Date: Wed, 10 May 2000 08:20:24 -0400 (EDT) Message-Id: <200005101220.IAA11797@indy.delorie.com> From: Eli Zaretskii To: Jason Green CC: djgpp AT delorie DOT com In-reply-to: (message from Jason Green on Tue, 09 May 2000 22:38:32 +0100) Subject: Re: Signal & setitimer References: Reply-To: djgpp AT delorie DOT com Errors-To: nobody AT delorie DOT com X-Mailing-List: djgpp AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk > From: Jason Green > Newsgroups: comp.os.msdos.djgpp > Date: Tue, 09 May 2000 22:38:32 +0100 > > Eli Zaretskii wrote: > > > On Sun, 7 May 2000, Jason Green wrote: > > > > > I'm not sure exactly what the expected behaviour of this program > > > *should* be (can someone clarify this) but the actual behaviour is > > > definitely not consistent. > > > > What you report *is* reproducible. The small patch I posted should > > fix it; please try that. > > I patched itimer.c and it fixes the problem when run normally in a DOS > box. It is too difficult to tell if general protection exception > problem still exists because this was rare anyway. > > The other reported results of the test program remain - immediate exit > when run with redir, and unusual behaviour when run from UltraEdit. I cannot reproduce the problem with `redir', neither on plain DOS nor on Windows 98. Perhaps `redir' picks up a different program: `test' is not a good name for a program, as there are standard programs and commands by that name (e.g., in Bash or in Sh-utils). What happens if you rename test.exe to some less-common name? Also, what version of `redir' did you use? Was it from DJGPP v2.03? As for UltraEdit, I don't use it, so I cannot look into that. But since UltraEdit is a Windows program, and thus runs subsidiary DJGPP programs in another VM, I'm not surprised: Windows does not deliver timer interrupts to a DOS box that's in background, so perhaps when the program runs under UltraEdit, it misses some of the timer ticks when UltraEdit gets control?