From aeb8fe0283d4d3b0f27a87c5f5c938e7324f7d8f Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Fri, 31 May 2024 21:45:00 +0200 Subject: bpf: Fix bpf_session_cookie BTF_ID in special_kfunc_set list The bpf_session_cookie is unavailable for !CONFIG_FPROBE as reported by Sebastian [1]. To fix that we remove CONFIG_FPROBE ifdef for session kfuncs, which is fine, because there's filter for session programs. Then based on bpf_trace.o dependency: obj-$(CONFIG_BPF_EVENTS) += bpf_trace.o we add bpf_session_cookie BTF_ID in special_kfunc_set list dependency on CONFIG_BPF_EVENTS. [1] https://lore.kernel.org/bpf/20240531071557.MvfIqkn7@linutronix.de/T/#m71c6d5ec71db2967288cb79acedc15cc5dbfeec5 Reported-by: Sebastian Andrzej Siewior Tested-by: Sebastian Andrzej Siewior Suggested-by: Alexei Starovoitov Fixes: 5c919acef8514 ("bpf: Add support for kprobe session cookie") Signed-off-by: Jiri Olsa Link: https://lore.kernel.org/r/20240531194500.2967187-1-jolsa@kernel.org Signed-off-by: Alexei Starovoitov --- kernel/trace/bpf_trace.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 6249dac61701..d1daeab1bbc1 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -3517,7 +3517,6 @@ static u64 bpf_uprobe_multi_entry_ip(struct bpf_run_ctx *ctx) } #endif /* CONFIG_UPROBES */ -#ifdef CONFIG_FPROBE __bpf_kfunc_start_defs(); __bpf_kfunc bool bpf_session_is_return(void) @@ -3566,4 +3565,3 @@ static int __init bpf_kprobe_multi_kfuncs_init(void) } late_initcall(bpf_kprobe_multi_kfuncs_init); -#endif -- cgit v1.2.3 From 42675b723b4842bca7bfb0f209aa9a493a10324a Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Mon, 3 Jun 2024 15:07:05 -0400 Subject: function_graph: Convert ret_stack to a series of longs In order to make it possible to have multiple callbacks registered with the function_graph tracer, the retstack needs to be converted from an array of ftrace_ret_stack structures to an array of longs. This will allow to store the list of callbacks on the stack for the return side of the functions. Link: https://lore.kernel.org/linux-trace-kernel/171509092742.162236.4427737821399314856.stgit@devnote2 Link: https://lore.kernel.org/linux-trace-kernel/20240603190821.073111754@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 136 ++++++++++++++++++++++++++++++-------------------- 1 file changed, 82 insertions(+), 54 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index a130b2d898f7..c62e6db718a0 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -25,6 +25,30 @@ #define ASSIGN_OPS_HASH(opsname, val) #endif +/* + * FGRAPH_FRAME_SIZE: Size in bytes of the meta data on the shadow stack + * FGRAPH_FRAME_OFFSET: Size in long words of the meta data frame + * SHADOW_STACK_SIZE: The size in bytes of the entire shadow stack + * SHADOW_STACK_OFFSET: The size in long words of the shadow stack + * SHADOW_STACK_MAX_OFFSET: The max offset of the stack for a new frame to be added + */ +#define FGRAPH_FRAME_SIZE sizeof(struct ftrace_ret_stack) +#define FGRAPH_FRAME_OFFSET (ALIGN(FGRAPH_FRAME_SIZE, sizeof(long)) / sizeof(long)) +#define SHADOW_STACK_SIZE (PAGE_SIZE) +#define SHADOW_STACK_OFFSET \ + (ALIGN(SHADOW_STACK_SIZE, sizeof(long)) / sizeof(long)) +/* Leave on a buffer at the end */ +#define SHADOW_STACK_MAX_INDEX (SHADOW_STACK_OFFSET - FGRAPH_FRAME_OFFSET) + +/* + * RET_STACK(): Return the frame from a given @offset from task @t + * RET_STACK_INC(): Reserve one frame size on the stack. + * RET_STACK_DEC(): Remove one frame size from the stack. + */ +#define RET_STACK(t, index) ((struct ftrace_ret_stack *)(&(t)->ret_stack[index])) +#define RET_STACK_INC(c) ({ c += FGRAPH_FRAME_OFFSET; }) +#define RET_STACK_DEC(c) ({ c -= FGRAPH_FRAME_OFFSET; }) + DEFINE_STATIC_KEY_FALSE(kill_ftrace_graph); int ftrace_graph_active; @@ -69,6 +93,7 @@ static int ftrace_push_return_trace(unsigned long ret, unsigned long func, unsigned long frame_pointer, unsigned long *retp) { + struct ftrace_ret_stack *ret_stack; unsigned long long calltime; int index; @@ -85,23 +110,25 @@ ftrace_push_return_trace(unsigned long ret, unsigned long func, smp_rmb(); /* The return trace stack is full */ - if (current->curr_ret_stack == FTRACE_RETFUNC_DEPTH - 1) { + if (current->curr_ret_stack >= SHADOW_STACK_MAX_INDEX) { atomic_inc(¤t->trace_overrun); return -EBUSY; } calltime = trace_clock_local(); - index = ++current->curr_ret_stack; + index = current->curr_ret_stack; + RET_STACK_INC(current->curr_ret_stack); + ret_stack = RET_STACK(current, index); barrier(); - current->ret_stack[index].ret = ret; - current->ret_stack[index].func = func; - current->ret_stack[index].calltime = calltime; + ret_stack->ret = ret; + ret_stack->func = func; + ret_stack->calltime = calltime; #ifdef HAVE_FUNCTION_GRAPH_FP_TEST - current->ret_stack[index].fp = frame_pointer; + ret_stack->fp = frame_pointer; #endif #ifdef HAVE_FUNCTION_GRAPH_RET_ADDR_PTR - current->ret_stack[index].retp = retp; + ret_stack->retp = retp; #endif return 0; } @@ -137,7 +164,7 @@ int function_graph_enter(unsigned long ret, unsigned long func, return 0; out_ret: - current->curr_ret_stack--; + RET_STACK_DEC(current->curr_ret_stack); out: current->curr_ret_depth--; return -EBUSY; @@ -148,11 +175,13 @@ static void ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret, unsigned long frame_pointer) { + struct ftrace_ret_stack *ret_stack; int index; index = current->curr_ret_stack; + RET_STACK_DEC(index); - if (unlikely(index < 0 || index >= FTRACE_RETFUNC_DEPTH)) { + if (unlikely(index < 0 || index > SHADOW_STACK_MAX_INDEX)) { ftrace_graph_stop(); WARN_ON(1); /* Might as well panic, otherwise we have no where to go */ @@ -160,6 +189,7 @@ ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret, return; } + ret_stack = RET_STACK(current, index); #ifdef HAVE_FUNCTION_GRAPH_FP_TEST /* * The arch may choose to record the frame pointer used @@ -175,22 +205,22 @@ ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret, * Note, -mfentry does not use frame pointers, and this test * is not needed if CC_USING_FENTRY is set. */ - if (unlikely(current->ret_stack[index].fp != frame_pointer)) { + if (unlikely(ret_stack->fp != frame_pointer)) { ftrace_graph_stop(); WARN(1, "Bad frame pointer: expected %lx, received %lx\n" " from func %ps return to %lx\n", current->ret_stack[index].fp, frame_pointer, - (void *)current->ret_stack[index].func, - current->ret_stack[index].ret); + (void *)ret_stack->func, + ret_stack->ret); *ret = (unsigned long)panic; return; } #endif - *ret = current->ret_stack[index].ret; - trace->func = current->ret_stack[index].func; - trace->calltime = current->ret_stack[index].calltime; + *ret = ret_stack->ret; + trace->func = ret_stack->func; + trace->calltime = ret_stack->calltime; trace->overrun = atomic_read(¤t->trace_overrun); trace->depth = current->curr_ret_depth--; /* @@ -251,7 +281,7 @@ static unsigned long __ftrace_return_to_handler(struct fgraph_ret_regs *ret_regs * curr_ret_stack is after that. */ barrier(); - current->curr_ret_stack--; + RET_STACK_DEC(current->curr_ret_stack); if (unlikely(!ret)) { ftrace_graph_stop(); @@ -294,12 +324,13 @@ unsigned long ftrace_return_to_handler(unsigned long frame_pointer) struct ftrace_ret_stack * ftrace_graph_get_ret_stack(struct task_struct *task, int idx) { - idx = task->curr_ret_stack - idx; + int index = task->curr_ret_stack; - if (idx >= 0 && idx <= task->curr_ret_stack) - return &task->ret_stack[idx]; + index -= FGRAPH_FRAME_OFFSET * (idx + 1); + if (index < 0) + return NULL; - return NULL; + return RET_STACK(task, index); } /** @@ -321,18 +352,20 @@ ftrace_graph_get_ret_stack(struct task_struct *task, int idx) unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, unsigned long ret, unsigned long *retp) { + struct ftrace_ret_stack *ret_stack; int index = task->curr_ret_stack; int i; if (ret != (unsigned long)dereference_kernel_function_descriptor(return_to_handler)) return ret; - if (index < 0) - return ret; + RET_STACK_DEC(index); - for (i = 0; i <= index; i++) - if (task->ret_stack[i].retp == retp) - return task->ret_stack[i].ret; + for (i = index; i >= 0; RET_STACK_DEC(i)) { + ret_stack = RET_STACK(task, i); + if (ret_stack->retp == retp) + return ret_stack->ret; + } return ret; } @@ -346,14 +379,15 @@ unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, return ret; task_idx = task->curr_ret_stack; + RET_STACK_DEC(task_idx); if (!task->ret_stack || task_idx < *idx) return ret; task_idx -= *idx; - (*idx)++; + RET_STACK_INC(*idx); - return task->ret_stack[task_idx].ret; + return RET_STACK(task, task_idx); } #endif /* HAVE_FUNCTION_GRAPH_RET_ADDR_PTR */ @@ -391,7 +425,7 @@ trace_func_graph_ent_t ftrace_graph_entry = ftrace_graph_entry_stub; static trace_func_graph_ent_t __ftrace_graph_entry = ftrace_graph_entry_stub; /* Try to assign a return stack array on FTRACE_RETSTACK_ALLOC_SIZE tasks. */ -static int alloc_retstack_tasklist(struct ftrace_ret_stack **ret_stack_list) +static int alloc_retstack_tasklist(unsigned long **ret_stack_list) { int i; int ret = 0; @@ -399,10 +433,7 @@ static int alloc_retstack_tasklist(struct ftrace_ret_stack **ret_stack_list) struct task_struct *g, *t; for (i = 0; i < FTRACE_RETSTACK_ALLOC_SIZE; i++) { - ret_stack_list[i] = - kmalloc_array(FTRACE_RETFUNC_DEPTH, - sizeof(struct ftrace_ret_stack), - GFP_KERNEL); + ret_stack_list[i] = kmalloc(SHADOW_STACK_SIZE, GFP_KERNEL); if (!ret_stack_list[i]) { start = 0; end = i; @@ -420,9 +451,9 @@ static int alloc_retstack_tasklist(struct ftrace_ret_stack **ret_stack_list) if (t->ret_stack == NULL) { atomic_set(&t->trace_overrun, 0); - t->curr_ret_stack = -1; + t->curr_ret_stack = 0; t->curr_ret_depth = -1; - /* Make sure the tasks see the -1 first: */ + /* Make sure the tasks see the 0 first: */ smp_wmb(); t->ret_stack = ret_stack_list[start++]; } @@ -442,6 +473,7 @@ ftrace_graph_probe_sched_switch(void *ignore, bool preempt, struct task_struct *next, unsigned int prev_state) { + struct ftrace_ret_stack *ret_stack; unsigned long long timestamp; int index; @@ -466,8 +498,11 @@ ftrace_graph_probe_sched_switch(void *ignore, bool preempt, */ timestamp -= next->ftrace_timestamp; - for (index = next->curr_ret_stack; index >= 0; index--) - next->ret_stack[index].calltime += timestamp; + for (index = next->curr_ret_stack - FGRAPH_FRAME_OFFSET; index >= 0; ) { + ret_stack = RET_STACK(next, index); + ret_stack->calltime += timestamp; + index -= FGRAPH_FRAME_OFFSET; + } } static int ftrace_graph_entry_test(struct ftrace_graph_ent *trace) @@ -510,10 +545,10 @@ void update_function_graph_func(void) ftrace_graph_entry = __ftrace_graph_entry; } -static DEFINE_PER_CPU(struct ftrace_ret_stack *, idle_ret_stack); +static DEFINE_PER_CPU(unsigned long *, idle_ret_stack); static void -graph_init_task(struct task_struct *t, struct ftrace_ret_stack *ret_stack) +graph_init_task(struct task_struct *t, unsigned long *ret_stack) { atomic_set(&t->trace_overrun, 0); t->ftrace_timestamp = 0; @@ -528,7 +563,7 @@ graph_init_task(struct task_struct *t, struct ftrace_ret_stack *ret_stack) */ void ftrace_graph_init_idle_task(struct task_struct *t, int cpu) { - t->curr_ret_stack = -1; + t->curr_ret_stack = 0; t->curr_ret_depth = -1; /* * The idle task has no parent, it either has its own @@ -538,14 +573,11 @@ void ftrace_graph_init_idle_task(struct task_struct *t, int cpu) WARN_ON(t->ret_stack != per_cpu(idle_ret_stack, cpu)); if (ftrace_graph_active) { - struct ftrace_ret_stack *ret_stack; + unsigned long *ret_stack; ret_stack = per_cpu(idle_ret_stack, cpu); if (!ret_stack) { - ret_stack = - kmalloc_array(FTRACE_RETFUNC_DEPTH, - sizeof(struct ftrace_ret_stack), - GFP_KERNEL); + ret_stack = kmalloc(SHADOW_STACK_SIZE, GFP_KERNEL); if (!ret_stack) return; per_cpu(idle_ret_stack, cpu) = ret_stack; @@ -559,15 +591,13 @@ void ftrace_graph_init_task(struct task_struct *t) { /* Make sure we do not use the parent ret_stack */ t->ret_stack = NULL; - t->curr_ret_stack = -1; + t->curr_ret_stack = 0; t->curr_ret_depth = -1; if (ftrace_graph_active) { - struct ftrace_ret_stack *ret_stack; + unsigned long *ret_stack; - ret_stack = kmalloc_array(FTRACE_RETFUNC_DEPTH, - sizeof(struct ftrace_ret_stack), - GFP_KERNEL); + ret_stack = kmalloc(SHADOW_STACK_SIZE, GFP_KERNEL); if (!ret_stack) return; graph_init_task(t, ret_stack); @@ -576,7 +606,7 @@ void ftrace_graph_init_task(struct task_struct *t) void ftrace_graph_exit_task(struct task_struct *t) { - struct ftrace_ret_stack *ret_stack = t->ret_stack; + unsigned long *ret_stack = t->ret_stack; t->ret_stack = NULL; /* NULL must become visible to IRQs before we free it: */ @@ -588,12 +618,10 @@ void ftrace_graph_exit_task(struct task_struct *t) /* Allocate a return stack for each task */ static int start_graph_tracing(void) { - struct ftrace_ret_stack **ret_stack_list; + unsigned long **ret_stack_list; int ret, cpu; - ret_stack_list = kmalloc_array(FTRACE_RETSTACK_ALLOC_SIZE, - sizeof(struct ftrace_ret_stack *), - GFP_KERNEL); + ret_stack_list = kmalloc(SHADOW_STACK_SIZE, GFP_KERNEL); if (!ret_stack_list) return -ENOMEM; -- cgit v1.2.3 From 59e5f04e4184181227889663618e01dce676e671 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Mon, 3 Jun 2024 15:07:06 -0400 Subject: fgraph: Use BUILD_BUG_ON() to make sure we have structures divisible by long Instead of using "ALIGN()", use BUILD_BUG_ON() as the structures should always be divisible by sizeof(long). Co-developed with Masami Hiramatsu: Link: https://lore.kernel.org/linux-trace-kernel/171509093949.162236.14518699447151894536.stgit@devnote2 Link: http://lkml.kernel.org/r/20190524111144.GI2589@hirez.programming.kicks-ass.net Link: https://lore.kernel.org/linux-trace-kernel/20240603190821.232168933@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Suggested-by: Peter Zijlstra Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index c62e6db718a0..fdb206aeffe3 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -33,7 +33,7 @@ * SHADOW_STACK_MAX_OFFSET: The max offset of the stack for a new frame to be added */ #define FGRAPH_FRAME_SIZE sizeof(struct ftrace_ret_stack) -#define FGRAPH_FRAME_OFFSET (ALIGN(FGRAPH_FRAME_SIZE, sizeof(long)) / sizeof(long)) +#define FGRAPH_FRAME_OFFSET DIV_ROUND_UP(FGRAPH_FRAME_SIZE, sizeof(long)) #define SHADOW_STACK_SIZE (PAGE_SIZE) #define SHADOW_STACK_OFFSET \ (ALIGN(SHADOW_STACK_SIZE, sizeof(long)) / sizeof(long)) @@ -103,6 +103,8 @@ ftrace_push_return_trace(unsigned long ret, unsigned long func, if (!current->ret_stack) return -EBUSY; + BUILD_BUG_ON(SHADOW_STACK_SIZE % sizeof(long)); + /* * We must make sure the ret_stack is tested before we read * anything else. @@ -326,6 +328,8 @@ ftrace_graph_get_ret_stack(struct task_struct *task, int idx) { int index = task->curr_ret_stack; + BUILD_BUG_ON(FGRAPH_FRAME_SIZE % sizeof(long)); + index -= FGRAPH_FRAME_OFFSET * (idx + 1); if (index < 0) return NULL; -- cgit v1.2.3 From 518d6804a865772735588c5b2203d6c0c0bf98a9 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Mon, 3 Jun 2024 15:07:07 -0400 Subject: function_graph: Add an array structure that will allow multiple callbacks Add an array structure that will eventually allow the function graph tracer to have up to 16 simultaneous callbacks attached. It's an array of 16 fgraph_ops pointers, that is assigned when one is registered. On entry of a function the entry of the first item in the array is called, and if it returns zero, then the callback returns non zero if it wants the return callback to be called on exit of the function. The array will simplify the process of having more than one callback attached to the same function, as its index into the array can be stored on the shadow stack. We need to only save the index, because this will allow the fgraph_ops to be freed before the function returns (which may happen if the function call schedule for a long time). Co-developed with Masami Hiramatsu: Link: https://lore.kernel.org/linux-trace-kernel/171509095075.162236.8272148192748284581.stgit@devnote2 Link: https://lore.kernel.org/linux-trace-kernel/20240603190821.392113213@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 114 +++++++++++++++++++++++++++++++++++--------------- 1 file changed, 81 insertions(+), 33 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index fdb206aeffe3..d2ce5d651cf0 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -52,6 +52,11 @@ DEFINE_STATIC_KEY_FALSE(kill_ftrace_graph); int ftrace_graph_active; +static int fgraph_array_cnt; +#define FGRAPH_ARRAY_SIZE 16 + +static struct fgraph_ops *fgraph_array[FGRAPH_ARRAY_SIZE]; + /* Both enabled by default (can be cleared by function_graph tracer flags */ static bool fgraph_sleep_time = true; @@ -75,6 +80,20 @@ int __weak ftrace_disable_ftrace_graph_caller(void) } #endif +int ftrace_graph_entry_stub(struct ftrace_graph_ent *trace) +{ + return 0; +} + +static void ftrace_graph_ret_stub(struct ftrace_graph_ret *trace) +{ +} + +static struct fgraph_ops fgraph_stub = { + .entryfunc = ftrace_graph_entry_stub, + .retfunc = ftrace_graph_ret_stub, +}; + /** * ftrace_graph_stop - set to permanently disable function graph tracing * @@ -161,7 +180,7 @@ int function_graph_enter(unsigned long ret, unsigned long func, goto out; /* Only trace if the calling function expects to */ - if (!ftrace_graph_entry(&trace)) + if (!fgraph_array[0]->entryfunc(&trace)) goto out_ret; return 0; @@ -276,7 +295,7 @@ static unsigned long __ftrace_return_to_handler(struct fgraph_ret_regs *ret_regs trace.retval = fgraph_ret_regs_return_value(ret_regs); #endif trace.rettime = trace_clock_local(); - ftrace_graph_return(&trace); + fgraph_array[0]->retfunc(&trace); /* * The ftrace_graph_return() may still access the current * ret_stack structure, we need to make sure the update of @@ -412,11 +431,6 @@ void ftrace_graph_sleep_time_control(bool enable) fgraph_sleep_time = enable; } -int ftrace_graph_entry_stub(struct ftrace_graph_ent *trace) -{ - return 0; -} - /* * Simply points to ftrace_stub, but with the proper protocol. * Defined by the linker script in linux/vmlinux.lds.h @@ -654,37 +668,54 @@ static int start_graph_tracing(void) int register_ftrace_graph(struct fgraph_ops *gops) { int ret = 0; + int i; mutex_lock(&ftrace_lock); - /* we currently allow only one tracer registered at a time */ - if (ftrace_graph_active) { + if (!fgraph_array[0]) { + /* The array must always have real data on it */ + for (i = 0; i < FGRAPH_ARRAY_SIZE; i++) + fgraph_array[i] = &fgraph_stub; + } + + /* Look for an available spot */ + for (i = 0; i < FGRAPH_ARRAY_SIZE; i++) { + if (fgraph_array[i] == &fgraph_stub) + break; + } + if (i >= FGRAPH_ARRAY_SIZE) { ret = -EBUSY; goto out; } - register_pm_notifier(&ftrace_suspend_notifier); + fgraph_array[i] = gops; + if (i + 1 > fgraph_array_cnt) + fgraph_array_cnt = i + 1; ftrace_graph_active++; - ret = start_graph_tracing(); - if (ret) { - ftrace_graph_active--; - goto out; - } - ftrace_graph_return = gops->retfunc; + if (ftrace_graph_active == 1) { + register_pm_notifier(&ftrace_suspend_notifier); + ret = start_graph_tracing(); + if (ret) { + ftrace_graph_active--; + goto out; + } + + ftrace_graph_return = gops->retfunc; - /* - * Update the indirect function to the entryfunc, and the - * function that gets called to the entry_test first. Then - * call the update fgraph entry function to determine if - * the entryfunc should be called directly or not. - */ - __ftrace_graph_entry = gops->entryfunc; - ftrace_graph_entry = ftrace_graph_entry_test; - update_function_graph_func(); + /* + * Update the indirect function to the entryfunc, and the + * function that gets called to the entry_test first. Then + * call the update fgraph entry function to determine if + * the entryfunc should be called directly or not. + */ + __ftrace_graph_entry = gops->entryfunc; + ftrace_graph_entry = ftrace_graph_entry_test; + update_function_graph_func(); - ret = ftrace_startup(&graph_ops, FTRACE_START_FUNC_RET); + ret = ftrace_startup(&graph_ops, FTRACE_START_FUNC_RET); + } out: mutex_unlock(&ftrace_lock); return ret; @@ -692,19 +723,36 @@ out: void unregister_ftrace_graph(struct fgraph_ops *gops) { + int i; + mutex_lock(&ftrace_lock); if (unlikely(!ftrace_graph_active)) goto out; - ftrace_graph_active--; - ftrace_graph_return = ftrace_stub_graph; - ftrace_graph_entry = ftrace_graph_entry_stub; - __ftrace_graph_entry = ftrace_graph_entry_stub; - ftrace_shutdown(&graph_ops, FTRACE_STOP_FUNC_RET); - unregister_pm_notifier(&ftrace_suspend_notifier); - unregister_trace_sched_switch(ftrace_graph_probe_sched_switch, NULL); + for (i = 0; i < fgraph_array_cnt; i++) + if (gops == fgraph_array[i]) + break; + if (i >= fgraph_array_cnt) + goto out; + fgraph_array[i] = &fgraph_stub; + if (i + 1 == fgraph_array_cnt) { + for (; i >= 0; i--) + if (fgraph_array[i] != &fgraph_stub) + break; + fgraph_array_cnt = i + 1; + } + + ftrace_graph_active--; + if (!ftrace_graph_active) { + ftrace_graph_return = ftrace_stub_graph; + ftrace_graph_entry = ftrace_graph_entry_stub; + __ftrace_graph_entry = ftrace_graph_entry_stub; + ftrace_shutdown(&graph_ops, FTRACE_STOP_FUNC_RET); + unregister_pm_notifier(&ftrace_suspend_notifier); + unregister_trace_sched_switch(ftrace_graph_probe_sched_switch, NULL); + } out: mutex_unlock(&ftrace_lock); } -- cgit v1.2.3 From 7aa1eaef9f4282c9acd39588b1fdc9dda7e73f34 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Mon, 3 Jun 2024 15:07:08 -0400 Subject: function_graph: Allow multiple users to attach to function graph Allow for multiple users to attach to function graph tracer at the same time. Only 16 simultaneous users can attach to the tracer. This is because there's an array that stores the pointers to the attached fgraph_ops. When a function being traced is entered, each of the ftrace_ops entryfunc is called and if it returns non zero, its index into the array will be added to the shadow stack. On exit of the function being traced, the shadow stack will contain the indexes of the ftrace_ops on the array that want their retfunc to be called. Because a function may sleep for a long time (if a task sleeps itself), the return of the function may be literally days later. If the ftrace_ops is removed, its place on the array is replaced with a ftrace_ops that contains the stub functions and that will be called when the function finally returns. If another ftrace_ops is added that happens to get the same index into the array, its return function may be called. But that's actually the way things current work with the old function graph tracer. If one tracer is removed and another is added, the new one will get the return calls of the function traced by the previous one, thus this is not a regression. This can be fixed by adding a counter to each time the array item is updated and save that on the shadow stack as well, such that it won't be called if the index saved does not match the index on the array. Note, being able to filter functions when both are called is not completely handled yet, but that shouldn't be too hard to manage. Co-developed with Masami Hiramatsu: Link: https://lore.kernel.org/linux-trace-kernel/171509096221.162236.8806372072523195752.stgit@devnote2 Link: https://lore.kernel.org/linux-trace-kernel/20240603190821.555493396@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 379 ++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 302 insertions(+), 77 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index d2ce5d651cf0..aae51f746828 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -7,6 +7,7 @@ * * Highly modified by Steven Rostedt (VMware). */ +#include #include #include #include @@ -28,35 +29,177 @@ /* * FGRAPH_FRAME_SIZE: Size in bytes of the meta data on the shadow stack * FGRAPH_FRAME_OFFSET: Size in long words of the meta data frame - * SHADOW_STACK_SIZE: The size in bytes of the entire shadow stack - * SHADOW_STACK_OFFSET: The size in long words of the shadow stack - * SHADOW_STACK_MAX_OFFSET: The max offset of the stack for a new frame to be added */ #define FGRAPH_FRAME_SIZE sizeof(struct ftrace_ret_stack) #define FGRAPH_FRAME_OFFSET DIV_ROUND_UP(FGRAPH_FRAME_SIZE, sizeof(long)) -#define SHADOW_STACK_SIZE (PAGE_SIZE) -#define SHADOW_STACK_OFFSET \ - (ALIGN(SHADOW_STACK_SIZE, sizeof(long)) / sizeof(long)) -/* Leave on a buffer at the end */ -#define SHADOW_STACK_MAX_INDEX (SHADOW_STACK_OFFSET - FGRAPH_FRAME_OFFSET) /* - * RET_STACK(): Return the frame from a given @offset from task @t - * RET_STACK_INC(): Reserve one frame size on the stack. - * RET_STACK_DEC(): Remove one frame size from the stack. + * On entry to a function (via function_graph_enter()), a new fgraph frame + * (ftrace_ret_stack) is pushed onto the stack as well as a word that + * holds a bitmask and a type (called "bitmap"). The bitmap is defined as: + * + * bits: 0 - 9 offset in words from the previous ftrace_ret_stack + * Currently, this will always be set to FGRAPH_FRAME_OFFSET + * to get to the fgraph frame. + * + * bits: 10 - 11 Type of storage + * 0 - reserved + * 1 - bitmap of fgraph_array index + * + * For type with "bitmap of fgraph_array index" (FGRAPH_TYPE_BITMAP): + * bits: 12 - 27 The bitmap of fgraph_ops fgraph_array index + * That is, it's a bitmask of 0-15 (16 bits) + * where if a corresponding ops in the fgraph_array[] + * expects a callback from the return of the function + * it's corresponding bit will be set. + * + * + * The top of the ret_stack (when not empty) will always have a reference + * word that points to the last fgraph frame that was saved. + * + * That is, at the end of function_graph_enter, if the first and forth + * fgraph_ops on the fgraph_array[] (index 0 and 3) needs their retfunc called + * on the return of the function being traced, this is what will be on the + * task's shadow ret_stack: (the stack grows upward) + * + * ret_stack[SHADOW_STACK_MAX_OFFSET] + * ... + * | | <- task->curr_ret_stack + * +--------------------------------------------+ + * | (9 << 12) | (1 << 10) | FGRAPH_FRAME_OFFSET| + * | *or put another way* | + * | (BIT(3)|BIT(0)) << FGRAPH_INDEX_SHIFT | \ | + * | FGRAPH_TYPE_BITMAP << FGRAPH_TYPE_SHIFT| \ | + * | (offset:FGRAPH_FRAME_OFFSET) | <- the offset is from here + * +--------------------------------------------+ + * | struct ftrace_ret_stack | + * | (stores the saved ret pointer) | <- the offset points here + * +--------------------------------------------+ + * | (X) | (N) | ( N words away from + * | | previous ret_stack) + * ... + * ret_stack[0] + * + * If a backtrace is required, and the real return pointer needs to be + * fetched, then it looks at the task's curr_ret_stack offset, if it + * is greater than zero (reserved, or right before popped), it would mask + * the value by FGRAPH_FRAME_OFFSET_MASK to get the offset of the + * ftrace_ret_stack structure stored on the shadow stack. + */ + +/* + * The following is for the top word on the stack: + * + * FGRAPH_FRAME_OFFSET (0-9) holds the offset delta to the fgraph frame + * FGRAPH_TYPE (10-11) holds the type of word this is. + * (RESERVED or BITMAP) + */ +#define FGRAPH_FRAME_OFFSET_BITS 10 +#define FGRAPH_FRAME_OFFSET_MASK GENMASK(FGRAPH_FRAME_OFFSET_BITS - 1, 0) + +#define FGRAPH_TYPE_BITS 2 +#define FGRAPH_TYPE_MASK GENMASK(FGRAPH_TYPE_BITS - 1, 0) +#define FGRAPH_TYPE_SHIFT FGRAPH_FRAME_OFFSET_BITS + +enum { + FGRAPH_TYPE_RESERVED = 0, + FGRAPH_TYPE_BITMAP = 1, +}; + +/* + * For BITMAP type: + * FGRAPH_INDEX (12-27) bits holding the gops index wanting return callback called + */ +#define FGRAPH_INDEX_BITS 16 +#define FGRAPH_INDEX_MASK GENMASK(FGRAPH_INDEX_BITS - 1, 0) +#define FGRAPH_INDEX_SHIFT (FGRAPH_TYPE_SHIFT + FGRAPH_TYPE_BITS) + +#define FGRAPH_ARRAY_SIZE FGRAPH_INDEX_BITS + +/* + * SHADOW_STACK_SIZE: The size in bytes of the entire shadow stack + * SHADOW_STACK_OFFSET: The size in long words of the shadow stack + * SHADOW_STACK_MAX_OFFSET: The max offset of the stack for a new frame to be added */ -#define RET_STACK(t, index) ((struct ftrace_ret_stack *)(&(t)->ret_stack[index])) -#define RET_STACK_INC(c) ({ c += FGRAPH_FRAME_OFFSET; }) -#define RET_STACK_DEC(c) ({ c -= FGRAPH_FRAME_OFFSET; }) +#define SHADOW_STACK_SIZE (PAGE_SIZE) +#define SHADOW_STACK_OFFSET (SHADOW_STACK_SIZE / sizeof(long)) +/* Leave on a buffer at the end */ +#define SHADOW_STACK_MAX_OFFSET (SHADOW_STACK_OFFSET - (FGRAPH_FRAME_OFFSET + 1)) + +/* RET_STACK(): Return the frame from a given @offset from task @t */ +#define RET_STACK(t, offset) ((struct ftrace_ret_stack *)(&(t)->ret_stack[offset])) DEFINE_STATIC_KEY_FALSE(kill_ftrace_graph); int ftrace_graph_active; static int fgraph_array_cnt; -#define FGRAPH_ARRAY_SIZE 16 static struct fgraph_ops *fgraph_array[FGRAPH_ARRAY_SIZE]; +/* Get the FRAME_OFFSET from the word from the @offset on ret_stack */ +static inline int get_frame_offset(struct task_struct *t, int offset) +{ + return t->ret_stack[offset] & FGRAPH_FRAME_OFFSET_MASK; +} + +/* Get FGRAPH_TYPE from the word from the @offset at ret_stack */ +static inline int get_fgraph_type(struct task_struct *t, int offset) +{ + return (t->ret_stack[offset] >> FGRAPH_TYPE_SHIFT) & FGRAPH_TYPE_MASK; +} + +/* For BITMAP type: get the bitmask from the @offset at ret_stack */ +static inline unsigned long +get_bitmap_bits(struct task_struct *t, int offset) +{ + return (t->ret_stack[offset] >> FGRAPH_INDEX_SHIFT) & FGRAPH_INDEX_MASK; +} + +/* Write the bitmap to the ret_stack at @offset (does index, offset and bitmask) */ +static inline void +set_bitmap(struct task_struct *t, int offset, unsigned long bitmap) +{ + t->ret_stack[offset] = (bitmap << FGRAPH_INDEX_SHIFT) | + (FGRAPH_TYPE_BITMAP << FGRAPH_TYPE_SHIFT) | FGRAPH_FRAME_OFFSET; +} + +/* + * @offset: The offset into @t->ret_stack to find the ret_stack entry + * @frame_offset: Where to place the offset into @t->ret_stack of that entry + * + * Returns a pointer to the previous ret_stack below @offset or NULL + * when it reaches the bottom of the stack. + * + * Calling this with: + * + * offset = task->curr_ret_stack; + * do { + * ret_stack = get_ret_stack(task, offset, &offset); + * } while (ret_stack); + * + * Will iterate through all the ret_stack entries from curr_ret_stack + * down to the first one. + */ +static inline struct ftrace_ret_stack * +get_ret_stack(struct task_struct *t, int offset, int *frame_offset) +{ + int offs; + + BUILD_BUG_ON(FGRAPH_FRAME_SIZE % sizeof(long)); + + if (unlikely(offset <= 0)) + return NULL; + + offs = get_frame_offset(t, --offset); + if (WARN_ON_ONCE(offs <= 0 || offs > offset)) + return NULL; + + offset -= offs; + + *frame_offset = offset; + return RET_STACK(t, offset); +} + /* Both enabled by default (can be cleared by function_graph tracer flags */ static bool fgraph_sleep_time = true; @@ -110,11 +253,13 @@ void ftrace_graph_stop(void) /* Add a function return address to the trace stack on thread info.*/ static int ftrace_push_return_trace(unsigned long ret, unsigned long func, - unsigned long frame_pointer, unsigned long *retp) + unsigned long frame_pointer, unsigned long *retp, + int fgraph_idx) { struct ftrace_ret_stack *ret_stack; unsigned long long calltime; - int index; + unsigned long val; + int offset; if (unlikely(ftrace_graph_is_dead())) return -EBUSY; @@ -124,24 +269,57 @@ ftrace_push_return_trace(unsigned long ret, unsigned long func, BUILD_BUG_ON(SHADOW_STACK_SIZE % sizeof(long)); + /* Set val to "reserved" with the delta to the new fgraph frame */ + val = (FGRAPH_TYPE_RESERVED << FGRAPH_TYPE_SHIFT) | FGRAPH_FRAME_OFFSET; + /* * We must make sure the ret_stack is tested before we read * anything else. */ smp_rmb(); - /* The return trace stack is full */ - if (current->curr_ret_stack >= SHADOW_STACK_MAX_INDEX) { + /* + * Check if there's room on the shadow stack to fit a fraph frame + * and a bitmap word. + */ + if (current->curr_ret_stack + FGRAPH_FRAME_OFFSET + 1 >= SHADOW_STACK_MAX_OFFSET) { atomic_inc(¤t->trace_overrun); return -EBUSY; } calltime = trace_clock_local(); - index = current->curr_ret_stack; - RET_STACK_INC(current->curr_ret_stack); - ret_stack = RET_STACK(current, index); + offset = READ_ONCE(current->curr_ret_stack); + ret_stack = RET_STACK(current, offset); + offset += FGRAPH_FRAME_OFFSET; + + /* ret offset = FGRAPH_FRAME_OFFSET ; type = reserved */ + current->ret_stack[offset] = val; + ret_stack->ret = ret; + /* + * The unwinders expect curr_ret_stack to point to either zero + * or an offset where to find the next ret_stack. Even though the + * ret stack might be bogus, we want to write the ret and the + * offset to find the ret_stack before we increment the stack point. + * If an interrupt comes in now before we increment the curr_ret_stack + * it may blow away what we wrote. But that's fine, because the + * offset will still be correct (even though the 'ret' won't be). + * What we worry about is the offset being correct after we increment + * the curr_ret_stack and before we update that offset, as if an + * interrupt comes in and does an unwind stack dump, it will need + * at least a correct offset! + */ barrier(); + WRITE_ONCE(current->curr_ret_stack, offset + 1); + /* + * This next barrier is to ensure that an interrupt coming in + * will not corrupt what we are about to write. + */ + barrier(); + + /* Still keep it reserved even if an interrupt came in */ + current->ret_stack[offset] = val; + ret_stack->ret = ret; ret_stack->func = func; ret_stack->calltime = calltime; @@ -151,7 +329,7 @@ ftrace_push_return_trace(unsigned long ret, unsigned long func, #ifdef HAVE_FUNCTION_GRAPH_RET_ADDR_PTR ret_stack->retp = retp; #endif - return 0; + return offset; } /* @@ -168,49 +346,67 @@ ftrace_push_return_trace(unsigned long ret, unsigned long func, # define MCOUNT_INSN_SIZE 0 #endif +/* If the caller does not use ftrace, call this function. */ int function_graph_enter(unsigned long ret, unsigned long func, unsigned long frame_pointer, unsigned long *retp) { struct ftrace_graph_ent trace; + unsigned long bitmap = 0; + int offset; + int i; trace.func = func; trace.depth = ++current->curr_ret_depth; - if (ftrace_push_return_trace(ret, func, frame_pointer, retp)) + offset = ftrace_push_return_trace(ret, func, frame_pointer, retp, 0); + if (offset < 0) goto out; - /* Only trace if the calling function expects to */ - if (!fgraph_array[0]->entryfunc(&trace)) + for (i = 0; i < fgraph_array_cnt; i++) { + struct fgraph_ops *gops = fgraph_array[i]; + + if (gops == &fgraph_stub) + continue; + + if (gops->entryfunc(&trace)) + bitmap |= BIT(i); + } + + if (!bitmap) goto out_ret; + /* + * Since this function uses fgraph_idx = 0 as a tail-call checking + * flag, set that bit always. + */ + set_bitmap(current, offset, bitmap | BIT(0)); + return 0; out_ret: - RET_STACK_DEC(current->curr_ret_stack); + current->curr_ret_stack -= FGRAPH_FRAME_OFFSET + 1; out: current->curr_ret_depth--; return -EBUSY; } /* Retrieve a function return address to the trace stack on thread info.*/ -static void +static struct ftrace_ret_stack * ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret, - unsigned long frame_pointer) + unsigned long frame_pointer, int *offset) { struct ftrace_ret_stack *ret_stack; - int index; - index = current->curr_ret_stack; - RET_STACK_DEC(index); + ret_stack = get_ret_stack(current, current->curr_ret_stack, offset); - if (unlikely(index < 0 || index > SHADOW_STACK_MAX_INDEX)) { + if (unlikely(!ret_stack)) { ftrace_graph_stop(); - WARN_ON(1); + WARN(1, "Bad function graph ret_stack pointer: %d", + current->curr_ret_stack); /* Might as well panic, otherwise we have no where to go */ *ret = (unsigned long)panic; - return; + return NULL; } - ret_stack = RET_STACK(current, index); #ifdef HAVE_FUNCTION_GRAPH_FP_TEST /* * The arch may choose to record the frame pointer used @@ -230,26 +426,29 @@ ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret, ftrace_graph_stop(); WARN(1, "Bad frame pointer: expected %lx, received %lx\n" " from func %ps return to %lx\n", - current->ret_stack[index].fp, + ret_stack->fp, frame_pointer, (void *)ret_stack->func, ret_stack->ret); *ret = (unsigned long)panic; - return; + return NULL; } #endif + *offset += FGRAPH_FRAME_OFFSET; *ret = ret_stack->ret; trace->func = ret_stack->func; trace->calltime = ret_stack->calltime; trace->overrun = atomic_read(¤t->trace_overrun); - trace->depth = current->curr_ret_depth--; + trace->depth = current->curr_ret_depth; /* * We still want to trace interrupts coming in if * max_depth is set to 1. Make sure the decrement is * seen before ftrace_graph_return. */ barrier(); + + return ret_stack; } /* @@ -287,30 +486,47 @@ struct fgraph_ret_regs; static unsigned long __ftrace_return_to_handler(struct fgraph_ret_regs *ret_regs, unsigned long frame_pointer) { + struct ftrace_ret_stack *ret_stack; struct ftrace_graph_ret trace; + unsigned long bitmap; unsigned long ret; + int offset; + int i; + + ret_stack = ftrace_pop_return_trace(&trace, &ret, frame_pointer, &offset); + + if (unlikely(!ret_stack)) { + ftrace_graph_stop(); + WARN_ON(1); + /* Might as well panic. What else to do? */ + return (unsigned long)panic; + } - ftrace_pop_return_trace(&trace, &ret, frame_pointer); + trace.rettime = trace_clock_local(); #ifdef CONFIG_FUNCTION_GRAPH_RETVAL trace.retval = fgraph_ret_regs_return_value(ret_regs); #endif - trace.rettime = trace_clock_local(); - fgraph_array[0]->retfunc(&trace); + + bitmap = get_bitmap_bits(current, offset); + for (i = 0; i < FGRAPH_ARRAY_SIZE; i++) { + struct fgraph_ops *gops = fgraph_array[i]; + + if (!(bitmap & BIT(i))) + continue; + if (gops == &fgraph_stub) + continue; + + gops->retfunc(&trace); + } + /* * The ftrace_graph_return() may still access the current * ret_stack structure, we need to make sure the update of * curr_ret_stack is after that. */ barrier(); - RET_STACK_DEC(current->curr_ret_stack); - - if (unlikely(!ret)) { - ftrace_graph_stop(); - WARN_ON(1); - /* Might as well panic. What else to do? */ - ret = (unsigned long)panic; - } - + current->curr_ret_stack -= FGRAPH_FRAME_OFFSET + 1; + current->curr_ret_depth--; return ret; } @@ -333,7 +549,7 @@ unsigned long ftrace_return_to_handler(unsigned long frame_pointer) /** * ftrace_graph_get_ret_stack - return the entry of the shadow stack - * @task: The task to read the shadow stack from + * @task: The task to read the shadow stack from. * @idx: Index down the shadow stack * * Return the ret_struct on the shadow stack of the @task at the @@ -345,15 +561,17 @@ unsigned long ftrace_return_to_handler(unsigned long frame_pointer) struct ftrace_ret_stack * ftrace_graph_get_ret_stack(struct task_struct *task, int idx) { - int index = task->curr_ret_stack; + struct ftrace_ret_stack *ret_stack = NULL; + int offset = task->curr_ret_stack; - BUILD_BUG_ON(FGRAPH_FRAME_SIZE % sizeof(long)); - - index -= FGRAPH_FRAME_OFFSET * (idx + 1); - if (index < 0) + if (offset < 0) return NULL; - return RET_STACK(task, index); + do { + ret_stack = get_ret_stack(task, offset, &offset); + } while (ret_stack && --idx >= 0); + + return ret_stack; } /** @@ -376,16 +594,15 @@ unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, unsigned long ret, unsigned long *retp) { struct ftrace_ret_stack *ret_stack; - int index = task->curr_ret_stack; - int i; + int i = task->curr_ret_stack; if (ret != (unsigned long)dereference_kernel_function_descriptor(return_to_handler)) return ret; - RET_STACK_DEC(index); - - for (i = index; i >= 0; RET_STACK_DEC(i)) { - ret_stack = RET_STACK(task, i); + while (i > 0) { + ret_stack = get_ret_stack(current, i, &i); + if (!ret_stack) + break; if (ret_stack->retp == retp) return ret_stack->ret; } @@ -396,21 +613,26 @@ unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, unsigned long ret, unsigned long *retp) { - int task_idx; + struct ftrace_ret_stack *ret_stack; + int offset = task->curr_ret_stack; + int i; if (ret != (unsigned long)dereference_kernel_function_descriptor(return_to_handler)) return ret; - task_idx = task->curr_ret_stack; - RET_STACK_DEC(task_idx); - - if (!task->ret_stack || task_idx < *idx) + if (!idx) return ret; - task_idx -= *idx; - RET_STACK_INC(*idx); + i = *idx; + do { + ret_stack = get_ret_stack(task, offset, &offset); + i--; + } while (i >= 0 && ret_stack); + + if (ret_stack) + return ret_stack->ret; - return RET_STACK(task, task_idx); + return ret; } #endif /* HAVE_FUNCTION_GRAPH_RET_ADDR_PTR */ @@ -493,7 +715,7 @@ ftrace_graph_probe_sched_switch(void *ignore, bool preempt, { struct ftrace_ret_stack *ret_stack; unsigned long long timestamp; - int index; + int offset; /* * Does the user want to count the time a function was asleep. @@ -516,10 +738,10 @@ ftrace_graph_probe_sched_switch(void *ignore, bool preempt, */ timestamp -= next->ftrace_timestamp; - for (index = next->curr_ret_stack - FGRAPH_FRAME_OFFSET; index >= 0; ) { - ret_stack = RET_STACK(next, index); - ret_stack->calltime += timestamp; - index -= FGRAPH_FRAME_OFFSET; + for (offset = next->curr_ret_stack; offset > 0; ) { + ret_stack = get_ret_stack(next, offset, &offset); + if (ret_stack) + ret_stack->calltime += timestamp; } } @@ -570,6 +792,8 @@ graph_init_task(struct task_struct *t, unsigned long *ret_stack) { atomic_set(&t->trace_overrun, 0); t->ftrace_timestamp = 0; + t->curr_ret_stack = 0; + t->curr_ret_depth = -1; /* make curr_ret_stack visible before we add the ret_stack */ smp_wmb(); t->ret_stack = ret_stack; @@ -691,6 +915,7 @@ int register_ftrace_graph(struct fgraph_ops *gops) fgraph_array[i] = gops; if (i + 1 > fgraph_array_cnt) fgraph_array_cnt = i + 1; + gops->idx = i; ftrace_graph_active++; -- cgit v1.2.3 From 375bb57292f49fa0956cc2739c81749b88e69510 Mon Sep 17 00:00:00 2001 From: "Masami Hiramatsu (Google)" Date: Mon, 3 Jun 2024 15:07:09 -0400 Subject: function_graph: Handle tail calls for stack unwinding For the tail-call, there would be 2 or more ftrace_ret_stacks on the ret_stack, which records "return_to_handler" as the return address except for the last one. But on the real stack, there should be 1 entry because tail-call reuses the return address on the stack and jump to the next function. In ftrace_graph_ret_addr() that is used for stack unwinding, skip tail calls as a real stack unwinder would do. Link: https://lore.kernel.org/linux-trace-kernel/171509096221.162236.8806372072523195752.stgit@devnote2 Link: https://lore.kernel.org/linux-trace-kernel/20240603190821.717065217@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index aae51f746828..8de2a2662281 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -594,16 +594,26 @@ unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, unsigned long ret, unsigned long *retp) { struct ftrace_ret_stack *ret_stack; + unsigned long return_handler = (unsigned long)dereference_kernel_function_descriptor(return_to_handler); int i = task->curr_ret_stack; - if (ret != (unsigned long)dereference_kernel_function_descriptor(return_to_handler)) + if (ret != return_handler) return ret; while (i > 0) { ret_stack = get_ret_stack(current, i, &i); if (!ret_stack) break; - if (ret_stack->retp == retp) + /* + * For the tail-call, there would be 2 or more ftrace_ret_stacks on + * the ret_stack, which records "return_to_handler" as the return + * address except for the last one. + * But on the real stack, there should be 1 entry because tail-call + * reuses the return address on the stack and jump to the next function. + * Thus we will continue to find real return address. + */ + if (ret_stack->retp == retp && + ret_stack->ret != return_handler) return ret_stack->ret; } @@ -614,10 +624,11 @@ unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, unsigned long ret, unsigned long *retp) { struct ftrace_ret_stack *ret_stack; + unsigned long return_handler = (unsigned long)dereference_kernel_function_descriptor(return_to_handler); int offset = task->curr_ret_stack; int i; - if (ret != (unsigned long)dereference_kernel_function_descriptor(return_to_handler)) + if (ret != return_handler) return ret; if (!idx) @@ -626,6 +637,8 @@ unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, i = *idx; do { ret_stack = get_ret_stack(task, offset, &offset); + if (ret_stack && ret_stack->ret == return_handler) + continue; i--; } while (i >= 0 && ret_stack); -- cgit v1.2.3 From 2fbb549983763b2cc32a1ab840fe59cc1822e06d Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Mon, 3 Jun 2024 15:07:10 -0400 Subject: function_graph: Remove logic around ftrace_graph_entry and return The function pointers ftrace_graph_entry and ftrace_graph_return are no longer called via the function_graph tracer. Instead, an array structure is now used that will allow for multiple users of the function_graph infrastructure. The variables are still used by the architecture code for non dynamic ftrace configs, where a test is made against them to see if they point to the default stub function or not. This is how the static function tracing knows to call into the function graph tracer infrastructure or not. Two new stub functions are made. entry_run() and return_run(). The ftrace_graph_entry and ftrace_graph_return are set to them respectively when the function graph tracer is enabled, and this will trigger the architecture specific function graph code to be executed. This also requires checking the global_ops hash for all calls into the function_graph tracer. Co-developed with Masami Hiramatsu: Link: https://lore.kernel.org/linux-trace-kernel/171509097408.162236.17387844142114638932.stgit@devnote2 Link: https://lore.kernel.org/linux-trace-kernel/20240603190821.872127216@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 67 ++++++++++-------------------------------- kernel/trace/ftrace.c | 2 -- kernel/trace/ftrace_internal.h | 2 -- 3 files changed, 15 insertions(+), 56 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index 8de2a2662281..2b52afa03ab4 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -163,6 +163,17 @@ set_bitmap(struct task_struct *t, int offset, unsigned long bitmap) (FGRAPH_TYPE_BITMAP << FGRAPH_TYPE_SHIFT) | FGRAPH_FRAME_OFFSET; } +/* ftrace_graph_entry set to this to tell some archs to run function graph */ +static int entry_run(struct ftrace_graph_ent *trace) +{ + return 0; +} + +/* ftrace_graph_return set to this to tell some archs to run function graph */ +static void return_run(struct ftrace_graph_ret *trace) +{ +} + /* * @offset: The offset into @t->ret_stack to find the ret_stack entry * @frame_offset: Where to place the offset into @t->ret_stack of that entry @@ -675,7 +686,6 @@ extern void ftrace_stub_graph(struct ftrace_graph_ret *); /* The callbacks that hook a function */ trace_func_graph_ret_t ftrace_graph_return = ftrace_stub_graph; trace_func_graph_ent_t ftrace_graph_entry = ftrace_graph_entry_stub; -static trace_func_graph_ent_t __ftrace_graph_entry = ftrace_graph_entry_stub; /* Try to assign a return stack array on FTRACE_RETSTACK_ALLOC_SIZE tasks. */ static int alloc_retstack_tasklist(unsigned long **ret_stack_list) @@ -758,46 +768,6 @@ ftrace_graph_probe_sched_switch(void *ignore, bool preempt, } } -static int ftrace_graph_entry_test(struct ftrace_graph_ent *trace) -{ - if (!ftrace_ops_test(&global_ops, trace->func, NULL)) - return 0; - return __ftrace_graph_entry(trace); -} - -/* - * The function graph tracer should only trace the functions defined - * by set_ftrace_filter and set_ftrace_notrace. If another function - * tracer ops is registered, the graph tracer requires testing the - * function against the global ops, and not just trace any function - * that any ftrace_ops registered. - */ -void update_function_graph_func(void) -{ - struct ftrace_ops *op; - bool do_test = false; - - /* - * The graph and global ops share the same set of functions - * to test. If any other ops is on the list, then - * the graph tracing needs to test if its the function - * it should call. - */ - do_for_each_ftrace_op(op, ftrace_ops_list) { - if (op != &global_ops && op != &graph_ops && - op != &ftrace_list_end) { - do_test = true; - /* in double loop, break out with goto */ - goto out; - } - } while_for_each_ftrace_op(op); - out: - if (do_test) - ftrace_graph_entry = ftrace_graph_entry_test; - else - ftrace_graph_entry = __ftrace_graph_entry; -} - static DEFINE_PER_CPU(unsigned long *, idle_ret_stack); static void @@ -939,18 +909,12 @@ int register_ftrace_graph(struct fgraph_ops *gops) ftrace_graph_active--; goto out; } - - ftrace_graph_return = gops->retfunc; - /* - * Update the indirect function to the entryfunc, and the - * function that gets called to the entry_test first. Then - * call the update fgraph entry function to determine if - * the entryfunc should be called directly or not. + * Some archs just test to see if these are not + * the default function */ - __ftrace_graph_entry = gops->entryfunc; - ftrace_graph_entry = ftrace_graph_entry_test; - update_function_graph_func(); + ftrace_graph_return = return_run; + ftrace_graph_entry = entry_run; ret = ftrace_startup(&graph_ops, FTRACE_START_FUNC_RET); } @@ -986,7 +950,6 @@ void unregister_ftrace_graph(struct fgraph_ops *gops) if (!ftrace_graph_active) { ftrace_graph_return = ftrace_stub_graph; ftrace_graph_entry = ftrace_graph_entry_stub; - __ftrace_graph_entry = ftrace_graph_entry_stub; ftrace_shutdown(&graph_ops, FTRACE_STOP_FUNC_RET); unregister_pm_notifier(&ftrace_suspend_notifier); unregister_trace_sched_switch(ftrace_graph_probe_sched_switch, NULL); diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 65208d3b5ed9..789950a4f977 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -235,8 +235,6 @@ static void update_ftrace_function(void) func = ftrace_ops_list_func; } - update_function_graph_func(); - /* If there's no change, then do nothing more here */ if (ftrace_trace_function == func) return; diff --git a/kernel/trace/ftrace_internal.h b/kernel/trace/ftrace_internal.h index 5012c04f92c0..19eddcb91584 100644 --- a/kernel/trace/ftrace_internal.h +++ b/kernel/trace/ftrace_internal.h @@ -42,10 +42,8 @@ ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip, void *regs) #ifdef CONFIG_FUNCTION_GRAPH_TRACER extern int ftrace_graph_active; -void update_function_graph_func(void); #else /* !CONFIG_FUNCTION_GRAPH_TRACER */ # define ftrace_graph_active 0 -static inline void update_function_graph_func(void) { } #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ #else /* !CONFIG_FUNCTION_TRACER */ -- cgit v1.2.3 From 37238abe3cb47b8daaa8706c9949f67b2a705cf1 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Mon, 3 Jun 2024 15:07:11 -0400 Subject: ftrace/function_graph: Pass fgraph_ops to function graph callbacks Pass the fgraph_ops structure to the function graph callbacks. This will allow callbacks to add a descriptor to a fgraph_ops private field that wil be added in the future and use it for the callbacks. This will be useful when more than one callback can be registered to the function graph tracer. Co-developed with Masami Hiramatsu: Link: https://lore.kernel.org/linux-trace-kernel/171509098588.162236.4787930115997357578.stgit@devnote2 Link: https://lore.kernel.org/linux-trace-kernel/20240603190822.035147698@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 16 +++++++++------- kernel/trace/ftrace.c | 6 ++++-- kernel/trace/trace.h | 4 ++-- kernel/trace/trace_functions_graph.c | 11 +++++++---- kernel/trace/trace_irqsoff.c | 6 ++++-- kernel/trace/trace_sched_wakeup.c | 6 ++++-- kernel/trace/trace_selftest.c | 5 +++-- 7 files changed, 33 insertions(+), 21 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index 2b52afa03ab4..54ed2ed2036b 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -164,13 +164,13 @@ set_bitmap(struct task_struct *t, int offset, unsigned long bitmap) } /* ftrace_graph_entry set to this to tell some archs to run function graph */ -static int entry_run(struct ftrace_graph_ent *trace) +static int entry_run(struct ftrace_graph_ent *trace, struct fgraph_ops *ops) { return 0; } /* ftrace_graph_return set to this to tell some archs to run function graph */ -static void return_run(struct ftrace_graph_ret *trace) +static void return_run(struct ftrace_graph_ret *trace, struct fgraph_ops *ops) { } @@ -234,12 +234,14 @@ int __weak ftrace_disable_ftrace_graph_caller(void) } #endif -int ftrace_graph_entry_stub(struct ftrace_graph_ent *trace) +int ftrace_graph_entry_stub(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops) { return 0; } -static void ftrace_graph_ret_stub(struct ftrace_graph_ret *trace) +static void ftrace_graph_ret_stub(struct ftrace_graph_ret *trace, + struct fgraph_ops *gops) { } @@ -379,7 +381,7 @@ int function_graph_enter(unsigned long ret, unsigned long func, if (gops == &fgraph_stub) continue; - if (gops->entryfunc(&trace)) + if (gops->entryfunc(&trace, gops)) bitmap |= BIT(i); } @@ -527,7 +529,7 @@ static unsigned long __ftrace_return_to_handler(struct fgraph_ret_regs *ret_regs if (gops == &fgraph_stub) continue; - gops->retfunc(&trace); + gops->retfunc(&trace, gops); } /* @@ -681,7 +683,7 @@ void ftrace_graph_sleep_time_control(bool enable) * Simply points to ftrace_stub, but with the proper protocol. * Defined by the linker script in linux/vmlinux.lds.h */ -extern void ftrace_stub_graph(struct ftrace_graph_ret *); +void ftrace_stub_graph(struct ftrace_graph_ret *trace, struct fgraph_ops *gops); /* The callbacks that hook a function */ trace_func_graph_ret_t ftrace_graph_return = ftrace_stub_graph; diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 789950a4f977..d18387c0642d 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -815,7 +815,8 @@ void ftrace_graph_graph_time_control(bool enable) fgraph_graph_time = enable; } -static int profile_graph_entry(struct ftrace_graph_ent *trace) +static int profile_graph_entry(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops) { struct ftrace_ret_stack *ret_stack; @@ -832,7 +833,8 @@ static int profile_graph_entry(struct ftrace_graph_ent *trace) return 1; } -static void profile_graph_return(struct ftrace_graph_ret *trace) +static void profile_graph_return(struct ftrace_graph_ret *trace, + struct fgraph_ops *gops) { struct ftrace_ret_stack *ret_stack; struct ftrace_profile_stat *stat; diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 749a182dab48..2575ec243350 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -679,8 +679,8 @@ void trace_latency_header(struct seq_file *m); void trace_default_header(struct seq_file *m); void print_trace_header(struct seq_file *m, struct trace_iterator *iter); -void trace_graph_return(struct ftrace_graph_ret *trace); -int trace_graph_entry(struct ftrace_graph_ent *trace); +void trace_graph_return(struct ftrace_graph_ret *trace, struct fgraph_ops *gops); +int trace_graph_entry(struct ftrace_graph_ent *trace, struct fgraph_ops *gops); void set_graph_array(struct trace_array *tr); void tracing_start_cmdline_record(void); diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index c35fbaab2a47..b7b142b65299 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -129,7 +129,8 @@ static inline int ftrace_graph_ignore_irqs(void) return in_hardirq(); } -int trace_graph_entry(struct ftrace_graph_ent *trace) +int trace_graph_entry(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops) { struct trace_array *tr = graph_array; struct trace_array_cpu *data; @@ -238,7 +239,8 @@ void __trace_graph_return(struct trace_array *tr, trace_buffer_unlock_commit_nostack(buffer, event); } -void trace_graph_return(struct ftrace_graph_ret *trace) +void trace_graph_return(struct ftrace_graph_ret *trace, + struct fgraph_ops *gops) { struct trace_array *tr = graph_array; struct trace_array_cpu *data; @@ -275,7 +277,8 @@ void set_graph_array(struct trace_array *tr) smp_mb(); } -static void trace_graph_thresh_return(struct ftrace_graph_ret *trace) +static void trace_graph_thresh_return(struct ftrace_graph_ret *trace, + struct fgraph_ops *gops) { ftrace_graph_addr_finish(trace); @@ -288,7 +291,7 @@ static void trace_graph_thresh_return(struct ftrace_graph_ret *trace) (trace->rettime - trace->calltime < tracing_thresh)) return; else - trace_graph_return(trace); + trace_graph_return(trace, gops); } static struct fgraph_ops funcgraph_thresh_ops = { diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index ba37f768e2f2..5478f4c4f708 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -175,7 +175,8 @@ static int irqsoff_display_graph(struct trace_array *tr, int set) return start_irqsoff_tracer(irqsoff_trace, set); } -static int irqsoff_graph_entry(struct ftrace_graph_ent *trace) +static int irqsoff_graph_entry(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops) { struct trace_array *tr = irqsoff_trace; struct trace_array_cpu *data; @@ -205,7 +206,8 @@ static int irqsoff_graph_entry(struct ftrace_graph_ent *trace) return ret; } -static void irqsoff_graph_return(struct ftrace_graph_ret *trace) +static void irqsoff_graph_return(struct ftrace_graph_ret *trace, + struct fgraph_ops *gops) { struct trace_array *tr = irqsoff_trace; struct trace_array_cpu *data; diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 0469a04a355f..49bcc812652c 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -112,7 +112,8 @@ static int wakeup_display_graph(struct trace_array *tr, int set) return start_func_tracer(tr, set); } -static int wakeup_graph_entry(struct ftrace_graph_ent *trace) +static int wakeup_graph_entry(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops) { struct trace_array *tr = wakeup_trace; struct trace_array_cpu *data; @@ -141,7 +142,8 @@ static int wakeup_graph_entry(struct ftrace_graph_ent *trace) return ret; } -static void wakeup_graph_return(struct ftrace_graph_ret *trace) +static void wakeup_graph_return(struct ftrace_graph_ret *trace, + struct fgraph_ops *gops) { struct trace_array *tr = wakeup_trace; struct trace_array_cpu *data; diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index e9c5058a8efd..56f269c0560a 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -762,7 +762,8 @@ trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr) static unsigned int graph_hang_thresh; /* Wrap the real function entry probe to avoid possible hanging */ -static int trace_graph_entry_watchdog(struct ftrace_graph_ent *trace) +static int trace_graph_entry_watchdog(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops) { /* This is harmlessly racy, we want to approximately detect a hang */ if (unlikely(++graph_hang_thresh > GRAPH_MAX_FUNC_TEST)) { @@ -776,7 +777,7 @@ static int trace_graph_entry_watchdog(struct ftrace_graph_ent *trace) return 0; } - return trace_graph_entry(trace); + return trace_graph_entry(trace, gops); } static struct fgraph_ops fgraph_ops __initdata = { -- cgit v1.2.3 From 26dda5631d1bb2f254f4c94aa87ee6c92a89cfdb Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Mon, 3 Jun 2024 15:07:12 -0400 Subject: ftrace: Allow function_graph tracer to be enabled in instances Now that function graph tracing can handle more than one user, allow it to be enabled in the ftrace instances. Note, the filtering of the functions is still joined by the top level set_ftrace_filter and friends, as well as the graph and nograph files. Co-developed with Masami Hiramatsu: Link: https://lore.kernel.org/linux-trace-kernel/171509099743.162236.1699959255446248163.stgit@devnote2 Link: https://lore.kernel.org/linux-trace-kernel/20240603190822.190630762@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ftrace.c | 1 + kernel/trace/trace.h | 13 +++++++- kernel/trace/trace_functions.c | 8 +++++ kernel/trace/trace_functions_graph.c | 65 ++++++++++++++++++++++-------------- kernel/trace/trace_selftest.c | 4 +-- 5 files changed, 63 insertions(+), 28 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index d18387c0642d..b85f00b0ffe7 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -7327,6 +7327,7 @@ __init void ftrace_init_global_array_ops(struct trace_array *tr) tr->ops = &global_ops; tr->ops->private = tr; ftrace_init_trace_array(tr); + init_array_fgraph_ops(tr); } void ftrace_init_array_ops(struct trace_array *tr, ftrace_func_t func) diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 2575ec243350..a5070f9b977b 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -397,6 +397,9 @@ struct trace_array { struct ftrace_ops *ops; struct trace_pid_list __rcu *function_pids; struct trace_pid_list __rcu *function_no_pids; +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + struct fgraph_ops *gops; +#endif #ifdef CONFIG_DYNAMIC_FTRACE /* All of these are protected by the ftrace_lock */ struct list_head func_probes; @@ -681,7 +684,6 @@ void print_trace_header(struct seq_file *m, struct trace_iterator *iter); void trace_graph_return(struct ftrace_graph_ret *trace, struct fgraph_ops *gops); int trace_graph_entry(struct ftrace_graph_ent *trace, struct fgraph_ops *gops); -void set_graph_array(struct trace_array *tr); void tracing_start_cmdline_record(void); void tracing_stop_cmdline_record(void); @@ -892,6 +894,9 @@ extern int __trace_graph_entry(struct trace_array *tr, extern void __trace_graph_return(struct trace_array *tr, struct ftrace_graph_ret *trace, unsigned int trace_ctx); +extern void init_array_fgraph_ops(struct trace_array *tr); +extern int allocate_fgraph_ops(struct trace_array *tr); +extern void free_fgraph_ops(struct trace_array *tr); #ifdef CONFIG_DYNAMIC_FTRACE extern struct ftrace_hash __rcu *ftrace_graph_hash; @@ -1004,6 +1009,12 @@ print_graph_function_flags(struct trace_iterator *iter, u32 flags) { return TRACE_TYPE_UNHANDLED; } +static inline void init_array_fgraph_ops(struct trace_array *tr) { } +static inline int allocate_fgraph_ops(struct trace_array *tr) +{ + return 0; +} +static inline void free_fgraph_ops(struct trace_array *tr) { } #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ extern struct list_head ftrace_pids; diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c index 9f1bfbe105e8..8e8da0d0ee52 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -80,6 +80,7 @@ void ftrace_free_ftrace_ops(struct trace_array *tr) int ftrace_create_function_files(struct trace_array *tr, struct dentry *parent) { + int ret; /* * The top level array uses the "global_ops", and the files are * created on boot up. @@ -90,6 +91,12 @@ int ftrace_create_function_files(struct trace_array *tr, if (!tr->ops) return -EINVAL; + ret = allocate_fgraph_ops(tr); + if (ret) { + kfree(tr->ops); + return ret; + } + ftrace_create_filter_files(tr->ops, parent); return 0; @@ -99,6 +106,7 @@ void ftrace_destroy_function_files(struct trace_array *tr) { ftrace_destroy_filter_files(tr->ops); ftrace_free_ftrace_ops(tr); + free_fgraph_ops(tr); } static ftrace_func_t select_trace_function(u32 flags_val) diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index b7b142b65299..9ccc904a7703 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -83,8 +83,6 @@ static struct tracer_flags tracer_flags = { .opts = trace_opts }; -static struct trace_array *graph_array; - /* * DURATION column is being also used to display IRQ signs, * following values are used by print_graph_irq and others @@ -132,7 +130,7 @@ static inline int ftrace_graph_ignore_irqs(void) int trace_graph_entry(struct ftrace_graph_ent *trace, struct fgraph_ops *gops) { - struct trace_array *tr = graph_array; + struct trace_array *tr = gops->private; struct trace_array_cpu *data; unsigned long flags; unsigned int trace_ctx; @@ -242,7 +240,7 @@ void __trace_graph_return(struct trace_array *tr, void trace_graph_return(struct ftrace_graph_ret *trace, struct fgraph_ops *gops) { - struct trace_array *tr = graph_array; + struct trace_array *tr = gops->private; struct trace_array_cpu *data; unsigned long flags; unsigned int trace_ctx; @@ -268,15 +266,6 @@ void trace_graph_return(struct ftrace_graph_ret *trace, local_irq_restore(flags); } -void set_graph_array(struct trace_array *tr) -{ - graph_array = tr; - - /* Make graph_array visible before we start tracing */ - - smp_mb(); -} - static void trace_graph_thresh_return(struct ftrace_graph_ret *trace, struct fgraph_ops *gops) { @@ -294,25 +283,53 @@ static void trace_graph_thresh_return(struct ftrace_graph_ret *trace, trace_graph_return(trace, gops); } -static struct fgraph_ops funcgraph_thresh_ops = { - .entryfunc = &trace_graph_entry, - .retfunc = &trace_graph_thresh_return, -}; - static struct fgraph_ops funcgraph_ops = { .entryfunc = &trace_graph_entry, .retfunc = &trace_graph_return, }; +int allocate_fgraph_ops(struct trace_array *tr) +{ + struct fgraph_ops *gops; + + gops = kzalloc(sizeof(*gops), GFP_KERNEL); + if (!gops) + return -ENOMEM; + + gops->entryfunc = &trace_graph_entry; + gops->retfunc = &trace_graph_return; + + tr->gops = gops; + gops->private = tr; + return 0; +} + +void free_fgraph_ops(struct trace_array *tr) +{ + kfree(tr->gops); +} + +__init void init_array_fgraph_ops(struct trace_array *tr) +{ + tr->gops = &funcgraph_ops; + funcgraph_ops.private = tr; +} + static int graph_trace_init(struct trace_array *tr) { int ret; - set_graph_array(tr); + tr->gops->entryfunc = trace_graph_entry; + if (tracing_thresh) - ret = register_ftrace_graph(&funcgraph_thresh_ops); + tr->gops->retfunc = trace_graph_thresh_return; else - ret = register_ftrace_graph(&funcgraph_ops); + tr->gops->retfunc = trace_graph_return; + + /* Make gops functions are visible before we start tracing */ + smp_mb(); + + ret = register_ftrace_graph(tr->gops); if (ret) return ret; tracing_start_cmdline_record(); @@ -323,10 +340,7 @@ static int graph_trace_init(struct trace_array *tr) static void graph_trace_reset(struct trace_array *tr) { tracing_stop_cmdline_record(); - if (tracing_thresh) - unregister_ftrace_graph(&funcgraph_thresh_ops); - else - unregister_ftrace_graph(&funcgraph_ops); + unregister_ftrace_graph(tr->gops); } static int graph_trace_update_thresh(struct trace_array *tr) @@ -1365,6 +1379,7 @@ static struct tracer graph_trace __tracer_data = { .print_header = print_graph_headers, .flags = &tracer_flags, .set_flag = func_graph_set_flag, + .allow_instances = true, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_function_graph, #endif diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index 56f269c0560a..f8f55fd79e53 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -813,7 +813,7 @@ trace_selftest_startup_function_graph(struct tracer *trace, * to detect and recover from possible hangs */ tracing_reset_online_cpus(&tr->array_buffer); - set_graph_array(tr); + fgraph_ops.private = tr; ret = register_ftrace_graph(&fgraph_ops); if (ret) { warn_failed_init_tracer(trace, ret); @@ -856,7 +856,7 @@ trace_selftest_startup_function_graph(struct tracer *trace, cond_resched(); tracing_reset_online_cpus(&tr->array_buffer); - set_graph_array(tr); + fgraph_ops.private = tr; /* * Some archs *cough*PowerPC*cough* add characters to the -- cgit v1.2.3 From 5fccc7552ccbc521bad61653ee739b1196b1bc53 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Mon, 3 Jun 2024 15:07:14 -0400 Subject: ftrace: Add subops logic to allow one ops to manage many There are cases where a single system will use a single function callback to handle multiple users. For example, to allow function_graph tracer to have multiple users where each can trace their own set of functions, it is useful to only have one ftrace_ops registered to ftrace that will call a function by the function_graph tracer to handle the multiplexing with the different registered function_graph tracers. Add a "subop_list" to the ftrace_ops that will hold a list of other ftrace_ops that the top ftrace_ops will manage. The function ftrace_startup_subops() that takes the manager ftrace_ops and a subop ftrace_ops it will manage. If there are no subops with the ftrace_ops yet, it will copy the ftrace_ops subop filters to the manager ftrace_ops and register that with ftrace_startup(), and adds the subop to its subop_list. If the manager ops already has something registered, it will then merge the new subop filters with what it has and enable the new functions that covers all the subops it has. To remove a subop, ftrace_shutdown_subops() is called which will use the subop_list of the manager ops to rebuild all the functions it needs to trace, and update the ftrace records to only call the functions it now has registered. If there are no more functions registered, it will then call ftrace_shutdown() to disable itself completely. Note, it is up to the manager ops callback to always make sure that the subops callbacks are called if its filter matches, as there are times in the update where the callback could be calling more functions than those that are currently registered. This could be updated to handle other systems other than function_graph, for example, fprobes could use this (but will need an interface to call ftrace_startup_subops()). Link: https://lore.kernel.org/linux-trace-kernel/20240603190822.508431129@goodmis.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 3 +- kernel/trace/ftrace.c | 401 ++++++++++++++++++++++++++++++++++++++++- kernel/trace/ftrace_internal.h | 1 + kernel/trace/trace.h | 1 + 4 files changed, 404 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index 54ed2ed2036b..e39042c40937 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -21,7 +21,8 @@ #ifdef CONFIG_DYNAMIC_FTRACE #define ASSIGN_OPS_HASH(opsname, val) \ .func_hash = val, \ - .local_hash.regex_lock = __MUTEX_INITIALIZER(opsname.local_hash.regex_lock), + .local_hash.regex_lock = __MUTEX_INITIALIZER(opsname.local_hash.regex_lock), \ + .subop_list = LIST_HEAD_INIT(opsname.subop_list), #else #define ASSIGN_OPS_HASH(opsname, val) #endif diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index b85f00b0ffe7..63b8d1fe1dd8 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -74,7 +74,8 @@ #ifdef CONFIG_DYNAMIC_FTRACE #define INIT_OPS_HASH(opsname) \ .func_hash = &opsname.local_hash, \ - .local_hash.regex_lock = __MUTEX_INITIALIZER(opsname.local_hash.regex_lock), + .local_hash.regex_lock = __MUTEX_INITIALIZER(opsname.local_hash.regex_lock), \ + .subop_list = LIST_HEAD_INIT(opsname.subop_list), #else #define INIT_OPS_HASH(opsname) #endif @@ -161,6 +162,7 @@ static inline void ftrace_ops_init(struct ftrace_ops *ops) #ifdef CONFIG_DYNAMIC_FTRACE if (!(ops->flags & FTRACE_OPS_FL_INITIALIZED)) { mutex_init(&ops->local_hash.regex_lock); + INIT_LIST_HEAD(&ops->subop_list); ops->func_hash = &ops->local_hash; ops->flags |= FTRACE_OPS_FL_INITIALIZED; } @@ -3164,6 +3166,403 @@ out: return 0; } +/* Simply make a copy of @src and return it */ +static struct ftrace_hash *copy_hash(struct ftrace_hash *src) +{ + if (ftrace_hash_empty(src)) + return EMPTY_HASH; + + return alloc_and_copy_ftrace_hash(src->size_bits, src); +} + +/* + * Append @new_hash entries to @hash: + * + * If @hash is the EMPTY_HASH then it traces all functions and nothing + * needs to be done. + * + * If @new_hash is the EMPTY_HASH, then make *hash the EMPTY_HASH so + * that it traces everything. + * + * Otherwise, go through all of @new_hash and add anything that @hash + * doesn't already have, to @hash. + * + * The filter_hash updates uses just the append_hash() function + * and the notrace_hash does not. + */ +static int append_hash(struct ftrace_hash **hash, struct ftrace_hash *new_hash) +{ + struct ftrace_func_entry *entry; + int size; + int i; + + /* An empty hash does everything */ + if (ftrace_hash_empty(*hash)) + return 0; + + /* If new_hash has everything make hash have everything */ + if (ftrace_hash_empty(new_hash)) { + free_ftrace_hash(*hash); + *hash = EMPTY_HASH; + return 0; + } + + size = 1 << new_hash->size_bits; + for (i = 0; i < size; i++) { + hlist_for_each_entry(entry, &new_hash->buckets[i], hlist) { + /* Only add if not already in hash */ + if (!__ftrace_lookup_ip(*hash, entry->ip) && + add_hash_entry(*hash, entry->ip) == NULL) + return -ENOMEM; + } + } + return 0; +} + +/* + * Add to @hash only those that are in both @new_hash1 and @new_hash2 + * + * The notrace_hash updates uses just the intersect_hash() function + * and the filter_hash does not. + */ +static int intersect_hash(struct ftrace_hash **hash, struct ftrace_hash *new_hash1, + struct ftrace_hash *new_hash2) +{ + struct ftrace_func_entry *entry; + int size; + int i; + + /* + * If new_hash1 or new_hash2 is the EMPTY_HASH then make the hash + * empty as well as empty for notrace means none are notraced. + */ + if (ftrace_hash_empty(new_hash1) || ftrace_hash_empty(new_hash2)) { + free_ftrace_hash(*hash); + *hash = EMPTY_HASH; + return 0; + } + + size = 1 << new_hash1->size_bits; + for (i = 0; i < size; i++) { + hlist_for_each_entry(entry, &new_hash1->buckets[i], hlist) { + /* Only add if in both @new_hash1 and @new_hash2 */ + if (__ftrace_lookup_ip(new_hash2, entry->ip) && + add_hash_entry(*hash, entry->ip) == NULL) + return -ENOMEM; + } + } + /* If nothing intersects, make it the empty set */ + if (ftrace_hash_empty(*hash)) { + free_ftrace_hash(*hash); + *hash = EMPTY_HASH; + } + return 0; +} + +/* Return a new hash that has a union of all @ops->filter_hash entries */ +static struct ftrace_hash *append_hashes(struct ftrace_ops *ops) +{ + struct ftrace_hash *new_hash; + struct ftrace_ops *subops; + int ret; + + new_hash = alloc_ftrace_hash(ops->func_hash->filter_hash->size_bits); + if (!new_hash) + return NULL; + + list_for_each_entry(subops, &ops->subop_list, list) { + ret = append_hash(&new_hash, subops->func_hash->filter_hash); + if (ret < 0) { + free_ftrace_hash(new_hash); + return NULL; + } + /* Nothing more to do if new_hash is empty */ + if (ftrace_hash_empty(new_hash)) + break; + } + return new_hash; +} + +/* Make @ops trace evenything except what all its subops do not trace */ +static struct ftrace_hash *intersect_hashes(struct ftrace_ops *ops) +{ + struct ftrace_hash *new_hash = NULL; + struct ftrace_ops *subops; + int size_bits; + int ret; + + list_for_each_entry(subops, &ops->subop_list, list) { + struct ftrace_hash *next_hash; + + if (!new_hash) { + size_bits = subops->func_hash->notrace_hash->size_bits; + new_hash = alloc_and_copy_ftrace_hash(size_bits, ops->func_hash->notrace_hash); + if (!new_hash) + return NULL; + continue; + } + size_bits = new_hash->size_bits; + next_hash = new_hash; + new_hash = alloc_ftrace_hash(size_bits); + ret = intersect_hash(&new_hash, next_hash, subops->func_hash->notrace_hash); + free_ftrace_hash(next_hash); + if (ret < 0) { + free_ftrace_hash(new_hash); + return NULL; + } + /* Nothing more to do if new_hash is empty */ + if (ftrace_hash_empty(new_hash)) + break; + } + return new_hash; +} + +static bool ops_equal(struct ftrace_hash *A, struct ftrace_hash *B) +{ + struct ftrace_func_entry *entry; + int size; + int i; + + if (ftrace_hash_empty(A)) + return ftrace_hash_empty(B); + + if (ftrace_hash_empty(B)) + return ftrace_hash_empty(A); + + if (A->count != B->count) + return false; + + size = 1 << A->size_bits; + for (i = 0; i < size; i++) { + hlist_for_each_entry(entry, &A->buckets[i], hlist) { + if (!__ftrace_lookup_ip(B, entry->ip)) + return false; + } + } + + return true; +} + +static int ftrace_hash_move_and_update_ops(struct ftrace_ops *ops, + struct ftrace_hash **orig_hash, + struct ftrace_hash *hash, + int enable); + +static int ftrace_update_ops(struct ftrace_ops *ops, struct ftrace_hash *filter_hash, + struct ftrace_hash *notrace_hash) +{ + int ret; + + if (!ops_equal(filter_hash, ops->func_hash->filter_hash)) { + ret = ftrace_hash_move_and_update_ops(ops, &ops->func_hash->filter_hash, + filter_hash, 1); + if (ret < 0) + return ret; + } + + if (!ops_equal(notrace_hash, ops->func_hash->notrace_hash)) { + ret = ftrace_hash_move_and_update_ops(ops, &ops->func_hash->notrace_hash, + notrace_hash, 0); + if (ret < 0) + return ret; + } + + return 0; +} + +/** + * ftrace_startup_subops - enable tracing for subops of an ops + * @ops: Manager ops (used to pick all the functions of its subops) + * @subops: A new ops to add to @ops + * @command: Extra commands to use to enable tracing + * + * The @ops is a manager @ops that has the filter that includes all the functions + * that its list of subops are tracing. Adding a new @subops will add the + * functions of @subops to @ops. + */ +int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command) +{ + struct ftrace_hash *filter_hash; + struct ftrace_hash *notrace_hash; + struct ftrace_hash *save_filter_hash; + struct ftrace_hash *save_notrace_hash; + int size_bits; + int ret; + + if (unlikely(ftrace_disabled)) + return -ENODEV; + + ftrace_ops_init(ops); + ftrace_ops_init(subops); + + if (WARN_ON_ONCE(subops->flags & FTRACE_OPS_FL_ENABLED)) + return -EBUSY; + + /* Make everything canonical (Just in case!) */ + if (!ops->func_hash->filter_hash) + ops->func_hash->filter_hash = EMPTY_HASH; + if (!ops->func_hash->notrace_hash) + ops->func_hash->notrace_hash = EMPTY_HASH; + if (!subops->func_hash->filter_hash) + subops->func_hash->filter_hash = EMPTY_HASH; + if (!subops->func_hash->notrace_hash) + subops->func_hash->notrace_hash = EMPTY_HASH; + + /* For the first subops to ops just enable it normally */ + if (list_empty(&ops->subop_list)) { + /* Just use the subops hashes */ + filter_hash = copy_hash(subops->func_hash->filter_hash); + notrace_hash = copy_hash(subops->func_hash->notrace_hash); + if (!filter_hash || !notrace_hash) { + free_ftrace_hash(filter_hash); + free_ftrace_hash(notrace_hash); + return -ENOMEM; + } + + save_filter_hash = ops->func_hash->filter_hash; + save_notrace_hash = ops->func_hash->notrace_hash; + + ops->func_hash->filter_hash = filter_hash; + ops->func_hash->notrace_hash = notrace_hash; + list_add(&subops->list, &ops->subop_list); + ret = ftrace_startup(ops, command); + if (ret < 0) { + list_del(&subops->list); + ops->func_hash->filter_hash = save_filter_hash; + ops->func_hash->notrace_hash = save_notrace_hash; + free_ftrace_hash(filter_hash); + free_ftrace_hash(notrace_hash); + } else { + free_ftrace_hash(save_filter_hash); + free_ftrace_hash(save_notrace_hash); + subops->flags |= FTRACE_OPS_FL_ENABLED; + } + return ret; + } + + /* + * Here there's already something attached. Here are the rules: + * o If either filter_hash is empty then the final stays empty + * o Otherwise, the final is a superset of both hashes + * o If either notrace_hash is empty then the final stays empty + * o Otherwise, the final is an intersection between the hashes + */ + if (ftrace_hash_empty(ops->func_hash->filter_hash) || + ftrace_hash_empty(subops->func_hash->filter_hash)) { + filter_hash = EMPTY_HASH; + } else { + size_bits = max(ops->func_hash->filter_hash->size_bits, + subops->func_hash->filter_hash->size_bits); + filter_hash = alloc_and_copy_ftrace_hash(size_bits, ops->func_hash->filter_hash); + if (!filter_hash) + return -ENOMEM; + ret = append_hash(&filter_hash, subops->func_hash->filter_hash); + if (ret < 0) { + free_ftrace_hash(filter_hash); + return ret; + } + } + + if (ftrace_hash_empty(ops->func_hash->notrace_hash) || + ftrace_hash_empty(subops->func_hash->notrace_hash)) { + notrace_hash = EMPTY_HASH; + } else { + size_bits = max(ops->func_hash->filter_hash->size_bits, + subops->func_hash->filter_hash->size_bits); + notrace_hash = alloc_ftrace_hash(size_bits); + if (!notrace_hash) { + free_ftrace_hash(filter_hash); + return -ENOMEM; + } + + ret = intersect_hash(¬race_hash, ops->func_hash->filter_hash, + subops->func_hash->filter_hash); + if (ret < 0) { + free_ftrace_hash(filter_hash); + free_ftrace_hash(notrace_hash); + return ret; + } + } + + list_add(&subops->list, &ops->subop_list); + + ret = ftrace_update_ops(ops, filter_hash, notrace_hash); + free_ftrace_hash(filter_hash); + free_ftrace_hash(notrace_hash); + if (ret < 0) + list_del(&subops->list); + else + subops->flags |= FTRACE_OPS_FL_ENABLED; + + return ret; +} + +/** + * ftrace_shutdown_subops - Remove a subops from a manager ops + * @ops: A manager ops to remove @subops from + * @subops: The subops to remove from @ops + * @command: Any extra command flags to add to modifying the text + * + * Removes the functions being traced by the @subops from @ops. Note, it + * will not affect functions that are being traced by other subops that + * still exist in @ops. + * + * If the last subops is removed from @ops, then @ops is shutdown normally. + */ +int ftrace_shutdown_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command) +{ + struct ftrace_hash *filter_hash; + struct ftrace_hash *notrace_hash; + int ret; + + if (unlikely(ftrace_disabled)) + return -ENODEV; + + if (WARN_ON_ONCE(!(subops->flags & FTRACE_OPS_FL_ENABLED))) + return -EINVAL; + + list_del(&subops->list); + + if (list_empty(&ops->subop_list)) { + /* Last one, just disable the current ops */ + + ret = ftrace_shutdown(ops, command); + if (ret < 0) { + list_add(&subops->list, &ops->subop_list); + return ret; + } + + subops->flags &= ~FTRACE_OPS_FL_ENABLED; + + free_ftrace_hash(ops->func_hash->filter_hash); + free_ftrace_hash(ops->func_hash->notrace_hash); + ops->func_hash->filter_hash = EMPTY_HASH; + ops->func_hash->notrace_hash = EMPTY_HASH; + + return 0; + } + + /* Rebuild the hashes without subops */ + filter_hash = append_hashes(ops); + notrace_hash = intersect_hashes(ops); + if (!filter_hash || !notrace_hash) { + free_ftrace_hash(filter_hash); + free_ftrace_hash(notrace_hash); + list_add(&subops->list, &ops->subop_list); + return -ENOMEM; + } + + ret = ftrace_update_ops(ops, filter_hash, notrace_hash); + if (ret < 0) + list_add(&subops->list, &ops->subop_list); + else + subops->flags &= ~FTRACE_OPS_FL_ENABLED; + + free_ftrace_hash(filter_hash); + free_ftrace_hash(notrace_hash); + return ret; +} + static u64 ftrace_update_time; unsigned long ftrace_update_tot_cnt; unsigned long ftrace_number_of_pages; diff --git a/kernel/trace/ftrace_internal.h b/kernel/trace/ftrace_internal.h index 19eddcb91584..cdfd12c44ab4 100644 --- a/kernel/trace/ftrace_internal.h +++ b/kernel/trace/ftrace_internal.h @@ -15,6 +15,7 @@ extern struct ftrace_ops global_ops; int ftrace_startup(struct ftrace_ops *ops, int command); int ftrace_shutdown(struct ftrace_ops *ops, int command); int ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip, void *regs); +int ftrace_shutdown_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command); #else /* !CONFIG_DYNAMIC_FTRACE */ diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index a5070f9b977b..9a70beb2cc46 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -1136,6 +1136,7 @@ extern int ftrace_set_filter(struct ftrace_ops *ops, unsigned char *buf, int len, int reset); extern int ftrace_set_notrace(struct ftrace_ops *ops, unsigned char *buf, int len, int reset); +extern int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command); #else struct ftrace_func_command; -- cgit v1.2.3 From d9bbfbd14f58d2955cc7a3efa8ae6d4e09ee5995 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Mon, 3 Jun 2024 15:07:15 -0400 Subject: ftrace: Allow subops filtering to be modified The subops filters use a "manager" ops to enable and disable its filters. The manager ops can handle more than one subops, and its filter is what controls what functions get set. Add a ftrace_hash_move_and_update_subops() function that will update the manager ops when the subops filters change. Link: https://lore.kernel.org/linux-trace-kernel/20240603190822.673932251@goodmis.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ftrace.c | 141 ++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 113 insertions(+), 28 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 63b8d1fe1dd8..cbb91b0afcc8 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -3343,10 +3343,28 @@ static bool ops_equal(struct ftrace_hash *A, struct ftrace_hash *B) return true; } -static int ftrace_hash_move_and_update_ops(struct ftrace_ops *ops, - struct ftrace_hash **orig_hash, - struct ftrace_hash *hash, - int enable); +static void ftrace_ops_update_code(struct ftrace_ops *ops, + struct ftrace_ops_hash *old_hash); + +static int __ftrace_hash_move_and_update_ops(struct ftrace_ops *ops, + struct ftrace_hash **orig_hash, + struct ftrace_hash *hash, + int enable) +{ + struct ftrace_ops_hash old_hash_ops; + struct ftrace_hash *old_hash; + int ret; + + old_hash = *orig_hash; + old_hash_ops.filter_hash = ops->func_hash->filter_hash; + old_hash_ops.notrace_hash = ops->func_hash->notrace_hash; + ret = ftrace_hash_move(ops, enable, orig_hash, hash); + if (!ret) { + ftrace_ops_update_code(ops, &old_hash_ops); + free_ftrace_hash_rcu(old_hash); + } + return ret; +} static int ftrace_update_ops(struct ftrace_ops *ops, struct ftrace_hash *filter_hash, struct ftrace_hash *notrace_hash) @@ -3354,15 +3372,15 @@ static int ftrace_update_ops(struct ftrace_ops *ops, struct ftrace_hash *filter_ int ret; if (!ops_equal(filter_hash, ops->func_hash->filter_hash)) { - ret = ftrace_hash_move_and_update_ops(ops, &ops->func_hash->filter_hash, - filter_hash, 1); + ret = __ftrace_hash_move_and_update_ops(ops, &ops->func_hash->filter_hash, + filter_hash, 1); if (ret < 0) return ret; } if (!ops_equal(notrace_hash, ops->func_hash->notrace_hash)) { - ret = ftrace_hash_move_and_update_ops(ops, &ops->func_hash->notrace_hash, - notrace_hash, 0); + ret = __ftrace_hash_move_and_update_ops(ops, &ops->func_hash->notrace_hash, + notrace_hash, 0); if (ret < 0) return ret; } @@ -3435,7 +3453,8 @@ int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int } else { free_ftrace_hash(save_filter_hash); free_ftrace_hash(save_notrace_hash); - subops->flags |= FTRACE_OPS_FL_ENABLED; + subops->flags |= FTRACE_OPS_FL_ENABLED | FTRACE_OPS_FL_SUBOP; + subops->managed = ops; } return ret; } @@ -3489,11 +3508,12 @@ int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int ret = ftrace_update_ops(ops, filter_hash, notrace_hash); free_ftrace_hash(filter_hash); free_ftrace_hash(notrace_hash); - if (ret < 0) + if (ret < 0) { list_del(&subops->list); - else - subops->flags |= FTRACE_OPS_FL_ENABLED; - + } else { + subops->flags |= FTRACE_OPS_FL_ENABLED | FTRACE_OPS_FL_SUBOP; + subops->managed = ops; + } return ret; } @@ -3538,6 +3558,8 @@ int ftrace_shutdown_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, in free_ftrace_hash(ops->func_hash->notrace_hash); ops->func_hash->filter_hash = EMPTY_HASH; ops->func_hash->notrace_hash = EMPTY_HASH; + subops->flags &= ~(FTRACE_OPS_FL_ENABLED | FTRACE_OPS_FL_SUBOP); + subops->managed = NULL; return 0; } @@ -3553,16 +3575,65 @@ int ftrace_shutdown_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, in } ret = ftrace_update_ops(ops, filter_hash, notrace_hash); - if (ret < 0) + if (ret < 0) { list_add(&subops->list, &ops->subop_list); - else - subops->flags &= ~FTRACE_OPS_FL_ENABLED; - + } else { + subops->flags &= ~(FTRACE_OPS_FL_ENABLED | FTRACE_OPS_FL_SUBOP); + subops->managed = NULL; + } free_ftrace_hash(filter_hash); free_ftrace_hash(notrace_hash); return ret; } +static int ftrace_hash_move_and_update_subops(struct ftrace_ops *subops, + struct ftrace_hash **orig_subhash, + struct ftrace_hash *hash, + int enable) +{ + struct ftrace_ops *ops = subops->managed; + struct ftrace_hash **orig_hash; + struct ftrace_hash *save_hash; + struct ftrace_hash *new_hash; + int ret; + + /* Manager ops can not be subops (yet) */ + if (WARN_ON_ONCE(!ops || ops->flags & FTRACE_OPS_FL_SUBOP)) + return -EINVAL; + + /* Move the new hash over to the subops hash */ + save_hash = *orig_subhash; + *orig_subhash = __ftrace_hash_move(hash); + if (!*orig_subhash) { + *orig_subhash = save_hash; + return -ENOMEM; + } + + /* Create a new_hash to hold the ops new functions */ + if (enable) { + orig_hash = &ops->func_hash->filter_hash; + new_hash = append_hashes(ops); + } else { + orig_hash = &ops->func_hash->notrace_hash; + new_hash = intersect_hashes(ops); + } + + /* Move the hash over to the new hash */ + ret = __ftrace_hash_move_and_update_ops(ops, orig_hash, new_hash, enable); + + free_ftrace_hash(new_hash); + + if (ret) { + /* Put back the original hash */ + free_ftrace_hash_rcu(*orig_subhash); + *orig_subhash = save_hash; + } else { + free_ftrace_hash_rcu(save_hash); + } + return ret; +} + + static u64 ftrace_update_time; unsigned long ftrace_update_tot_cnt; unsigned long ftrace_number_of_pages; @@ -4779,19 +4850,33 @@ static int ftrace_hash_move_and_update_ops(struct ftrace_ops *ops, struct ftrace_hash *hash, int enable) { - struct ftrace_ops_hash old_hash_ops; - struct ftrace_hash *old_hash; - int ret; + if (ops->flags & FTRACE_OPS_FL_SUBOP) + return ftrace_hash_move_and_update_subops(ops, orig_hash, hash, enable); - old_hash = *orig_hash; - old_hash_ops.filter_hash = ops->func_hash->filter_hash; - old_hash_ops.notrace_hash = ops->func_hash->notrace_hash; - ret = ftrace_hash_move(ops, enable, orig_hash, hash); - if (!ret) { - ftrace_ops_update_code(ops, &old_hash_ops); - free_ftrace_hash_rcu(old_hash); + /* + * If this ops is not enabled, it could be sharing its filters + * with a subop. If that's the case, update the subop instead of + * this ops. Shared filters are only allowed to have one ops set + * at a time, and if we update the ops that is not enabled, + * it will not affect subops that share it. + */ + if (!(ops->flags & FTRACE_OPS_FL_ENABLED)) { + struct ftrace_ops *op; + + /* Check if any other manager subops maps to this hash */ + do_for_each_ftrace_op(op, ftrace_ops_list) { + struct ftrace_ops *subops; + + list_for_each_entry(subops, &op->subop_list, list) { + if ((subops->flags & FTRACE_OPS_FL_ENABLED) && + subops->func_hash == ops->func_hash) { + return ftrace_hash_move_and_update_subops(subops, orig_hash, hash, enable); + } + } + } while_for_each_ftrace_op(op); } - return ret; + + return __ftrace_hash_move_and_update_ops(ops, orig_hash, hash, enable); } static bool module_exists(const char *module) -- cgit v1.2.3 From c132be2c4fcc1150ad0791c2a85dd4c9ad0bd0c8 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Mon, 3 Jun 2024 15:07:16 -0400 Subject: function_graph: Have the instances use their own ftrace_ops for filtering Allow for instances to have their own ftrace_ops part of the fgraph_ops that makes the funtion_graph tracer filter on the set_ftrace_filter file of the instance and not the top instance. This uses the new ftrace_startup_subops(), by using graph_ops as the "manager ops" that defines the callback function and adds the functions defined by the filters of the ops for each trace instance. The callback defined by the manager ops will call the registered fgraph ops that were added to the fgraph_array. Co-developed with Masami Hiramatsu: Link: https://lore.kernel.org/linux-trace-kernel/171509102088.162236.15758883237657317789.stgit@devnote2 Link: https://lore.kernel.org/linux-trace-kernel/20240603190822.832946261@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 81 +++++++++++++++++++++++------------- kernel/trace/ftrace.c | 2 +- kernel/trace/trace.h | 15 +++---- kernel/trace/trace_functions.c | 2 +- kernel/trace/trace_functions_graph.c | 8 +++- 5 files changed, 67 insertions(+), 41 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index e39042c40937..3ef6db53c0bf 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -18,15 +18,6 @@ #include "ftrace_internal.h" #include "trace.h" -#ifdef CONFIG_DYNAMIC_FTRACE -#define ASSIGN_OPS_HASH(opsname, val) \ - .func_hash = val, \ - .local_hash.regex_lock = __MUTEX_INITIALIZER(opsname.local_hash.regex_lock), \ - .subop_list = LIST_HEAD_INIT(opsname.subop_list), -#else -#define ASSIGN_OPS_HASH(opsname, val) -#endif - /* * FGRAPH_FRAME_SIZE: Size in bytes of the meta data on the shadow stack * FGRAPH_FRAME_OFFSET: Size in long words of the meta data frame @@ -156,6 +147,13 @@ get_bitmap_bits(struct task_struct *t, int offset) return (t->ret_stack[offset] >> FGRAPH_INDEX_SHIFT) & FGRAPH_INDEX_MASK; } +/* For BITMAP type: set the bits in the bitmap bitmask at @offset on ret_stack */ +static inline void +set_bitmap_bits(struct task_struct *t, int offset, unsigned long bitmap) +{ + t->ret_stack[offset] |= (bitmap << FGRAPH_INDEX_SHIFT); +} + /* Write the bitmap to the ret_stack at @offset (does index, offset and bitmask) */ static inline void set_bitmap(struct task_struct *t, int offset, unsigned long bitmap) @@ -382,7 +380,8 @@ int function_graph_enter(unsigned long ret, unsigned long func, if (gops == &fgraph_stub) continue; - if (gops->entryfunc(&trace, gops)) + if (ftrace_ops_test(&gops->ops, func, NULL) && + gops->entryfunc(&trace, gops)) bitmap |= BIT(i); } @@ -665,16 +664,28 @@ unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, static struct ftrace_ops graph_ops = { .func = ftrace_graph_func, - .flags = FTRACE_OPS_FL_INITIALIZED | - FTRACE_OPS_FL_PID | - FTRACE_OPS_GRAPH_STUB, + .flags = FTRACE_OPS_GRAPH_STUB, #ifdef FTRACE_GRAPH_TRAMP_ADDR .trampoline = FTRACE_GRAPH_TRAMP_ADDR, /* trampoline_size is only needed for dynamically allocated tramps */ #endif - ASSIGN_OPS_HASH(graph_ops, &global_ops.local_hash) }; +void fgraph_init_ops(struct ftrace_ops *dst_ops, + struct ftrace_ops *src_ops) +{ + dst_ops->flags = FTRACE_OPS_FL_PID | FTRACE_OPS_GRAPH_STUB; + +#ifdef CONFIG_DYNAMIC_FTRACE + if (src_ops) { + dst_ops->func_hash = &src_ops->local_hash; + mutex_init(&dst_ops->local_hash.regex_lock); + INIT_LIST_HEAD(&dst_ops->subop_list); + dst_ops->flags |= FTRACE_OPS_FL_INITIALIZED; + } +#endif +} + void ftrace_graph_sleep_time_control(bool enable) { fgraph_sleep_time = enable; @@ -877,6 +888,7 @@ static int start_graph_tracing(void) int register_ftrace_graph(struct fgraph_ops *gops) { + int command = 0; int ret = 0; int i; @@ -894,7 +906,7 @@ int register_ftrace_graph(struct fgraph_ops *gops) break; } if (i >= FGRAPH_ARRAY_SIZE) { - ret = -EBUSY; + ret = -ENOSPC; goto out; } @@ -908,18 +920,22 @@ int register_ftrace_graph(struct fgraph_ops *gops) if (ftrace_graph_active == 1) { register_pm_notifier(&ftrace_suspend_notifier); ret = start_graph_tracing(); - if (ret) { - ftrace_graph_active--; - goto out; - } + if (ret) + goto error; /* * Some archs just test to see if these are not * the default function */ ftrace_graph_return = return_run; ftrace_graph_entry = entry_run; + command = FTRACE_START_FUNC_RET; + } - ret = ftrace_startup(&graph_ops, FTRACE_START_FUNC_RET); + ret = ftrace_startup_subops(&graph_ops, &gops->ops, command); +error: + if (ret) { + fgraph_array[i] = &fgraph_stub; + ftrace_graph_active--; } out: mutex_unlock(&ftrace_lock); @@ -928,6 +944,7 @@ out: void unregister_ftrace_graph(struct fgraph_ops *gops) { + int command = 0; int i; mutex_lock(&ftrace_lock); @@ -935,25 +952,29 @@ void unregister_ftrace_graph(struct fgraph_ops *gops) if (unlikely(!ftrace_graph_active)) goto out; - for (i = 0; i < fgraph_array_cnt; i++) - if (gops == fgraph_array[i]) - break; - if (i >= fgraph_array_cnt) + if (unlikely(gops->idx < 0 || gops->idx >= fgraph_array_cnt)) goto out; - fgraph_array[i] = &fgraph_stub; - if (i + 1 == fgraph_array_cnt) { - for (; i >= 0; i--) - if (fgraph_array[i] != &fgraph_stub) - break; + WARN_ON_ONCE(fgraph_array[gops->idx] != gops); + + fgraph_array[gops->idx] = &fgraph_stub; + if (gops->idx + 1 == fgraph_array_cnt) { + i = gops->idx; + while (i >= 0 && fgraph_array[i] == &fgraph_stub) + i--; fgraph_array_cnt = i + 1; } ftrace_graph_active--; + + if (!ftrace_graph_active) + command = FTRACE_STOP_FUNC_RET; + + ftrace_shutdown_subops(&graph_ops, &gops->ops, command); + if (!ftrace_graph_active) { ftrace_graph_return = ftrace_stub_graph; ftrace_graph_entry = ftrace_graph_entry_stub; - ftrace_shutdown(&graph_ops, FTRACE_STOP_FUNC_RET); unregister_pm_notifier(&ftrace_suspend_notifier); unregister_trace_sched_switch(ftrace_graph_probe_sched_switch, NULL); } diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index cbb91b0afcc8..58e0f4bc0241 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -7811,7 +7811,7 @@ __init void ftrace_init_global_array_ops(struct trace_array *tr) tr->ops = &global_ops; tr->ops->private = tr; ftrace_init_trace_array(tr); - init_array_fgraph_ops(tr); + init_array_fgraph_ops(tr, tr->ops); } void ftrace_init_array_ops(struct trace_array *tr, ftrace_func_t func) diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 9a70beb2cc46..f06b5ddd3580 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -894,8 +894,8 @@ extern int __trace_graph_entry(struct trace_array *tr, extern void __trace_graph_return(struct trace_array *tr, struct ftrace_graph_ret *trace, unsigned int trace_ctx); -extern void init_array_fgraph_ops(struct trace_array *tr); -extern int allocate_fgraph_ops(struct trace_array *tr); +extern void init_array_fgraph_ops(struct trace_array *tr, struct ftrace_ops *ops); +extern int allocate_fgraph_ops(struct trace_array *tr, struct ftrace_ops *ops); extern void free_fgraph_ops(struct trace_array *tr); #ifdef CONFIG_DYNAMIC_FTRACE @@ -1003,18 +1003,19 @@ static inline bool ftrace_graph_ignore_func(struct ftrace_graph_ent *trace) (fgraph_max_depth && trace->depth >= fgraph_max_depth); } +void fgraph_init_ops(struct ftrace_ops *dst_ops, + struct ftrace_ops *src_ops); + #else /* CONFIG_FUNCTION_GRAPH_TRACER */ static inline enum print_line_t print_graph_function_flags(struct trace_iterator *iter, u32 flags) { return TRACE_TYPE_UNHANDLED; } -static inline void init_array_fgraph_ops(struct trace_array *tr) { } -static inline int allocate_fgraph_ops(struct trace_array *tr) -{ - return 0; -} static inline void free_fgraph_ops(struct trace_array *tr) { } +/* ftrace_ops may not be defined */ +#define init_array_fgraph_ops(tr, ops) do { } while (0) +#define allocate_fgraph_ops(tr, ops) ({ 0; }) #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ extern struct list_head ftrace_pids; diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c index 8e8da0d0ee52..13bf2415245d 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -91,7 +91,7 @@ int ftrace_create_function_files(struct trace_array *tr, if (!tr->ops) return -EINVAL; - ret = allocate_fgraph_ops(tr); + ret = allocate_fgraph_ops(tr, tr->ops); if (ret) { kfree(tr->ops); return ret; diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index 9ccc904a7703..7f30652f0e97 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -288,7 +288,7 @@ static struct fgraph_ops funcgraph_ops = { .retfunc = &trace_graph_return, }; -int allocate_fgraph_ops(struct trace_array *tr) +int allocate_fgraph_ops(struct trace_array *tr, struct ftrace_ops *ops) { struct fgraph_ops *gops; @@ -301,6 +301,9 @@ int allocate_fgraph_ops(struct trace_array *tr) tr->gops = gops; gops->private = tr; + + fgraph_init_ops(&gops->ops, ops); + return 0; } @@ -309,10 +312,11 @@ void free_fgraph_ops(struct trace_array *tr) kfree(tr->gops); } -__init void init_array_fgraph_ops(struct trace_array *tr) +__init void init_array_fgraph_ops(struct trace_array *tr, struct ftrace_ops *ops) { tr->gops = &funcgraph_ops; funcgraph_ops.private = tr; + fgraph_init_ops(&tr->gops->ops, ops); } static int graph_trace_init(struct trace_array *tr) -- cgit v1.2.3 From df3ec5da6a1e7f6e142680d7c5266d3af187170b Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Mon, 3 Jun 2024 15:07:17 -0400 Subject: function_graph: Add pid tracing back to function graph tracer Now that the function_graph has a main callback that handles the function graph subops tracing, it no longer honors the pid filtering of ftrace. Add back this logic in the function_graph code to update the gops callback for the entry function to test if it should trace the current task or not. Link: https://lore.kernel.org/linux-trace-kernel/20240603190822.991720703@goodmis.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 40 ++++++++++++++++++++++++++++++++++++++++ kernel/trace/ftrace.c | 5 +++-- kernel/trace/ftrace_internal.h | 2 ++ 3 files changed, 45 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index 3ef6db53c0bf..30bed20c655f 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -854,6 +854,41 @@ void ftrace_graph_exit_task(struct task_struct *t) kfree(ret_stack); } +static int fgraph_pid_func(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops) +{ + struct trace_array *tr = gops->ops.private; + int pid; + + if (tr) { + pid = this_cpu_read(tr->array_buffer.data->ftrace_ignore_pid); + if (pid == FTRACE_PID_IGNORE) + return 0; + if (pid != FTRACE_PID_TRACE && + pid != current->pid) + return 0; + } + + return gops->saved_func(trace, gops); +} + +void fgraph_update_pid_func(void) +{ + struct fgraph_ops *gops; + struct ftrace_ops *op; + + if (!(graph_ops.flags & FTRACE_OPS_FL_INITIALIZED)) + return; + + list_for_each_entry(op, &graph_ops.subop_list, list) { + if (op->flags & FTRACE_OPS_FL_PID) { + gops = container_of(op, struct fgraph_ops, ops); + gops->entryfunc = ftrace_pids_enabled(op) ? + fgraph_pid_func : gops->saved_func; + } + } +} + /* Allocate a return stack for each task */ static int start_graph_tracing(void) { @@ -931,11 +966,15 @@ int register_ftrace_graph(struct fgraph_ops *gops) command = FTRACE_START_FUNC_RET; } + /* Always save the function, and reset at unregistering */ + gops->saved_func = gops->entryfunc; + ret = ftrace_startup_subops(&graph_ops, &gops->ops, command); error: if (ret) { fgraph_array[i] = &fgraph_stub; ftrace_graph_active--; + gops->saved_func = NULL; } out: mutex_unlock(&ftrace_lock); @@ -979,5 +1018,6 @@ void unregister_ftrace_graph(struct fgraph_ops *gops) unregister_trace_sched_switch(ftrace_graph_probe_sched_switch, NULL); } out: + gops->saved_func = NULL; mutex_unlock(&ftrace_lock); } diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 58e0f4bc0241..da7e6abf48b4 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -100,7 +100,7 @@ struct ftrace_ops *function_trace_op __read_mostly = &ftrace_list_end; /* What to set function_trace_op to */ static struct ftrace_ops *set_function_trace_op; -static bool ftrace_pids_enabled(struct ftrace_ops *ops) +bool ftrace_pids_enabled(struct ftrace_ops *ops) { struct trace_array *tr; @@ -402,10 +402,11 @@ static void ftrace_update_pid_func(void) if (op->flags & FTRACE_OPS_FL_PID) { op->func = ftrace_pids_enabled(op) ? ftrace_pid_func : op->saved_func; - ftrace_update_trampoline(op); } } while_for_each_ftrace_op(op); + fgraph_update_pid_func(); + update_ftrace_function(); } diff --git a/kernel/trace/ftrace_internal.h b/kernel/trace/ftrace_internal.h index cdfd12c44ab4..bfba10c2fcf1 100644 --- a/kernel/trace/ftrace_internal.h +++ b/kernel/trace/ftrace_internal.h @@ -43,8 +43,10 @@ ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip, void *regs) #ifdef CONFIG_FUNCTION_GRAPH_TRACER extern int ftrace_graph_active; +extern void fgraph_update_pid_func(void); #else /* !CONFIG_FUNCTION_GRAPH_TRACER */ # define ftrace_graph_active 0 +static inline void fgraph_update_pid_func(void) {} #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ #else /* !CONFIG_FUNCTION_TRACER */ -- cgit v1.2.3 From 6d4786592ac88aa31f45fde6bfaad3162e3a92a4 Mon Sep 17 00:00:00 2001 From: "Masami Hiramatsu (Google)" Date: Mon, 3 Jun 2024 15:07:18 -0400 Subject: function_graph: Use a simple LRU for fgraph_array index number Since the fgraph_array index is used for the bitmap on the shadow stack, it may leave some entries after a function_graph instance is removed. Thus if another instance reuses the fgraph_array index soon after releasing it, the fgraph may confuse to call the newer callback for the entries which are pushed by the older instance. To avoid reusing the fgraph_array index soon after releasing, introduce a simple LRU table for managing the index number. This will reduce the possibility of this confusion. Link: https://lore.kernel.org/linux-trace-kernel/171509103267.162236.6885097397289135378.stgit@devnote2 Link: https://lore.kernel.org/linux-trace-kernel/20240603190823.147421545@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 71 ++++++++++++++++++++++++++++++++++++--------------- 1 file changed, 50 insertions(+), 21 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index 30bed20c655f..7fd9b03bd170 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -124,10 +124,48 @@ enum { DEFINE_STATIC_KEY_FALSE(kill_ftrace_graph); int ftrace_graph_active; -static int fgraph_array_cnt; - static struct fgraph_ops *fgraph_array[FGRAPH_ARRAY_SIZE]; +/* LRU index table for fgraph_array */ +static int fgraph_lru_table[FGRAPH_ARRAY_SIZE]; +static int fgraph_lru_next; +static int fgraph_lru_last; + +/* Initialize fgraph_lru_table with unused index */ +static void fgraph_lru_init(void) +{ + int i; + + for (i = 0; i < FGRAPH_ARRAY_SIZE; i++) + fgraph_lru_table[i] = i; +} + +/* Release the used index to the LRU table */ +static int fgraph_lru_release_index(int idx) +{ + if (idx < 0 || idx >= FGRAPH_ARRAY_SIZE || + WARN_ON_ONCE(fgraph_lru_table[fgraph_lru_last] != -1)) + return -1; + + fgraph_lru_table[fgraph_lru_last] = idx; + fgraph_lru_last = (fgraph_lru_last + 1) % FGRAPH_ARRAY_SIZE; + return 0; +} + +/* Allocate a new index from LRU table */ +static int fgraph_lru_alloc_index(void) +{ + int idx = fgraph_lru_table[fgraph_lru_next]; + + /* No id is available */ + if (idx == -1) + return -1; + + fgraph_lru_table[fgraph_lru_next] = -1; + fgraph_lru_next = (fgraph_lru_next + 1) % FGRAPH_ARRAY_SIZE; + return idx; +} + /* Get the FRAME_OFFSET from the word from the @offset on ret_stack */ static inline int get_frame_offset(struct task_struct *t, int offset) { @@ -374,7 +412,7 @@ int function_graph_enter(unsigned long ret, unsigned long func, if (offset < 0) goto out; - for (i = 0; i < fgraph_array_cnt; i++) { + for (i = 0; i < FGRAPH_ARRAY_SIZE; i++) { struct fgraph_ops *gops = fgraph_array[i]; if (gops == &fgraph_stub) @@ -925,7 +963,7 @@ int register_ftrace_graph(struct fgraph_ops *gops) { int command = 0; int ret = 0; - int i; + int i = -1; mutex_lock(&ftrace_lock); @@ -933,21 +971,16 @@ int register_ftrace_graph(struct fgraph_ops *gops) /* The array must always have real data on it */ for (i = 0; i < FGRAPH_ARRAY_SIZE; i++) fgraph_array[i] = &fgraph_stub; + fgraph_lru_init(); } - /* Look for an available spot */ - for (i = 0; i < FGRAPH_ARRAY_SIZE; i++) { - if (fgraph_array[i] == &fgraph_stub) - break; - } - if (i >= FGRAPH_ARRAY_SIZE) { + i = fgraph_lru_alloc_index(); + if (i < 0 || WARN_ON_ONCE(fgraph_array[i] != &fgraph_stub)) { ret = -ENOSPC; goto out; } fgraph_array[i] = gops; - if (i + 1 > fgraph_array_cnt) - fgraph_array_cnt = i + 1; gops->idx = i; ftrace_graph_active++; @@ -975,6 +1008,7 @@ error: fgraph_array[i] = &fgraph_stub; ftrace_graph_active--; gops->saved_func = NULL; + fgraph_lru_release_index(i); } out: mutex_unlock(&ftrace_lock); @@ -984,25 +1018,20 @@ out: void unregister_ftrace_graph(struct fgraph_ops *gops) { int command = 0; - int i; mutex_lock(&ftrace_lock); if (unlikely(!ftrace_graph_active)) goto out; - if (unlikely(gops->idx < 0 || gops->idx >= fgraph_array_cnt)) + if (unlikely(gops->idx < 0 || gops->idx >= FGRAPH_ARRAY_SIZE || + fgraph_array[gops->idx] != gops)) goto out; - WARN_ON_ONCE(fgraph_array[gops->idx] != gops); + if (fgraph_lru_release_index(gops->idx) < 0) + goto out; fgraph_array[gops->idx] = &fgraph_stub; - if (gops->idx + 1 == fgraph_array_cnt) { - i = gops->idx; - while (i >= 0 && fgraph_array[i] == &fgraph_stub) - i--; - fgraph_array_cnt = i + 1; - } ftrace_graph_active--; -- cgit v1.2.3 From 4497412a1f7b5d9e0849f125652f2cc58cdba562 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Mon, 3 Jun 2024 15:07:19 -0400 Subject: function_graph: Add "task variables" per task for fgraph_ops Add a "task variables" array on the tasks shadow ret_stack that is the size of longs for each possible registered fgraph_ops. That's a total of 16, taking up 8 * 16 = 128 bytes (out of a page size 4k). This will allow for fgraph_ops to do specific features on a per task basis having a way to maintain state for each task. Co-developed with Masami Hiramatsu: Link: https://lore.kernel.org/linux-trace-kernel/171509104383.162236.12239656156685718550.stgit@devnote2 Link: https://lore.kernel.org/linux-trace-kernel/20240603190823.308806126@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 74 ++++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 73 insertions(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index 7fd9b03bd170..baa08249ffd5 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -54,6 +54,10 @@ * on the return of the function being traced, this is what will be on the * task's shadow ret_stack: (the stack grows upward) * + * ret_stack[SHADOW_STACK_OFFSET] + * | SHADOW_STACK_TASK_VARS(ret_stack)[15] | + * ... + * | SHADOW_STACK_TASK_VARS(ret_stack)[0] | * ret_stack[SHADOW_STACK_MAX_OFFSET] * ... * | | <- task->curr_ret_stack @@ -116,11 +120,19 @@ enum { #define SHADOW_STACK_SIZE (PAGE_SIZE) #define SHADOW_STACK_OFFSET (SHADOW_STACK_SIZE / sizeof(long)) /* Leave on a buffer at the end */ -#define SHADOW_STACK_MAX_OFFSET (SHADOW_STACK_OFFSET - (FGRAPH_FRAME_OFFSET + 1)) +#define SHADOW_STACK_MAX_OFFSET \ + (SHADOW_STACK_OFFSET - (FGRAPH_FRAME_OFFSET + 1 + FGRAPH_ARRAY_SIZE)) /* RET_STACK(): Return the frame from a given @offset from task @t */ #define RET_STACK(t, offset) ((struct ftrace_ret_stack *)(&(t)->ret_stack[offset])) +/* + * Each fgraph_ops has a reservered unsigned long at the end (top) of the + * ret_stack to store task specific state. + */ +#define SHADOW_STACK_TASK_VARS(ret_stack) \ + ((unsigned long *)(&(ret_stack)[SHADOW_STACK_OFFSET - FGRAPH_ARRAY_SIZE])) + DEFINE_STATIC_KEY_FALSE(kill_ftrace_graph); int ftrace_graph_active; @@ -211,6 +223,44 @@ static void return_run(struct ftrace_graph_ret *trace, struct fgraph_ops *ops) { } +static void ret_stack_set_task_var(struct task_struct *t, int idx, long val) +{ + unsigned long *gvals = SHADOW_STACK_TASK_VARS(t->ret_stack); + + gvals[idx] = val; +} + +static unsigned long * +ret_stack_get_task_var(struct task_struct *t, int idx) +{ + unsigned long *gvals = SHADOW_STACK_TASK_VARS(t->ret_stack); + + return &gvals[idx]; +} + +static void ret_stack_init_task_vars(unsigned long *ret_stack) +{ + unsigned long *gvals = SHADOW_STACK_TASK_VARS(ret_stack); + + memset(gvals, 0, sizeof(*gvals) * FGRAPH_ARRAY_SIZE); +} + +/** + * fgraph_get_task_var - retrieve a task specific state variable + * @gops: The ftrace_ops that owns the task specific variable + * + * Every registered fgraph_ops has a task state variable + * reserved on the task's ret_stack. This function returns the + * address to that variable. + * + * Returns the address to the fgraph_ops @gops tasks specific + * unsigned long variable. + */ +unsigned long *fgraph_get_task_var(struct fgraph_ops *gops) +{ + return ret_stack_get_task_var(current, gops->idx); +} + /* * @offset: The offset into @t->ret_stack to find the ret_stack entry * @frame_offset: Where to place the offset into @t->ret_stack of that entry @@ -766,6 +816,7 @@ static int alloc_retstack_tasklist(unsigned long **ret_stack_list) if (t->ret_stack == NULL) { atomic_set(&t->trace_overrun, 0); + ret_stack_init_task_vars(ret_stack_list[start]); t->curr_ret_stack = 0; t->curr_ret_depth = -1; /* Make sure the tasks see the 0 first: */ @@ -826,6 +877,7 @@ static void graph_init_task(struct task_struct *t, unsigned long *ret_stack) { atomic_set(&t->trace_overrun, 0); + ret_stack_init_task_vars(ret_stack); t->ftrace_timestamp = 0; t->curr_ret_stack = 0; t->curr_ret_depth = -1; @@ -959,6 +1011,24 @@ static int start_graph_tracing(void) return ret; } +static void init_task_vars(int idx) +{ + struct task_struct *g, *t; + int cpu; + + for_each_online_cpu(cpu) { + if (idle_task(cpu)->ret_stack) + ret_stack_set_task_var(idle_task(cpu), idx, 0); + } + + read_lock(&tasklist_lock); + for_each_process_thread(g, t) { + if (t->ret_stack) + ret_stack_set_task_var(t, idx, 0); + } + read_unlock(&tasklist_lock); +} + int register_ftrace_graph(struct fgraph_ops *gops) { int command = 0; @@ -997,6 +1067,8 @@ int register_ftrace_graph(struct fgraph_ops *gops) ftrace_graph_return = return_run; ftrace_graph_entry = entry_run; command = FTRACE_START_FUNC_RET; + } else { + init_task_vars(gops->idx); } /* Always save the function, and reset at unregistering */ -- cgit v1.2.3 From 12117f3307b63f287756d7ec8cc4f11b94e1206a Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Mon, 3 Jun 2024 15:07:20 -0400 Subject: function_graph: Move set_graph_function tests to shadow stack global var The use of the task->trace_recursion for the logic used for the set_graph_function was a bit of an abuse of that variable. Now that there exists global vars that are per stack for registered graph traces, use that instead. Link: https://lore.kernel.org/linux-trace-kernel/171509105520.162236.10339831553995971290.stgit@devnote2 Link: https://lore.kernel.org/linux-trace-kernel/20240603190823.472955399@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.h | 32 +++++++++++++++++++++----------- kernel/trace/trace_functions_graph.c | 6 +++--- kernel/trace/trace_irqsoff.c | 4 ++-- kernel/trace/trace_sched_wakeup.c | 4 ++-- 4 files changed, 28 insertions(+), 18 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index f06b5ddd3580..73919129e57c 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -898,11 +898,16 @@ extern void init_array_fgraph_ops(struct trace_array *tr, struct ftrace_ops *ops extern int allocate_fgraph_ops(struct trace_array *tr, struct ftrace_ops *ops); extern void free_fgraph_ops(struct trace_array *tr); +enum { + TRACE_GRAPH_FL = 1, +}; + #ifdef CONFIG_DYNAMIC_FTRACE extern struct ftrace_hash __rcu *ftrace_graph_hash; extern struct ftrace_hash __rcu *ftrace_graph_notrace_hash; -static inline int ftrace_graph_addr(struct ftrace_graph_ent *trace) +static inline int +ftrace_graph_addr(unsigned long *task_var, struct ftrace_graph_ent *trace) { unsigned long addr = trace->func; int ret = 0; @@ -924,12 +929,11 @@ static inline int ftrace_graph_addr(struct ftrace_graph_ent *trace) } if (ftrace_lookup_ip(hash, addr)) { - /* * This needs to be cleared on the return functions * when the depth is zero. */ - trace_recursion_set(TRACE_GRAPH_BIT); + *task_var |= TRACE_GRAPH_FL; trace_recursion_set_depth(trace->depth); /* @@ -949,11 +953,14 @@ out: return ret; } -static inline void ftrace_graph_addr_finish(struct ftrace_graph_ret *trace) +static inline void +ftrace_graph_addr_finish(struct fgraph_ops *gops, struct ftrace_graph_ret *trace) { - if (trace_recursion_test(TRACE_GRAPH_BIT) && + unsigned long *task_var = fgraph_get_task_var(gops); + + if ((*task_var & TRACE_GRAPH_FL) && trace->depth == trace_recursion_depth()) - trace_recursion_clear(TRACE_GRAPH_BIT); + *task_var &= ~TRACE_GRAPH_FL; } static inline int ftrace_graph_notrace_addr(unsigned long addr) @@ -979,7 +986,7 @@ static inline int ftrace_graph_notrace_addr(unsigned long addr) return ret; } #else -static inline int ftrace_graph_addr(struct ftrace_graph_ent *trace) +static inline int ftrace_graph_addr(unsigned long *task_var, struct ftrace_graph_ent *trace) { return 1; } @@ -988,17 +995,20 @@ static inline int ftrace_graph_notrace_addr(unsigned long addr) { return 0; } -static inline void ftrace_graph_addr_finish(struct ftrace_graph_ret *trace) +static inline void ftrace_graph_addr_finish(struct fgraph_ops *gops, struct ftrace_graph_ret *trace) { } #endif /* CONFIG_DYNAMIC_FTRACE */ extern unsigned int fgraph_max_depth; -static inline bool ftrace_graph_ignore_func(struct ftrace_graph_ent *trace) +static inline bool +ftrace_graph_ignore_func(struct fgraph_ops *gops, struct ftrace_graph_ent *trace) { + unsigned long *task_var = fgraph_get_task_var(gops); + /* trace it when it is-nested-in or is a function enabled. */ - return !(trace_recursion_test(TRACE_GRAPH_BIT) || - ftrace_graph_addr(trace)) || + return !((*task_var & TRACE_GRAPH_FL) || + ftrace_graph_addr(task_var, trace)) || (trace->depth < 0) || (fgraph_max_depth && trace->depth >= fgraph_max_depth); } diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index 7f30652f0e97..66cce73e94f8 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -160,7 +160,7 @@ int trace_graph_entry(struct ftrace_graph_ent *trace, if (!ftrace_trace_task(tr)) return 0; - if (ftrace_graph_ignore_func(trace)) + if (ftrace_graph_ignore_func(gops, trace)) return 0; if (ftrace_graph_ignore_irqs()) @@ -247,7 +247,7 @@ void trace_graph_return(struct ftrace_graph_ret *trace, long disabled; int cpu; - ftrace_graph_addr_finish(trace); + ftrace_graph_addr_finish(gops, trace); if (trace_recursion_test(TRACE_GRAPH_NOTRACE_BIT)) { trace_recursion_clear(TRACE_GRAPH_NOTRACE_BIT); @@ -269,7 +269,7 @@ void trace_graph_return(struct ftrace_graph_ret *trace, static void trace_graph_thresh_return(struct ftrace_graph_ret *trace, struct fgraph_ops *gops) { - ftrace_graph_addr_finish(trace); + ftrace_graph_addr_finish(gops, trace); if (trace_recursion_test(TRACE_GRAPH_NOTRACE_BIT)) { trace_recursion_clear(TRACE_GRAPH_NOTRACE_BIT); diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index 5478f4c4f708..fce064e20570 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -184,7 +184,7 @@ static int irqsoff_graph_entry(struct ftrace_graph_ent *trace, unsigned int trace_ctx; int ret; - if (ftrace_graph_ignore_func(trace)) + if (ftrace_graph_ignore_func(gops, trace)) return 0; /* * Do not trace a function if it's filtered by set_graph_notrace. @@ -214,7 +214,7 @@ static void irqsoff_graph_return(struct ftrace_graph_ret *trace, unsigned long flags; unsigned int trace_ctx; - ftrace_graph_addr_finish(trace); + ftrace_graph_addr_finish(gops, trace); if (!func_prolog_dec(tr, &data, &flags)) return; diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 49bcc812652c..130ca7e7787e 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -120,7 +120,7 @@ static int wakeup_graph_entry(struct ftrace_graph_ent *trace, unsigned int trace_ctx; int ret = 0; - if (ftrace_graph_ignore_func(trace)) + if (ftrace_graph_ignore_func(gops, trace)) return 0; /* * Do not trace a function if it's filtered by set_graph_notrace. @@ -149,7 +149,7 @@ static void wakeup_graph_return(struct ftrace_graph_ret *trace, struct trace_array_cpu *data; unsigned int trace_ctx; - ftrace_graph_addr_finish(trace); + ftrace_graph_addr_finish(gops, trace); if (!func_prolog_preempt_disable(tr, &data, &trace_ctx)) return; -- cgit v1.2.3 From 068da098eb504469dc195137ae35eeacfe0c8de9 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Mon, 3 Jun 2024 15:07:21 -0400 Subject: function_graph: Move graph depth stored data to shadow stack global var The use of the task->trace_recursion for the logic used for the function graph depth was a bit of an abuse of that variable. Now that there exists global vars that are per stack for registered graph traces, use that instead. Link: https://lore.kernel.org/linux-trace-kernel/171509106728.162236.2398372644430125344.stgit@devnote2 Link: https://lore.kernel.org/linux-trace-kernel/20240603190823.634870264@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.h | 34 ++++++++++++++++++++++++++++++++-- 1 file changed, 32 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 73919129e57c..82d879dc63ff 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -900,8 +900,38 @@ extern void free_fgraph_ops(struct trace_array *tr); enum { TRACE_GRAPH_FL = 1, + + /* + * In the very unlikely case that an interrupt came in + * at a start of graph tracing, and we want to trace + * the function in that interrupt, the depth can be greater + * than zero, because of the preempted start of a previous + * trace. In an even more unlikely case, depth could be 2 + * if a softirq interrupted the start of graph tracing, + * followed by an interrupt preempting a start of graph + * tracing in the softirq, and depth can even be 3 + * if an NMI came in at the start of an interrupt function + * that preempted a softirq start of a function that + * preempted normal context!!!! Luckily, it can't be + * greater than 3, so the next two bits are a mask + * of what the depth is when we set TRACE_GRAPH_FL + */ + + TRACE_GRAPH_DEPTH_START_BIT, + TRACE_GRAPH_DEPTH_END_BIT, }; +static inline unsigned long ftrace_graph_depth(unsigned long *task_var) +{ + return (*task_var >> TRACE_GRAPH_DEPTH_START_BIT) & 3; +} + +static inline void ftrace_graph_set_depth(unsigned long *task_var, int depth) +{ + *task_var &= ~(3 << TRACE_GRAPH_DEPTH_START_BIT); + *task_var |= (depth & 3) << TRACE_GRAPH_DEPTH_START_BIT; +} + #ifdef CONFIG_DYNAMIC_FTRACE extern struct ftrace_hash __rcu *ftrace_graph_hash; extern struct ftrace_hash __rcu *ftrace_graph_notrace_hash; @@ -934,7 +964,7 @@ ftrace_graph_addr(unsigned long *task_var, struct ftrace_graph_ent *trace) * when the depth is zero. */ *task_var |= TRACE_GRAPH_FL; - trace_recursion_set_depth(trace->depth); + ftrace_graph_set_depth(task_var, trace->depth); /* * If no irqs are to be traced, but a set_graph_function @@ -959,7 +989,7 @@ ftrace_graph_addr_finish(struct fgraph_ops *gops, struct ftrace_graph_ret *trace unsigned long *task_var = fgraph_get_task_var(gops); if ((*task_var & TRACE_GRAPH_FL) && - trace->depth == trace_recursion_depth()) + trace->depth == ftrace_graph_depth(task_var)) *task_var &= ~TRACE_GRAPH_FL; } -- cgit v1.2.3 From b84214890a9bc56f0fe4ec4fc72f2307ed05096d Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Mon, 3 Jun 2024 15:07:22 -0400 Subject: function_graph: Move graph notrace bit to shadow stack global var The use of the task->trace_recursion for the logic used for the function graph no-trace was a bit of an abuse of that variable. Now that there exists global vars that are per stack for registered graph traces, use that instead. Link: https://lore.kernel.org/linux-trace-kernel/171509107907.162236.6564679266777519065.stgit@devnote2 Link: https://lore.kernel.org/linux-trace-kernel/20240603190823.796709456@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.h | 9 +++++++++ kernel/trace/trace_functions_graph.c | 10 ++++++---- 2 files changed, 15 insertions(+), 4 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 82d879dc63ff..b37402e3f0c9 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -919,8 +919,17 @@ enum { TRACE_GRAPH_DEPTH_START_BIT, TRACE_GRAPH_DEPTH_END_BIT, + + /* + * To implement set_graph_notrace, if this bit is set, we ignore + * function graph tracing of called functions, until the return + * function is called to clear it. + */ + TRACE_GRAPH_NOTRACE_BIT, }; +#define TRACE_GRAPH_NOTRACE (1 << TRACE_GRAPH_NOTRACE_BIT) + static inline unsigned long ftrace_graph_depth(unsigned long *task_var) { return (*task_var >> TRACE_GRAPH_DEPTH_START_BIT) & 3; diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index 66cce73e94f8..13d0387ac6a6 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -130,6 +130,7 @@ static inline int ftrace_graph_ignore_irqs(void) int trace_graph_entry(struct ftrace_graph_ent *trace, struct fgraph_ops *gops) { + unsigned long *task_var = fgraph_get_task_var(gops); struct trace_array *tr = gops->private; struct trace_array_cpu *data; unsigned long flags; @@ -138,7 +139,7 @@ int trace_graph_entry(struct ftrace_graph_ent *trace, int ret; int cpu; - if (trace_recursion_test(TRACE_GRAPH_NOTRACE_BIT)) + if (*task_var & TRACE_GRAPH_NOTRACE) return 0; /* @@ -149,7 +150,7 @@ int trace_graph_entry(struct ftrace_graph_ent *trace, * returning from the function. */ if (ftrace_graph_notrace_addr(trace->func)) { - trace_recursion_set(TRACE_GRAPH_NOTRACE_BIT); + *task_var |= TRACE_GRAPH_NOTRACE_BIT; /* * Need to return 1 to have the return called * that will clear the NOTRACE bit. @@ -240,6 +241,7 @@ void __trace_graph_return(struct trace_array *tr, void trace_graph_return(struct ftrace_graph_ret *trace, struct fgraph_ops *gops) { + unsigned long *task_var = fgraph_get_task_var(gops); struct trace_array *tr = gops->private; struct trace_array_cpu *data; unsigned long flags; @@ -249,8 +251,8 @@ void trace_graph_return(struct ftrace_graph_ret *trace, ftrace_graph_addr_finish(gops, trace); - if (trace_recursion_test(TRACE_GRAPH_NOTRACE_BIT)) { - trace_recursion_clear(TRACE_GRAPH_NOTRACE_BIT); + if (*task_var & TRACE_GRAPH_NOTRACE) { + *task_var &= ~TRACE_GRAPH_NOTRACE; return; } -- cgit v1.2.3 From 91c46b0aa917546432b5b219494859cda0edc39e Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Mon, 3 Jun 2024 15:07:23 -0400 Subject: function_graph: Implement fgraph_reserve_data() and fgraph_retrieve_data() Added functions that can be called by a fgraph_ops entryfunc and retfunc to store state between the entry of the function being traced to the exit of the same function. The fgraph_ops entryfunc() may call fgraph_reserve_data() to store up to 32 words onto the task's shadow ret_stack and this then can be retrieved by fgraph_retrieve_data() called by the corresponding retfunc(). Co-developed with Masami Hiramatsu: Link: https://lore.kernel.org/linux-trace-kernel/171509109089.162236.11372474169781184034.stgit@devnote2 Link: https://lore.kernel.org/linux-trace-kernel/20240603190823.959703050@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 191 +++++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 183 insertions(+), 8 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index baa08249ffd5..f207b7ae5f46 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -31,12 +31,11 @@ * holds a bitmask and a type (called "bitmap"). The bitmap is defined as: * * bits: 0 - 9 offset in words from the previous ftrace_ret_stack - * Currently, this will always be set to FGRAPH_FRAME_OFFSET - * to get to the fgraph frame. * * bits: 10 - 11 Type of storage * 0 - reserved * 1 - bitmap of fgraph_array index + * 2 - reserved data * * For type with "bitmap of fgraph_array index" (FGRAPH_TYPE_BITMAP): * bits: 12 - 27 The bitmap of fgraph_ops fgraph_array index @@ -49,10 +48,15 @@ * The top of the ret_stack (when not empty) will always have a reference * word that points to the last fgraph frame that was saved. * + * For reserved data: + * bits: 12 - 17 The size in words that is stored + * bits: 18 - 23 The index of fgraph_array, which shows who is stored + * * That is, at the end of function_graph_enter, if the first and forth * fgraph_ops on the fgraph_array[] (index 0 and 3) needs their retfunc called - * on the return of the function being traced, this is what will be on the - * task's shadow ret_stack: (the stack grows upward) + * on the return of the function being traced, and the forth fgraph_ops + * stored two words of data, this is what will be on the task's shadow + * ret_stack: (the stack grows upward) * * ret_stack[SHADOW_STACK_OFFSET] * | SHADOW_STACK_TASK_VARS(ret_stack)[15] | @@ -62,11 +66,21 @@ * ... * | | <- task->curr_ret_stack * +--------------------------------------------+ + * | (3 << 12) | (3 << 10) | FGRAPH_FRAME_OFFSET| + * | *or put another way* | + * | (3 << FGRAPH_DATA_INDEX_SHIFT)| \ | This is for fgraph_ops[3]. + * | ((2 - 1) << FGRAPH_DATA_SHIFT)| \ | The data size is 2 words. + * | (FGRAPH_TYPE_DATA << FGRAPH_TYPE_SHIFT)| \ | + * | (offset2:FGRAPH_FRAME_OFFSET+3) | <- the offset2 is from here + * +--------------------------------------------+ ( It is 4 words from the ret_stack) + * | STORED DATA WORD 2 | + * | STORED DATA WORD 1 | + * +--------------------------------------------+ * | (9 << 12) | (1 << 10) | FGRAPH_FRAME_OFFSET| * | *or put another way* | * | (BIT(3)|BIT(0)) << FGRAPH_INDEX_SHIFT | \ | * | FGRAPH_TYPE_BITMAP << FGRAPH_TYPE_SHIFT| \ | - * | (offset:FGRAPH_FRAME_OFFSET) | <- the offset is from here + * | (offset1:FGRAPH_FRAME_OFFSET) | <- the offset1 is from here * +--------------------------------------------+ * | struct ftrace_ret_stack | * | (stores the saved ret pointer) | <- the offset points here @@ -100,6 +114,7 @@ enum { FGRAPH_TYPE_RESERVED = 0, FGRAPH_TYPE_BITMAP = 1, + FGRAPH_TYPE_DATA = 2, }; /* @@ -110,6 +125,26 @@ enum { #define FGRAPH_INDEX_MASK GENMASK(FGRAPH_INDEX_BITS - 1, 0) #define FGRAPH_INDEX_SHIFT (FGRAPH_TYPE_SHIFT + FGRAPH_TYPE_BITS) +/* + * For DATA type: + * FGRAPH_DATA (12-17) bits hold the size of data (in words) + * FGRAPH_INDEX (18-23) bits hold the index for which gops->idx the data is for + * + * Note: + * data_size == 0 means 1 word, and 31 (=2^5 - 1) means 32 words. + */ +#define FGRAPH_DATA_BITS 5 +#define FGRAPH_DATA_MASK GENMASK(FGRAPH_DATA_BITS - 1, 0) +#define FGRAPH_DATA_SHIFT (FGRAPH_TYPE_SHIFT + FGRAPH_TYPE_BITS) +#define FGRAPH_MAX_DATA_SIZE (sizeof(long) * (1 << FGRAPH_DATA_BITS)) + +#define FGRAPH_DATA_INDEX_BITS 4 +#define FGRAPH_DATA_INDEX_MASK GENMASK(FGRAPH_DATA_INDEX_BITS - 1, 0) +#define FGRAPH_DATA_INDEX_SHIFT (FGRAPH_DATA_SHIFT + FGRAPH_DATA_BITS) + +#define FGRAPH_MAX_INDEX \ + ((FGRAPH_INDEX_SIZE << FGRAPH_DATA_BITS) + FGRAPH_RET_INDEX) + #define FGRAPH_ARRAY_SIZE FGRAPH_INDEX_BITS /* @@ -178,16 +213,46 @@ static int fgraph_lru_alloc_index(void) return idx; } +/* Get the offset to the fgraph frame from a ret_stack value */ +static inline int __get_offset(unsigned long val) +{ + return val & FGRAPH_FRAME_OFFSET_MASK; +} + +/* Get the type of word from a ret_stack value */ +static inline int __get_type(unsigned long val) +{ + return (val >> FGRAPH_TYPE_SHIFT) & FGRAPH_TYPE_MASK; +} + +/* Get the data_index for a DATA type ret_stack word */ +static inline int __get_data_index(unsigned long val) +{ + return (val >> FGRAPH_DATA_INDEX_SHIFT) & FGRAPH_DATA_INDEX_MASK; +} + +/* Get the data_size for a DATA type ret_stack word */ +static inline int __get_data_size(unsigned long val) +{ + return ((val >> FGRAPH_DATA_SHIFT) & FGRAPH_DATA_MASK) + 1; +} + +/* Get the word from the ret_stack at @offset */ +static inline unsigned long get_fgraph_entry(struct task_struct *t, int offset) +{ + return t->ret_stack[offset]; +} + /* Get the FRAME_OFFSET from the word from the @offset on ret_stack */ static inline int get_frame_offset(struct task_struct *t, int offset) { - return t->ret_stack[offset] & FGRAPH_FRAME_OFFSET_MASK; + return __get_offset(t->ret_stack[offset]); } /* Get FGRAPH_TYPE from the word from the @offset at ret_stack */ static inline int get_fgraph_type(struct task_struct *t, int offset) { - return (t->ret_stack[offset] >> FGRAPH_TYPE_SHIFT) & FGRAPH_TYPE_MASK; + return __get_type(t->ret_stack[offset]); } /* For BITMAP type: get the bitmask from the @offset at ret_stack */ @@ -212,6 +277,25 @@ set_bitmap(struct task_struct *t, int offset, unsigned long bitmap) (FGRAPH_TYPE_BITMAP << FGRAPH_TYPE_SHIFT) | FGRAPH_FRAME_OFFSET; } +/* For DATA type: get the data saved under the ret_stack word at @offset */ +static inline void *get_data_type_data(struct task_struct *t, int offset) +{ + unsigned long val = t->ret_stack[offset]; + + if (__get_type(val) != FGRAPH_TYPE_DATA) + return NULL; + offset -= __get_data_size(val); + return (void *)&t->ret_stack[offset]; +} + +/* Create the ret_stack word for a DATA type */ +static inline unsigned long make_data_type_val(int idx, int size, int offset) +{ + return (idx << FGRAPH_DATA_INDEX_SHIFT) | + ((size - 1) << FGRAPH_DATA_SHIFT) | + (FGRAPH_TYPE_DATA << FGRAPH_TYPE_SHIFT) | offset; +} + /* ftrace_graph_entry set to this to tell some archs to run function graph */ static int entry_run(struct ftrace_graph_ent *trace, struct fgraph_ops *ops) { @@ -245,6 +329,91 @@ static void ret_stack_init_task_vars(unsigned long *ret_stack) memset(gvals, 0, sizeof(*gvals) * FGRAPH_ARRAY_SIZE); } +/** + * fgraph_reserve_data - Reserve storage on the task's ret_stack + * @idx: The index of fgraph_array + * @size_bytes: The size in bytes to reserve + * + * Reserves space of up to FGRAPH_MAX_DATA_SIZE bytes on the + * task's ret_stack shadow stack, for a given fgraph_ops during + * the entryfunc() call. If entryfunc() returns zero, the storage + * is discarded. An entryfunc() can only call this once per iteration. + * The fgraph_ops retfunc() can retrieve this stored data with + * fgraph_retrieve_data(). + * + * Returns: On success, a pointer to the data on the stack. + * Otherwise, NULL if there's not enough space left on the + * ret_stack for the data, or if fgraph_reserve_data() was called + * more than once for a single entryfunc() call. + */ +void *fgraph_reserve_data(int idx, int size_bytes) +{ + unsigned long val; + void *data; + int curr_ret_stack = current->curr_ret_stack; + int data_size; + + if (size_bytes > FGRAPH_MAX_DATA_SIZE) + return NULL; + + /* Convert the data size to number of longs. */ + data_size = (size_bytes + sizeof(long) - 1) >> (sizeof(long) == 4 ? 2 : 3); + + val = get_fgraph_entry(current, curr_ret_stack - 1); + data = ¤t->ret_stack[curr_ret_stack]; + + curr_ret_stack += data_size + 1; + if (unlikely(curr_ret_stack >= SHADOW_STACK_MAX_OFFSET)) + return NULL; + + val = make_data_type_val(idx, data_size, __get_offset(val) + data_size + 1); + + /* Set the last word to be reserved */ + current->ret_stack[curr_ret_stack - 1] = val; + + /* Make sure interrupts see this */ + barrier(); + current->curr_ret_stack = curr_ret_stack; + /* Again sync with interrupts, and reset reserve */ + current->ret_stack[curr_ret_stack - 1] = val; + + return data; +} + +/** + * fgraph_retrieve_data - Retrieve stored data from fgraph_reserve_data() + * @idx: the index of fgraph_array (fgraph_ops::idx) + * @size_bytes: pointer to retrieved data size. + * + * This is to be called by a fgraph_ops retfunc(), to retrieve data that + * was stored by the fgraph_ops entryfunc() on the function entry. + * That is, this will retrieve the data that was reserved on the + * entry of the function that corresponds to the exit of the function + * that the fgraph_ops retfunc() is called on. + * + * Returns: The stored data from fgraph_reserve_data() called by the + * matching entryfunc() for the retfunc() this is called from. + * Or NULL if there was nothing stored. + */ +void *fgraph_retrieve_data(int idx, int *size_bytes) +{ + int offset = current->curr_ret_stack - 1; + unsigned long val; + + val = get_fgraph_entry(current, offset); + while (__get_type(val) == FGRAPH_TYPE_DATA) { + if (__get_data_index(val) == idx) + goto found; + offset -= __get_data_size(val) + 1; + val = get_fgraph_entry(current, offset); + } + return NULL; +found: + if (size_bytes) + *size_bytes = __get_data_size(val) * sizeof(long); + return get_data_type_data(current, offset); +} + /** * fgraph_get_task_var - retrieve a task specific state variable * @gops: The ftrace_ops that owns the task specific variable @@ -464,13 +633,18 @@ int function_graph_enter(unsigned long ret, unsigned long func, for (i = 0; i < FGRAPH_ARRAY_SIZE; i++) { struct fgraph_ops *gops = fgraph_array[i]; + int save_curr_ret_stack; if (gops == &fgraph_stub) continue; + save_curr_ret_stack = current->curr_ret_stack; if (ftrace_ops_test(&gops->ops, func, NULL) && gops->entryfunc(&trace, gops)) bitmap |= BIT(i); + else + /* Clear out any saved storage */ + current->curr_ret_stack = save_curr_ret_stack; } if (!bitmap) @@ -626,7 +800,8 @@ static unsigned long __ftrace_return_to_handler(struct fgraph_ret_regs *ret_regs * curr_ret_stack is after that. */ barrier(); - current->curr_ret_stack -= FGRAPH_FRAME_OFFSET + 1; + current->curr_ret_stack = offset - FGRAPH_FRAME_OFFSET; + current->curr_ret_depth--; return ret; } -- cgit v1.2.3 From 47c3c70aa36971c90e32e91f9254110195d67a02 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Mon, 3 Jun 2024 15:07:24 -0400 Subject: function_graph: Add selftest for passing local variables Add boot up selftest that passes variables from a function entry to a function exit, and make sure that they do get passed around. Co-developed with Masami Hiramatsu: Link: https://lore.kernel.org/linux-trace-kernel/171509110271.162236.11047551496319744627.stgit@devnote2 Link: https://lore.kernel.org/linux-trace-kernel/20240603190824.122952310@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace_selftest.c | 169 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 169 insertions(+) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index f8f55fd79e53..fcdc744c245e 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -756,6 +756,173 @@ trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr) #ifdef CONFIG_FUNCTION_GRAPH_TRACER +#ifdef CONFIG_DYNAMIC_FTRACE + +#define BYTE_NUMBER 123 +#define SHORT_NUMBER 12345 +#define WORD_NUMBER 1234567890 +#define LONG_NUMBER 1234567890123456789LL + +static int fgraph_store_size __initdata; +static const char *fgraph_store_type_name __initdata; +static char *fgraph_error_str __initdata; +static char fgraph_error_str_buf[128] __initdata; + +static __init int store_entry(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops) +{ + const char *type = fgraph_store_type_name; + int size = fgraph_store_size; + void *p; + + p = fgraph_reserve_data(gops->idx, size); + if (!p) { + snprintf(fgraph_error_str_buf, sizeof(fgraph_error_str_buf), + "Failed to reserve %s\n", type); + fgraph_error_str = fgraph_error_str_buf; + return 0; + } + + switch (fgraph_store_size) { + case 1: + *(char *)p = BYTE_NUMBER; + break; + case 2: + *(short *)p = SHORT_NUMBER; + break; + case 4: + *(int *)p = WORD_NUMBER; + break; + case 8: + *(long long *)p = LONG_NUMBER; + break; + } + + return 1; +} + +static __init void store_return(struct ftrace_graph_ret *trace, + struct fgraph_ops *gops) +{ + const char *type = fgraph_store_type_name; + long long expect = 0; + long long found = -1; + int size; + char *p; + + p = fgraph_retrieve_data(gops->idx, &size); + if (!p) { + snprintf(fgraph_error_str_buf, sizeof(fgraph_error_str_buf), + "Failed to retrieve %s\n", type); + fgraph_error_str = fgraph_error_str_buf; + return; + } + if (fgraph_store_size > size) { + snprintf(fgraph_error_str_buf, sizeof(fgraph_error_str_buf), + "Retrieved size %d is smaller than expected %d\n", + size, (int)fgraph_store_size); + fgraph_error_str = fgraph_error_str_buf; + return; + } + + switch (fgraph_store_size) { + case 1: + expect = BYTE_NUMBER; + found = *(char *)p; + break; + case 2: + expect = SHORT_NUMBER; + found = *(short *)p; + break; + case 4: + expect = WORD_NUMBER; + found = *(int *)p; + break; + case 8: + expect = LONG_NUMBER; + found = *(long long *)p; + break; + } + + if (found != expect) { + snprintf(fgraph_error_str_buf, sizeof(fgraph_error_str_buf), + "%s returned not %lld but %lld\n", type, expect, found); + fgraph_error_str = fgraph_error_str_buf; + return; + } + fgraph_error_str = NULL; +} + +static struct fgraph_ops store_bytes __initdata = { + .entryfunc = store_entry, + .retfunc = store_return, +}; + +static int __init test_graph_storage_type(const char *name, int size) +{ + char *func_name; + int len; + int ret; + + fgraph_store_type_name = name; + fgraph_store_size = size; + + snprintf(fgraph_error_str_buf, sizeof(fgraph_error_str_buf), + "Failed to execute storage %s\n", name); + fgraph_error_str = fgraph_error_str_buf; + + pr_cont("PASSED\n"); + pr_info("Testing fgraph storage of %d byte%s: ", size, size > 1 ? "s" : ""); + + func_name = "*" __stringify(DYN_FTRACE_TEST_NAME); + len = strlen(func_name); + + ret = ftrace_set_filter(&store_bytes.ops, func_name, len, 1); + if (ret && ret != -ENODEV) { + pr_cont("*Could not set filter* "); + return -1; + } + + ret = register_ftrace_graph(&store_bytes); + if (ret) { + pr_warn("Failed to init store_bytes fgraph tracing\n"); + return -1; + } + + DYN_FTRACE_TEST_NAME(); + + unregister_ftrace_graph(&store_bytes); + + if (fgraph_error_str) { + pr_cont("*** %s ***", fgraph_error_str); + return -1; + } + + return 0; +} +/* Test the storage passed across function_graph entry and return */ +static __init int test_graph_storage(void) +{ + int ret; + + ret = test_graph_storage_type("byte", 1); + if (ret) + return ret; + ret = test_graph_storage_type("short", 2); + if (ret) + return ret; + ret = test_graph_storage_type("word", 4); + if (ret) + return ret; + ret = test_graph_storage_type("long long", 8); + if (ret) + return ret; + return 0; +} +#else +static inline int test_graph_storage(void) { return 0; } +#endif /* CONFIG_DYNAMIC_FTRACE */ + /* Maximum number of functions to trace before diagnosing a hang */ #define GRAPH_MAX_FUNC_TEST 100000000 @@ -913,6 +1080,8 @@ trace_selftest_startup_function_graph(struct tracer *trace, ftrace_set_global_filter(NULL, 0, 1); #endif + ret = test_graph_storage(); + /* Don't test dynamic tracing, the function tracer already did */ out: /* Stop it if we failed */ -- cgit v1.2.3 From dd120af2d5f8f3d2d742a64cefc4a529d382ab06 Mon Sep 17 00:00:00 2001 From: "Masami Hiramatsu (Google)" Date: Mon, 3 Jun 2024 15:07:25 -0400 Subject: ftrace: Add multiple fgraph storage selftest Add a selftest for multiple function graph tracer with storage on a same function. In this case, the shadow stack entry will be shared among those fgraph with different data storage. So this will ensure the fgraph will not mixed those storage data. Link: https://lore.kernel.org/linux-trace-kernel/171509111465.162236.3795819216426570800.stgit@devnote2 Link: https://lore.kernel.org/linux-trace-kernel/20240603190824.284049716@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Masami Hiramatsu (Google) Suggested-by: Steven Rostedt (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace_selftest.c | 171 +++++++++++++++++++++++++++++++----------- 1 file changed, 126 insertions(+), 45 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index fcdc744c245e..369efc569238 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -762,28 +762,32 @@ trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr) #define SHORT_NUMBER 12345 #define WORD_NUMBER 1234567890 #define LONG_NUMBER 1234567890123456789LL - -static int fgraph_store_size __initdata; -static const char *fgraph_store_type_name __initdata; -static char *fgraph_error_str __initdata; -static char fgraph_error_str_buf[128] __initdata; +#define ERRSTR_BUFLEN 128 + +struct fgraph_fixture { + struct fgraph_ops gops; + int store_size; + const char *store_type_name; + char error_str_buf[ERRSTR_BUFLEN]; + char *error_str; +}; static __init int store_entry(struct ftrace_graph_ent *trace, struct fgraph_ops *gops) { - const char *type = fgraph_store_type_name; - int size = fgraph_store_size; + struct fgraph_fixture *fixture = container_of(gops, struct fgraph_fixture, gops); + const char *type = fixture->store_type_name; + int size = fixture->store_size; void *p; p = fgraph_reserve_data(gops->idx, size); if (!p) { - snprintf(fgraph_error_str_buf, sizeof(fgraph_error_str_buf), + snprintf(fixture->error_str_buf, ERRSTR_BUFLEN, "Failed to reserve %s\n", type); - fgraph_error_str = fgraph_error_str_buf; return 0; } - switch (fgraph_store_size) { + switch (size) { case 1: *(char *)p = BYTE_NUMBER; break; @@ -804,7 +808,8 @@ static __init int store_entry(struct ftrace_graph_ent *trace, static __init void store_return(struct ftrace_graph_ret *trace, struct fgraph_ops *gops) { - const char *type = fgraph_store_type_name; + struct fgraph_fixture *fixture = container_of(gops, struct fgraph_fixture, gops); + const char *type = fixture->store_type_name; long long expect = 0; long long found = -1; int size; @@ -812,20 +817,18 @@ static __init void store_return(struct ftrace_graph_ret *trace, p = fgraph_retrieve_data(gops->idx, &size); if (!p) { - snprintf(fgraph_error_str_buf, sizeof(fgraph_error_str_buf), + snprintf(fixture->error_str_buf, ERRSTR_BUFLEN, "Failed to retrieve %s\n", type); - fgraph_error_str = fgraph_error_str_buf; return; } - if (fgraph_store_size > size) { - snprintf(fgraph_error_str_buf, sizeof(fgraph_error_str_buf), + if (fixture->store_size > size) { + snprintf(fixture->error_str_buf, ERRSTR_BUFLEN, "Retrieved size %d is smaller than expected %d\n", - size, (int)fgraph_store_size); - fgraph_error_str = fgraph_error_str_buf; + size, (int)fixture->store_size); return; } - switch (fgraph_store_size) { + switch (fixture->store_size) { case 1: expect = BYTE_NUMBER; found = *(char *)p; @@ -845,45 +848,44 @@ static __init void store_return(struct ftrace_graph_ret *trace, } if (found != expect) { - snprintf(fgraph_error_str_buf, sizeof(fgraph_error_str_buf), + snprintf(fixture->error_str_buf, ERRSTR_BUFLEN, "%s returned not %lld but %lld\n", type, expect, found); - fgraph_error_str = fgraph_error_str_buf; return; } - fgraph_error_str = NULL; + fixture->error_str = NULL; } -static struct fgraph_ops store_bytes __initdata = { - .entryfunc = store_entry, - .retfunc = store_return, -}; - -static int __init test_graph_storage_type(const char *name, int size) +static int __init init_fgraph_fixture(struct fgraph_fixture *fixture) { char *func_name; int len; - int ret; - fgraph_store_type_name = name; - fgraph_store_size = size; + snprintf(fixture->error_str_buf, ERRSTR_BUFLEN, + "Failed to execute storage %s\n", fixture->store_type_name); + fixture->error_str = fixture->error_str_buf; - snprintf(fgraph_error_str_buf, sizeof(fgraph_error_str_buf), - "Failed to execute storage %s\n", name); - fgraph_error_str = fgraph_error_str_buf; + func_name = "*" __stringify(DYN_FTRACE_TEST_NAME); + len = strlen(func_name); + + return ftrace_set_filter(&fixture->gops.ops, func_name, len, 1); +} + +/* Test fgraph storage for each size */ +static int __init test_graph_storage_single(struct fgraph_fixture *fixture) +{ + int size = fixture->store_size; + int ret; pr_cont("PASSED\n"); pr_info("Testing fgraph storage of %d byte%s: ", size, size > 1 ? "s" : ""); - func_name = "*" __stringify(DYN_FTRACE_TEST_NAME); - len = strlen(func_name); - - ret = ftrace_set_filter(&store_bytes.ops, func_name, len, 1); + ret = init_fgraph_fixture(fixture); if (ret && ret != -ENODEV) { pr_cont("*Could not set filter* "); return -1; } - ret = register_ftrace_graph(&store_bytes); + ret = register_ftrace_graph(&fixture->gops); if (ret) { pr_warn("Failed to init store_bytes fgraph tracing\n"); return -1; @@ -891,30 +893,109 @@ static int __init test_graph_storage_type(const char *name, int size) DYN_FTRACE_TEST_NAME(); - unregister_ftrace_graph(&store_bytes); + unregister_ftrace_graph(&fixture->gops); - if (fgraph_error_str) { - pr_cont("*** %s ***", fgraph_error_str); + if (fixture->error_str) { + pr_cont("*** %s ***", fixture->error_str); return -1; } return 0; } + +static struct fgraph_fixture store_bytes[4] __initdata = { + [0] = { + .gops = { + .entryfunc = store_entry, + .retfunc = store_return, + }, + .store_size = 1, + .store_type_name = "byte", + }, + [1] = { + .gops = { + .entryfunc = store_entry, + .retfunc = store_return, + }, + .store_size = 2, + .store_type_name = "short", + }, + [2] = { + .gops = { + .entryfunc = store_entry, + .retfunc = store_return, + }, + .store_size = 4, + .store_type_name = "word", + }, + [3] = { + .gops = { + .entryfunc = store_entry, + .retfunc = store_return, + }, + .store_size = 8, + .store_type_name = "long long", + }, +}; + +static __init int test_graph_storage_multi(void) +{ + struct fgraph_fixture *fixture; + bool printed = false; + int i, ret; + + pr_cont("PASSED\n"); + pr_info("Testing multiple fgraph storage on a function: "); + + for (i = 0; i < ARRAY_SIZE(store_bytes); i++) { + fixture = &store_bytes[i]; + ret = init_fgraph_fixture(fixture); + if (ret && ret != -ENODEV) { + pr_cont("*Could not set filter* "); + printed = true; + goto out; + } + + ret = register_ftrace_graph(&fixture->gops); + if (ret) { + pr_warn("Failed to init store_bytes fgraph tracing\n"); + printed = true; + goto out; + } + } + + DYN_FTRACE_TEST_NAME(); +out: + while (--i >= 0) { + fixture = &store_bytes[i]; + unregister_ftrace_graph(&fixture->gops); + + if (fixture->error_str && !printed) { + pr_cont("*** %s ***", fixture->error_str); + printed = true; + } + } + return printed ? -1 : 0; +} + /* Test the storage passed across function_graph entry and return */ static __init int test_graph_storage(void) { int ret; - ret = test_graph_storage_type("byte", 1); + ret = test_graph_storage_single(&store_bytes[0]); + if (ret) + return ret; + ret = test_graph_storage_single(&store_bytes[1]); if (ret) return ret; - ret = test_graph_storage_type("short", 2); + ret = test_graph_storage_single(&store_bytes[2]); if (ret) return ret; - ret = test_graph_storage_type("word", 4); + ret = test_graph_storage_single(&store_bytes[3]); if (ret) return ret; - ret = test_graph_storage_type("long long", 8); + ret = test_graph_storage_multi(); if (ret) return ret; return 0; -- cgit v1.2.3 From 420e1354bcb6f006f183a1b6fe5dd21f60a457ef Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Mon, 3 Jun 2024 15:07:26 -0400 Subject: function_graph: Use for_each_set_bit() in __ftrace_return_to_handler() Instead of iterating through the entire fgraph_array[] and seeing if one of the bitmap bits are set to know to call the array's retfunc() function, use for_each_set_bit() on the bitmap itself. This will only iterate for the number of set bits. Link: https://lore.kernel.org/linux-trace-kernel/20240603190824.447448026@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index f207b7ae5f46..0827b67f746d 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -783,11 +783,10 @@ static unsigned long __ftrace_return_to_handler(struct fgraph_ret_regs *ret_regs #endif bitmap = get_bitmap_bits(current, offset); - for (i = 0; i < FGRAPH_ARRAY_SIZE; i++) { + + for_each_set_bit(i, &bitmap, sizeof(bitmap) * BITS_PER_BYTE) { struct fgraph_ops *gops = fgraph_array[i]; - if (!(bitmap & BIT(i))) - continue; if (gops == &fgraph_stub) continue; -- cgit v1.2.3 From a5b6d4da0218a0539c36ad6794c624c2c6ca7b32 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Mon, 3 Jun 2024 15:07:27 -0400 Subject: function_graph: Use bitmask to loop on fgraph entry Instead of looping through all the elements of fgraph_array[] to see if there's an gops attached to one and then calling its gops->func(). Create a fgraph_array_bitmask that sets bits when an index in the array is reserved (via the simple lru algorithm). Then only the bits set in this bitmask needs to be looked at where only elements in the array that have ops registered need to be looked at. Note, we do not care about races. If a bit is set before the gops is assigned, it only wastes time looking at the element and ignoring it (as it did before this bitmask is added). Link: https://lore.kernel.org/linux-trace-kernel/20240603190824.604448781@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index 0827b67f746d..4d566a0a741d 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -172,6 +172,7 @@ DEFINE_STATIC_KEY_FALSE(kill_ftrace_graph); int ftrace_graph_active; static struct fgraph_ops *fgraph_array[FGRAPH_ARRAY_SIZE]; +static unsigned long fgraph_array_bitmask; /* LRU index table for fgraph_array */ static int fgraph_lru_table[FGRAPH_ARRAY_SIZE]; @@ -196,6 +197,8 @@ static int fgraph_lru_release_index(int idx) fgraph_lru_table[fgraph_lru_last] = idx; fgraph_lru_last = (fgraph_lru_last + 1) % FGRAPH_ARRAY_SIZE; + + clear_bit(idx, &fgraph_array_bitmask); return 0; } @@ -210,6 +213,8 @@ static int fgraph_lru_alloc_index(void) fgraph_lru_table[fgraph_lru_next] = -1; fgraph_lru_next = (fgraph_lru_next + 1) % FGRAPH_ARRAY_SIZE; + + set_bit(idx, &fgraph_array_bitmask); return idx; } @@ -631,7 +636,8 @@ int function_graph_enter(unsigned long ret, unsigned long func, if (offset < 0) goto out; - for (i = 0; i < FGRAPH_ARRAY_SIZE; i++) { + for_each_set_bit(i, &fgraph_array_bitmask, + sizeof(fgraph_array_bitmask) * BITS_PER_BYTE) { struct fgraph_ops *gops = fgraph_array[i]; int save_curr_ret_stack; -- cgit v1.2.3 From cc60ee813b50334b32343861057dc9e981e9c7f0 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Mon, 3 Jun 2024 15:07:28 -0400 Subject: function_graph: Use static_call and branch to optimize entry function In most cases function graph is used by a single user. Instead of calling a loop to call function graph callbacks in this case, call the function entry callback directly. Add a static_key that will be used to set the function graph logic to either do the loop (when more than one callback is registered) or to call the callback directly if there is only one registered callback. Link: https://lore.kernel.org/linux-trace-kernel/20240603190824.766858241@goodmis.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 77 +++++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 66 insertions(+), 11 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index 4d566a0a741d..7c3b0261b1bb 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -11,6 +11,7 @@ #include #include #include +#include #include #include @@ -511,6 +512,10 @@ static struct fgraph_ops fgraph_stub = { .retfunc = ftrace_graph_ret_stub, }; +static struct fgraph_ops *fgraph_direct_gops = &fgraph_stub; +DEFINE_STATIC_CALL(fgraph_func, ftrace_graph_entry_stub); +DEFINE_STATIC_KEY_TRUE(fgraph_do_direct); + /** * ftrace_graph_stop - set to permanently disable function graph tracing * @@ -636,21 +641,34 @@ int function_graph_enter(unsigned long ret, unsigned long func, if (offset < 0) goto out; - for_each_set_bit(i, &fgraph_array_bitmask, - sizeof(fgraph_array_bitmask) * BITS_PER_BYTE) { - struct fgraph_ops *gops = fgraph_array[i]; - int save_curr_ret_stack; - - if (gops == &fgraph_stub) - continue; +#ifdef CONFIG_HAVE_STATIC_CALL + if (static_branch_likely(&fgraph_do_direct)) { + int save_curr_ret_stack = current->curr_ret_stack; - save_curr_ret_stack = current->curr_ret_stack; - if (ftrace_ops_test(&gops->ops, func, NULL) && - gops->entryfunc(&trace, gops)) - bitmap |= BIT(i); + if (static_call(fgraph_func)(&trace, fgraph_direct_gops)) + bitmap |= BIT(fgraph_direct_gops->idx); else /* Clear out any saved storage */ current->curr_ret_stack = save_curr_ret_stack; + } else +#endif + { + for_each_set_bit(i, &fgraph_array_bitmask, + sizeof(fgraph_array_bitmask) * BITS_PER_BYTE) { + struct fgraph_ops *gops = fgraph_array[i]; + int save_curr_ret_stack; + + if (gops == &fgraph_stub) + continue; + + save_curr_ret_stack = current->curr_ret_stack; + if (ftrace_ops_test(&gops->ops, func, NULL) && + gops->entryfunc(&trace, gops)) + bitmap |= BIT(i); + else + /* Clear out any saved storage */ + current->curr_ret_stack = save_curr_ret_stack; + } } if (!bitmap) @@ -1155,6 +1173,8 @@ void fgraph_update_pid_func(void) gops = container_of(op, struct fgraph_ops, ops); gops->entryfunc = ftrace_pids_enabled(op) ? fgraph_pid_func : gops->saved_func; + if (ftrace_graph_active == 1) + static_call_update(fgraph_func, gops->entryfunc); } } } @@ -1209,6 +1229,32 @@ static void init_task_vars(int idx) read_unlock(&tasklist_lock); } +static void ftrace_graph_enable_direct(bool enable_branch) +{ + trace_func_graph_ent_t func = NULL; + int i; + + for_each_set_bit(i, &fgraph_array_bitmask, + sizeof(fgraph_array_bitmask) * BITS_PER_BYTE) { + func = fgraph_array[i]->entryfunc; + fgraph_direct_gops = fgraph_array[i]; + } + if (WARN_ON_ONCE(!func)) + return; + + static_call_update(fgraph_func, func); + if (enable_branch) + static_branch_disable(&fgraph_do_direct); +} + +static void ftrace_graph_disable_direct(bool disable_branch) +{ + if (disable_branch) + static_branch_disable(&fgraph_do_direct); + static_call_update(fgraph_func, ftrace_graph_entry_stub); + fgraph_direct_gops = &fgraph_stub; +} + int register_ftrace_graph(struct fgraph_ops *gops) { int command = 0; @@ -1235,7 +1281,11 @@ int register_ftrace_graph(struct fgraph_ops *gops) ftrace_graph_active++; + if (ftrace_graph_active == 2) + ftrace_graph_disable_direct(true); + if (ftrace_graph_active == 1) { + ftrace_graph_enable_direct(false); register_pm_notifier(&ftrace_suspend_notifier); ret = start_graph_tracing(); if (ret) @@ -1292,6 +1342,11 @@ void unregister_ftrace_graph(struct fgraph_ops *gops) ftrace_shutdown_subops(&graph_ops, &gops->ops, command); + if (ftrace_graph_active == 1) + ftrace_graph_enable_direct(true); + else if (!ftrace_graph_active) + ftrace_graph_disable_direct(false); + if (!ftrace_graph_active) { ftrace_graph_return = ftrace_stub_graph; ftrace_graph_entry = ftrace_graph_entry_stub; -- cgit v1.2.3 From fe835e3ca40e172aa8ad12f4ed2898c181fafab0 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Mon, 3 Jun 2024 15:07:29 -0400 Subject: function_graph: Use static_call and branch to optimize return function In most cases function graph is used by a single user. Instead of calling a loop to call function graph callbacks in this case, call the function return callback directly. Use the static_key that is set when the function graph tracer has less than 2 callbacks registered. It will do the direct call in that case, and will do the loop over all callers when there are 2 or more callbacks registered. Link: https://lore.kernel.org/linux-trace-kernel/20240603190824.921460797@goodmis.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Alexei Starovoitov Cc: Florent Revest Cc: Martin KaFai Lau Cc: bpf Cc: Sven Schnelle Cc: Alexei Starovoitov Cc: Jiri Olsa Cc: Arnaldo Carvalho de Melo Cc: Daniel Borkmann Cc: Alan Maguire Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Guo Ren Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 23 ++++++++++++++++++----- 1 file changed, 18 insertions(+), 5 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index 7c3b0261b1bb..4bf91eebbb08 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -514,6 +514,7 @@ static struct fgraph_ops fgraph_stub = { static struct fgraph_ops *fgraph_direct_gops = &fgraph_stub; DEFINE_STATIC_CALL(fgraph_func, ftrace_graph_entry_stub); +DEFINE_STATIC_CALL(fgraph_retfunc, ftrace_graph_ret_stub); DEFINE_STATIC_KEY_TRUE(fgraph_do_direct); /** @@ -808,13 +809,21 @@ static unsigned long __ftrace_return_to_handler(struct fgraph_ret_regs *ret_regs bitmap = get_bitmap_bits(current, offset); - for_each_set_bit(i, &bitmap, sizeof(bitmap) * BITS_PER_BYTE) { - struct fgraph_ops *gops = fgraph_array[i]; +#ifdef CONFIG_HAVE_STATIC_CALL + if (static_branch_likely(&fgraph_do_direct)) { + if (test_bit(fgraph_direct_gops->idx, &bitmap)) + static_call(fgraph_retfunc)(&trace, fgraph_direct_gops); + } else +#endif + { + for_each_set_bit(i, &bitmap, sizeof(bitmap) * BITS_PER_BYTE) { + struct fgraph_ops *gops = fgraph_array[i]; - if (gops == &fgraph_stub) - continue; + if (gops == &fgraph_stub) + continue; - gops->retfunc(&trace, gops); + gops->retfunc(&trace, gops); + } } /* @@ -1232,17 +1241,20 @@ static void init_task_vars(int idx) static void ftrace_graph_enable_direct(bool enable_branch) { trace_func_graph_ent_t func = NULL; + trace_func_graph_ret_t retfunc = NULL; int i; for_each_set_bit(i, &fgraph_array_bitmask, sizeof(fgraph_array_bitmask) * BITS_PER_BYTE) { func = fgraph_array[i]->entryfunc; + retfunc = fgraph_array[i]->retfunc; fgraph_direct_gops = fgraph_array[i]; } if (WARN_ON_ONCE(!func)) return; static_call_update(fgraph_func, func); + static_call_update(fgraph_retfunc, retfunc); if (enable_branch) static_branch_disable(&fgraph_do_direct); } @@ -1252,6 +1264,7 @@ static void ftrace_graph_disable_direct(bool disable_branch) if (disable_branch) static_branch_disable(&fgraph_do_direct); static_call_update(fgraph_func, ftrace_graph_entry_stub); + static_call_update(fgraph_retfunc, ftrace_graph_ret_stub); fgraph_direct_gops = &fgraph_stub; } -- cgit v1.2.3 From 24311962484385fa3a06c4fc1f2864b613df5af6 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 5 Jun 2024 20:53:37 -0400 Subject: ftrace: Add back ftrace_update_trampoline() to ftrace_update_pid_func() The update to the ops trampoline done by the function ftrace_update_trampoline() was accidentally removed from ftrace_update_pid_func(). Add it back. Link: https://lore.kernel.org/linux-trace-kernel/20240605205337.6115e9a5@gandalf.local.home Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Fixes: df3ec5da6a1e ("function_graph: Add pid tracing back to function graph tracer") Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ftrace.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index da7e6abf48b4..897d7541041c 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -402,6 +402,7 @@ static void ftrace_update_pid_func(void) if (op->flags & FTRACE_OPS_FL_PID) { op->func = ftrace_pids_enabled(op) ? ftrace_pid_func : op->saved_func; + ftrace_update_trampoline(op); } } while_for_each_ftrace_op(op); -- cgit v1.2.3 From 22b639253ec046d66c69c54b9d28bb1bd30f3a7a Mon Sep 17 00:00:00 2001 From: Jeff Johnson Date: Mon, 6 May 2024 07:22:33 -0700 Subject: tracing: Fix trace_pid_list_free() kernel-doc make C=1 reports: kernel/trace/pid_list.c:458: warning: Function parameter or struct member 'pid_list' not described in 'trace_pid_list_free' Add the missing parameter to the trace_pid_list_free() kernel-doc. Link: https://lore.kernel.org/linux-trace-kernel/20240506-trace_pid_list_free-kdoc-v1-1-c70f0ae29144@quicinc.com Cc: Masami Hiramatsu Cc: Mathieu Desnoyers Signed-off-by: Jeff Johnson Signed-off-by: Steven Rostedt (Google) --- kernel/trace/pid_list.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel/trace') diff --git a/kernel/trace/pid_list.c b/kernel/trace/pid_list.c index 95106d02b32d..19b271a12c99 100644 --- a/kernel/trace/pid_list.c +++ b/kernel/trace/pid_list.c @@ -451,6 +451,7 @@ struct trace_pid_list *trace_pid_list_alloc(void) /** * trace_pid_list_free - Frees an allocated pid_list. + * @pid_list: The pid list to free. * * Frees the memory for a pid_list that was allocated. */ -- cgit v1.2.3 From 3afd801f4264c7a6c84957cd3ab653cfd8f61d3e Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 5 Jun 2024 14:03:35 -0400 Subject: ftrace: Rename dup_hash() and comment it The name "dup_hash()" is a misnomer as it does not duplicate the hash that is passed in, but instead moves its entities from that hash to a newly allocated one. Rename it to "__move_hash()" (using starting underscores as it is an internal function), and add some comments about what it does. Link: https://lore.kernel.org/linux-trace-kernel/20240605180408.537723591@goodmis.org Cc: Mathieu Desnoyers Cc: Andrew Morton Acked-by: Masami Hiramatsu (Google) Acked-by: Mark Rutland Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ftrace.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 897d7541041c..f4b253d20df8 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -1392,7 +1392,11 @@ ftrace_hash_rec_enable_modify(struct ftrace_ops *ops, int filter_hash); static int ftrace_hash_ipmodify_update(struct ftrace_ops *ops, struct ftrace_hash *new_hash); -static struct ftrace_hash *dup_hash(struct ftrace_hash *src, int size) +/* + * Allocate a new hash and remove entries from @src and move them to the new hash. + * On success, the @src hash will be empty and should be freed. + */ +static struct ftrace_hash *__move_hash(struct ftrace_hash *src, int size) { struct ftrace_func_entry *entry; struct ftrace_hash *new_hash; @@ -1439,7 +1443,7 @@ __ftrace_hash_move(struct ftrace_hash *src) if (ftrace_hash_empty(src)) return EMPTY_HASH; - return dup_hash(src, size); + return __move_hash(src, size); } static int -- cgit v1.2.3 From 07bbe0833ed62f48785dffa8e429f35c1d906415 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 5 Jun 2024 14:03:36 -0400 Subject: ftrace: Remove "filter_hash" parameter from __ftrace_hash_rec_update() While adding comments to the function __ftrace_hash_rec_update() and trying to describe in detail what the parameter for "filter_hash" does, I realized that it basically does exactly the same thing (but differently) if it is set or not! If it is set, the idea was the ops->filter_hash was being updated, and the code should focus on the functions that are in the ops->filter_hash and add them. But it still had to pay attention to the functions in the ops->notrace_hash, to ignore them. If it was cleared, it focused on the ops->notrace_hash, and would add functions that were not in the ops->notrace_hash but would still keep functions in the "ops->filter_hash". Basically doing the same thing. In reality, the __ftrace_hash_rec_update() only needs to either remove the functions associated to the give ops (if "inc" is set) or remove them (if "inc" is cleared). It has to pay attention to both the filter_hash and notrace_hash regardless. Remove the "filter_hash" parameter from __filter_hash_rec_update() and comment the function for what it really is doing. Link: https://lore.kernel.org/linux-trace-kernel/20240605180408.691995506@goodmis.org Cc: Mathieu Desnoyers Cc: Andrew Morton Acked-by: Masami Hiramatsu (Google) Reviewed-by: Mark Rutland Tested-by: Mark Rutland Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ftrace.c | 102 +++++++++++++++++++------------------------------- 1 file changed, 38 insertions(+), 64 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index f4b253d20df8..abcdffa4d572 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -1384,10 +1384,8 @@ alloc_and_copy_ftrace_hash(int size_bits, struct ftrace_hash *hash) return NULL; } -static void -ftrace_hash_rec_disable_modify(struct ftrace_ops *ops, int filter_hash); -static void -ftrace_hash_rec_enable_modify(struct ftrace_ops *ops, int filter_hash); +static void ftrace_hash_rec_disable_modify(struct ftrace_ops *ops); +static void ftrace_hash_rec_enable_modify(struct ftrace_ops *ops); static int ftrace_hash_ipmodify_update(struct ftrace_ops *ops, struct ftrace_hash *new_hash); @@ -1475,11 +1473,11 @@ ftrace_hash_move(struct ftrace_ops *ops, int enable, * Remove the current set, update the hash and add * them back. */ - ftrace_hash_rec_disable_modify(ops, enable); + ftrace_hash_rec_disable_modify(ops); rcu_assign_pointer(*dst, new_hash); - ftrace_hash_rec_enable_modify(ops, enable); + ftrace_hash_rec_enable_modify(ops); return 0; } @@ -1702,12 +1700,21 @@ static bool skip_record(struct dyn_ftrace *rec) !(rec->flags & FTRACE_FL_ENABLED); } +/* + * This is the main engine to the ftrace updates to the dyn_ftrace records. + * + * It will iterate through all the available ftrace functions + * (the ones that ftrace can have callbacks to) and set the flags + * in the associated dyn_ftrace records. + * + * @inc: If true, the functions associated to @ops are added to + * the dyn_ftrace records, otherwise they are removed. + */ static bool __ftrace_hash_rec_update(struct ftrace_ops *ops, - int filter_hash, bool inc) { struct ftrace_hash *hash; - struct ftrace_hash *other_hash; + struct ftrace_hash *notrace_hash; struct ftrace_page *pg; struct dyn_ftrace *rec; bool update = false; @@ -1719,35 +1726,16 @@ static bool __ftrace_hash_rec_update(struct ftrace_ops *ops, return false; /* - * In the filter_hash case: * If the count is zero, we update all records. * Otherwise we just update the items in the hash. - * - * In the notrace_hash case: - * We enable the update in the hash. - * As disabling notrace means enabling the tracing, - * and enabling notrace means disabling, the inc variable - * gets inversed. */ - if (filter_hash) { - hash = ops->func_hash->filter_hash; - other_hash = ops->func_hash->notrace_hash; - if (ftrace_hash_empty(hash)) - all = true; - } else { - inc = !inc; - hash = ops->func_hash->notrace_hash; - other_hash = ops->func_hash->filter_hash; - /* - * If the notrace hash has no items, - * then there's nothing to do. - */ - if (ftrace_hash_empty(hash)) - return false; - } + hash = ops->func_hash->filter_hash; + notrace_hash = ops->func_hash->notrace_hash; + if (ftrace_hash_empty(hash)) + all = true; do_for_each_ftrace_rec(pg, rec) { - int in_other_hash = 0; + int in_notrace_hash = 0; int in_hash = 0; int match = 0; @@ -1759,26 +1747,17 @@ static bool __ftrace_hash_rec_update(struct ftrace_ops *ops, * Only the filter_hash affects all records. * Update if the record is not in the notrace hash. */ - if (!other_hash || !ftrace_lookup_ip(other_hash, rec->ip)) + if (!notrace_hash || !ftrace_lookup_ip(notrace_hash, rec->ip)) match = 1; } else { in_hash = !!ftrace_lookup_ip(hash, rec->ip); - in_other_hash = !!ftrace_lookup_ip(other_hash, rec->ip); + in_notrace_hash = !!ftrace_lookup_ip(notrace_hash, rec->ip); /* - * If filter_hash is set, we want to match all functions - * that are in the hash but not in the other hash. - * - * If filter_hash is not set, then we are decrementing. - * That means we match anything that is in the hash - * and also in the other_hash. That is, we need to turn - * off functions in the other hash because they are disabled - * by this hash. + * We want to match all functions that are in the hash but + * not in the other hash. */ - if (filter_hash && in_hash && !in_other_hash) - match = 1; - else if (!filter_hash && in_hash && - (in_other_hash || ftrace_hash_empty(other_hash))) + if (in_hash && !in_notrace_hash) match = 1; } if (!match) @@ -1884,24 +1863,21 @@ static bool __ftrace_hash_rec_update(struct ftrace_ops *ops, return update; } -static bool ftrace_hash_rec_disable(struct ftrace_ops *ops, - int filter_hash) +static bool ftrace_hash_rec_disable(struct ftrace_ops *ops) { - return __ftrace_hash_rec_update(ops, filter_hash, 0); + return __ftrace_hash_rec_update(ops, 0); } -static bool ftrace_hash_rec_enable(struct ftrace_ops *ops, - int filter_hash) +static bool ftrace_hash_rec_enable(struct ftrace_ops *ops) { - return __ftrace_hash_rec_update(ops, filter_hash, 1); + return __ftrace_hash_rec_update(ops, 1); } -static void ftrace_hash_rec_update_modify(struct ftrace_ops *ops, - int filter_hash, int inc) +static void ftrace_hash_rec_update_modify(struct ftrace_ops *ops, int inc) { struct ftrace_ops *op; - __ftrace_hash_rec_update(ops, filter_hash, inc); + __ftrace_hash_rec_update(ops, inc); if (ops->func_hash != &global_ops.local_hash) return; @@ -1915,20 +1891,18 @@ static void ftrace_hash_rec_update_modify(struct ftrace_ops *ops, if (op == ops) continue; if (op->func_hash == &global_ops.local_hash) - __ftrace_hash_rec_update(op, filter_hash, inc); + __ftrace_hash_rec_update(op, inc); } while_for_each_ftrace_op(op); } -static void ftrace_hash_rec_disable_modify(struct ftrace_ops *ops, - int filter_hash) +static void ftrace_hash_rec_disable_modify(struct ftrace_ops *ops) { - ftrace_hash_rec_update_modify(ops, filter_hash, 0); + ftrace_hash_rec_update_modify(ops, 0); } -static void ftrace_hash_rec_enable_modify(struct ftrace_ops *ops, - int filter_hash) +static void ftrace_hash_rec_enable_modify(struct ftrace_ops *ops) { - ftrace_hash_rec_update_modify(ops, filter_hash, 1); + ftrace_hash_rec_update_modify(ops, 1); } /* @@ -3051,7 +3025,7 @@ int ftrace_startup(struct ftrace_ops *ops, int command) return ret; } - if (ftrace_hash_rec_enable(ops, 1)) + if (ftrace_hash_rec_enable(ops)) command |= FTRACE_UPDATE_CALLS; ftrace_startup_enable(command); @@ -3093,7 +3067,7 @@ int ftrace_shutdown(struct ftrace_ops *ops, int command) /* Disabling ipmodify never fails */ ftrace_hash_ipmodify_disable(ops); - if (ftrace_hash_rec_disable(ops, 1)) + if (ftrace_hash_rec_disable(ops)) command |= FTRACE_UPDATE_CALLS; ops->flags &= ~FTRACE_OPS_FL_ENABLED; -- cgit v1.2.3 From da73f6d49007d1b6889d5cb3320dc9f90cf3ccbe Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 5 Jun 2024 14:03:37 -0400 Subject: ftrace: Add comments to ftrace_hash_rec_disable/enable() Add comments to describe what the functions ftrace_hash_rec_disable() and ftrace_hash_rec_enable() do. Also change the passing of the "inc" variable to __ftrace_hash_rec_update() to a boolean value as that is what it is supposed to take. Link: https://lore.kernel.org/linux-trace-kernel/20240605180408.857333430@goodmis.org Cc: Mathieu Desnoyers Cc: Andrew Morton Acked-by: Mark Rutland Acked-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ftrace.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index abcdffa4d572..4140d0ce25f1 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -1863,14 +1863,24 @@ static bool __ftrace_hash_rec_update(struct ftrace_ops *ops, return update; } +/* + * This is called when an ops is removed from tracing. It will decrement + * the counters of the dyn_ftrace records for all the functions that + * the @ops attached to. + */ static bool ftrace_hash_rec_disable(struct ftrace_ops *ops) { - return __ftrace_hash_rec_update(ops, 0); + return __ftrace_hash_rec_update(ops, false); } +/* + * This is called when an ops is added to tracing. It will increment + * the counters of the dyn_ftrace records for all the functions that + * the @ops attached to. + */ static bool ftrace_hash_rec_enable(struct ftrace_ops *ops) { - return __ftrace_hash_rec_update(ops, 1); + return __ftrace_hash_rec_update(ops, true); } static void ftrace_hash_rec_update_modify(struct ftrace_ops *ops, int inc) -- cgit v1.2.3 From 1a88c071679496907ba3e62ec1190d198c20ace8 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 5 Jun 2024 14:03:38 -0400 Subject: ftrace: Convert "inc" parameter to bool in ftrace_hash_rec_update_modify() The parameter "inc" in the function ftrace_hash_rec_update_modify() is boolean. Change it to be such. Also add documentation to what the function does. Link: https://lore.kernel.org/linux-trace-kernel/20240605180409.021080462@goodmis.org Cc: Mathieu Desnoyers Cc: Andrew Morton Acked-by: Masami Hiramatsu (Google) Acked-by: Mark Rutland Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ftrace.c | 23 ++++++++++++++++++++--- 1 file changed, 20 insertions(+), 3 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 4140d0ce25f1..256b5e07c39a 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -1883,7 +1883,24 @@ static bool ftrace_hash_rec_enable(struct ftrace_ops *ops) return __ftrace_hash_rec_update(ops, true); } -static void ftrace_hash_rec_update_modify(struct ftrace_ops *ops, int inc) +/* + * This function will update what functions @ops traces when its filter + * changes. + * + * The @inc states if the @ops callbacks are going to be added or removed. + * When one of the @ops hashes are updated to a "new_hash" the dyn_ftrace + * records are update via: + * + * ftrace_hash_rec_disable_modify(ops); + * ops->hash = new_hash + * ftrace_hash_rec_enable_modify(ops); + * + * Where the @ops is removed from all the records it is tracing using + * its old hash. The @ops hash is updated to the new hash, and then + * the @ops is added back to the records so that it is tracing all + * the new functions. + */ +static void ftrace_hash_rec_update_modify(struct ftrace_ops *ops, bool inc) { struct ftrace_ops *op; @@ -1907,12 +1924,12 @@ static void ftrace_hash_rec_update_modify(struct ftrace_ops *ops, int inc) static void ftrace_hash_rec_disable_modify(struct ftrace_ops *ops) { - ftrace_hash_rec_update_modify(ops, 0); + ftrace_hash_rec_update_modify(ops, false); } static void ftrace_hash_rec_enable_modify(struct ftrace_ops *ops) { - ftrace_hash_rec_update_modify(ops, 1); + ftrace_hash_rec_update_modify(ops, true); } /* -- cgit v1.2.3 From d66bb33479e594ba7cf72fa2c0f848e3d0bc9eb1 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 5 Jun 2024 14:03:39 -0400 Subject: ftrace: Add comments to ftrace_hash_move() and friends Describe what ftrace_hash_move() does and add some more comments to some other functions to make it easier to understand. Link: https://lore.kernel.org/linux-trace-kernel/20240605180409.179520305@goodmis.org Cc: Mathieu Desnoyers Cc: Andrew Morton Acked-by: Mark Rutland Acked-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ftrace.c | 24 +++++++++++++++++++++++- 1 file changed, 23 insertions(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 256b5e07c39a..9c4d01b1bb68 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -169,6 +169,7 @@ static inline void ftrace_ops_init(struct ftrace_ops *ops) #endif } +/* Call this function for when a callback filters on set_ftrace_pid */ static void ftrace_pid_func(unsigned long ip, unsigned long parent_ip, struct ftrace_ops *op, struct ftrace_regs *fregs) { @@ -1318,7 +1319,7 @@ static struct ftrace_hash *alloc_ftrace_hash(int size_bits) return hash; } - +/* Used to save filters on functions for modules not loaded yet */ static int ftrace_add_mod(struct trace_array *tr, const char *func, const char *module, int enable) @@ -1430,6 +1431,7 @@ static struct ftrace_hash *__move_hash(struct ftrace_hash *src, int size) return new_hash; } +/* Move the @src entries to a newly allocated hash */ static struct ftrace_hash * __ftrace_hash_move(struct ftrace_hash *src) { @@ -1444,6 +1446,26 @@ __ftrace_hash_move(struct ftrace_hash *src) return __move_hash(src, size); } +/** + * ftrace_hash_move - move a new hash to a filter and do updates + * @ops: The ops with the hash that @dst points to + * @enable: True if for the filter hash, false for the notrace hash + * @dst: Points to the @ops hash that should be updated + * @src: The hash to update @dst with + * + * This is called when an ftrace_ops hash is being updated and the + * the kernel needs to reflect this. Note, this only updates the kernel + * function callbacks if the @ops is enabled (not to be confused with + * @enable above). If the @ops is enabled, its hash determines what + * callbacks get called. This function gets called when the @ops hash + * is updated and it requires new callbacks. + * + * On success the elements of @src is moved to @dst, and @dst is updated + * properly, as well as the functions determined by the @ops hashes + * are now calling the @ops callback function. + * + * Regardless of return type, @src should be freed with free_ftrace_hash(). + */ static int ftrace_hash_move(struct ftrace_ops *ops, int enable, struct ftrace_hash **dst, struct ftrace_hash *src) -- cgit v1.2.3 From 1f51ba905e792a4f6d7b313756a4f99386990ddf Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 5 Jun 2024 16:26:46 -0400 Subject: ftrace: Assign ftrace_list_end to ftrace_ops_list type cast to RCU Use a type cast to convert ftrace_list_end to RCU when assigning ftrace_ops_list. This will quiet the sparse warning: kernel/trace/ftrace.c:125:59: warning: incorrect type in initializer (different address spaces) kernel/trace/ftrace.c:125:59: expected struct ftrace_ops [noderef] __rcu *[addressable] [toplevel] ftrace_ops_list kernel/trace/ftrace.c:125:59: got struct ftrace_ops * Link: https://lore.kernel.org/linux-trace-kernel/20240605202708.450784356@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Acked-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ftrace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 9c4d01b1bb68..034242675e7b 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -122,7 +122,7 @@ static int ftrace_disabled __read_mostly; DEFINE_MUTEX(ftrace_lock); -struct ftrace_ops __rcu *ftrace_ops_list __read_mostly = &ftrace_list_end; +struct ftrace_ops __rcu *ftrace_ops_list __read_mostly = (struct ftrace_ops __rcu *)&ftrace_list_end; ftrace_func_t ftrace_trace_function __read_mostly = ftrace_stub; struct ftrace_ops global_ops; -- cgit v1.2.3 From 0ddef5d601ff992430b7874074c3549aa66f2c85 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 5 Jun 2024 16:26:47 -0400 Subject: ftrace: Assign RCU list variable with rcu_assign_ptr() Use rcu_assign_ptr() to assign the list pointer as it is marked as RCU, and this quiets the sparse warning: kernel/trace/ftrace.c:313:23: warning: incorrect type in assignment (different address spaces) kernel/trace/ftrace.c:313:23: expected struct ftrace_ops [noderef] __rcu * kernel/trace/ftrace.c:313:23: got struct ftrace_ops * Link: https://lore.kernel.org/linux-trace-kernel/20240605202708.613471310@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Acked-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ftrace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 034242675e7b..4aeb1183ea9f 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -311,7 +311,7 @@ static int remove_ftrace_ops(struct ftrace_ops __rcu **list, lockdep_is_held(&ftrace_lock)) == ops && rcu_dereference_protected(ops->next, lockdep_is_held(&ftrace_lock)) == &ftrace_list_end) { - *list = &ftrace_list_end; + rcu_assign_pointer(*list, &ftrace_list_end); return 0; } -- cgit v1.2.3 From 86b49970e071921490ebab3d982d93dd616cd967 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 5 Jun 2024 16:26:48 -0400 Subject: ftrace: Fix prototypes for ftrace_startup/shutdown_subops() The ftrace_startup_subops() was in the wrong header, and both functions were not defined on !CONFIG_DYNAMIC_FTRACE. Link: https://lore.kernel.org/linux-trace-kernel/20240605202708.773583114@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Fixes: 5fccc7552ccbc ("ftrace: Add subops logic to allow one ops to manage many") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202406051524.a12JqLqx-lkp@intel.com/ Acked-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ftrace_internal.h | 9 +++++++++ kernel/trace/trace.h | 1 - 2 files changed, 9 insertions(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace_internal.h b/kernel/trace/ftrace_internal.h index bfba10c2fcf1..4bb1e881154a 100644 --- a/kernel/trace/ftrace_internal.h +++ b/kernel/trace/ftrace_internal.h @@ -15,6 +15,7 @@ extern struct ftrace_ops global_ops; int ftrace_startup(struct ftrace_ops *ops, int command); int ftrace_shutdown(struct ftrace_ops *ops, int command); int ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip, void *regs); +int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command); int ftrace_shutdown_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command); #else /* !CONFIG_DYNAMIC_FTRACE */ @@ -39,6 +40,14 @@ ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip, void *regs) { return 1; } +static inline int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command) +{ + return -EINVAL; +} +static inline int ftrace_shutdown_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command) +{ + return -EINVAL; +} #endif /* CONFIG_DYNAMIC_FTRACE */ #ifdef CONFIG_FUNCTION_GRAPH_TRACER diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index b37402e3f0c9..8783bebd0562 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -1186,7 +1186,6 @@ extern int ftrace_set_filter(struct ftrace_ops *ops, unsigned char *buf, int len, int reset); extern int ftrace_set_notrace(struct ftrace_ops *ops, unsigned char *buf, int len, int reset); -extern int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command); #else struct ftrace_func_command; -- cgit v1.2.3 From 0c4d8cbb2ce20df3a72b5b3189d93b5c0cc3e968 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 5 Jun 2024 16:26:49 -0400 Subject: function_graph: Make fgraph_do_direct static key static The static branch key "fgraph_do_direct" was not declared static but is only used in one file. Change it to a static variable. Link: https://lore.kernel.org/linux-trace-kernel/20240605202708.936515302@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Fixes: cc60ee813b503 ("function_graph: Use static_call and branch to optimize entry function") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202406051711.dS1sQZ9n-lkp@intel.com/ Acked-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index 4bf91eebbb08..63d828054c79 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -515,7 +515,7 @@ static struct fgraph_ops fgraph_stub = { static struct fgraph_ops *fgraph_direct_gops = &fgraph_stub; DEFINE_STATIC_CALL(fgraph_func, ftrace_graph_entry_stub); DEFINE_STATIC_CALL(fgraph_retfunc, ftrace_graph_ret_stub); -DEFINE_STATIC_KEY_TRUE(fgraph_do_direct); +static DEFINE_STATIC_KEY_TRUE(fgraph_do_direct); /** * ftrace_graph_stop - set to permanently disable function graph tracing -- cgit v1.2.3 From 4057fd2cddaea0d6f6c5f8af4a455741f5632259 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 5 Jun 2024 16:26:50 -0400 Subject: function_graph: Do not update pid func if CONFIG_DYNAMIC_FTRACE not enabled The ftrace subops is only defined if CONFIG_DYNAMIC_FTRACE is enabled. If it is not, function tracing is extremely limited, and the subops in the ftrace_ops structure is not defined (and will fail to compile). If DYNAMIC_FTRACE is not enabled, then function graph filtering will not work (as it shouldn't). Link: https://lore.kernel.org/linux-trace-kernel/20240605202709.096020676@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Fixes: df3ec5da6a1e7 ("function_graph: Add pid tracing back to function graph tracer") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202406051855.9VIYXbTB-lkp@intel.com/ Acked-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index 63d828054c79..c0e428c87ea5 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -1177,6 +1177,7 @@ void fgraph_update_pid_func(void) if (!(graph_ops.flags & FTRACE_OPS_FL_INITIALIZED)) return; +#ifdef CONFIG_DYNAMIC_FTRACE list_for_each_entry(op, &graph_ops.subop_list, list) { if (op->flags & FTRACE_OPS_FL_PID) { gops = container_of(op, struct fgraph_ops, ops); @@ -1186,6 +1187,7 @@ void fgraph_update_pid_func(void) static_call_update(fgraph_func, gops->entryfunc); } } +#endif } /* Allocate a return stack for each task */ -- cgit v1.2.3 From 6c1f7f0acabd35ad73525bd2766da4669c18a9e6 Mon Sep 17 00:00:00 2001 From: Tatsuya S Date: Tue, 28 May 2024 12:24:14 +0900 Subject: ftrace: Hide one more entry in stack trace when ftrace_pid is enabled On setting set_ftrace_pid, a extra entry generated by ftrace_pid_func() is shown on stack trace(CONFIG_UNWINDER_FRAME_POINTER=y). [004] ..... 68.459382: => 0xffffffffa00090af => ksys_read => __x64_sys_read => x64_sys_call => do_syscall_64 => entry_SYSCALL_64_after_hwframe To resolve this issue, increment skip count in function_stack_trace_call() if pids are set. Link: https://lore.kernel.org/linux-trace-kernel/20240528032604.6813-3-tatsuya.s2862@gmail.com Signed-off-by: Tatsuya S [ Rebased to current tree ] Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace_functions.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c index 13bf2415245d..3b0cea37e029 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -231,6 +231,7 @@ function_stack_trace_call(unsigned long ip, unsigned long parent_ip, long disabled; int cpu; unsigned int trace_ctx; + int skip = STACK_SKIP; if (unlikely(!tr->function_enabled)) return; @@ -247,7 +248,11 @@ function_stack_trace_call(unsigned long ip, unsigned long parent_ip, if (likely(disabled == 1)) { trace_ctx = tracing_gen_ctx_flags(flags); trace_function(tr, ip, parent_ip, trace_ctx); - __trace_stack(tr, trace_ctx, STACK_SKIP); +#ifdef CONFIG_UNWINDER_FRAME_POINTER + if (ftrace_pids_enabled(op)) + skip++; +#endif + __trace_stack(tr, trace_ctx, skip); } atomic_dec(&data->disabled); -- cgit v1.2.3 From 9a2a3aab7329f1e62982531559866b90d5a5208b Mon Sep 17 00:00:00 2001 From: Jiapeng Chong Date: Thu, 6 Jun 2024 10:10:53 +0800 Subject: fgraph: Remove some unused functions These functions are defined in the fgraph.c file, but not called elsewhere, so delete these unused functions. kernel/trace/fgraph.c:273:1: warning: unused function 'set_bitmap_bits'. kernel/trace/fgraph.c:259:19: warning: unused function 'get_fgraph_type'. Link: https://lore.kernel.org/linux-trace-kernel/20240606021053.27783-1-jiapeng.chong@linux.alibaba.com Reported-by: Abaci Robot Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=9289 Signed-off-by: Jiapeng Chong Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 13 ------------- 1 file changed, 13 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index c0e428c87ea5..a13551a023aa 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -255,12 +255,6 @@ static inline int get_frame_offset(struct task_struct *t, int offset) return __get_offset(t->ret_stack[offset]); } -/* Get FGRAPH_TYPE from the word from the @offset at ret_stack */ -static inline int get_fgraph_type(struct task_struct *t, int offset) -{ - return __get_type(t->ret_stack[offset]); -} - /* For BITMAP type: get the bitmask from the @offset at ret_stack */ static inline unsigned long get_bitmap_bits(struct task_struct *t, int offset) @@ -268,13 +262,6 @@ get_bitmap_bits(struct task_struct *t, int offset) return (t->ret_stack[offset] >> FGRAPH_INDEX_SHIFT) & FGRAPH_INDEX_MASK; } -/* For BITMAP type: set the bits in the bitmap bitmask at @offset on ret_stack */ -static inline void -set_bitmap_bits(struct task_struct *t, int offset, unsigned long bitmap) -{ - t->ret_stack[offset] |= (bitmap << FGRAPH_INDEX_SHIFT); -} - /* Write the bitmap to the ret_stack at @offset (does index, offset and bitmask) */ static inline void set_bitmap(struct task_struct *t, int offset, unsigned long bitmap) -- cgit v1.2.3 From 2f6b884dfcc55065b76d2bf1e2424b93991ae92d Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Thu, 6 Jun 2024 08:18:46 -0400 Subject: function_graph: Rename BYTE_NUMBER to CHAR_NUMBER in selftests The function_graph selftests checks various size variables to pass from the entry of the function to the exit. It tests 1, 2, 4 and 8 byte words. The 1 byte macro was called BYTE_NUMBER but that is used in the sh architecture: arch/sh/include/asm/bitops-op32.h Just rename the macro to CHAR_NUMBER. Link: https://lore.kernel.org/linux-trace-kernel/20240606081846.4cb82dc4@gandalf.local.home Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Fixes: 47c3c70aa3697 ("function_graph: Add selftest for passing local variables") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202406061744.rZDXfRrG-lkp@intel.com/ Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace_selftest.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index 369efc569238..adf0f436d84b 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -758,7 +758,7 @@ trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr) #ifdef CONFIG_DYNAMIC_FTRACE -#define BYTE_NUMBER 123 +#define CHAR_NUMBER 123 #define SHORT_NUMBER 12345 #define WORD_NUMBER 1234567890 #define LONG_NUMBER 1234567890123456789LL @@ -789,7 +789,7 @@ static __init int store_entry(struct ftrace_graph_ent *trace, switch (size) { case 1: - *(char *)p = BYTE_NUMBER; + *(char *)p = CHAR_NUMBER; break; case 2: *(short *)p = SHORT_NUMBER; @@ -830,7 +830,7 @@ static __init void store_return(struct ftrace_graph_ret *trace, switch (fixture->store_size) { case 1: - expect = BYTE_NUMBER; + expect = CHAR_NUMBER; found = *(char *)p; break; case 2: -- cgit v1.2.3 From 4267fda4afd9fcf1f92b6bfa0259295f140c8ecd Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Fri, 7 Jun 2024 09:48:33 -0400 Subject: function_graph: Make fgraph_update_pid_func() a stub for !DYNAMIC_FTRACE When CONFIG_DYNAMIC_FTRACE is not set, the function fgraph_update_pid_func() doesn't do anything. Currently, most of its logic is within a "#ifdef CONFIG_DYNAMIC_FTRACE" block, but its variables were declared outside that, and when DYNAMIC_FTRACE is not set, it produces unused variable warnings. Instead, just place it (and the helper function fgraph_pid_func()) within the #ifdef block and have the header file use a empty stub function for when DYNAMIC_FTRACE is not defined. Link: https://lore.kernel.org/linux-trace-kernel/20240607094833.6a787d73@rorschach.local.home Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202406071806.BRjaC5FF-lkp@intel.com/ Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 4 ++-- kernel/trace/ftrace_internal.h | 4 ++++ 2 files changed, 6 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index a13551a023aa..63d0c2f84ce1 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -1138,6 +1138,7 @@ void ftrace_graph_exit_task(struct task_struct *t) kfree(ret_stack); } +#ifdef CONFIG_DYNAMIC_FTRACE static int fgraph_pid_func(struct ftrace_graph_ent *trace, struct fgraph_ops *gops) { @@ -1164,7 +1165,6 @@ void fgraph_update_pid_func(void) if (!(graph_ops.flags & FTRACE_OPS_FL_INITIALIZED)) return; -#ifdef CONFIG_DYNAMIC_FTRACE list_for_each_entry(op, &graph_ops.subop_list, list) { if (op->flags & FTRACE_OPS_FL_PID) { gops = container_of(op, struct fgraph_ops, ops); @@ -1174,8 +1174,8 @@ void fgraph_update_pid_func(void) static_call_update(fgraph_func, gops->entryfunc); } } -#endif } +#endif /* Allocate a return stack for each task */ static int start_graph_tracing(void) diff --git a/kernel/trace/ftrace_internal.h b/kernel/trace/ftrace_internal.h index 4bb1e881154a..3235470e61b3 100644 --- a/kernel/trace/ftrace_internal.h +++ b/kernel/trace/ftrace_internal.h @@ -52,7 +52,11 @@ static inline int ftrace_shutdown_subops(struct ftrace_ops *ops, struct ftrace_o #ifdef CONFIG_FUNCTION_GRAPH_TRACER extern int ftrace_graph_active; +# ifdef CONFIG_DYNAMIC_FTRACE extern void fgraph_update_pid_func(void); +# else +static inline void fgraph_update_pid_func(void) {} +# endif #else /* !CONFIG_FUNCTION_GRAPH_TRACER */ # define ftrace_graph_active 0 static inline void fgraph_update_pid_func(void) {} -- cgit v1.2.3 From 29c1c24a2707a579bba078d8b73b70fef4451b34 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Mon, 10 Jun 2024 23:09:35 -0400 Subject: function_graph: Fix up ftrace_graph_ret_addr() Yang Li sent a patch to fix the kerneldoc of ftrace_graph_ret_addr(). While reviewing it, I realized that the comments in the entire function header needed a rewrite. When doing that, I realized that @idx parameter was being ignored. Every time this was called by the unwinder, it would start the loop at the top of the shadow stack and look for the matching stack pointer. When it found it, it would return it. When the unwinder asked for the next function, it would search from the beginning again. In reality, it should start from where it left off. That was the reason for the @idx parameter in the first place. The first time the unwinder calls this function, the @idx pointer would contain zero. That would mean to start from the top of the stack. The function was supposed to update the @idx with the index where it found the return address, so that the next time the unwinder calls this function it doesn't have to search through the previous addresses it found (making it O(n^2)!). This speeds up the unwinder's use of ftrace_graph_ret_addr() by an order of magnitude. Link: https://lore.kernel.org/linux-trace-kernel/20240610181746.656e3759@gandalf.local.home/ Link: https://lore.kernel.org/linux-trace-kernel/20240611031737.821995106@goodmis.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Jonathan Corbet Cc: Catalin Marinas Cc: Will Deacon Cc: Guo Ren Cc: Huacai Chen Cc: WANG Xuerui Cc: Michael Ellerman Cc: Nicholas Piggin Cc: Christophe Leroy Cc: "Naveen N. Rao" Cc: Paul Walmsley Cc: Palmer Dabbelt Cc: Albert Ou Cc: Heiko Carstens Cc: Vasily Gorbik Cc: Alexander Gordeev Cc: Christian Borntraeger Cc: Sven Schnelle Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc: Dave Hansen Cc: "H. Peter Anvin" Reported-by: Yang Li Fixes: 7aa1eaef9f428 ("function_graph: Allow multiple users to attach to function graph") Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 28 ++++++++++++++++++++-------- 1 file changed, 20 insertions(+), 8 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index 63d0c2f84ce1..91f1eef256af 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -870,18 +870,24 @@ ftrace_graph_get_ret_stack(struct task_struct *task, int idx) } /** - * ftrace_graph_ret_addr - convert a potentially modified stack return address - * to its original value + * ftrace_graph_ret_addr - return the original value of the return address + * @task: The task the unwinder is being executed on + * @idx: An initialized pointer to the next stack index to use + * @ret: The current return address (likely pointing to return_handler) + * @retp: The address on the stack of the current return location * * This function can be called by stack unwinding code to convert a found stack - * return address ('ret') to its original value, in case the function graph + * return address (@ret) to its original value, in case the function graph * tracer has modified it to be 'return_to_handler'. If the address hasn't - * been modified, the unchanged value of 'ret' is returned. + * been modified, the unchanged value of @ret is returned. * - * 'idx' is a state variable which should be initialized by the caller to zero - * before the first call. + * @idx holds the last index used to know where to start from. It should be + * initialized to zero for the first iteration as that will mean to start + * at the top of the shadow stack. If the location is found, this pointer + * will be assigned that location so that if called again, it will continue + * where it left off. * - * 'retp' is a pointer to the return address on the stack. It's ignored if + * @retp is a pointer to the return address on the stack. It's ignored if * the arch doesn't have HAVE_FUNCTION_GRAPH_RET_ADDR_PTR defined. */ #ifdef HAVE_FUNCTION_GRAPH_RET_ADDR_PTR @@ -895,6 +901,10 @@ unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, if (ret != return_handler) return ret; + if (!idx) + return ret; + + i = *idx ? : task->curr_ret_stack; while (i > 0) { ret_stack = get_ret_stack(current, i, &i); if (!ret_stack) @@ -908,8 +918,10 @@ unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, * Thus we will continue to find real return address. */ if (ret_stack->retp == retp && - ret_stack->ret != return_handler) + ret_stack->ret != return_handler) { + *idx = i; return ret_stack->ret; + } } return ret; -- cgit v1.2.3 From 5f7fb89a115d53b4a10bf7ba2733e78df281e98d Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Mon, 10 Jun 2024 23:09:36 -0400 Subject: function_graph: Everyone uses HAVE_FUNCTION_GRAPH_RET_ADDR_PTR, remove it All architectures that implement function graph also implements HAVE_FUNCTION_GRAPH_RET_ADDR_PTR. Remove it, as it is no longer a differentiator. Link: https://lore.kernel.org/linux-trace-kernel/20240611031737.982047614@goodmis.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Jonathan Corbet Cc: Catalin Marinas Cc: Will Deacon Cc: Guo Ren Cc: Huacai Chen Cc: WANG Xuerui Cc: Michael Ellerman Cc: Nicholas Piggin Cc: Christophe Leroy Cc: "Naveen N. Rao" Cc: Paul Walmsley Cc: Palmer Dabbelt Cc: Albert Ou Cc: Heiko Carstens Cc: Vasily Gorbik Cc: Alexander Gordeev Cc: Christian Borntraeger Cc: Sven Schnelle Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc: Dave Hansen Cc: "H. Peter Anvin" Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 35 +---------------------------------- 1 file changed, 1 insertion(+), 34 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index 91f1eef256af..8317d1a7f43a 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -593,9 +593,7 @@ ftrace_push_return_trace(unsigned long ret, unsigned long func, #ifdef HAVE_FUNCTION_GRAPH_FP_TEST ret_stack->fp = frame_pointer; #endif -#ifdef HAVE_FUNCTION_GRAPH_RET_ADDR_PTR ret_stack->retp = retp; -#endif return offset; } @@ -887,10 +885,8 @@ ftrace_graph_get_ret_stack(struct task_struct *task, int idx) * will be assigned that location so that if called again, it will continue * where it left off. * - * @retp is a pointer to the return address on the stack. It's ignored if - * the arch doesn't have HAVE_FUNCTION_GRAPH_RET_ADDR_PTR defined. + * @retp is a pointer to the return address on the stack. */ -#ifdef HAVE_FUNCTION_GRAPH_RET_ADDR_PTR unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, unsigned long ret, unsigned long *retp) { @@ -926,35 +922,6 @@ unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, return ret; } -#else /* !HAVE_FUNCTION_GRAPH_RET_ADDR_PTR */ -unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, - unsigned long ret, unsigned long *retp) -{ - struct ftrace_ret_stack *ret_stack; - unsigned long return_handler = (unsigned long)dereference_kernel_function_descriptor(return_to_handler); - int offset = task->curr_ret_stack; - int i; - - if (ret != return_handler) - return ret; - - if (!idx) - return ret; - - i = *idx; - do { - ret_stack = get_ret_stack(task, offset, &offset); - if (ret_stack && ret_stack->ret == return_handler) - continue; - i--; - } while (i >= 0 && ret_stack); - - if (ret_stack) - return ret_stack->ret; - - return ret; -} -#endif /* HAVE_FUNCTION_GRAPH_RET_ADDR_PTR */ static struct ftrace_ops graph_ops = { .func = ftrace_graph_func, -- cgit v1.2.3 From 9b5a45eb639c46c0374b5e040e6e6db386909676 Mon Sep 17 00:00:00 2001 From: Marilene A Garcia Date: Thu, 6 Jun 2024 10:25:20 -0300 Subject: ftrace: Add missing kerneldoc parameters to unregister_ftrace_direct() Add the description to the parameters addr and free_filters of the function unregister_ftrace_direct(). Link: https://lore.kernel.org/linux-trace-kernel/20240606132520.1397567-1-marilene.agarcia@gmail.com Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Shuah Khan Cc: Javier Carrasco Signed-off-by: Marilene A Garcia Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ftrace.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 4aeb1183ea9f..f44229294e9d 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -5988,6 +5988,8 @@ EXPORT_SYMBOL_GPL(register_ftrace_direct); * unregister_ftrace_direct - Remove calls to custom trampoline * previously registered by register_ftrace_direct for @ops object. * @ops: The address of the struct ftrace_ops object + * @addr: The address of the direct function that is called by the @ops functions + * @free_filters: Set to true to remove all filters for the ftrace_ops, false otherwise * * This is used to remove a direct calls to @addr from the nop locations * of the functions registered in @ops (with by ftrace_set_filter_ip -- cgit v1.2.3 From 3572bd5689b0812b161b40279e39ca5b66d73e88 Mon Sep 17 00:00:00 2001 From: "Masami Hiramatsu (Google)" Date: Tue, 11 Jun 2024 22:30:37 +0900 Subject: tracing: Build event generation tests only as modules The kprobes and synth event generation test modules add events and lock (get a reference) those event file reference in module init function, and unlock and delete it in module exit function. This is because those are designed for playing as modules. If we make those modules as built-in, those events are left locked in the kernel, and never be removed. This causes kprobe event self-test failure as below. [ 97.349708] ------------[ cut here ]------------ [ 97.353453] WARNING: CPU: 3 PID: 1 at kernel/trace/trace_kprobe.c:2133 kprobe_trace_self_tests_init+0x3f1/0x480 [ 97.357106] Modules linked in: [ 97.358488] CPU: 3 PID: 1 Comm: swapper/0 Not tainted 6.9.0-g699646734ab5-dirty #14 [ 97.361556] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014 [ 97.363880] RIP: 0010:kprobe_trace_self_tests_init+0x3f1/0x480 [ 97.365538] Code: a8 24 08 82 e9 ae fd ff ff 90 0f 0b 90 48 c7 c7 e5 aa 0b 82 e9 ee fc ff ff 90 0f 0b 90 48 c7 c7 2d 61 06 82 e9 8e fd ff ff 90 <0f> 0b 90 48 c7 c7 33 0b 0c 82 89 c6 e8 6e 03 1f ff 41 ff c7 e9 90 [ 97.370429] RSP: 0000:ffffc90000013b50 EFLAGS: 00010286 [ 97.371852] RAX: 00000000fffffff0 RBX: ffff888005919c00 RCX: 0000000000000000 [ 97.373829] RDX: ffff888003f40000 RSI: ffffffff8236a598 RDI: ffff888003f40a68 [ 97.375715] RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000000 [ 97.377675] R10: ffffffff811c9ae5 R11: ffffffff8120c4e0 R12: 0000000000000000 [ 97.379591] R13: 0000000000000001 R14: 0000000000000015 R15: 0000000000000000 [ 97.381536] FS: 0000000000000000(0000) GS:ffff88807dcc0000(0000) knlGS:0000000000000000 [ 97.383813] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 97.385449] CR2: 0000000000000000 CR3: 0000000002244000 CR4: 00000000000006b0 [ 97.387347] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 97.389277] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 97.391196] Call Trace: [ 97.391967] [ 97.392647] ? __warn+0xcc/0x180 [ 97.393640] ? kprobe_trace_self_tests_init+0x3f1/0x480 [ 97.395181] ? report_bug+0xbd/0x150 [ 97.396234] ? handle_bug+0x3e/0x60 [ 97.397311] ? exc_invalid_op+0x1a/0x50 [ 97.398434] ? asm_exc_invalid_op+0x1a/0x20 [ 97.399652] ? trace_kprobe_is_busy+0x20/0x20 [ 97.400904] ? tracing_reset_all_online_cpus+0x15/0x90 [ 97.402304] ? kprobe_trace_self_tests_init+0x3f1/0x480 [ 97.403773] ? init_kprobe_trace+0x50/0x50 [ 97.404972] do_one_initcall+0x112/0x240 [ 97.406113] do_initcall_level+0x95/0xb0 [ 97.407286] ? kernel_init+0x1a/0x1a0 [ 97.408401] do_initcalls+0x3f/0x70 [ 97.409452] kernel_init_freeable+0x16f/0x1e0 [ 97.410662] ? rest_init+0x1f0/0x1f0 [ 97.411738] kernel_init+0x1a/0x1a0 [ 97.412788] ret_from_fork+0x39/0x50 [ 97.413817] ? rest_init+0x1f0/0x1f0 [ 97.414844] ret_from_fork_asm+0x11/0x20 [ 97.416285] [ 97.417134] irq event stamp: 13437323 [ 97.418376] hardirqs last enabled at (13437337): [] console_unlock+0x11c/0x150 [ 97.421285] hardirqs last disabled at (13437370): [] console_unlock+0x101/0x150 [ 97.423838] softirqs last enabled at (13437366): [] handle_softirqs+0x23f/0x2a0 [ 97.426450] softirqs last disabled at (13437393): [] __irq_exit_rcu+0x66/0xd0 [ 97.428850] ---[ end trace 0000000000000000 ]--- And also, since we can not cleanup dynamic_event file, ftracetest are failed too. To avoid these issues, build these tests only as modules. Link: https://lore.kernel.org/all/171811263754.85078.5877446624311852525.stgit@devnote2/ Fixes: 9fe41efaca08 ("tracing: Add synth event generation test module") Fixes: 64836248dda2 ("tracing: Add kprobe event command generation test module") Signed-off-by: Masami Hiramatsu (Google) Reviewed-by: Steven Rostedt (Google) --- kernel/trace/Kconfig | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index 166ad5444eea..721c3b221048 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -1136,7 +1136,7 @@ config PREEMPTIRQ_DELAY_TEST config SYNTH_EVENT_GEN_TEST tristate "Test module for in-kernel synthetic event generation" - depends on SYNTH_EVENTS + depends on SYNTH_EVENTS && m help This option creates a test module to check the base functionality of in-kernel synthetic event definition and @@ -1149,7 +1149,7 @@ config SYNTH_EVENT_GEN_TEST config KPROBE_EVENT_GEN_TEST tristate "Test module for in-kernel kprobe event generation" - depends on KPROBE_EVENTS + depends on KPROBE_EVENTS && m help This option creates a test module to check the base functionality of in-kernel kprobe event definition. -- cgit v1.2.3 From 41051daa38a778dd6da49f854442260ebc029894 Mon Sep 17 00:00:00 2001 From: "Masami Hiramatsu (Google)" Date: Tue, 11 Jun 2024 22:30:46 +0900 Subject: tracing/kprobe: Integrate test warnings into WARN_ONCE Cleanup the redundant WARN_ON_ONCE(cond) + pr_warn(msg) into WARN_ONCE(cond, msg). Also add some WARN_ONCE() for hitcount check. These WARN_ONCE() errors makes it easy to handle errors from ktest. Link: https://lore.kernel.org/all/171811264685.85078.8068819097047430463.stgit@devnote2/ Suggested-by: Steven Rostedt Signed-off-by: Masami Hiramatsu (Google) Reviewed-by: Steven Rostedt (Google) --- kernel/trace/trace_kprobe.c | 54 ++++++++++++++++----------------------------- 1 file changed, 19 insertions(+), 35 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 16383247bdbf..8c5816c04bd2 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -2023,19 +2023,16 @@ static __init int kprobe_trace_self_tests_init(void) pr_info("Testing kprobe tracing: "); ret = create_or_delete_trace_kprobe("p:testprobe kprobe_trace_selftest_target $stack $stack0 +0($stack)"); - if (WARN_ON_ONCE(ret)) { - pr_warn("error on probing function entry.\n"); + if (WARN_ONCE(ret, "error on probing function entry.")) { warn++; } else { /* Enable trace point */ tk = find_trace_kprobe("testprobe", KPROBE_EVENT_SYSTEM); - if (WARN_ON_ONCE(tk == NULL)) { - pr_warn("error on getting new probe.\n"); + if (WARN_ONCE(tk == NULL, "error on probing function entry.")) { warn++; } else { file = find_trace_probe_file(tk, top_trace_array()); - if (WARN_ON_ONCE(file == NULL)) { - pr_warn("error on getting probe file.\n"); + if (WARN_ONCE(file == NULL, "error on getting probe file.")) { warn++; } else enable_trace_kprobe( @@ -2044,19 +2041,16 @@ static __init int kprobe_trace_self_tests_init(void) } ret = create_or_delete_trace_kprobe("r:testprobe2 kprobe_trace_selftest_target $retval"); - if (WARN_ON_ONCE(ret)) { - pr_warn("error on probing function return.\n"); + if (WARN_ONCE(ret, "error on probing function return.")) { warn++; } else { /* Enable trace point */ tk = find_trace_kprobe("testprobe2", KPROBE_EVENT_SYSTEM); - if (WARN_ON_ONCE(tk == NULL)) { - pr_warn("error on getting 2nd new probe.\n"); + if (WARN_ONCE(tk == NULL, "error on getting 2nd new probe.")) { warn++; } else { file = find_trace_probe_file(tk, top_trace_array()); - if (WARN_ON_ONCE(file == NULL)) { - pr_warn("error on getting probe file.\n"); + if (WARN_ONCE(file == NULL, "error on getting probe file.")) { warn++; } else enable_trace_kprobe( @@ -2079,18 +2073,15 @@ static __init int kprobe_trace_self_tests_init(void) /* Disable trace points before removing it */ tk = find_trace_kprobe("testprobe", KPROBE_EVENT_SYSTEM); - if (WARN_ON_ONCE(tk == NULL)) { - pr_warn("error on getting test probe.\n"); + if (WARN_ONCE(tk == NULL, "error on getting test probe.")) { warn++; } else { - if (trace_kprobe_nhit(tk) != 1) { - pr_warn("incorrect number of testprobe hits\n"); + if (WARN_ONCE(trace_kprobe_nhit(tk) != 1, + "incorrect number of testprobe hits.")) warn++; - } file = find_trace_probe_file(tk, top_trace_array()); - if (WARN_ON_ONCE(file == NULL)) { - pr_warn("error on getting probe file.\n"); + if (WARN_ONCE(file == NULL, "error on getting probe file.")) { warn++; } else disable_trace_kprobe( @@ -2098,18 +2089,15 @@ static __init int kprobe_trace_self_tests_init(void) } tk = find_trace_kprobe("testprobe2", KPROBE_EVENT_SYSTEM); - if (WARN_ON_ONCE(tk == NULL)) { - pr_warn("error on getting 2nd test probe.\n"); + if (WARN_ONCE(tk == NULL, "error on getting 2nd test probe.")) { warn++; } else { - if (trace_kprobe_nhit(tk) != 1) { - pr_warn("incorrect number of testprobe2 hits\n"); + if (WARN_ONCE(trace_kprobe_nhit(tk) != 1, + "incorrect number of testprobe2 hits.")) warn++; - } file = find_trace_probe_file(tk, top_trace_array()); - if (WARN_ON_ONCE(file == NULL)) { - pr_warn("error on getting probe file.\n"); + if (WARN_ONCE(file == NULL, "error on getting probe file.")) { warn++; } else disable_trace_kprobe( @@ -2117,23 +2105,19 @@ static __init int kprobe_trace_self_tests_init(void) } ret = create_or_delete_trace_kprobe("-:testprobe"); - if (WARN_ON_ONCE(ret)) { - pr_warn("error on deleting a probe.\n"); + if (WARN_ONCE(ret, "error on deleting a probe.")) warn++; - } ret = create_or_delete_trace_kprobe("-:testprobe2"); - if (WARN_ON_ONCE(ret)) { - pr_warn("error on deleting a probe.\n"); + if (WARN_ONCE(ret, "error on deleting a probe.")) warn++; - } + end: ret = dyn_events_release_all(&trace_kprobe_ops); - if (WARN_ON_ONCE(ret)) { - pr_warn("error on cleaning up probes.\n"); + if (WARN_ONCE(ret, "error on cleaning up probes.")) warn++; - } + /* * Wait for the optimizer work to finish. Otherwise it might fiddle * with probes in already freed __init text. -- cgit v1.2.3 From 3eddb031965ae9a95ba098ae6eb81b082e024c65 Mon Sep 17 00:00:00 2001 From: "Masami Hiramatsu (Google)" Date: Tue, 11 Jun 2024 22:30:56 +0900 Subject: tracing/kprobe: Remove cleanup code unrelated to selftest This cleanup all kprobe events code is not related to the selftest itself, and it can fail by the reason unrelated to this test. If the test is successful, the generated events are cleaned up. And if not, we cannot guarantee that the kprobe events will work correctly. So, anyway, there is no need to clean it up. Link: https://lore.kernel.org/all/171811265627.85078.16897867213512435822.stgit@devnote2/ Signed-off-by: Masami Hiramatsu (Google) Reviewed-by: Steven Rostedt (Google) --- kernel/trace/trace_kprobe.c | 4 ---- 1 file changed, 4 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 8c5816c04bd2..7fd0f8576e4c 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -2114,10 +2114,6 @@ static __init int kprobe_trace_self_tests_init(void) end: - ret = dyn_events_release_all(&trace_kprobe_ops); - if (WARN_ONCE(ret, "error on cleaning up probes.")) - warn++; - /* * Wait for the optimizer work to finish. Otherwise it might fiddle * with probes in already freed __init text. -- cgit v1.2.3 From 2b8dd87332cd2782b5b3f0c423bd6693e487ed30 Mon Sep 17 00:00:00 2001 From: Daniel Xu Date: Wed, 12 Jun 2024 09:58:30 -0600 Subject: bpf: Make bpf_session_cookie() kfunc return long * We will soon be generating kfunc prototypes from BTF. As part of that, we need to align the manual signatures in bpf_kfuncs.h with the actual kfunc definitions. There is currently a conflicting signature for bpf_session_cookie() w.r.t. return type. The original intent was to return long * and not __u64 *. You can see evidence of that intent in a3a5113393cc ("selftests/bpf: Add kprobe session cookie test"). Fix conflict by changing kfunc definition. Fixes: 5c919acef851 ("bpf: Add support for kprobe session cookie") Signed-off-by: Daniel Xu Link: https://lore.kernel.org/r/7043e1c251ab33151d6e3830f8ea1902ed2604ac.1718207789.git.dxu@dxuuu.xyz Signed-off-by: Alexei Starovoitov --- kernel/trace/bpf_trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index d1daeab1bbc1..bc16e21a2a44 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -3527,7 +3527,7 @@ __bpf_kfunc bool bpf_session_is_return(void) return session_ctx->is_return; } -__bpf_kfunc __u64 *bpf_session_cookie(void) +__bpf_kfunc long *bpf_session_cookie(void) { struct bpf_session_run_ctx *session_ctx; -- cgit v1.2.3 From cce4c40b960673f9e020835def310f1e89d3a940 Mon Sep 17 00:00:00 2001 From: Daniel Xu Date: Wed, 12 Jun 2024 09:58:33 -0600 Subject: bpf: treewide: Align kfunc signatures to prog point-of-view Previously, kfunc declarations in bpf_kfuncs.h (and others) used "user facing" types for kfuncs prototypes while the actual kfunc definitions used "kernel facing" types. More specifically: bpf_dynptr vs bpf_dynptr_kern, __sk_buff vs sk_buff, and xdp_md vs xdp_buff. It wasn't an issue before, as the verifier allows aliased types. However, since we are now generating kfunc prototypes in vmlinux.h (in addition to keeping bpf_kfuncs.h around), this conflict creates compilation errors. Fix this conflict by using "user facing" types in kfunc definitions. This results in more casts, but otherwise has no additional runtime cost. Note, similar to 5b268d1ebcdc ("bpf: Have bpf_rdonly_cast() take a const pointer"), we also make kfuncs take const arguments where appropriate in order to make the kfunc more permissive. Signed-off-by: Daniel Xu Link: https://lore.kernel.org/r/b58346a63a0e66bc9b7504da751b526b0b189a67.1718207789.git.dxu@dxuuu.xyz Signed-off-by: Alexei Starovoitov --- kernel/trace/bpf_trace.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index bc16e21a2a44..4b3fda456299 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -1369,8 +1369,8 @@ __bpf_kfunc void bpf_key_put(struct bpf_key *bkey) #ifdef CONFIG_SYSTEM_DATA_VERIFICATION /** * bpf_verify_pkcs7_signature - verify a PKCS#7 signature - * @data_ptr: data to verify - * @sig_ptr: signature of the data + * @data_p: data to verify + * @sig_p: signature of the data * @trusted_keyring: keyring with keys trusted for signature verification * * Verify the PKCS#7 signature *sig_ptr* against the supplied *data_ptr* @@ -1378,10 +1378,12 @@ __bpf_kfunc void bpf_key_put(struct bpf_key *bkey) * * Return: 0 on success, a negative value on error. */ -__bpf_kfunc int bpf_verify_pkcs7_signature(struct bpf_dynptr_kern *data_ptr, - struct bpf_dynptr_kern *sig_ptr, +__bpf_kfunc int bpf_verify_pkcs7_signature(struct bpf_dynptr *data_p, + struct bpf_dynptr *sig_p, struct bpf_key *trusted_keyring) { + struct bpf_dynptr_kern *data_ptr = (struct bpf_dynptr_kern *)data_p; + struct bpf_dynptr_kern *sig_ptr = (struct bpf_dynptr_kern *)sig_p; const void *data, *sig; u32 data_len, sig_len; int ret; @@ -1444,7 +1446,7 @@ __bpf_kfunc_start_defs(); * bpf_get_file_xattr - get xattr of a file * @file: file to get xattr from * @name__str: name of the xattr - * @value_ptr: output buffer of the xattr value + * @value_p: output buffer of the xattr value * * Get xattr *name__str* of *file* and store the output in *value_ptr*. * @@ -1453,8 +1455,9 @@ __bpf_kfunc_start_defs(); * Return: 0 on success, a negative value on error. */ __bpf_kfunc int bpf_get_file_xattr(struct file *file, const char *name__str, - struct bpf_dynptr_kern *value_ptr) + struct bpf_dynptr *value_p) { + struct bpf_dynptr_kern *value_ptr = (struct bpf_dynptr_kern *)value_p; struct dentry *dentry; u32 value_len; void *value; -- cgit v1.2.3 From 63a8dfb889112ab4a065aa60a9a1b590b410d055 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Thu, 13 Jun 2024 09:52:23 -0400 Subject: function_graph: Add READ_ONCE() when accessing fgraph_array[] In function_graph_enter() there's a loop that looks at fgraph_array[] elements which are fgraph_ops. It first tests if it is a fgraph_stub op, and if so skips it, as that's just there as a place holder. Then it checks the fgraph_ops filters to see if the ops wants to trace the current function. But if the compiler reloads the fgraph_array[] after the check against fgraph_stub, it could race with the fgraph_array[] being updated with the fgraph_stub. That would cause the stub to be processed. But the stub has a null "func_hash" field which will cause a NULL pointer dereference. Add a READ_ONCE() so that the gops that is compared against the fgraph_stub is also the gops that is processed later. Link: https://lore.kernel.org/all/CA+G9fYsSVJQZH=nM=1cjTc94PgSnMF9y65BnOv6XSoCG_b6wmw@mail.gmail.com/ Link: https://lore.kernel.org/linux-trace-kernel/20240613095223.1f07e3a4@rorschach.local.home Cc: Mark Rutland Cc: Mathieu Desnoyers Fixes: cc60ee813b503 ("function_graph: Use static_call and branch to optimize entry function") Reported-by: Naresh Kamboju Signed-off-by: Steven Rostedt (Google) Acked-by: Masami Hiramatsu (Google) --- kernel/trace/fgraph.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index 8317d1a7f43a..fc205ad167a9 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -641,7 +641,7 @@ int function_graph_enter(unsigned long ret, unsigned long func, { for_each_set_bit(i, &fgraph_array_bitmask, sizeof(fgraph_array_bitmask) * BITS_PER_BYTE) { - struct fgraph_ops *gops = fgraph_array[i]; + struct fgraph_ops *gops = READ_ONCE(fgraph_array[i]); int save_curr_ret_stack; if (gops == &fgraph_stub) -- cgit v1.2.3 From 717d6313bba1b3179f0bf1026aaec6b7e26f484e Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Wed, 19 Jun 2024 10:16:24 +0200 Subject: bpf: Change bpf_session_cookie return value to __u64 * This reverts [1] and changes return value for bpf_session_cookie in bpf selftests. Having long * might lead to problems on 32-bit architectures. Fixes: 2b8dd87332cd ("bpf: Make bpf_session_cookie() kfunc return long *") Suggested-by: Andrii Nakryiko Signed-off-by: Jiri Olsa Signed-off-by: Daniel Borkmann Acked-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20240619081624.1620152-1-jolsa@kernel.org --- kernel/trace/bpf_trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 4b3fda456299..cd098846e251 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -3530,7 +3530,7 @@ __bpf_kfunc bool bpf_session_is_return(void) return session_ctx->is_return; } -__bpf_kfunc long *bpf_session_cookie(void) +__bpf_kfunc __u64 *bpf_session_cookie(void) { struct bpf_session_run_ctx *session_ctx; -- cgit v1.2.3 From 7e1f4eb9a60d40dd17a97d9b76818682a024a127 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 4 Apr 2024 12:04:54 +0200 Subject: kallsyms: rework symbol lookup return codes Building with W=1 in some configurations produces a false positive warning for kallsyms: kernel/kallsyms.c: In function '__sprint_symbol.isra': kernel/kallsyms.c:503:17: error: 'strcpy' source argument is the same as destination [-Werror=restrict] 503 | strcpy(buffer, name); | ^~~~~~~~~~~~~~~~~~~~ This originally showed up while building with -O3, but later started happening in other configurations as well, depending on inlining decisions. The underlying issue is that the local 'name' variable is always initialized to the be the same as 'buffer' in the called functions that fill the buffer, which gcc notices while inlining, though it could see that the address check always skips the copy. The calling conventions here are rather unusual, as all of the internal lookup functions (bpf_address_lookup, ftrace_mod_address_lookup, ftrace_func_address_lookup, module_address_lookup and kallsyms_lookup_buildid) already use the provided buffer and either return the address of that buffer to indicate success, or NULL for failure, but the callers are written to also expect an arbitrary other buffer to be returned. Rework the calling conventions to return the length of the filled buffer instead of its address, which is simpler and easier to follow as well as avoiding the warning. Leave only the kallsyms_lookup() calling conventions unchanged, since that is called from 16 different functions and adapting this would be a much bigger change. Link: https://lore.kernel.org/lkml/20200107214042.855757-1-arnd@arndb.de/ Link: https://lore.kernel.org/lkml/20240326130647.7bfb1d92@gandalf.local.home/ Tested-by: Geert Uytterhoeven Reviewed-by: Luis Chamberlain Acked-by: Steven Rostedt (Google) Signed-off-by: Arnd Bergmann --- kernel/trace/ftrace.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 65208d3b5ed9..eacab4020508 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -6969,7 +6969,7 @@ allocate_ftrace_mod_map(struct module *mod, return mod_map; } -static const char * +static int ftrace_func_address_lookup(struct ftrace_mod_map *mod_map, unsigned long addr, unsigned long *size, unsigned long *off, char *sym) @@ -6990,21 +6990,18 @@ ftrace_func_address_lookup(struct ftrace_mod_map *mod_map, *size = found_func->size; if (off) *off = addr - found_func->ip; - if (sym) - strscpy(sym, found_func->name, KSYM_NAME_LEN); - - return found_func->name; + return strscpy(sym, found_func->name, KSYM_NAME_LEN); } - return NULL; + return 0; } -const char * +int ftrace_mod_address_lookup(unsigned long addr, unsigned long *size, unsigned long *off, char **modname, char *sym) { struct ftrace_mod_map *mod_map; - const char *ret = NULL; + int ret = 0; /* mod_map is freed via call_rcu() */ preempt_disable(); -- cgit v1.2.3 From c40583e19eebae50f6b024d8285d2703e2522364 Mon Sep 17 00:00:00 2001 From: "Luis Claudio R. Goncalves" Date: Mon, 10 Jun 2024 10:23:14 -0300 Subject: rtla/osnoise: set the default threshold to 1us Change the default threshold for osnoise to 1us, so that any noise equal or above this value is recorded. Let the user set a higher threshold if necessary. Link: https://lore.kernel.org/linux-trace-kernel/Zmb-QhiiiI6jM9To@uudg.org Cc: Masami Hiramatsu Cc: Mathieu Desnoyers Cc: Jonathan Corbet Suggested-by: Daniel Bristot de Oliveira Reviewed-by: Clark Williams Signed-off-by: Luis Claudio R. Goncalves Acked-by: Daniel Bristot de Oliveira Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace_osnoise.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_osnoise.c b/kernel/trace/trace_osnoise.c index a8e28f9b9271..66a871553d4a 100644 --- a/kernel/trace/trace_osnoise.c +++ b/kernel/trace/trace_osnoise.c @@ -1444,9 +1444,9 @@ static int run_osnoise(void) save_osn_sample_stats(osn_var, &s); /* - * if threshold is 0, use the default value of 5 us. + * if threshold is 0, use the default value of 1 us. */ - threshold = tracing_thresh ? : 5000; + threshold = tracing_thresh ? : 1000; /* * Apply PREEMPT and IRQ disabled options. -- cgit v1.2.3 From b576d375b536568c85d42c15a189f6b6fdd75b74 Mon Sep 17 00:00:00 2001 From: Jiapeng Chong Date: Tue, 18 Jun 2024 15:20:14 +0800 Subject: fgraph: Use str_plural() in test_graph_storage_single() Use existing str_plural() function rather than duplicating its implementation. ./kernel/trace/trace_selftest.c:880:56-60: opportunity for str_plural(size). Link: https://lore.kernel.org/linux-trace-kernel/20240618072014.20855-1-jiapeng.chong@linux.alibaba.com Reported-by: Abaci Robot Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=9349 Signed-off-by: Jiapeng Chong Acked-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace_selftest.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index adf0f436d84b..97f1e4bc47dc 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -877,7 +877,7 @@ static int __init test_graph_storage_single(struct fgraph_fixture *fixture) int ret; pr_cont("PASSED\n"); - pr_info("Testing fgraph storage of %d byte%s: ", size, size > 1 ? "s" : ""); + pr_info("Testing fgraph storage of %d byte%s: ", size, str_plural(size)); ret = init_fgraph_fixture(fixture); if (ret && ret != -ENODEV) { -- cgit v1.2.3 From c02525a33969000fa7b595b743deb4d79804916b Mon Sep 17 00:00:00 2001 From: Ilya Leoshkevich Date: Fri, 21 Jun 2024 13:34:45 +0200 Subject: ftrace: unpoison ftrace_regs in ftrace_ops_list_func() Patch series "kmsan: Enable on s390", v7. Architectures use assembly code to initialize ftrace_regs and call ftrace_ops_list_func(). Therefore, from the KMSAN's point of view, ftrace_regs is poisoned on ftrace_ops_list_func entry(). This causes KMSAN warnings when running the ftrace testsuite. Fix by trusting the architecture-specific assembly code and always unpoisoning ftrace_regs in ftrace_ops_list_func. The issue was not encountered on x86_64 so far only by accident: assembly-allocated ftrace_regs was overlapping a stale partially unpoisoned stack frame. Poisoning stack frames before returns [1] makes the issue appear on x86_64 as well. [1] https://github.com/iii-i/llvm-project/commits/msan-poison-allocas-before-returning-2024-06-12/ Link: https://lkml.kernel.org/r/20240621113706.315500-1-iii@linux.ibm.com Link: https://lkml.kernel.org/r/20240621113706.315500-2-iii@linux.ibm.com Signed-off-by: Ilya Leoshkevich Reviewed-by: Alexander Potapenko Acked-by: Steven Rostedt (Google) Cc: Alexander Gordeev Cc: Christian Borntraeger Cc: Christoph Lameter Cc: David Rientjes Cc: Dmitry Vyukov Cc: Heiko Carstens Cc: Hyeonggon Yoo <42.hyeyoo@gmail.com> Cc: Joonsoo Kim Cc: Cc: Marco Elver Cc: Mark Rutland Cc: Masami Hiramatsu (Google) Cc: Pekka Enberg Cc: Roman Gushchin Cc: Sven Schnelle Cc: Vasily Gorbik Cc: Vlastimil Babka Signed-off-by: Andrew Morton --- kernel/trace/ftrace.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index eacab4020508..f1150f081d6b 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -7404,6 +7404,7 @@ out: void arch_ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip, struct ftrace_ops *op, struct ftrace_regs *fregs) { + kmsan_unpoison_memory(fregs, sizeof(*fregs)); __ftrace_ops_list_func(ip, parent_ip, NULL, fregs); } #else -- cgit v1.2.3 From 9d8616034f161222a4ac166c1b42b6d79961c005 Mon Sep 17 00:00:00 2001 From: "Masami Hiramatsu (Google)" Date: Fri, 5 Jul 2024 16:11:25 +0900 Subject: tracing/kprobes: Add symbol counting check when module loads Currently, kprobe event checks whether the target symbol name is unique or not, so that it does not put a probe on an unexpected place. But this skips the check if the target is on a module because the module may not be loaded. To fix this issue, this patch checks the number of probe target symbols in a target module when the module is loaded. If the probe is not on the unique name symbols in the module, it will be rejected at that point. Note that the symbol which has a unique name in the target module, it will be accepted even if there are same-name symbols in the kernel or other modules, Link: https://lore.kernel.org/all/172016348553.99543.2834679315611882137.stgit@devnote2/ Signed-off-by: Masami Hiramatsu (Google) Reviewed-by: Steven Rostedt (Google) --- kernel/trace/trace_kprobe.c | 125 ++++++++++++++++++++++++++++---------------- 1 file changed, 81 insertions(+), 44 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 7fd0f8576e4c..4cee3442bcce 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -678,6 +678,21 @@ end: } #ifdef CONFIG_MODULES +static int validate_module_probe_symbol(const char *modname, const char *symbol); + +static int register_module_trace_kprobe(struct module *mod, struct trace_kprobe *tk) +{ + const char *p; + int ret = 0; + + p = strchr(trace_kprobe_symbol(tk), ':'); + if (p) + ret = validate_module_probe_symbol(module_name(mod), p + 1); + if (!ret) + ret = __register_trace_kprobe(tk); + return ret; +} + /* Module notifier call back, checking event on the module */ static int trace_kprobe_module_callback(struct notifier_block *nb, unsigned long val, void *data) @@ -696,7 +711,7 @@ static int trace_kprobe_module_callback(struct notifier_block *nb, if (trace_kprobe_within_module(tk, mod)) { /* Don't need to check busy - this should have gone. */ __unregister_trace_kprobe(tk); - ret = __register_trace_kprobe(tk); + ret = register_module_trace_kprobe(mod, tk); if (ret) pr_warn("Failed to re-register probe %s on %s: %d\n", trace_probe_name(&tk->tp), @@ -747,17 +762,68 @@ static int count_mod_symbols(void *data, const char *name, unsigned long unused) return 0; } -static unsigned int number_of_same_symbols(char *func_name) +static unsigned int number_of_same_symbols(const char *mod, const char *func_name) { struct sym_count_ctx ctx = { .count = 0, .name = func_name }; - kallsyms_on_each_match_symbol(count_symbols, func_name, &ctx.count); + if (!mod) + kallsyms_on_each_match_symbol(count_symbols, func_name, &ctx.count); - module_kallsyms_on_each_symbol(NULL, count_mod_symbols, &ctx); + module_kallsyms_on_each_symbol(mod, count_mod_symbols, &ctx); return ctx.count; } +static int validate_module_probe_symbol(const char *modname, const char *symbol) +{ + unsigned int count = number_of_same_symbols(modname, symbol); + + if (count > 1) { + /* + * Users should use ADDR to remove the ambiguity of + * using KSYM only. + */ + return -EADDRNOTAVAIL; + } else if (count == 0) { + /* + * We can return ENOENT earlier than when register the + * kprobe. + */ + return -ENOENT; + } + return 0; +} + +static int validate_probe_symbol(char *symbol) +{ + struct module *mod = NULL; + char *modname = NULL, *p; + int ret = 0; + + p = strchr(symbol, ':'); + if (p) { + modname = symbol; + symbol = p + 1; + *p = '\0'; + /* Return 0 (defer) if the module does not exist yet. */ + rcu_read_lock_sched(); + mod = find_module(modname); + if (mod && !try_module_get(mod)) + mod = NULL; + rcu_read_unlock_sched(); + if (!mod) + goto out; + } + + ret = validate_module_probe_symbol(modname, symbol); +out: + if (p) + *p = ':'; + if (mod) + module_put(mod); + return ret; +} + static int trace_kprobe_entry_handler(struct kretprobe_instance *ri, struct pt_regs *regs); @@ -881,6 +947,14 @@ static int __trace_kprobe_create(int argc, const char *argv[]) trace_probe_log_err(0, BAD_PROBE_ADDR); goto parse_error; } + ret = validate_probe_symbol(symbol); + if (ret) { + if (ret == -EADDRNOTAVAIL) + trace_probe_log_err(0, NON_UNIQ_SYMBOL); + else + trace_probe_log_err(0, BAD_PROBE_ADDR); + goto parse_error; + } if (is_return) ctx.flags |= TPARG_FL_RETURN; ret = kprobe_on_func_entry(NULL, symbol, offset); @@ -893,31 +967,6 @@ static int __trace_kprobe_create(int argc, const char *argv[]) } } - if (symbol && !strchr(symbol, ':')) { - unsigned int count; - - count = number_of_same_symbols(symbol); - if (count > 1) { - /* - * Users should use ADDR to remove the ambiguity of - * using KSYM only. - */ - trace_probe_log_err(0, NON_UNIQ_SYMBOL); - ret = -EADDRNOTAVAIL; - - goto error; - } else if (count == 0) { - /* - * We can return ENOENT earlier than when register the - * kprobe. - */ - trace_probe_log_err(0, BAD_PROBE_ADDR); - ret = -ENOENT; - - goto error; - } - } - trace_probe_log_set_index(0); if (event) { ret = traceprobe_parse_event_name(&event, &group, gbuf, @@ -1835,21 +1884,9 @@ create_local_trace_kprobe(char *func, void *addr, unsigned long offs, char *event; if (func) { - unsigned int count; - - count = number_of_same_symbols(func); - if (count > 1) - /* - * Users should use addr to remove the ambiguity of - * using func only. - */ - return ERR_PTR(-EADDRNOTAVAIL); - else if (count == 0) - /* - * We can return ENOENT earlier than when register the - * kprobe. - */ - return ERR_PTR(-ENOENT); + ret = validate_probe_symbol(func); + if (ret) + return ERR_PTR(ret); } /* -- cgit v1.2.3 From b10545b6b86b7a0b3e26b4c2a5c99b72d49bc4de Mon Sep 17 00:00:00 2001 From: "Masami Hiramatsu (Google)" Date: Wed, 10 Jul 2024 08:36:31 +0900 Subject: tracing/kprobes: Fix build error when find_module() is not available The kernel test robot reported that the find_module() is not available if CONFIG_MODULES=n. Fix this error by hiding find_modules() in #ifdef CONFIG_MODULES with related rcu locks as try_module_get_by_name(). Link: https://lore.kernel.org/all/172056819167.201571.250053007194508038.stgit@devnote2/ Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202407070744.RcLkn8sq-lkp@intel.com/ Closes: https://lore.kernel.org/oe-kbuild-all/202407070917.VVUCBlaS-lkp@intel.com/ Signed-off-by: Masami Hiramatsu (Google) --- kernel/trace/trace_kprobe.c | 25 +++++++++++++++++++------ 1 file changed, 19 insertions(+), 6 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 4cee3442bcce..61a6da808203 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -794,6 +794,24 @@ static int validate_module_probe_symbol(const char *modname, const char *symbol) return 0; } +#ifdef CONFIG_MODULES +/* Return NULL if the module is not loaded or under unloading. */ +static struct module *try_module_get_by_name(const char *name) +{ + struct module *mod; + + rcu_read_lock_sched(); + mod = find_module(name); + if (mod && !try_module_get(mod)) + mod = NULL; + rcu_read_unlock_sched(); + + return mod; +} +#else +#define try_module_get_by_name(name) (NULL) +#endif + static int validate_probe_symbol(char *symbol) { struct module *mod = NULL; @@ -805,12 +823,7 @@ static int validate_probe_symbol(char *symbol) modname = symbol; symbol = p + 1; *p = '\0'; - /* Return 0 (defer) if the module does not exist yet. */ - rcu_read_lock_sched(); - mod = find_module(modname); - if (mod && !try_module_get(mod)) - mod = NULL; - rcu_read_unlock_sched(); + mod = try_module_get_by_name(modname); if (!mod) goto out; } -- cgit v1.2.3 From 7dc836187f7c6f70a82b4521503e9f9f96194581 Mon Sep 17 00:00:00 2001 From: "levi.yun" Date: Thu, 4 Jul 2024 16:02:26 +0100 Subject: trace/pid_list: Change gfp flags in pid_list_fill_irq() pid_list_fill_irq() runs via irq_work. When CONFIG_PREEMPT_RT is disabled, it would run in irq_context. so it shouldn't sleep while memory allocation. Change gfp flags from GFP_KERNEL to GFP_NOWAIT to prevent sleep in irq_work. This change wouldn't impact functionality in practice because the worst-size is 2K. Cc: stable@goodmis.org Fixes: 8d6e90983ade2 ("tracing: Create a sparse bitmask for pid filtering") Link: https://lore.kernel.org/20240704150226.1359936-1-yeoreum.yun@arm.com Acked-by: Masami Hiramatsu (Google) Signed-off-by: levi.yun Signed-off-by: Steven Rostedt (Google) --- kernel/trace/pid_list.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/pid_list.c b/kernel/trace/pid_list.c index 19b271a12c99..4966e6bbdf6f 100644 --- a/kernel/trace/pid_list.c +++ b/kernel/trace/pid_list.c @@ -354,7 +354,7 @@ static void pid_list_refill_irq(struct irq_work *iwork) while (upper_count-- > 0) { union upper_chunk *chunk; - chunk = kzalloc(sizeof(*chunk), GFP_KERNEL); + chunk = kzalloc(sizeof(*chunk), GFP_NOWAIT); if (!chunk) break; *upper_next = chunk; @@ -365,7 +365,7 @@ static void pid_list_refill_irq(struct irq_work *iwork) while (lower_count-- > 0) { union lower_chunk *chunk; - chunk = kzalloc(sizeof(*chunk), GFP_KERNEL); + chunk = kzalloc(sizeof(*chunk), GFP_NOWAIT); if (!chunk) break; *lower_next = chunk; -- cgit v1.2.3 From 78eb4ea25cd5fdbdae7eb9fdf87b99195ff67508 Mon Sep 17 00:00:00 2001 From: Joel Granados Date: Wed, 24 Jul 2024 20:59:29 +0200 Subject: sysctl: treewide: constify the ctl_table argument of proc_handlers MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit const qualify the struct ctl_table argument in the proc_handler function signatures. This is a prerequisite to moving the static ctl_table structs into .rodata data which will ensure that proc_handler function pointers cannot be modified. This patch has been generated by the following coccinelle script: ``` virtual patch @r1@ identifier ctl, write, buffer, lenp, ppos; identifier func !~ "appldata_(timer|interval)_handler|sched_(rt|rr)_handler|rds_tcp_skbuf_handler|proc_sctp_do_(hmac_alg|rto_min|rto_max|udp_port|alpha_beta|auth|probe_interval)"; @@ int func( - struct ctl_table *ctl + const struct ctl_table *ctl ,int write, void *buffer, size_t *lenp, loff_t *ppos); @r2@ identifier func, ctl, write, buffer, lenp, ppos; @@ int func( - struct ctl_table *ctl + const struct ctl_table *ctl ,int write, void *buffer, size_t *lenp, loff_t *ppos) { ... } @r3@ identifier func; @@ int func( - struct ctl_table * + const struct ctl_table * ,int , void *, size_t *, loff_t *); @r4@ identifier func, ctl; @@ int func( - struct ctl_table *ctl + const struct ctl_table *ctl ,int , void *, size_t *, loff_t *); @r5@ identifier func, write, buffer, lenp, ppos; @@ int func( - struct ctl_table * + const struct ctl_table * ,int write, void *buffer, size_t *lenp, loff_t *ppos); ``` * Code formatting was adjusted in xfs_sysctl.c to comply with code conventions. The xfs_stats_clear_proc_handler, xfs_panic_mask_proc_handler and xfs_deprecated_dointvec_minmax where adjusted. * The ctl_table argument in proc_watchdog_common was const qualified. This is called from a proc_handler itself and is calling back into another proc_handler, making it necessary to change it as part of the proc_handler migration. Co-developed-by: Thomas Weißschuh Signed-off-by: Thomas Weißschuh Co-developed-by: Joel Granados Signed-off-by: Joel Granados --- kernel/trace/ftrace.c | 2 +- kernel/trace/trace.c | 2 +- kernel/trace/trace_events_user.c | 2 +- kernel/trace/trace_stack.c | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 0f579430f02a..4c28dd177ca6 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -8735,7 +8735,7 @@ static bool is_permanent_ops_registered(void) } static int -ftrace_enable_sysctl(struct ctl_table *table, int write, +ftrace_enable_sysctl(const struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos) { int ret = -ENODEV; diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 578a49ff5c32..10cd38bce2f1 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2767,7 +2767,7 @@ static void output_printk(struct trace_event_buffer *fbuffer) raw_spin_unlock_irqrestore(&tracepoint_iter_lock, flags); } -int tracepoint_printk_sysctl(struct ctl_table *table, int write, +int tracepoint_printk_sysctl(const struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos) { diff --git a/kernel/trace/trace_events_user.c b/kernel/trace/trace_events_user.c index 3a2b46847c8b..42b0d998d103 100644 --- a/kernel/trace/trace_events_user.c +++ b/kernel/trace/trace_events_user.c @@ -2885,7 +2885,7 @@ err: return -ENODEV; } -static int set_max_user_events_sysctl(struct ctl_table *table, int write, +static int set_max_user_events_sysctl(const struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos) { int ret; diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index 5a48dba912ea..7f9572a37333 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -514,7 +514,7 @@ static const struct file_operations stack_trace_filter_fops = { #endif /* CONFIG_DYNAMIC_FTRACE */ int -stack_trace_sysctl(struct ctl_table *table, int write, void *buffer, +stack_trace_sysctl(const struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos) { int was_enabled; -- cgit v1.2.3 From 1a251f52cfdc417c84411a056bc142cbd77baef4 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 28 Jul 2024 15:49:18 -0700 Subject: minmax: make generic MIN() and MAX() macros available everywhere This just standardizes the use of MIN() and MAX() macros, with the very traditional semantics. The goal is to use these for C constant expressions and for top-level / static initializers, and so be able to simplify the min()/max() macros. These macro names were used by various kernel code - they are very traditional, after all - and all such users have been fixed up, with a few different approaches: - trivial duplicated macro definitions have been removed Note that 'trivial' here means that it's obviously kernel code that already included all the major kernel headers, and thus gets the new generic MIN/MAX macros automatically. - non-trivial duplicated macro definitions are guarded with #ifndef This is the "yes, they define their own versions, but no, the include situation is not entirely obvious, and maybe they don't get the generic version automatically" case. - strange use case #1 A couple of drivers decided that the way they want to describe their versioning is with #define MAJ 1 #define MIN 2 #define DRV_VERSION __stringify(MAJ) "." __stringify(MIN) which adds zero value and I just did my Alexander the Great impersonation, and rewrote that pointless Gordian knot as #define DRV_VERSION "1.2" instead. - strange use case #2 A couple of drivers thought that it's a good idea to have a random 'MIN' or 'MAX' define for a value or index into a table, rather than the traditional macro that takes arguments. These values were re-written as C enum's instead. The new function-line macros only expand when followed by an open parenthesis, and thus don't clash with enum use. Happily, there weren't really all that many of these cases, and a lot of users already had the pattern of using '#ifndef' guarding (or in one case just using '#undef MIN') before defining their own private version that does the same thing. I left such cases alone. Cc: David Laight Cc: Lorenzo Stoakes Signed-off-by: Linus Torvalds --- kernel/trace/preemptirq_delay_test.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/preemptirq_delay_test.c b/kernel/trace/preemptirq_delay_test.c index cb0871fbdb07..314ffc143039 100644 --- a/kernel/trace/preemptirq_delay_test.c +++ b/kernel/trace/preemptirq_delay_test.c @@ -34,8 +34,6 @@ MODULE_PARM_DESC(cpu_affinity, "Cpu num test is running on"); static struct completion done; -#define MIN(x, y) ((x) < (y) ? (x) : (y)) - static void busy_wait(ulong time) { u64 start, end; -- cgit v1.2.3 From b1560408692cd0ab0370cfbe9deb03ce97ab3f6d Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Tue, 30 Jul 2024 11:06:57 -0400 Subject: tracing: Have format file honor EVENT_FILE_FL_FREED MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When eventfs was introduced, special care had to be done to coordinate the freeing of the file meta data with the files that are exposed to user space. The file meta data would have a ref count that is set when the file is created and would be decremented and freed after the last user that opened the file closed it. When the file meta data was to be freed, it would set a flag (EVENT_FILE_FL_FREED) to denote that the file is freed, and any new references made (like new opens or reads) would fail as it is marked freed. This allowed other meta data to be freed after this flag was set (under the event_mutex). All the files that were dynamically created in the events directory had a pointer to the file meta data and would call event_release() when the last reference to the user space file was closed. This would be the time that it is safe to free the file meta data. A shortcut was made for the "format" file. It's i_private would point to the "call" entry directly and not point to the file's meta data. This is because all format files are the same for the same "call", so it was thought there was no reason to differentiate them. The other files maintain state (like the "enable", "trigger", etc). But this meant if the file were to disappear, the "format" file would be unaware of it. This caused a race that could be trigger via the user_events test (that would create dynamic events and free them), and running a loop that would read the user_events format files: In one console run: # cd tools/testing/selftests/user_events # while true; do ./ftrace_test; done And in another console run: # cd /sys/kernel/tracing/ # while true; do cat events/user_events/__test_event/format; done 2>/dev/null With KASAN memory checking, it would trigger a use-after-free bug report (which was a real bug). This was because the format file was not checking the file's meta data flag "EVENT_FILE_FL_FREED", so it would access the event that the file meta data pointed to after the event was freed. After inspection, there are other locations that were found to not check the EVENT_FILE_FL_FREED flag when accessing the trace_event_file. Add a new helper function: event_file_file() that will make sure that the event_mutex is held, and will return NULL if the trace_event_file has the EVENT_FILE_FL_FREED flag set. Have the first reference of the struct file pointer use event_file_file() and check for NULL. Later uses can still use the event_file_data() helper function if the event_mutex is still held and was not released since the event_file_file() call. Link: https://lore.kernel.org/all/20240719204701.1605950-1-minipli@grsecurity.net/ Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mathieu Desnoyers Cc: Ajay Kaher Cc: Ilkka Naulapää Cc: Linus Torvalds Cc: Al Viro Cc: Dan Carpenter Cc: Beau Belgrave Cc: Florian Fainelli Cc: Alexey Makhalov Cc: Vasavi Sirnapalli Link: https://lore.kernel.org/20240730110657.3b69d3c1@gandalf.local.home Fixes: b63db58e2fa5d ("eventfs/tracing: Add callback for release of an eventfs_inode") Reported-by: Mathias Krause Tested-by: Mathias Krause Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.h | 23 +++++++++++++++++++++++ kernel/trace/trace_events.c | 33 ++++++++++++++++++++------------- kernel/trace/trace_events_hist.c | 4 ++-- kernel/trace/trace_events_inject.c | 2 +- kernel/trace/trace_events_trigger.c | 6 +++--- 5 files changed, 49 insertions(+), 19 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 8783bebd0562..bd3e3069300e 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -1634,6 +1634,29 @@ static inline void *event_file_data(struct file *filp) extern struct mutex event_mutex; extern struct list_head ftrace_events; +/* + * When the trace_event_file is the filp->i_private pointer, + * it must be taken under the event_mutex lock, and then checked + * if the EVENT_FILE_FL_FREED flag is set. If it is, then the + * data pointed to by the trace_event_file can not be trusted. + * + * Use the event_file_file() to access the trace_event_file from + * the filp the first time under the event_mutex and check for + * NULL. If it is needed to be retrieved again and the event_mutex + * is still held, then the event_file_data() can be used and it + * is guaranteed to be valid. + */ +static inline struct trace_event_file *event_file_file(struct file *filp) +{ + struct trace_event_file *file; + + lockdep_assert_held(&event_mutex); + file = READ_ONCE(file_inode(filp)->i_private); + if (!file || file->flags & EVENT_FILE_FL_FREED) + return NULL; + return file; +} + extern const struct file_operations event_trigger_fops; extern const struct file_operations event_hist_fops; extern const struct file_operations event_hist_debug_fops; diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 6ef29eba90ce..f08fbaf8cad6 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -1386,12 +1386,12 @@ event_enable_read(struct file *filp, char __user *ubuf, size_t cnt, char buf[4] = "0"; mutex_lock(&event_mutex); - file = event_file_data(filp); + file = event_file_file(filp); if (likely(file)) flags = file->flags; mutex_unlock(&event_mutex); - if (!file || flags & EVENT_FILE_FL_FREED) + if (!file) return -ENODEV; if (flags & EVENT_FILE_FL_ENABLED && @@ -1424,8 +1424,8 @@ event_enable_write(struct file *filp, const char __user *ubuf, size_t cnt, case 1: ret = -ENODEV; mutex_lock(&event_mutex); - file = event_file_data(filp); - if (likely(file && !(file->flags & EVENT_FILE_FL_FREED))) { + file = event_file_file(filp); + if (likely(file)) { ret = tracing_update_buffers(file->tr); if (ret < 0) { mutex_unlock(&event_mutex); @@ -1540,7 +1540,8 @@ enum { static void *f_next(struct seq_file *m, void *v, loff_t *pos) { - struct trace_event_call *call = event_file_data(m->private); + struct trace_event_file *file = event_file_data(m->private); + struct trace_event_call *call = file->event_call; struct list_head *common_head = &ftrace_common_fields; struct list_head *head = trace_get_fields(call); struct list_head *node = v; @@ -1572,7 +1573,8 @@ static void *f_next(struct seq_file *m, void *v, loff_t *pos) static int f_show(struct seq_file *m, void *v) { - struct trace_event_call *call = event_file_data(m->private); + struct trace_event_file *file = event_file_data(m->private); + struct trace_event_call *call = file->event_call; struct ftrace_event_field *field; const char *array_descriptor; @@ -1627,12 +1629,14 @@ static int f_show(struct seq_file *m, void *v) static void *f_start(struct seq_file *m, loff_t *pos) { + struct trace_event_file *file; void *p = (void *)FORMAT_HEADER; loff_t l = 0; /* ->stop() is called even if ->start() fails */ mutex_lock(&event_mutex); - if (!event_file_data(m->private)) + file = event_file_file(m->private); + if (!file) return ERR_PTR(-ENODEV); while (l < *pos && p) @@ -1706,8 +1710,8 @@ event_filter_read(struct file *filp, char __user *ubuf, size_t cnt, trace_seq_init(s); mutex_lock(&event_mutex); - file = event_file_data(filp); - if (file && !(file->flags & EVENT_FILE_FL_FREED)) + file = event_file_file(filp); + if (file) print_event_filter(file, s); mutex_unlock(&event_mutex); @@ -1736,9 +1740,13 @@ event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, return PTR_ERR(buf); mutex_lock(&event_mutex); - file = event_file_data(filp); - if (file) - err = apply_event_filter(file, buf); + file = event_file_file(filp); + if (file) { + if (file->flags & EVENT_FILE_FL_FREED) + err = -ENODEV; + else + err = apply_event_filter(file, buf); + } mutex_unlock(&event_mutex); kfree(buf); @@ -2485,7 +2493,6 @@ static int event_callback(const char *name, umode_t *mode, void **data, if (strcmp(name, "format") == 0) { *mode = TRACE_MODE_READ; *fops = &ftrace_event_format_fops; - *data = call; return 1; } diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c index 6ece1308d36a..5f9119eb7c67 100644 --- a/kernel/trace/trace_events_hist.c +++ b/kernel/trace/trace_events_hist.c @@ -5601,7 +5601,7 @@ static int hist_show(struct seq_file *m, void *v) mutex_lock(&event_mutex); - event_file = event_file_data(m->private); + event_file = event_file_file(m->private); if (unlikely(!event_file)) { ret = -ENODEV; goto out_unlock; @@ -5880,7 +5880,7 @@ static int hist_debug_show(struct seq_file *m, void *v) mutex_lock(&event_mutex); - event_file = event_file_data(m->private); + event_file = event_file_file(m->private); if (unlikely(!event_file)) { ret = -ENODEV; goto out_unlock; diff --git a/kernel/trace/trace_events_inject.c b/kernel/trace/trace_events_inject.c index 8650562bdaa9..a8f076809db4 100644 --- a/kernel/trace/trace_events_inject.c +++ b/kernel/trace/trace_events_inject.c @@ -299,7 +299,7 @@ event_inject_write(struct file *filp, const char __user *ubuf, size_t cnt, strim(buf); mutex_lock(&event_mutex); - file = event_file_data(filp); + file = event_file_file(filp); if (file) { call = file->event_call; size = parse_entry(buf, call, &entry); diff --git a/kernel/trace/trace_events_trigger.c b/kernel/trace/trace_events_trigger.c index 4bec043c8690..a5e3d6acf1e1 100644 --- a/kernel/trace/trace_events_trigger.c +++ b/kernel/trace/trace_events_trigger.c @@ -159,7 +159,7 @@ static void *trigger_start(struct seq_file *m, loff_t *pos) /* ->stop() is called even if ->start() fails */ mutex_lock(&event_mutex); - event_file = event_file_data(m->private); + event_file = event_file_file(m->private); if (unlikely(!event_file)) return ERR_PTR(-ENODEV); @@ -213,7 +213,7 @@ static int event_trigger_regex_open(struct inode *inode, struct file *file) mutex_lock(&event_mutex); - if (unlikely(!event_file_data(file))) { + if (unlikely(!event_file_file(file))) { mutex_unlock(&event_mutex); return -ENODEV; } @@ -293,7 +293,7 @@ static ssize_t event_trigger_regex_write(struct file *file, strim(buf); mutex_lock(&event_mutex); - event_file = event_file_data(file); + event_file = event_file_file(file); if (unlikely(!event_file)) { mutex_unlock(&event_mutex); kfree(buf); -- cgit v1.2.3 From 6e2fdceffdc6bd7b8ba314a1d1b976721533c8f9 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 26 Jul 2024 14:42:08 -0400 Subject: tracing: Use refcount for trace_event_file reference counter Instead of using an atomic counter for the trace_event_file reference counter, use the refcount interface. It has various checks to make sure the reference counting is correct, and will warn if it detects an error (like refcount_inc() on '0'). Cc: Mathieu Desnoyers Link: https://lore.kernel.org/20240726144208.687cce24@rorschach.local.home Acked-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace_events.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index f08fbaf8cad6..7266ec2a4eea 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -992,18 +992,18 @@ static void remove_subsystem(struct trace_subsystem_dir *dir) void event_file_get(struct trace_event_file *file) { - atomic_inc(&file->ref); + refcount_inc(&file->ref); } void event_file_put(struct trace_event_file *file) { - if (WARN_ON_ONCE(!atomic_read(&file->ref))) { + if (WARN_ON_ONCE(!refcount_read(&file->ref))) { if (file->flags & EVENT_FILE_FL_FREED) kmem_cache_free(file_cachep, file); return; } - if (atomic_dec_and_test(&file->ref)) { + if (refcount_dec_and_test(&file->ref)) { /* Count should only go to zero when it is freed */ if (WARN_ON_ONCE(!(file->flags & EVENT_FILE_FL_FREED))) return; @@ -3003,7 +3003,7 @@ trace_create_new_event(struct trace_event_call *call, atomic_set(&file->tm_ref, 0); INIT_LIST_HEAD(&file->triggers); list_add(&file->list, &tr->events); - event_file_get(file); + refcount_set(&file->ref, 1); return file; } -- cgit v1.2.3 From 604b72b32522d548f855ed82842d2e49bf384edb Mon Sep 17 00:00:00 2001 From: Petr Pavlu Date: Sat, 3 Aug 2024 15:09:26 +0200 Subject: function_graph: Fix the ret_stack used by ftrace_graph_ret_addr() When ftrace_graph_ret_addr() is invoked to convert a found stack return address to its original value, the function can end up producing the following crash: [ 95.442712] BUG: kernel NULL pointer dereference, address: 0000000000000028 [ 95.442720] #PF: supervisor read access in kernel mode [ 95.442724] #PF: error_code(0x0000) - not-present page [ 95.442727] PGD 0 P4D 0- [ 95.442731] Oops: Oops: 0000 [#1] PREEMPT SMP PTI [ 95.442736] CPU: 1 UID: 0 PID: 2214 Comm: insmod Kdump: loaded Tainted: G OE K 6.11.0-rc1-default #1 67c62a3b3720562f7e7db5f11c1fdb40b7a2857c [ 95.442747] Tainted: [O]=OOT_MODULE, [E]=UNSIGNED_MODULE, [K]=LIVEPATCH [ 95.442750] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.16.2-3-gd478f380-rebuilt.opensuse.org 04/01/2014 [ 95.442754] RIP: 0010:ftrace_graph_ret_addr+0x42/0xc0 [ 95.442766] Code: [...] [ 95.442773] RSP: 0018:ffff979b80ff7718 EFLAGS: 00010006 [ 95.442776] RAX: ffffffff8ca99b10 RBX: ffff979b80ff7760 RCX: ffff979b80167dc0 [ 95.442780] RDX: ffffffff8ca99b10 RSI: ffff979b80ff7790 RDI: 0000000000000005 [ 95.442783] RBP: 0000000000000001 R08: 0000000000000005 R09: 0000000000000000 [ 95.442786] R10: 0000000000000005 R11: 0000000000000000 R12: ffffffff8e9491e0 [ 95.442790] R13: ffffffff8d6f70f0 R14: ffff979b80167da8 R15: ffff979b80167dc8 [ 95.442793] FS: 00007fbf83895740(0000) GS:ffff8a0afdd00000(0000) knlGS:0000000000000000 [ 95.442797] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 95.442800] CR2: 0000000000000028 CR3: 0000000005070002 CR4: 0000000000370ef0 [ 95.442806] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 95.442809] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 95.442816] Call Trace: [ 95.442823] [ 95.442896] unwind_next_frame+0x20d/0x830 [ 95.442905] arch_stack_walk_reliable+0x94/0xe0 [ 95.442917] stack_trace_save_tsk_reliable+0x7d/0xe0 [ 95.442922] klp_check_and_switch_task+0x55/0x1a0 [ 95.442931] task_call_func+0xd3/0xe0 [ 95.442938] klp_try_switch_task.part.5+0x37/0x150 [ 95.442942] klp_try_complete_transition+0x79/0x2d0 [ 95.442947] klp_enable_patch+0x4db/0x890 [ 95.442960] do_one_initcall+0x41/0x2e0 [ 95.442968] do_init_module+0x60/0x220 [ 95.442975] load_module+0x1ebf/0x1fb0 [ 95.443004] init_module_from_file+0x88/0xc0 [ 95.443010] idempotent_init_module+0x190/0x240 [ 95.443015] __x64_sys_finit_module+0x5b/0xc0 [ 95.443019] do_syscall_64+0x74/0x160 [ 95.443232] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ 95.443236] RIP: 0033:0x7fbf82f2c709 [ 95.443241] Code: [...] [ 95.443247] RSP: 002b:00007fffd5ea3b88 EFLAGS: 00000246 ORIG_RAX: 0000000000000139 [ 95.443253] RAX: ffffffffffffffda RBX: 000056359c48e750 RCX: 00007fbf82f2c709 [ 95.443257] RDX: 0000000000000000 RSI: 000056356ed4efc5 RDI: 0000000000000003 [ 95.443260] RBP: 000056356ed4efc5 R08: 0000000000000000 R09: 00007fffd5ea3c10 [ 95.443263] R10: 0000000000000003 R11: 0000000000000246 R12: 0000000000000000 [ 95.443267] R13: 000056359c48e6f0 R14: 0000000000000000 R15: 0000000000000000 [ 95.443272] [ 95.443274] Modules linked in: [...] [ 95.443385] Unloaded tainted modules: intel_uncore_frequency(E):1 isst_if_common(E):1 skx_edac(E):1 [ 95.443414] CR2: 0000000000000028 The bug can be reproduced with kselftests: cd linux/tools/testing/selftests make TARGETS='ftrace livepatch' (cd ftrace; ./ftracetest test.d/ftrace/fgraph-filter.tc) (cd livepatch; ./test-livepatch.sh) The problem is that ftrace_graph_ret_addr() is supposed to operate on the ret_stack of a selected task but wrongly accesses the ret_stack of the current task. Specifically, the above NULL dereference occurs when task->curr_ret_stack is non-zero, but current->ret_stack is NULL. Correct ftrace_graph_ret_addr() to work with the right ret_stack. Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Reported-by: Miroslav Benes Link: https://lore.kernel.org/20240803131211.17255-1-petr.pavlu@suse.com Fixes: 7aa1eaef9f42 ("function_graph: Allow multiple users to attach to function graph") Signed-off-by: Petr Pavlu Signed-off-by: Steven Rostedt (Google) --- kernel/trace/fgraph.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index fc205ad167a9..d1d5ea2d0a1b 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -902,7 +902,7 @@ unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, i = *idx ? : task->curr_ret_stack; while (i > 0) { - ret_stack = get_ret_stack(current, i, &i); + ret_stack = get_ret_stack(task, i, &i); if (!ret_stack) break; /* -- cgit v1.2.3 From bcf86c01ca4676316557dd482c8416ece8c2e143 Mon Sep 17 00:00:00 2001 From: Tze-nan Wu Date: Mon, 5 Aug 2024 13:59:22 +0800 Subject: tracing: Fix overflow in get_free_elt() "tracing_map->next_elt" in get_free_elt() is at risk of overflowing. Once it overflows, new elements can still be inserted into the tracing_map even though the maximum number of elements (`max_elts`) has been reached. Continuing to insert elements after the overflow could result in the tracing_map containing "tracing_map->max_size" elements, leaving no empty entries. If any attempt is made to insert an element into a full tracing_map using `__tracing_map_insert()`, it will cause an infinite loop with preemption disabled, leading to a CPU hang problem. Fix this by preventing any further increments to "tracing_map->next_elt" once it reaches "tracing_map->max_elt". Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Fixes: 08d43a5fa063e ("tracing: Add lock-free tracing_map") Co-developed-by: Cheng-Jui Wang Link: https://lore.kernel.org/20240805055922.6277-1-Tze-nan.Wu@mediatek.com Signed-off-by: Cheng-Jui Wang Signed-off-by: Tze-nan Wu Signed-off-by: Steven Rostedt (Google) --- kernel/trace/tracing_map.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/tracing_map.c b/kernel/trace/tracing_map.c index a4dcf0f24352..3a56e7c8aa4f 100644 --- a/kernel/trace/tracing_map.c +++ b/kernel/trace/tracing_map.c @@ -454,7 +454,7 @@ static struct tracing_map_elt *get_free_elt(struct tracing_map *map) struct tracing_map_elt *elt = NULL; int idx; - idx = atomic_inc_return(&map->next_elt); + idx = atomic_fetch_add_unless(&map->next_elt, 1, map->max_elts); if (idx < map->max_elts) { elt = *(TRACING_MAP_ELT(map->elts, idx)); if (map->ops && map->ops->elt_init) @@ -699,7 +699,7 @@ void tracing_map_clear(struct tracing_map *map) { unsigned int i; - atomic_set(&map->next_elt, -1); + atomic_set(&map->next_elt, 0); atomic64_set(&map->hits, 0); atomic64_set(&map->drops, 0); @@ -783,7 +783,7 @@ struct tracing_map *tracing_map_create(unsigned int map_bits, map->map_bits = map_bits; map->max_elts = (1 << map_bits); - atomic_set(&map->next_elt, -1); + atomic_set(&map->next_elt, 0); map->map_size = (1 << (map_bits + 1)); map->ops = ops; -- cgit v1.2.3 From 58f7e4d7ba32758b861807e77535853cacc1f426 Mon Sep 17 00:00:00 2001 From: Jianhui Zhou <912460177@qq.com> Date: Mon, 5 Aug 2024 19:36:31 +0800 Subject: ring-buffer: Remove unused function ring_buffer_nr_pages() Because ring_buffer_nr_pages() is not an inline function and user accesses buffer->buffers[cpu]->nr_pages directly, the function ring_buffer_nr_pages is removed. Signed-off-by: Jianhui Zhou <912460177@qq.com> Link: https://lore.kernel.org/tencent_F4A7E9AB337F44E0F4B858D07D19EF460708@qq.com Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ring_buffer.c | 12 ------------ 1 file changed, 12 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 28853966aa9a..cebd879a30cb 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -692,18 +692,6 @@ u64 ring_buffer_event_time_stamp(struct trace_buffer *buffer, return ts; } -/** - * ring_buffer_nr_pages - get the number of buffer pages in the ring buffer - * @buffer: The ring_buffer to get the number of pages from - * @cpu: The cpu of the ring_buffer to get the number of pages from - * - * Returns the number of pages used by a per_cpu buffer of the ring buffer. - */ -size_t ring_buffer_nr_pages(struct trace_buffer *buffer, int cpu) -{ - return buffer->buffers[cpu]->nr_pages; -} - /** * ring_buffer_nr_dirty_pages - get the number of used pages in the ring buffer * @buffer: The ring_buffer to get the number of pages from -- cgit v1.2.3