aboutsummaryrefslogtreecommitdiff
path: root/tools
AgeCommit message (Collapse)AuthorFilesLines
2022-07-13KVM: selftests: Use vCPU's CPUID directly in Hyper-V testSean Christopherson3-89/+64
Use the vCPU's persistent CPUID array directly when manipulating the set of exposed Hyper-V CPUID features. Drop set_cpuid() to route all future modification through the vCPU helpers; the Hyper-V features test was the last user. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Use vcpu_get_cpuid_entry() in PV features test (sort of)Sean Christopherson3-11/+13
Add a new helper, vcpu_clear_cpuid_entry(), to do a RMW operation on the vCPU's CPUID model to clear a given CPUID entry, and use it to clear KVM's paravirt feature instead of operating on kvm_get_supported_cpuid()'s static "cpuid" variable. This also eliminates a user of the soon-be-defunct set_cpuid() helper. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Use vcpu_clear_cpuid_feature() in monitor_mwait_testSean Christopherson1-9/+1
Use vcpu_clear_cpuid_feature() to the MONITOR/MWAIT CPUID feature bit in the MONITOR/MWAIT quirk test. Signed-off-by: Sean Christopherson <[email protected]>
2022-07-13KVM: selftests: Add and use helper to set vCPU's CPUID maxphyaddrSean Christopherson3-9/+11
Add a helper to set a vCPU's guest.MAXPHYADDR, and use it in the test that verifies the emulator returns an error on an unknown instruction when KVM emulates in response to an EPT violation with a GPA that is legal in hardware but illegal with respect to the guest's MAXPHYADDR. Add a helper even though there's only a single user at this time. Before its removal, mmu_role_test also stuffed guest.MAXPHYADDR, and the helper provides a small amount of clarity. More importantly, this eliminates a set_cpuid() user and an instance of modifying kvm_get_supported_cpuid()'s static "cpuid". Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Use vm->pa_bits to generate reserved PA bitsSean Christopherson1-6/+2
Use vm->pa_bits to generate the mask of physical address bits that are reserved in page table entries. vm->pa_bits is set when the VM is created, i.e. it's guaranteed to be valid when populating page tables. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Add helpers to get and modify a vCPU's CPUID entriesSean Christopherson2-0/+48
Add helpers to get a specific CPUID entry for a given vCPU, and to toggle a specific CPUID-based feature for a vCPU. The helpers will reduce the amount of boilerplate code needed to tweak a vCPU's CPUID model, improve code clarity, and most importantly move tests away from modifying the static "cpuid" returned by kvm_get_supported_cpuid(). Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Use get_cpuid_entry() in kvm_get_supported_cpuid_index()Sean Christopherson2-39/+7
Use get_cpuid_entry() in kvm_get_supported_cpuid_index() to replace functionally identical code. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Rename and tweak get_cpuid() to get_cpuid_entry()Sean Christopherson3-10/+9
Rename get_cpuid() to get_cpuid_entry() to better reflect its behavior. Leave set_cpuid() as is to avoid unnecessary churn, that helper will soon be removed entirely. Oppurtunistically tweak the implementation to avoid using a temporary variable in anticipation of taggin the input @cpuid with "const". No functional change intended. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Don't use a static local in vcpu_get_supported_hv_cpuid()Sean Christopherson1-3/+1
Don't use a static variable for the Hyper-V supported CPUID array, the helper unconditionally reallocates the array on every invocation (and all callers free the array immediately after use). The array is intentionally recreated and refilled because the set of supported CPUID features is dependent on vCPU state, e.g. whether or not eVMCS has been enabled. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Cache CPUID in struct kvm_vcpuSean Christopherson11-50/+69
Cache a vCPU's CPUID information in "struct kvm_vcpu" to allow fixing the mess where tests, often unknowingly, modify the global/static "cpuid" allocated by kvm_get_supported_cpuid(). Add vcpu_init_cpuid() to handle stuffing an entirely different CPUID model, e.g. during vCPU creation or when switching to the Hyper-V enabled CPUID model. Automatically refresh the cache on vcpu_set_cpuid() so that any adjustments made by KVM are always reflected in the cache. Drop vcpu_get_cpuid() entirely to force tests to use the cache, and to allow adding e.g. vcpu_get_cpuid_entry() in the future without creating a conflicting set of APIs where vcpu_get_cpuid() does KVM_GET_CPUID2, but vcpu_get_cpuid_entry() does not. Opportunistically convert the VMX nested state test and KVM PV test to manipulating the vCPU's CPUID (because it's easy), but use vcpu_init_cpuid() for the Hyper-V features test and "emulator error" test to effectively retain their current behavior as they're less trivial to convert. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Split out kvm_cpuid2_size() from allocate_kvm_cpuid2()Sean Christopherson2-41/+30
Split out the computation of the effective size of a kvm_cpuid2 struct from allocate_kvm_cpuid2(), and modify both to take an arbitrary number of entries. Future commits will add caching of a vCPU's CPUID model, and will (a) be able to precisely size the entries array, and (b) will need to know the effective size of the struct in order to copy to/from the cache. Expose the helpers so that the Hyper-V Features test can use them in the (somewhat distant) future. The Hyper-V test very, very subtly relies on propagating CPUID info across vCPU instances, and will need to make a copy of the previous vCPU's CPUID information when it switches to using the per-vCPU cache. Alternatively, KVM could provide helpers to duplicate and/or copy a kvm_cpuid2 instance, but each is literally a single line of code if the helpers are exposed, and it's not like the size of kvm_cpuid2 is secret knowledge. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Verify that kvm_cpuid2.entries layout is unchanged by KVMSean Christopherson1-28/+21
In the CPUID test, verify that KVM doesn't modify the kvm_cpuid2.entries layout, i.e. that the order of entries and their flags is identical between what the test provides via KVM_SET_CPUID2 and what KVM returns via KVM_GET_CPUID2. Asserting that the layouts match simplifies the test as there's no need to iterate over both arrays. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Use kvm_cpu_has() for nSVM soft INT injection testSean Christopherson2-8/+2
Use kvm_cpu_has() to query for NRIPS support instead of open coding equivalent functionality using kvm_get_supported_cpuid_entry(). Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Use kvm_cpu_has() for KVM's PV steal timeSean Christopherson2-3/+23
Use kvm_cpu_has() in the stea-ltime test instead of open coding equivalent functionality using kvm_get_supported_cpuid_entry(). Opportunistically define all of KVM's paravirt CPUID-based features. No functional change intended. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Remove the obsolete/dead MMU role testSean Christopherson4-142/+0
Remove the MMU role test, which was made obsolete by KVM commit feb627e8d6f6 ("KVM: x86: Forbid KVM_SET_CPUID{,2} after KVM_RUN"). The ongoing costs of keeping the test updated far outweigh any benefits, e.g. the test _might_ be useful as an example or for documentation purposes, but otherwise the test is dead weight. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Use kvm_cpu_has() for XSAVE in cr4_cpuid_sync_testSean Christopherson1-3/+1
Use kvm_cpu_has() in the CR4/CPUID sync test instead of open coding equivalent functionality using kvm_get_supported_cpuid_entry(). No functional change intended. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Use kvm_cpu_has() in AMX testSean Christopherson2-11/+9
Use kvm_cpu_has() in the AMX test instead of open coding equivalent functionality using kvm_get_supported_cpuid_entry() and kvm_get_supported_cpuid_index(). No functional change intended. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Check for _both_ XTILE data and cfg in AMX testSean Christopherson1-1/+2
Check for _both_ XTILE data and cfg support in the AMX test instead of checking for _either_ feature. Practically speaking, no sane CPU or vCPU will support one but not the other, but the effective "or" behavior is subtle and technically incorrect. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Use kvm_cpu_has() for XSAVES in XSS MSR testSean Christopherson2-7/+2
Use kvm_cpu_has() in the XSS MSR test instead of open coding equivalent functionality using kvm_get_supported_cpuid_index(). No functional change intended. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Drop redundant vcpu_set_cpuid() from PMU selftestSean Christopherson1-3/+0
Drop a redundant vcpu_set_cpuid() from the PMU test. The vCPU's CPUID is set to KVM's supported CPUID by vm_create_with_one_vcpu(), which was also true back when the helper was named vm_create_default(). Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Use kvm_cpu_has() to query PDCM in PMU selftestSean Christopherson2-5/+3
Use kvm_cpu_has() in the PMU test to query PDCM support instead of open coding equivalent functionality using kvm_get_supported_cpuid_index(). No functional change intended. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Use kvm_cpu_has() for nested VMX checksSean Christopherson16-29/+15
Use kvm_cpu_has() to check for nested VMX support, and drop the helpers now that their functionality is trivial to implement. No functional change intended. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Use kvm_cpu_has() for nested SVM checksSean Christopherson8-21/+6
Use kvm_cpu_has() to check for nested SVM support, and drop the helpers now that their functionality is trivial to implement. No functional change intended. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Use kvm_cpu_has() in the SEV migration testSean Christopherson2-11/+4
Use kvm_cpu_has() in the SEV migration test instead of open coding equivalent functionality using kvm_get_supported_cpuid_entry(). No functional change intended. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Add framework to query KVM CPUID bitsSean Christopherson3-28/+128
Add X86_FEATURE_* magic in the style of KVM-Unit-Tests' implementation, where the CPUID function, index, output register, and output bit position are embedded in the macro value. Add kvm_cpu_has() to query KVM's supported CPUID and use it set_sregs_test, which is the most prolific user of manual feature querying. Opportunstically rename calc_cr4_feature_bits() to calc_supported_cr4_feature_bits() to better capture how the CR4 bits are chosen. Link: https://lore.kernel.org/all/[email protected] Suggested-by: Paolo Bonzini <[email protected]> Suggested-by: Jim Mattson <[email protected]> Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: sefltests: Use CPUID_* instead of X86_FEATURE_* for one-off usageSean Christopherson5-15/+12
Rename X86_FEATURE_* macros to CPUID_* in various tests to free up the X86_FEATURE_* names for KVM-Unit-Tests style CPUID automagic where the function, leaf, register, and bit for the feature is embedded in its macro value. No functional change intended. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Set KVM's supported CPUID as vCPU's CPUID during recreateSean Christopherson7-7/+26
On x86-64, set KVM's supported CPUID as the vCPU's CPUID when recreating a VM+vCPU to deduplicate code for state save/restore tests, and to provide symmetry of sorts with respect to vm_create_with_one_vcpu(). The extra KVM_SET_CPUID2 call is wasteful for Hyper-V, but ultimately is nothing more than an expensive nop, and overriding the vCPU's CPUID with the Hyper-V CPUID information is the only known scenario where a state save/restore test wouldn't need/want the default CPUID. Opportunistically use __weak for the default vm_compute_max_gfn(), it's provided by tools' compiler.h. Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Fix filename reporting in guest assertsColton Lewis41-102/+68
Fix filename reporting in guest asserts by ensuring the GUEST_ASSERT macro records __FILE__ and substituting REPORT_GUEST_ASSERT for many repetitive calls to TEST_FAIL. Previously filename was reported by using __FILE__ directly in the selftest, wrongly assuming it would always be the same as where the assertion failed. Signed-off-by: Colton Lewis <[email protected]> Reported-by: Ricardo Koller <[email protected]> Fixes: 4e18bccc2e5544f0be28fc1c4e6be47a469d6c60 Link: https://lore.kernel.org/r/[email protected] [sean: convert more TEST_FAIL => REPORT_GUEST_ASSERT instances] Signed-off-by: Sean Christopherson <[email protected]>
2022-07-13KVM: selftests: Write REPORT_GUEST_ASSERT macros to pair with GUEST_ASSERTColton Lewis1-0/+42
Write REPORT_GUEST_ASSERT macros to pair with GUEST_ASSERT to abstract and make consistent all guest assertion reporting. Every report includes an explanatory string, a filename, and a line number. Signed-off-by: Colton Lewis <[email protected]> Reviewed-by: Andrew Jones <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Sean Christopherson <[email protected]>
2022-07-13KVM: selftests: Increase UCALL_MAX_ARGS to 7Colton Lewis1-1/+1
Increase UCALL_MAX_ARGS to 7 to allow GUEST_ASSERT_4 to pass 3 builtin ucall arguments specified in guest_assert_builtin_args plus 4 user-specified arguments. Signed-off-by: Colton Lewis <[email protected]> Reviewed-by: Andrew Jones <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Sean Christopherson <[email protected]>
2022-07-13KVM: selftests: enumerate GUEST_ASSERT argumentsColton Lewis1-0/+8
Enumerate GUEST_ASSERT arguments to avoid magic indices to ucall.args. Signed-off-by: Colton Lewis <[email protected]> Reviewed-by: Andrew Jones <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Sean Christopherson <[email protected]>
2022-07-13KVM: x86: Tweak name of MONITOR/MWAIT #UD quirk to make it #UD specificSean Christopherson1-1/+1
Add a "UD" clause to KVM_X86_QUIRK_MWAIT_NEVER_FAULTS to make it clear that the quirk only controls the #UD behavior of MONITOR/MWAIT. KVM doesn't currently enforce fault checks when MONITOR/MWAIT are supported, but that could change in the future. SVM also has a virtualization hole in that it checks all faults before intercepts, and so "never faults" is already a lie when running on SVM. Fixes: bfbcc81bb82c ("KVM: x86: Add a quirk for KVM's "MONITOR/MWAIT are NOPs!" behavior") Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13KVM: selftests: Use "a" and "d" to set EAX/EDX for wrmsr_safe()Vitaly Kuznetsov1-1/+1
Do not use GCC's "A" constraint to load EAX:EDX in wrmsr_safe(). Per GCC's documenation on x86-specific constraints, "A" will not actually load a 64-bit value into EAX:EDX on x86-64. The a and d registers. This class is used for instructions that return double word results in the ax:dx register pair. Single word values will be allocated either in ax or dx. For example on i386 the following implements rdtsc: unsigned long long rdtsc (void) { unsigned long long tick; __asm__ __volatile__("rdtsc":"=A"(tick)); return tick; } This is not correct on x86-64 as it would allocate tick in either ax or dx. You have to use the following variant instead: unsigned long long rdtsc (void) { unsigned int tickl, tickh; __asm__ __volatile__("rdtsc":"=a"(tickl),"=d"(tickh)); return ((unsigned long long)tickh << 32)|tickl; } Because a u64 fits in a single 64-bit register, using "A" for selftests, which are 64-bit only, results in GCC loading the value into either RAX or RDX instead of splitting it across EAX:EDX. E.g.: kvm_exit: reason MSR_WRITE rip 0x402919 info 0 0 kvm_msr: msr_write 40000118 = 0x60000000001 (#GP) ... With "A": 48 8b 43 08 mov 0x8(%rbx),%rax 49 b9 ba da ca ba 0a movabs $0xabacadaba,%r9 00 00 00 4c 8d 15 07 00 00 00 lea 0x7(%rip),%r10 # 402f44 <guest_msr+0x34> 4c 8d 1d 06 00 00 00 lea 0x6(%rip),%r11 # 402f4a <guest_msr+0x3a> 0f 30 wrmsr With "a"/"d": 48 8b 53 08 mov 0x8(%rbx),%rdx 89 d0 mov %edx,%eax 48 c1 ea 20 shr $0x20,%rdx 49 b9 ba da ca ba 0a movabs $0xabacadaba,%r9 00 00 00 4c 8d 15 07 00 00 00 lea 0x7(%rip),%r10 # 402fc3 <guest_msr+0xb3> 4c 8d 1d 06 00 00 00 lea 0x6(%rip),%r11 # 402fc9 <guest_msr+0xb9> 0f 30 wrmsr Fixes: 3b23054cd3f5 ("KVM: selftests: Add x86-64 support for exception fixup") Signed-off-by: Vitaly Kuznetsov <[email protected]> Link: https://gcc.gnu.org/onlinedocs/gcc/Machine-Constraints.html#Machine-Constraints [sean: use "& -1u", provide GCC blurb and link to documentation] Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-13libbpf: Error out when binary_path is NULL for uprobe and USDTHengqi Chen1-6/+7
binary_path is a required non-null parameter for bpf_program__attach_usdt and bpf_program__attach_uprobe_opts. Check it against NULL to prevent coredump on strchr. Signed-off-by: Hengqi Chen <[email protected]> Signed-off-by: Andrii Nakryiko <[email protected]> Link: https://lore.kernel.org/bpf/[email protected]
2022-07-12selftests: mptcp: add MPC backup testsPaolo Abeni1-0/+30
Add a couple of test-cases covering the newly introduced features - priority update for the MPC subflow. Reviewed-by: Mat Martineau <[email protected]> Signed-off-by: Paolo Abeni <[email protected]> Signed-off-by: Mat Martineau <[email protected]> Signed-off-by: Jakub Kicinski <[email protected]>
2022-07-12KVM: selftests: Provide valid inputs for MONITOR/MWAIT regsSean Christopherson1-2/+6
Provide valid inputs for RAX, RCX, and RDX when testing whether or not KVM injects a #UD on MONITOR/MWAIT. SVM has a virtualization hole and checks for _all_ faults before checking for intercepts, e.g. MONITOR with an unsupported RCX will #GP before KVM gets a chance to intercept and emulate. Fixes: 2325d4dd7321 ("KVM: selftests: Add MONITOR/MWAIT quirk test") Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-12KVM: selftests: Test MONITOR and MWAIT, not just MONITOR for quirkSean Christopherson1-1/+1
Fix a copy+paste error in monitor_mwait_test by switching one of the two "monitor" instructions to an "mwait". The intent of the test is very much to verify the quirk handles both MONITOR and MWAIT. Fixes: 2325d4dd7321 ("KVM: selftests: Add MONITOR/MWAIT quirk test") Reported-by: Yuan Yao <[email protected]> Signed-off-by: Sean Christopherson <[email protected]> Link: https://lore.kernel.org/r/[email protected]
2022-07-12selftests/bpf: add a ksym iter subtestAlan Maguire3-0/+97
add subtest verifying BPF ksym iter behaviour. The BPF ksym iter program shows an example of dumping a format different to /proc/kallsyms. It adds KIND and MAX_SIZE fields which represent the kind of symbol (core kernel, module, ftrace, bpf, or kprobe) and the maximum size the symbol can be. The latter is calculated from the difference between current symbol value and the next symbol value. The key benefit for this iterator will likely be supporting in-kernel data-gathering rather than dumping symbol details to userspace and parsing the results. Signed-off-by: Alan Maguire <[email protected]> Acked-by: Yonghong Song <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Alexei Starovoitov <[email protected]>
2022-07-12perf lock: Look up callchain for the contended locksNamhyung Kim1-4/+156
The lock contention tracepoints don't provide lock names. All we can do is to get stack traces and show the caller instead. To minimize the overhead it's limited to up to 8 stack traces and display the first non-lock function symbol name as a caller. $ perf lock report -F acquired,contended,avg_wait,wait_total Name acquired contended avg wait total wait update_blocked_a... 40 40 3.61 us 144.45 us kernfs_fop_open+... 5 5 3.64 us 18.18 us _nohz_idle_balance 3 3 2.65 us 7.95 us tick_do_update_j... 1 1 6.04 us 6.04 us ep_scan_ready_list 1 1 3.93 us 3.93 us ... Signed-off-by: Namhyung Kim <[email protected]> Acked-by: Ian Rogers <[email protected]> Cc: Boqun Feng <[email protected]> Cc: Davidlohr Bueso <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Jiri Olsa <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Waiman Long <[email protected]> Cc: Will Deacon <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2022-07-12perf record: Allow to specify max stack depth of fp callchainNamhyung Kim2-6/+17
Currently it has no interface to specify the max stack depth for perf record. Extend the command line parameter to accept a number after 'fp' to specify the depth like '--call-graph fp,32'. Signed-off-by: Namhyung Kim <[email protected]> Cc: Boqun Feng <[email protected]> Cc: Davidlohr Bueso <[email protected]> Cc: Ian Rogers <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Jiri Olsa <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Waiman Long <[email protected]> Cc: Will Deacon <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2022-07-12perf lock: Handle lock contention tracepointsNamhyung Kim1-0/+137
When the lock contention events are used, there's no tracking of acquire and release. So the state machine is simplified to use UNINITIALIZED -> CONTENDED -> ACQUIRED only. Note that CONTENDED state is re-entrant since mutex locks can hit two or more consecutive contention_begin events for optimistic spinning and sleep. Signed-off-by: Namhyung Kim <[email protected]> Acked-by: Ian Rogers <[email protected]> Cc: Boqun Feng <[email protected]> Cc: Davidlohr Bueso <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Jiri Olsa <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Waiman Long <[email protected]> Cc: Will Deacon <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2022-07-12perf lock: Add lock contention tracepoints record supportNamhyung Kim1-7/+69
When LOCKDEP and LOCK_STAT events are not available, it falls back to record the new lock contention tracepoints. Signed-off-by: Namhyung Kim <[email protected]> Cc: Boqun Feng <[email protected]> Cc: Davidlohr Bueso <[email protected]> Cc: Ian Rogers <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Jiri Olsa <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Waiman Long <[email protected]> Cc: Will Deacon <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2022-07-12perf lock: Skip print_bad_events() if nothing badNamhyung Kim1-0/+7
The debug output is meaningful when there are bad lock sequences. Skip it unless there's one or -v option is given. Signed-off-by: Namhyung Kim <[email protected]> Cc: Boqun Feng <[email protected]> Cc: Davidlohr Bueso <[email protected]> Cc: Ian Rogers <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Jiri Olsa <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Waiman Long <[email protected]> Cc: Will Deacon <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2022-07-12perf lock: Allow to use different kernel symbolsNamhyung Kim2-0/+11
Add --vmlinux and --kallsyms options to support data file from different kernels. Signed-off-by: Namhyung Kim <[email protected]> Cc: Boqun Feng <[email protected]> Cc: Davidlohr Bueso <[email protected]> Cc: Ian Rogers <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Jiri Olsa <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Waiman Long <[email protected]> Cc: Will Deacon <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2022-07-12perf lock: Print wait times with unitNamhyung Kim1-8/+40
Currently it only prints the time in nsec but it's a bit hard to read and takes longer in the screen. We can change it to use different units and keep the number small to save the space. Before: $ perf lock report Name acquired contended avg wait (ns) total wait (ns) max wait (ns) min wait (ns) jiffies_lock 433 32 2778 88908 13570 692 &lruvec->lru_lock 747 5 11254 56272 18317 1412 slock-AF_INET6 7 1 23543 23543 23543 23543 &newf->file_lock 706 15 1025 15388 2279 618 slock-AF_INET6 8 1 10379 10379 10379 10379 &rq->__lock 2143 5 2037 10185 3462 939 After: Name acquired contended avg wait total wait max wait min wait jiffies_lock 433 32 2.78 us 88.91 us 13.57 us 692 ns &lruvec->lru_lock 747 5 11.25 us 56.27 us 18.32 us 1.41 us slock-AF_INET6 7 1 23.54 us 23.54 us 23.54 us 23.54 us &newf->file_lock 706 15 1.02 us 15.39 us 2.28 us 618 ns slock-AF_INET6 8 1 10.38 us 10.38 us 10.38 us 10.38 us &rq->__lock 2143 5 2.04 us 10.19 us 3.46 us 939 ns Signed-off-by: Namhyung Kim <[email protected]> Acked-by: Ian Rogers <[email protected]> Cc: Boqun Feng <[email protected]> Cc: Davidlohr Bueso <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Jiri Olsa <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Waiman Long <[email protected]> Cc: Will Deacon <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2022-07-12perf test: Add test for branch stack samplingGerman Gomez1-0/+114
Add a self test for branch stack sampling, to check that we get the expected branch types, and filters behave as expected. Suggested-by: Anshuman Khandual <[email protected]> Signed-off-by: German Gomez <[email protected]> Acked-by: Namhyung Kim <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Ian Rogers <[email protected]> Cc: Mark Rutland <[email protected]> Link: https://lore.kernel.org/r/[email protected] Tested-by: Jiri Olsa <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2022-07-12perf test: Make all metrics test more tolerantIan Rogers1-16/+31
Metric names are truncated so don't try to match all of one. Allow AMX metrics to skip as floating point ones do. Metrics for optane memory can also skip rather than fail. Add a system wide check for uncore metrics. Restructure code to avoid extensive nesting. Some impetus for this in: https://lore.kernel.org/lkml/[email protected]/ Suggested-by: John Garry <[email protected]> Signed-off-by: Ian Rogers <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Andi Kleen <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Jiri Olsa <[email protected]> Cc: John Garry <[email protected]> Cc: Kan Liang <[email protected]> Cc: Mark Rutland <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Stephane Eranian <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2022-07-12perf test: Add debug line to diagnose broken metricsIan Rogers1-0/+1
Printing out the metric name and architecture makes finding the source of a failure easier. Signed-off-by: Ian Rogers <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Andi Kleen <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Jiri Olsa <[email protected]> Cc: John Garry <[email protected]> Cc: Kan Liang <[email protected]> Cc: Mark Rutland <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Stephane Eranian <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2022-07-11selftests: tls: add test for NoPad getsockoptJakub Kicinski1-0/+51
Make sure setsockopt / getsockopt behave as expected. Signed-off-by: Jakub Kicinski <[email protected]>
2022-07-11selftest: net: add tun to .gitignoreJakub Kicinski1-0/+1
Add missing .gitignore entry. Fixes: 839b92fede7b ("selftest: tun: add test for NAPI dismantle") Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Jakub Kicinski <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.