From ea1a525fa369dcc60de04a0aef8e332c21b4e830 Mon Sep 17 00:00:00 2001 From: Ivo Anjo Date: Mon, 2 Dec 2024 10:09:01 +0000 Subject: [PATCH] Fix outdated comment It's no longer true that we don't use the new Ruby profiler on Ruby 2.5; what's true is that we enable the "no signals workaround" to avoid the potentially-problematic code path. --- ext/datadog_profiling_native_extension/private_vm_api_access.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/ext/datadog_profiling_native_extension/private_vm_api_access.c b/ext/datadog_profiling_native_extension/private_vm_api_access.c index dbbebd531d0..59e5c644882 100644 --- a/ext/datadog_profiling_native_extension/private_vm_api_access.c +++ b/ext/datadog_profiling_native_extension/private_vm_api_access.c @@ -158,7 +158,7 @@ bool is_current_thread_holding_the_gvl(void) { // // Thus an incorrect `is_current_thread_holding_the_gvl` result may lead to issues inside `rb_postponed_job_register_one`. // - // For this reason we currently do not enable the new Ruby profiler on Ruby 2.5 by default, and we print a + // For this reason we default to use the "no signals workaround" on Ruby 2.5 by default, and we print a // warning when customers force-enable it. bool gvl_acquired = vm->gvl.acquired != 0; rb_thread_t *current_owner = vm->running_thread;