lkml.org 
[lkml]   [2018]   [Mar]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patches in this message
/
From
SubjectRe: [PATCH v9 04/24] mm: Prepare for FAULT_FLAG_SPECULATIVE
Date
On 25/03/2018 23:50, David Rientjes wrote:
> On Tue, 13 Mar 2018, Laurent Dufour wrote:
>
>> From: Peter Zijlstra <peterz@infradead.org>
>>
>> When speculating faults (without holding mmap_sem) we need to validate
>> that the vma against which we loaded pages is still valid when we're
>> ready to install the new PTE.
>>
>> Therefore, replace the pte_offset_map_lock() calls that (re)take the
>> PTL with pte_map_lock() which can fail in case we find the VMA changed
>> since we started the fault.
>>
>
> Based on how its used, I would have suspected this to be named
> pte_map_trylock().
>
>> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
>>
>> [Port to 4.12 kernel]
>> [Remove the comment about the fault_env structure which has been
>> implemented as the vm_fault structure in the kernel]
>> [move pte_map_lock()'s definition upper in the file]
>> Signed-off-by: Laurent Dufour <ldufour@linux.vnet.ibm.com>
>> ---
>> include/linux/mm.h | 1 +
>> mm/memory.c | 56 ++++++++++++++++++++++++++++++++++++++----------------
>> 2 files changed, 41 insertions(+), 16 deletions(-)
>>
>> diff --git a/include/linux/mm.h b/include/linux/mm.h
>> index 4d02524a7998..2f3e98edc94a 100644
>> --- a/include/linux/mm.h
>> +++ b/include/linux/mm.h
>> @@ -300,6 +300,7 @@ extern pgprot_t protection_map[16];
>> #define FAULT_FLAG_USER 0x40 /* The fault originated in userspace */
>> #define FAULT_FLAG_REMOTE 0x80 /* faulting for non current tsk/mm */
>> #define FAULT_FLAG_INSTRUCTION 0x100 /* The fault was during an instruction fetch */
>> +#define FAULT_FLAG_SPECULATIVE 0x200 /* Speculative fault, not holding mmap_sem */
>>
>> #define FAULT_FLAG_TRACE \
>> { FAULT_FLAG_WRITE, "WRITE" }, \
>
> I think FAULT_FLAG_SPECULATIVE should be introduced in the patch that
> actually uses it.

I think you're right, I'll move down this define in the series.

>> diff --git a/mm/memory.c b/mm/memory.c
>> index e0ae4999c824..8ac241b9f370 100644
>> --- a/mm/memory.c
>> +++ b/mm/memory.c
>> @@ -2288,6 +2288,13 @@ int apply_to_page_range(struct mm_struct *mm, unsigned long addr,
>> }
>> EXPORT_SYMBOL_GPL(apply_to_page_range);
>>
>> +static bool pte_map_lock(struct vm_fault *vmf)
>
> inline?

Agreed.

>> +{
>> + vmf->pte = pte_offset_map_lock(vmf->vma->vm_mm, vmf->pmd,
>> + vmf->address, &vmf->ptl);
>> + return true;
>> +}
>> +
>> /*
>> * handle_pte_fault chooses page fault handler according to an entry which was
>> * read non-atomically. Before making any commitment, on those architectures
>> @@ -2477,6 +2484,7 @@ static int wp_page_copy(struct vm_fault *vmf)
>> const unsigned long mmun_start = vmf->address & PAGE_MASK;
>> const unsigned long mmun_end = mmun_start + PAGE_SIZE;
>> struct mem_cgroup *memcg;
>> + int ret = VM_FAULT_OOM;
>>
>> if (unlikely(anon_vma_prepare(vma)))
>> goto oom;
>> @@ -2504,7 +2512,11 @@ static int wp_page_copy(struct vm_fault *vmf)
>> /*
>> * Re-check the pte - we dropped the lock
>> */
>> - vmf->pte = pte_offset_map_lock(mm, vmf->pmd, vmf->address, &vmf->ptl);
>> + if (!pte_map_lock(vmf)) {
>> + mem_cgroup_cancel_charge(new_page, memcg, false);
>> + ret = VM_FAULT_RETRY;
>> + goto oom_free_new;
>> + }
>
> Ugh, but we aren't oom here, so maybe rename oom_free_new so that it makes
> sense for return values other than VM_FAULT_OOM?

You're right, now this label name is not correct, I'll rename it to
"out_free_new" and rename also the label "oom" to "out" since it is generic too
now.

>> if (likely(pte_same(*vmf->pte, vmf->orig_pte))) {
>> if (old_page) {
>> if (!PageAnon(old_page)) {
>> @@ -2596,7 +2608,7 @@ static int wp_page_copy(struct vm_fault *vmf)
>> oom:
>> if (old_page)
>> put_page(old_page);
>> - return VM_FAULT_OOM;
>> + return ret;
>> }
>>
>> /**
>> @@ -2617,8 +2629,8 @@ static int wp_page_copy(struct vm_fault *vmf)
>> int finish_mkwrite_fault(struct vm_fault *vmf)
>> {
>> WARN_ON_ONCE(!(vmf->vma->vm_flags & VM_SHARED));
>> - vmf->pte = pte_offset_map_lock(vmf->vma->vm_mm, vmf->pmd, vmf->address,
>> - &vmf->ptl);
>> + if (!pte_map_lock(vmf))
>> + return VM_FAULT_RETRY;
>> /*
>> * We might have raced with another page fault while we released the
>> * pte_offset_map_lock.
>> @@ -2736,8 +2748,11 @@ static int do_wp_page(struct vm_fault *vmf)
>> get_page(vmf->page);
>> pte_unmap_unlock(vmf->pte, vmf->ptl);
>> lock_page(vmf->page);
>> - vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd,
>> - vmf->address, &vmf->ptl);
>> + if (!pte_map_lock(vmf)) {
>> + unlock_page(vmf->page);
>> + put_page(vmf->page);
>> + return VM_FAULT_RETRY;
>> + }
>> if (!pte_same(*vmf->pte, vmf->orig_pte)) {
>> unlock_page(vmf->page);
>> pte_unmap_unlock(vmf->pte, vmf->ptl);
>> @@ -2947,8 +2962,10 @@ int do_swap_page(struct vm_fault *vmf)
>> * Back out if somebody else faulted in this pte
>> * while we released the pte lock.
>> */
>
> Comment needs updating, pte_same() isn't the only reason to bail out here.

I'll update it to :
/*
* Back out if the VMA has changed in our back during
* a speculative page fault or if somebody else
* faulted in this pte while we released the pte lock.
*/

>
>> - vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd,
>> - vmf->address, &vmf->ptl);
>> + if (!pte_map_lock(vmf)) {
>> + delayacct_clear_flag(DELAYACCT_PF_SWAPIN);
>> + return VM_FAULT_RETRY;
>> + }
>> if (likely(pte_same(*vmf->pte, vmf->orig_pte)))
>> ret = VM_FAULT_OOM;
>> delayacct_clear_flag(DELAYACCT_PF_SWAPIN);
>
> Not crucial, but it would be nice if this could do goto out instead,
> otherwise this is the first mid function return.

ok will do.

>
>> @@ -3003,8 +3020,11 @@ int do_swap_page(struct vm_fault *vmf)
>> /*
>> * Back out if somebody else already faulted in this pte.
>> */
>
> Same as above.

Ok changing as above.

>
>> - vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd, vmf->address,
>> - &vmf->ptl);
>> + if (!pte_map_lock(vmf)) {
>> + ret = VM_FAULT_RETRY;
>> + mem_cgroup_cancel_charge(page, memcg, false);
>> + goto out_page;
>> + }
>> if (unlikely(!pte_same(*vmf->pte, vmf->orig_pte)))
>> goto out_nomap;
>>
>
> mem_cgroup_try_charge() is done before grabbing pte_offset_map_lock(), why
> does the out_nomap exit path do mem_cgroup_cancel_charge();
> pte_unmap_unlock()? If the pte lock can be droppde first, there's no need
> to embed the mem_cgroup_cancel_charge() here.

I think we can safely invert the call to mem_cgroup_cancel_charge() and to
pte_unmap_unlock(), and then introduce a new label and jump in if
pte_map_lock() failed.
Something like this:

@@ -3001,10 +3020,13 @@ int do_swap_page(struct vm_fault *vmf)
}

/*
- * Back out if somebody else already faulted in this pte.
+ * Back out if the VMA has changed in our back during a speculative
+ * page fault or if somebody else already faulted in this pte.
*/
- vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd, vmf->address,
- &vmf->ptl);
+ if (!pte_map_lock(vmf)) {
+ ret = VM_FAULT_RETRY;
+ goto out_cancel_cgroup;
+ }
if (unlikely(!pte_same(*vmf->pte, vmf->orig_pte)))
goto out_nomap;

@@ -3082,8 +3104,9 @@ int do_swap_page(struct vm_fault *vmf)
out:
return ret;
out_nomap:
- mem_cgroup_cancel_charge(page, memcg, false);
pte_unmap_unlock(vmf->pte, vmf->ptl);
+out_cancel_cgroup:
+ mem_cgroup_cancel_charge(page, memcg, false);
out_page:
unlock_page(page);
out_release:


>> @@ -3133,8 +3153,8 @@ static int do_anonymous_page(struct vm_fault *vmf)
>> !mm_forbids_zeropage(vma->vm_mm)) {
>> entry = pte_mkspecial(pfn_pte(my_zero_pfn(vmf->address),
>> vma->vm_page_prot));
>> - vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd,
>> - vmf->address, &vmf->ptl);
>> + if (!pte_map_lock(vmf))
>> + return VM_FAULT_RETRY;
>> if (!pte_none(*vmf->pte))
>> goto unlock;
>> ret = check_stable_address_space(vma->vm_mm);
>> @@ -3169,8 +3189,11 @@ static int do_anonymous_page(struct vm_fault *vmf)
>> if (vma->vm_flags & VM_WRITE)
>> entry = pte_mkwrite(pte_mkdirty(entry));
>>
>> - vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd, vmf->address,
>> - &vmf->ptl);
>> + if (!pte_map_lock(vmf)) {
>> + mem_cgroup_cancel_charge(page, memcg, false);
>> + put_page(page);
>> + return VM_FAULT_RETRY;
>> + }
>> if (!pte_none(*vmf->pte))
>> goto release;
>>
>
> This is more spaghetti, can the exit path be fixed up so we order things
> consistently for all gotos?

I do agree, this was due to inverted calls to mem_cgroup_cancel_charge() and
pte_unmap_unlock().

This will become:
@@ -3170,14 +3193,16 @@ static int do_anonymous_page(struct vm_fault *vmf)
if (vma->vm_flags & VM_WRITE)
entry = pte_mkwrite(pte_mkdirty(entry));
- vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd, vmf->address,
- &vmf->ptl);
- if (!pte_none(*vmf->pte))
+ if (!pte_map_lock(vmf)) {
+ ret = VM_FAULT_RETRY;
goto release;
+ }
+ if (!pte_none(*vmf->pte))
+ goto unlock_and_release;
ret = check_stable_address_space(vma->vm_mm);
if (ret)
- goto release;
+ goto unlock_and_release;
/* Deliver the page fault to userland, check inside PT lock */
if (userfaultfd_missing(vma)) {
@@ -3199,10 +3224,12 @@ static int do_anonymous_page(struct vm_fault *vmf)
unlock:
pte_unmap_unlock(vmf->pte, vmf->ptl);
return ret;
+unlock_and_release:
+ pte_unmap_unlock(vmf->pte, vmf->ptl);
release:
mem_cgroup_cancel_charge(page, memcg, false);
put_page(page);
- goto unlock;
+ return ret;
oom_free_page:
put_page(page);
oom:
Thanks,
Laurent.

>
>> @@ -3294,8 +3317,9 @@ static int pte_alloc_one_map(struct vm_fault *vmf)
>> * pte_none() under vmf->ptl protection when we return to
>> * alloc_set_pte().
>> */
>> - vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd, vmf->address,
>> - &vmf->ptl);
>> + if (!pte_map_lock(vmf))
>> + return VM_FAULT_RETRY;
>> +
>> return 0;
>> }
>>
>

\
 
 \ /
  Last update: 2018-03-28 12:28    [W:0.250 / U:2.400 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site