From a8e537fa4d61b29c5c0ab1395918ad63f7752b25 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Alex=20Benn=C3=A9e?= Date: Mon, 29 Nov 2021 14:09:29 +0000 Subject: [PATCH] gdbstub: handle a potentially racing TaskState MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When dealing with multi-threaded userspace programs there is a race condition with the addition of cpu->opaque (aka TaskState). This is due to cpu_copy calling cpu_create which updates the global vCPU list. However the task state isn't set until later. This shouldn't be a problem because the new thread can't have executed anything yet but the gdbstub code does liberally iterate through the CPU list in various places. This sticking plaster ensure the not yet fully realized vCPU is given an pid of -1 which should be enough to ensure it doesn't show up anywhere else. In the longer term I think the code that manages the association between vCPUs and attached GDB processes could do with a clean-up and re-factor. Signed-off-by: Alex Bennée Tested-by: Richard Henderson Reviewed-by: Richard Henderson Cc: Richard Henderson Resolves: https://gitlab.com/qemu-project/qemu/-/issues/730 Message-Id: <20211129140932.4115115-6-alex.bennee@linaro.org> --- gdbstub.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/gdbstub.c b/gdbstub.c index 23baaef40e..141d7bc4ec 100644 --- a/gdbstub.c +++ b/gdbstub.c @@ -94,7 +94,7 @@ static inline int cpu_gdb_index(CPUState *cpu) { #if defined(CONFIG_USER_ONLY) TaskState *ts = (TaskState *) cpu->opaque; - return ts->ts_tid; + return ts ? ts->ts_tid : -1; #else return cpu->cpu_index + 1; #endif