On Oct 5 12:26, Bob Byrnes wrote: > But there is a strange twist: When a read is pending on an empty pipe, > then WriteQuotaAvailable is also decremented! I can't imagine why this > would be the case, but it is easy to demonstrate using a pair of small > test programs that I wrote to experiment with pipe buffering.
Oh boy! > I think my speculations about POSIX atomicity requirements related to > PIPE_BUF were a red herring. Ditto for hypothetical bizarre pipe > buffering behavior. The real problem here is NtQueryInformationFile. > > I want to first try reversing the direction of the pipe. We don't really > care which end is the server or client, so we could easily modify our pipe > creation code to make an outbound pipe, and if WriteQuotaAvailable from > NtQueryInformationFile behaves sanely, then we're done. Yep, that sounds interesting. Thanks, Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader mailto:[EMAIL PROTECTED] Red Hat, Inc.
