From 0cdf5640e4f6940bdbbefee4bb0adb7dffb185ec Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Fri, 2 Oct 2015 18:42:00 +0200 Subject: ebpf: include perf_event only where really needed Commit ea317b267e9d ("bpf: Add new bpf map type to store the pointer to struct perf_event") added perf_event.h to the main eBPF header, so it gets included for all users. perf_event.h is actually only needed from array map side, so lets sanitize this a bit. Signed-off-by: Daniel Borkmann Cc: Kaixu Xia Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- kernel/bpf/arraymap.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel/bpf/arraymap.c') diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 29ace107f236..2fecc4aed119 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -15,6 +15,7 @@ #include #include #include +#include /* Called from syscall */ static struct bpf_map *array_map_alloc(union bpf_attr *attr) -- cgit v1.2.3 From aaac3ba95e4c8b496d22f68bd1bc01cfbf525eca Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Wed, 7 Oct 2015 22:23:22 -0700 Subject: bpf: charge user for creation of BPF maps and programs since eBPF programs and maps use kernel memory consider it 'locked' memory from user accounting point of view and charge it against RLIMIT_MEMLOCK limit. This limit is typically set to 64Kbytes by distros, so almost all bpf+tracing programs would need to increase it, since they use maps, but kernel charges maximum map size upfront. For example the hash map of 1024 elements will be charged as 64Kbyte. It's inconvenient for current users and changes current behavior for root, but probably worth doing to be consistent root vs non-root. Similar accounting logic is done by mmap of perf_event. Signed-off-by: Alexei Starovoitov Signed-off-by: David S. Miller --- kernel/bpf/arraymap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/bpf/arraymap.c') diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 2fecc4aed119..f2d9e698c753 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -49,7 +49,7 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr) array->map.key_size = attr->key_size; array->map.value_size = attr->value_size; array->map.max_entries = attr->max_entries; - + array->map.pages = round_up(array_size, PAGE_SIZE) >> PAGE_SHIFT; array->elem_size = elem_size; return &array->map; -- cgit v1.2.3 From a43eec304259a6c637f4014a6d4767159b6a3aa3 Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Tue, 20 Oct 2015 20:02:34 -0700 Subject: bpf: introduce bpf_perf_event_output() helper This helper is used to send raw data from eBPF program into special PERF_TYPE_SOFTWARE/PERF_COUNT_SW_BPF_OUTPUT perf_event. User space needs to perf_event_open() it (either for one or all cpus) and store FD into perf_event_array (similar to bpf_perf_event_read() helper) before eBPF program can send data into it. Today the programs triggered by kprobe collect the data and either store it into the maps or print it via bpf_trace_printk() where latter is the debug facility and not suitable to stream the data. This new helper replaces such bpf_trace_printk() usage and allows programs to have dedicated channel into user space for post-processing of the raw data collected. Signed-off-by: Alexei Starovoitov Signed-off-by: David S. Miller --- kernel/bpf/arraymap.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel/bpf/arraymap.c') diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index f2d9e698c753..e3cfe46b074f 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -295,6 +295,8 @@ static void *perf_event_fd_array_get_ptr(struct bpf_map *map, int fd) return (void *)attr; if (attr->type != PERF_TYPE_RAW && + !(attr->type == PERF_TYPE_SOFTWARE && + attr->config == PERF_COUNT_SW_BPF_OUTPUT) && attr->type != PERF_TYPE_HARDWARE) { perf_event_release_kernel(event); return ERR_PTR(-EINVAL); -- cgit v1.2.3 From 62544ce8e01c1879d420ba309f7f319d24c0f4e6 Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Thu, 22 Oct 2015 17:10:14 -0700 Subject: bpf: fix bpf_perf_event_read() helper Fix safety checks for bpf_perf_event_read(): - only non-inherited events can be added to perf_event_array map (do this check statically at map insertion time) - dynamically check that event is local and !pmu->count Otherwise buggy bpf program can cause kernel splat. Also fix error path after perf_event_attrs() and remove redundant 'extern'. Fixes: 35578d798400 ("bpf: Implement function bpf_perf_event_read() that get the selected hardware PMU conuter") Signed-off-by: Alexei Starovoitov Tested-by: Wang Nan Signed-off-by: David S. Miller --- kernel/bpf/arraymap.c | 25 ++++++++++++++++--------- 1 file changed, 16 insertions(+), 9 deletions(-) (limited to 'kernel/bpf/arraymap.c') diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index e3cfe46b074f..3f4c99e06c6b 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -292,16 +292,23 @@ static void *perf_event_fd_array_get_ptr(struct bpf_map *map, int fd) attr = perf_event_attrs(event); if (IS_ERR(attr)) - return (void *)attr; + goto err; - if (attr->type != PERF_TYPE_RAW && - !(attr->type == PERF_TYPE_SOFTWARE && - attr->config == PERF_COUNT_SW_BPF_OUTPUT) && - attr->type != PERF_TYPE_HARDWARE) { - perf_event_release_kernel(event); - return ERR_PTR(-EINVAL); - } - return event; + if (attr->inherit) + goto err; + + if (attr->type == PERF_TYPE_RAW) + return event; + + if (attr->type == PERF_TYPE_HARDWARE) + return event; + + if (attr->type == PERF_TYPE_SOFTWARE && + attr->config == PERF_COUNT_SW_BPF_OUTPUT) + return event; +err: + perf_event_release_kernel(event); + return ERR_PTR(-EINVAL); } static void perf_event_fd_array_put_ptr(void *ptr) -- cgit v1.2.3