On a 6.1/amd64 NFS server, I got the following:
uvm_fault(0xfe82cd149228, 0x0, 1) - e
fatal page fault in supervisor mode
trap type 6 code 0 rip 803fc5a1 cs 8 rflags 10246 cr2 c8 cpl 0 rsp
fe82aaa38660
kernel: page fault trap, code=0
Stopped in pid 9953.3 (nfsd) at
On Mon, May 26, 2014 at 12:24:56PM +0200, Martin Husemann wrote:
On Mon, May 26, 2014 at 11:39:09AM +0200, Edgar Fuß wrote:
On a 6.1/amd64 NFS server, I got the following:
uvm_fault(0xfe82cd149228, 0x0, 1) - e
fatal page fault in supervisor mode
trap type 6 code 0 rip
Hi,
while watching the thread about the performance in the newest
kernel, i experience my own performance riddle.
On a VM with 1 GiB reserved host RAM (which it does not exhaust)
and elsewise idle host system, i see poor performance of a virtual
DOS-MBR-partition as device file, while it
On Mon, May 26, 2014 at 10:11:16PM +0200, Thomas Schmitt wrote:
netbsd# dd bs=2048 count=131072 if=/dev/wd1f of=/dev/null
The short answer is: use rwd1f, not wd1f. The latter goes through the
buffer cache, but unlike when you mount it uses the old (non-unified)
buffer cache for everything...
On Mon, May 26, 2014 at 12:43:32PM +0200, Manuel Bouyer wrote:
if ((error = VFS_VGET(mp, ufhp-ufid_ino, nvp)) != 0) {
*vpp = NULLVP;
return (error);
}
ip = VTOI(nvp);
KASSERT(ip !=
Hi,
David Holland:
The short answer is: use rwd1f, not wd1f.
I tried this already. Re-try on 6.99.40:
netbsd# dd bs=2048 count=131072 if=/dev/rwd1f of=/dev/null
131072+0 records in
131072+0 records out
268435456 bytes transferred in 36.982 secs (7258543 bytes/sec)
In any case the
Hello,
As we are trying to bring more parallelism in our network stack, I would
like to make IPv4 and IPv6 input queues lockless. This is implemented by
replacing struct ifqueue and macros around it with a pktqueue interface.
This interface also abstracts and handles network ISR scheduling and