> > For my analyzer I generate the experim.h file from the odb.
This issue is still open. Shouldn't midas.h provide the 'new' data types as typedefs like  

typedef int INT32;

etc. Of course you would need to deal with all the supported targets and wrap it accordingly.

A.S.

> > 
> > Before midas commit 13c3b2b this generates structs with INT data types. compiles fine with my analysis code (using the old mana.cpp)
> > 
> > newer midas versions generate INT32, ... types. I get a 
> > 
> > ‘INT32’ does not name a type   
> > 
> > although I include midas.h 
> > 
> > how to fix this?
> 
> You could run experim.h through "sed" to replace the "wrong" data types with the correct data types.
> 
> You can also #define the "wrong" data types before doing #include experim.h.
> 
> I put your bug report into our bug tracker, but for myself I am very busy
> with the alpha-g experiment and cannot promise to fix this quickly.
> 
> https://bitbucket.org/tmidas/midas/issues/289/int32-types-in-experimh
> 
> Here is an example to substitute things using "sed" (it can also do "in-place" editing, "man sed" and google sed examples)
> sed "sZshm_unlink(.*)Zshm_unlink(SHM)Zg"
> 
> K.O.

Ok, I added

/* define integer types with explicit widths */
#ifndef NO_INT_TYPES_DEFINE
typedef unsigned char      UINT8;
typedef char               INT8;
typedef unsigned short     UINT16;
typedef short              INT16;
typedef unsigned int       UINT32;
typedef int                INT32;
typedef unsigned long long UINT64;
typedef long long          INT64;
#endif

to cover all new types. If there is a collision with user defined types, compile your program with -DNO_INT_TYPES_DEFINE and you remove the 
above definition. I hope there are no other conflicts.

Stefan

works!

> Ok, I added
> 
> /* define integer types with explicit widths */
> #ifndef NO_INT_TYPES_DEFINE
> typedef unsigned char      UINT8;
> typedef char               INT8;
> typedef unsigned short     UINT16;
> typedef short              INT16;
> typedef unsigned int       UINT32;
> typedef int                INT32;
> typedef unsigned long long UINT64;
> typedef long long          INT64;
> #endif
> 
> to cover all new types. If there is a collision with user defined types, compile your program with -DNO_INT_TYPES_DEFINE and you remove the 
> above definition. I hope there are no other conflicts.
> 
> Stefan

> > 
> > /* define integer types with explicit widths */
> > #ifndef NO_INT_TYPES_DEFINE
> > typedef unsigned char      UINT8;
> > typedef char               INT8;
> > typedef unsigned short     UINT16;
> > typedef short              INT16;
> > typedef unsigned int       UINT32;
> > typedef int                INT32;
> > typedef unsigned long long UINT64;
> > typedef long long          INT64;
> > #endif
> > 

NIH at work. In C and C++ the standard fixed bit length data types are available
in #include <stdint.h> as uint8_t, uint16_t, uint32_t, uint64_t & co.

BTW, the definition of UINT32 as "unsigned int" is technically incorrect, on 16-bit machines
"int" is 16-bit wide and on some 64-bit machines "int" is 64-bit wide.

K.O.

Hello,

I'm co-organizing the upcoming Real Time Conference, which covers also the field of data acquisition, so it might be interesting for people working 
with MIDAS. If you have something to report, you could also consider to send an abstract to this conference. It will be located in Nara, Japan. The conference
site is now open at http://rt2014.rcnp.osaka-u.ac.jp/

Best regards,
Stefan Ritt

Hello,

I'm co-organizing the upcoming Real Time Conference, which covers also the field of data acquisition, so it might be interesting for people working 
with MIDAS. If you have something to report, you could also consider to send an abstract to this conference.  It will be nicely located in Berkeley, 
California. We plan excursions to San Francisco and to Napa Valley.

Best regards,
Stefan Ritt

---------------------------

18th Real Time Conference
June 11 – 15, 2012
Berkeley, CA

We invite you to the Hotel Shattuck Plaza in downtown Berkeley, California for
the 2012 Real-Time Conference (RT2012).   It will take place Monday, June 11
through Friday, June 15, 2012, with optional pre-conference tutorials Saturday
and Sunday, June 9-10.

Like the previous editions, RT2012 will be a multidisciplinary conference
devoted to the latest developments on realtime techniques in the fields of
plasma and nuclear fusion, particle physics, nuclear physics and astrophysics,
space science, accelerators, medical physics, nuclear power instrumentation and
other radiation instrumentation.

Abstract submission is open as of 18 January (deadline 2 March). Please visit

http://www.npss-confs.org/rtc/welcome.asp?flag=44675.77&Retry=1 to submit an

abstract.

Call for Abstracts 

RT 2012 is an interdisciplinary conference on realtime data acquisition and
computing applications in the physical sciences. These applications include:

* High energy physics 
* Nuclear physics 
* Astrophysics and astroparticle physics 
* Nuclear fusion 
* Medical physics 
* Space instrumentation 
* Nuclear power instrumentation 
* Realtime security and safety 
* General Radiation Instrumentation 

Specific topics include (but are certainly not limited to) the list shown below.
We welcome correspondence to see how your research fits our venue.   

Key Dates

* Abstract submission opened:  January 18, 2012 
* Abstract deadline:  March 2, 2012 
* Program available: April 2 

Suggested Topics

* Realtime system architectures 
* Intelligent signal processing 
* Programmable devices 
* Fast data transfer links and networks 
* Trigger systems 
* Data acquisition 
* Processing farms 
* Control, monitoring, and test systems 
* Upgrades 
* Emerging realtime technologies 
* New standards 
* Realtime safety and security 
* Feedback on experiences 

Contact Information

If you have a question or wish to opt in for occasional e-mail updates about
RT2012, send us a message at RT2012@lbl.gov. To view full conference
information, visit http://rt2012.lbl.gov/index.html

I'm looking into using MIDAS for an experiment that creates one large event
(20MB or more) every second.

Q1: It looks like I should use EQ_FRAGMENTED. Has this feature been in use
recently? Is it known to work/not work?

More specifically, the computer should initiate a 1 second data taking, start to
such the data out of the electronics (which may take a while), change some
experimental parameters, and start over. 

Q2: What's the best way to do this? EQ_PERIODIC? 
I cannot guarantee that the time required to read the hardware has an upper bound.
In a standalone-prog I would simply use a big loop and let the machine execute
it as fast as it can: 1.1s, 1.5s, 1.1s, 1.3s, 2.5s, ..... depending on the HW
deadtimes.
Will this work with EQ_PERIODIC?

(Sorry for these maybe stupid questions, but I have so far only used MIDAS for
externally generated events, with <32kB event size).


Thanks a lot,

Randolf

> I'm looking into using MIDAS for an experiment that creates one large event
> (20MB or more) every second.
> 
> Q1: It looks like I should use EQ_FRAGMENTED. Has this feature been in use
> recently? Is it known to work/not work?
> 
> More specifically, the computer should initiate a 1 second data taking, start to
> such the data out of the electronics (which may take a while), change some
> experimental parameters, and start over. 
> 
> Q2: What's the best way to do this? EQ_PERIODIC? 
> I cannot guarantee that the time required to read the hardware has an upper bound.
> In a standalone-prog I would simply use a big loop and let the machine execute
> it as fast as it can: 1.1s, 1.5s, 1.1s, 1.3s, 2.5s, ..... depending on the HW
> deadtimes.
> Will this work with EQ_PERIODIC?
> 
> (Sorry for these maybe stupid questions, but I have so far only used MIDAS for
> externally generated events, with <32kB event size).
> 
> 
> Thanks a lot,
> 
> Randolf

Hi Randolf,

EQ_FRAGMENTED is kind of historically, when computers had a few MB of memory and you have to play special tricks to get large data buffers through. Today I 
would just use EQ_PERIODIC and increase the midas maximal event size to your needs. For details look here:

https://midas.triumf.ca/MidasWiki/index.php/Event_Buffer

The front-end scheduler is asynchronous, which means that your readout is called when the given period (1 second) is elapsed. If the readout takes longer 
than 1s, the schedule will (hopefully) call your readout immediately after the event has been sent. So you get automatically your maximal data rate. At MEG, we 
use 2 MB events with 10 Hz, so a 20 MB/sec data rate should not be a problem on decent computers.

Best,
Stefan

> I'm looking into using MIDAS for an experiment that creates one large event
> (20MB or more) every second.

Hi, there - 20 Mbyte event at 1/sec is not so large these days. (Well, depending on your hardware).

Using typical 1-2 year old PC hardware, 20 M/sec to local disk should work right away. Sending data from a 
remote front end (through the mserver), or writing to a remote disk (NFS, etc) - will of course requre a GigE 
network connection.

By default, MIDAS is configured for using about 1-2 Mbyte events, so for your case, you will need to:

- increase the event size limits in your frontend,
- increase /Experiment/MAX_EVENT_SIZE in ODB
- increase the size of the SYSTEM event buffer (/Experiment/Buffer sizes/SYSTEM in ODB)

I generally recommend sizing the SYSTEM event buffer to hold a few seconds worth of data (ot 
accommodate any delays in writing to local disk - competing  reads, internal delays of the disks, etc).

So for 20 M/s, the SYSTEM buffer size should be about 40-60 Mbytes.

For your case, you also want to buffer 3-5-10 events, so the SYSTEM buffer size would be between 100 and 
200 MBytes.

Assuming you have between 8-16-32 GBytes of RAM, this should not be a problem.

One the other hand, if you are running on a low-power ("green") ARM system with 1 Gbyte of RAM and a 
1GHz CPU, you should be able to handle the data rate of 20 Mbytes/sec, as long as your network and 
storage can handle it - I see GigE ethernet running at about 30-40 Mbytes/sec, so you should be okey,
but local storage to SD flash is only about 10 Mbytes/sec - too slow. You can try USB-attached HDD or SSD, 
this should run at up to 30-40 Mbytes/sec. I would expect no problems with this rate from MIDAS, as long 
as you can fit into your 1 GByte of RAM - obviously your SYSTEM buffer will have to be a little smaller than 
on a full-featured PC.

More information on MIDAS event size limits is here (as already reported by Stefan)
https://midas.triumf.ca/MidasWiki/index.php/Event_Buffer

Let us know how it works out for you.

K.O.

Works, and here is how I did it. The attached example is based on the standard MIDAS
example in "src/midas/examples/experiment". 

My somewhat unsorted notes, haven't really tweaked the numbers. But it WORKS.

(1) mlogger writes "last.xml" (hard-coded!) which takes an awful amount of time
    as it writes the complete ODB containing the 10MB bank!
    just outcomment 
       // odb_save("last.xml");
    in mlogger.cxx, function 
    INT tr_start(INT run_number, char *error)
    (line ~3870 in mlogger rev. 5377, .cxx-file included)

(2) frontend.c:
     * the most important declarations are

/* BIG_DATA_BYTES is the data in 1 bank
   BIG_EVENT_SIZE is the event size. It's a bit larger than the bank size
                  because MIDAS needs to add some header bytes, I think
 */

#define BIG_DATA_BYTES  (10*1024*1024)   // 10 MB
#define BIG_EVENT_SIZE  (BIG_DATA_BYTES + 100)

/* maximum event size produced by this frontend */
INT max_event_size = BIG_EVENT_SIZE;

/* maximum event size for fragmented events (EQ_FRAGMENTED) */
INT max_event_size_frag = 5 * BIG_EVENT_SIZE;

/* buffer size to hold 10 events */
INT event_buffer_size = 10 * BIG_EVENT_SIZE;


     * bk_init() can hold at most 32kByte size events! Use bk_init32() instead.

     * complete frontend.c is attached

(3) in an xterm do
    # . setup.sh
    # odbedit -s 41943040
          (first invocation of odbedit must create large enough odb,
           otherwise you'll get "odb full" errors)
(4) odbedit> load big.odb      
    (attached). Essentials are:

    /Experiment/MAX_EVENT_SIZE = 20971520
    /Experiment/Buffer sizes/SYSTEM = 41943040   <- at least 2 events!

    To avoid excessive latecies when starting/stopping a run, do
    /Logger/ODB Dump = no 
    /Logger/Channels/0/Settings/ODB Dump = no 
     
    and create an Equipment Tree to make the mlogger happy

(5) a few more xterms, always ". setup.sh":
    # mlogger_patched (see (1))
    # ./frontend  (attched)

(6) in your odbedit (4) say "start". You should fill your disk rather quickly.

> Works, and here is how I did it. The attached example is based on the standard MIDAS
> example in "src/midas/examples/experiment". 

If you have such huge events, it does not make sense to put them into the ODB. The size needs to be increased (as 
you realized correctly) and the run stop takes long if you write last.xml. So just remove the RO_ODB flag in the 
frontend program and you won't have these problems.

/Stefan

I want to let the run stop and start periodically. But I looked through the ODB
and didn't find anything may help. I also checked the FAQ online and didn't find
answer either. Who can help me? Thank you.

Hi Exaos,

This may help: https://ladd00.triumf.ca/elog/Midas/645

You need to set the following keys:
/Logger/Run duration
/Logger/Auto restart
/Logger/Auto restart delay

Regards,
Jimmy


> I want to let the run stop and start periodically. But I looked through the ODB
> and didn't find anything may help. I also checked the FAQ online and didn't find
> answer either. Who can help me? Thank you.

If there's some template for writing a client to access event data, that would be 
very useful (and you can probably just ignore the context I gave below in that 
case).


Some context:

Quite a while ago, I wrote the attached "data pipeline" client whose job was to 
listen for events, copy their data, and pipe them to a python script. I believe I 
just stole bits and pieces from mdump.cxx to accomplish this. Later I wrote the 
attached wrapper class "MidasConnector.cpp" and a main.cpp to generalize
data_pipeline.cxx a bit. There were a lot of iterations to the code where I had the 
below problems; so don't take the logic in the attached code as the exact code that 
caused the issues below.

However, I'm unable to resolve a couple issues:

1. If a timeout is set, everything will work until that timeout is reached. Then 
regardless of what kind of logic I tried to implement (retry receiving event, 
disconnect and reconnect client, etc.) the client would refuse to receive more data.

2. When I ctrl-C main, it hangs; this is expected because it's stuck in a while 
loop. But because I can't set a timeout I have to ctrl-C twice; this would 
occasionally corrupt the ODB which was not ideal. I was able to get around this with 
some impractical solution involving ncurses I believe.


Thanks,
Jack

> If there's some template for writing a client to access event data, that would be 
> very useful (and you can probably just ignore the context I gave below in that 
> case).
> 
> 
> Some context:
> 
> Quite a while ago, I wrote the attached "data pipeline" client whose job was to 
> listen for events, copy their data, and pipe them to a python script. I believe I 
> just stole bits and pieces from mdump.cxx to accomplish this. Later I wrote the 
> attached wrapper class "MidasConnector.cpp" and a main.cpp to generalize
> data_pipeline.cxx a bit. There were a lot of iterations to the code where I had the 
> below problems; so don't take the logic in the attached code as the exact code that 
> caused the issues below.
> 
> However, I'm unable to resolve a couple issues:
> 
> 1. If a timeout is set, everything will work until that timeout is reached. Then 
> regardless of what kind of logic I tried to implement (retry receiving event, 
> disconnect and reconnect client, etc.) the client would refuse to receive more data.
> 
> 2. When I ctrl-C main, it hangs; this is expected because it's stuck in a while 
> loop. But because I can't set a timeout I have to ctrl-C twice; this would 
> occasionally corrupt the ODB which was not ideal. I was able to get around this with 
> some impractical solution involving ncurses I believe.
> 
> 
> Thanks,
> Jack

midas/examples/lowlevel/consume.cxx might be what you're looking for, but I think all 
you're missing is a call to cm_yield() in your loop, so your midas client doesn't get 
killed when the timeout is reached (and also so you can act on shutdown requests from 
midas)

Something like 
      int status = cm_yield(100);
      if (status == SS_ABORT || status == RPC_SHUTDOWN)
         break;

There might be a recommended way to handle the ctrl-c and disconnect from the ODB, but 
off the top of my head I don't remember it. 

Also check out Ben's new(ish) python library, midas/python/examples/event_receiver.py 
might be a much easier solution. And you can use the context manager, which will take 
care of safely disconnecting from midas after you ctrl-C.

unsigned int EXPRT mvme_read_value(MVME_INTERFACE * vme, mvme_addr_t vme_addr);

Read single data from VME bus. Useful for register access. See example in
mvme_open()

Parameters:
    	*vme 	  VME structure
    	vme_addr  source address (VME location).

Returns:
    MVME_SUCCESS

int EXPRT mvme_read_value(MVME_INTERFACE *mvme, mvme_addr_t vme_addr, unsigned int *var);

unsigned int EXPRT mvme_read_value(MVME_INTERFACE * vme, mvme_addr_t vme_addr);

Read single data from VME bus. Useful for register access. See example in
mvme_open()

Parameters:
    	*vme 	  VME structure
    	vme_addr  source address (VME location).

Returns:
    MVME_SUCCESS

int EXPRT mvme_read_value(MVME_INTERFACE *mvme, mvme_addr_t vme_addr, unsigned int *var);

  printf("%d\n", mvme_read_value(...)); 

  unsigned long d;
  mvme_read_value(..., &d);
  printf("%d\n", d);

  status = mvme_read(..., &d, adr, 4); 

> uint32_t mvme_read_value(MVME_INTERFACE * vme, mvme_addr_t vme_addr);
> Question: How to distinguish the status and value returned?

On VME interfaces using the Universe and tsi148 PCI-VME bridges, your question has no meaning.

The VME address space is directly mapped into the PCI address space, then mmap()ed into your 
program address space. Internally mvme_read_value() is "return *(uint32_t*)(mmap_base + vme_addr);" 
and there is no such thing as "status".

Physically on the VME bus, for single-word VME cycles (mvme_read_value), there are only 2 error 
conditions, an AS or a DS timeout, and these bridges return the bit pattern 0xFFFFFFFF for either error, 
the same as the traditional VME bus always worked (i.e. before PCI, before ISA, back when the VME bus 
*was* the main CPU-memory-IO bus).

So the answer to your question is "yes". If mvme_read_value() returned 0xFFFFFFFF, there was a VME 
bus timeout because the board you are trying to address a) is not installed, b) was unplugged, c) does 
not decode the address you tried to access (maybe you used the wrong AM code or the wrong data 
width).

With Universe and tsi148 PCI-VME bridges, the mvme_read() call runs the DMA interface that can issue 
block transfer cycles on the VME bus. These DMA interfaces have interesting error handling, but 
basically, they only tell you the estimated VME address at which the AS or DS timeout or BERR has 
occurred. For sane VME boards, DMA errors mean very basic breakage of the VME crate and VME board.

With non-directly attached VME interfaces, i.e. the SIS3100, you can also have communication errors. I 
do not know how those are reported by the SIS3100 Linux drivers, and I do not know how the MIDAS 
driver reports them. But I do know that if you see those errors, your interface is very broken and VME 
bus errors are the least of your worries.

P.S. There also exist PCI bus errors, they also return the bit pattern 0xFFFFFFFF and mean basic 
breakage inside your computer. PCI-PCI and PCI-host bridges have special registers you can read to 
find out the exact cause of the error ("your computer is broken").

K.O.

Big thanks to Peter Kunz - specifically when using the CAEN libraries:

>
> After upgrading to the lastes MIDAS version I got the DAQ frontend of my application running by
> changing all compiler directives from cc to g++ and using
>
> #include "mfe.h"
>
> extern HNDLE hDB
> 
>  extern "C" { 
>  #include <CAENComm.h>  
>  }
>
> With these changes everything seems to work fine.
>


K.O.



> > To convert a MIDAS frontend to C++ follow this checklist:
> 
> Pierre A.-A. reminded me that include files for CAEN libraries have  to
> use "extern C" brackets:
> 
> some 3rd party libraries (CAEN, etc) are written in C (or require C linkage),
> if their include files are not C++ compatible (do not have "extern C" brackets
> for all exported symbols), the experiment frontend code must say something like this:
> 
> extern "C" {
> #include "3rd-party-c-library.h"
> }
> 
> Note: "#ifdef cplusplus" is not needed because we already know we are C++, not C.
> 
> K.O.