On Wed, 20 Feb 2019 14:08:45 -0300
Martin Pieuchot wrote:
> When choosing the initial CPU for a new thread to run on, the scheduler
> do not consider that CPU0 is handling interrupts.
>
> Threads that are created early, like the network taskq, always end up
> scheduled on CPU0. If the machine
> In sched_choosecpu_fork(), we see best_run is INT_MAX, the comparison below
> actually reduces to a assignment ==> choice = ci;
> - if (choice == NULL || run < best_run ||
> - (run == best_run & < best_load)) {
> - choice = ci;
> -
Thanks.
ok yasuoka
On Wed, 20 Feb 2019 18:28:40 +0100
Denis Fondras wrote:
> Documentation better be accurate when syntax is not checked :)
>
> Index: npppd/npppd-users.5
> ===
> RCS file:
right now we add vlan_input as a possible input handler on the parent
interface, and if the packet is for a vlan we take it and pretend we
received it on the vlan interface by calling if_input against that mbuf.
as mpi notes, the if input queue stuff looks like a lot of work,
especially for a
On Tue, 19 Feb 2019 23:48:03 +0100, Klemens Nanni wrote:
> Given this feedback and the fact that the majority of implementations
> already behave that way: any OKs or objections for my diff?
OK millert@
- todd
On 20/02/19(Wed) 14:44, David Gwynne wrote:
> Index: sys/net/if.c
> ===
> RCS file: /cvs/src/sys/net/if.c,v
> retrieving revision 1.571
> diff -u -p -r1.571 if.c
> --- sys/net/if.c 9 Jan 2019 01:14:21 - 1.571
> +++
On Wed, Feb 20 2019, "Todd C. Miller" wrote:
> On Tue, 19 Feb 2019 23:48:03 +0100, Klemens Nanni wrote:
>
>> Given this feedback and the fact that the majority of implementations
>> already behave that way: any OKs or objections for my diff?
>
> OK millert@
Also ok jca@
--
jca | PGP :
When choosing the initial CPU for a new thread to run on, the scheduler
do not consider that CPU0 is handling interrupts.
Threads that are created early, like the network taskq, always end up
scheduled on CPU0. If the machine isn't running many programs, like
simple firewalls, there won't be
Documentation better be accurate when syntax is not checked :)
Index: npppd/npppd-users.5
===
RCS file: /cvs/src/usr.sbin/npppd/npppd/npppd-users.5,v
retrieving revision 1.7
diff -u -p -r1.7 npppd-users.5
--- npppd/npppd-users.5 13
iwm(4) has been using a firmware-based approach to retrying failed
Tx attempts with successively lower data rates. This is suboptimal
for our kernel's Tx rate selection algorithms, which are called
MiRA (for 11n) and AMRR (for 11a/b/g).
I have observed a mismatch between the Tx rate reported by
On Wed, Feb 20, 2019 at 04:04:13PM +0100, Stefan Sperling wrote:
> iwm(4) has been using a firmware-based approach to retrying failed
> Tx attempts with successively lower data rates. This is suboptimal
> for our kernel's Tx rate selection algorithms, which are called
> MiRA (for 11n) and AMRR
Hi.
When malloc fails we should return like the MMAP case does.
Greetings Ben
Index: nm.c
===
RCS file: /cvs/src/usr.bin/nm/nm.c,v
retrieving revision 1.53
diff -u -p -u -C10 -r1.53 nm.c
*** nm.c27 Oct 2017 16:47:08 -
Hi Stefan,
* Stefan Sperling wrote:
>
> When testing this, please compare throughput before and after this diff.
> I run tcpbench -s on a host behind the AP and start a tcpbench client on
> my iwm laptop. With this diff, and an 11ac AP on channel 149, tcpbench
> measures up to 26 Mpbs in my
UFS dirhash has a limit on how much memory is used to prevent runaway
allocations. I believe the default limit (2MB) is rather conservative.
It was picked to avoid causing trouble back when machines were tiny and
dirhash was new.
For example, a maildir with 100K files takes about 3MB of memory.
Hi,
In July 2018, henning@ updated src/sys/net/pfvar.h to lift the default
state table limit to 10:
https://cvsweb.openbsd.org/src/sys/net/pfvar.h?rev=1.480=text/x-cvsweb-markup
The patch below brings the pf options page inline with this change.
Cheers,
Jason.
Index: faq/pf/options.html
> On 20 Feb 2019, at 11:21 pm, Martin Pieuchot wrote:
>
> On 20/02/19(Wed) 14:44, David Gwynne wrote:
>> Index: sys/net/if.c
>> ===
>> RCS file: /cvs/src/sys/net/if.c,v
>> retrieving revision 1.571
>> diff -u -p -r1.571 if.c
>>
Jason Tubnor(ja...@tubnor.net) on 2019.02.21 09:13:46 +1100:
> Hi,
>
> In July 2018, henning@ updated src/sys/net/pfvar.h to lift the default
> state table limit to 10:
>
> https://cvsweb.openbsd.org/src/sys/net/pfvar.h?rev=1.480=text/x-cvsweb-markup
>
> The patch below brings the pf
Hi all,
Just replying to myself for the sake of completeness and anyone else
searching the list archive.
Le 06/04/2018 10:55, Adrian Close a écrit :
I'm having some issues getting an mpw(4) MPLS "pseudowire" setup
working reliably and it seems to come down to ARP issues between my
"PE" and
On Tue, Feb 12, 2019 at 04:32:03PM +0100, Claudio Jeker wrote:
> While it has been possible to do wildcard matching on communities and
> large communities for a long time, ext-communities did not offer this.
> The result is more complex rulesets when cleaning ext-communities.
> This diff allows
On Wed, Feb 20, 2019 at 09:26:35AM +0100, Claudio Jeker wrote:
> On Tue, Feb 12, 2019 at 04:32:03PM +0100, Claudio Jeker wrote:
> > While it has been possible to do wildcard matching on communities and
> > large communities for a long time, ext-communities did not offer this.
> > The result is
20 matches
Mail list logo