Hi Corinna,
On Wed, 11 Mar 2026 16:11:08 +0100
Corinna Vinschen wrote:
> On Mar 10 17:50, Takashi Yano wrote:
> > The current code waits for `sendsig` by `for` loop in sigproc.cc,
> > however, the wait time might be insufficient for recent CPU.
> > The current code is as follows.
> >
> > for (int i = 0; !p->sendsig && i < 10000; i++)
> > yield ();
> >
> > Due to this problem, in tcsh, the following command occasionally
> > cannot be terminated by Ctrl-C. This is because, SIGCONT does not
> > wake-up `sleep` process correctly.
> >
> > $ cat | sleep 100 &
> > $ fg
> > $ (type Ctrl-C)
> >
> > With this patch, the wait time for `sendsig` is guaranteed to be
> > up to 100ms instead of looping for 10000 times.
> >
> > Fixes: d584454c8231 ("* sigproc.cc (sig_send): Wait for dwProcessId to be
> > non-zero as well as sendsig.")
> > Signed-off-by: Takashi Yano <[email protected]>
> > Reviewed-by:
> > ---
> > winsup/cygwin/sigproc.cc | 3 ++-
> > 1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/winsup/cygwin/sigproc.cc b/winsup/cygwin/sigproc.cc
> > index 30779cf8e..0fd7ed3ba 100644
> > --- a/winsup/cygwin/sigproc.cc
> > +++ b/winsup/cygwin/sigproc.cc
> > @@ -646,7 +646,8 @@ sig_send (_pinfo *p, siginfo_t& si, _cygtls *tls)
> > {
> > HANDLE dupsig;
> > DWORD dwProcessId;
> > - for (int i = 0; !p->sendsig && i < 10000; i++)
> > + DWORD t0 = GetTickCount ();
>
> Again a case where GetTickCount is sufficient? I'd suggest
> to use GetTickCount64 instead.
>
> Other than that, LGTM.
Thanks for reviewing.
The usage of GetTickCount() like this is safe, because
DWORD (unsigned integer) wraps on overflow. For example:
If the first call of GetTickCount() returns 0xFFFFFFF0,
and the second one returns 0x00000010, the result of
subtranction is:
0x00000010 - 0xFFFFFFF0 = 0x00000020
Therefore, if yield() returns within 49 days, the result
will be as expected.
Am I overlooking something?
--
Takashi Yano <[email protected]>
