lkml.org 
[lkml]   [2022]   [May]   [16]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
SubjectRe: [PATCH] mapletree-vs-khugepaged
Date
Liam Howlett <liam.howlett@oracle.com> writes:

> * Sven Schnelle <svens@linux.ibm.com> [220516 11:37]:
>> Hi Liam,
>>
>> Liam Howlett <liam.howlett@oracle.com> writes:
>>
>> > * Sven Schnelle <svens@linux.ibm.com> [220515 16:02]:
>> >
>> > I tried the above on my qemu s390 with kernel 5.18.0-rc6-next-20220513,
>> > but it runs without issue, return code is 0. Is there something the VM
>> > needs to have for this to trigger?
>>
>> A coworker said the same. Reason for this seems to be that i've run the
>> code in a unittest environment which seems to make a difference. When
>> compiling the code above with gcc on my system it also doesn't crash.
>> So i have to figure out what makes this unittest binary special.
>>
>> >> I've added a few debug statements to the maple tree code:
>> >>
>> >> [ 27.769641] mas_next_entry: offset=14
>> >> [ 27.769642] mas_next_nentry: entry = 0e00000000000000, slots=0000000090249f80, mas->offset=15 count=14
>> >
>> > Where exactly are you printing this?
>>
>> I added a lot of debug statements to the code trying to understand
>> it. I'll attach it to this mail.
>
> Thanks. Can you check to see if that diff you sent was the correct
> file? It appears to be the git stats and not the changes themselves.

No, it wasn't. I did git stash show with -p, which doesn't make sense.
I've attached the correct diff.

>> > If this is the case, then I think any task that has more than 14 VMAs
>> > would have issues. I also use mas_next_entry() in mas_find() which is
>> > used for the mas_for_each() macro/iterator. Can you please enable
>> > CONFIG_DEBUG_VM_MAPLE_TREE ? mmap.c tests the tree after pretty much
>> > any change and will dump useful information if there is an issue -
>> > including the entire tree. See validate_mm_mt() for details.
>> >
>> > You can find CONFIG_DEBUG_VM_MAPLE_TREE in the config:
>> > kernel hacking -> Memory debugging -> Debug VM -> Debug VM maple trees
>>
>> I have both DEBUG_MAPPLE_TREE and DEBUG_VM_MAPLE_TREE enabled, but don't
>> see anything printed.
>
> I suspect that this means the issue is actually in the mmap code and not
> the tree. It may be sensitive to the task-specific layout. Do you have
> randomization on as well? I'm thinking maybe I return NULL because I'm
> asking for the next element after the last VMA and not checking that
> correctly or similar.


> Does ./scripts/faddr2line work for you? What does it say about
> mmap_region+0x19e/0x848 ?

2629 next = mas_next(&mas, ULONG_MAX);
2630 prev = mas_prev(&mas, 0);
2631 if (vm_flags & VM_SPECIAL)
2632 goto cannot_expand;
2633
2634 /* Attempt to expand an old mapping */
2635 /* Check next */
2636 if (next && next->vm_start == end && !vma_policy(next) &&

next above is 0x0e00000000000000 and next->vm_start will trigger the crash.

2637 can_vma_merge_before(next, vm_flags, NULL, file, pgoff+pglen,
2638 NULL_VM_UFFD_CTX, NULL)) {
2639 merge_end = next->vm_end;
2640 vma = next;
2641 vm_pgoff = next->vm_pgoff - pglen;
2642 }
diff --git a/lib/maple_tree.c b/lib/maple_tree.c
index 967631055210..a621d17e872d 100644
--- a/lib/maple_tree.c
+++ b/lib/maple_tree.c
@@ -140,6 +140,7 @@ struct maple_subtree_state {
struct maple_big_node *bn;
};

+extern int mas_debug;
/* Functions */
static inline struct maple_node *mt_alloc_one(gfp_t gfp)
{
@@ -4550,6 +4551,9 @@ static inline void *mas_next_nentry(struct ma_state *mas,
while (mas->offset < count) {
pivot = pivots[mas->offset];
entry = mas_slot(mas, slots, mas->offset);
+ if (mas_debug)
+ pr_err("%s: entry = %px, slots=%px, mas->offset=%d\n",
+ __func__, entry, slots, mas->offset);
if (ma_dead_node(node))
return NULL;

@@ -4570,6 +4574,10 @@ static inline void *mas_next_nentry(struct ma_state *mas,

pivot = mas_safe_pivot(mas, pivots, mas->offset, type);
entry = mas_slot(mas, slots, mas->offset);
+ if (mas_debug)
+ pr_err("%s: entry = %px, slots=%px, mas->offset=%d count=%d\n",
+ __func__, entry, slots, mas->offset, count);
+
if (ma_dead_node(node))
return NULL;

@@ -4580,6 +4588,8 @@ static inline void *mas_next_nentry(struct ma_state *mas,
return NULL;

found:
+ if (mas_debug)
+ pr_err("found pivot = %lx, entry = %px\n", pivot, entry);
mas->last = pivot;
return entry;
}
@@ -4618,6 +4628,9 @@ static inline void *mas_next_entry(struct ma_state *mas, unsigned long limit)
unsigned long last;
enum maple_type mt;

+ if (mas_debug)
+ pr_err("%s: entry\n", __func__);
+
last = mas->last;
retry:
offset = mas->offset;
@@ -4625,10 +4638,17 @@ static inline void *mas_next_entry(struct ma_state *mas, unsigned long limit)
node = mas_mn(mas);
mt = mte_node_type(mas->node);
mas->offset++;
- if (unlikely(mas->offset >= mt_slots[mt]))
+ if (mas_debug)
+ pr_err("%s: offset=%d\n", __func__, offset);
+ if (unlikely(mas->offset >= mt_slots[mt])) {
+ if (mas_debug)
+ pr_err("%s: next node\n", __func__);
goto next_node;
+ }

while (!mas_is_none(mas)) {
+ if (mas_debug)
+ pr_err("%s: !none\n", __func__);
entry = mas_next_nentry(mas, node, limit, mt);
if (unlikely(ma_dead_node(node))) {
mas_rewalk(mas, last);
@@ -4656,6 +4676,8 @@ static inline void *mas_next_entry(struct ma_state *mas, unsigned long limit)
mas->index = mas->last = limit;
mas->offset = offset;
mas->node = prev_node;
+ if (mas_debug)
+ pr_err("%s: return NULL, mas->node = %px\n", __func__, prev_node);
return NULL;
}

@@ -4914,6 +4936,8 @@ static inline bool mas_anode_descend(struct ma_state *mas, unsigned long size)
void *mas_walk(struct ma_state *mas)
{
void *entry;
+ if (mas_debug)
+ pr_err("%s\n", __func__);

retry:
entry = mas_state_walk(mas);
@@ -5838,7 +5862,12 @@ EXPORT_SYMBOL_GPL(mas_pause);
*/
void *mas_find(struct ma_state *mas, unsigned long max)
{
+ if (mas_debug)
+ pr_err("%s: max=%lx\n", __func__, max);
+
if (unlikely(mas_is_paused(mas))) {
+ if (mas_debug)
+ pr_err("%s: paused\n", __func__);
if (unlikely(mas->last == ULONG_MAX)) {
mas->node = MAS_NONE;
return NULL;
@@ -5848,6 +5877,8 @@ void *mas_find(struct ma_state *mas, unsigned long max)
}

if (unlikely(mas_is_start(mas))) {
+ if (mas_debug)
+ pr_err("%s: start\n", __func__);
/* First run or continue */
void *entry;

@@ -5859,8 +5890,10 @@ void *mas_find(struct ma_state *mas, unsigned long max)
return entry;
}

- if (unlikely(!mas_searchable(mas)))
+ if (unlikely(!mas_searchable(mas))) {
+ pr_err("%s: not searchable\n", __func__);
return NULL;
+ }

/* Retries on dead nodes handled by mas_next_entry */
return mas_next_entry(mas, max);
diff --git a/mm/mmap.c b/mm/mmap.c
index 2049500931ae..16ee834c0a4c 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -57,6 +57,7 @@

#include "internal.h"

+int mas_debug;
#ifndef arch_mmap_check
#define arch_mmap_check(addr, len, flags) (0)
#endif
@@ -2375,6 +2376,10 @@ do_mas_align_munmap(struct ma_state *mas, struct vm_area_struct *vma,
int count = 0;
int error = -ENOMEM;
MA_STATE(mas_detach, &mt_detach, start, end - 1);
+
+ if (mas_debug)
+ pr_err("%s:%d\n", __func__, __LINE__);
+
mt_init_flags(&mt_detach, MM_MT_FLAGS);
mt_set_external_lock(&mt_detach, &mm->mmap_lock);

@@ -2541,28 +2546,37 @@ do_mas_align_munmap(struct ma_state *mas, struct vm_area_struct *vma,
*
* Returns: -EINVAL on failure, 1 on success and unlock, 0 otherwise.
*/
+
+
+
int do_mas_munmap(struct ma_state *mas, struct mm_struct *mm,
unsigned long start, size_t len, struct list_head *uf,
bool downgrade)
{
unsigned long end;
struct vm_area_struct *vma;
+ if (mas_debug)
+ pr_err("%s: %lx %lx\n", __func__, start, len);

if ((offset_in_page(start)) || start > TASK_SIZE || len > TASK_SIZE-start)
return -EINVAL;
-
+ if (mas_debug)
+ pr_err("%s:%d\n", __func__, __LINE__);
end = start + PAGE_ALIGN(len);
if (end == start)
return -EINVAL;
-
+ if (mas_debug)
+ pr_err("%s:%d\n", __func__, __LINE__);
/* arch_unmap() might do unmaps itself. */
arch_unmap(mm, start, end);
-
+ if (mas_debug)
+ pr_err("%s:%d\n", __func__, __LINE__);
/* Find the first overlapping VMA */
vma = mas_find(mas, end - 1);
if (!vma)
return 0;
-
+ if (mas_debug)
+ pr_err("vma=%px\n", vma);
return do_mas_align_munmap(mas, vma, mm, start, end, uf, downgrade);
}

@@ -2594,6 +2608,7 @@ unsigned long mmap_region(struct file *file, unsigned long addr,
pgoff_t vm_pgoff;
int error;
MA_STATE(mas, &mm->mm_mt, addr, end - 1);
+ mas_debug = (addr == (1UL << 54));

/* Check against address space limit. */
if (!may_expand_vm(mm, vm_flags, len >> PAGE_SHIFT)) {
@@ -2609,23 +2624,30 @@ unsigned long mmap_region(struct file *file, unsigned long addr,
(len >> PAGE_SHIFT) - nr_pages))
return -ENOMEM;
}
-
/* Unmap any existing mapping in the area */
- if (do_mas_munmap(&mas, mm, addr, len, uf, false))
+
+ if (do_mas_munmap(&mas, mm, addr, len, uf, false)) {
+ mas_debug = 0;
return -ENOMEM;
+ }

/*
* Private writable mapping: check memory availability
*/
if (accountable_mapping(file, vm_flags)) {
charged = len >> PAGE_SHIFT;
- if (security_vm_enough_memory_mm(mm, charged))
+ if (security_vm_enough_memory_mm(mm, charged)) {
+ mas_debug = 0;
return -ENOMEM;
+ }
vm_flags |= VM_ACCOUNT;
}

next = mas_next(&mas, ULONG_MAX);
prev = mas_prev(&mas, 0);
+ if (mas_debug)
+ pr_err("%s: next %px\n", __func__, next);
+ mas_debug = 0;
if (vm_flags & VM_SPECIAL)
goto cannot_expand;
\
 
 \ /
  Last update: 2022-05-16 19:12    [W:0.091 / U:1.080 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site