[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20240807082508.358322-3-21cnbao@gmail.com>
Date: Wed, 7 Aug 2024 20:25:08 +1200
From: Barry Song <21cnbao@...il.com>
To: akpm@...ux-foundation.org,
linux-mm@...ck.org
Cc: justinjiang@...o.com,
chrisl@...nel.org,
david@...hat.com,
hughd@...gle.com,
kaleshsingh@...gle.com,
kasong@...cent.com,
linux-kernel@...r.kernel.org,
ryan.roberts@....com,
v-songbaohua@...o.com,
ying.huang@...el.com
Subject: [PATCH v2 2/2] mm: attempt to batch free swap entries for zap_pte_range()
From: Barry Song <v-songbaohua@...o.com>
Zhiguo reported that swap release could be a serious bottleneck
during process exits[1]. With mTHP, we have the opportunity to
batch free swaps.
Thanks to the work of Chris and Kairui[2], I was able to achieve
this optimization with minimal code changes by building on their
efforts.
If swap_count is 1, which is likely true as most anon memory are
private, we can free all contiguous swap slots all together.
Ran the below test program for measuring the bandwidth of munmap
using zRAM and 64KiB mTHP:
#include <sys/mman.h>
#include <sys/time.h>
#include <stdlib.h>
unsigned long long tv_to_ms(struct timeval tv)
{
return tv.tv_sec * 1000 + tv.tv_usec / 1000;
}
main()
{
struct timeval tv_b, tv_e;
int i;
#define SIZE 1024*1024*1024
void *p = mmap(NULL, SIZE, PROT_READ | PROT_WRITE,
MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
if (!p) {
perror("fail to get memory");
exit(-1);
}
madvise(p, SIZE, MADV_HUGEPAGE);
memset(p, 0x11, SIZE); /* write to get mem */
madvise(p, SIZE, MADV_PAGEOUT);
gettimeofday(&tv_b, NULL);
munmap(p, SIZE);
gettimeofday(&tv_e, NULL);
printf("munmap in bandwidth: %ld bytes/ms\n",
SIZE/(tv_to_ms(tv_e) - tv_to_ms(tv_b)));
}
The result is as below (munmap bandwidth):
mm-unstable mm-unstable-with-patch
round1 21053761 63161283
round2 21053761 63161283
round3 21053761 63161283
round4 20648881 67108864
round5 20648881 67108864
munmap bandwidth becomes 3X faster.
[1] https://lore.kernel.org/linux-mm/20240731133318.527-1-justinjiang@vivo.com/
[2] https://lore.kernel.org/linux-mm/20240730-swap-allocator-v5-0-cb9c148b9297@kernel.org/
Cc: Kairui Song <kasong@...cent.com>
Cc: Chris Li <chrisl@...nel.org>
Cc: "Huang, Ying" <ying.huang@...el.com>
Cc: Hugh Dickins <hughd@...gle.com>
Cc: Kalesh Singh <kaleshsingh@...gle.com>
Cc: Ryan Roberts <ryan.roberts@....com>
Cc: David Hildenbrand <david@...hat.com>
Signed-off-by: Barry Song <v-songbaohua@...o.com>
---
mm/swapfile.c | 78 +++++++++++++++++++++++++++++++++++++++++++--------
1 file changed, 67 insertions(+), 11 deletions(-)
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 35cb58373493..25c3f98fa8d5 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -156,6 +156,25 @@ static bool swap_is_has_cache(struct swap_info_struct *si,
return true;
}
+static bool swap_is_last_map(struct swap_info_struct *si,
+ unsigned long offset, int nr_pages,
+ bool *has_cache)
+{
+ unsigned char *map = si->swap_map + offset;
+ unsigned char *map_end = map + nr_pages;
+ bool cached = false;
+
+ do {
+ if ((*map & ~SWAP_HAS_CACHE) != 1)
+ return false;
+ if (*map & SWAP_HAS_CACHE)
+ cached = true;
+ } while (++map < map_end);
+
+ *has_cache = cached;
+ return true;
+}
+
/*
* returns number of pages in the folio that backs the swap entry. If positive,
* the folio was reclaimed. If negative, the folio was not reclaimed. If 0, no
@@ -1469,6 +1488,53 @@ static unsigned char __swap_entry_free(struct swap_info_struct *si,
return usage;
}
+static bool __swap_entries_free(struct swap_info_struct *si,
+ swp_entry_t entry, int nr)
+{
+ unsigned long offset = swp_offset(entry);
+ unsigned int type = swp_type(entry);
+ struct swap_cluster_info *ci;
+ bool has_cache = false;
+ unsigned char count;
+ bool can_batch;
+ int i;
+
+ if (nr <= 1 || swap_count(data_race(si->swap_map[offset])) != 1)
+ goto fallback;
+ /* cross into another cluster */
+ if (nr > SWAPFILE_CLUSTER - offset % SWAPFILE_CLUSTER)
+ goto fallback;
+
+ ci = lock_cluster_or_swap_info(si, offset);
+ can_batch = swap_is_last_map(si, offset, nr, &has_cache);
+ if (can_batch) {
+ for (i = 0; i < nr; i++)
+ WRITE_ONCE(si->swap_map[offset + i], SWAP_HAS_CACHE);
+ }
+ unlock_cluster_or_swap_info(si, ci);
+
+ if (!can_batch)
+ goto fallback;
+ if (!has_cache) {
+ spin_lock(&si->lock);
+ swap_entry_range_free(si, entry, nr);
+ spin_unlock(&si->lock);
+ }
+ return has_cache;
+
+fallback:
+ for (i = 0; i < nr; i++) {
+ if (data_race(si->swap_map[offset + i])) {
+ count = __swap_entry_free(si, swp_entry(type, offset + i));
+ if (count == SWAP_HAS_CACHE)
+ has_cache = true;
+ } else {
+ WARN_ON_ONCE(1);
+ }
+ }
+ return has_cache;
+}
+
/*
* Drop the last HAS_CACHE flag of swap entries, caller have to
* ensure all entries belong to the same cgroup.
@@ -1792,11 +1858,9 @@ void free_swap_and_cache_nr(swp_entry_t entry, int nr)
{
const unsigned long start_offset = swp_offset(entry);
const unsigned long end_offset = start_offset + nr;
- unsigned int type = swp_type(entry);
struct swap_info_struct *si;
bool any_only_cache = false;
unsigned long offset;
- unsigned char count;
if (non_swap_entry(entry))
return;
@@ -1811,15 +1875,7 @@ void free_swap_and_cache_nr(swp_entry_t entry, int nr)
/*
* First free all entries in the range.
*/
- for (offset = start_offset; offset < end_offset; offset++) {
- if (data_race(si->swap_map[offset])) {
- count = __swap_entry_free(si, swp_entry(type, offset));
- if (count == SWAP_HAS_CACHE)
- any_only_cache = true;
- } else {
- WARN_ON_ONCE(1);
- }
- }
+ any_only_cache = __swap_entries_free(si, entry, nr);
/*
* Short-circuit the below loop if none of the entries had their
--
2.34.1
Powered by blists - more mailing lists