On Wed, Jul 27, 2022 at 01:37:30PM -0500, Scott Cheloha wrote:
> On Wed, Jul 27, 2022 at 07:31:11AM +0100, Jason McIntyre wrote:
> > On Tue, Jul 26, 2022 at 09:18:47PM -0500, Scott Cheloha wrote:
> > > A few improvements I want to make to the sleep(1) manpage.
> > >
> > > DESCRIPTION
> > >
> > >
The diff below removes support for 386SX/DX processors. We already claim
we don't support anything older than a Pentium so there's no point to keep
this code.
The main code change is in locore0.S and is to stop checking if the CPU
we're on has the alignment check (PSL_AC) flag.
The rest of the
On Wed, Jul 27, 2022 at 04:15:36PM +0200, Alexander Hall wrote:
>
>
> On July 27, 2022 3:47:56 PM GMT+02:00, Andrew Hewus Fresh
> wrote:
> >On Wed, Jul 27, 2022 at 11:06:39AM +0200, Alexander Hall wrote:
> >> I think replacing '*' with '*([![:cntrl:]])' can be the droid your looking
> >> for.
I've always loved these little feedback loops in the random subsystem,
where it relies upon it's own random state to subtly change the meaning
of future events.
Damien Miller wrote:
> On Thu, 28 Jul 2022, Damien Miller wrote:
>
> > On Wed, 27 Jul 2022, Theo de Raadt wrote:
> >
> > > + rs
On Thu, 28 Jul 2022, Damien Miller wrote:
> On Wed, 27 Jul 2022, Theo de Raadt wrote:
>
> > + rs->rs_count += rekey_fuzz & (REKEY_BASE - 1);
> >
> > I mean, why not use % here
>
> Sure, that's reasonable.
revised:
Index: crypt/arc4random.c
===
> because we can't use arc4random_uniform() in this context.
But who says it has to be uniform?? It is code which wasn't even there
before. Or put it a different way, this additional you are adding used to
be uniformly zero, or ununiformly zero...
On Wed, 27 Jul 2022, Theo de Raadt wrote:
> + rs->rs_count += rekey_fuzz & (REKEY_BASE - 1);
>
> I mean, why not use % here
Sure, that's reasonable.
> And then, set the default to a pow2.
>
> But if someone changes it to not pow2, it still works.
>
> The & is premature hand-optimization
On Wed, 27 Jul 2022, Theo de Raadt wrote:
> I love it.
>
> > +#define REKEY_BASE (1<<20) /* NB. *must* be a power of 2 */
>
> Why insist on that?
Because I need to do this later:
+ rs->rs_count += rekey_fuzz & (REKEY_BASE - 1);
because we can't use arc4random_uniform() in this context.
+ rs->rs_count += rekey_fuzz & (REKEY_BASE - 1);
I mean, why not use % here
And then, set the default to a pow2.
But if someone changes it to not pow2, it still works.
The & is premature hand-optimization, let a compiler do it if it can.
I love it.
> +#define REKEY_BASE (1<<20) /* NB. *must* be a power of 2 */
Why insist on that?
Also, I would prefer (1024*1024), it is quicker to read.
Hi,
arc4random() rekeys currently rekeys from the kernel every 1.6MB.
It costs us almost nothing to make this interval non-deterministic,
so let's do that.
With the below it will rekey randomly somewhere between 1MB and 2MB.
ok?
Index: crypt/arc4random.c
On Sat, Jul 16, 2022 at 10:28:56AM +, Visa Hankala wrote:
> The index of a network interface is assigned in if_idxmap_insert().
> ifq_init() and ifiq_init() use if_index before it has its final value.
> As a consequence, interfaces tend to use net_tq(0) as their first
> softnet task queue even
Hi,
There exist functions to print routes from the kernel, but they
were not accessible from ddb. Implement "show all routes" to print
routing tables, and "show route 0xfd807e9b" for a single
route entry.
Note that the rtable id is not part of a route entry, so it make
no sense to print
> Date: Wed, 27 Jul 2022 00:11:19 +0200
> From: Alexander Bluhm
>
> On Tue, Jul 26, 2022 at 11:19:27PM +0200, Mark Kettenis wrote:
> > > Date: Tue, 26 Jul 2022 18:11:01 +0200
> > > From: Alexander Bluhm
> > >
> > > On Fri, Jul 22, 2022 at 06:13:04PM +0200, Alexander Bluhm wrote:
> > > > But I am
On Wed, Jul 27, 2022 at 07:31:11AM +0100, Jason McIntyre wrote:
> On Tue, Jul 26, 2022 at 09:18:47PM -0500, Scott Cheloha wrote:
> > A few improvements I want to make to the sleep(1) manpage.
> >
> > DESCRIPTION
> >
> > - "for a minimum of" is better said "for at least".
> >
>
> hi.
>
> i can;
On Wed, Jul 27, 2022 at 06:36:22PM +0200, Theo Buehler wrote:
> On Wed, Jul 27, 2022 at 05:41:11PM +0200, Claudio Jeker wrote:
> > My last try was not successful because kr_tofull() did not return RTP_MINE
> > and so some checks for RTP_MINE instead of F_KERNEL did not work.
> >
> > This diff does
On Wed, Jul 27, 2022 at 05:41:11PM +0200, Claudio Jeker wrote:
> My last try was not successful because kr_tofull() did not return RTP_MINE
> and so some checks for RTP_MINE instead of F_KERNEL did not work.
>
> This diff does two things. It replaces the F_KERNEL checks with !F_BGPD
> checks. F_KE
My last try was not successful because kr_tofull() did not return RTP_MINE
and so some checks for RTP_MINE instead of F_KERNEL did not work.
This diff does two things. It replaces the F_KERNEL checks with !F_BGPD
checks. F_KERNEL and F_BGPD are mutual exclusive.
On top of that keep kr->priority at
On July 27, 2022 3:47:56 PM GMT+02:00, Andrew Hewus Fresh
wrote:
>On Wed, Jul 27, 2022 at 11:06:39AM +0200, Alexander Hall wrote:
>> I think replacing '*' with '*([![:cntrl:]])' can be the droid your looking
>> for.
>
>As I was falling asleep last night I was trying to figure out how to get
>
On Wed, Jul 27, 2022 at 11:06:39AM +0200, Alexander Hall wrote:
> I think replacing '*' with '*([![:cntrl:]])' can be the droid your looking
> for.
As I was falling asleep last night I was trying to figure out how to get
'*([!$_nl])' into the match and couldn't think of a good one. Your
solution
Dear tech@,
the scan-build(1) tool from the llvm port on sbin/pfctl reported a
memory leak in the error path of load_feedback_profile.
There are more places that could be investigated.
Would it make sense to look into these?
Independent of that I was wondering if it makes sense to warn if
pfctl_
On Wed, Jul 27, 2022 at 01:55:58PM +0200, Alexander Hall wrote:
> On a sidenote, the
>
> _nl=$(echo)
>
> in fw_update.sh is AFAICT deceitful and moot, since ksh strips away trailing
> linefeeds. Use the uglier
>
> _nl='
> '
>
> instead.
Correct:
$ _nl=$(echo)
$ printf
On July 27, 2022 11:06:39 AM GMT+02:00, Alexander Hall
wrote:
>On July 27, 2022 12:48:29 AM GMT+02:00, Andrew Hewus Fresh
> wrote:
>>On Sun, Jul 24, 2022 at 10:34:04AM -0700, Andrew Hewus Fresh wrote:
>>> On Sun, Jul 24, 2022 at 09:14:30AM -0700, Andrew Hewus Fresh wrote:
>>> > On Sun, Jul 24
On July 27, 2022 12:48:29 AM GMT+02:00, Andrew Hewus Fresh
wrote:
>On Sun, Jul 24, 2022 at 10:34:04AM -0700, Andrew Hewus Fresh wrote:
>> On Sun, Jul 24, 2022 at 09:14:30AM -0700, Andrew Hewus Fresh wrote:
>> > On Sun, Jul 24, 2022 at 10:01:26AM -0600, Theo de Raadt wrote:
>> > > Jonathan Gray
24 matches
Mail list logo
