soft real-time on linux

Do you have a question? Post it now! No Registration Necessary

Translate This Thread From English to

Threaded View
Hello All,

I have a requirement to port the MSTP physical layer (basically RS-485)
of the BACnet protocol stack to linux. The protocol requires certain
timing to be met, for example, the response to a "Poll For Master" must
be less than 20 milliseconds.

In order to develop this in as short an amount of time, I'd like to use
simple threads and read()/write() commands. However, the standard linux
kernel is not real-time, so any thread could be blocked for an unknown
amount of time. Thus for example a thread that is listening for a "Poll
For Master" packet might not run for more than 20 milliseconds after the
packet was received, thus missing the poll.

But missing a poll is not the end of the world. Eventually (and fairly
quickly) another PFM will be sent out and the device would have another
chance to respond.

Let us also presume such is the case for other response/time violations
throughout the protocol.

The question is, could such an application be written that was "close
enough to real-time" on a standard (non-real-time) kernel?  

I was told by someone on the ##kernel channel that if the kernel is
compiled with "preemptive" enabled, there is a way to set a thread's
priority to "real-time" and it will not only have priority over other
user-land threads but also will preempt much of the kernel's internal
threads. This sounds promising.

If this isn't "good enough" (which is not really well-defined at
this point), there are two alternatives that I can think of:

  1. Switch to the real-time kernel.

  2. Write our own interrupt handler/driver.

I'm not very confident as to which direction to go. Any input would be
appreciated.
--  
Randy Yates, DSP/Embedded Firmware Developer
Digital Signal Labs
We've slightly trimmed the long signature. Click to see the full one.
Re: soft real-time on linux

Quoted text here. Click to load it

In my experience (on a 400MHz ARM9), if you build the kernel with all
the pre-emptable options, and run your task at the highest real-time
priority then latency between an external event that causes an
interrupt and your user task waking up and running is going to be
somwhere in the 100-300 microsecond range.  [Assuming the interrupt
routine and device driver require minimal processing.]

That, of course, depends on the exact set of drivers you have enabled.
All it takes is one long critical section in a poorly written driver
to run that into many hundreds of microseconds.

I would be shocked if it ever got up to 1ms.

Unless you disable the rx FIFO, I would guess that the vast majority
of the latency is going to be in the UART itself.  With the rxFIFO
enabled, there's often a delay of several tens of bit-times between
the last data byte and the receive interrupt.

--  
Grant Edwards               grant.b.edwards        Yow! Zippy's brain cells
                                  at               are straining to bridge
We've slightly trimmed the long signature. Click to see the full one.
Re: soft real-time on linux
On Sat, 12 Nov 2016 08:53:09 -0500, Randy Yates wrote:

Quoted text here. Click to load it


I did something like that.  We  had a microcontroller sending packets of  
a couple of hundred bytes via a USB-serial link every 5ms containing  
state of the input hardware, and accepting similar packets meant to  
control output hardware.  We used straight-up Debian Linux on a VIA Epia  
board (chosen not for real-time ability but for low power consumption,  
reasonable speed, and lots of Ethernet and USB ports.)  Application was  
written in C with one process for each major sub-department, SysV IPC  
shared storage for common working data and configuration blocks, and SysV  
IPC messaging for coordination.  The most urgent processes (serial I/O  
and dispatching for one, and the core of the application for the other)  
ran with an assigned  niceness of -10 to get the dispatches they needed;  
what was left was enough for the other processes, given the processor we  
chose.  Everything marched to the 5ms beat of the input packets.  Testing  
showed all input being processed and responded to in the next output  
packet sent 5ms latter.

I've heard since that they've quickened the drum beat to 2ms, but I  
haven't heard whether they needed to change much code to get that done.


Re: soft real-time on linux
On Sat, 12 Nov 2016 08:53:09 -0500, Randy Yates

Quoted text here. Click to load it

Not very hard requirements for any modern kernel (2.6 and later).  

The only thing that I would warn against is the RS-485 data direction
(Rx/Tx) control.  Do not try to use the RTS pin driven by a user mode
driver. Use proper hardware control for the Tx/Rx switching.

Of course with any half duplex protocols, a large pequest/response
latency will drop the effective throughput well below the theoretical
throughput.


Re: soft real-time on linux
On 11/12/2016 6:53 AM, Randy Yates wrote:
Quoted text here. Click to load it

If you don't want to rely on timeliness guarantees from the software,
then rely on those implicit in the hardware.  I.e., build a layer in
front of the existing RxISR that operates a trivial state machine that
does your PFM recognition.  You *know* that the Rx ISR will be serviced
before N characters have been received, at maximum bit rate (otherwise
characters are dropped and your application has bigger problems to address,
needs faster hardware, a re-tooling of the rest of the system tasks, etc.)
so that gives you an upper bound on the latency incurred by your modified
ISR.  This has the advantage of being somewhat portable (to different
OS's, different versions of the OS, different hardware capabilities, etc.).

And, as you're not doing ALL of your protocol handling/decoding in the
ISR, it's not as brittle as it would be, otherwise.

[It's akin to decoding XON/XOFF characters (a two bit FSM) in the Rx ISR
because the latency involved in doing it in userland would make it
impractical]

Re: soft real-time on linux
Xenomai does pretty much this, and it seems to work fairly well in my exper
ience! It doesn't necessarily come with all the hardware drivers but in the
 worst case you can get away with doing some memmap'd I/O in userland if yo
u don't want to spend the time to write formal device drivers.

On Saturday, November 12, 2016 at 12:53:31 PM UTC-8, Don Y wrote:

Quoted text here. Click to load it


Re: soft real-time on linux
Quoted text here. Click to load it

Unless other processes are contending for the cpu, this shouldn't be too
much of a problem.  On a 300 mhz arm7 board a while back, I remember
timing a thread race and finding it could switch threads at around 20 khz.
It should be simple enough to set up such a test on your own system.

Re: soft real-time on linux
On 11/12/2016 08:53 AM, Randy Yates wrote:
Quoted text here. Click to load it

The Linux thread scheduler is horribly broken, or was last time I used  
it (a while ago, admittedly).  The info pages will tell you all about  
how to specify priorities, scheduling algorithms, and so forth, but  
leave out two vital items:

1. You can only hack those things if you're running as root, and
2. You can't have both real-time and normal threads in the same process.

Both (1) and (2) are easy in Windows (or the late lamented OS/2, which  
is where I learned multithreaded programming, 1992ish).

Note that for my purposes I'd have been totally happy if I could even  
_reduce_ the priority of some threads in my process to make clear which  
ones were less important than others.  Not trying to hog other users'  
gjhresources, no sir.  Still not doable as an ordinary user.

On the other hand, I think you can set processor affinities so as to  
reserve one or more cores for just your own code.  I hope so, because  
I'm looking at a similar requirement myself soon. ;)

Cheers

Phil Hobbs

Re: soft real-time on linux

o:
Quoted text here. Click to load it

the priority of your processes can be modified with "nice".
If you want to change some kernel config of course you need to be root (sam
e thing in any other OS)

Quoted text here. Click to load it

since threads are created/destroyed dinamically, and the OS can't know what
 the fuck is going on in the thread, it's pretty normal that the priority i
s process based and not thread based.

If you want to prioritize a particular thread in a process, you need to do  
it at programming time.

good old stackoverflow: http://stackoverflow.com/questions/3649281/how-to-i
ncrease-thread-priority-in-pthreads

Quoted text here. Click to load it

the (1) is "easy" in win because a normal user is automatically Administrat
or, if aou use a user without admin privileges, well you can change the pri
ority of your processes, but not others.

Bye Jack

Re: soft real-time on linux
On 09/02/17 09:16, Jack wrote:

Quoted text here. Click to load it

And the (2) is "easy" in Windows because windows is poor at process
handling, and compensates by being quite good at thread handling.  Linux
(like other *nix) has always prioritised good process handling - it does
not /need/ such flexibility in thread prioritising, but that is simply
not the way you design programs in Linux.  If you want a program that
involves hard (or as hard as possible) real-time control of a motor,
soft real-time logging of the data, and non-real-time display on a gui,
you organise it differently on Windows and Linux.  With Windows, you put
everything in one process and have separate threads, because passing
data between processes is inefficient.  With Linux, you use three
separate processes with different priorities.



Re: soft real-time on linux
On 02/09/2017 03:16 AM, Jack wrote:

Quoted text here. Click to load it

Only process-by-process.
Quoted text here. Click to load it

Not true in Windows, for instance.  It's a serious handicap in my case,  
where I have compute threads and comms threads in a clusterized  
simulator: another box can totally stall waiting for a comms thread on  
the next box to finish its job.  It trashes the scaling--Windows boxes  
scale nearly linearly up to 25 hosts or so, whereas Linux ones start  
getting less efficient at around 8 hosts.

On a single box, the thread scheduler has a global view of what's going  
on, so it isn't as big a deal.

Quoted text here. Click to load it
Only works as root, and if you want any thread to be real-time, you have  
to put the whole compute-bound process in the RT class, which brings the  
system to its knees.  What I need is a RT comms thread that is usually  
blocked on a semaphore and then runs furiously about 1% of the time.
Quoted text here. Click to load it

But not the relative priority of threads within the process, whereas you  
can in Windows.  As I said originally, I'd be quite happy to be able  
just to _reduce_ the priority of the compute threads vs. the comms  
threads,  but _noooooooo_.  Broken as designed.

Cheers

Phil Hobbs


--  
Dr Philip C D Hobbs
Principal Consultant
We've slightly trimmed the long signature. Click to see the full one.
Re: soft real-time on linux
On Wed, 8 Feb 2017 21:17:35 -0500, Phil Hobbs


Quoted text here. Click to load it

It has improved quite a bit in the recent kernels.  
However ...


Quoted text here. Click to load it

All of this still is true.  

The scheduler historically had problems dealing with overused and
underused cores, and it has been greatly improved of late.  However
control over scheduling has not changed significantly since 2.6.24
(circa 2008).


Quoted text here. Click to load it

Have you tried using clone(2)?
http://man7.org/linux/man-pages/man2/clone.2.html


Clone is a low level call that can create either new processes or new
(kernel) threads in the same process.  For new processes it gives
quite fine control over the execution environment.

You can, e.g., create a new process that *acts* as though it were a
thread: sharing memory (and/or other environment) with the parent, but
having a different priority, scheduling policy, process group, etc.

It is the best Linux can manage.


Quoted text here. Click to load it

You can choose which cores will execute your code, but it is very
difficult to prevent other processes from also using your chosen
cores.

You can try to futz with group scheduling to achieve what you want,
but that is clumsy and difficult because it can involve manipulating
process *owners* as well as parent/child relationships and priorities.


I think the best you can do without inviting a whole lot of headache
is to try to refactor your (POSIX?) threaded application into a set of
clone processes.


Quoted text here. Click to load it

George

Re: soft real-time on linux
On 02/11/2017 01:33 AM, George Neuner wrote:
Quoted text here. Click to load it
=
Quoted text here. Click to load it
Interesting, thanks.  For most of my purposes nowadays I can run the  
simulator on one of my 24-core Opteron servers, running a single kernel  
instance, which greatly reduces the problem.  Back when I had a rack  
full of pizza-box servers, it was ugly.

Cheers

Phil Hobbs

--  
Dr Philip C D Hobbs
Principal Consultant
We've slightly trimmed the long signature. Click to see the full one.

Site Timeline