[<<] [<] Page 1 of 1 [>] [>>] | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Subject:
nxFlushReq
From: "Robbie" ####@####.#### Date: 12 May 2006 16:12:07 +0100 Message-Id: <20060512151200.71F28BD67B@xprdmxin.myway.com> I've been getting this error message occationally, could someone tell me what this is related to? Oops! nxFlushReq() timed out, exiting Does this somehow tell me a timer expired and shouldn't have? thanks. _______________________________________________ Join Excite! - http://www.excite.com The most personalized portal on the Web! | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Subject:
Re: [nanogui] nxFlushReq
From: "Greg Haerr" ####@####.#### Date: 15 May 2006 01:57:25 +0100 Message-Id: <1f2a01c677b9$de26aed0$6401a8c0@winXP> : Oops! nxFlushReq() timed out, exiting : : Does this somehow tell me a timer expired and shouldn't have? Not likely. However, I can't find this in the source. Can you grep your entire tree and find where this message is generated? Regards, Greg | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Subject:
Re: [nanogui] nxFlushReq
From: "Robbie" ####@####.#### Date: 15 May 2006 15:02:58 +0100 Message-Id: <20060515140256.6CDED8B32D@xprdmxin.myway.com> --- On Sun 05/14, Greg Haerr < ####@####.#### > wrote: From: Greg Haerr [mailto: ####@####.#### To: ####@####.#### ####@####.#### Date: Sun, 14 May 2006 18:52:39 -0600 Subject: Re: [nanogui] nxFlushReq >: Oops! nxFlushReq() timed out, exiting > Does this somehow tell me a timer expired and shouldn't have? >Not likely. However, I can't find this in the source. Can<br>you grep your entire tree and find where this message is generated? grep source: src/nanox/client.c: nxFlushReq(0L,0); src/nanox/client.c: EPRINTF("Oops! nxFlushReq() timed out, exiting\n"); src/nanox/client.c:/* Vladimir Cotfas: hang in GrFlush() --> nxFlushReq(0L,1); */ src/nanox/client.c: nxFlushReq(0L,1); src/nanox/nxproto.c: nxFlushReq(aligned_len,1); src/nanox/nxproto.c: EPRINTF("nxFlushReq: Can't allocate initial request buffer\n"); src/nanox/nxproto.c: EPRINTF("nxFlushReq: write failed: %m\n"); src/nanox/nxproto.c:nxFlushReq(long newsize, int reply_needed) src/nanox/nxproto.c: EPRINTF("nxFlushReq: shm region too small\n"); src/nanox/nxproto.c: EPRINTF("nxFlushReq: Can't reallocate request buffer\n"); src/nanox/nxproto.h:void nxFlushReq(long newsize, int reply_needed); _______________________________________________ Join Excite! - http://www.excite.com The most personalized portal on the Web! | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Subject:
Re: [nanogui] nxFlushReq
From: "Greg Haerr" ####@####.#### Date: 15 May 2006 22:00:04 +0100 Message-Id: <21ce01c67862$7ce0c030$6401a8c0@winXP> : >: Oops! nxFlushReq() timed out, exiting This message comes from a hack in nanox/client.c, where the SIG_ALARM signal is caught by the client process and exits. This was a hack to fix a hanging close() in GrClose(). I suggest you set GR_CLOSE_FIX to 0 in nanox/client.c for your system, since your appliction seems to be using SIG_ALARM. In looking at this hack, I don't like it. I can't quite remember the reason for the contribution, other than the GrClose would hang on occaision. Perhaps this should always be left OFF unless required. Regards, Greg | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
[<<] [<] Page 1 of 1 [>] [>>] |