Am 31.01.2014 um 15:24 schrieb Kenneth Lerman :
> On 1/31/2014 4:53 AM, Michael Haberler wrote:
>> Am 30.01.2014 um 21:47 schrieb Charles Steinkuehler
>> :
>>
>>> On 1/30/2014 2:14 PM, Michael Haberler wrote:
Once ethernet level packet I/O is settled as fast enough, we can turn
to add
On 1/31/2014 4:53 AM, Michael Haberler wrote:
> Am 30.01.2014 um 21:47 schrieb Charles Steinkuehler
> :
>
>> On 1/30/2014 2:14 PM, Michael Haberler wrote:
>>> Once ethernet level packet I/O is settled as fast enough, we can turn
>>> to add the minimal IP/UDP framing needed. This will be bare bones
Am 30.01.2014 um 21:47 schrieb Charles Steinkuehler :
> On 1/30/2014 2:14 PM, Michael Haberler wrote:
>>
>> Once ethernet level packet I/O is settled as fast enough, we can turn
>> to add the minimal IP/UDP framing needed. This will be bare bones (no
>> ping, arp, and other amenenities) but shou
I have been helping test the mesa 7i80 (ethernet). I started with the
RTnet setup but what a pain. Supported nics are very limited. When
micges switched to rt_preempt there was a watchdog bug that bit the 7i80
(pun intended..) but through all the testing I don't think I once got a
realtime d
On 1/30/2014 2:14 PM, Michael Haberler wrote:
>
> Once ethernet level packet I/O is settled as fast enough, we can turn
> to add the minimal IP/UDP framing needed. This will be bare bones (no
> ping, arp, and other amenenities) but should have pretty much the
> same timing behavior and good enough
Am 30.01.2014 um 14:17 schrieb Michał Geszkiewicz :
> So now is good time for me to follow MHaberler searching of best udp
> stack or layer or library to decrease this terrible overhead while
> reading data from 7i80. It took about 400 to 700us on each 1ms thread entry.
well these figures sugg
W dniu 30.01.2014 16:38, andy pugh pisze:
> On 30 January 2014 13:17, Michał Geszkiewicz wrote:
>
>> Peter has driver
>> working for 3 weeks 24/7, Skunkworks can flashing leds and I can use
>> sserial slave cards without breaking connection.
> As a rather more extreme test, I have used it software
On 30 January 2014 13:17, Michał Geszkiewicz wrote:
> Peter has driver
> working for 3 weeks 24/7, Skunkworks can flashing leds and I can use
> sserial slave cards without breaking connection.
As a rather more extreme test, I have used it software-commutation of
a brushless motor, with no appare
Hi,
Basic idea of making hm2_eth driver under rt-preempt was to get basic
architecture up and running, there was to much 'I don't know, must test
it' areas.
I started to use rtnet becouse it is realtime and it is supported in
RTAI and Xenomai. Unfortunately specific driver architecture (part of
Am 30.01.2014 um 12:55 schrieb Lars Segerlund :
> 2014-01-30 Michael Haberler :
>>
>> Am 30.01.2014 um 11:42 schrieb Lars Segerlund :
>>
>>> It depends on the driver implementation, ie. if the kernel code is
>>> using preempt disable or disable interrupts.
>>
>> The kernel execution path is _m
2014-01-30 Michael Haberler :
>
> Am 30.01.2014 um 11:42 schrieb Lars Segerlund :
>
>> It depends on the driver implementation, ie. if the kernel code is
>> using preempt disable or disable interrupts.
>
> The kernel execution path is _much_ longer than just the driver, and
> depending on the syst
Am 30.01.2014 um 11:42 schrieb Lars Segerlund :
> It depends on the driver implementation, ie. if the kernel code is
> using preempt disable or disable interrupts.
The kernel execution path is _much_ longer than just the driver, and depending
on the system call all sorts of issues happen well b
It depends on the driver implementation, ie. if the kernel code is
using preempt disable or disable interrupts.
ALL preempt-rt performance is to a large extent dependent on the
quality of the device drivers.
If you look at osadl.org there is an effort on realtime device drivers ...
/ Lars
Michal is trying hard to get the 7i80/hm2_eth.c driver working for RT-PREEMPT
using normal socket I/O from an RT thread.
The hopes with RT-PREEMPT are obviously pegged on the assumption: 'the kernel
is hardened, so I'm free to use any system calls from an RT thread and still
get decent latency,
14 matches
Mail list logo
