Merge tag 'bpf-next-6.19' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next

Pull bpf updates from Alexei Starovoitov:

 - Convert selftests/bpf/test_tc_edt and test_tc_tunnel from .sh to
   test_progs runner (Alexis Lothoré)

 - Convert selftests/bpf/test_xsk to test_progs runner (Bastien
   Curutchet)

 - Replace bpf memory allocator with kmalloc_nolock() in
   bpf_local_storage (Amery Hung), and in bpf streams and range tree
   (Puranjay Mohan)

 - Introduce support for indirect jumps in BPF verifier and x86 JIT
   (Anton Protopopov) and arm64 JIT (Puranjay Mohan)

 - Remove runqslower bpf tool (Hoyeon Lee)

 - Fix corner cases in the verifier to close several syzbot reports
   (Eduard Zingerman, KaFai Wan)

 - Several improvements in deadlock detection in rqspinlock (Kumar
   Kartikeya Dwivedi)

 - Implement "jmp" mode for BPF trampoline and corresponding
   DYNAMIC_FTRACE_WITH_JMP. It improves "fexit" program type performance
   from 80 M/s to 136 M/s. With Steven's Ack. (Menglong Dong)

 - Add ability to test non-linear skbs in BPF_PROG_TEST_RUN (Paul
   Chaignon)

 - Do not let BPF_PROG_TEST_RUN emit invalid GSO types to stack (Daniel
   Borkmann)

 - Generalize buildid reader into bpf_dynptr (Mykyta Yatsenko)

 - Optimize bpf_map_update_elem() for map-in-map types (Ritesh
   Oedayrajsingh Varma)

 - Introduce overwrite mode for BPF ring buffer (Xu Kuohai)

* tag 'bpf-next-6.19' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next: (169 commits)
  bpf: optimize bpf_map_update_elem() for map-in-map types
  bpf: make kprobe_multi_link_prog_run always_inline
  selftests/bpf: do not hardcode target rate in test_tc_edt BPF program
  selftests/bpf: remove test_tc_edt.sh
  selftests/bpf: integrate test_tc_edt into test_progs
  selftests/bpf: rename test_tc_edt.bpf.c section to expose program type
  selftests/bpf: Add success stats to rqspinlock stress test
  rqspinlock: Precede non-head waiter queueing with AA check
  rqspinlock: Disable spinning for trylock fallback
  rqspinlock: Use trylock fallback when per-CPU rqnode is busy
  rqspinlock: Perform AA checks immediately
  rqspinlock: Enclose lock/unlock within lock entry acquisitions
  bpf: Remove runqslower tool
  selftests/bpf: Remove usage of lsm/file_alloc_security in selftest
  bpf: Disable file_alloc_security hook
  bpf: check for insn arrays in check_ptr_alignment
  bpf: force BPF_F_RDONLY_PROG on insn array creation
  bpf: Fix exclusive map memory leak
  selftests/bpf: Make CS length configurable for rqspinlock stress test
  selftests/bpf: Add lock wait time stats to rqspinlock stress test
  ...

1 files changed, 17 insertions, 5 deletions

diff --git a/tools/testing/selftests/bpf/prog_tests/perf_branches.c b/tools/testing/selftests/bpf/prog_tests/perf_branches.c
index bc24f83339d6..0a7ef770c487 100644
--- a/tools/testing/selftests/bpf/prog_tests/perf_branches.c
+++ b/tools/testing/selftests/bpf/prog_tests/perf_branches.c
@@ -15,6 +15,10 @@ static void check_good_sample(struct test_perf_branches *skel)
 	int pbe_size = sizeof(struct perf_branch_entry);
 	int duration = 0;
 
+	if (CHECK(!skel->bss->run_cnt, "invalid run_cnt",
+		  "checked sample validity before prog run"))
+		return;
+
 	if (CHECK(!skel->bss->valid, "output not valid",
 		 "no valid sample from prog"))
 		return;
@@ -45,6 +49,10 @@ static void check_bad_sample(struct test_perf_branches *skel)
 	int written_stack = skel->bss->written_stack_out;
 	int duration = 0;
 
+	if (CHECK(!skel->bss->run_cnt, "invalid run_cnt",
+		  "checked sample validity before prog run"))
+		return;
+
 	if (CHECK(!skel->bss->valid, "output not valid",
 		 "no valid sample from prog"))
 		return;
@@ -83,8 +91,12 @@ static void test_perf_branches_common(int perf_fd,
 	err = pthread_setaffinity_np(pthread_self(), sizeof(cpu_set), &cpu_set);
 	if (CHECK(err, "set_affinity", "cpu #0, err %d\n", err))
 		goto out_destroy;
-	/* spin the loop for a while (random high number) */
-	for (i = 0; i < 1000000; ++i)
+
+	/* Spin the loop for a while by using a high iteration count, and by
+	 * checking whether the specific run count marker has been explicitly
+	 * incremented at least once by the backing perf_event BPF program.
+	 */
+	for (i = 0; i < 100000000 && !*(volatile int *)&skel->bss->run_cnt; ++i)
 		++j;
 
 	test_perf_branches__detach(skel);
@@ -116,11 +128,11 @@ static void test_perf_branches_hw(void)
 	pfd = syscall(__NR_perf_event_open, &attr, -1, 0, -1, PERF_FLAG_FD_CLOEXEC);
 
 	/*
-	 * Some setups don't support branch records (virtual machines, !x86),
-	 * so skip test in this case.
+	 * Some setups don't support LBR (virtual machines, !x86, AMD Milan Zen
+	 * 3 which only supports BRS), so skip test in this case.
 	 */
 	if (pfd < 0) {
-		if (errno == ENOENT || errno == EOPNOTSUPP) {
+		if (errno == ENOENT || errno == EOPNOTSUPP || errno == EINVAL) {
 			printf("%s:SKIP:no PERF_SAMPLE_BRANCH_STACK\n",
 			       __func__);
 			test__skip();