On 2011-06-14 05:46, Alexandre Raymond wrote:
> This fix effectively blocks, in the main thread, the signals handled
> by signalfd or the compatibility signal thread.
>
> This way, such signals are received synchronously in the main thread
> through sigfd_handler() instead of triggering the signal handler
> directly, asynchronously.
>
> Signed-off-by: Alexandre Raymond
> ---
> cpus.c |1 +
> 1 files changed, 1 insertions(+), 0 deletions(-)
>
> diff --git a/cpus.c b/cpus.c
> index 4ab76f0..b6366b5 100644
> --- a/cpus.c
> +++ b/cpus.c
> @@ -411,6 +411,7 @@ static int qemu_signal_init(void)
> sigaddset(&set, SIGIO);
> sigaddset(&set, SIGALRM);
> }
> +pthread_sigmask(SIG_BLOCK, &set, NULL);
> #endif
>
> sigfd = qemu_signalfd(&set);
Good catch. Probably remained unnoticed as most of us have signalfd
support on their boxes so that the window for async delivery was rather
small.
But please pull the now common pthread_sigmask out of the #ifdef.
Jan
signature.asc
Description: OpenPGP digital signature
