On Thu, Feb 04, 2021 at 01:02:53PM +0530, Prathu Baronia wrote: > From: Ira Weiny <[email protected]> > > kmap_atomic() is being deprecated in favor of kmap_local_page(). > > Replace the uses of kmap_atomic() within the highmem code. > > On profiling clear_huge_page() using ftrace an improvement > of 62% was observed on the below setup. > > Setup:- > Below data has been collected on Qualcomm's SM7250 SoC THP enabled > (kernel v4.19.113) with only CPU-0(Cortex-A55) and CPU-7(Cortex-A76) > switched on and set to max frequency, also DDR set to perf governor. > > FTRACE Data:- > > Base data:- > Number of iterations: 48 > Mean of allocation time: 349.5 us > std deviation: 74.5 us > > v4 data:- > Number of iterations: 48 > Mean of allocation time: 131 us > std deviation: 32.7 us > > The following simple userspace experiment to allocate > 100MB(BUF_SZ) of pages and writing to it gave us a good insight, > we observed an improvement of 42% in allocation and writing timings. > ------------------------------------------------------------- > Test code snippet > ------------------------------------------------------------- > clock_start(); > buf = malloc(BUF_SZ); /* Allocate 100 MB of memory */ > > for(i=0; i < BUF_SZ_PAGES; i++) > { > *((int *)(buf + (i*PAGE_SIZE))) = 1; > } > clock_end(); > ------------------------------------------------------------- > > Malloc test timings for 100MB anon allocation:- > > Base data:- > Number of iterations: 100 > Mean of allocation time: 31831 us > std deviation: 4286 us > > v4 data:- > Number of iterations: 100 > Mean of allocation time: 18193 us > std deviation: 4915 us > > Signed-off-by: Ira Weiny <[email protected]>
This already has my signed off by so I'm not going to 'review'. With Prathu's testing information I hope this can land. Andrew did you see this patch? Thanks, Ira > Signed-off-by: Prathu Baronia <[email protected]> > [Updated commit text with test data] > --- > include/linux/highmem.h | 28 ++++++++++++++-------------- > 1 file changed, 14 insertions(+), 14 deletions(-) > > diff --git a/include/linux/highmem.h b/include/linux/highmem.h > index d2c70d3772a3..9a202c7e4e26 100644 > --- a/include/linux/highmem.h > +++ b/include/linux/highmem.h > @@ -146,9 +146,9 @@ static inline void invalidate_kernel_vmap_range(void > *vaddr, int size) > #ifndef clear_user_highpage > static inline void clear_user_highpage(struct page *page, unsigned long > vaddr) > { > - void *addr = kmap_atomic(page); > + void *addr = kmap_local_page(page); > clear_user_page(addr, vaddr, page); > - kunmap_atomic(addr); > + kunmap_local(addr); > } > #endif > > @@ -199,9 +199,9 @@ alloc_zeroed_user_highpage_movable(struct vm_area_struct > *vma, > > static inline void clear_highpage(struct page *page) > { > - void *kaddr = kmap_atomic(page); > + void *kaddr = kmap_local_page(page); > clear_page(kaddr); > - kunmap_atomic(kaddr); > + kunmap_local(kaddr); > } > > /* > @@ -216,7 +216,7 @@ static inline void zero_user_segments(struct page *page, > unsigned start1, unsigned end1, > unsigned start2, unsigned end2) > { > - void *kaddr = kmap_atomic(page); > + void *kaddr = kmap_local_page(page); > unsigned int i; > > BUG_ON(end1 > page_size(page) || end2 > page_size(page)); > @@ -227,7 +227,7 @@ static inline void zero_user_segments(struct page *page, > if (end2 > start2) > memset(kaddr + start2, 0, end2 - start2); > > - kunmap_atomic(kaddr); > + kunmap_local(kaddr); > for (i = 0; i < compound_nr(page); i++) > flush_dcache_page(page + i); > } > @@ -252,11 +252,11 @@ static inline void copy_user_highpage(struct page *to, > struct page *from, > { > char *vfrom, *vto; > > - vfrom = kmap_atomic(from); > - vto = kmap_atomic(to); > + vfrom = kmap_local_page(from); > + vto = kmap_local_page(to); > copy_user_page(vto, vfrom, vaddr, to); > - kunmap_atomic(vto); > - kunmap_atomic(vfrom); > + kunmap_local(vto); > + kunmap_local(vfrom); > } > > #endif > @@ -267,11 +267,11 @@ static inline void copy_highpage(struct page *to, > struct page *from) > { > char *vfrom, *vto; > > - vfrom = kmap_atomic(from); > - vto = kmap_atomic(to); > + vfrom = kmap_local_page(from); > + vto = kmap_local_page(to); > copy_page(vto, vfrom); > - kunmap_atomic(vto); > - kunmap_atomic(vfrom); > + kunmap_local(vto); > + kunmap_local(vfrom); > } > > #endif > -- > 2.17.1 >
