The following reply was made to PR kern/46325; it has been noted by GNATS.
From: Christoph Egger<Christoph_Egger%gmx.de@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: kern-bug-people%netbsd.org@localhost, gnats-admin%netbsd.org@localhost,
netbsd-bugs%netbsd.org@localhost
Subject: Re: kern/46325: wapbl + disk io = temporary system freeze
Date: Thu, 12 Apr 2012 13:01:11 +0200
On 04/12/12 12:45, Christoph Egger wrote:
> The following reply was made to PR kern/46325; it has been noted by GNATS.
>
> From: Christoph Egger<Christoph_Egger%gmx.de@localhost>
> To: gnats-bugs%NetBSD.org@localhost
> Cc: Martin Husemann<martin%duskware.de@localhost>,
kern-bug-people%netbsd.org@localhost,
> gnats-admin%netbsd.org@localhost, netbsd-bugs%netbsd.org@localhost
> Subject: Re: kern/46325: wapbl + disk io = temporary system freeze
> Date: Thu, 12 Apr 2012 12:40:50 +0200
>
> On 04/12/12 11:50, Martin Husemann wrote:
> > The following reply was made to PR kern/46325; it has been noted by
GNATS.
> >
> > From: Martin Husemann<martin%duskware.de@localhost>
> > To: gnats-bugs%NetBSD.org@localhost
> > Cc:
> > Subject: Re: kern/46325: wapbl + disk io = temporary system freeze
> > Date: Thu, 12 Apr 2012 11:47:51 +0200
> >
> > Stupid questions, just to make sure we are not barking up the
wrong tree:
> >
> > - your underlying disk io speed is OK? (dmesg excerpts or
bonnie++ runs
> > would give hints)
>
> wd0 at atabus1 drive 0
> wd0:<ST3500413AS>
> wd0: drive supports 16-sector PIO transfers, LBA48 addressing
> wd0: 465 GB, 969021 cyl, 16 head, 63 sec, 512 bytes/sect x 976773168
sectors
> wd0: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 6 (Ultra/133)
>
> that's ok.
>
> pciide0 at pci0 dev 17 function 0: AMD Hudson SATA Controller (rev. 0x40)
> pciide0: bus-master DMA support present, but unused (no driver support)
> pciide0: primary channel configured to native-PCI mode
> pciide0: using ioapic0 pin 19 for native-PCI interrupt
> atabus0 at pciide0 channel 0
> pciide0: secondary channel configured to native-PCI mode
> atabus1 at pciide0 channel 1
>
> eek, DMA support is missing in pciide0.
Problem seems to be solved: I changed the BIOS SATA mode from IDE to AHCI.
load averages: 0.95, 0.47, 0.19; up 0+00:04:52
12:58:40
135 threads: 18 idle, 111 sleeping, 2 zombie, 4 on CPU
CPU0 states: 0.0% user, 0.0% nice, 9.4% system, 5.6% interrupt,
85.0% idle
CPU1 states: 1.6% user, 0.0% nice, 4.4% system, 0.0% interrupt,
94.0% idle
CPU2 states: 0.2% user, 0.0% nice, 0.0% system, 0.0% interrupt,
99.8% idle
CPU3 states: 6.6% user, 0.0% nice, 68.1% system, 0.0% interrupt,
25.3% idle
Memory: 783M Act, 16K Wired, 88M Exec, 177M File, 6852M Free
Swap: 8171M Total, 8171M Free
PID LID UID PRI STATE TIME WCPU CPU NAME COMMAND
483 1 0 26 CPU/3 0:36 74.88% 69.34% - cvs
0 63 0 124 syncer/1 0:02 6.15% 6.15% ioflush [system]
I am still puzzled why so much 'system' time is spent.