xiaomi-sm8450/android_kernel_xiaomi_sm8450
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

FROMLIST: mm: speculative page fault handler return VMA

Browse Source 
When the speculative page fault handler is returning VM_RETRY, there is a
chance that VMA fetched without grabbing the mmap_sem can be reused by the
legacy page fault handler.  By reusing it, we avoid calling find_vma()
again. To achieve, that we must ensure that the VMA structure will not be
freed in our back. This is done by getting the reference on it (get_vma())
and by assuming that the caller will call the new service
can_reuse_spf_vma() once it has grabbed the mmap_sem.

can_reuse_spf_vma() is first checking that the VMA is still in the RB tree
, and then that the VMA's boundaries matched the passed address and release
the reference on the VMA so that it can be freed if needed.

In the case the VMA is freed, can_reuse_spf_vma() will have returned false
as the VMA is no more in the RB tree.

In the architecture page fault handler, the call to the new service
reuse_spf_or_find_vma() should be made in place of find_vma(), this will
handle the check on the spf_vma and if needed call find_vma().

Signed-off-by: Laurent Dufour <ldufour@linux.vnet.ibm.com>
Change-Id: Ia56dcf807e8bddf6788fd696dd80372db35476f0
Link: https://lore.kernel.org/lkml/1523975611-15978-23-git-send-email-ldufour@linux.vnet.ibm.com/
Bug: 161210518
Signed-off-by: Vinayak Menon <vinmenon@codeaurora.org>
This commit is contained in:
Laurent Dufour
2018-04-17 16:33:28 +02:00
committed by Suren Baghdasaryan
parent 736ae8bde8
commit 99e15a0799
2 changed files with 102 additions and 59 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
include/linux/mm.h
View File

@@ -1770,25 +1770,37 @@ extern int fixup_user_fault(struct mm_struct *mm,
#ifdef CONFIG_SPECULATIVE_PAGE_FAULT
extern int __handle_speculative_fault(struct mm_struct *mm,
unsigned long address,
unsigned int flags);
unsigned int flags,
struct vm_area_struct **vma);
static inline int handle_speculative_fault(struct mm_struct *mm,
unsigned long address,
unsigned int flags)
unsigned int flags,
struct vm_area_struct **vma)
{
/*
* Try speculative page fault for multithreaded user space task only.
*/
if (!(flags & FAULT_FLAG_USER) || atomic_read(&mm->mm_users) == 1)
if (!(flags & FAULT_FLAG_USER) || atomic_read(&mm->mm_users) == 1) {
*vma = NULL;
return VM_FAULT_RETRY;
return __handle_speculative_fault(mm, address, flags);
}
return __handle_speculative_fault(mm, address, flags, vma);
}
extern bool can_reuse_spf_vma(struct vm_area_struct *vma,
unsigned long address);
#else
static inline int handle_speculative_fault(struct mm_struct *mm,
unsigned long address,
unsigned int flags)
unsigned int flags,
struct vm_area_struct **vma)
{
return VM_FAULT_RETRY;
}
static inline bool can_reuse_spf_vma(struct vm_area_struct *vma,
unsigned long address)
{
return false;
}
#endif /* CONFIG_SPECULATIVE_PAGE_FAULT */
void unmap_mapping_pages(struct address_space *mapping,