LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
* Re: Running Linux on FPGA
       [not found] <20070120234237.49126.qmail@web7912.mail.in.yahoo.com>
@ 2007-01-21  0:14 ` Ralf Baechle
  2007-01-21  9:59   ` Sunil Naidu
  2007-01-21 21:31   ` Jan Engelhardt
  0 siblings, 2 replies; 5+ messages in thread
From: Ralf Baechle @ 2007-01-21  0:14 UTC (permalink / raw)
  To: sathesh babu; +Cc: linux-mips, linux-kernel

On Sat, Jan 20, 2007 at 11:42:37PM +0000, sathesh babu wrote:

> Hi,
>   I am trying to run Linux-2.6.18.2 ( with preemption enable) kernel on FPGA board which has MIPS24KE processor runs at 12 MHZ. Programmed the timer to give interrupt at every 10msec.
>   I am seeing some inconsistence behavior during boot up processor. Some times it stops after "NET: Registered protocol family 17" and "VFS: Mounted root (jffs2 filesystem).".
>   Could some give some pointers why the behavior is random.
>   Is it OK to program the timer to 10 msec? or should it be more.

The overhead of timer interrupts at this low clockrate is significant
so I recommend to minimize the timer interrupt rate as far as possible.
This is really a tradeoff between latency and overhead and matters
much less on hardcores which run at hundreds of MHz.  For power sensitive
applications lowering the interrupt rate can also help.  And that's alredy
pretty much what you need to know, that is a 10ms  timer is fine.

Btw, is this coincidentally on a CoreFPGA 2 or 3 CPU card on a Malta board?

  Ralf

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Running Linux on FPGA
  2007-01-21  0:14 ` Running Linux on FPGA Ralf Baechle
@ 2007-01-21  9:59   ` Sunil Naidu
  2007-01-21 11:22     ` Kevin D. Kissell
  2007-01-21 21:31   ` Jan Engelhardt
  1 sibling, 1 reply; 5+ messages in thread
From: Sunil Naidu @ 2007-01-21  9:59 UTC (permalink / raw)
  To: Ralf Baechle; +Cc: linux-kernel, sathesh babu, linux-mips

On 1/21/07, Ralf Baechle <ralf@linux-mips.org> wrote:
>
> The overhead of timer interrupts at this low clockrate is significant
> so I recommend to minimize the timer interrupt rate as far as possible.
> This is really a tradeoff between latency and overhead and matters
> much less on hardcores which run at hundreds of MHz.  For power sensitive
> applications lowering the interrupt rate can also help.  And that's alredy
> pretty much what you need to know, that is a 10ms  timer is fine.
>

I have worked with FPGA Linux system which is reconfigurable
on-the-fly by the 200Mhz ARM9 CPU running Debian Linux, Altera Cyclone
II FPGA is included on my TS-7300 board. Advantage is, Altera FPGA and
a dedicated high-speed bus between the CPU and FPGA provides a good
design scope to provide many solutions.

Coming to boot up (by an USB 1GB SD card), by doing enough software
tuning bootup to a Linux prompt takes just 1.69 seconds after
power-up. If I remember correctly, SD image will look at the state of
jumper 6 (should be put ON), the full Debian bootup will be bypassed
and the system will instead drop straight to a shell prompt. 1.69
seconds after power-on the serial console prompt is active and 2.41
seconds after power-on the video console is displayed.

This software is based on Debian & has a vendor supplied Linux boot
loader. Currently am working (slowly in free time) to bring the whole
thing to FC6.  Shall post the progres...

>
>   Ralf

~Akula2

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Running Linux on FPGA
  2007-01-21  9:59   ` Sunil Naidu
@ 2007-01-21 11:22     ` Kevin D. Kissell
  0 siblings, 0 replies; 5+ messages in thread
From: Kevin D. Kissell @ 2007-01-21 11:22 UTC (permalink / raw)
  To: Sunil Naidu, Ralf Baechle; +Cc: linux-kernel, sathesh babu, linux-mips

> > The overhead of timer interrupts at this low clockrate is significant
> > so I recommend to minimize the timer interrupt rate as far as possible.
> > This is really a tradeoff between latency and overhead and matters
> > much less on hardcores which run at hundreds of MHz.  For power sensitive
> > applications lowering the interrupt rate can also help.  And that's alredy
> > pretty much what you need to know, that is a 10ms  timer is fine.
> >
> 
> I have worked with FPGA Linux system which is reconfigurable
> on-the-fly by the 200Mhz ARM9 CPU running Debian Linux, Altera Cyclone
> II FPGA is included on my TS-7300 board. Advantage is, Altera FPGA and
> a dedicated high-speed bus between the CPU and FPGA provides a good
> design scope to provide many solutions.

What's your point here?  A 200MHz hard ore won't see the issues 
under discussion.  We're talking about systems where the CPU itself
is "soft" and implemented in an FPGA.

            Regards,

            Kevin K.


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Running Linux on FPGA
  2007-01-21  0:14 ` Running Linux on FPGA Ralf Baechle
  2007-01-21  9:59   ` Sunil Naidu
@ 2007-01-21 21:31   ` Jan Engelhardt
  2007-01-22 16:58     ` Kevin D. Kissell
  1 sibling, 1 reply; 5+ messages in thread
From: Jan Engelhardt @ 2007-01-21 21:31 UTC (permalink / raw)
  To: Ralf Baechle; +Cc: sathesh babu, linux-mips, linux-kernel


On Jan 21 2007 00:14, Ralf Baechle wrote:
>On Sat, Jan 20, 2007 at 11:42:37PM +0000, sathesh babu wrote:
>
>>   I am trying to run Linux-2.6.18.2 ( with preemption enable)
>>   kernel on FPGA board which has MIPS24KE processor runs at 12
>>   MHZ. Programmed the timer to give interrupt at every 10msec. I
>>   am seeing some inconsistence behavior during boot up processor.
>>   Some times it stops after "NET: Registered protocol family 17"
>>   and "VFS: Mounted root (jffs2 filesystem).". Could some give
>>   some pointers why the behavior is random. Is it OK to program
>>   the timer to 10 msec? or should it be more.
>
>The overhead of timer interrupts at this low clockrate is
>significant so I recommend to minimize the timer interrupt rate as
>far as possible. This is really a tradeoff between latency and
>overhead and matters much less on hardcores which run at hundreds of
>MHz.

Hm I've been running 2.6.13 on a 10/20 MHz (switchable) i386 @ 100 Hz
before without any hangs during boot or operation.


	-`J'
-- 

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Running Linux on FPGA
  2007-01-21 21:31   ` Jan Engelhardt
@ 2007-01-22 16:58     ` Kevin D. Kissell
  0 siblings, 0 replies; 5+ messages in thread
From: Kevin D. Kissell @ 2007-01-22 16:58 UTC (permalink / raw)
  To: Jan Engelhardt, Ralf Baechle; +Cc: sathesh babu, linux-mips, linux-kernel

> On Jan 21 2007 00:14, Ralf Baechle wrote:
> >On Sat, Jan 20, 2007 at 11:42:37PM +0000, sathesh babu wrote:
> >
> >>   I am trying to run Linux-2.6.18.2 ( with preemption enable)
> >>   kernel on FPGA board which has MIPS24KE processor runs at 12
> >>   MHZ. Programmed the timer to give interrupt at every 10msec. I
> >>   am seeing some inconsistence behavior during boot up processor.
> >>   Some times it stops after "NET: Registered protocol family 17"
> >>   and "VFS: Mounted root (jffs2 filesystem).". Could some give
> >>   some pointers why the behavior is random. Is it OK to program
> >>   the timer to 10 msec? or should it be more.
> >
> >The overhead of timer interrupts at this low clockrate is
> >significant so I recommend to minimize the timer interrupt rate as
> >far as possible. This is really a tradeoff between latency and
> >overhead and matters much less on hardcores which run at hundreds of
> >MHz.
> 
> Hm I've been running 2.6.13 on a 10/20 MHz (switchable) i386 @ 100 Hz
> before without any hangs during boot or operation.

Interrupt service overhead varies a bit between architectures, but your
observation isn't too surprising.  While the 1000Hz Linux 2.6 default
is just bad craziness for embedded cores and FPGA prototypes, I've
only seen 100Hz be truly unusable for sub-megahertz hardware simulators,
and then only when running virtual SMP kernels, where multiple virtual
"CPUs" on the same core all had to perform timer interrupt service
every clock interval, which multiplies the proportion of available cycles
consumed.  But even if the system boots and runs, it's pretty scary
to look at the proportion of time that a 20MHz core spends in interrupt
service with a HZ value of 100 or more.

So on one hand I agree with Ralf that on slow systems, especially
FPGA systems, one wants to keep the clock interrupt frequency
down to no more than 100Hz as a general rule (less than 100
wouldn't compile on 2.6.9 without some minor patches, which
took the minimum down to HZ=48, below which the various macros
that depend on HZ start generating divide-by-zero problems),
while on the other hand I agree with Jan that it's by no means certain
that Satesh's problem is really one of too many clock interrupts.

            Regards,

            Kevin K.


^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2007-01-22 16:50 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
     [not found] <20070120234237.49126.qmail@web7912.mail.in.yahoo.com>
2007-01-21  0:14 ` Running Linux on FPGA Ralf Baechle
2007-01-21  9:59   ` Sunil Naidu
2007-01-21 11:22     ` Kevin D. Kissell
2007-01-21 21:31   ` Jan Engelhardt
2007-01-22 16:58     ` Kevin D. Kissell

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).