From 5dcf6334ae1af158ec149794c3f35e459b98df5e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Alex=20Benn=C3=A9e?= Date: Fri, 1 Dec 2023 09:36:27 +0000 Subject: gdbstub: use a better signal when we halt for IO reasons MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The gdb description GDB_SIGNAL_IO is "I/O possible" and by default gdb will try and restart the guest, getting us nowhere. Report GDB_SIGNAL_STOP instead which should at least halt the session at the failure point. Reviewed-by: Luis Machado Reviewed-by: Richard Henderson Acked-by: Stefan Hajnoczi Signed-off-by: Alex Bennée Message-Id: <20231201093633.2551497-2-alex.bennee@linaro.org> --- gdbstub/system.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'gdbstub/system.c') diff --git a/gdbstub/system.c b/gdbstub/system.c index 783ac140b9..83fd452800 100644 --- a/gdbstub/system.c +++ b/gdbstub/system.c @@ -183,7 +183,7 @@ static void gdb_vm_state_change(void *opaque, bool running, RunState state) break; case RUN_STATE_IO_ERROR: trace_gdbstub_hit_io_error(); - ret = GDB_SIGNAL_IO; + ret = GDB_SIGNAL_STOP; break; case RUN_STATE_WATCHDOG: trace_gdbstub_hit_watchdog(); -- cgit v1.2.3