- Linux-kselftest-mirror - lists.linaro.org

[PATCH -next 1/2] mm/memfd: make F_SEAL_FUTURE_WRITE seal more robust

by Joel Fernandes (Google)

A better way to do F_SEAL_FUTURE_WRITE seal was discussed [1] last week where we don't need to modify core VFS structures to get the same behavior of the seal. This solves several side-effects pointed out by Andy [2]. [1] https://lore.kernel.org/lkml/20181111173650.GA256781@google.com/ [2] https://lore.kernel.org/lkml/69CE06CC-E47C-4992-848A-66EB23EE6C74@amacapita… Suggested-by: Andy Lutomirski <luto(a)kernel.org> Fixes: 5e653c2923fd ("mm: Add an F_SEAL_FUTURE_WRITE seal to memfd") Signed-off-by: Joel Fernandes (Google) <joel(a)joelfernandes.org> --- fs/hugetlbfs/inode.c | 2 +- mm/memfd.c | 19 ------------------- mm/shmem.c | 24 +++++++++++++++++++++--- 3 files changed, 22 insertions(+), 23 deletions(-) diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 762028994f47..5b54bf893a67 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -558,7 +558,7 @@ static long hugetlbfs_punch_hole(struct inode *inode, loff_t offset, loff_t len) inode_lock(inode); /* protected by i_mutex */ - if (info->seals & F_SEAL_WRITE) { + if (info->seals & (F_SEAL_WRITE | F_SEAL_FUTURE_WRITE)) { inode_unlock(inode); return -EPERM; } diff --git a/mm/memfd.c b/mm/memfd.c index 63fff5e77114..650e65a46b9c 100644 --- a/mm/memfd.c +++ b/mm/memfd.c @@ -201,25 +201,6 @@ static int memfd_add_seals(struct file *file, unsigned int seals) } } - if ((seals & F_SEAL_FUTURE_WRITE) && - !(*file_seals & F_SEAL_FUTURE_WRITE)) { - /* - * The FUTURE_WRITE seal also prevents growing and shrinking - * so we need them to be already set, or requested now. - */ - int test_seals = (seals | *file_seals) & - (F_SEAL_GROW | F_SEAL_SHRINK); - - if (test_seals != (F_SEAL_GROW | F_SEAL_SHRINK)) { - error = -EINVAL; - goto unlock; - } - - spin_lock(&file->f_lock); - file->f_mode &= ~(FMODE_WRITE | FMODE_PWRITE); - spin_unlock(&file->f_lock); - } - *file_seals |= seals; error = 0; diff --git a/mm/shmem.c b/mm/shmem.c index 32eb29bd72c6..cee9878c87f1 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -2121,6 +2121,23 @@ int shmem_lock(struct file *file, int lock, struct user_struct *user) static int shmem_mmap(struct file *file, struct vm_area_struct *vma) { + struct shmem_inode_info *info = SHMEM_I(file_inode(file)); + + /* + * New PROT_READ and MAP_SHARED mmaps are not allowed when "future + * write" seal active. + */ + if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_WRITE) && + (info->seals & F_SEAL_FUTURE_WRITE)) + return -EPERM; + + /* + * Since the F_SEAL_FUTURE_WRITE seals allow for a MAP_SHARED read-only + * mapping, take care to not allow mprotect to revert protections. + */ + if (info->seals & F_SEAL_FUTURE_WRITE) + vma->vm_flags &= ~(VM_MAYWRITE); + file_accessed(file); vma->vm_ops = &shmem_vm_ops; if (IS_ENABLED(CONFIG_TRANSPARENT_HUGE_PAGECACHE) && @@ -2346,8 +2363,9 @@ shmem_write_begin(struct file *file, struct address_space *mapping, pgoff_t index = pos >> PAGE_SHIFT; /* i_mutex is held by caller */ - if (unlikely(info->seals & (F_SEAL_WRITE | F_SEAL_GROW))) { - if (info->seals & F_SEAL_WRITE) + if (unlikely(info->seals & (F_SEAL_GROW | + F_SEAL_WRITE | F_SEAL_FUTURE_WRITE))) { + if (info->seals & (F_SEAL_WRITE | F_SEAL_FUTURE_WRITE)) return -EPERM; if ((info->seals & F_SEAL_GROW) && pos + len > inode->i_size) return -EPERM; @@ -2610,7 +2628,7 @@ static long shmem_fallocate(struct file *file, int mode, loff_t offset, DECLARE_WAIT_QUEUE_HEAD_ONSTACK(shmem_falloc_waitq); /* protected by i_mutex */ - if (info->seals & F_SEAL_WRITE) { + if (info->seals & (F_SEAL_WRITE | F_SEAL_FUTURE_WRITE)) { error = -EPERM; goto out; } -- 2.19.1.1215.g8438c0b245-goog

6 years, 5 months

7
15
0 0

[PATCH][next] tools/bpf: fix spelling mistake "memeory" -> "memory"

by Colin King

From: Colin Ian King <colin.king(a)canonical.com> The CHECK message contains a spelling mistake, fix it. Signed-off-by: Colin Ian King <colin.king(a)canonical.com> --- tools/testing/selftests/bpf/test_btf.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/bpf/test_btf.c b/tools/testing/selftests/bpf/test_btf.c index 7b1b160d6e67..bcbda7037840 100644 --- a/tools/testing/selftests/bpf/test_btf.c +++ b/tools/testing/selftests/bpf/test_btf.c @@ -2573,7 +2573,7 @@ static int do_test_file(unsigned int test_num) } func_info = malloc(info.func_info_cnt * rec_size); - if (CHECK(!func_info, "out of memeory")) { + if (CHECK(!func_info, "out of memory")) { err = -1; goto done; } @@ -3299,7 +3299,7 @@ static int do_test_func_type(int test_num) } func_info = malloc(info.func_info_cnt * rec_size); - if (CHECK(!func_info, "out of memeory")) { + if (CHECK(!func_info, "out of memory")) { err = -1; goto done; } -- 2.19.1

6 years, 5 months

2
1
0 0

selftests/bpf :get_cgroup_id_user: File not found: /sys/kernel/debug/tracing/events/syscalls/sys_enter_nanosleep/id

by Naresh Kamboju

Kselftest bpf get_cgroup_id_user is failed on all devices. selftests: bpf: get_cgroup_id_user main:PASS:setup_cgroup_environment main:PASS:create_and_get_cgroup main:PASS:join_cgroup main:PASS:bpf_prog_load main:PASS:bpf_find_map main:PASS:bpf_find_map main:FAIL:open err -1 errno 2 not ok 1..15 selftests: bpf: get_cgroup_id_user [FAIL] The strace output shows, This expected file not found, "/sys/kernel/debug/tracing/events/syscalls/sys_enter_nanosleep/id" bpf(BPF_MAP_UPDATE_ELEM, {map_fd=5, key=0x7fff0c68c138, value=0x7fff0c68c13c, flags=BPF_ANY}, 72) = 0 open(\"/sys/kernel/debug/tracing/events/syscalls/sys_enter_nanosleep/id\", O_RDONLY) = -1 ENOENT (No such file or directory) write(1, \"main:FAIL:open err -1 errno 2\n\", 30main:FAIL:open err -1 errno 2 Am I missing any pre-requirement ? Best regards Naresh Kamboju

6 years, 5 months

3
2
0 0

[PATCH v2] sysctl: kselftests: fix test_modprobe issue

by Lei Yang

when CONFIG_TEST_SYSCTL=y, there is no "/sys/module/test_sysctl/" when CONFIG_TEST_SYSCTL=m, checking /sys/module/test_sysctl/ is before kernel module loading you'll get below error message root@intel-x86-64:/tmp/sysctl# ./sysctl.sh Checking production write strict setting ... ok ./sysctl.sh: /sys/module/test_sysctl/ not present You must have the following enabled in your kernel: This patch will fix this issue. when CONFIG_TEST_SYSCTL=y, it has no chance to check "/sys/module/test_sysctl/" when CONFIG_TEST_SYSCTL=m, it will load kernel module first before checking sys interface. Signed-off-by: Lei Yang <Lei.Yang(a)windriver.com> --- tools/testing/selftests/sysctl/sysctl.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/sysctl/sysctl.sh b/tools/testing/selftests/sysctl/sysctl.sh index 584eb8e..08dc995 100755 --- a/tools/testing/selftests/sysctl/sysctl.sh +++ b/tools/testing/selftests/sysctl/sysctl.sh @@ -120,6 +120,7 @@ test_reqs() function load_req_mod() { + trap "test_modprobe" EXIT if [ ! -d $DIR ]; then if ! modprobe -q -n $TEST_DRIVER; then echo "$0: module $TEST_DRIVER not found [SKIP]" @@ -770,7 +771,6 @@ function parse_args() test_reqs allow_user_defaults check_production_sysctl_writes_strict -test_modprobe load_req_mod trap "test_finish" EXIT -- 1.9.1

6 years, 5 months

3
2
0 0

selftests/bpf: libbpf: object file doesn't contain bpf program

by Naresh Kamboju

The listed bpf test case failed due to, libbpf: object file doesn't contain bpf program We are cross compiling, installing selftests on device under test and running tests by using run_kselftest.sh script file. selftests: bpf: test_maps libbpf: object file doesn't contain bpf program Failed to load SK_SKB parse prog not ok 1..3 selftests: bpf: test_maps [FAIL] selftests: bpf: test_progs libbpf: object file doesn't contain bpf program libbpf: object file doesn't contain bpf program libbpf: object file doesn't contain bpf program libbpf: object file doesn't contain bpf program libbpf: ./test_tcp_estats.o doesn't provide kernel version libbpf: object file doesn't contain bpf program selftests: bpf: test_tcpbpf_user libbpf: object file doesn't contain bpf program FAILED: load_bpf_file failed for: test_tcpbpf_kern.o not ok 1..10 selftests: bpf: test_tcpbpf_user [FAIL] selftests: bpf: test_tcpnotify_user libbpf: object file doesn't contain bpf program FAILED: load_bpf_file failed for: test_tcpnotify_kern.o not ok 1..21 selftests: bpf: test_tcpnotify_user [FAIL] selftests: bpf: test_flow_dissector.sh bpffs not mounted. Mounting... libbpf: object file doesn't contain bpf program ./flow_dissector_load: bpf_prog_load bpf_flow.o selftests: test_flow_dissector [FAILED] Do you see problem when running on target devices ? Best regards Naresh Kamboju

6 years, 5 months

1
0
0 0

[PATCH -manpage 1/2] fcntl.2: Update manpage with new memfd F_SEAL_FUTURE_WRITE seal

by Joel Fernandes (Google)

More details of the seal can be found in the LKML patch: https://lore.kernel.org/lkml/20181120052137.74317-1-joel@joelfernandes.org/… Signed-off-by: Joel Fernandes (Google) <joel(a)joelfernandes.org> --- man2/fcntl.2 | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/man2/fcntl.2 b/man2/fcntl.2 index 03533d65b49d..54772f94964c 100644 --- a/man2/fcntl.2 +++ b/man2/fcntl.2 @@ -1525,6 +1525,21 @@ Furthermore, if there are any asynchronous I/O operations .RB ( io_submit (2)) pending on the file, all outstanding writes will be discarded. +.TP +.BR F_SEAL_FUTURE_WRITE +If this seal is set, the contents of the file can be modified only from +existing writeable mappings that were created prior to the seal being set. +Any attempt to create a new writeable mapping on the memfd via +.BR mmap (2) +will fail with +.BR EPERM. +Also any attempts to write to the memfd via +.BR write (2) +will fail with +.BR EPERM. +This is useful in situations where existing writable mapped regions need to be +kept intact while preventing any future writes. For example, to share a +read-only memory buffer to other processes that only the sender can write to. .\" .SS File read/write hints Write lifetime hints can be used to inform the kernel about the relative -- 2.19.1.1215.g8438c0b245-goog

6 years, 5 months

1
1
0 0

[PATCH net-next 00/18] selftests: Add tests for VXLAN at an 802.1d bridge

by Ido Schimmel

Petr says: This patchset adds several tests for VXLAN attached to an 802.1d bridge and fixes a related bug. First patch #1 fixes a bug in propagating SKB already-forwarded marks over veth to bridges, where they are irrelevant. This bug causes the vxlan_bridge_1d test suite from this patchset to fail as the packets aren't forwarded by br2. In patches #2 and #3, lib.sh is extended to support network namespaces. The use of namespaces is necessitated by VXLAN, which allows only one VXLAN device with a given VNI per namespace. Thus to host full topology on a single box for selftests, the "remote" endpoints need to be in namespaces. In patches #4-#6, lib.sh is extended in other ways to facilitate the following patches. In patches #7-#15, first the skeleton, and later the generic tests themselves are added. Patch #16 then adds another test that serves as a wrapper around the previous one, and runs it with a non-default port number. Patches #17 and #18 add mlxsw-specific tests. About those, Ido writes: The first test creates various configurations with regards to the VxLAN and bridge devices and makes sure the driver correctly forbids unsupported configuration and permits supported ones. It also verifies that the driver correctly sets the offload indication on FDB entries and the local route used for VxLAN decapsulation. The second test verifies that the driver correctly configures the singly linked list used to flood BUM traffic and that traffic is flooded as expected. Ido Schimmel (2): selftests: mlxsw: Add a test for VxLAN configuration selftests: mlxsw: Add a test for VxLAN flooding Petr Machata (16): net: skb_scrub_packet(): Scrub offload_fwd_mark selftests: forwarding: lib: Support NUM_NETIFS of 0 selftests: forwarding: lib: Add in_ns() selftests: forwarding: ping{6,}_test(): Add description argument selftests: forwarding: ping{6,}_do(): Allow passing ping arguments selftests: forwarding: lib: Add link_stats_rx_errors_get() selftests: forwarding: Add a skeleton of vxlan_bridge_1d selftests: forwarding: vxlan_bridge_1d: Add ping test selftests: forwarding: vxlan_bridge_1d: Add flood test selftests: forwarding: vxlan_bridge_1d: Add unicast test selftests: forwarding: vxlan_bridge_1d: Reconfigure & rerun tests selftests: forwarding: vxlan_bridge_1d: Add a TTL test selftests: forwarding: vxlan_bridge_1d: Add a TOS test selftests: forwarding: vxlan_bridge_1d: Add an ECN encap test selftests: forwarding: vxlan_bridge_1d: Add an ECN decap test selftests: forwarding: vxlan_bridge_1d_port_8472: New test net/core/skbuff.c | 5 + .../selftests/drivers/net/mlxsw/vxlan.sh | 664 +++++++++++++++++ .../drivers/net/mlxsw/vxlan_flooding.sh | 309 ++++++++ tools/testing/selftests/net/forwarding/lib.sh | 42 +- .../net/forwarding/vxlan_bridge_1d.sh | 678 ++++++++++++++++++ .../forwarding/vxlan_bridge_1d_port_8472.sh | 10 + 6 files changed, 1700 insertions(+), 8 deletions(-) create mode 100755 tools/testing/selftests/drivers/net/mlxsw/vxlan.sh create mode 100755 tools/testing/selftests/drivers/net/mlxsw/vxlan_flooding.sh create mode 100755 tools/testing/selftests/net/forwarding/vxlan_bridge_1d.sh create mode 100755 tools/testing/selftests/net/forwarding/vxlan_bridge_1d_port_8472.sh -- 2.19.1

6 years, 5 months

4
24
0 0

Re: [PATCH v4] selftests: membarrier: reorganized test for LTS supportability

by Shuah Khan

On 09/03/2018 01:04 PM, Rafael David Tinoco wrote: > This commit re-organizes membarrier test, solving issues when testing > LTS kernels. Now, the code: > > - always run the same amount of tests (even on older kernels). > - allows each test to succeed, fail or be skipped independently. > - allows testing features even when explicitly unsupported (force=1). > - able to consider different return codes for diff kernel versions. > - checks false positive/negative by checking ret code and errno. > - can be extended easily: to expand an array with commands. > > Link: https://bugs.linaro.org/show_bug.cgi?id=3771 > Signed-off-by: Rafael David Tinoco <rafael.tinoco(a)linaro.org> > --- > .../selftests/membarrier/membarrier_test.c | 482 +++++++++--------- > 1 file changed, 241 insertions(+), 241 deletions(-) > > diff --git a/tools/testing/selftests/membarrier/membarrier_test.c b/tools/testing/selftests/membarrier/membarrier_test.c > index 6793f8ecc8e7..151bc8a944a3 100644 > --- a/tools/testing/selftests/membarrier/membarrier_test.c > +++ b/tools/testing/selftests/membarrier/membarrier_test.c > @@ -1,6 +1,7 @@ > // SPDX-License-Identifier: GPL-2.0 > #define _GNU_SOURCE > #include <linux/membarrier.h> > +#include <sys/utsname.h> > #include <syscall.h> > #include <stdio.h> > #include <errno.h> > @@ -8,305 +9,304 @@ > > #include "../kselftest.h" > > -static int sys_membarrier(int cmd, int flags) > +#define ARRAY_SIZE(arr) (sizeof(arr) / sizeof((arr)[0])) > +#define KERNEL_VERSION(a, b, c) (((a) << 16) + ((b) << 8) + (c)) > + > +struct memb_tests { > + char testname[80]; > + int command; > + int flags; > + int exp_ret; > + int exp_errno; > + int enabled; > + int force; > + int force_exp_errno; > + int above; > + int bellow; > +}; > + > +struct memb_tests mbt[] = { > + { > + .testname = "cmd_fail\0", > + .command = -1, > + .exp_ret = -1, > + .exp_errno = EINVAL, > + .enabled = 1, > + }, > + { > + .testname = "cmd_flags_fail\0", > + .command = MEMBARRIER_CMD_QUERY, > + .flags = 1, > + .exp_ret = -1, > + .exp_errno = EINVAL, > + .enabled = 1, > + }, > + { > + .testname = "cmd_global_success\0", > + .command = MEMBARRIER_CMD_GLOBAL, > + .flags = 0, > + .exp_ret = 0, > + }, > + /* > + * PRIVATE EXPEDITED (forced) > + */ > + { > + .testname = "cmd_private_expedited_fail\0", > + .command = MEMBARRIER_CMD_PRIVATE_EXPEDITED, > + .flags = 0, > + .exp_ret = -1, > + .exp_errno = EPERM, > + .force = 1, > + .force_exp_errno = EINVAL, > + .bellow = KERNEL_VERSION(4, 10, 0), > + }, > + { > + .testname = "cmd_private_expedited_fail\0", > + .command = MEMBARRIER_CMD_PRIVATE_EXPEDITED, > + .flags = 0, > + .exp_ret = -1, > + .exp_errno = EPERM, > + .force = 1, > + .force_exp_errno = EPERM, > + .above = KERNEL_VERSION(4, 10, 0), > + }, > + { > + .testname = "cmd_register_private_expedited_success\0", > + .command = MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED, > + .flags = 0, > + .exp_ret = 0, > + .force = 1, > + .force_exp_errno = EINVAL, > + }, > + { > + .testname = "cmd_private_expedited_success\0", > + .command = MEMBARRIER_CMD_PRIVATE_EXPEDITED, > + .flags = 0, > + .exp_ret = 0, > + .force = 1, > + .force_exp_errno = EINVAL, > + }, > + /* > + * PRIVATE EXPEDITED SYNC CORE > + */ > + { > + .testname = "cmd_private_expedited_sync_core_fail\0", > + .command = MEMBARRIER_CMD_PRIVATE_EXPEDITED_SYNC_CORE, > + .flags = 0, > + .exp_ret = -1, > + .exp_errno = EPERM, > + }, > + { > + .testname = "cmd_register_private_expedited_sync_core_success\0", > + .command = MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED_SYNC_CORE, > + .flags = 0, > + .exp_ret = 0, > + }, > + { > + .testname = "cmd_private_expedited_sync_core_success\0", > + .command = MEMBARRIER_CMD_PRIVATE_EXPEDITED, > + .flags = 0, > + .exp_ret = 0, > + }, > + /* > + * GLOBAL EXPEDITED > + * global membarrier from a non-registered process is valid > + */ > + { > + .testname = "cmd_global_expedited_success\0", > + .command = MEMBARRIER_CMD_GLOBAL_EXPEDITED, > + .flags = 0, > + .exp_ret = 0, > + }, > + { > + .testname = "cmd_register_global_expedited_success\0", > + .command = MEMBARRIER_CMD_REGISTER_GLOBAL_EXPEDITED, > + .flags = 0, > + .exp_ret = 0, > + }, > + { > + .testname = "cmd_global_expedited_success\0", > + .command = MEMBARRIER_CMD_GLOBAL_EXPEDITED, > + .flags = 0, > + .exp_ret = 0, > + }, > +}; > + > +static void > +info_passed_ok(struct memb_tests test) > { > - return syscall(__NR_membarrier, cmd, flags); > + ksft_test_result_pass("sys_membarrier(): %s succeeded.\n", > + test.testname); > } > Why do we need to add new routines for these conditions. Why can't handle these strings in array. For example you can define an array of strings for passed unexpectedly etc. and the pass the string to appropriate ksft_* interface instead of adding of these routines. Also it is hard to review the code this way. I do like the direction though. Also please run get_maintainer.pl and cc everybody it suggests. thanks, -- Shuah

6 years, 5 months

2
2
0 0

KVM selftests are failing

by Ahmed Soliman

Hello, KVM Self tests located at tools/testing/selftests/kvm seams to be failing. I have tried: - dirty_log_test - x86_64/cr4_cpuid_sync_test - x86_64/evmcs_test - x86_64/platform_info_test - x86_64/set_sregs_test - x86_64/state_test - x86_64/sync_regs_test - x86_64/vmx_tsc_adjust_test All of them triggered the following assert: ==== Test Assertion Failure ==== lib/kvm_util.c:596: ret == 0 pid=24507 tid=24507 - Invalid argument 1 0x000056360774e468: ?? ??:0 2 0x000056360774e609: ?? ??:0 3 0x00005636077537a0: ?? ??:0 4 0x000056360774d35e: ?? ??:0 5 0x00007f3117571ae6: ?? ??:0 6 0x000056360774d4a9: ?? ??:0 madvise failed, addr: 0x7f3117350000 length: 0x200000 src_type: 0 I have tried the tests on both kvm(next) branch commit 22a7cdcae6a4a3c89748 as well as torvald's master 4.20-rc2. -- Ahmed, Junior Researcher, IoT and Cyber Security lab, SmartCI, Alexandria University, & CIS @ VMI

6 years, 5 months

5
14
0 0

[PATCH v3 6/7] kselftests: cgroup: add freezer controller self-tests

by Roman Gushchin

This patch implements six tests for the freezer controller for cgroup v2: 1) a simple test, which aims to freeze and unfreeze a cgroup with 100 processes 2) a more complicated tree test, which creates a hierarchy of cgroups, puts some processes in some cgroups, and tries to freeze and unfreeze different parts of the subtree 3) a forkbomb test: the test aims to freeze a forkbomb running in a cgroup, kill all tasks in the cgroup and remove the cgroup without the unfreezing. 4) rmdir test: the test creates two nested cgroups, freezes the parent one, checks that the child can be successfully removed, and a new child can be created 5) migration tests: the test checks migration of a task between frozen cgroups: from a frozen to a running, from a running to a frozen, and from a frozen to a frozen. 6) ptrace test: the test checks that it's possible to attach to a process in a frozen cgroup, get some information and detach, and the cgroup will remain frozen. Expected output: $ ./test_freezer ok 1 test_cgfreezer_simple ok 2 test_cgfreezer_tree ok 3 test_cgfreezer_forkbomb ok 4 test_cgrreezer_rmdir ok 5 test_cgfreezer_migrate ok 6 test_cgfreezer_ptrace Signed-off-by: Roman Gushchin <guro(a)fb.com> Cc: Shuah Khan <shuah(a)kernel.org> Cc: Tejun Heo <tj(a)kernel.org> Cc: kernel-team(a)fb.com Cc: linux-kselftest(a)vger.kernel.org --- tools/testing/selftests/cgroup/.gitignore | 1 + tools/testing/selftests/cgroup/Makefile | 2 + tools/testing/selftests/cgroup/cgroup_util.c | 81 ++- tools/testing/selftests/cgroup/cgroup_util.h | 7 + tools/testing/selftests/cgroup/test_freezer.c | 685 ++++++++++++++++++ 5 files changed, 775 insertions(+), 1 deletion(-) create mode 100644 tools/testing/selftests/cgroup/test_freezer.c diff --git a/tools/testing/selftests/cgroup/.gitignore b/tools/testing/selftests/cgroup/.gitignore index adacda50a4b2..7f9835624793 100644 --- a/tools/testing/selftests/cgroup/.gitignore +++ b/tools/testing/selftests/cgroup/.gitignore @@ -1,2 +1,3 @@ test_memcontrol test_core +test_freezer diff --git a/tools/testing/selftests/cgroup/Makefile b/tools/testing/selftests/cgroup/Makefile index 23fbaa4a9630..8d369b6a2069 100644 --- a/tools/testing/selftests/cgroup/Makefile +++ b/tools/testing/selftests/cgroup/Makefile @@ -5,8 +5,10 @@ all: TEST_GEN_PROGS = test_memcontrol TEST_GEN_PROGS += test_core +TEST_GEN_PROGS += test_freezer include ../lib.mk $(OUTPUT)/test_memcontrol: cgroup_util.c $(OUTPUT)/test_core: cgroup_util.c +$(OUTPUT)/test_freezer: cgroup_util.c diff --git a/tools/testing/selftests/cgroup/cgroup_util.c b/tools/testing/selftests/cgroup/cgroup_util.c index eba06f94433b..e9cdad673901 100644 --- a/tools/testing/selftests/cgroup/cgroup_util.c +++ b/tools/testing/selftests/cgroup/cgroup_util.c @@ -74,6 +74,16 @@ char *cg_name_indexed(const char *root, const char *name, int index) return ret; } +char *cg_control(const char *cgroup, const char *control) +{ + size_t len = strlen(cgroup) + strlen(control) + 2; + char *ret = malloc(len); + + snprintf(ret, len, "%s/%s", cgroup, control); + + return ret; +} + int cg_read(const char *cgroup, const char *control, char *buf, size_t len) { char path[PATH_MAX]; @@ -196,7 +206,59 @@ int cg_create(const char *cgroup) return mkdir(cgroup, 0644); } -static int cg_killall(const char *cgroup) +int cg_for_all_procs(const char *cgroup, int (*fn)(int pid, void *arg), + void *arg) +{ + char buf[PAGE_SIZE]; + char *ptr = buf; + int ret; + + if (cg_read(cgroup, "cgroup.procs", buf, sizeof(buf))) + return -1; + + while (ptr < buf + sizeof(buf)) { + int pid = strtol(ptr, &ptr, 10); + + if (pid == 0) + break; + if (*ptr) + ptr++; + else + break; + ret = fn(pid, arg); + if (ret) + return ret; + } + + return 0; +} + +int cg_wait_for_proc_count(const char *cgroup, int count) +{ + char buf[10 * PAGE_SIZE] = {0}; + int attempts; + char *ptr; + + for (attempts = 10; attempts >= 0; attempts--) { + int nr = 0; + + if (cg_read(cgroup, "cgroup.procs", buf, sizeof(buf))) + break; + + for (ptr = buf; *ptr; ptr++) + if (*ptr == '\n') + nr++; + + if (nr >= count) + return 0; + + usleep(100000); + } + + return -1; +} + +int cg_killall(const char *cgroup) { char buf[PAGE_SIZE]; char *ptr = buf; @@ -238,6 +300,14 @@ int cg_destroy(const char *cgroup) return ret; } +int cg_enter(const char *cgroup, int pid) +{ + char pidbuf[64]; + + snprintf(pidbuf, sizeof(pidbuf), "%d", pid); + return cg_write(cgroup, "cgroup.procs", pidbuf); +} + int cg_enter_current(const char *cgroup) { char pidbuf[64]; @@ -367,3 +437,12 @@ int set_oom_adj_score(int pid, int score) close(fd); return 0; } + +char proc_read_text(int pid, const char *item, char *buf, size_t size) +{ + char path[PATH_MAX]; + + snprintf(path, sizeof(path), "/proc/%d/%s", pid, item); + + return read_text(path, buf, size); +} diff --git a/tools/testing/selftests/cgroup/cgroup_util.h b/tools/testing/selftests/cgroup/cgroup_util.h index 9ac8b7958f83..8ee63c00a668 100644 --- a/tools/testing/selftests/cgroup/cgroup_util.h +++ b/tools/testing/selftests/cgroup/cgroup_util.h @@ -18,6 +18,7 @@ static inline int values_close(long a, long b, int err) extern int cg_find_unified_root(char *root, size_t len); extern char *cg_name(const char *root, const char *name); extern char *cg_name_indexed(const char *root, const char *name, int index); +extern char *cg_control(const char *cgroup, const char *control); extern int cg_create(const char *cgroup); extern int cg_destroy(const char *cgroup); extern int cg_read(const char *cgroup, const char *control, @@ -32,6 +33,7 @@ extern int cg_write(const char *cgroup, const char *control, char *buf); extern int cg_run(const char *cgroup, int (*fn)(const char *cgroup, void *arg), void *arg); +extern int cg_enter(const char *cgroup, int pid); extern int cg_enter_current(const char *cgroup); extern int cg_run_nowait(const char *cgroup, int (*fn)(const char *cgroup, void *arg), @@ -41,3 +43,8 @@ extern int alloc_pagecache(int fd, size_t size); extern int alloc_anon(const char *cgroup, void *arg); extern int is_swap_enabled(void); extern int set_oom_adj_score(int pid, int score); +extern int cg_for_all_procs(const char *cgroup, int (*fn)(int pid, void *arg), + void *arg); +extern int cg_wait_for_proc_count(const char *cgroup, int count); +extern int cg_killall(const char *cgroup); +extern char proc_read_text(int pid, const char *item, char *buf, size_t size); diff --git a/tools/testing/selftests/cgroup/test_freezer.c b/tools/testing/selftests/cgroup/test_freezer.c new file mode 100644 index 000000000000..0c4106658360 --- /dev/null +++ b/tools/testing/selftests/cgroup/test_freezer.c @@ -0,0 +1,685 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#include <stdbool.h> +#include <linux/limits.h> +#include <sys/ptrace.h> +#include <sys/types.h> +#include <sys/mman.h> +#include <unistd.h> +#include <stdio.h> +#include <errno.h> +#include <poll.h> +#include <stdlib.h> +#include <sys/inotify.h> +#include <string.h> +#include <sys/types.h> +#include <sys/wait.h> + +#include "../kselftest.h" +#include "cgroup_util.h" + +#define DEBUG +#ifdef DEBUG +#define debug(args...) fprintf(stderr, args) +#else +#define debug(args...) +#endif + +/* + * Freeze the given cgroup and wait for the inotify signal. + * If there is no signal in 10 seconds, treat this as an error. + */ +static int cg_freeze_wait(const char *cgroup, bool freeze) +{ + int fd, wd; + struct pollfd fds; + int ret = -1; + + fd = inotify_init1(IN_NONBLOCK); + if (fd == -1) + return fd; + + wd = inotify_add_watch(fd, cg_control(cgroup, "cgroup.events"), + IN_MODIFY); + if (wd == -1) { + close(fd); + return wd; + } + fds.fd = fd; + fds.events = POLLIN; + + ret = cg_write(cgroup, "cgroup.freeze", freeze ? "1" : "0"); + if (ret) { + close(fd); + return ret; + } + + while (true) { + wd = poll(&fds, 1, 10000); + + if (wd == -1 && errno == EINTR) + continue; + + if (wd == 1 && fds.revents & POLLIN) + ret = 0; + + break; + } + + close(fd); + + return ret; +} + +/* + * Check if the process is frozen and parked in a proper place. + */ +static int proc_check_frozen(int pid, void *arg) +{ + char buf[PAGE_SIZE]; + int len; + + len = proc_read_text(pid, "stat", buf, sizeof(buf)); + if (len == -1) { + debug("Can't get %d stat\n", pid); + return -1; + } + + if (strstr(buf, "(test_freezer) S ") == NULL) { + debug("Process %d in the unexpected state: %s\n", pid, buf); + return -1; + } + + len = proc_read_text(pid, "stack", buf, sizeof(buf)); + if (len == -1) { + debug("Can't get stack of the process %d\n", pid); + return -1; + } + + if (strstr(buf, "[<0>] cgroup_enter_frozen") != buf) { + debug("Process %d has unexpected stacktrace: %s\n", pid, buf); + return -1; + } + + return 0; +} + +/* + * Check if the cgroup is frozen and all belonging processes are + * parked in a proper place. + */ +static int cg_check_frozen(const char *cgroup, bool frozen) +{ + if (frozen) { + /* + * Check the cgroup.events::frozen value. + */ + if (cg_read_strstr(cgroup, "cgroup.events", "frozen 1") != 0) { + debug("Cgroup %s isn't unexpectedly frozen\n", cgroup); + return -1; + } + + /* + * Check that all processes are parked in the proper place. + */ + if (cg_for_all_procs(cgroup, proc_check_frozen, NULL)) { + debug("Some processes of cgroup %s are not frozen\n", + cgroup); + return -1; + } + } else { + /* + * Check the cgroup.events::frozen value. + */ + if (cg_read_strstr(cgroup, "cgroup.events", "frozen 0") != 0) { + debug("Cgroup %s is unexpectedly frozen\n", cgroup); + return -1; + } + } + + return 0; +} + +/* + * A simple process running in a sleep loop until being + * re-parented. + */ +static int child_fn(const char *cgroup, void *arg) +{ + int ppid = getppid(); + + while (getppid() == ppid) + usleep(1000); + + return getppid() == ppid; +} + +/* + * A simple test for the cgroup freezer: populated the cgroup with 100 + * running processes and freeze it. Then unfreeze it. Then it kills all + * processes and destroys the cgroup. + */ +static int test_cgfreezer_simple(const char *root) +{ + int ret = KSFT_FAIL; + char *cgroup = NULL; + int i; + + cgroup = cg_name(root, "cg_test"); + if (!cgroup) + goto cleanup; + + if (cg_create(cgroup)) + goto cleanup; + + for (i = 0; i < 100; i++) + cg_run_nowait(cgroup, child_fn, NULL); + + if (cg_wait_for_proc_count(cgroup, 100)) + goto cleanup; + + if (cg_check_frozen(cgroup, false)) + goto cleanup; + + if (cg_freeze_wait(cgroup, true)) + goto cleanup; + + if (cg_check_frozen(cgroup, true)) + goto cleanup; + + if (cg_freeze_wait(cgroup, false)) + goto cleanup; + + if (cg_check_frozen(cgroup, false)) + goto cleanup; + + ret = KSFT_PASS; + +cleanup: + if (cgroup) + cg_destroy(cgroup); + free(cgroup); + return ret; +} + +/* + * The test creates the following hierarchy: + * A + * / / \ \ + * B E I K + * /\ | + * C D F + * | + * G + * | + * H + * + * with a process in C, H and 3 processes in K. + * Then it tries to freeze and unfreeze the whole tree. + */ +static int test_cgfreezer_tree(const char *root) +{ + char *cgroup[10] = {0}; + int ret = KSFT_FAIL; + int i; + + cgroup[0] = cg_name(root, "cg_test_A"); + if (!cgroup[0]) + goto cleanup; + + cgroup[1] = cg_name(cgroup[0], "cg_test_B"); + if (!cgroup[1]) + goto cleanup; + + cgroup[2] = cg_name(cgroup[1], "cg_test_C"); + if (!cgroup[2]) + goto cleanup; + + cgroup[3] = cg_name(cgroup[1], "cg_test_D"); + if (!cgroup[3]) + goto cleanup; + + cgroup[4] = cg_name(cgroup[0], "cg_test_E"); + if (!cgroup[4]) + goto cleanup; + + cgroup[5] = cg_name(cgroup[4], "cg_test_F"); + if (!cgroup[5]) + goto cleanup; + + cgroup[6] = cg_name(cgroup[5], "cg_test_G"); + if (!cgroup[6]) + goto cleanup; + + cgroup[7] = cg_name(cgroup[6], "cg_test_H"); + if (!cgroup[7]) + goto cleanup; + + cgroup[8] = cg_name(cgroup[0], "cg_test_I"); + if (!cgroup[8]) + goto cleanup; + + cgroup[9] = cg_name(cgroup[0], "cg_test_K"); + if (!cgroup[9]) + goto cleanup; + + for (i = 0; i < 10; i++) + if (cg_create(cgroup[i])) + goto cleanup; + + cg_run_nowait(cgroup[2], child_fn, NULL); + cg_run_nowait(cgroup[7], child_fn, NULL); + cg_run_nowait(cgroup[9], child_fn, NULL); + cg_run_nowait(cgroup[9], child_fn, NULL); + cg_run_nowait(cgroup[9], child_fn, NULL); + + /* + * Wait until all child processes will enter + * corresponding cgroups. + */ + + if (cg_wait_for_proc_count(cgroup[2], 1) || + cg_wait_for_proc_count(cgroup[7], 1) || + cg_wait_for_proc_count(cgroup[9], 3)) + goto cleanup; + + /* + * Freeze B. + */ + if (cg_freeze_wait(cgroup[1], true)) + goto cleanup; + + if (cg_check_frozen(cgroup[1], true)) + goto cleanup; + + /* + * Freeze F. + */ + if (cg_freeze_wait(cgroup[5], true)) + goto cleanup; + + if (cg_check_frozen(cgroup[5], true)) + goto cleanup; + + /* + * Freeze G. + */ + if (cg_freeze_wait(cgroup[6], true)) + goto cleanup; + + if (cg_check_frozen(cgroup[6], true)) + goto cleanup; + + /* + * Check that A and E are not frozen. + */ + if (cg_check_frozen(cgroup[0], false)) + goto cleanup; + + if (cg_check_frozen(cgroup[4], false)) + goto cleanup; + + /* + * Freeze A. Check that A, B and E are frozen. + */ + if (cg_freeze_wait(cgroup[0], true)) + goto cleanup; + + if (cg_check_frozen(cgroup[0], true)) + goto cleanup; + + if (cg_check_frozen(cgroup[1], true)) + goto cleanup; + + if (cg_check_frozen(cgroup[4], true)) + goto cleanup; + + /* + * Unfreeze B, F and G + */ + if (cg_freeze_wait(cgroup[1], false)) + goto cleanup; + + if (cg_freeze_wait(cgroup[5], false)) + goto cleanup; + + if (cg_freeze_wait(cgroup[6], false)) + goto cleanup; + + /* + * Check that C and H are still frozen. + */ + if (cg_check_frozen(cgroup[2], true)) + goto cleanup; + + if (cg_check_frozen(cgroup[7], true)) + goto cleanup; + + /* + * Unfreezing A failed. Check that A, C and K are not frozen. + */ + if (cg_freeze_wait(cgroup[0], false)) + goto cleanup; + + if (cg_check_frozen(cgroup[0], false)) + goto cleanup; + + if (cg_check_frozen(cgroup[2], false)) + goto cleanup; + + if (cg_check_frozen(cgroup[9], false)) + goto cleanup; + + ret = KSFT_PASS; + +cleanup: + for (i = 9; i >= 0 && cgroup[i]; i--) { + cg_destroy(cgroup[i]); + free(cgroup[i]); + } + + return ret; +} + +/* + * A fork bomb emulator. + */ +static int forkbomb_fn(const char *cgroup, void *arg) +{ + int ppid; + + fork(); + fork(); + + ppid = getppid(); + + while (getppid() == ppid) + usleep(1000); + + return getppid() == ppid; +} + +/* + * The test runs a fork bomb in a cgroup and tries to freeze it. + * Then it kills all processes and checks that cgroup isn't populated + * anymore. + */ +static int test_cgfreezer_forkbomb(const char *root) +{ + int ret = KSFT_FAIL; + char *cgroup = NULL; + + cgroup = cg_name(root, "cg_forkbomb_test"); + if (!cgroup) + goto cleanup; + + if (cg_create(cgroup)) + goto cleanup; + + cg_run_nowait(cgroup, forkbomb_fn, NULL); + + usleep(100000); + + if (cg_freeze_wait(cgroup, true)) + goto cleanup; + + if (cg_check_frozen(cgroup, true)) + goto cleanup; + + if (cg_killall(cgroup)) + goto cleanup; + + if (cg_wait_for_proc_count(cgroup, 0)) + goto cleanup; + + ret = KSFT_PASS; + +cleanup: + if (cgroup) + cg_destroy(cgroup); + free(cgroup); + return ret; +} + +/* + * The test creates two nested cgroups, freezes the parent + * and removes the child. Then it checks that the parent cgroup + * remains frozen and it's possible to create a new child + * without unfreezing. The new child is frozen too. + */ +static int test_cgfreezer_rmdir(const char *root) +{ + int ret = KSFT_FAIL; + char *parent, *child = NULL; + + parent = cg_name(root, "cg_test_A"); + if (!parent) + goto cleanup; + + child = cg_name(parent, "cg_test_B"); + if (!child) + goto cleanup; + + if (cg_create(parent)) + goto cleanup; + + if (cg_create(child)) + goto cleanup; + + if (cg_freeze_wait(parent, true)) + goto cleanup; + + if (cg_check_frozen(parent, true)) + goto cleanup; + + if (cg_destroy(child)) + goto cleanup; + + if (cg_check_frozen(parent, true)) + goto cleanup; + + if (cg_create(child)) + goto cleanup; + + if (cg_check_frozen(child, true)) + goto cleanup; + + ret = KSFT_PASS; + +cleanup: + if (child) + cg_destroy(child); + free(child); + if (parent) + cg_destroy(parent); + free(parent); + return ret; +} + +/* + * The test creates two cgroups: A and B. The it runs a process in A, + * and performs several migrations: + * 1) A (running) -> B (frozen) + * 2) B (frozen) -> A (running) + * 3) A (frozen) -> B (frozen) + * + * One each step it checks that the actual state of cgroups matches + * the expected state. + */ +static int test_cgfreezer_migrate(const char *root) +{ + int ret = KSFT_FAIL; + char *cgroup[2] = {0}; + int pid; + + cgroup[0] = cg_name(root, "cg_test_A"); + if (!cgroup[0]) + goto cleanup; + + cgroup[1] = cg_name(root, "cg_test_B"); + if (!cgroup[1]) + goto cleanup; + + if (cg_create(cgroup[0])) + goto cleanup; + + if (cg_create(cgroup[1])) + goto cleanup; + + pid = cg_run_nowait(cgroup[0], child_fn, NULL); + if (pid < 0) + goto cleanup; + + if (cg_wait_for_proc_count(cgroup[0], 1)) + goto cleanup; + + /* + * Migrate from A (running) to B (frozen) + */ + if (cg_freeze_wait(cgroup[1], true)) + goto cleanup; + + if (cg_check_frozen(cgroup[1], true)) + goto cleanup; + + if (cg_enter(cgroup[1], pid)) + goto cleanup; + + if (cg_check_frozen(cgroup[0], false)) + goto cleanup; + + if (cg_check_frozen(cgroup[1], true)) + goto cleanup; + + /* + * Migrate from B (frozen) to A (running) + */ + if (cg_enter(cgroup[0], pid)) + goto cleanup; + + if (cg_check_frozen(cgroup[0], false)) + goto cleanup; + + if (cg_check_frozen(cgroup[1], true)) + goto cleanup; + + /* + * Migrate from A (frozen) to B (frozen) + */ + if (cg_freeze_wait(cgroup[0], true)) + goto cleanup; + + if (cg_check_frozen(cgroup[0], true)) + goto cleanup; + + if (cg_enter(cgroup[1], pid)) + goto cleanup; + + if (cg_check_frozen(cgroup[0], true)) + goto cleanup; + + if (cg_check_frozen(cgroup[1], true)) + goto cleanup; + + ret = KSFT_PASS; + +cleanup: + if (cgroup[0]) + cg_destroy(cgroup[0]); + free(cgroup[0]); + if (cgroup[1]) + cg_destroy(cgroup[1]); + free(cgroup[1]); + return ret; +} + +/* + * The test checks that ptrace works with a tracing process in a frozen cgroup. + */ +static int test_cgfreezer_ptrace(const char *root) +{ + int ret = KSFT_FAIL; + char *cgroup = NULL; + siginfo_t siginfo; + int pid; + + cgroup = cg_name(root, "cg_test"); + if (!cgroup) + goto cleanup; + + if (cg_create(cgroup)) + goto cleanup; + + pid = cg_run_nowait(cgroup, child_fn, NULL); + if (pid < 0) + goto cleanup; + + if (cg_wait_for_proc_count(cgroup, 1)) + goto cleanup; + + if (cg_freeze_wait(cgroup, true)) + goto cleanup; + + if (cg_check_frozen(cgroup, true)) + goto cleanup; + + if (ptrace(PTRACE_SEIZE, pid, NULL, NULL)) + goto cleanup; + + if (ptrace(PTRACE_INTERRUPT, pid, NULL, NULL)) + goto cleanup; + + waitpid(pid, NULL, 0); + + if (ptrace(PTRACE_GETSIGINFO, pid, NULL, &siginfo)) + goto cleanup; + + if (ptrace(PTRACE_DETACH, pid, NULL, NULL)) + goto cleanup; + + ret = KSFT_PASS; + +cleanup: + if (cgroup) + cg_destroy(cgroup); + free(cgroup); + return ret; +} + +#define T(x) { x, #x } +struct cgfreezer_test { + int (*fn)(const char *root); + const char *name; +} tests[] = { + T(test_cgfreezer_simple), + T(test_cgfreezer_tree), + T(test_cgfreezer_forkbomb), + T(test_cgfreezer_rmdir), + T(test_cgfreezer_migrate), + T(test_cgfreezer_ptrace), +}; +#undef T + +int main(int argc, char *argv[]) +{ + char root[PATH_MAX]; + int i, ret = EXIT_SUCCESS; + + if (cg_find_unified_root(root, sizeof(root))) + ksft_exit_skip("cgroup v2 isn't mounted\n"); + for (i = 0; i < ARRAY_SIZE(tests); i++) { + switch (tests[i].fn(root)) { + case KSFT_PASS: + ksft_test_result_pass("%s\n", tests[i].name); + break; + case KSFT_SKIP: + ksft_test_result_skip("%s\n", tests[i].name); + break; + default: + ret = EXIT_FAILURE; + ksft_test_result_fail("%s\n", tests[i].name); + break; + } + } + + return ret; +} -- 2.17.2

6 years, 5 months

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-kselftest-mirror