| delorie.com/archives/browse.cgi | search |
On Tue, Oct 15, 2002 at 11:11:09PM -0400, Pierre A. Humblet wrote: >The program below demonstrates the issue. After the second ioctl, >getchar() becomes non-blocking and returns -1 even though it should >(? I am not an expert on termio ...) wait for a character because of > sbuf.c_cc[VMIN] = 1; >It works as expected with 3-12. This is on WinME. > >Incidentally what's happening is that ispell is generating infinitely >many "bells" while getting getchar() == -1. They can be heard on the >console. Under rxvt this eventually causes fflush(stdout) to freeze. Mega thanks for the test case. That allowed me to duplicate the problem. This was caused by some recent changes to Cygwin to properly (?) deal with VMIN and VTIME. The description of how this is supposed to work is pretty straightforward in the linux man page. You wouldn't think that implementing things so that they worked that way would be so hard. But it is. I'm generating a new snapshot which should fix this. I've downloaded ispell from franken.de and verified that it was previously broken but seems to work now. I should probably incorporate this test into the cygwin test suite. PLEASE TRY THE SNAPSHOT and verify that I really did fix this. Thanks again for the test suite. That always helps the debugging process immeasurably. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
| webmaster | delorie software privacy |
| Copyright © 2019 by DJ Delorie | Updated Jul 2019 |