% uname -a SunOS abc 5.10 Generic_125100-09 sun4v sparc SUNW,Sun-Fire-T1000
It's easy... Just blast my system with 10,000 pps of 100 octets each. If I set the sq_max_size to 0 (unlimited sync queues), then at 15,000 pps the machine would run out of memory and panic. Thanks, Leonti -----Original Message----- From: Peter Memishian [mailto:[EMAIL PROTECTED] Sent: Friday, June 22, 2007 8:31 PM To: Leonti Dailis Cc: [email protected] Subject: re: [osol-code] Putnext and qfill_syncq question > Is my conclusion correct? If yes, would creating a separate M4 instead > of M1 solve this (so that the qfill_sync is not invoked?) There are > several reasons I don't want this behavior: normal STREAMS flow control > is not functioning since the normal STREAMS queue is not used to back up > extra traffic. Also invoking qfill_syncq on every message is probably > costly (though CPU utilization is almost 0) and sync queue can grow very > quickly before it is drained, eventually causing QFULL condition. I'm a bit surprised by this -- what release are you running on? On OpenSolaris, sq_max_size is 10000, so it should be relatively hard to end up with qfill_syncq() leading to QFULL. -- meem - Scanned for SPAM by SS8 IT-- _______________________________________________ opensolaris-code mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/opensolaris-code
