On 26 October 2011 17:20, Andreas Färber <afaer...@suse.de> wrote: >> +DECLARE_TLS(CPUState *,tls_cpu_single_env); > > So this effectively becomes tls__tls_cpu_single_env iiuc. > For compile-checking for missed conversions, > > DECLARE_TLS(CPUState *,cpu_single_env); > > should do, no?
Yes. (That does impose the silent requirement that DEFINE_TLS(type, x) is defined such that it only uses its parameter 'x' with the ## concatenation operator, though, otherwise when exec.c invokes DEFINE_TLS(CPUState *,cpu_single_env) we'd get an unwanted extra expansion. I don't know if we care about that.) >> diff --git a/darwin-user/main.c b/darwin-user/main.c >> index 1a881a0..c0f14f8 100644 >> --- a/darwin-user/main.c >> +++ b/darwin-user/main.c >> @@ -729,8 +729,6 @@ static void usage(void) >> >> /* XXX: currently only used for async signals (see signal.c) */ >> CPUState *global_env; >> -/* used only if single thread */ >> -CPUState *cpu_single_env = NULL; >> >> /* used to free thread contexts */ >> TaskState *first_task_state; > > This is not making anything local, it's apparently removing an unused > variable. Can we then have that in a patch saying so, please? Sure. -- PMM
