @@ static unsigned long do_brk(unsigned lon
out:
perf_event_mmap(vma);
mm->total_vm += len >> PAGE_SHIFT;
+ mm->data_vm += len >> PAGE_SHIFT;
if (flags & VM_LOCKED)
mm->locked_vm += (len >> PAGE_SHIFT);
vma->vm_flags |= VM_SOFTDIRTY;
@@ -2995,9 +2977,28 @@ out:
* Return true if the calling process may expand its vm space by the passed
* number of pages
*/
-int may_expand_vm(struct mm_struct *mm, unsigned long npages)
+bool may_expand_vm(struct mm_struct *mm, vm_flags_t flags, unsigned long npages)
{
- return mm->total_vm + npages <= rlimit(RLIMIT_AS) >> PAGE_SHIFT;
+ if (mm->total_vm + npages > rlimit(RLIMIT_AS) >> PAGE_SHIFT)
+ return false;
+
+ if ((flags & (VM_WRITE | VM_SHARED | (VM_STACK_FLAGS &
+ (VM_GROWSUP | VM_GROWSDOWN)))) == VM_WRITE)
+ return mm->data_vm + npages <= rlimit(RLIMIT_DATA);
+
+ return true;
+}
+
+void vm_stat_account(struct mm_struct *mm, vm_flags_t flags, long npages)
+{
+ mm->total_vm += npages;
+
+ if ((flags & (VM_EXEC | VM_WRITE)) == VM_EXEC)
+ mm->exec_vm += npages;
+ else if (flags & (VM_STACK_FLAGS & (VM_GROWSUP | VM_GROWSDOWN)))
+ mm->stack_vm += npages;
+ else if ((flags & (VM_WRITE | VM_SHARED)) == VM_WRITE)
+ mm->data_vm += npages;
}
static int special_mapping_fault(struct vm_area_struct *vma,
@@ -3079,7 +3080,7 @@ static struct vm_area_struct *__install_
if (ret)
goto out;
- mm->total_vm += len >> PAGE_SHIFT;
+ vm_stat_account(mm, vma->vm_flags, len >> PAGE_SHIFT);
perf_event_mmap(vma);
diff -puN mm/mprotect.c~mm-rework-virtual-memory-accounting mm/mprotect.c
--- a/mm/mprotect.c~mm-rework-virtual-memory-accounting
+++ a/mm/mprotect.c
@@ -278,6 +278,10 @@ mprotect_fixup(struct vm_area_struct *vm
* even if read-only so there is no need to account for them here
*/
if (newflags & VM_WRITE) {
+ /* Check space limits when area turns into data. */
+ if (!may_expand_vm(mm, newflags, nrpages) &&
+ may_expand_vm(mm, oldflags, nrpages))
+ return -ENOMEM;
if (!(oldflags & (VM_ACCOUNT|VM_WRITE|VM_HUGETLB|
VM_SHARED|VM_NORESERVE))) {
charged = nrpages;
@@ -334,8 +338,8 @@ success:
populate_vma_page_range(vma, start, end, NULL);
}
- vm_stat_account(mm, oldflags, vma->vm_file, -nrpages);
- vm_stat_account(mm, newflags, vma->vm_file, nrpages);
+ vm_stat_account(mm, oldflags, -nrpages);
+ vm_stat_account(mm, newflags, nrpages);
perf_event_mmap(vma);
return 0;
diff -puN mm/mremap.c~mm-rework-virtual-memory-accounting mm/mremap.c
--- a/mm/mremap.c~mm-rework-virtual-memory-accounting
+++ a/mm/mremap.c
@@ -317,7 +317,7 @@ static unsigned long move_vma(struct vm_
* If this were a serious issue, we'd add a flag to do_munmap().
*/
hiwater_vm = mm->hiwater_vm;
- vm_stat_account(mm, vma->vm_flags, vma->vm_file, new_len>>PAGE_SHIFT);
+ vm_stat_account(mm, vma->vm_flags, new_len >> PAGE_SHIFT);
if (do_munmap(mm, old_addr, old_len) < 0) {
/* OOM: unable to split vma, just get accounts right */
@@ -379,7 +379,8 @@ static struct vm_area_struct *vma_to_res
return ERR_PTR(-EAGAIN);
}
- if (!may_expand_vm(mm, (new_len - old_len) >> PAGE_SHIFT))
+ if (!may_expand_vm(mm, vma->vm_flags,
+ (new_len - old_len) >> PAGE_SHIFT))
return ERR_PTR(-ENOMEM);
if (vma->vm_flags & VM_ACCOUNT) {
@@ -541,7 +542,7 @@ SYSCALL_DEFINE5(mremap, unsigned long, a
goto out;
}
- vm_stat_account(mm, vma->vm_flags, vma->vm_file, pages);
+ vm_stat_account(mm, vma->vm_flags, pages);
if (vma->vm_flags & VM_LOCKED) {
mm->locked_vm += pages;
locked = true;
_
Patches currently in -mm which might be from koct9i@xxxxxxxxx are
mm-rework-virtual-memory-accounting.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
