bpf: enable stackmap with build_id in nmi context

Currently, we cannot parse build_id in nmi context because of
up_read(&current->mm->mmap_sem), this makes stackmap with build_id
less useful. This patch enables parsing build_id in nmi by putting
the up_read() call in irq_work. To avoid memory allocation in nmi
context, we use per cpu variable for the irq_work. As a result, only
one irq_work per cpu is allowed. If the irq_work is in-use, we
fallback to only report ips.

Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>

diff --git a/init/Kconfig b/init/Kconfig
index f013afc74b1139e2fe24f1dcc7df17bb91810ab4..480a4f2713d9d948d043f9697bc80d0a529d537d 100644 (file)
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -1391,6 +1391,7 @@ config BPF_SYSCALL
        bool "Enable bpf() system call"
        select ANON_INODES
        select BPF
+       select IRQ_WORK
        default n
        help
          Enable the bpf() system call that allows to manipulate eBPF

diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c
index 3ba102b4151251c3d3918ad51359c0457fc39197..b59ace0f0f0912724a2b826a77a306f31fd7fb83 100644 (file)
--- a/kernel/bpf/stackmap.c
+++ b/kernel/bpf/stackmap.c
@@ -11,6 +11,7 @@
 #include <linux/perf_event.h>
 #include <linux/elf.h>
 #include <linux/pagemap.h>
+#include <linux/irq_work.h>
 #include "percpu_freelist.h"
 
 #define STACK_CREATE_FLAG_MASK                                 \
@@ -32,6 +33,23 @@ struct bpf_stack_map {
        struct stack_map_bucket *buckets[];
 };
 
+/* irq_work to run up_read() for build_id lookup in nmi context */
+struct stack_map_irq_work {
+       struct irq_work irq_work;
+       struct rw_semaphore *sem;
+};
+
+static void do_up_read(struct irq_work *entry)
+{
+       struct stack_map_irq_work *work;
+
+       work = container_of(entry, struct stack_map_irq_work, irq_work);
+       up_read(work->sem);
+       work->sem = NULL;
+}
+
+static DEFINE_PER_CPU(struct stack_map_irq_work, up_read_work);
+
 static inline bool stack_map_use_build_id(struct bpf_map *map)
 {
        return (map->map_flags & BPF_F_STACK_BUILD_ID);
@@ -267,17 +285,27 @@ static void stack_map_get_build_id_offset(struct bpf_stack_build_id *id_offs,
 {
        int i;
        struct vm_area_struct *vma;
+       bool in_nmi_ctx = in_nmi();
+       bool irq_work_busy = false;
+       struct stack_map_irq_work *work;
+
+       if (in_nmi_ctx) {
+               work = this_cpu_ptr(&up_read_work);
+               if (work->irq_work.flags & IRQ_WORK_BUSY)
+                       /* cannot queue more up_read, fallback */
+                       irq_work_busy = true;
+       }
 
        /*
-        * We cannot do up_read() in nmi context, so build_id lookup is
-        * only supported for non-nmi events. If at some point, it is
-        * possible to run find_vma() without taking the semaphore, we
-        * would like to allow build_id lookup in nmi context.
+        * We cannot do up_read() in nmi context. To do build_id lookup
+        * in nmi context, we need to run up_read() in irq_work. We use
+        * a percpu variable to do the irq_work. If the irq_work is
+        * already used by another lookup, we fall back to report ips.
         *
         * Same fallback is used for kernel stack (!user) on a stackmap
         * with build_id.
         */
-       if (!user || !current || !current->mm || in_nmi() ||
+       if (!user || !current || !current->mm || irq_work_busy ||
            down_read_trylock(&current->mm->mmap_sem) == 0) {
                /* cannot access current->mm, fall back to ips */
                for (i = 0; i < trace_nr; i++) {
@@ -299,7 +327,13 @@ static void stack_map_get_build_id_offset(struct bpf_stack_build_id *id_offs,
                        - vma->vm_start;
                id_offs[i].status = BPF_STACK_BUILD_ID_VALID;
        }
-       up_read(&current->mm->mmap_sem);
+
+       if (!in_nmi_ctx) {
+               up_read(&current->mm->mmap_sem);
+       } else {
+               work->sem = &current->mm->mmap_sem;
+               irq_work_queue(&work->irq_work);
+       }
 }
 
 BPF_CALL_3(bpf_get_stackid, struct pt_regs *, regs, struct bpf_map *, map,
@@ -575,3 +609,16 @@ const struct bpf_map_ops stack_map_ops = {
        .map_update_elem = stack_map_update_elem,
        .map_delete_elem = stack_map_delete_elem,
 };
+
+static int __init stack_map_init(void)
+{
+       int cpu;
+       struct stack_map_irq_work *work;
+
+       for_each_possible_cpu(cpu) {
+               work = per_cpu_ptr(&up_read_work, cpu);
+               init_irq_work(&work->irq_work, do_up_read);
+       }
+       return 0;
+}
+subsys_initcall(stack_map_init);