From 575570f02761bd680ba5731c1dfd4701062e7fb2 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Tue, 27 Jul 2010 16:06:34 +0800 Subject: tracing: Fix an unallocated memory access in function_graph With CONFIG_DEBUG_PAGEALLOC, I observed an unallocated memory access in function_graph trace. It appears we find a small size entry in ring buffer, but we access it as a big size entry. The access overflows the page size and touches an unallocated page. Cc: Signed-off-by: Shaohua Li LKML-Reference: <1280217994.32400.76.camel@sli10-desk.sh.intel.com> [ Added a comment to explain the problem - SDR ] Signed-off-by: Steven Rostedt --- kernel/trace/trace_functions_graph.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index 79f4bac99a94..b4c179ae4e45 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -507,7 +507,15 @@ get_return_for_leaf(struct trace_iterator *iter, * if the output fails. */ data->ent = *curr; - data->ret = *next; + /* + * If the next event is not a return type, then + * we only care about what type it is. Otherwise we can + * safely copy the entire event. + */ + if (next->ent.type == TRACE_GRAPH_RET) + data->ret = *next; + else + data->ret.ent.type = next->ent.type; } } -- cgit v1.2.3 From 18fab912d4fa70133df164d2dcf3310be0c38c34 Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 28 Jul 2010 14:14:01 +0800 Subject: tracing: Fix ring_buffer_read_page reading out of page boundary With the configuration: CONFIG_DEBUG_PAGEALLOC=y and Shaohua's patch: [PATCH]x86: make spurious_fault check correct pte bit Function call graph trace with the following will trigger a page fault. # cd /sys/kernel/debug/tracing/ # echo function_graph > current_tracer # cat per_cpu/cpu1/trace_pipe_raw > /dev/null BUG: unable to handle kernel paging request at ffff880006e99000 IP: [] rb_event_length+0x1/0x3f PGD 1b19063 PUD 1b1d063 PMD 3f067 PTE 6e99160 Oops: 0000 [#1] SMP DEBUG_PAGEALLOC last sysfs file: /sys/devices/virtual/net/lo/operstate CPU 1 Modules linked in: Pid: 1982, comm: cat Not tainted 2.6.35-rc6-aes+ #300 /Bochs RIP: 0010:[] [] rb_event_length+0x1/0x3f RSP: 0018:ffff880006475e38 EFLAGS: 00010006 RAX: 0000000000000ff0 RBX: ffff88000786c630 RCX: 000000000000001d RDX: ffff880006e98000 RSI: 0000000000000ff0 RDI: ffff880006e99000 RBP: ffff880006475eb8 R08: 000000145d7008bd R09: 0000000000000000 R10: 0000000000008000 R11: ffffffff815d9336 R12: ffff880006d08000 R13: ffff880006e605d8 R14: 0000000000000000 R15: 0000000000000018 FS: 00007f2b83e456f0(0000) GS:ffff880002100000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: ffff880006e99000 CR3: 00000000064a8000 CR4: 00000000000006e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process cat (pid: 1982, threadinfo ffff880006474000, task ffff880006e40770) Stack: ffff880006475eb8 ffffffff8108730f 0000000000000ff0 000000145d7008bd <0> ffff880006e98010 ffff880006d08010 0000000000000296 ffff88000786c640 <0> ffffffff81002956 0000000000000000 ffff8800071f4680 ffff8800071f4680 Call Trace: [] ? ring_buffer_read_page+0x15a/0x24a [] ? return_to_handler+0x15/0x2f [] tracing_buffers_read+0xb9/0x164 [] vfs_read+0xaf/0x150 [] return_to_handler+0x0/0x2f [] __bad_area_nosemaphore+0x17e/0x1a1 [] return_to_handler+0x0/0x2f [] bad_area_nosemaphore+0x13/0x15 Code: 80 25 b2 16 b3 00 fe c9 c3 55 48 89 e5 f0 80 0d a4 16 b3 00 02 c9 c3 55 31 c0 48 89 e5 48 83 3d 94 16 b3 00 01 c9 0f 94 c0 c3 55 <8a> 0f 48 89 e5 83 e1 1f b8 08 00 00 00 0f b6 d1 83 fa 1e 74 27 RIP [] rb_event_length+0x1/0x3f RSP CR2: ffff880006e99000 ---[ end trace a6877bb92ccb36bb ]--- The root cause is that ring_buffer_read_page() may read out of page boundary, because the boundary checking is done after reading. This is fixed via doing boundary checking before reading. Reported-by: Shaohua Li Cc: Signed-off-by: Huang Ying LKML-Reference: <1280297641.2771.307.camel@yhuang-dev> Signed-off-by: Steven Rostedt --- kernel/trace/ring_buffer.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'kernel/trace') diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 1da7b6ea8b85..5ec8f1d1480e 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -3868,6 +3868,9 @@ int ring_buffer_read_page(struct ring_buffer *buffer, rpos = reader->read; pos += size; + if (rpos >= commit) + break; + event = rb_reader_event(cpu_buffer); size = rb_event_length(event); } while (len > size); -- cgit v1.2.3 From 2a37a3df57c44e947271758a1aa4bea7bff9feab Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Thu, 3 Jun 2010 15:21:34 -0400 Subject: tracing/events: Convert format output to seq_file Two new events were added that broke the current format output. Both from the SCSI system: scsi_dispatch_cmd_done and scsi_dispatch_cmd_timeout The reason is that their print_fmt exceeded a page size. Since the output of the format used simple_read_from_buffer and trace_seq, it was limited to a page size in output. This patch converts the printing of the format of an event into seq_file, which allows greater than a page size to be shown. I diffed all event formats comparing the output with and without this patch. All matched except for the above two, which showed just: FORMAT TOO BIG without this patch, but now properly displays the output with this patch. v2: Remove updating *pos in seq start function. [ Thanks to Li Zefan for pointing that out ] Reviewed-by: Li Zefan Cc: Martin K. Petersen Cc: Kei Tokunaga Cc: James Bottomley Cc: Tomohiro Kusumi Cc: Xiao Guangrong Signed-off-by: Steven Rostedt --- kernel/trace/trace_events.c | 208 ++++++++++++++++++++++++++++++-------------- 1 file changed, 141 insertions(+), 67 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 53cffc0b0801..45a8968707aa 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -29,6 +29,8 @@ DEFINE_MUTEX(event_mutex); LIST_HEAD(ftrace_events); +#define COMMON_FIELD_COUNT 5 + struct list_head * trace_get_fields(struct ftrace_event_call *event_call) { @@ -544,85 +546,155 @@ out: return ret; } -static ssize_t -event_format_read(struct file *filp, char __user *ubuf, size_t cnt, - loff_t *ppos) +enum { + FORMAT_HEADER = 1, + FORMAT_PRINTFMT = 2, +}; + +static void *f_next(struct seq_file *m, void *v, loff_t *pos) { - struct ftrace_event_call *call = filp->private_data; + struct ftrace_event_call *call = m->private; struct ftrace_event_field *field; struct list_head *head; - struct trace_seq *s; - int common_field_count = 5; - char *buf; - int r = 0; - - if (*ppos) - return 0; + loff_t index = *pos; - s = kmalloc(sizeof(*s), GFP_KERNEL); - if (!s) - return -ENOMEM; + (*pos)++; - trace_seq_init(s); + head = trace_get_fields(call); - trace_seq_printf(s, "name: %s\n", call->name); - trace_seq_printf(s, "ID: %d\n", call->event.type); - trace_seq_printf(s, "format:\n"); + switch ((unsigned long)v) { + case FORMAT_HEADER: - head = trace_get_fields(call); - list_for_each_entry_reverse(field, head, link) { - /* - * Smartly shows the array type(except dynamic array). - * Normal: - * field:TYPE VAR - * If TYPE := TYPE[LEN], it is shown: - * field:TYPE VAR[LEN] - */ - const char *array_descriptor = strchr(field->type, '['); - - if (!strncmp(field->type, "__data_loc", 10)) - array_descriptor = NULL; - - if (!array_descriptor) { - r = trace_seq_printf(s, "\tfield:%s %s;\toffset:%u;" - "\tsize:%u;\tsigned:%d;\n", - field->type, field->name, field->offset, - field->size, !!field->is_signed); - } else { - r = trace_seq_printf(s, "\tfield:%.*s %s%s;\toffset:%u;" - "\tsize:%u;\tsigned:%d;\n", - (int)(array_descriptor - field->type), - field->type, field->name, - array_descriptor, field->offset, - field->size, !!field->is_signed); - } + if (unlikely(list_empty(head))) + return NULL; - if (--common_field_count == 0) - r = trace_seq_printf(s, "\n"); + field = list_entry(head->prev, struct ftrace_event_field, link); + return field; - if (!r) - break; + case FORMAT_PRINTFMT: + /* all done */ + return NULL; } - if (r) - r = trace_seq_printf(s, "\nprint fmt: %s\n", - call->print_fmt); + /* + * To separate common fields from event fields, the + * LSB is set on the first event field. Clear it in case. + */ + v = (void *)((unsigned long)v & ~1L); - if (!r) { - /* - * ug! The format output is bigger than a PAGE!! - */ - buf = "FORMAT TOO BIG\n"; - r = simple_read_from_buffer(ubuf, cnt, ppos, - buf, strlen(buf)); - goto out; + field = v; + if (field->link.prev == head) + return (void *)FORMAT_PRINTFMT; + + field = list_entry(field->link.prev, struct ftrace_event_field, link); + + /* Set the LSB to notify f_show to print an extra newline */ + if (index == COMMON_FIELD_COUNT) + field = (struct ftrace_event_field *) + ((unsigned long)field | 1); + + return field; +} + +static void *f_start(struct seq_file *m, loff_t *pos) +{ + loff_t l = 0; + void *p; + + /* Start by showing the header */ + if (!*pos) + return (void *)FORMAT_HEADER; + + p = (void *)FORMAT_HEADER; + do { + p = f_next(m, p, &l); + } while (p && l < *pos); + + return p; +} + +static int f_show(struct seq_file *m, void *v) +{ + struct ftrace_event_call *call = m->private; + struct ftrace_event_field *field; + const char *array_descriptor; + + switch ((unsigned long)v) { + case FORMAT_HEADER: + seq_printf(m, "name: %s\n", call->name); + seq_printf(m, "ID: %d\n", call->event.type); + seq_printf(m, "format:\n"); + return 0; + + case FORMAT_PRINTFMT: + seq_printf(m, "\nprint fmt: %s\n", + call->print_fmt); + return 0; } - r = simple_read_from_buffer(ubuf, cnt, ppos, - s->buffer, s->len); - out: - kfree(s); - return r; + /* + * To separate common fields from event fields, the + * LSB is set on the first event field. Clear it and + * print a newline if it is set. + */ + if ((unsigned long)v & 1) { + seq_putc(m, '\n'); + v = (void *)((unsigned long)v & ~1L); + } + + field = v; + + /* + * Smartly shows the array type(except dynamic array). + * Normal: + * field:TYPE VAR + * If TYPE := TYPE[LEN], it is shown: + * field:TYPE VAR[LEN] + */ + array_descriptor = strchr(field->type, '['); + + if (!strncmp(field->type, "__data_loc", 10)) + array_descriptor = NULL; + + if (!array_descriptor) + seq_printf(m, "\tfield:%s %s;\toffset:%u;\tsize:%u;\tsigned:%d;\n", + field->type, field->name, field->offset, + field->size, !!field->is_signed); + else + seq_printf(m, "\tfield:%.*s %s%s;\toffset:%u;\tsize:%u;\tsigned:%d;\n", + (int)(array_descriptor - field->type), + field->type, field->name, + array_descriptor, field->offset, + field->size, !!field->is_signed); + + return 0; +} + +static void f_stop(struct seq_file *m, void *p) +{ +} + +static const struct seq_operations trace_format_seq_ops = { + .start = f_start, + .next = f_next, + .stop = f_stop, + .show = f_show, +}; + +static int trace_format_open(struct inode *inode, struct file *file) +{ + struct ftrace_event_call *call = inode->i_private; + struct seq_file *m; + int ret; + + ret = seq_open(file, &trace_format_seq_ops); + if (ret < 0) + return ret; + + m = file->private_data; + m->private = call; + + return 0; } static ssize_t @@ -820,8 +892,10 @@ static const struct file_operations ftrace_enable_fops = { }; static const struct file_operations ftrace_event_format_fops = { - .open = tracing_open_generic, - .read = event_format_read, + .open = trace_format_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, }; static const struct file_operations ftrace_event_id_fops = { -- cgit v1.2.3 From 1aa54bca6ee0d07ebcafb8ca8074b624d80724aa Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Wed, 28 Jul 2010 01:18:01 +0200 Subject: tracing: Sanitize value returned from write(trace_marker, "...", len) When userspace code writes non-new-line-terminated string to trace_marker file, write handler appends new-line and returns number of bytes written to trace buffer, so write(fd, "abc", 3) will return 4 That's unexpected and unfortunately it confuses glibc's fprintf function. Example: int main() { fprintf(stderr, "abc"); return 0; } $ gcc test.c -o test $ echo mmiotrace > /sys/kernel/debug/tracing/current_tracer $ ./test 2>/sys/kernel/debug/tracing/trace_marker results in infinite loop: write(fd, "abc", 3) = 4 write(fd, "", 1) = 0 write(fd, "", 1) = 0 write(fd, "", 1) = 0 write(fd, "", 1) = 0 write(fd, "", 1) = 0 write(fd, "", 1) = 0 write(fd, "", 1) = 0 (...) ...and kernel trace buffer full of empty markers. Fix it by sanitizing write return value. Signed-off-by: Marcin Slusarz LKML-Reference: <20100727231801.GB2826@joi.lan> Cc: Frederic Weisbecker Cc: Ingo Molnar Signed-off-by: Steven Rostedt --- kernel/trace/trace.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 086d36316805..88b42d14d32d 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -3498,6 +3498,7 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *fpos) { char *buf; + size_t written; if (tracing_disabled) return -EINVAL; @@ -3519,11 +3520,15 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, } else buf[cnt] = '\0'; - cnt = mark_printk("%s", buf); + written = mark_printk("%s", buf); kfree(buf); - *fpos += cnt; + *fpos += written; - return cnt; + /* don't tell userspace we wrote more - it might confuse them */ + if (written > cnt) + written = cnt; + + return written; } static int tracing_clock_show(struct seq_file *m, void *v) -- cgit v1.2.3 From 151772dbfad4dbe81721e40f9b3d588ea77bb7aa Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Wed, 25 Aug 2010 11:32:38 +1000 Subject: tracing/trace_stack: Fix stack trace on ppc64 save_stack_trace() stores the instruction pointer, not the function descriptor. On ppc64 the trace stack code currently dereferences the instruction pointer and shows 8 bytes of instructions in our backtraces: # cat /sys/kernel/debug/tracing/stack_trace Depth Size Location (26 entries) ----- ---- -------- 0) 5424 112 0x6000000048000004 1) 5312 160 0x60000000ebad01b0 2) 5152 160 0x2c23000041c20030 3) 4992 240 0x600000007c781b79 4) 4752 160 0xe84100284800000c 5) 4592 192 0x600000002fa30000 6) 4400 256 0x7f1800347b7407e0 7) 4144 208 0xe89f0108f87f0070 8) 3936 272 0xe84100282fa30000 Since we aren't dealing with function descriptors, use %pS instead of %pF to fix it: # cat /sys/kernel/debug/tracing/stack_trace Depth Size Location (26 entries) ----- ---- -------- 0) 5424 112 ftrace_call+0x4/0x8 1) 5312 160 .current_io_context+0x28/0x74 2) 5152 160 .get_io_context+0x48/0xa0 3) 4992 240 .cfq_set_request+0x94/0x4c4 4) 4752 160 .elv_set_request+0x60/0x84 5) 4592 192 .get_request+0x2d4/0x468 6) 4400 256 .get_request_wait+0x7c/0x258 7) 4144 208 .__make_request+0x49c/0x610 8) 3936 272 .generic_make_request+0x390/0x434 Signed-off-by: Anton Blanchard Cc: rostedt@goodmis.org Cc: fweisbec@gmail.com LKML-Reference: <20100825013238.GE28360@kryten> Signed-off-by: Ingo Molnar --- kernel/trace/trace_stack.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index 056468eae7cf..a6b7e0e0f3eb 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -249,7 +249,7 @@ static int trace_lookup_stack(struct seq_file *m, long i) { unsigned long addr = stack_dump_trace[i]; - return seq_printf(m, "%pF\n", (void *)addr); + return seq_printf(m, "%pS\n", (void *)addr); } static void print_disabled(struct seq_file *m) -- cgit v1.2.3 From 3aaba20f26f58843e8f20611e5c0b1c06954310f Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Mon, 23 Aug 2010 16:50:12 +0800 Subject: tracing: Fix a race in function profile While we are reading trace_stat/functionX and someone just disabled function_profile at that time, we can trigger this: divide error: 0000 [#1] PREEMPT SMP ... EIP is at function_stat_show+0x90/0x230 ... This fix just takes the ftrace_profile_lock and checks if rec->counter is 0. If it's 0, we know the profile buffer has been reset. Signed-off-by: Li Zefan Cc: stable@kernel.org LKML-Reference: <4C723644.4040708@cn.fujitsu.com> Signed-off-by: Steven Rostedt --- kernel/trace/ftrace.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 0d88ce9b9fb8..7cb1f45a1de1 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -381,12 +381,19 @@ static int function_stat_show(struct seq_file *m, void *v) { struct ftrace_profile *rec = v; char str[KSYM_SYMBOL_LEN]; + int ret = 0; #ifdef CONFIG_FUNCTION_GRAPH_TRACER - static DEFINE_MUTEX(mutex); static struct trace_seq s; unsigned long long avg; unsigned long long stddev; #endif + mutex_lock(&ftrace_profile_lock); + + /* we raced with function_profile_reset() */ + if (unlikely(rec->counter == 0)) { + ret = -EBUSY; + goto out; + } kallsyms_lookup(rec->ip, NULL, NULL, NULL, str); seq_printf(m, " %-30.30s %10lu", str, rec->counter); @@ -408,7 +415,6 @@ static int function_stat_show(struct seq_file *m, void *v) do_div(stddev, (rec->counter - 1) * 1000); } - mutex_lock(&mutex); trace_seq_init(&s); trace_print_graph_duration(rec->time, &s); trace_seq_puts(&s, " "); @@ -416,11 +422,12 @@ static int function_stat_show(struct seq_file *m, void *v) trace_seq_puts(&s, " "); trace_print_graph_duration(stddev, &s); trace_print_seq(m, &s); - mutex_unlock(&mutex); #endif seq_putc(m, '\n'); +out: + mutex_unlock(&ftrace_profile_lock); - return 0; + return ret; } static void ftrace_profile_reset(struct ftrace_profile_stat *stat) -- cgit v1.2.3 From b3bd3de66f60df4c9a2076e2886a622458929056 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 10 Aug 2010 14:17:51 -0700 Subject: gcc-4.6: kernel/*: Fix unused but set warnings No real bugs I believe, just some dead code. Signed-off-by: Andi Kleen Cc: Peter Zijlstra Cc: andi@firstfloor.org Signed-off-by: Andrew Morton Signed-off-by: Ingo Molnar --- kernel/trace/ring_buffer.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 19cccc3c3028..492197e2f86c 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -2985,13 +2985,11 @@ static void rb_advance_reader(struct ring_buffer_per_cpu *cpu_buffer) static void rb_advance_iter(struct ring_buffer_iter *iter) { - struct ring_buffer *buffer; struct ring_buffer_per_cpu *cpu_buffer; struct ring_buffer_event *event; unsigned length; cpu_buffer = iter->cpu_buffer; - buffer = cpu_buffer->buffer; /* * Check if we are at the end of the buffer. -- cgit v1.2.3 From 9c55cb12c1c172e2d51e85fbb5a4796ca86b77e7 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 8 Sep 2010 11:20:37 -0400 Subject: tracing: Do not allow llseek to set_ftrace_filter Reading the file set_ftrace_filter does three things. 1) shows whether or not filters are set for the function tracer 2) shows what functions are set for the function tracer 3) shows what triggers are set on any functions 3 is independent from 1 and 2. The way this file currently works is that it is a state machine, and as you read it, it may change state. But this assumption breaks when you use lseek() on the file. The state machine gets out of sync and the t_show() may use the wrong pointer and cause a kernel oops. Luckily, this will only kill the app that does the lseek, but the app dies while holding a mutex. This prevents anyone else from using the set_ftrace_filter file (or any other function tracing file for that matter). A real fix for this is to rewrite the code, but that is too much for a -rc release or stable. This patch simply disables llseek on the set_ftrace_filter() file for now, and we can do the proper fix for the next major release. Reported-by: Robert Swiecki Cc: Chris Wright Cc: Tavis Ormandy Cc: Eugene Teo Cc: vendor-sec@lst.de Cc: Signed-off-by: Steven Rostedt --- 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 7cb1f45a1de1..83a16e9ee518 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -2416,7 +2416,7 @@ static const struct file_operations ftrace_filter_fops = { .open = ftrace_filter_open, .read = seq_read, .write = ftrace_filter_write, - .llseek = ftrace_regex_lseek, + .llseek = no_llseek, .release = ftrace_filter_release, }; -- cgit v1.2.3 From 61a527362234ac3352a91ac67c50c6f7cd248eb1 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Fri, 27 Aug 2010 20:38:46 +0900 Subject: tracing/kprobe: Fix a memory leak in error case Fix a memory leak which happens when a field name conflicts with others. In error case, free_trace_probe() will free all arguments until nr_args, so this increments nr_args the begining of the loop instead of the end. Cc: Steven Rostedt Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Mathieu Desnoyers LKML-Reference: <20100827113846.22882.12670.stgit@ltc236.sdl.hitachi.co.jp> Signed-off-by: Masami Hiramatsu Signed-off-by: Arnaldo Carvalho de Melo --- kernel/trace/trace_kprobe.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 8b27c9849b42..0116c038b0bc 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -992,6 +992,9 @@ static int create_trace_probe(int argc, char **argv) /* parse arguments */ ret = 0; for (i = 0; i < argc && i < MAX_TRACE_ARGS; i++) { + /* Increment count for freeing args in error case */ + tp->nr_args++; + /* Parse argument name */ arg = strchr(argv[i], '='); if (arg) @@ -1021,11 +1024,8 @@ static int create_trace_probe(int argc, char **argv) ret = parse_probe_arg(arg, tp, &tp->args[i], is_return); if (ret) { pr_info("Parse error at argument%d. (%d)\n", i, ret); - kfree(tp->args[i].name); goto error; } - - tp->nr_args++; } ret = register_trace_probe(tp); -- cgit v1.2.3 From aba91595cfcebd193425e20aabc407531526a1c5 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Fri, 27 Aug 2010 20:39:06 +0900 Subject: tracing/kprobes: Fix handling of argument names Set "argN" name for each argument automatically if it has no specified name. Since dynamic trace event(kprobe_events) accepts special characters for its argument, its format can show those special characters (e.g. '$', '%', '+'). However, perf can't parse those format because of the character (especially '%') mess up the format. This sets "argX" name for those arguments if user omitted the argument names. E.g. # echo 'p do_fork %ax IP=%ip $stack' > tracing/kprobe_events # cat tracing/kprobe_events p:kprobes/p_do_fork_0 do_fork arg1=%ax IP=%ip arg3=$stack Reported-by: Srikar Dronamraju Cc: Steven Rostedt Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Mathieu Desnoyers LKML-Reference: <20100827113906.22882.59312.stgit@ltc236.sdl.hitachi.co.jp> Signed-off-by: Masami Hiramatsu Signed-off-by: Arnaldo Carvalho de Melo --- kernel/trace/trace_kprobe.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 0116c038b0bc..a39251ef1a7b 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -997,15 +997,18 @@ static int create_trace_probe(int argc, char **argv) /* Parse argument name */ arg = strchr(argv[i], '='); - if (arg) + if (arg) { *arg++ = '\0'; - else + tp->args[i].name = kstrdup(argv[i], GFP_KERNEL); + } else { arg = argv[i]; + /* If argument name is omitted, set "argN" */ + snprintf(buf, MAX_EVENT_NAME_LEN, "arg%d", i + 1); + tp->args[i].name = kstrdup(buf, GFP_KERNEL); + } - tp->args[i].name = kstrdup(argv[i], GFP_KERNEL); if (!tp->args[i].name) { - pr_info("Failed to allocate argument%d name '%s'.\n", - i, argv[i]); + pr_info("Failed to allocate argument[%d] name.\n", i); ret = -ENOMEM; goto error; } @@ -1014,7 +1017,7 @@ static int create_trace_probe(int argc, char **argv) *tmp = '_'; /* convert : to _ */ if (conflict_field_name(tp->args[i].name, tp->args, i)) { - pr_info("Argument%d name '%s' conflicts with " + pr_info("Argument[%d] name '%s' conflicts with " "another field.\n", i, argv[i]); ret = -EINVAL; goto error; @@ -1023,7 +1026,7 @@ static int create_trace_probe(int argc, char **argv) /* Parse fetch argument */ ret = parse_probe_arg(arg, tp, &tp->args[i], is_return); if (ret) { - pr_info("Parse error at argument%d. (%d)\n", i, ret); + pr_info("Parse error at argument[%d]. (%d)\n", i, ret); goto error; } } -- cgit v1.2.3 From da34634fd39958725310d2c30c9b4543945f968b Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Fri, 27 Aug 2010 20:39:12 +0900 Subject: tracing/kprobe: Fix handling of C-unlike argument names Check the argument name whether it is invalid (not C-like symbol name). This makes event format simple. Reported-by: Srikar Dronamraju Cc: Steven Rostedt Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Mathieu Desnoyers LKML-Reference: <20100827113912.22882.62313.stgit@ltc236.sdl.hitachi.co.jp> Signed-off-by: Masami Hiramatsu Signed-off-by: Arnaldo Carvalho de Melo --- kernel/trace/trace_kprobe.c | 20 ++++++++++++-------- 1 file changed, 12 insertions(+), 8 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index a39251ef1a7b..544301d29dee 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -514,8 +514,8 @@ static int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs); static int kretprobe_dispatcher(struct kretprobe_instance *ri, struct pt_regs *regs); -/* Check the name is good for event/group */ -static int check_event_name(const char *name) +/* Check the name is good for event/group/fields */ +static int is_good_name(const char *name) { if (!isalpha(*name) && *name != '_') return 0; @@ -557,7 +557,7 @@ static struct trace_probe *alloc_trace_probe(const char *group, else tp->rp.kp.pre_handler = kprobe_dispatcher; - if (!event || !check_event_name(event)) { + if (!event || !is_good_name(event)) { ret = -EINVAL; goto error; } @@ -567,7 +567,7 @@ static struct trace_probe *alloc_trace_probe(const char *group, if (!tp->call.name) goto error; - if (!group || !check_event_name(group)) { + if (!group || !is_good_name(group)) { ret = -EINVAL; goto error; } @@ -883,7 +883,7 @@ static int create_trace_probe(int argc, char **argv) int i, ret = 0; int is_return = 0, is_delete = 0; char *symbol = NULL, *event = NULL, *group = NULL; - char *arg, *tmp; + char *arg; unsigned long offset = 0; void *addr = NULL; char buf[MAX_EVENT_NAME_LEN]; @@ -1012,9 +1012,13 @@ static int create_trace_probe(int argc, char **argv) ret = -ENOMEM; goto error; } - tmp = strchr(tp->args[i].name, ':'); - if (tmp) - *tmp = '_'; /* convert : to _ */ + + if (!is_good_name(tp->args[i].name)) { + pr_info("Invalid argument[%d] name: %s\n", + i, tp->args[i].name); + ret = -EINVAL; + goto error; + } if (conflict_field_name(tp->args[i].name, tp->args, i)) { pr_info("Argument[%d] name '%s' conflicts with " -- cgit v1.2.3 From 9cb627d5f38830ca19aa0dca52d1d3a633018bf7 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Wed, 1 Sep 2010 12:58:43 +0200 Subject: perf, trace: Fix module leak Commit 1c024eca (perf, trace: Optimize tracepoints by using per-tracepoint-per-cpu hlist to track events) caused a module refcount leak. Reported-And-Tested-by: Avi Kivity Signed-off-by: Peter Zijlstra LKML-Reference: <4C7E1F12.8030304@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/trace/trace_event_perf.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c index 000e6e85b445..31cc4cb0dbf2 100644 --- a/kernel/trace/trace_event_perf.c +++ b/kernel/trace/trace_event_perf.c @@ -91,6 +91,8 @@ int perf_trace_init(struct perf_event *p_event) tp_event->class && tp_event->class->reg && try_module_get(tp_event->mod)) { ret = perf_trace_event_init(tp_event, p_event); + if (ret) + module_put(tp_event->mod); break; } } @@ -146,6 +148,7 @@ void perf_trace_destroy(struct perf_event *p_event) } } out: + module_put(tp_event->mod); mutex_unlock(&event_mutex); } -- cgit v1.2.3 From df09162550fbb53354f0c88e85b5d0e6129ee9cc Mon Sep 17 00:00:00 2001 From: Chris Wright Date: Thu, 9 Sep 2010 16:34:59 -0700 Subject: tracing: t_start: reset FTRACE_ITER_HASH in case of seek/pread Be sure to avoid entering t_show() with FTRACE_ITER_HASH set without having properly started the iterator to iterate the hash. This case is degenerate and, as discovered by Robert Swiecki, can cause t_hash_show() to misuse a pointer. This causes a NULL ptr deref with possible security implications. Tracked as CVE-2010-3079. Cc: Robert Swiecki Cc: Eugene Teo Cc: Signed-off-by: Chris Wright Signed-off-by: Steven Rostedt --- 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 83a16e9ee518..fa7ece649fe1 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -1510,6 +1510,8 @@ static void *t_start(struct seq_file *m, loff_t *pos) if (*pos > 0) return t_hash_start(m, pos); iter->flags |= FTRACE_ITER_PRINTALL; + /* reset in case of seek/pread */ + iter->flags &= ~FTRACE_ITER_HASH; return iter; } -- cgit v1.2.3