This brings arplookup() in sync with nd6_lookup() and is necessary for
upcoming struct rtentry reference count.
Ok?
Index: netinet/if_ether.c
===
RCS file: /cvs/src/sys/netinet/if_ether.c,v
retrieving revision 1.160
diff -u -p
Ultimately my goal is to use rt_ifa_{add,del}() instead of
nd6_prefix_{on,off}link() but right now I need to remove the
rt-ref_cnt--.
This diff does that and reduce the differences between these functions.
Note that nd6_prefix_onlink()'s error are always logged so I doubt we
need two messages.
On 11/08/15(Tue) 16:25, Alexander Bluhm wrote:
On Mon, Aug 10, 2015 at 10:43:46PM +0200, Martin Pieuchot wrote:
In general these messages do not help. Here we have two cases of
messages, either logged at LOG_INFO or LOG_ERR.
Yes, multiple logs for the same error are bad. But there is a
Hello Richard,
On 17/08/15(Mon) 17:39, Richard Procter wrote:
Hi,
This series of 29 small diffs slims pf.o by 2640 bytes and pf.c by
113 non-comment lines.
We generally discuss one diff per mail. It makes it easier for people
to comment and as you can imagine deal with possible conflicts
Here a route lookup is done to find an existing ``ifa'' in order to
attach a new route.
This code runs under KERNEL_LOCK and returns a route entry found in
the table (RT_VALID), so rtfree(9) will not free the route.
Note that it is safe to rtfree() a route entry while keeping a pointer
to a
On 17/08/15(Mon) 13:38, Alexey Suslikov wrote:
Martin Pieuchot mpi at openbsd.org writes:
On 17/08/15(Mon) 17:39, Richard Procter wrote:
Hi,
This series of 29 small diffs slims pf.o by 2640 bytes and pf.c by
113 non-comment lines.
We generally discuss one diff per mail. It
On Mon, Aug 17, 2015 at 01:06:32PM +0200, Martin Pieuchot wrote:
On 11/08/15(Tue) 16:25, Alexander Bluhm wrote:
On Mon, Aug 10, 2015 at 10:43:46PM +0200, Martin Pieuchot wrote:
In general these messages do not help. Here we have two cases of
messages, either logged at LOG_INFO or
Martin Pieuchot mpi at openbsd.org writes:
On 17/08/15(Mon) 17:39, Richard Procter wrote:
Hi,
This series of 29 small diffs slims pf.o by 2640 bytes and pf.c by
113 non-comment lines.
We generally discuss one diff per mail. It makes it easier for people
to comment and as you can
armv7 kernel will fail to compile with -DSDHC_DEBUG as it lacks a
definition for ommc_dump_regs, so add it.
I previously mailed this patch a while ago[1], but my mail client fudged
tabs/whitespace, here is a working patch.
Ian
[1] http://marc.info/?l=openbsd-techm=143580188423401w=2
Index:
Hi,
http(1) is a drop-in substitute for ftp(1)'s AUTO-FETCHING FILES
mode. It defaults to HTTP method when the url doesn't specify the
protocol and supports transferring files from HTTP(S) and FTP
servers. The FTP support is limited to file transfer and doesn't
do command interpretation. It
The current sed(1) will not process a given line until it has read a
character of the following line (to detect EOF conditions). This means
that the following command
(echo foo; sleep 2; echo bar; sleep 2; echo baz; sleep 2; echo quux) | sed
Will wait two seconds before printing anything, and
I spent some time today figuting out why the binutils update broke ld
-Z on powerpc. Turns out to be a fairly thorny issue.
The new binutils discard empty setions. As a result the .gotpad0 and
.gotpad1 sections have disappeared. And a s a consequence the
__got_start and __got_end symbols
On Mon, Aug 17, 2015 at 12:34:13PM +0200, Martin Pieuchot wrote:
Ultimately my goal is to use rt_ifa_{add,del}() instead of
nd6_prefix_{on,off}link() but right now I need to remove the
rt-ref_cnt--.
@@ -1861,26 +1856,11 @@ nd6_prefix_onlink(struct nd_prefix *pr)
On 17/08/15(Mon) 18:25, Alexander Bluhm wrote:
On Mon, Aug 17, 2015 at 12:34:13PM +0200, Martin Pieuchot wrote:
Ultimately my goal is to use rt_ifa_{add,del}() instead of
nd6_prefix_{on,off}link() but right now I need to remove the
rt-ref_cnt--.
@@ -1861,26 +1856,11 @@
On Mon, Aug 17, 2015 at 07:03:55PM +0200, Martin Pieuchot wrote:
On 17/08/15(Mon) 18:25, Alexander Bluhm wrote:
On Mon, Aug 17, 2015 at 12:34:13PM +0200, Martin Pieuchot wrote:
Ultimately my goal is to use rt_ifa_{add,del}() instead of
nd6_prefix_{on,off}link() but right now I need to
On Mon, Aug 17, 2015 at 06:06:17PM +0200, Sebastien Marie wrote:
Hi,
I start reading your code, and I have a first remark.
I see in main.c (at line 142 and next) that on redirection, you trust
the server for the filename. I am not sure it is a good thing to do.
If the user request
The diff below is a first step towards running the em(4) interrupt
handler without grabbing the kernel lock. It runs the rx completion
path without the lock, which is the important bit to be able to test
the work that has been going on to make the network stack run on
multiple CPUs. But the
Hi,
I attached 3 patchs:
- retry.patch: move name of output file outside the retry loop. Should
keep the filename defined by user, instead of pike one from server.
- spell.patch: correct a small spelling error
- url-user.patch: a small change in url_parse() in order to copte with
@
On Mon, Aug 17, 2015 at 12:31:20PM +0200, Martin Pieuchot wrote:
This brings arplookup() in sync with nd6_lookup() and is necessary for
upcoming struct rtentry reference count.
@@ -633,9 +633,10 @@ in_arpinput(struct mbuf *m)
itaddr = myaddr;
goto reply;
}
Hello all,
I would like to suggest these changes to remove unused variables
and a respectively unnecessary call of the gettimeofday, also removes
a casting in the malloc, that seems to be unnecessary.
Index: worm.c
===
RCS file:
On Mon, Aug 17, 2015 at 12:29:46PM +0200, Martin Pieuchot wrote:
Here a route lookup is done to find an existing ``ifa'' in order to
attach a new route.
This code runs under KERNEL_LOCK and returns a route entry found in
the table (RT_VALID), so rtfree(9) will not free the route.
Note
Actually, ignore this, it seems to have been fixed (by accident?) by the
recent implementation of the -i flag in CURRENT.
--
\ Troels
/\ Henriksen
Hi,
When running my pf regression tests, I triggered netcat hanging in
write(2). This is bit strange as poll(2) should check that the socket
is writeable. However making the network socket non-blocking
keeps the data flowing. Note that the code to handle EAGAIN is
already there.
To trigger
Hello,
I'm sending finalized patch. The final shape has been discussed with bluhm@,
who was kind enough to do thorough testing on OpenBSD.
The patch solves the problem for deployment as follows:
Client -- MTU_1 -- PF -- MTU_1 -- Router -- MTU_2 -- Server
MTU_2 = MTU_1/2
PF does
Date: Wed, 12 Aug 2015 15:48:57 +0200 (CEST)
From: Mark Kettenis mark.kette...@xs4all.nl
I spent some time today figuting out why the binutils update broke ld
-Z on powerpc. Turns out to be a fairly thorny issue.
The new binutils discard empty setions. As a result the .gotpad0 and
Hi,
I start reading your code, and I have a first remark.
I see in main.c (at line 142 and next) that on redirection, you trust
the server for the filename. I am not sure it is a good thing to do.
If the user request 'http://www.example.com/a_filename' (without -o),
the file created should be
26 matches
Mail list logo