2020-06-26 22:32:41

by Peter Xu

Subject: [PATCH 12/26] mm/nds32: Use general page fault accounting

Use the general page fault accounting by passing regs into handle_mm_fault().
It naturally solve the issue of multiple page fault accounting when page fault
retry happened.

Fix PERF_COUNT_SW_PAGE_FAULTS perf event manually for page fault retries, by
moving it before taking mmap_sem.

CC: Nick Hu <[email protected]>
CC: Greentime Hu <[email protected]>
CC: Vincent Chen <[email protected]>
Signed-off-by: Peter Xu <[email protected]>
---
arch/nds32/mm/fault.c | 19 +++----------------
1 file changed, 3 insertions(+), 16 deletions(-)

diff --git a/arch/nds32/mm/fault.c b/arch/nds32/mm/fault.c
index 22527129025c..e7344440623c 100644
--- a/arch/nds32/mm/fault.c
+++ b/arch/nds32/mm/fault.c
@@ -122,6 +122,8 @@ void do_page_fault(unsigned long entry, unsigned long addr,
if (unlikely(faulthandler_disabled() || !mm))
goto no_context;

+ perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, addr);
+
/*
* As per x86, we may deadlock here. However, since the kernel only
* validly references user space from well defined areas of the code,
@@ -207,7 +209,7 @@ void do_page_fault(unsigned long entry, unsigned long addr,
* the fault.
*/

- fault = handle_mm_fault(vma, addr, flags, NULL);
+ fault = handle_mm_fault(vma, addr, flags, regs);

/*
* If we need to retry but a fatal signal is pending, handle the
@@ -229,22 +231,7 @@ void do_page_fault(unsigned long entry, unsigned long addr,
goto bad_area;
}

- /*
- * Major/minor page fault accounting is only done on the initial
- * attempt. If we go through a retry, it is extremely likely that the
- * page will be found in page cache at that point.
- */
- perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, addr);
if (flags & FAULT_FLAG_ALLOW_RETRY) {
- if (fault & VM_FAULT_MAJOR) {
- tsk->maj_flt++;
- perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ,
- 1, regs, addr);
- } else {
- tsk->min_flt++;
- perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN,
- 1, regs, addr);
- }
if (fault & VM_FAULT_RETRY) {
flags |= FAULT_FLAG_TRIED;

--
2.26.2

2020-06-27 03:02:55

by Greentime Hu

Subject: Re: [PATCH 12/26] mm/nds32: Use general page fault accounting

Peter Xu <[email protected]> 於 2020年6月27日 週六 上午6:31寫道：
>
> Use the general page fault accounting by passing regs into handle_mm_fault().
> It naturally solve the issue of multiple page fault accounting when page fault
> retry happened.
>
> Fix PERF_COUNT_SW_PAGE_FAULTS perf event manually for page fault retries, by
> moving it before taking mmap_sem.
>
> CC: Nick Hu <[email protected]>
> CC: Greentime Hu <[email protected]>
> CC: Vincent Chen <[email protected]>
> Signed-off-by: Peter Xu <[email protected]>
> ---
> arch/nds32/mm/fault.c | 19 +++----------------
> 1 file changed, 3 insertions(+), 16 deletions(-)
>
> diff --git a/arch/nds32/mm/fault.c b/arch/nds32/mm/fault.c
> index 22527129025c..e7344440623c 100644
> --- a/arch/nds32/mm/fault.c
> +++ b/arch/nds32/mm/fault.c
> @@ -122,6 +122,8 @@ void do_page_fault(unsigned long entry, unsigned long addr,
> if (unlikely(faulthandler_disabled() || !mm))
> goto no_context;
>
> + perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, addr);
> +
> /*
> * As per x86, we may deadlock here. However, since the kernel only
> * validly references user space from well defined areas of the code,
> @@ -207,7 +209,7 @@ void do_page_fault(unsigned long entry, unsigned long addr,
> * the fault.
> */
>
> - fault = handle_mm_fault(vma, addr, flags, NULL);
> + fault = handle_mm_fault(vma, addr, flags, regs);
>
> /*
> * If we need to retry but a fatal signal is pending, handle the
> @@ -229,22 +231,7 @@ void do_page_fault(unsigned long entry, unsigned long addr,
> goto bad_area;
> }
>
> - /*
> - * Major/minor page fault accounting is only done on the initial
> - * attempt. If we go through a retry, it is extremely likely that the
> - * page will be found in page cache at that point.
> - */
> - perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, addr);
> if (flags & FAULT_FLAG_ALLOW_RETRY) {
> - if (fault & VM_FAULT_MAJOR) {
> - tsk->maj_flt++;
> - perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ,
> - 1, regs, addr);
> - } else {
> - tsk->min_flt++;
> - perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN,
> - 1, regs, addr);
> - }
> if (fault & VM_FAULT_RETRY) {
> flags |= FAULT_FLAG_TRIED;
>

Hi Peter,

Thank you.
Acked-by: Greentime Hu <[email protected]>