Date: Fri, 4 Nov 2005 20:49:05 +1100 From: Peter Jeremy <PeterJeremy@optushome.com.au> To: kamal kc <kamal_ckk@yahoo.com> Cc: freebsd <freebsd-hackers@freebsd.org>, Giorgos Keramidas <keramida@linux.gr> Subject: Re: allocating 14KB memory per packet compression/decompression results in vm_fault Message-ID: <20051104094904.GL39882@cirb503493.alcatel.com.au> In-Reply-To: <20051104065630.9592.qmail@web35704.mail.mud.yahoo.com> References: <20051103145729.GA2088@flame.pc> <20051104065630.9592.qmail@web35704.mail.mud.yahoo.com>
next in thread | previous in thread | raw e-mail | index | archive | help
[dropping -net] On Thu, 2005-Nov-03 22:56:30 -0800, kamal kc wrote: >as i said before the compression/decompression works >fine. but soon the kernel would panic with one >of the vm_fault: error message. What's the exact panic and traceback? Have you enabled the various sanity checks (WITNESS, DEBUG_MEMGUARD, INVARIANTS)? >what would be the best possible way to >allocate/deallocate 14KB memory per packet without >causing vm_faults ?? The most efficient way would be to statically allocate the dictionary and string tables. The downside is that you then need to serialise the [de]compression. You could look at zone(9) as an alternative to malloc(9). >is there anything i am missing ?? - Are you correctly allocating the mbuf for the outgoing packet correctly? - Are you correctly freeing the mbuf for the incoming packet? - Are you sure that you're not writing outside the bounds of one of the memory blocks you're allocating? -- Peter Jeremy
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20051104094904.GL39882>