On Fri, Jan 08, 2016 at 07:37:37PM -0800, Gleb Smirnoff wrote:
> On Wed, Jan 06, 2016 at 12:50:52PM -0800, Adrian Chadd wrote:
> A> hah, that's very pretty. The downside of a completely direct-dispatch
> A> network stack. :)
>
> The stack is big, but shouldn't eat 4 pages anyway. Back in the netgr
On Wed, Jan 06, 2016 at 12:50:52PM -0800, Adrian Chadd wrote:
A> hah, that's very pretty. The downside of a completely direct-dispatch
A> network stack. :)
The stack is big, but shouldn't eat 4 pages anyway. Back in the netgraph
hacking times, we used to have stack exhaustion with much longer back
hah, that's very pretty. The downside of a completely direct-dispatch
network stack. :)
-a
On 6 January 2016 at 10:00, Chagin Dmitry wrote:
> Hi, all.
>
> Here's a panic that occurs during the massive network loads,
> as far as I can see network segment has eaten the whole stack, so
> only 0xe
07.01.2016 1:00, Chagin Dmitry пишет:
Hi, all.
Here's a panic that occurs during the massive network loads,
as far as I can see network segment has eaten the whole stack, so
only 0xeb0 bytes left for interrupt hanlder.
For stable/10, sys/amd64/include/param.h defines KSTACK_PAGES=4 as default
Hi, all.
Here's a panic that occurs during the massive network loads,
as far as I can see network segment has eaten the whole stack, so
only 0xeb0 bytes left for interrupt hanlder.
#0 doadump (textdump=0x8151cd90) at /home/git/head/sys/kern/kern_shutdown.c:296
#1 0x803d4715 in db_fncal
