On Sat, Mar 19, 2016 at 01:53:07PM +0100, Martin Pieuchot wrote:
> Applications using multiple threads often call sched_yield(2) to
> indicate that one of the threads cannot make any progress because
> it is waiting for a resource held by another one.
>
> One example of this scenario is the _spinl
Michal Mazurek wrote:
> On 17:19:39, 2.03.16, Martin Natano wrote:
> > On Wed, Mar 02, 2016 at 05:07:21PM +0100, Michal Mazurek wrote:
> > > kern_sched.c:
> > > - remove unused functions
> > > - mark static functions as static
> >
> > Functions shouldn't be static in the kernel. "In userland,
> >
On Wed, Mar 16, 2016 at 14:42 +0900, Masao Uebayashi wrote:
> This doesn't use atomic operation, because:
>
> - hardclock() is the only writer
> - Clock interrupt doesn't run simultaneously
> - Reading int should be atomic on all architectures
>
Well, in reality both the new location and timeou
Index: cchar.c
===
RCS file: /cvs/src/bin/stty/cchar.c,v
retrieving revision 1.11
diff -u -p -u -r1.11 cchar.c
--- cchar.c27 Oct 2009 23:59:22 -1.11
+++ cchar.c16 Mar 2016 23:53:58 -
@@ -30,6 +30,7 @@
* SUCH DAMA
Hello there,
Here is my patch that adds support for creating IPv6-only or
IPv4-only bridges. This is different from simply blocking one of the
protocols via PF - it allows you to create a setup where IPv4 is routed
and IPv6 is bridged (or vice versa). Both of them being filtered by th
On Sat, Mar 19, 2016 at 09:56:48AM +1100, Jonathan Gray wrote:
> Instead of armv4 tuned for strongarm switch to an armv5te default to
> allow additional instructions to be used (clz etc).
>
> armish and zaurus are both xscale (armv5te with additional extensions).
>
> The builtin gcc defines chang
Hi,
related to past cleanup for some of the sys/arch/x/x/conf.c.
-Artturi
Index: sys/arch/alpha/alpha/autoconf.c
===
RCS file: /cvs/src/sys/arch/alpha/alpha/autoconf.c,v
retrieving revision 1.35
diff -u -p -r1.35 autoconf.c
--- sys
I guess I could have organized alphabetically. Still can if preferred.
Index: funcs.h
===
RCS file: /cvs/src/usr.sbin/cron/funcs.h,v
retrieving revision 1.28
diff -u -p -u -r1.28 funcs.h
--- funcs.h14 Nov 2015 13:09:14 -
On Tue, Mar 15, 2016 at 03:05:47PM +0100, Michal Mazurek wrote:
> On 16:28:33, 14.03.16, Martin Pieuchot wrote:
> > > The number of calls to yield() dropped to 4,576.
> >
> > This is really similar to what I observed with Firefox and Chrome.
> >
> > > This is where I get stuck, I don't know how t
On Sun, Mar 20, 2016 at 01:33:06AM +0100, Alexander Hall wrote:
> Hi,
>
> As far as I can tell, there is no point in having these files lying
> around in /etc as they are not needed by tcsh. They're all comments
> and don't even hold an example.
>
> Thus, remove them and make them optional in /e
On Sat, Mar 19, 2016 at 09:27:19AM +1100, Jonathan Gray wrote:
> On Fri, Mar 18, 2016 at 04:43:08PM +0100, Patrick Wildt wrote:
> > Hi,
> >
> > next up on the list is the StrongARM and IXP12x0.
> >
> > This diff removes most, but keeps the SA1100 headers,
> > which are still used by zaurus.
> >
Hi,
As far as I can tell, there is no point in having these files lying
around in /etc as they are not needed by tcsh. They're all comments
and don't even hold an example.
Thus, remove them and make them optional in /etc/mtree/special.
Objections from tcsh users?
OK?
/Alexander
Index: Makef
While tethering to my phone on a train, I'm seeing a lot of these lines
in dmesg:
urndis0: usb errors on rx: IOERROR
urndis0: usb error on tx: IOERROR
So many in fact that if I run 'dmesg' that's all I see, over and over.
However, network mostly works, and works particularly well given the right
Bob Beck wrote:
> this is cool .. but
>
> I would be interested in someone comparing server workloads, as
> opposed to interactive GUI response, using this.
>
> I wouldn't be surprised that inspiriation from BFS would produce
> better interactive response, my bigger concern would be does this
> a
> p_p->ps_threads (and p_thr_link) can only be modified by threads in
> this process. The result is that the test here can't claim the
> process is single-threaded when it's multi-threaded. (An "is MT"
> result can go stale if there's no locking and the only other thread in
> the process exits,
On Sat, Mar 19, 2016 at 01:53:07PM +0100, Martin Pieuchot wrote:
> Applications using multiple threads often call sched_yield(2) to
> indicate that one of the threads cannot make any progress because
> it is waiting for a resource held by another one.
>
> One example of this scenario is the _spinl
One correction:
Edgar Pettijohn wrote:
> Index: cchar.c
> ===
> RCS file: /cvs/src/bin/stty/cchar.c,v
> retrieving revision 1.11
> diff -u -p -u -r1.11 cchar.c
> --- cchar.c27 Oct 2009 23:59:22 -1.11
> +++ cchar.c16 Ma
The recent uvm commits fixed hangs because machines went out of memory
because of using too much space for amap slots.
It's possible to shrink memory requirements for amaps even more,
but the current code needs some simplifications first. So here's another
cleanup diff. There'll be one or two more
this is cool .. but
I would be interested in someone comparing server workloads, as
opposed to interactive GUI response, using this.
I wouldn't be surprised that inspiriation from BFS would produce
better interactive response, my bigger concern
would be does this adversely impact how we deal with
This is the intended use, avoid busy-polling of BIOS PC console if
running on HV. Avoid 100% CPU usage at boot prompt on hypervisors.
Originally I unconditionally reverted the polling part, which was
introduced only for Intel Mac Mini. After brief discussion with mikeb@,
I decided to leave that
Hi,
To prevent attacks on the hash buckets of the syn cache, our TCP
stack reseeds the hash function every time the cache is empty.
Unfortunatly the attacker can prevent the reseeding by sending
unanswered SYN packes periodically.
I fix this by having an active syn cache that gets new entries and
I'm also interested in hearing from more people using multi-threaded
applications.
I applied the patch to my old core duo p7570 running CURRENT/amd64.
Firefox is more responsive and youtube videos, previously impossible to
watch, run smoothly, even in full screen. Sumultaneously did some phot
On Fri, Mar 18, 2016 at 01:02:43PM +0100, m...@openbsd.org wrote:
>
> Could you test the diff below and tell me if it also helps? It is
> adapted from a submission from Dominic Marks:
> https://marc.info/?l=openbsd-ppc&m=142836014007157&w=2
Indeed it does more or less the same thing.
So,
On Thu, Mar 17, 2016 at 03:31:45PM +0900, Masao Uebayashi wrote:
> On Wed, Mar 16, 2016 at 11:26:39PM -0700, Mike Larkin wrote:
> > On Thu, Mar 17, 2016 at 03:15:07PM +0900, Masao Uebayashi wrote:
> > > This is the intended use, avoid busy-polling of BIOS PC console if
> > > running on HV. Avoid 1
On Wed, Mar 16, 2016 at 11:49:32PM -0700, Mike Larkin wrote:
> On Thu, Mar 17, 2016 at 03:38:20PM +0900, Masao Uebayashi wrote:
> > On Wed, Mar 16, 2016 at 11:25:11PM -0700, Mike Larkin wrote:
> > > On Thu, Mar 17, 2016 at 09:44:22AM +0900, Masao Uebayashi wrote:
> > > > Implement delay() using TSC
Implement delay() using TSC
- Calculate delay using 64-bit RDTSC instruction
- Enable tsc_delay() as delay(9) backend
- Use tsc_delay() only when TSC is invariant
- Configure tsc_delay() after primary CPU is identified
- Assume all CPUs are identical
diff --git a/sys/arch/amd64/amd64/cpu.c b/sys/
I've recently acquired a usb3.0->gigabit ethernet adapter. It did not
attach reliably, pass traffic reliably, and it made my machine panic
when I unplugged it. Takahiro HAYASHI suggested that the reset
code doesn't do anything and that it should initialize the chip. Indeed
that does seem to help. I
On Thu, Mar 17, 2016 at 03:38:20PM +0900, Masao Uebayashi wrote:
> On Wed, Mar 16, 2016 at 11:25:11PM -0700, Mike Larkin wrote:
> > On Thu, Mar 17, 2016 at 09:44:22AM +0900, Masao Uebayashi wrote:
> > > Implement delay() using TSC
> > >
> > > - Calculate delay using 64-bit RDTSC instruction
> > >
s/approves/improves/g
``ticks'' is in my opinion a really badly named global variable. But we
all know that finding names is hard ;)
Since its popularity seems to be really high lately, I'd suggest to
rename it. I chose ``hcticks'' for "hardclock ticks". I believe this
would help auditing the possible existing and/o
On Thu, Mar 17, 2016 at 04:03:25PM +0900, Masao Uebayashi wrote:
> On Wed, Mar 16, 2016 at 11:49:32PM -0700, Mike Larkin wrote:
> > On Thu, Mar 17, 2016 at 03:38:20PM +0900, Masao Uebayashi wrote:
> > > On Wed, Mar 16, 2016 at 11:25:11PM -0700, Mike Larkin wrote:
> > > > On Thu, Mar 17, 2016 at 09:
Michal Mazurek [akf...@jasminek.net] wrote:
> On 16:28:33, 14.03.16, Martin Pieuchot wrote:
> > > The number of calls to yield() dropped to 4,576.
> >
> > This is really similar to what I observed with Firefox and Chrome.
> >
> > > This is where I get stuck, I don't know how to replace the call t
I've had this sitting around for a while and I figure it's time to get it in
the tree. Octeon II sensibly got rid of the DWC USB controller and put in ehci
and ohci compliant controllers instead, so all we need to do here is set up a
few registers in the usb controller interface and attach the sta
The java based Netbeans IDE performs a lot of things in the background
using threads like indexing sources, downloading artifacts, unpacking
archives, compiling sources in addition to the normal java background
tasks like JIT compiling. This change approves UI responsiveness on my T60.
ok mmcc@
Edgar Pettijohn wrote:
> Index: cchar.c
> ===
> RCS file: /cvs/src/bin/stty/cchar.c,v
> retrieving revision 1.11
> diff -u -p -u -r1.11 cchar.c
> --- cchar.c27 Oct 2009 23:59:22 -1.11
> +++ cchar.c16 Mar 2016
On Sat, Mar 19, 2016 at 01:53:07PM +0100, Martin Pieuchot wrote:
> I experimented with various values for "p_priority" and this one is
> the one that generates fewer # IPIs when watching a HD video on firefox.
> Because yes, with this diff, now I can.
YouTube on Firefox on ThinkPad T420: It's def
ok mmcc@
Edgar Pettijohn wrote:
> Index: buf.c
> ===
> RCS file: /cvs/src/bin/ed/buf.c,v
> retrieving revision 1.22
> diff -u -p -u -r1.22 buf.c
> --- buf.c9 Oct 2015 19:47:02 -1.22
> +++ buf.c16 Mar 2016 23:23:53 -000
This tiny driver is only meant for a backup watchdog configuration on
QEMU, whose ICH watchdog, supported by ichwdt(4), seems broken.
ib700wdt(4) supports only 0-30 second timeouts but in practice it should
be good enough.
diff --git a/sys/arch/amd64/conf/GENERIC b/sys/arch/amd64/conf/GENERIC
ind
Factor out CPUID code in mdrandom(), invoke it once, then save the
result. I'll use it to switch behavior depending on HV or not.
efiboot is not tested.
Comments?
>From 104cb04bbbd0f7e40758938cc3103b2370a2285c Mon Sep 17 00:00:00 2001
From: Masao Uebayashi
Date: Thu, 10 Mar 2016 21:03:07 +0900
Bob Beck [b...@openbsd.org] wrote:
> this is cool .. but
>
> I would be interested in someone comparing server workloads, as
> opposed to interactive GUI response, using this.
>
> I wouldn't be surprised that inspiriation from BFS would produce
> better interactive response, my bigger concern
> w
On Tue, Mar 15, 2016 at 03:05:47PM +0100, Michal Mazurek wrote:
>
> Please test, and let me know if the performance of something else
> degrades.
>
With your diff firefox consumes twice less cpu (watched the same
video with and without the diff). This suggests firefox spins
somewhere and your d
On Fri, Mar 18, 2016 at 08:18:30PM +0100, Mark Kettenis wrote:
> > Date: Thu, 17 Mar 2016 14:28:04 +0900
> > From: Masao Uebayashi
> >
> > This tiny driver is only meant for a backup watchdog configuration on
> > QEMU, whose ICH watchdog, supported by ichwdt(4), seems broken.
> >
> > ib700wdt(4)
On 17/03/16(Thu) 14:28, Masao Uebayashi wrote:
> This tiny driver is only meant for a backup watchdog configuration on
> QEMU, whose ICH watchdog, supported by ichwdt(4), seems broken.
What's the problem? Is it in our code or in QEMU?
Edd Barrett wrote:
> On Sat, Mar 19, 2016 at 09:06:26AM +0100, Alexandre Ratchov wrote:
> > The browsers problems seem caused by the way pthreads behave;
> > browsers appear to spin. With the proposed scheduler they spin
> > less. But the real question is why they spin at all?
>
> Inspired by th
The compiler is also smart enough to recognize that this is redundant.
The resulting code on amd64 is basically equivalent (slightly different
register allocation and instruction scheduling).
ok?
Index: uvm/uvm_amap.c
===
RCS file: /
On Thu, Mar 17, 2016 at 03:15:07PM +0900, Masao Uebayashi wrote:
> This is the intended use, avoid busy-polling of BIOS PC console if
> running on HV. Avoid 100% CPU usage at boot prompt on hypervisors.
>
Which hypervisors have you tested this on?
-ml
> Originally I unconditionally reverted th
On Fri, Mar 18, 2016 at 08:44:11AM +0100, Marcus MERIGHI wrote:
> Hello,
>
> "The patch seems to work fine on an EdgeRouter Pro.
> OK visa@"
> (http://marc.info/?l=openbsd-tech&m=145822792814191)
>
> EdgeRouter Pro is not on octeon.html, put it there?
>
> Bye, Marcus
Last time I checked the ethe
A recent diff from dlg@ made me looks at where the global variable
``ticks'' is used and how. I find this *really* difficult because
there's a lot of noise. In order to reduce the noise and help for
this audit here's a first diff. It renames all the local variables
from ``ticks'' to ``nticks'' wh
On Sun, Mar 20, 2016 at 03:36:45AM +1100, Jonathan Gray wrote:
> On Wed, Mar 02, 2016 at 03:31:57PM +0100, Patrick Wildt wrote:
> > Hi,
> >
> > currently we have a rather big list of bits that we remove and set in
> > the System Control Register. Those differ per processor, so not every
> > armv7
It isn't possible to leave all of the existing code intact,
at least not for touchpads, which form the most complex and
critical case here. Either all of them run with the modified
synaptics driver, or they don't. I don't see a reasonable
solution in between (wsconscomm.c might become a little mon
On Sat, Mar 19, 2016 at 9:18 AM, Miod Vallat wrote:
>
>> I'd like to know if dereferencing ``p_p'' is safe without holding the
>> KERNEL_LOCK.
>
> SCHED_LOCK is enough to access p_p->ps_threads,
Uh, not quite.
p_p itself is immutable for the life of a thread and can be
dereferenced without locks
On Wed, Mar 16, 2016 at 23:47 -0700, Mike Larkin wrote:
> On Thu, Mar 17, 2016 at 03:31:45PM +0900, Masao Uebayashi wrote:
> > On Wed, Mar 16, 2016 at 11:26:39PM -0700, Mike Larkin wrote:
> > > On Thu, Mar 17, 2016 at 03:15:07PM +0900, Masao Uebayashi wrote:
> > > > This is the intended use, avoid
Hi Michal,
On Fri Mar 18 2016 10:03, Edd Barrett wrote:
> On Thu, Mar 17, 2016 at 09:26:08PM +0100, Henrik Friedrichsen wrote:
> > On Tue, Mar 15, 2016 at 03:05:47PM +0100, Michal Mazurek wrote:
> > > Chrome still isn't smooth.
> > >
> > > Please test, and let me know if the performance of someth
What's the plan regarding this? libc and locale(1) don't use
LOCALE_CHARSETS anymore, we only support UTF-8 and I doubt that we'll
change our mind about that.
The diff below removes support for non-UTF-8 LC_CTYPE files. It
survived make release.
Thoughts?
Index: lib/libc/locale/Makefile.inc
=
On Fri, Mar 18, 2016 at 08:31:36AM -0600, Bob Beck wrote:
> > But it officially requires support for IPP version 1.0, which used
> > SSLv3.
> > I assume that there are printers, perhaps many were sold, which did use
> > version 1.0. That version used SSLv3 for encrypted communication. Which
> > is
On Wed, Mar 02, 2016 at 03:31:57PM +0100, Patrick Wildt wrote:
> Hi,
>
> currently we have a rather big list of bits that we remove and set in
> the System Control Register. Those differ per processor, so not every
> armv7/armv8 processor has the same bits, although they tend to be
> rather simil
> I'd like to know if dereferencing ``p_p'' is safe without holding the
> KERNEL_LOCK.
SCHED_LOCK is enough to access p_p->ps_threads,
The construct used to decide whether the process is multi-threaded
already appears twice in sys/kern¸ and your diff adds a third
instance. It is probably worth t
On 18/03/2016 12:02, Martin Pieuchot wrote:
Could you test the diff below and tell me if it also helps? It is
adapted from a submission from Dominic Marks:
https://marc.info/?l=openbsd-ppc&m=142836014007157&w=2
It definitely helps here.
As the kernel works its way through probing vario
http://ftp.openbsd.org/pub/OpenBSD/patches/5.9/common/002_in6bind.patch.sig
002: SECURITY FIX: March 16, 2016 All architectures
Insufficient checks in IPv6 socket binding and UDP IPv6 option processing allow
a local user to send UDP packets with a source (IPv6 address + port) already
reserved
> But it officially requires support for IPP version 1.0, which used
> SSLv3.
> I assume that there are printers, perhaps many were sold, which did use
> version 1.0. That version used SSLv3 for encrypted communication. Which
> is now gone.
Almost certainly.
>
> How should we deal with this probl
OK uebayasi@
On Sat, Mar 19, 2016 at 02:51:56PM +, Visa Hankala wrote:
> dwctwo(4) should be attached only on those Octeon models that have the
> DWC2 controller. The list below is not complete but it covers every
> model that has any chance of working with cnmac(4).
>
> OK?
>
> Index: arch/
Fri, 18 Mar 2016 13:04:49 -0600 Theo de Raadt
> > So I'm skeptical about this BFS scheduler. Don't get me wrong; I do
> > think the scheduler needs attention. But I'm not sure a scheduler
> > designed for interactive desktop use is the best option for OpenBSD.
> > I'm willing to be proven wrong.
dwctwo(4) should be attached only on those Octeon models that have the
DWC2 controller. The list below is not complete but it covers every
model that has any chance of working with cnmac(4).
OK?
Index: arch/octeon/dev/octdwctwo.c
===
On Sat, Mar 19, 2016 at 01:53:07PM +0100, Martin Pieuchot wrote:
> I experimented with various values for "p_priority" and this one is
> the one that generates fewer # IPIs when watching a HD video on firefox.
> Because yes, with this diff, now I can.
Same result here: Video in firefox plays nice
The common part of rtwn(4) needs help from attachment drivers to
determine the chip type. The PCI/USB driver parts must detect the
chip type based on PCI/USB vendor and device IDs. There seems to be no
other way of doing this. Even the Linux vendor driver does it this way.
The common part is still
On Sat, Mar 19, 2016 at 01:53:07PM +0100, Martin Pieuchot wrote:
> I experimented with various values for "p_priority" and this one is
> the one that generates fewer # IPIs when watching a HD video on firefox.
> Because yes, with this diff, now I can.
Works well here in firefox too.
Will run thi
A sparc64 machine of mine crashes every few weeks because the kernel
runs out of memory.
The crash usually happens in a malloc call in the vnet(4) ioctl code path.
I have a cronjob which downs/up vnet interfaces in case they get stuck in
OACTIVE (another bug which, AFAIK, is not solved yet). So th
> So I'm skeptical about this BFS scheduler. Don't get me wrong; I do
> think the scheduler needs attention. But I'm not sure a scheduler
> designed for interactive desktop use is the best option for OpenBSD.
> I'm willing to be proven wrong. But that means we need careful
> benchmarking on a wi
On Thu, Mar 17, 2016 at 09:26:08PM +0100, Henrik Friedrichsen wrote:
> On Tue, Mar 15, 2016 at 03:05:47PM +0100, Michal Mazurek wrote:
> > Chrome still isn't smooth.
> >
> > Please test, and let me know if the performance of something else
> > degrades.
>
> While Chrome may not be 100% smooth yet
> Factor out CPUID code in mdrandom(), invoke it once, then save the
> result. I'll use it to switch behavior depending on HV or not.
>
> efiboot is not tested.
>
> Comments?
I don't understand the purpose of the refactoring.
Applications using multiple threads often call sched_yield(2) to
indicate that one of the threads cannot make any progress because
it is waiting for a resource held by another one.
One example of this scenario is the _spinlock() implementation of
our librthread. But if you look on https://codesea
On Sat, Mar 19, 2016 at 09:06:26AM +0100, Alexandre Ratchov wrote:
> The browsers problems seem caused by the way pthreads behave;
> browsers appear to spin. With the proposed scheduler they spin
> less. But the real question is why they spin at all?
Inspired by this comment, I set out to see if
On Thu, Mar 17, 2016 at 11:51:19AM +1000, Jonathan Matthew wrote:
> I've had this sitting around for a while and I figure it's time to get it in
> the tree. Octeon II sensibly got rid of the DWC USB controller and put in
> ehci
> and ohci compliant controllers instead, so all we need to do here i
> Date: Thu, 17 Mar 2016 21:02:03 +0100
> From: Martin Pieuchot
>
> ``ticks'' is in my opinion a really badly named global variable. But we
> all know that finding names is hard ;)
>
> Since its popularity seems to be really high lately, I'd suggest to
> rename it. I chose ``hcticks'' for "har
The process I see here the submission of 2 seperate diffs to
the bootblocks, requiring people to take two steps.
Then readers of tech@ will realize that the proposition is to
test their bootblocks right away, many will balk.
The diffs also do not come with a report as to what machines were
te
On 31/01/16(Sun) 23:47, Olivier Cherrier wrote:
> Hi,
>
> It may not be the proper solution to control the fans on macppc smu (G5)
> but the following patch allows me to work with my iMac G5 without
> losing my hearing capacity ...
>
>
>
> $ sysctl hw.sensors
> hw.sensors.smu0.temp0=44.72
On Wed, Mar 16, 2016 at 22:18 -0400, Michael McConville wrote:
> Michal Mazurek wrote:
> > On 17:19:39, 2.03.16, Martin Natano wrote:
> > > On Wed, Mar 02, 2016 at 05:07:21PM +0100, Michal Mazurek wrote:
> > > > kern_sched.c:
> > > > - remove unused functions
> > > > - mark static functions as sta
Index: alloc.c
===
RCS file: /cvs/src/bin/csh/alloc.c,v
retrieving revision 1.17
diff -u -p -u -r1.17 alloc.c
--- alloc.c26 Dec 2015 13:48:38 -1.17
+++ alloc.c16 Mar 2016 22:47:56 -
@@ -30,12 +30,18 @@
* SUCH DAM
On 18 March 2016 at 18:02, Michal Mazurek wrote:
> On 09:22:18, 18.03.16, Chris Cappuccio wrote:
>> These are really rough observations. This box spawns lots of dirty
>> perl processes and also lots of fping processes for monitoring.
>
> The next step I had planned was related to juggling processe
> Date: Wed, 16 Mar 2016 22:18:41 -0400
> From: Michael McConville
>
> Michal Mazurek wrote:
> > On 17:19:39, 2.03.16, Martin Natano wrote:
> > > On Wed, Mar 02, 2016 at 05:07:21PM +0100, Michal Mazurek wrote:
> > > > kern_sched.c:
> > > > - remove unused functions
> > > > - mark static function
On 16/03/16(Wed) 20:54, Momtchil Momtchev wrote:
> [..]
> Here is my patch that adds support for creating IPv6-only or IPv4-only
> bridges. This is different from simply blocking one of the protocols via PF
> - it allows you to create a setup where IPv4 is routed and IPv6 is bridged
> (or vice
On Fri, Mar 18, 2016 at 6:02 PM, Michal Mazurek wrote:
> BFS has one shared queue for all CPUs, maybe there is a very good reason
> for that, we'll see.
Michal,
first of all congrats to optimistic results in interactive workloads.
Honestly I'm a little bit worried about your attempts since I thi
There are whitespace errors in the diffs. In case anyone
wants to test them, please use the new versions in the
attachment.
On 03/13/2016 06:37 PM, Ulf Brosziewski wrote:
The diffs below are a rewrite of the input-processing part of wsmouse. It
adds support for multitouch input.
[...]
mtsuppo
Hello,
Recently I bought a cheap PCIe UART adapter.
It uses ASMedia ASM1083 PCIe-PCI bridge and TXIC TX382B UART controller.
There is no details/datasheets about TX382B, but a Chinese seller
saids that it is similar to WCH351Q (aka WinChipHead's CH351Q). [1]
So I add new entry to pucdata.c like C
On 09:22:18, 18.03.16, Chris Cappuccio wrote:
> These are really rough observations. This box spawns lots of dirty
> perl processes and also lots of fping processes for monitoring.
The next step I had planned was related to juggling processes between
cpus. Right now that code is untouched, other t
Index: buf.c
===
RCS file: /cvs/src/bin/ed/buf.c,v
retrieving revision 1.22
diff -u -p -u -r1.22 buf.c
--- buf.c9 Oct 2015 19:47:02 -1.22
+++ buf.c16 Mar 2016 23:23:53 -
@@ -30,6 +30,15 @@
*/
#include
+#includ
On 18/03/16(Fri) 13:04, Theo de Raadt wrote:
> > So I'm skeptical about this BFS scheduler. Don't get me wrong; I do
> > think the scheduler needs attention. But I'm not sure a scheduler
> > designed for interactive desktop use is the best option for OpenBSD.
> > I'm willing to be proven wrong.
On Thu, Mar 17, 2016 at 11:43:03AM +0900, Masao Uebayashi wrote:
> A bit surprised that code does more than I imagined by reading your
> description :) but it looks very good to me!
>
> A few comments:
>
> - `struct octuctl_softc' is not using `sc_ehci' and `sc_ohci'?
These were leftovers from w
A bit surprised that code does more than I imagined by reading your
description :) but it looks very good to me!
A few comments:
- `struct octuctl_softc' is not using `sc_ehci' and `sc_ohci'?
- Some magic numbers in `octuctl_clock_setup' could be #define'ed
I've just bought EdgeRouter Pro. Hope
Hello,
The diff is committed (ok by deraadt@), and I am testing the behavior of
TX382B's FIFO.
> Here is a dmesg.
>
> ppb1 at pci1 dev 0 function 0 "ASMedia ASM1083/1085 PCIE-PCI" rev 0x04
> pci2 at ppb1 bus 2
> puc0 at pci2 dev 0 function 0 "TXIC TX382B" rev 0x10: ports: 2 com
> com4 at
Apologies for my disappearance for a while but I have been brutally sick.
Fine now.
IPP is now using version 1.1 and new products also.
But it officially requires support for IPP version 1.0, which used
SSLv3.
I assume that there are printers, perhaps many were sold, which did use
version 1.0. Th
On Thu, Mar 17, 2016 at 09:02:03PM +0100, Martin Pieuchot wrote:
> ``ticks'' is in my opinion a really badly named global variable. But we
> all know that finding names is hard ;)
>
> Since its popularity seems to be really high lately, I'd suggest to
> rename it. I chose ``hcticks'' for "hardcl
Hello,
"The patch seems to work fine on an EdgeRouter Pro.
OK visa@"
(http://marc.info/?l=openbsd-tech&m=145822792814191)
EdgeRouter Pro is not on octeon.html, put it there?
Bye, Marcus
Index: octeon.html
===
RCS file: /cvs/www/oct
On Fri, Mar 18, 2016 at 08:00:35PM +0100, Mark Kettenis wrote:
> > From: Bob Beck
> > Date: Fri, 18 Mar 2016 09:20:35 -0600
> >
> > this is cool .. but
> >
> > I would be interested in someone comparing server workloads, as
> > opposed to interactive GUI response, using this.
> >
> > I wouldn't
On Thu, Mar 17, 2016 at 09:44:22AM +0900, Masao Uebayashi wrote:
> Implement delay() using TSC
>
> - Calculate delay using 64-bit RDTSC instruction
> - Enable tsc_delay() as delay(9) backend
> - Use tsc_delay() only when TSC is invariant
> - Configure tsc_delay() after primary CPU is identified
>
With this diff on my X200, YouTube used to be a stuttering mess on both
chrome and firefox, and is now buttery smooth, even at 1080p. Thanks!
On Wed, Mar 16, 2016 at 11:26:39PM -0700, Mike Larkin wrote:
> On Thu, Mar 17, 2016 at 03:15:07PM +0900, Masao Uebayashi wrote:
> > This is the intended use, avoid busy-polling of BIOS PC console if
> > running on HV. Avoid 100% CPU usage at boot prompt on hypervisors.
> >
>
> Which hypervisors
Hey,
On Tue, Mar 15, 2016 at 03:05:47PM +0100, Michal Mazurek wrote:
> Chrome still isn't smooth.
>
> Please test, and let me know if the performance of something else
> degrades.
While Chrome may not be 100% smooth yet, the system is a lot more
interactive. I can now play YouTube videos without
Hi,
based on the SA1 diff, this change also removes IXP425.
Patrick
diff --git sys/arch/arm/arm/cpu.c sys/arch/arm/arm/cpu.c
index 593bdac..5c7e385 100644
--- sys/arch/arm/arm/cpu.c
+++ sys/arch/arm/arm/cpu.c
@@ -146,13 +146,6 @@ static const char * const pxa27x_steppings[16] = {
"rev 12
Octeon firmware provides a list of usable memory regions. I would like
to make the memory setup routine use the list instead of hardcoded
assumptions. EdgeRouter Pro has a memory layout that does not match
with the current code, which causes a kernel crash.
I have tested this patch on EdgeRouter L
1 - 100 of 102 matches
Mail list logo
