summaryrefslogtreecommitdiff
path: root/tools
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2022-05-27 21:40:49 +0300
committerLinus Torvalds <torvalds@linux-foundation.org>2022-05-27 21:40:49 +0300
commit8291eaafed36f575f23951f3ce18407f480e9ecf (patch)
tree279b61422ba2df7b8579af8ccc81331de80affa8 /tools
parent77fb622de1393b1d54f24f4f7ed98f84feeda502 (diff)
parentfa020a2b87d24016723fff4a4237deb612478a32 (diff)
downloadlinux-8291eaafed36f575f23951f3ce18407f480e9ecf.tar.xz
Merge tag 'mm-stable-2022-05-27' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Pull more MM updates from Andrew Morton: - Two follow-on fixes for the post-5.19 series "Use pageblock_order for cma and alloc_contig_range alignment", from Zi Yan. - A series of z3fold cleanups and fixes from Miaohe Lin. - Some memcg selftests work from Michal Koutný <mkoutny@suse.com> - Some swap fixes and cleanups from Miaohe Lin - Several individual minor fixups * tag 'mm-stable-2022-05-27' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (25 commits) mm/shmem.c: suppress shift warning mm: Kconfig: reorganize misplaced mm options mm: kasan: fix input of vmalloc_to_page() mm: fix is_pinnable_page against a cma page mm: filter out swapin error entry in shmem mapping mm/shmem: fix infinite loop when swap in shmem error at swapoff time mm/madvise: free hwpoison and swapin error entry in madvise_free_pte_range mm/swapfile: fix lost swap bits in unuse_pte() mm/swapfile: unuse_pte can map random data if swap read fails selftests: memcg: factor out common parts of memory.{low,min} tests selftests: memcg: remove protection from top level memcg selftests: memcg: adjust expected reclaim values of protected cgroups selftests: memcg: expect no low events in unprotected sibling selftests: memcg: fix compilation mm/z3fold: fix z3fold_page_migrate races with z3fold_map mm/z3fold: fix z3fold_reclaim_page races with z3fold_free mm/z3fold: always clear PAGE_CLAIMED under z3fold page lock mm/z3fold: put z3fold page back into unbuddied list when reclaim or migration fails revert "mm/z3fold.c: allow __GFP_HIGHMEM in z3fold_alloc" mm/z3fold: throw warning on failure of trylock_page in z3fold_alloc ...
Diffstat (limited to 'tools')
-rw-r--r--tools/testing/selftests/cgroup/memcg_protection.m89
-rw-r--r--tools/testing/selftests/cgroup/test_memcontrol.c247
2 files changed, 151 insertions, 185 deletions
diff --git a/tools/testing/selftests/cgroup/memcg_protection.m b/tools/testing/selftests/cgroup/memcg_protection.m
new file mode 100644
index 000000000000..051daa3477b6
--- /dev/null
+++ b/tools/testing/selftests/cgroup/memcg_protection.m
@@ -0,0 +1,89 @@
+% SPDX-License-Identifier: GPL-2.0
+%
+% run as: octave-cli memcg_protection.m
+%
+% This script simulates reclaim protection behavior on a single level of memcg
+% hierarchy to illustrate how overcommitted protection spreads among siblings
+% (as it depends also on their current consumption).
+%
+% Simulation assumes siblings consumed the initial amount of memory (w/out
+% reclaim) and then the reclaim starts, all memory is reclaimable, i.e. treated
+% same. It simulates only non-low reclaim and assumes all memory.min = 0.
+%
+% Input configurations
+% --------------------
+% E number parent effective protection
+% n vector nominal protection of siblings set at the given level (memory.low)
+% c vector current consumption -,,- (memory.current)
+
+% example from testcase (values in GB)
+E = 50 / 1024;
+n = [75 25 0 500 ] / 1024;
+c = [50 50 50 0] / 1024;
+
+% Reclaim parameters
+% ------------------
+
+% Minimal reclaim amount (GB)
+cluster = 32*4 / 2**20;
+
+% Reclaim coefficient (think as 0.5^sc->priority)
+alpha = .1
+
+% Simulation parameters
+% ---------------------
+epsilon = 1e-7;
+timeout = 1000;
+
+% Simulation loop
+% ---------------
+
+ch = [];
+eh = [];
+rh = [];
+
+for t = 1:timeout
+ % low_usage
+ u = min(c, n);
+ siblings = sum(u);
+
+ % effective_protection()
+ protected = min(n, c); % start with nominal
+ e = protected * min(1, E / siblings); % normalize overcommit
+
+ % recursive protection
+ unclaimed = max(0, E - siblings);
+ parent_overuse = sum(c) - siblings;
+ if (unclaimed > 0 && parent_overuse > 0)
+ overuse = max(0, c - protected);
+ e += unclaimed * (overuse / parent_overuse);
+ endif
+
+ % get_scan_count()
+ r = alpha * c; % assume all memory is in a single LRU list
+
+ % commit 1bc63fb1272b ("mm, memcg: make scan aggression always exclude protection")
+ sz = max(e, c);
+ r .*= (1 - (e+epsilon) ./ (sz+epsilon));
+
+ % uncomment to debug prints
+ % e, c, r
+
+ % nothing to reclaim, reached equilibrium
+ if max(r) < epsilon
+ break;
+ endif
+
+ % SWAP_CLUSTER_MAX roundup
+ r = max(r, (r > epsilon) .* cluster);
+ % XXX here I do parallel reclaim of all siblings
+ % in reality reclaim is serialized and each sibling recalculates own residual
+ c = max(c - r, 0);
+
+ ch = [ch ; c];
+ eh = [eh ; e];
+ rh = [rh ; r];
+endfor
+
+t
+c, e
diff --git a/tools/testing/selftests/cgroup/test_memcontrol.c b/tools/testing/selftests/cgroup/test_memcontrol.c
index 44a974ec472c..8833359556f3 100644
--- a/tools/testing/selftests/cgroup/test_memcontrol.c
+++ b/tools/testing/selftests/cgroup/test_memcontrol.c
@@ -190,13 +190,6 @@ cleanup:
return ret;
}
-static int alloc_pagecache_50M(const char *cgroup, void *arg)
-{
- int fd = (long)arg;
-
- return alloc_pagecache(fd, MB(50));
-}
-
static int alloc_pagecache_50M_noexit(const char *cgroup, void *arg)
{
int fd = (long)arg;
@@ -247,33 +240,39 @@ static int cg_test_proc_killed(const char *cgroup)
/*
* First, this test creates the following hierarchy:
- * A memory.min = 50M, memory.max = 200M
- * A/B memory.min = 50M, memory.current = 50M
+ * A memory.min = 0, memory.max = 200M
+ * A/B memory.min = 50M
* A/B/C memory.min = 75M, memory.current = 50M
* A/B/D memory.min = 25M, memory.current = 50M
* A/B/E memory.min = 0, memory.current = 50M
* A/B/F memory.min = 500M, memory.current = 0
*
- * Usages are pagecache, but the test keeps a running
+ * (or memory.low if we test soft protection)
+ *
+ * Usages are pagecache and the test keeps a running
* process in every leaf cgroup.
* Then it creates A/G and creates a significant
- * memory pressure in it.
+ * memory pressure in A.
*
+ * Then it checks actual memory usages and expects that:
* A/B memory.current ~= 50M
- * A/B/C memory.current ~= 33M
- * A/B/D memory.current ~= 17M
- * A/B/F memory.current ~= 0
+ * A/B/C memory.current ~= 29M
+ * A/B/D memory.current ~= 21M
+ * A/B/E memory.current ~= 0
+ * A/B/F memory.current = 0
+ * (for origin of the numbers, see model in memcg_protection.m.)
*
* After that it tries to allocate more than there is
- * unprotected memory in A available, and checks
- * checks that memory.min protects pagecache even
- * in this case.
+ * unprotected memory in A available, and checks that:
+ * a) memory.min protects pagecache even in this case,
+ * b) memory.low allows reclaiming page cache with low events.
*/
-static int test_memcg_min(const char *root)
+static int test_memcg_protection(const char *root, bool min)
{
- int ret = KSFT_FAIL;
+ int ret = KSFT_FAIL, rc;
char *parent[3] = {NULL};
char *children[4] = {NULL};
+ const char *attribute = min ? "memory.min" : "memory.low";
long c[4];
int i, attempts;
int fd;
@@ -297,8 +296,10 @@ static int test_memcg_min(const char *root)
if (cg_create(parent[0]))
goto cleanup;
- if (cg_read_long(parent[0], "memory.min")) {
- ret = KSFT_SKIP;
+ if (cg_read_long(parent[0], attribute)) {
+ /* No memory.min on older kernels is fine */
+ if (min)
+ ret = KSFT_SKIP;
goto cleanup;
}
@@ -335,17 +336,15 @@ static int test_memcg_min(const char *root)
(void *)(long)fd);
}
- if (cg_write(parent[0], "memory.min", "50M"))
+ if (cg_write(parent[1], attribute, "50M"))
goto cleanup;
- if (cg_write(parent[1], "memory.min", "50M"))
+ if (cg_write(children[0], attribute, "75M"))
goto cleanup;
- if (cg_write(children[0], "memory.min", "75M"))
+ if (cg_write(children[1], attribute, "25M"))
goto cleanup;
- if (cg_write(children[1], "memory.min", "25M"))
+ if (cg_write(children[2], attribute, "0"))
goto cleanup;
- if (cg_write(children[2], "memory.min", "0"))
- goto cleanup;
- if (cg_write(children[3], "memory.min", "500M"))
+ if (cg_write(children[3], attribute, "500M"))
goto cleanup;
attempts = 0;
@@ -365,170 +364,35 @@ static int test_memcg_min(const char *root)
for (i = 0; i < ARRAY_SIZE(children); i++)
c[i] = cg_read_long(children[i], "memory.current");
- if (!values_close(c[0], MB(33), 10))
+ if (!values_close(c[0], MB(29), 10))
goto cleanup;
- if (!values_close(c[1], MB(17), 10))
+ if (!values_close(c[1], MB(21), 10))
goto cleanup;
if (c[3] != 0)
goto cleanup;
- if (!cg_run(parent[2], alloc_anon, (void *)MB(170)))
- goto cleanup;
-
- if (!values_close(cg_read_long(parent[1], "memory.current"), MB(50), 3))
- goto cleanup;
-
- ret = KSFT_PASS;
-
-cleanup:
- for (i = ARRAY_SIZE(children) - 1; i >= 0; i--) {
- if (!children[i])
- continue;
-
- cg_destroy(children[i]);
- free(children[i]);
- }
-
- for (i = ARRAY_SIZE(parent) - 1; i >= 0; i--) {
- if (!parent[i])
- continue;
-
- cg_destroy(parent[i]);
- free(parent[i]);
- }
- close(fd);
- return ret;
-}
-
-/*
- * First, this test creates the following hierarchy:
- * A memory.low = 50M, memory.max = 200M
- * A/B memory.low = 50M, memory.current = 50M
- * A/B/C memory.low = 75M, memory.current = 50M
- * A/B/D memory.low = 25M, memory.current = 50M
- * A/B/E memory.low = 0, memory.current = 50M
- * A/B/F memory.low = 500M, memory.current = 0
- *
- * Usages are pagecache.
- * Then it creates A/G an creates a significant
- * memory pressure in it.
- *
- * Then it checks actual memory usages and expects that:
- * A/B memory.current ~= 50M
- * A/B/ memory.current ~= 33M
- * A/B/D memory.current ~= 17M
- * A/B/F memory.current ~= 0
- *
- * After that it tries to allocate more than there is
- * unprotected memory in A available,
- * and checks low and oom events in memory.events.
- */
-static int test_memcg_low(const char *root)
-{
- int ret = KSFT_FAIL;
- char *parent[3] = {NULL};
- char *children[4] = {NULL};
- long low, oom;
- long c[4];
- int i;
- int fd;
-
- fd = get_temp_fd();
- if (fd < 0)
- goto cleanup;
-
- parent[0] = cg_name(root, "memcg_test_0");
- if (!parent[0])
- goto cleanup;
-
- parent[1] = cg_name(parent[0], "memcg_test_1");
- if (!parent[1])
- goto cleanup;
-
- parent[2] = cg_name(parent[0], "memcg_test_2");
- if (!parent[2])
- goto cleanup;
-
- if (cg_create(parent[0]))
- goto cleanup;
-
- if (cg_read_long(parent[0], "memory.low"))
- goto cleanup;
-
- if (cg_write(parent[0], "cgroup.subtree_control", "+memory"))
- goto cleanup;
-
- if (cg_write(parent[0], "memory.max", "200M"))
- goto cleanup;
-
- if (cg_write(parent[0], "memory.swap.max", "0"))
- goto cleanup;
-
- if (cg_create(parent[1]))
- goto cleanup;
-
- if (cg_write(parent[1], "cgroup.subtree_control", "+memory"))
+ rc = cg_run(parent[2], alloc_anon, (void *)MB(170));
+ if (min && !rc)
goto cleanup;
-
- if (cg_create(parent[2]))
+ else if (!min && rc) {
+ fprintf(stderr,
+ "memory.low prevents from allocating anon memory\n");
goto cleanup;
-
- for (i = 0; i < ARRAY_SIZE(children); i++) {
- children[i] = cg_name_indexed(parent[1], "child_memcg", i);
- if (!children[i])
- goto cleanup;
-
- if (cg_create(children[i]))
- goto cleanup;
-
- if (i > 2)
- continue;
-
- if (cg_run(children[i], alloc_pagecache_50M, (void *)(long)fd))
- goto cleanup;
}
- if (cg_write(parent[0], "memory.low", "50M"))
- goto cleanup;
- if (cg_write(parent[1], "memory.low", "50M"))
- goto cleanup;
- if (cg_write(children[0], "memory.low", "75M"))
- goto cleanup;
- if (cg_write(children[1], "memory.low", "25M"))
- goto cleanup;
- if (cg_write(children[2], "memory.low", "0"))
- goto cleanup;
- if (cg_write(children[3], "memory.low", "500M"))
- goto cleanup;
-
- if (cg_run(parent[2], alloc_anon, (void *)MB(148)))
- goto cleanup;
-
if (!values_close(cg_read_long(parent[1], "memory.current"), MB(50), 3))
goto cleanup;
- for (i = 0; i < ARRAY_SIZE(children); i++)
- c[i] = cg_read_long(children[i], "memory.current");
-
- if (!values_close(c[0], MB(33), 10))
- goto cleanup;
-
- if (!values_close(c[1], MB(17), 10))
- goto cleanup;
-
- if (c[3] != 0)
- goto cleanup;
-
- if (cg_run(parent[2], alloc_anon, (void *)MB(166))) {
- fprintf(stderr,
- "memory.low prevents from allocating anon memory\n");
+ if (min) {
+ ret = KSFT_PASS;
goto cleanup;
}
for (i = 0; i < ARRAY_SIZE(children); i++) {
- int no_low_events_index = has_recursiveprot ? 2 : 1;
+ int no_low_events_index = 1;
+ long low, oom;
oom = cg_read_key_long(children[i], "memory.events", "oom ");
low = cg_read_key_long(children[i], "memory.events", "low ");
@@ -564,6 +428,16 @@ cleanup:
return ret;
}
+static int test_memcg_min(const char *root)
+{
+ return test_memcg_protection(root, true);
+}
+
+static int test_memcg_low(const char *root)
+{
+ return test_memcg_protection(root, false);
+}
+
static int alloc_pagecache_max_30M(const char *cgroup, void *arg)
{
size_t size = MB(50);
@@ -1241,7 +1115,16 @@ static int test_memcg_oom_group_leaf_events(const char *root)
if (cg_read_key_long(child, "memory.events", "oom_kill ") <= 0)
goto cleanup;
- if (cg_read_key_long(parent, "memory.events", "oom_kill ") <= 0)
+ parent_oom_events = cg_read_key_long(
+ parent, "memory.events", "oom_kill ");
+ /*
+ * If memory_localevents is not enabled (the default), the parent should
+ * count OOM events in its children groups. Otherwise, it should not
+ * have observed any events.
+ */
+ if (has_localevents && parent_oom_events != 0)
+ goto cleanup;
+ else if (!has_localevents && parent_oom_events <= 0)
goto cleanup;
ret = KSFT_PASS;
@@ -1349,20 +1232,14 @@ static int test_memcg_oom_group_score_events(const char *root)
if (!cg_run(memcg, alloc_anon, (void *)MB(100)))
goto cleanup;
- parent_oom_events = cg_read_key_long(
- parent, "memory.events", "oom_kill ");
- /*
- * If memory_localevents is not enabled (the default), the parent should
- * count OOM events in its children groups. Otherwise, it should not
- * have observed any events.
- */
- if ((has_localevents && parent_oom_events == 0) ||
- parent_oom_events > 0)
- ret = KSFT_PASS;
+ if (cg_read_key_long(memcg, "memory.events", "oom_kill ") != 3)
+ goto cleanup;
if (kill(safe_pid, SIGKILL))
goto cleanup;
+ ret = KSFT_PASS;
+
cleanup:
if (memcg)
cg_destroy(memcg);