From: Alistair Popple <apopple@nvidia.com>
To: linux-mm@kvack.org, cgroups@vger.kernel.org
Cc: linux-kernel@vger.kernel.org, jgg@nvidia.com,
jhubbard@nvidia.com, tjmercier@google.com, hannes@cmpxchg.org,
surenb@google.com, mkoutny@suse.com, daniel@ffwll.ch,
Alistair Popple <apopple@nvidia.com>,
Shuah Khan <shuah@kernel.org>,
linux-kselftest@vger.kernel.org
Subject: [RFC PATCH 19/19] selftests/vm: Add pins-cgroup selftest for mlock/mmap
Date: Tue, 24 Jan 2023 16:42:48 +1100 [thread overview]
Message-ID: <9e24fb108c8fae43c18e0189d86cbb978848fbb0.1674538665.git-series.apopple@nvidia.com> (raw)
In-Reply-To: <cover.f52b9eb2792bccb8a9ecd6bc95055705cfe2ae03.1674538665.git-series.apopple@nvidia.com>
Add some basic tests of mlock/mmap cgroup accounting for pinned
memory.
Signed-off-by: Alistair Popple <apopple@nvidia.com>
Cc: Shuah Khan <shuah@kernel.org>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mm@kvack.org
Cc: linux-kselftest@vger.kernel.org
Cc: cgroups@vger.kernel.org
---
MAINTAINERS | 1 +-
tools/testing/selftests/vm/Makefile | 1 +-
tools/testing/selftests/vm/pins-cgroup.c | 271 ++++++++++++++++++++++++-
3 files changed, 273 insertions(+)
create mode 100644 tools/testing/selftests/vm/pins-cgroup.c
diff --git a/MAINTAINERS b/MAINTAINERS
index f8526e2..4c4eed9 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -5387,6 +5387,7 @@ L: cgroups@vger.kernel.org
L: linux-mm@kvack.org
S: Maintained
F: mm/pins_cgroup.c
+F: tools/testing/selftests/vm/pins-cgroup.c
CORETEMP HARDWARE MONITORING DRIVER
M: Fenghua Yu <fenghua.yu@intel.com>
diff --git a/tools/testing/selftests/vm/Makefile b/tools/testing/selftests/vm/Makefile
index 89c14e4..0653720 100644
--- a/tools/testing/selftests/vm/Makefile
+++ b/tools/testing/selftests/vm/Makefile
@@ -56,6 +56,7 @@ TEST_GEN_PROGS += soft-dirty
TEST_GEN_PROGS += split_huge_page_test
TEST_GEN_FILES += ksm_tests
TEST_GEN_PROGS += ksm_functional_tests
+TEST_GEN_FILES += pins-cgroup
ifeq ($(MACHINE),x86_64)
CAN_BUILD_I386 := $(shell ./../x86/check_cc.sh "$(CC)" ../x86/trivial_32bit_program.c -m32)
diff --git a/tools/testing/selftests/vm/pins-cgroup.c b/tools/testing/selftests/vm/pins-cgroup.c
new file mode 100644
index 0000000..c2eabc2
--- /dev/null
+++ b/tools/testing/selftests/vm/pins-cgroup.c
@@ -0,0 +1,271 @@
+// SPDX-License-Identifier: GPL-2.0
+#include "../kselftest_harness.h"
+
+#define _GNU_SOURCE
+#include <fcntl.h>
+#include <assert.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/mman.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/ioctl.h>
+#include <sys/prctl.h>
+#include <sys/resource.h>
+#include <sys/capability.h>
+#include <unistd.h>
+
+#define CGROUP_TEMP "/sys/fs/cgroup/pins_XXXXXX"
+#define PINS_MAX (-1UL)
+
+FIXTURE(pins_cg)
+{
+ char *cg_path;
+ long page_size;
+};
+
+static char *cgroup_new(void)
+{
+ char *cg;
+
+ cg = malloc(sizeof(CGROUP_TEMP));
+ strcpy(cg, CGROUP_TEMP);
+ if (!mkdtemp(cg)) {
+ perror("Failed to create cgroup");
+ return NULL;
+ }
+
+ return cg;
+}
+
+static int cgroup_add_proc(char *cg, pid_t pid)
+{
+ char *cg_proc;
+ FILE *f;
+ int ret = 0;
+
+ if (asprintf(&cg_proc, "%s/cgroup.procs", cg) < 0)
+ return -1;
+
+ f = fopen(cg_proc, "w");
+ free(cg_proc);
+ if (!f)
+ return -1;
+
+ if (fprintf(f, "%ld\n", (long) pid) < 0)
+ ret = -1;
+
+ fclose(f);
+ return ret;
+}
+
+static int cgroup_set_limit(char *cg, unsigned long limit)
+{
+ char *cg_pins_max;
+ FILE *f;
+ int ret = 0;
+
+ if (asprintf(&cg_pins_max, "%s/pins.max", cg) < 0)
+ return -1;
+
+ f = fopen(cg_pins_max, "w");
+ free(cg_pins_max);
+ if (!f)
+ return -1;
+
+ if (limit != PINS_MAX) {
+ if (fprintf(f, "%ld\n", limit) < 0)
+ ret = -1;
+ } else {
+ if (fprintf(f, "max\n") < 0)
+ ret = -1;
+ }
+
+ fclose(f);
+ return ret;
+}
+
+FIXTURE_SETUP(pins_cg)
+{
+ char *cg_subtree_control;
+ FILE *f;
+
+ if (asprintf(&cg_subtree_control,
+ "/sys/fs/cgroup/cgroup.subtree_control") < 0)
+ return;
+
+ f = fopen(cg_subtree_control, "w");
+ free(cg_subtree_control);
+ if (!f)
+ return;
+
+ fprintf(f, "+pins\n");
+ fclose(f);
+
+ self->cg_path = cgroup_new();
+ self->page_size = sysconf(_SC_PAGE_SIZE);
+}
+
+FIXTURE_TEARDOWN(pins_cg)
+{
+ cgroup_add_proc("/sys/fs/cgroup", getpid());
+
+ rmdir(self->cg_path);
+ free(self->cg_path);
+}
+
+static long cgroup_pins(char *cg)
+{
+ long pin_count;
+ char *cg_pins_current;
+ FILE *f;
+ int ret;
+
+ if (asprintf(&cg_pins_current, "%s/pins.current", cg) < 0)
+ return -1;
+
+ f = fopen(cg_pins_current, "r");
+ if (!f) {
+ printf("Can't open %s\n", cg_pins_current);
+ getchar();
+ free(cg_pins_current);
+ return -2;
+ }
+
+ free(cg_pins_current);
+
+ if (fscanf(f, "%ld", &pin_count) == EOF)
+ ret = -3;
+ else
+ ret = pin_count;
+
+ fclose(f);
+ return ret;
+}
+
+static int set_rlim_memlock(unsigned long size)
+{
+ struct rlimit rlim_memlock = {
+ .rlim_cur = size,
+ .rlim_max = size,
+ };
+ cap_t cap;
+ cap_value_t capability[1] = { CAP_IPC_LOCK };
+
+ /*
+ * Many of the rlimit checks are skipped if a process has
+ * CAP_IP_LOCK. As this test should be run as root we need to
+ * explicitly drop it.
+ */
+ cap = cap_get_proc();
+ if (!cap)
+ return -1;
+ if (cap_set_flag(cap, CAP_EFFECTIVE, 1, capability, CAP_CLEAR))
+ return -1;
+ if (cap_set_proc(cap))
+ return -1;
+ return setrlimit(RLIMIT_MEMLOCK, &rlim_memlock);
+}
+
+TEST_F(pins_cg, basic)
+{
+ pid_t child_pid;
+ long page_size = self->page_size;
+ char *p;
+
+ ASSERT_EQ(cgroup_add_proc(self->cg_path, getpid()), 0);
+ p = mmap(NULL, 32*page_size, PROT_READ | PROT_WRITE,
+ MAP_ANONYMOUS | MAP_PRIVATE, -1, 0);
+ ASSERT_NE(p, MAP_FAILED);
+
+ ASSERT_EQ(cgroup_pins(self->cg_path), 0);
+ memset(p, 0, 16*page_size);
+ ASSERT_EQ(mlock(p, page_size), 0);
+ ASSERT_EQ(cgroup_pins(self->cg_path), 1);
+ ASSERT_EQ(mlock(p + page_size, page_size), 0);
+ ASSERT_EQ(cgroup_pins(self->cg_path), 2);
+ ASSERT_EQ(mlock(p, page_size), 0);
+ ASSERT_EQ(cgroup_pins(self->cg_path), 2);
+ ASSERT_EQ(mlock(p, 4*page_size), 0);
+ ASSERT_EQ(cgroup_pins(self->cg_path), 4);
+ ASSERT_EQ(munlock(p + 2*page_size, 2*page_size), 0);
+ ASSERT_EQ(cgroup_pins(self->cg_path), 2);
+ ASSERT_EQ(cgroup_set_limit(self->cg_path, 8), 0);
+ ASSERT_EQ(mlock(p, 16*page_size), -1);
+ ASSERT_EQ(errno, ENOMEM);
+ ASSERT_EQ(cgroup_pins(self->cg_path), 2);
+ ASSERT_EQ(cgroup_set_limit(self->cg_path, PINS_MAX), 0);
+
+ /* check mremap() a locked region correctly accounts locked pages */
+ ASSERT_EQ(mlock(p, 32*page_size), 0);
+ ASSERT_EQ(cgroup_pins(self->cg_path), 32);
+ p = mremap(p, 32*page_size, 64*page_size, MREMAP_MAYMOVE);
+ ASSERT_NE(p, MAP_FAILED);
+ ASSERT_EQ(cgroup_pins(self->cg_path), 64);
+ ASSERT_EQ(munmap(p + 32*page_size, 32*page_size), 0)
+ ASSERT_EQ(cgroup_pins(self->cg_path), 32);
+ p = mremap(p, 32*page_size, 32*page_size, MREMAP_MAYMOVE | MREMAP_DONTUNMAP);
+ ASSERT_NE(p, MAP_FAILED);
+ ASSERT_EQ(cgroup_pins(self->cg_path), 32);
+ ASSERT_EQ(munlock(p, 32*page_size), 0);
+
+ /* mremap() a locked region should fail if limit exceeded */
+ ASSERT_EQ(set_rlim_memlock(32*page_size), 0);
+ ASSERT_EQ(mlock(p, 32*page_size), 0);
+ ASSERT_EQ(mremap(p, 32*page_size, 64*page_size, 0), MAP_FAILED);
+ ASSERT_EQ(munlock(p, 32*page_size), 0);
+
+ /* Exceeds rlimit, expected to fail */
+ ASSERT_EQ(set_rlim_memlock(16*page_size), 0);
+ ASSERT_EQ(mlock(p, 32*page_size), -1);
+ ASSERT_EQ(errno, ENOMEM);
+
+ /* memory in the child isn't locked so shouldn't increase pin_cg count */
+ ASSERT_EQ(mlock(p, 16*page_size), 0);
+ child_pid = fork();
+ if (!child_pid) {
+ ASSERT_EQ(cgroup_pins(self->cg_path), 16);
+ ASSERT_EQ(mlock(p, 16*page_size), 0);
+ ASSERT_EQ(cgroup_pins(self->cg_path), 32);
+ return;
+
+ }
+ waitpid(child_pid, NULL, 0);
+
+ /* check that child exit uncharged the pins */
+ ASSERT_EQ(cgroup_pins(self->cg_path), 16);
+}
+
+TEST_F(pins_cg, mmap)
+{
+ char *p;
+
+ ASSERT_EQ(cgroup_add_proc(self->cg_path, getpid()), 0);
+ p = mmap(NULL, 4*self->page_size, PROT_READ | PROT_WRITE,
+ MAP_ANONYMOUS | MAP_PRIVATE | MAP_LOCKED, -1, 0);
+ ASSERT_NE(p, MAP_FAILED);
+ ASSERT_EQ(cgroup_pins(self->cg_path), 4);
+}
+
+/*
+ * Test moving to a different cgroup.
+ */
+TEST_F(pins_cg, move_cg)
+{
+ char *p, *new_cg;
+
+ ASSERT_EQ(cgroup_add_proc(self->cg_path, getpid()), 0);
+ p = mmap(NULL, 16*self->page_size, PROT_READ | PROT_WRITE,
+ MAP_ANONYMOUS | MAP_PRIVATE, -1, 0);
+ ASSERT_NE(p, MAP_FAILED);
+ memset(p, 0, 16*self->page_size);
+ ASSERT_EQ(mlock(p, 16*self->page_size), 0);
+ ASSERT_EQ(cgroup_pins(self->cg_path), 16);
+ ASSERT_NE(new_cg = cgroup_new(), NULL);
+ ASSERT_EQ(cgroup_add_proc(new_cg, getpid()), 0);
+ ASSERT_EQ(cgroup_pins(new_cg), 16);
+ ASSERT_EQ(cgroup_add_proc(self->cg_path, getpid()), 0);
+ rmdir(new_cg);
+}
+TEST_HARNESS_MAIN
--
git-series 0.9.1
next prev parent reply other threads:[~2023-01-24 5:50 UTC|newest]
Thread overview: 56+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-01-24 5:42 [RFC PATCH 00/19] mm: Introduce a cgroup to limit the amount of locked and pinned memory Alistair Popple
2023-01-24 5:42 ` [RFC PATCH 01/19] mm: Introduce vm_account Alistair Popple
2023-01-24 6:29 ` Christoph Hellwig
2023-01-24 14:32 ` Jason Gunthorpe
2023-01-30 11:36 ` Alistair Popple
2023-01-31 14:00 ` David Hildenbrand
2023-01-24 5:42 ` [RFC PATCH 02/19] drivers/vhost: Convert to use vm_account Alistair Popple
2023-01-24 5:55 ` Michael S. Tsirkin
2023-01-30 10:43 ` Alistair Popple
2023-01-24 14:34 ` Jason Gunthorpe
2023-01-24 5:42 ` [RFC PATCH 03/19] drivers/vdpa: Convert vdpa to use the new vm_structure Alistair Popple
2023-01-24 14:35 ` Jason Gunthorpe
2023-01-24 5:42 ` [RFC PATCH 04/19] infiniband/umem: Convert to use vm_account Alistair Popple
2023-01-24 5:42 ` [RFC PATCH 05/19] RMDA/siw: " Alistair Popple
2023-01-24 14:37 ` Jason Gunthorpe
2023-01-24 15:22 ` Bernard Metzler
2023-01-24 15:56 ` Bernard Metzler
2023-01-30 11:34 ` Alistair Popple
2023-01-30 13:27 ` Bernard Metzler
2023-01-24 5:42 ` [RFC PATCH 06/19] RDMA/usnic: convert " Alistair Popple
2023-01-24 14:41 ` Jason Gunthorpe
2023-01-30 11:10 ` Alistair Popple
2023-01-24 5:42 ` [RFC PATCH 07/19] vfio/type1: Charge pinned pages to pinned_vm instead of locked_vm Alistair Popple
2023-01-24 5:42 ` [RFC PATCH 08/19] vfio/spapr_tce: Convert accounting to pinned_vm Alistair Popple
2023-01-24 5:42 ` [RFC PATCH 09/19] io_uring: convert to use vm_account Alistair Popple
2023-01-24 14:44 ` Jason Gunthorpe
2023-01-30 11:12 ` Alistair Popple
2023-01-30 13:21 ` Jason Gunthorpe
2023-01-24 5:42 ` [RFC PATCH 10/19] net: skb: Switch to using vm_account Alistair Popple
2023-01-24 14:51 ` Jason Gunthorpe
2023-01-30 11:17 ` Alistair Popple
2023-02-06 4:36 ` Alistair Popple
2023-02-06 13:14 ` Jason Gunthorpe
2023-01-24 5:42 ` [RFC PATCH 11/19] xdp: convert to use vm_account Alistair Popple
2023-01-24 5:42 ` [RFC PATCH 12/19] kvm/book3s_64_vio: Convert account_locked_vm() to vm_account_pinned() Alistair Popple
2023-01-24 5:42 ` [RFC PATCH 13/19] fpga: dfl: afu: convert to use vm_account Alistair Popple
2023-01-24 5:42 ` [RFC PATCH 14/19] mm: Introduce a cgroup for pinned memory Alistair Popple
2023-01-27 21:44 ` Tejun Heo
2023-01-30 13:20 ` Jason Gunthorpe
2023-01-24 5:42 ` [RFC PATCH 15/19] mm/util: Extend vm_account to charge pages against the pin cgroup Alistair Popple
2023-01-24 5:42 ` [RFC PATCH 16/19] mm/util: Refactor account_locked_vm Alistair Popple
2023-01-24 5:42 ` [RFC PATCH 17/19] mm: Convert mmap and mlock to use account_locked_vm Alistair Popple
2023-01-24 5:42 ` [RFC PATCH 18/19] mm/mmap: Charge locked memory to pins cgroup Alistair Popple
2023-01-24 5:42 ` Alistair Popple [this message]
2023-01-24 18:26 ` [RFC PATCH 00/19] mm: Introduce a cgroup to limit the amount of locked and pinned memory Yosry Ahmed
2023-01-31 0:54 ` Alistair Popple
2023-01-31 5:14 ` Yosry Ahmed
2023-01-31 11:22 ` Alistair Popple
2023-01-31 19:49 ` Yosry Ahmed
2023-01-24 20:12 ` Jason Gunthorpe
2023-01-31 13:57 ` David Hildenbrand
2023-01-31 14:03 ` Jason Gunthorpe
2023-01-31 14:06 ` David Hildenbrand
2023-01-31 14:10 ` Jason Gunthorpe
2023-01-31 14:15 ` David Hildenbrand
2023-01-31 14:21 ` Jason Gunthorpe
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=9e24fb108c8fae43c18e0189d86cbb978848fbb0.1674538665.git-series.apopple@nvidia.com \
--to=apopple@nvidia.com \
--cc=cgroups@vger.kernel.org \
--cc=daniel@ffwll.ch \
--cc=hannes@cmpxchg.org \
--cc=jgg@nvidia.com \
--cc=jhubbard@nvidia.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mkoutny@suse.com \
--cc=shuah@kernel.org \
--cc=surenb@google.com \
--cc=tjmercier@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).