Commit f397c077 authored by Kautuk Consul's avatar Kautuk Consul Committed by Michal Simek

microblaze: Port OOM changes to do_page_fault

Commit d065bd81
(mm: retry page fault when blocking on disk transfer) and
commit 37b23e05
(x86,mm: make pagefault killable)

The above commits introduced changes into the x86 pagefault handler
for making the page fault handler retryable as well as killable.

These changes reduce the mmap_sem hold time, which is crucial
during OOM killer invocation.

Port these changes to microblaze.
Signed-off-by: default avatarKautuk Consul <consul.kautuk@gmail.com>
parent 59516b07
......@@ -92,6 +92,8 @@ void do_page_fault(struct pt_regs *regs, unsigned long address,
int code = SEGV_MAPERR;
int is_write = error_code & ESR_S;
int fault;
unsigned int flags = FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE |
(is_write ? FAULT_FLAG_WRITE : 0);
regs->ear = address;
regs->esr = error_code;
......@@ -138,6 +140,7 @@ void do_page_fault(struct pt_regs *regs, unsigned long address,
if (kernel_mode(regs) && !search_exception_tables(regs->pc))
goto bad_area_nosemaphore;
retry:
down_read(&mm->mmap_sem);
}
......@@ -210,7 +213,11 @@ void do_page_fault(struct pt_regs *regs, unsigned long address,
* make sure we exit gracefully rather than endlessly redo
* the fault.
*/
fault = handle_mm_fault(mm, vma, address, is_write ? FAULT_FLAG_WRITE : 0);
fault = handle_mm_fault(mm, vma, address, flags);
if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
return;
if (unlikely(fault & VM_FAULT_ERROR)) {
if (fault & VM_FAULT_OOM)
goto out_of_memory;
......@@ -218,11 +225,27 @@ void do_page_fault(struct pt_regs *regs, unsigned long address,
goto do_sigbus;
BUG();
}
if (flags & FAULT_FLAG_ALLOW_RETRY) {
if (unlikely(fault & VM_FAULT_MAJOR))
current->maj_flt++;
else
current->min_flt++;
if (fault & VM_FAULT_RETRY) {
flags &= ~FAULT_FLAG_ALLOW_RETRY;
/*
* No need to up_read(&mm->mmap_sem) as we would
* have already released it in __lock_page_or_retry
* in mm/filemap.c.
*/
goto retry;
}
}
up_read(&mm->mmap_sem);
/*
* keep track of tlb+htab misses that are good addrs but
* just need pte's created via handle_mm_fault()
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment