On Wed, Jan 23, 2013 at 08:10:36PM +0100, Marc-André Lureau wrote: > Avoid the spice-x from hanging in WaitForInputIdle(), although the > client itself might not be ready, not even started... If I understand things correctly, this patch makes sure we still process Windows messages while waiting for pi.hProcess to be signalled, otherwise this will interact badly if an activex launches the debug-helper process? If I understood this correctly, ACK, adding something like this to the commit log may make sense as imo this makes things clearer. Christophe > > https://bugzilla.redhat.com/show_bug.cgi?id=903190 > --- > src/debug-helper.c | 16 ++++++++++++++-- > 1 file changed, 14 insertions(+), 2 deletions(-) > > diff --git a/src/debug-helper.c b/src/debug-helper.c > index 0a9a923..01909da 100644 > --- a/src/debug-helper.c > +++ b/src/debug-helper.c > @@ -24,6 +24,7 @@ int > main(int argc, char *argv[]) > { > char pipe[2048]; > + MSG msg; > STARTUPINFO si = { 0, }; > PROCESS_INFORMATION pi = { 0, }; > gchar *program_path = get_program_path(); > @@ -55,8 +56,19 @@ main(int argc, char *argv[]) > goto end; > } > > - // Wait until child process exits > - WaitForSingleObject(pi.hProcess, INFINITE); > + > + while (1) { > + DWORD reason = MsgWaitForMultipleObjects(1, &pi.hProcess, FALSE, > + INFINITE, QS_ALLINPUT); > + if (reason == WAIT_OBJECT_0) > + break; > + else { > + if (PeekMessage(&msg, NULL, 0, 0, PM_REMOVE)) { > + TranslateMessage(&msg); > + DispatchMessage(&msg); > + } > + } > + } > > // Close process and thread handles > CloseHandle(pi.hProcess); > -- > 1.8.1.rc1.17.g75ed918 > > _______________________________________________ > virt-tools-list mailing list > virt-tools-list@xxxxxxxxxx > https://www.redhat.com/mailman/listinfo/virt-tools-list
Attachment:
pgp6cCbx29RX6.pgp
Description: PGP signature
_______________________________________________ virt-tools-list mailing list virt-tools-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/virt-tools-list
