| Message ID | 20210826221306.2280066-2-songliubraving@fb.com (mailing list archive) |
|---|---|
| State | Superseded |
| Delegated to: | BPF |
| Headers | show |
| Series | bpf: introduce bpf_get_branch_snapshot | expand |
On Thu, Aug 26, 2021 at 03:13:04PM -0700, Song Liu wrote: > +int dummy_perf_snapshot_branch_stack(struct perf_branch_snapshot *br_snapshot); > + > +DECLARE_STATIC_CALL(perf_snapshot_branch_stack, dummy_perf_snapshot_branch_stack); > + > #endif /* _LINUX_PERF_EVENT_H */ > diff --git a/kernel/events/core.c b/kernel/events/core.c > index 011cc5069b7ba..c53fe90e630ac 100644 > --- a/kernel/events/core.c > +++ b/kernel/events/core.c > @@ -13437,3 +13437,6 @@ struct cgroup_subsys perf_event_cgrp_subsys = { > .threaded = true, > }; > #endif /* CONFIG_CGROUP_PERF */ > + > +DEFINE_STATIC_CALL_NULL(perf_snapshot_branch_stack, > + dummy_perf_snapshot_branch_stack); This isn't right... The whole dummy_perf_snapshot_branch_stack() thing is a declaration only and used as a typedef. Also, DEFINE_STATIC_CALL_NULL() and static_call_cond() rely on a void return value, which it doesn't have. Did you want: DECLARE_STATIC_CALL(perf_snapshot_branch_stack, void (*)(struct perf_branch_snapshot *)); DEFINE_STATIC_CALL_NULL(perf_snapshot_branch_stack, void (*)(struct perf_branch_snapshot *)); static_call_cond(perf_snapshot_branch_stack)(...); *OR*, do you actually need that return value, in which case you're probably looking for: DECLARE_STATIC_CALL(perf_snapshot_branch_stack, int (*)(struct perf_branch_snapshot *)); DEFINE_STATIC_CALL_RET0(perf_snapshot_branch_stack, int (*)(struct perf_branch_snapshot *)); ret = static_call(perf_snapshot_branch_stack)(...); ?
On Thu, Aug 26, 2021 at 03:13:04PM -0700, Song Liu wrote: > Some data on intel_pmu_lbr_disable_all() and perf_pmu_disable(). > > With this patch, when fexit program triggers, intel_pmu_lbr_disable_all is > used to stop the LBR, and the LBR is stopped after 6 extra branch records > (see the full trace below). If we replace intel_pmu_lbr_disable_all in > intel_pmu_snapshot_branch_stack() with perf_pmu_disable, the LBR is stopped > after 19 extra branch records. This is still acceptable for systems with 32 > LBR entries. But for systems with fewer entries, all the entries before > fexit are flushed. Therefore, I suggest we take the short cut and stop LBR > asap. > > > LBR snapshot captured when we use intel_pmu_lbr_disable_all(): > > ID: 0 from intel_pmu_lbr_disable_all.part.10+37 to intel_pmu_lbr_disable_all.part.10+72 > ID: 1 from intel_pmu_lbr_disable_all.part.10+33 to intel_pmu_lbr_disable_all.part.10+37 > ID: 2 from intel_pmu_snapshot_branch_stack+51 to intel_pmu_lbr_disable_all.part.10+0 > ID: 3 from __bpf_prog_enter+53 to intel_pmu_snapshot_branch_stack+0 > ID: 4 from __bpf_prog_enter+8 to __bpf_prog_enter+38 > ID: 5 from __brk_limit+473903158 to __bpf_prog_enter+0 > ID: 6 from bpf_fexit_loop_test1+22 to __brk_limit+473903139 > ID: 7 from bpf_fexit_loop_test1+20 to bpf_fexit_loop_test1+13 > ID: 8 from bpf_fexit_loop_test1+20 to bpf_fexit_loop_test1+13 > ID: 9 from bpf_fexit_loop_test1+20 to bpf_fexit_loop_test1+13 > > > LBR snapshot captured when we use perf_pmu_disable(): > > ID: 0 from intel_pmu_lbr_disable_all+58 to intel_pmu_lbr_disable_all+93 > ID: 1 from intel_pmu_lbr_disable_all+54 to intel_pmu_lbr_disable_all+58 > ID: 2 from intel_pmu_disable_all+15 to intel_pmu_lbr_disable_all+0 > ID: 3 from intel_pmu_pebs_disable_all+30 to intel_pmu_disable_all+15 > ID: 4 from intel_pmu_disable_all+10 to intel_pmu_pebs_disable_all+0 > ID: 5 from __intel_pmu_disable_all+49 to intel_pmu_disable_all+10 > ID: 6 from intel_pmu_disable_all+5 to __intel_pmu_disable_all+0 > ID: 7 from x86_pmu_disable+61 to intel_pmu_disable_all+0 > ID: 8 from x86_pmu_disable+38 to x86_pmu_disable+41 > ID: 9 from __x86_indirect_thunk_rax+16 to x86_pmu_disable+0 > ID: 10 from __x86_indirect_thunk_rax+0 to __x86_indirect_thunk_rax+12 > ID: 11 from perf_pmu_disable.part.122+4 to __x86_indirect_thunk_rax+0 > ID: 12 from perf_pmu_disable+23 to perf_pmu_disable.part.122+0 > ID: 13 from intel_pmu_snapshot_branch_stack+45 to perf_pmu_disable+0 > ID: 14 from x86_get_pmu+35 to intel_pmu_snapshot_branch_stack+39 > ID: 15 from intel_pmu_snapshot_branch_stack+34 to x86_get_pmu+0 > ID: 16 from __bpf_prog_enter+53 to intel_pmu_snapshot_branch_stack+0 > ID: 17 from __bpf_prog_enter+8 to __bpf_prog_enter+38 > ID: 18 from __brk_limit+478056502 to __bpf_prog_enter+0 > ID: 19 from bpf_fexit_loop_test1+22 to __brk_limit+478056483 > ID: 20 from bpf_fexit_loop_test1+20 to bpf_fexit_loop_test1+13 > ID: 21 from bpf_fexit_loop_test1+20 to bpf_fexit_loop_test1+13 Well, if you're willing to do something like: > diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c > index ac6fd2dabf6a2..a29649e7241cc 100644 > --- a/arch/x86/events/intel/core.c > +++ b/arch/x86/events/intel/core.c > @@ -6283,8 +6283,11 @@ __init int intel_pmu_init(void) > x86_pmu.lbr_nr = 0; > } > > - if (x86_pmu.lbr_nr) > + if (x86_pmu.lbr_nr) { > pr_cont("%d-deep LBR, ", x86_pmu.lbr_nr); if (x86_pmu.disable_all == intel_pmu_disable_all) > + static_call_update(perf_snapshot_branch_stack, > + intel_pmu_snapshot_branch_stack); > + } > > intel_pmu_check_extra_regs(x86_pmu.extra_regs); > > diff --git a/arch/x86/events/intel/lbr.c b/arch/x86/events/intel/lbr.c > index 9e6d6eaeb4cb6..7d4fe1d6e79ff 100644 > --- a/arch/x86/events/intel/lbr.c > +++ b/arch/x86/events/intel/lbr.c > @@ -1862,3 +1862,16 @@ EXPORT_SYMBOL_GPL(x86_perf_get_lbr); > struct event_constraint vlbr_constraint = > __EVENT_CONSTRAINT(INTEL_FIXED_VLBR_EVENT, (1ULL << INTEL_PMC_IDX_FIXED_VLBR), > FIXED_EVENT_FLAGS, 1, 0, PERF_X86_EVENT_LBR_SELECT); > + > +int intel_pmu_snapshot_branch_stack(struct perf_branch_snapshot *br_snapshot) > +{ > + struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); > + > + intel_pmu_lbr_disable_all(); > + intel_pmu_lbr_read(); > + memcpy(br_snapshot->entries, cpuc->lbr_entries, > + sizeof(struct perf_branch_entry) * x86_pmu.lbr_nr); > + br_snapshot->nr = x86_pmu.lbr_nr; > + intel_pmu_lbr_enable_all(false); > + return 0; > +} Then the above can assume perfmon > v2 and we can either inline __intel_pmu_disable_all() or simply do the wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL). One thing that needs checking, intel_pmu_disable_all() also clears MSR_IA32_PEBS_ENABLE, is that really needed if we just want to inhibit PMIs ? That is, will the PEBS machinery still trigger PMI if GLOBAL_CTRL == 0 ?
> On Aug 30, 2021, at 3:22 AM, Peter Zijlstra <peterz@infradead.org> wrote: > > On Thu, Aug 26, 2021 at 03:13:04PM -0700, Song Liu wrote: >> +int dummy_perf_snapshot_branch_stack(struct perf_branch_snapshot *br_snapshot); >> + >> +DECLARE_STATIC_CALL(perf_snapshot_branch_stack, dummy_perf_snapshot_branch_stack); >> + >> #endif /* _LINUX_PERF_EVENT_H */ >> diff --git a/kernel/events/core.c b/kernel/events/core.c >> index 011cc5069b7ba..c53fe90e630ac 100644 >> --- a/kernel/events/core.c >> +++ b/kernel/events/core.c >> @@ -13437,3 +13437,6 @@ struct cgroup_subsys perf_event_cgrp_subsys = { >> .threaded = true, >> }; >> #endif /* CONFIG_CGROUP_PERF */ >> + >> +DEFINE_STATIC_CALL_NULL(perf_snapshot_branch_stack, >> + dummy_perf_snapshot_branch_stack); > > This isn't right... > > The whole dummy_perf_snapshot_branch_stack() thing is a declaration only > and used as a typedef. Also, DEFINE_STATIC_CALL_NULL() and > static_call_cond() rely on a void return value, which it doesn't have. > > Did you want: > > DECLARE_STATIC_CALL(perf_snapshot_branch_stack, void (*)(struct perf_branch_snapshot *)); > > DEFINE_STATIC_CALL_NULL(perf_snapshot_branch_stack, void (*)(struct perf_branch_snapshot *)); > > static_call_cond(perf_snapshot_branch_stack)(...); > > *OR*, do you actually need that return value, in which case you're > probably looking for: > > DECLARE_STATIC_CALL(perf_snapshot_branch_stack, int (*)(struct perf_branch_snapshot *)); > > DEFINE_STATIC_CALL_RET0(perf_snapshot_branch_stack, int (*)(struct perf_branch_snapshot *)); > > ret = static_call(perf_snapshot_branch_stack)(...); > > ? Thanks for these information! I did get confused these macros for quite a while. Let me try with the _RET0 version. Song
> On Aug 30, 2021, at 3:43 AM, Peter Zijlstra <peterz@infradead.org> wrote: > > On Thu, Aug 26, 2021 at 03:13:04PM -0700, Song Liu wrote: > >> Some data on intel_pmu_lbr_disable_all() and perf_pmu_disable(). >> >> With this patch, when fexit program triggers, intel_pmu_lbr_disable_all is >> used to stop the LBR, and the LBR is stopped after 6 extra branch records >> (see the full trace below). If we replace intel_pmu_lbr_disable_all in >> intel_pmu_snapshot_branch_stack() with perf_pmu_disable, the LBR is stopped >> after 19 extra branch records. This is still acceptable for systems with 32 >> LBR entries. But for systems with fewer entries, all the entries before >> fexit are flushed. Therefore, I suggest we take the short cut and stop LBR >> asap. >> >> >> LBR snapshot captured when we use intel_pmu_lbr_disable_all(): >> >> ID: 0 from intel_pmu_lbr_disable_all.part.10+37 to intel_pmu_lbr_disable_all.part.10+72 >> ID: 1 from intel_pmu_lbr_disable_all.part.10+33 to intel_pmu_lbr_disable_all.part.10+37 >> ID: 2 from intel_pmu_snapshot_branch_stack+51 to intel_pmu_lbr_disable_all.part.10+0 >> ID: 3 from __bpf_prog_enter+53 to intel_pmu_snapshot_branch_stack+0 >> ID: 4 from __bpf_prog_enter+8 to __bpf_prog_enter+38 >> ID: 5 from __brk_limit+473903158 to __bpf_prog_enter+0 >> ID: 6 from bpf_fexit_loop_test1+22 to __brk_limit+473903139 >> ID: 7 from bpf_fexit_loop_test1+20 to bpf_fexit_loop_test1+13 >> ID: 8 from bpf_fexit_loop_test1+20 to bpf_fexit_loop_test1+13 >> ID: 9 from bpf_fexit_loop_test1+20 to bpf_fexit_loop_test1+13 >> >> >> LBR snapshot captured when we use perf_pmu_disable(): >> >> ID: 0 from intel_pmu_lbr_disable_all+58 to intel_pmu_lbr_disable_all+93 >> ID: 1 from intel_pmu_lbr_disable_all+54 to intel_pmu_lbr_disable_all+58 >> ID: 2 from intel_pmu_disable_all+15 to intel_pmu_lbr_disable_all+0 >> ID: 3 from intel_pmu_pebs_disable_all+30 to intel_pmu_disable_all+15 >> ID: 4 from intel_pmu_disable_all+10 to intel_pmu_pebs_disable_all+0 >> ID: 5 from __intel_pmu_disable_all+49 to intel_pmu_disable_all+10 >> ID: 6 from intel_pmu_disable_all+5 to __intel_pmu_disable_all+0 >> ID: 7 from x86_pmu_disable+61 to intel_pmu_disable_all+0 >> ID: 8 from x86_pmu_disable+38 to x86_pmu_disable+41 >> ID: 9 from __x86_indirect_thunk_rax+16 to x86_pmu_disable+0 >> ID: 10 from __x86_indirect_thunk_rax+0 to __x86_indirect_thunk_rax+12 >> ID: 11 from perf_pmu_disable.part.122+4 to __x86_indirect_thunk_rax+0 >> ID: 12 from perf_pmu_disable+23 to perf_pmu_disable.part.122+0 >> ID: 13 from intel_pmu_snapshot_branch_stack+45 to perf_pmu_disable+0 >> ID: 14 from x86_get_pmu+35 to intel_pmu_snapshot_branch_stack+39 >> ID: 15 from intel_pmu_snapshot_branch_stack+34 to x86_get_pmu+0 >> ID: 16 from __bpf_prog_enter+53 to intel_pmu_snapshot_branch_stack+0 >> ID: 17 from __bpf_prog_enter+8 to __bpf_prog_enter+38 >> ID: 18 from __brk_limit+478056502 to __bpf_prog_enter+0 >> ID: 19 from bpf_fexit_loop_test1+22 to __brk_limit+478056483 >> ID: 20 from bpf_fexit_loop_test1+20 to bpf_fexit_loop_test1+13 >> ID: 21 from bpf_fexit_loop_test1+20 to bpf_fexit_loop_test1+13 > > Well, if you're willing to do something like: > >> diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c >> index ac6fd2dabf6a2..a29649e7241cc 100644 >> --- a/arch/x86/events/intel/core.c >> +++ b/arch/x86/events/intel/core.c >> @@ -6283,8 +6283,11 @@ __init int intel_pmu_init(void) >> x86_pmu.lbr_nr = 0; >> } >> >> - if (x86_pmu.lbr_nr) >> + if (x86_pmu.lbr_nr) { >> pr_cont("%d-deep LBR, ", x86_pmu.lbr_nr); > > if (x86_pmu.disable_all == intel_pmu_disable_all) > >> + static_call_update(perf_snapshot_branch_stack, >> + intel_pmu_snapshot_branch_stack); >> + } >> >> intel_pmu_check_extra_regs(x86_pmu.extra_regs); >> >> diff --git a/arch/x86/events/intel/lbr.c b/arch/x86/events/intel/lbr.c >> index 9e6d6eaeb4cb6..7d4fe1d6e79ff 100644 >> --- a/arch/x86/events/intel/lbr.c >> +++ b/arch/x86/events/intel/lbr.c >> @@ -1862,3 +1862,16 @@ EXPORT_SYMBOL_GPL(x86_perf_get_lbr); >> struct event_constraint vlbr_constraint = >> __EVENT_CONSTRAINT(INTEL_FIXED_VLBR_EVENT, (1ULL << INTEL_PMC_IDX_FIXED_VLBR), >> FIXED_EVENT_FLAGS, 1, 0, PERF_X86_EVENT_LBR_SELECT); >> + >> +int intel_pmu_snapshot_branch_stack(struct perf_branch_snapshot *br_snapshot) >> +{ >> + struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); >> + >> + intel_pmu_lbr_disable_all(); >> + intel_pmu_lbr_read(); >> + memcpy(br_snapshot->entries, cpuc->lbr_entries, >> + sizeof(struct perf_branch_entry) * x86_pmu.lbr_nr); >> + br_snapshot->nr = x86_pmu.lbr_nr; >> + intel_pmu_lbr_enable_all(false); >> + return 0; >> +} > > Then the above can assume perfmon > v2 and we can either inline > __intel_pmu_disable_all() or simply do the > wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL). I think can do perfmon > v2 only. > > One thing that needs checking, intel_pmu_disable_all() also clears > MSR_IA32_PEBS_ENABLE, is that really needed if we just want to inhibit > PMIs ? That is, will the PEBS machinery still trigger PMI if GLOBAL_CTRL > == 0 ? Actually, can we do something like: static void intel_pmu_disable_all(void) { intel_pmu_lbr_disable_all(); /* moved to the beginning */ __intel_pmu_disable_all(); intel_pmu_pebs_disable_all(); } int intel_pmu_snapshot_branch_stack(struct perf_branch_snapshot *br_snapshot) { struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); intel_pmu_disable_all(); /* call full pmu_disable */ intel_pmu_lbr_read(); memcpy(br_snapshot->entries, cpuc->lbr_entries, sizeof(struct perf_branch_entry) * x86_pmu.lbr_nr); br_snapshot->nr = x86_pmu.lbr_nr; intel_pmu_enable_all(false); return 0; } In this way, we still call intel_pmu_disable_all(), but since LBR is disabled at the beginning of it, we would not flush too many LBR entries. Thanks, Song
On Mon, Aug 30, 2021 at 03:25:44PM +0000, Song Liu wrote: > Thanks for these information! I did get confused these macros for quite a > while. Let me try with the _RET0 version. Does you kernel have: 9ae6ab27f44e ("static_call: Update API documentation") ? With that included, the comment at the top of static_call.h reads like the below. Please let me know where you think this can be improved. /* * Static call support * * Static calls use code patching to hard-code function pointers into direct * branch instructions. They give the flexibility of function pointers, but * with improved performance. This is especially important for cases where * retpolines would otherwise be used, as retpolines can significantly impact * performance. * * * API overview: * * DECLARE_STATIC_CALL(name, func); * DEFINE_STATIC_CALL(name, func); * DEFINE_STATIC_CALL_NULL(name, typename); * DEFINE_STATIC_CALL_RET0(name, typename); * * __static_call_return0; * * static_call(name)(args...); * static_call_cond(name)(args...); * static_call_update(name, func); * static_call_query(name); * * EXPORT_STATIC_CALL{,_TRAMP}{,_GPL}() * * Usage example: * * # Start with the following functions (with identical prototypes): * int func_a(int arg1, int arg2); * int func_b(int arg1, int arg2); * * # Define a 'my_name' reference, associated with func_a() by default * DEFINE_STATIC_CALL(my_name, func_a); * * # Call func_a() * static_call(my_name)(arg1, arg2); * * # Update 'my_name' to point to func_b() * static_call_update(my_name, &func_b); * * # Call func_b() * static_call(my_name)(arg1, arg2); * * * Implementation details: * * This requires some arch-specific code (CONFIG_HAVE_STATIC_CALL). * Otherwise basic indirect calls are used (with function pointers). * * Each static_call() site calls into a trampoline associated with the name. * The trampoline has a direct branch to the default function. Updates to a * name will modify the trampoline's branch destination. * * If the arch has CONFIG_HAVE_STATIC_CALL_INLINE, then the call sites * themselves will be patched at runtime to call the functions directly, * rather than calling through the trampoline. This requires objtool or a * compiler plugin to detect all the static_call() sites and annotate them * in the .static_call_sites section. * * * Notes on NULL function pointers: * * Static_call()s support NULL functions, with many of the caveats that * regular function pointers have. * * Clearly calling a NULL function pointer is 'BAD', so too for * static_call()s (although when HAVE_STATIC_CALL it might not be immediately * fatal). A NULL static_call can be the result of: * * DECLARE_STATIC_CALL_NULL(my_static_call, void (*)(int)); * * which is equivalent to declaring a NULL function pointer with just a * typename: * * void (*my_func_ptr)(int arg1) = NULL; * * or using static_call_update() with a NULL function. In both cases the * HAVE_STATIC_CALL implementation will patch the trampoline with a RET * instruction, instead of an immediate tail-call JMP. HAVE_STATIC_CALL_INLINE * architectures can patch the trampoline call to a NOP. * * In all cases, any argument evaluation is unconditional. Unlike a regular * conditional function pointer call: * * if (my_func_ptr) * my_func_ptr(arg1) * * where the argument evaludation also depends on the pointer value. * * When calling a static_call that can be NULL, use: * * static_call_cond(name)(arg1); * * which will include the required value tests to avoid NULL-pointer * dereferences. * * To query which function is currently set to be called, use: * * func = static_call_query(name); * * * DEFINE_STATIC_CALL_RET0 / __static_call_return0: * * Just like how DEFINE_STATIC_CALL_NULL() / static_call_cond() optimize the * conditional void function call, DEFINE_STATIC_CALL_RET0 / * __static_call_return0 optimize the do nothing return 0 function. * * This feature is strictly UB per the C standard (since it casts a function * pointer to a different signature) and relies on the architecture ABI to * make things work. In particular it relies on Caller Stack-cleanup and the * whole return register being clobbered for short return values. All normal * CDECL style ABIs conform. * * In particular the x86_64 implementation replaces the 5 byte CALL * instruction at the callsite with a 5 byte clear of the RAX register, * completely eliding any function call overhead. * * Notably argument setup is unconditional. * * * EXPORT_STATIC_CALL() vs EXPORT_STATIC_CALL_TRAMP(): * * The difference is that the _TRAMP variant tries to only export the * trampoline with the result that a module can use static_call{,_cond}() but * not static_call_update(). * */
> On Aug 30, 2021, at 9:06 AM, Peter Zijlstra <peterz@infradead.org> wrote: > > On Mon, Aug 30, 2021 at 03:25:44PM +0000, Song Liu wrote: >> Thanks for these information! I did get confused these macros for quite a >> while. Let me try with the _RET0 version. > > Does you kernel have: > > 9ae6ab27f44e ("static_call: Update API documentation") > > ? > > With that included, the comment at the top of static_call.h reads like > the below. Please let me know where you think this can be improved. Aha, my kernel has the code for _RET0 part, but not the documentation. > > /* > * Static call support > * > * Static calls use code patching to hard-code function pointers into direct > * branch instructions. They give the flexibility of function pointers, but > * with improved performance. This is especially important for cases where > * retpolines would otherwise be used, as retpolines can significantly impact > * performance. > * [...] > * > * Notes on NULL function pointers: > * > * Static_call()s support NULL functions, with many of the caveats that > * regular function pointers have. > * > * Clearly calling a NULL function pointer is 'BAD', so too for > * static_call()s (although when HAVE_STATIC_CALL it might not be immediately > * fatal). A NULL static_call can be the result of: > * Probably add: * /* for function that returns NULL */ > * DECLARE_STATIC_CALL_NULL(my_static_call, void (*)(int)); * or * /* for function that returns int */ * DECLARE_STATIC_CALL_RET0(my_static_call, int (*)(int)); * So it is clear that we need two different macros. IIUC, the number and type of arguments doesn't matter? Also, the default return int function has to return 0, right? Can we let it return -EOPNOSUPP? > * > * which is equivalent to declaring a NULL function pointer with just a > * typename: > * > * void (*my_func_ptr)(int arg1) = NULL; > * [...] > * which will include the required value tests to avoid NULL-pointer > * dereferences. > * > * To query which function is currently set to be called, use: > * > * func = static_call_query(name); Maybe move above two lines to "Usage example:" section? Thanks, Song
> On Aug 30, 2021, at 3:22 AM, Peter Zijlstra <peterz@infradead.org> wrote: > > On Thu, Aug 26, 2021 at 03:13:04PM -0700, Song Liu wrote: >> +int dummy_perf_snapshot_branch_stack(struct perf_branch_snapshot *br_snapshot); >> + >> +DECLARE_STATIC_CALL(perf_snapshot_branch_stack, dummy_perf_snapshot_branch_stack); >> + >> #endif /* _LINUX_PERF_EVENT_H */ >> diff --git a/kernel/events/core.c b/kernel/events/core.c >> index 011cc5069b7ba..c53fe90e630ac 100644 >> --- a/kernel/events/core.c >> +++ b/kernel/events/core.c >> @@ -13437,3 +13437,6 @@ struct cgroup_subsys perf_event_cgrp_subsys = { >> .threaded = true, >> }; >> #endif /* CONFIG_CGROUP_PERF */ >> + >> +DEFINE_STATIC_CALL_NULL(perf_snapshot_branch_stack, >> + dummy_perf_snapshot_branch_stack); > > This isn't right... > > The whole dummy_perf_snapshot_branch_stack() thing is a declaration only > and used as a typedef. Also, DEFINE_STATIC_CALL_NULL() and > static_call_cond() rely on a void return value, which it doesn't have. > > Did you want: > > DECLARE_STATIC_CALL(perf_snapshot_branch_stack, void (*)(struct perf_branch_snapshot *)); > > DEFINE_STATIC_CALL_NULL(perf_snapshot_branch_stack, void (*)(struct perf_branch_snapshot *)); > > static_call_cond(perf_snapshot_branch_stack)(...); > > *OR*, do you actually need that return value, in which case you're > probably looking for: > > DECLARE_STATIC_CALL(perf_snapshot_branch_stack, int (*)(struct perf_branch_snapshot *)); > > DEFINE_STATIC_CALL_RET0(perf_snapshot_branch_stack, int (*)(struct perf_branch_snapshot *)); > > ret = static_call(perf_snapshot_branch_stack)(...); > > ? Hmmm... something doesn't work here. I have: /* include/linux/perf_event.h */ DECLARE_STATIC_CALL(perf_snapshot_branch_stack, int (*)(struct perf_branch_snapshot *)); /* kernel/events/core.c */ DEFINE_STATIC_CALL_RET0(perf_snapshot_branch_stack, int (*)(struct perf_branch_snapshot *)); /* kernel/bpf/trampoline.c */ if (prog->call_get_branch) static_call(perf_snapshot_branch_stack)( this_cpu_ptr(&bpf_perf_branch_snapshot)); /* arch/x86/events/intel/core.c */ if (x86_pmu.disable_all == intel_pmu_disable_all) static_call_update(perf_snapshot_branch_stack, intel_pmu_snapshot_branch_stack); And the compiler keeps complain with: arch/x86/events/intel/core.c: In function ‘intel_pmu_init’: ./include/linux/static_call.h:121:41: error: initialization of ‘int (**)(struct perf_branch_snapshot *)’ from incompatible pointer type ‘int (*)(struct perf_branch_snapshot *)’ [-Werror=incompatible-pointer-type ] typeof(&STATIC_CALL_TRAMP(name)) __F = (func); \ ^ arch/x86/events/intel/core.c:6305:4: note: in expansion of macro ‘static_call_update’ static_call_update(perf_snapshot_branch_stack, ^~~~~~~~~~~~~~~~~~ Something like typedef int (perf_snapshot_branch_stack_t)(struct perf_branch_snapshot *); DECLARE_STATIC_CALL(perf_snapshot_branch_stack, perf_snapshot_branch_stack_t); seems to work fine. Thanks, Song
On Mon, Aug 30, 2021 at 05:41:46PM +0000, Song Liu wrote: > DECLARE_STATIC_CALL(perf_snapshot_branch_stack, > int (*)(struct perf_branch_snapshot *)); > Something like > > typedef int (perf_snapshot_branch_stack_t)(struct perf_branch_snapshot *); > DECLARE_STATIC_CALL(perf_snapshot_branch_stack, perf_snapshot_branch_stack_t); > > seems to work fine. Oh urg, indeed. It wants a function type, not a function pointer type. I've been bitten by that before. Go with the typedef, that's the sanest.
On Mon, Aug 30, 2021 at 04:36:44PM +0000, Song Liu wrote: > > /* > > * Static call support > > * > > * Static calls use code patching to hard-code function pointers into direct > > * branch instructions. They give the flexibility of function pointers, but > > * with improved performance. This is especially important for cases where > > * retpolines would otherwise be used, as retpolines can significantly impact > > * performance. > > * > > [...] > > > * > > * Notes on NULL function pointers: > > * > > * Static_call()s support NULL functions, with many of the caveats that > > * regular function pointers have. > > * > > * Clearly calling a NULL function pointer is 'BAD', so too for > > * static_call()s (although when HAVE_STATIC_CALL it might not be immediately > > * fatal). A NULL static_call can be the result of: > > * > > Probably add: > > * /* for function that returns NULL */ > > * DECLARE_STATIC_CALL_NULL(my_static_call, void (*)(int)); > > > * or > * /* for function that returns int */ > * DECLARE_STATIC_CALL_RET0(my_static_call, int (*)(int)); > * No, anything that uses static_call_cond() must have void return. Note that static_call_cond() replaces: if (func_ptr) func_ptr(args); which is a statement, not an expression, and as such can't function as an rvalue. > So it is clear that we need two different macros. IIUC, the number and > type of arguments doesn't matter? Right, arguments are irrelevant, provided CDECL ABI, which mandates that arguments that are pushed on the stack are cleaned up by the caller, not the callee. > Also, the default return int function has to return 0, right? Can we let > it return -EOPNOSUPP? Difficult, in principle we can patch any value that fits in a single instruction, but the more variants we have, the harder it gets.
On Mon, Aug 30, 2021 at 08:07:24PM +0200, Peter Zijlstra wrote: > On Mon, Aug 30, 2021 at 05:41:46PM +0000, Song Liu wrote: > > DECLARE_STATIC_CALL(perf_snapshot_branch_stack, > > int (*)(struct perf_branch_snapshot *)); > > > Something like > > > > typedef int (perf_snapshot_branch_stack_t)(struct perf_branch_snapshot *); > > DECLARE_STATIC_CALL(perf_snapshot_branch_stack, perf_snapshot_branch_stack_t); > > > > seems to work fine. > > Oh urg, indeed. It wants a function type, not a function pointer type. > I've been bitten by that before. Go with the typedef, that's the sanest. The below is the best I can make of it... it's a little inconsistent and somewhat tricky, but at least the compiler yells hard if you get it wrong. I can *almost* get to: DEFINE_STATIC_CALL(foo, &func), except for ARCH_DEFINE_STATIC_CALL_TRAMP() which needs the actual function name string for the ASM :-( The rest can do with a function pointer type and have it work. --- diff --git a/arch/arm/include/asm/paravirt.h b/arch/arm/include/asm/paravirt.h index 95d5b0d625cd..1094b3abd910 100644 --- a/arch/arm/include/asm/paravirt.h +++ b/arch/arm/include/asm/paravirt.h @@ -9,9 +9,7 @@ struct static_key; extern struct static_key paravirt_steal_enabled; extern struct static_key paravirt_steal_rq_enabled; -u64 dummy_steal_clock(int cpu); - -DECLARE_STATIC_CALL(pv_steal_clock, dummy_steal_clock); +DECLARE_STATIC_CALL(pv_steal_clock, u64 (*)(int)); static inline u64 paravirt_steal_clock(int cpu) { diff --git a/arch/arm64/include/asm/paravirt.h b/arch/arm64/include/asm/paravirt.h index 9aa193e0e8f2..26539c1c277a 100644 --- a/arch/arm64/include/asm/paravirt.h +++ b/arch/arm64/include/asm/paravirt.h @@ -9,9 +9,7 @@ struct static_key; extern struct static_key paravirt_steal_enabled; extern struct static_key paravirt_steal_rq_enabled; -u64 dummy_steal_clock(int cpu); - -DECLARE_STATIC_CALL(pv_steal_clock, dummy_steal_clock); +DECLARE_STATIC_CALL(pv_steal_clock, u64 (*)(int)); static inline u64 paravirt_steal_clock(int cpu) { diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c index 2a57dbed4894..0c3b302026dc 100644 --- a/arch/x86/events/core.c +++ b/arch/x86/events/core.c @@ -60,35 +60,35 @@ DEFINE_STATIC_KEY_FALSE(perf_is_hybrid); * This here uses DEFINE_STATIC_CALL_NULL() to get a static_call defined * from just a typename, as opposed to an actual function. */ -DEFINE_STATIC_CALL_NULL(x86_pmu_handle_irq, *x86_pmu.handle_irq); -DEFINE_STATIC_CALL_NULL(x86_pmu_disable_all, *x86_pmu.disable_all); -DEFINE_STATIC_CALL_NULL(x86_pmu_enable_all, *x86_pmu.enable_all); -DEFINE_STATIC_CALL_NULL(x86_pmu_enable, *x86_pmu.enable); -DEFINE_STATIC_CALL_NULL(x86_pmu_disable, *x86_pmu.disable); +DEFINE_STATIC_CALL_NULL(x86_pmu_handle_irq, x86_pmu.handle_irq); +DEFINE_STATIC_CALL_NULL(x86_pmu_disable_all, x86_pmu.disable_all); +DEFINE_STATIC_CALL_NULL(x86_pmu_enable_all, x86_pmu.enable_all); +DEFINE_STATIC_CALL_NULL(x86_pmu_enable, x86_pmu.enable); +DEFINE_STATIC_CALL_NULL(x86_pmu_disable, x86_pmu.disable); -DEFINE_STATIC_CALL_NULL(x86_pmu_add, *x86_pmu.add); -DEFINE_STATIC_CALL_NULL(x86_pmu_del, *x86_pmu.del); -DEFINE_STATIC_CALL_NULL(x86_pmu_read, *x86_pmu.read); +DEFINE_STATIC_CALL_NULL(x86_pmu_add, x86_pmu.add); +DEFINE_STATIC_CALL_NULL(x86_pmu_del, x86_pmu.del); +DEFINE_STATIC_CALL_NULL(x86_pmu_read, x86_pmu.read); -DEFINE_STATIC_CALL_NULL(x86_pmu_schedule_events, *x86_pmu.schedule_events); -DEFINE_STATIC_CALL_NULL(x86_pmu_get_event_constraints, *x86_pmu.get_event_constraints); -DEFINE_STATIC_CALL_NULL(x86_pmu_put_event_constraints, *x86_pmu.put_event_constraints); +DEFINE_STATIC_CALL_NULL(x86_pmu_schedule_events, x86_pmu.schedule_events); +DEFINE_STATIC_CALL_NULL(x86_pmu_get_event_constraints, x86_pmu.get_event_constraints); +DEFINE_STATIC_CALL_NULL(x86_pmu_put_event_constraints, x86_pmu.put_event_constraints); -DEFINE_STATIC_CALL_NULL(x86_pmu_start_scheduling, *x86_pmu.start_scheduling); -DEFINE_STATIC_CALL_NULL(x86_pmu_commit_scheduling, *x86_pmu.commit_scheduling); -DEFINE_STATIC_CALL_NULL(x86_pmu_stop_scheduling, *x86_pmu.stop_scheduling); +DEFINE_STATIC_CALL_NULL(x86_pmu_start_scheduling, x86_pmu.start_scheduling); +DEFINE_STATIC_CALL_NULL(x86_pmu_commit_scheduling, x86_pmu.commit_scheduling); +DEFINE_STATIC_CALL_NULL(x86_pmu_stop_scheduling, x86_pmu.stop_scheduling); -DEFINE_STATIC_CALL_NULL(x86_pmu_sched_task, *x86_pmu.sched_task); -DEFINE_STATIC_CALL_NULL(x86_pmu_swap_task_ctx, *x86_pmu.swap_task_ctx); +DEFINE_STATIC_CALL_NULL(x86_pmu_sched_task, x86_pmu.sched_task); +DEFINE_STATIC_CALL_NULL(x86_pmu_swap_task_ctx, x86_pmu.swap_task_ctx); -DEFINE_STATIC_CALL_NULL(x86_pmu_drain_pebs, *x86_pmu.drain_pebs); -DEFINE_STATIC_CALL_NULL(x86_pmu_pebs_aliases, *x86_pmu.pebs_aliases); +DEFINE_STATIC_CALL_NULL(x86_pmu_drain_pebs, x86_pmu.drain_pebs); +DEFINE_STATIC_CALL_NULL(x86_pmu_pebs_aliases, x86_pmu.pebs_aliases); /* * This one is magic, it will get called even when PMU init fails (because * there is no PMU), in which case it should simply return NULL. */ -DEFINE_STATIC_CALL_RET0(x86_pmu_guest_get_msrs, *x86_pmu.guest_get_msrs); +DEFINE_STATIC_CALL_RET0(x86_pmu_guest_get_msrs, x86_pmu.guest_get_msrs); u64 __read_mostly hw_cache_event_ids [PERF_COUNT_HW_CACHE_MAX] diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index af6ce8d4c86a..d383bda4316e 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -1503,7 +1503,7 @@ extern bool __read_mostly enable_apicv; extern struct kvm_x86_ops kvm_x86_ops; #define KVM_X86_OP(func) \ - DECLARE_STATIC_CALL(kvm_x86_##func, *(((struct kvm_x86_ops *)0)->func)); + DECLARE_STATIC_CALL(kvm_x86_##func, (((struct kvm_x86_ops *)0)->func)); #define KVM_X86_OP_NULL KVM_X86_OP #include <asm/kvm-x86-ops.h> diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index da3a1ac82be5..3db2237e9a8d 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -18,11 +18,8 @@ #include <linux/static_call_types.h> #include <asm/frame.h> -u64 dummy_steal_clock(int cpu); -u64 dummy_sched_clock(void); - -DECLARE_STATIC_CALL(pv_steal_clock, dummy_steal_clock); -DECLARE_STATIC_CALL(pv_sched_clock, dummy_sched_clock); +DECLARE_STATIC_CALL(pv_steal_clock, u64 (*)(int)); +DECLARE_STATIC_CALL(pv_sched_clock, u64 (*)(void)); void paravirt_set_sched_clock(u64 (*func)(void)); diff --git a/arch/x86/include/asm/preempt.h b/arch/x86/include/asm/preempt.h index fe5efbcba824..65b2e6ec87a7 100644 --- a/arch/x86/include/asm/preempt.h +++ b/arch/x86/include/asm/preempt.h @@ -117,7 +117,7 @@ extern asmlinkage void preempt_schedule_notrace_thunk(void); #ifdef CONFIG_PREEMPT_DYNAMIC -DECLARE_STATIC_CALL(preempt_schedule, __preempt_schedule_func); +DECLARE_STATIC_CALL(preempt_schedule, &__preempt_schedule_func); #define __preempt_schedule() \ do { \ @@ -125,7 +125,7 @@ do { \ asm volatile ("call " STATIC_CALL_TRAMP_STR(preempt_schedule) : ASM_CALL_CONSTRAINT); \ } while (0) -DECLARE_STATIC_CALL(preempt_schedule_notrace, __preempt_schedule_notrace_func); +DECLARE_STATIC_CALL(preempt_schedule_notrace, &__preempt_schedule_notrace_func); #define __preempt_schedule_notrace() \ do { \ diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index e5d5c5ed7dd4..940c17099fcf 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -125,7 +125,7 @@ EXPORT_SYMBOL_GPL(kvm_x86_ops); #define KVM_X86_OP(func) \ DEFINE_STATIC_CALL_NULL(kvm_x86_##func, \ - *(((struct kvm_x86_ops *)0)->func)); + (((struct kvm_x86_ops *)0)->func)); #define KVM_X86_OP_NULL KVM_X86_OP #include <asm/kvm-x86-ops.h> EXPORT_STATIC_CALL_GPL(kvm_x86_get_cs_db_l_bits); diff --git a/include/linux/entry-common.h b/include/linux/entry-common.h index 2e2b8d6140ed..d4106a3f3243 100644 --- a/include/linux/entry-common.h +++ b/include/linux/entry-common.h @@ -456,7 +456,7 @@ irqentry_state_t noinstr irqentry_enter(struct pt_regs *regs); */ void irqentry_exit_cond_resched(void); #ifdef CONFIG_PREEMPT_DYNAMIC -DECLARE_STATIC_CALL(irqentry_exit_cond_resched, irqentry_exit_cond_resched); +DECLARE_STATIC_CALL(irqentry_exit_cond_resched, &irqentry_exit_cond_resched); #endif /** diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 1b2f0a7e00d6..58aa80015db6 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -97,7 +97,7 @@ extern int __cond_resched(void); extern int __cond_resched(void); -DECLARE_STATIC_CALL(might_resched, __cond_resched); +DECLARE_STATIC_CALL(might_resched, &__cond_resched); static __always_inline void might_resched(void) { diff --git a/include/linux/sched.h b/include/linux/sched.h index 1780260f237b..93aad9c6fad1 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2012,7 +2012,7 @@ extern int __cond_resched(void); #ifdef CONFIG_PREEMPT_DYNAMIC -DECLARE_STATIC_CALL(cond_resched, __cond_resched); +DECLARE_STATIC_CALL(cond_resched, &__cond_resched); static __always_inline int _cond_resched(void) { diff --git a/include/linux/static_call.h b/include/linux/static_call.h index 3e56a9751c06..18ee529b4937 100644 --- a/include/linux/static_call.h +++ b/include/linux/static_call.h @@ -14,16 +14,16 @@ * * API overview: * - * DECLARE_STATIC_CALL(name, func); - * DEFINE_STATIC_CALL(name, func); - * DEFINE_STATIC_CALL_NULL(name, typename); - * DEFINE_STATIC_CALL_RET0(name, typename); + * DECLARE_STATIC_CALL(name, &func); + * DEFINE_STATIC_CALL(name, func); // needs an actual function + * DEFINE_STATIC_CALL_NULL(name, &func); + * DEFINE_STATIC_CALL_RET0(name, &func); * * __static_call_return0; * * static_call(name)(args...); * static_call_cond(name)(args...); - * static_call_update(name, func); + * static_call_update(name, &func); * static_call_query(name); * * EXPORT_STATIC_CALL{,_TRAMP}{,_GPL}() @@ -46,6 +46,9 @@ * # Call func_b() * static_call(my_name)(arg1, arg2); * + * @ To query which function is currently set to be called, use: + * func = static_call_query(name); + * * * Implementation details: * @@ -66,7 +69,8 @@ * Notes on NULL function pointers: * * Static_call()s support NULL functions, with many of the caveats that - * regular function pointers have. + * regular function pointers have and a few extra. In particular they rely on + * the function return type being void. * * Clearly calling a NULL function pointer is 'BAD', so too for * static_call()s (although when HAVE_STATIC_CALL it might not be immediately @@ -79,10 +83,11 @@ * * void (*my_func_ptr)(int arg1) = NULL; * - * or using static_call_update() with a NULL function. In both cases the - * HAVE_STATIC_CALL implementation will patch the trampoline with a RET - * instruction, instead of an immediate tail-call JMP. HAVE_STATIC_CALL_INLINE - * architectures can patch the trampoline call to a NOP. + * or using static_call_update() with a NULL function pointer. In both cases + * the HAVE_STATIC_CALL implementation will patch the trampoline with a RET + * instruction, instead of an immediate tail-call JMP. + * HAVE_STATIC_CALL_INLINE architectures can patch the trampoline call to a + * NOP. * * In all cases, any argument evaluation is unconditional. Unlike a regular * conditional function pointer call: @@ -97,11 +102,8 @@ * static_call_cond(name)(arg1); * * which will include the required value tests to avoid NULL-pointer - * dereferences. - * - * To query which function is currently set to be called, use: - * - * func = static_call_query(name); + * dereferences. Note that this is a statement, not an expression, hence the + * requirement for a void return value. * * * DEFINE_STATIC_CALL_RET0 / __static_call_return0: @@ -122,6 +124,14 @@ * * Notably argument setup is unconditional. * + * For example: + * + * DEFINE_STATIC_CALL_RET0(my_ret_func, int (*)(int)); + * + * ret = static_call(my_ret_func)(5); + * + * will, unless static_call_update() is used, return 0. + * * * EXPORT_STATIC_CALL() vs EXPORT_STATIC_CALL_TRAMP(): * @@ -180,16 +190,16 @@ extern int static_call_text_reserved(void *start, void *end); extern long __static_call_return0(void); -#define __DEFINE_STATIC_CALL(name, _func, _func_init) \ - DECLARE_STATIC_CALL(name, _func); \ +#define __DEFINE_STATIC_CALL(name, func_ptr, func_init) \ + DECLARE_STATIC_CALL(name, func_ptr); \ struct static_call_key STATIC_CALL_KEY(name) = { \ - .func = _func_init, \ + .func = func_init, \ .type = 1, \ }; \ - ARCH_DEFINE_STATIC_CALL_TRAMP(name, _func_init) + ARCH_DEFINE_STATIC_CALL_TRAMP(name, func_init) -#define DEFINE_STATIC_CALL_NULL(name, _func) \ - DECLARE_STATIC_CALL(name, _func); \ +#define DEFINE_STATIC_CALL_NULL(name, func_ptr) \ + DECLARE_STATIC_CALL(name, func_ptr); \ struct static_call_key STATIC_CALL_KEY(name) = { \ .func = NULL, \ .type = 1, \ @@ -217,15 +227,15 @@ extern long __static_call_return0(void); static inline int static_call_init(void) { return 0; } -#define __DEFINE_STATIC_CALL(name, _func, _func_init) \ - DECLARE_STATIC_CALL(name, _func); \ +#define __DEFINE_STATIC_CALL(name, func_ptr, func_init) \ + DECLARE_STATIC_CALL(name, func_ptr); \ struct static_call_key STATIC_CALL_KEY(name) = { \ - .func = _func_init, \ + .func = func_init, \ }; \ - ARCH_DEFINE_STATIC_CALL_TRAMP(name, _func_init) + ARCH_DEFINE_STATIC_CALL_TRAMP(name, func_init) -#define DEFINE_STATIC_CALL_NULL(name, _func) \ - DECLARE_STATIC_CALL(name, _func); \ +#define DEFINE_STATIC_CALL_NULL(name, func_ptr) \ + DECLARE_STATIC_CALL(name, func_ptr); \ struct static_call_key STATIC_CALL_KEY(name) = { \ .func = NULL, \ }; \ @@ -275,14 +285,14 @@ static inline long __static_call_return0(void) return 0; } -#define __DEFINE_STATIC_CALL(name, _func, _func_init) \ - DECLARE_STATIC_CALL(name, _func); \ +#define __DEFINE_STATIC_CALL(name, func_ptr, func_init) \ + DECLARE_STATIC_CALL(name, func_ptr); \ struct static_call_key STATIC_CALL_KEY(name) = { \ - .func = _func_init, \ + .func = func_init, \ } -#define DEFINE_STATIC_CALL_NULL(name, _func) \ - DECLARE_STATIC_CALL(name, _func); \ +#define DEFINE_STATIC_CALL_NULL(name, func_ptr) \ + DECLARE_STATIC_CALL(name, func_ptr); \ struct static_call_key STATIC_CALL_KEY(name) = { \ .func = NULL, \ } @@ -327,10 +337,10 @@ static inline int static_call_text_reserved(void *start, void *end) #endif /* CONFIG_HAVE_STATIC_CALL */ -#define DEFINE_STATIC_CALL(name, _func) \ - __DEFINE_STATIC_CALL(name, _func, _func) +#define DEFINE_STATIC_CALL(name, func) \ + __DEFINE_STATIC_CALL(name, &func, func) -#define DEFINE_STATIC_CALL_RET0(name, _func) \ - __DEFINE_STATIC_CALL(name, _func, __static_call_return0) +#define DEFINE_STATIC_CALL_RET0(name, func_ptr) \ + __DEFINE_STATIC_CALL(name, func_ptr, __static_call_return0) #endif /* _LINUX_STATIC_CALL_H */ diff --git a/include/linux/static_call_types.h b/include/linux/static_call_types.h index 5a00b8b2cf9f..3f87aa682e14 100644 --- a/include/linux/static_call_types.h +++ b/include/linux/static_call_types.h @@ -34,9 +34,15 @@ struct static_call_site { s32 key; }; -#define DECLARE_STATIC_CALL(name, func) \ +/* + * Type mismatch on __SCP__ functions is due to mismatched function vs function + * pointer arguments between DECLARE_STATIC_CALL() and DEFINE_STATIC_CALL*() + * variants. + */ +#define DECLARE_STATIC_CALL(name, func_ptr) \ extern struct static_call_key STATIC_CALL_KEY(name); \ - extern typeof(func) STATIC_CALL_TRAMP(name); + extern typeof(func_ptr) __SCP__##name; \ + extern typeof(*__SCP__##name) STATIC_CALL_TRAMP(name); #ifdef CONFIG_HAVE_STATIC_CALL diff --git a/include/linux/tracepoint.h b/include/linux/tracepoint.h index ab58696d0ddd..118b78fffc9d 100644 --- a/include/linux/tracepoint.h +++ b/include/linux/tracepoint.h @@ -240,7 +240,7 @@ static inline struct tracepoint *tracepoint_ptr_deref(tracepoint_ptr_t *p) */ #define __DECLARE_TRACE(name, proto, args, cond, data_proto) \ extern int __traceiter_##name(data_proto); \ - DECLARE_STATIC_CALL(tp_func_##name, __traceiter_##name); \ + DECLARE_STATIC_CALL(tp_func_##name, &__traceiter_##name); \ extern struct tracepoint __tracepoint_##name; \ static inline void trace_##name(proto) \ { \ diff --git a/kernel/sched/core.c b/kernel/sched/core.c index c4462c454ab9..def3aa224ae5 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -8171,10 +8171,10 @@ EXPORT_SYMBOL(__cond_resched); #endif #ifdef CONFIG_PREEMPT_DYNAMIC -DEFINE_STATIC_CALL_RET0(cond_resched, __cond_resched); +DEFINE_STATIC_CALL_RET0(cond_resched, &__cond_resched); EXPORT_STATIC_CALL_TRAMP(cond_resched); -DEFINE_STATIC_CALL_RET0(might_resched, __cond_resched); +DEFINE_STATIC_CALL_RET0(might_resched, &__cond_resched); EXPORT_STATIC_CALL_TRAMP(might_resched); #endif diff --git a/kernel/static_call.c b/kernel/static_call.c index 43ba0b1e0edb..b652c32a7250 100644 --- a/kernel/static_call.c +++ b/kernel/static_call.c @@ -517,6 +517,12 @@ static int func_b(int x) } DEFINE_STATIC_CALL(sc_selftest, func_a); +DEFINE_STATIC_CALL_NULL(sc_null, void (*)(int *)); + +static void null_a(int *arg) +{ + *arg = 1; +} static struct static_call_data { int (*func)(int); @@ -530,18 +536,26 @@ static struct static_call_data { static int __init test_static_call_init(void) { - int i; + int i; - for (i = 0; i < ARRAY_SIZE(static_call_data); i++ ) { - struct static_call_data *scd = &static_call_data[i]; + for (i = 0; i < ARRAY_SIZE(static_call_data); i++ ) { + struct static_call_data *scd = &static_call_data[i]; - if (scd->func) - static_call_update(sc_selftest, scd->func); + if (scd->func) + static_call_update(sc_selftest, scd->func); - WARN_ON(static_call(sc_selftest)(scd->val) != scd->expect); - } + WARN_ON(static_call(sc_selftest)(scd->val) != scd->expect); + } - return 0; + i = 5; + static_call_cond(sc_null)(&i); + WARN_ON(i != 5); + + static_call_update(sc_null, &null_a); + static_call_cond(sc_null)(&i); + WARN_ON(i != 1); + + return 0; } early_initcall(test_static_call_init); diff --git a/security/keys/trusted-keys/trusted_core.c b/security/keys/trusted-keys/trusted_core.c index d5c891d8d353..1d91eb6b242e 100644 --- a/security/keys/trusted-keys/trusted_core.c +++ b/security/keys/trusted-keys/trusted_core.c @@ -35,13 +35,13 @@ static const struct trusted_key_source trusted_key_sources[] = { #endif }; -DEFINE_STATIC_CALL_NULL(trusted_key_init, *trusted_key_sources[0].ops->init); -DEFINE_STATIC_CALL_NULL(trusted_key_seal, *trusted_key_sources[0].ops->seal); +DEFINE_STATIC_CALL_NULL(trusted_key_init, trusted_key_sources[0].ops->init); +DEFINE_STATIC_CALL_NULL(trusted_key_seal, trusted_key_sources[0].ops->seal); DEFINE_STATIC_CALL_NULL(trusted_key_unseal, - *trusted_key_sources[0].ops->unseal); + trusted_key_sources[0].ops->unseal); DEFINE_STATIC_CALL_NULL(trusted_key_get_random, - *trusted_key_sources[0].ops->get_random); -DEFINE_STATIC_CALL_NULL(trusted_key_exit, *trusted_key_sources[0].ops->exit); + trusted_key_sources[0].ops->get_random); +DEFINE_STATIC_CALL_NULL(trusted_key_exit, trusted_key_sources[0].ops->exit); static unsigned char migratable; enum { diff --git a/tools/include/linux/static_call_types.h b/tools/include/linux/static_call_types.h index 5a00b8b2cf9f..3f87aa682e14 100644 --- a/tools/include/linux/static_call_types.h +++ b/tools/include/linux/static_call_types.h @@ -34,9 +34,15 @@ struct static_call_site { s32 key; }; -#define DECLARE_STATIC_CALL(name, func) \ +/* + * Type mismatch on __SCP__ functions is due to mismatched function vs function + * pointer arguments between DECLARE_STATIC_CALL() and DEFINE_STATIC_CALL*() + * variants. + */ +#define DECLARE_STATIC_CALL(name, func_ptr) \ extern struct static_call_key STATIC_CALL_KEY(name); \ - extern typeof(func) STATIC_CALL_TRAMP(name); + extern typeof(func_ptr) __SCP__##name; \ + extern typeof(*__SCP__##name) STATIC_CALL_TRAMP(name); #ifdef CONFIG_HAVE_STATIC_CALL
On Wed, Sep 01, 2021 at 07:12:24PM +0200, Peter Zijlstra wrote: > On Mon, Aug 30, 2021 at 08:07:24PM +0200, Peter Zijlstra wrote: > > On Mon, Aug 30, 2021 at 05:41:46PM +0000, Song Liu wrote: > > > DECLARE_STATIC_CALL(perf_snapshot_branch_stack, > > > int (*)(struct perf_branch_snapshot *)); > > > > > Something like > > > > > > typedef int (perf_snapshot_branch_stack_t)(struct perf_branch_snapshot *); > > > DECLARE_STATIC_CALL(perf_snapshot_branch_stack, perf_snapshot_branch_stack_t); > > > > > > seems to work fine. > > > > Oh urg, indeed. It wants a function type, not a function pointer type. > > I've been bitten by that before. Go with the typedef, that's the sanest. > > The below is the best I can make of it... it's a little inconsistent and > somewhat tricky, but at least the compiler yells hard if you get it > wrong. > > I can *almost* get to: DEFINE_STATIC_CALL(foo, &func), except for > ARCH_DEFINE_STATIC_CALL_TRAMP() which needs the actual function name > string for the ASM :-( > > The rest can do with a function pointer type and have it work. Seems reasonable to me.
diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index ac6fd2dabf6a2..a29649e7241cc 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -6283,8 +6283,11 @@ __init int intel_pmu_init(void) x86_pmu.lbr_nr = 0; } - if (x86_pmu.lbr_nr) + if (x86_pmu.lbr_nr) { pr_cont("%d-deep LBR, ", x86_pmu.lbr_nr); + static_call_update(perf_snapshot_branch_stack, + intel_pmu_snapshot_branch_stack); + } intel_pmu_check_extra_regs(x86_pmu.extra_regs); diff --git a/arch/x86/events/intel/lbr.c b/arch/x86/events/intel/lbr.c index 9e6d6eaeb4cb6..7d4fe1d6e79ff 100644 --- a/arch/x86/events/intel/lbr.c +++ b/arch/x86/events/intel/lbr.c @@ -1862,3 +1862,16 @@ EXPORT_SYMBOL_GPL(x86_perf_get_lbr); struct event_constraint vlbr_constraint = __EVENT_CONSTRAINT(INTEL_FIXED_VLBR_EVENT, (1ULL << INTEL_PMC_IDX_FIXED_VLBR), FIXED_EVENT_FLAGS, 1, 0, PERF_X86_EVENT_LBR_SELECT); + +int intel_pmu_snapshot_branch_stack(struct perf_branch_snapshot *br_snapshot) +{ + struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); + + intel_pmu_lbr_disable_all(); + intel_pmu_lbr_read(); + memcpy(br_snapshot->entries, cpuc->lbr_entries, + sizeof(struct perf_branch_entry) * x86_pmu.lbr_nr); + br_snapshot->nr = x86_pmu.lbr_nr; + intel_pmu_lbr_enable_all(false); + return 0; +} diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h index e3ac05c97b5e5..0f4ca25d10bf1 100644 --- a/arch/x86/events/perf_event.h +++ b/arch/x86/events/perf_event.h @@ -1379,6 +1379,8 @@ void intel_pmu_pebs_data_source_skl(bool pmem); int intel_pmu_setup_lbr_filter(struct perf_event *event); +int intel_pmu_snapshot_branch_stack(struct perf_branch_snapshot *br_snapshot); + void intel_pt_interrupt(void); int intel_bts_interrupt(void); diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index fe156a8170aa3..f029eb4b2ce40 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -57,6 +57,7 @@ struct perf_guest_info_callbacks { #include <linux/cgroup.h> #include <linux/refcount.h> #include <linux/security.h> +#include <linux/static_call.h> #include <asm/local.h> struct perf_callchain_entry { @@ -1612,4 +1613,32 @@ extern void __weak arch_perf_update_userpage(struct perf_event *event, extern __weak u64 arch_perf_get_page_size(struct mm_struct *mm, unsigned long addr); #endif +/* + * Snapshot branch stack on software events. + * + * Branch stack can be very useful in understanding software events. For + * example, when a long function, e.g. sys_perf_event_open, returns an + * errno, it is not obvious why the function failed. Branch stack could + * provide very helpful information in this type of scenarios. + * + * On software event, it is necessary to stop the hardware branch recorder + * fast. Otherwise, the hardware register/buffer will be flushed with + * entries af the triggering event. Therefore, static call is used to + * stop the hardware recorder. + * + * To use the snapshot: + * 1) After the event triggers, call perf_snapshot_branch_stack asap; + * 2) On the same cpu, access the snapshot with perf_read_branch_snapshot; + */ +#define MAX_BRANCH_SNAPSHOT 32 + +struct perf_branch_snapshot { + unsigned int nr; + struct perf_branch_entry entries[MAX_BRANCH_SNAPSHOT]; +}; + +int dummy_perf_snapshot_branch_stack(struct perf_branch_snapshot *br_snapshot); + +DECLARE_STATIC_CALL(perf_snapshot_branch_stack, dummy_perf_snapshot_branch_stack); + #endif /* _LINUX_PERF_EVENT_H */ diff --git a/kernel/events/core.c b/kernel/events/core.c index 011cc5069b7ba..c53fe90e630ac 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -13437,3 +13437,6 @@ struct cgroup_subsys perf_event_cgrp_subsys = { .threaded = true, }; #endif /* CONFIG_CGROUP_PERF */ + +DEFINE_STATIC_CALL_NULL(perf_snapshot_branch_stack, + dummy_perf_snapshot_branch_stack);
The typical way to access branch record (e.g. Intel LBR) is via hardware perf_event. For CPUs with FREEZE_LBRS_ON_PMI support, PMI could capture reliable LBR. On the other hand, LBR could also be useful in non-PMI scenario. For example, in kretprobe or bpf fexit program, LBR could provide a lot of information on what happened with the function. Add API to use branch record for software use. Note that, when the software event triggers, it is necessary to stop the branch record hardware asap. Therefore, static_call is used to remove some branch instructions in this process. Signed-off-by: Song Liu <songliubraving@fb.com> --- Some data on intel_pmu_lbr_disable_all() and perf_pmu_disable(). With this patch, when fexit program triggers, intel_pmu_lbr_disable_all is used to stop the LBR, and the LBR is stopped after 6 extra branch records (see the full trace below). If we replace intel_pmu_lbr_disable_all in intel_pmu_snapshot_branch_stack() with perf_pmu_disable, the LBR is stopped after 19 extra branch records. This is still acceptable for systems with 32 LBR entries. But for systems with fewer entries, all the entries before fexit are flushed. Therefore, I suggest we take the short cut and stop LBR asap. LBR snapshot captured when we use intel_pmu_lbr_disable_all(): ID: 0 from intel_pmu_lbr_disable_all.part.10+37 to intel_pmu_lbr_disable_all.part.10+72 ID: 1 from intel_pmu_lbr_disable_all.part.10+33 to intel_pmu_lbr_disable_all.part.10+37 ID: 2 from intel_pmu_snapshot_branch_stack+51 to intel_pmu_lbr_disable_all.part.10+0 ID: 3 from __bpf_prog_enter+53 to intel_pmu_snapshot_branch_stack+0 ID: 4 from __bpf_prog_enter+8 to __bpf_prog_enter+38 ID: 5 from __brk_limit+473903158 to __bpf_prog_enter+0 ID: 6 from bpf_fexit_loop_test1+22 to __brk_limit+473903139 ID: 7 from bpf_fexit_loop_test1+20 to bpf_fexit_loop_test1+13 ID: 8 from bpf_fexit_loop_test1+20 to bpf_fexit_loop_test1+13 ID: 9 from bpf_fexit_loop_test1+20 to bpf_fexit_loop_test1+13 LBR snapshot captured when we use perf_pmu_disable(): ID: 0 from intel_pmu_lbr_disable_all+58 to intel_pmu_lbr_disable_all+93 ID: 1 from intel_pmu_lbr_disable_all+54 to intel_pmu_lbr_disable_all+58 ID: 2 from intel_pmu_disable_all+15 to intel_pmu_lbr_disable_all+0 ID: 3 from intel_pmu_pebs_disable_all+30 to intel_pmu_disable_all+15 ID: 4 from intel_pmu_disable_all+10 to intel_pmu_pebs_disable_all+0 ID: 5 from __intel_pmu_disable_all+49 to intel_pmu_disable_all+10 ID: 6 from intel_pmu_disable_all+5 to __intel_pmu_disable_all+0 ID: 7 from x86_pmu_disable+61 to intel_pmu_disable_all+0 ID: 8 from x86_pmu_disable+38 to x86_pmu_disable+41 ID: 9 from __x86_indirect_thunk_rax+16 to x86_pmu_disable+0 ID: 10 from __x86_indirect_thunk_rax+0 to __x86_indirect_thunk_rax+12 ID: 11 from perf_pmu_disable.part.122+4 to __x86_indirect_thunk_rax+0 ID: 12 from perf_pmu_disable+23 to perf_pmu_disable.part.122+0 ID: 13 from intel_pmu_snapshot_branch_stack+45 to perf_pmu_disable+0 ID: 14 from x86_get_pmu+35 to intel_pmu_snapshot_branch_stack+39 ID: 15 from intel_pmu_snapshot_branch_stack+34 to x86_get_pmu+0 ID: 16 from __bpf_prog_enter+53 to intel_pmu_snapshot_branch_stack+0 ID: 17 from __bpf_prog_enter+8 to __bpf_prog_enter+38 ID: 18 from __brk_limit+478056502 to __bpf_prog_enter+0 ID: 19 from bpf_fexit_loop_test1+22 to __brk_limit+478056483 ID: 20 from bpf_fexit_loop_test1+20 to bpf_fexit_loop_test1+13 ID: 21 from bpf_fexit_loop_test1+20 to bpf_fexit_loop_test1+13 --- arch/x86/events/intel/core.c | 5 ++++- arch/x86/events/intel/lbr.c | 13 +++++++++++++ arch/x86/events/perf_event.h | 2 ++ include/linux/perf_event.h | 29 +++++++++++++++++++++++++++++ kernel/events/core.c | 3 +++ 5 files changed, 51 insertions(+), 1 deletion(-) -- 2.30.2