<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=Windows-1252">
</head>
<body>
<div dir="auto" style="direction: ltr; margin: 0; padding: 0; font-family: sans-serif; font-size: 11pt; color: black; ">
yeah, it's a bit more complicated:<br>
<br>
</div>
<div dir="auto" style="direction: ltr; margin: 0; padding: 0; font-family: sans-serif; font-size: 11pt; color: black; ">
VME CPU is Intel/little endian, but the VME modules are all big endian...so the data readout out remains in big endian unless the Intel CPU byte swaps it (an easy task for it) - there should also be a bit set in the EVIO bank header indicating big or little
endian front end data type.<br>
<br>
</div>
<div dir="auto" style="direction: ltr; margin: 0; padding: 0; font-family: sans-serif; font-size: 11pt; color: black; ">
VTP is a bit different...Sure, it has an ARM CPU (which is little endian - ARM, btw, isn't committed to big or little - depends on the CPU model), but we're using the FPGA to readout the data which doesn Cate about endless and we picked a format to match the
Intel CPU default big endian format. We can easily change the VTP to have a config setting tonight it whichever way you want....so if little endian is really desired then we can add the support for this soon I think (something me/Dave/Bryan will need to do
for the various readout lists). The main thing is we're consist though and that we also indicate the correct endianness flag in the evio header that matches the data (I still don't see this being a huge CPU cycle cost as long as it's done remotely efficiently,
but just giving it in the right endianness doesn't seem like a big deal so let's shoot for that if folks are onboard/okay with that<span id="ms-outlook-android-cursor"></span>).</div>
<hr style="display:inline-block;width:98%" tabindex="-1">
<div id="divRplyFwdMsg" dir="ltr"><font face="Calibri, sans-serif" style="font-size:11pt" color="#000000"><b>From:</b> Sbs_daq <sbs_daq-bounces@jlab.org> on behalf of Alexandre Camsonne <camsonne@jlab.org><br>
<b>Sent:</b> Sunday, October 3, 2021 2:21:57 PM<br>
<b>To:</b> Paul King <pking@jlab.org><br>
<b>Cc:</b> sbs_daq@jlab.org <sbs_daq@jlab.org><br>
<b>Subject:</b> [Sbs_daq] [EXTERNAL] Re: Big endian raw data?</font>
<div> </div>
</div>
<div>
<div dir="auto">
<div>Everything is intel besided VTP. Though Dave mentionned VME was big endian.</div>
<div dir="auto"><br>
</div>
<div dir="auto">Alexandre<br>
<br>
<div class="x_gmail_quote" dir="auto">
<div dir="ltr" class="x_gmail_attr">On Sun, Oct 3, 2021, 13:56 Paul King <<a href="mailto:pking@jlab.org">pking@jlab.org</a>> wrote:<br>
</div>
<blockquote class="x_gmail_quote" style="margin:0 0 0 .8ex; border-left:1px #ccc solid; padding-left:1ex">
<div>
<div dir="auto" style="direction:ltr; margin:0; padding:0; font-family:sans-serif; font-size:11pt; color:black">
I can comment that when I wrote the helicity scaler library, I found that I needed to byte swap the data words (module data and diagnostic counters) on the crate in order to be decoded correctly.
<br>
</div>
<div dir="auto" style="direction:ltr; margin:0; padding:0; font-family:sans-serif; font-size:11pt; color:black">
I'm not sure if halladaq8 is an Intel or arm cpu.<br>
<br>
</div>
<div dir="auto" style="direction:ltr; margin:0; padding:0; font-family:sans-serif; font-size:11pt; color:black">
Does Podd or evio2xml do a dynamic check of endianness and then byteswap, or is that explicitly enabled?
<span id="x_m_-188557246818857469ms-outlook-android-cursor"></span></div>
<div><br>
</div>
<div id="x_m_-188557246818857469ms-outlook-mobile-signature">Sent from my mobile device.<br>
Get <a href="https://urldefense.proofpoint.com/v2/url?u=https-3A__aka.ms_ghei36&d=DwMFaQ&c=CJqEzB1piLOyyvZjb8YUQw&r=3WIiK4AiegFcRbgilX3vTw&m=qCvgogjK_XShaAzw6j_bVBbsMwegZv0ftnZS5u5C3VQ&s=DlHzZcyFxELhv7NPiwWjh8WrrALgcW1pFGAaTXpPHVE&e=" target="_blank" rel="noreferrer">
Outlook for Android</a></div>
<hr style="display:inline-block; width:98%">
<div id="x_m_-188557246818857469divRplyFwdMsg" dir="ltr"><font face="Calibri, sans-serif" color="#000000" style="font-size:11pt"><b>From:</b> Sbs_daq <<a href="mailto:sbs_daq-bounces@jlab.org" target="_blank" rel="noreferrer">sbs_daq-bounces@jlab.org</a>> on
behalf of Andrew Puckett <<a href="mailto:puckett@jlab.org" target="_blank" rel="noreferrer">puckett@jlab.org</a>><br>
<b>Sent:</b> Sunday, October 3, 2021 1:31:05 PM<br>
<b>To:</b> Robert Michaels <<a href="mailto:rom@jlab.org" target="_blank" rel="noreferrer">rom@jlab.org</a>>; Ole Hansen <<a href="mailto:ole@jlab.org" target="_blank" rel="noreferrer">ole@jlab.org</a>>;
<a href="mailto:sbs_daq@jlab.org" target="_blank" rel="noreferrer">sbs_daq@jlab.org</a> <<a href="mailto:sbs_daq@jlab.org" target="_blank" rel="noreferrer">sbs_daq@jlab.org</a>><br>
<b>Subject:</b> Re: [Sbs_daq] Big endian raw data?</font>
<div> </div>
</div>
<div lang="EN-US" style="word-wrap:break-word">
<div>
<p>Interesting. So perhaps I’m being naďve here, but other than the byte-swapping inefficiency Ole pointed out in processing the raw data on the compute farm nodes, is there an actual problem here? Do we need to check/care about this in the software in writing
our raw data decoders?</p>
<p> </p>
<p>The cause of Bradley’s crash while processing GRINCH data doesn’t necessarily seem related to this…
</p>
<p> </p>
<p>Andrew </p>
<p> </p>
<div style="border:none; border-top:solid #b5c4df 1.0pt; padding:3.0pt 0in 0in 0in">
<p style="margin-bottom:12.0pt"><b><span style="font-size:12.0pt; color:black">From:
</span></b><span style="font-size:12.0pt; color:black">Robert Michaels <<a href="mailto:rom@jlab.org" target="_blank" rel="noreferrer">rom@jlab.org</a>><br>
<b>Date: </b>Sunday, October 3, 2021 at 1:21 PM<br>
<b>To: </b>Ole Hansen <<a href="mailto:ole@jlab.org" target="_blank" rel="noreferrer">ole@jlab.org</a>>, Andrew Puckett <<a href="mailto:puckett@jlab.org" target="_blank" rel="noreferrer">puckett@jlab.org</a>>,
<a href="mailto:sbs_daq@jlab.org" target="_blank" rel="noreferrer">sbs_daq@jlab.org</a> <<a href="mailto:sbs_daq@jlab.org" target="_blank" rel="noreferrer">sbs_daq@jlab.org</a>><br>
<b>Subject: </b>Re: [Sbs_daq] Big endian raw data?</span></p>
</div>
<div>
<p><span style="font-size:12.0pt; color:black">I believe there are byte-swapping routines available in the DAQ libraries which allow to put the bytes in the right state and be consistent. But the DAQ expert needs to make this happen. Below is a snippet of
an email from Dave Abbott about a year ago when I was having some trouble, which I think is relevant.. Dave is a good person to ask. Can ask Bryan Moffit or Alexandre, too.</span></p>
</div>
<div>
<p><span style="font-size:12.0pt; color:black"> </span></p>
</div>
<div>
<p><span style="font-size:12.0pt; color:black">---------------------- snippet of email from Dave Abbott ------------------------</span></p>
</div>
<div>
<p><span style="font-size:12.0pt; color:black"> </span></p>
</div>
<div>
<p><span style="font-size:12.0pt; color:black; background:white">The CODA data files are written from a Java Event Builder. JAVA is inherently Big Endian. The EVIO</span><span style="font-size:12.0pt; color:black">
</span></p>
<div>
<p style="background:white"><span style="font-size:12.0pt; color:black">files will be by default in big endian.</span></p>
</div>
<div>
<p style="background:white"><span style="font-size:12.0pt; color:black"> </span></p>
</div>
<div>
<p style="background:white"><span style="font-size:12.0pt; color:black">However, ALL Banks of User data - created in your readout list - will NOT be swapped. They will stay</span></p>
</div>
<div>
<p style="background:white"><span style="font-size:12.0pt; color:black">whatever Endian it was when it was written.</span></p>
</div>
<div>
<p style="background:white"><span style="font-size:12.0pt; color:black"> </span></p>
</div>
<div>
<p style="background:white"><span style="font-size:12.0pt; color:black">Typically the ROC will run in Linux on Intel which is Little Endian. Therefore the Data banks you create will stay</span></p>
</div>
<div>
<p style="background:white"><span style="font-size:12.0pt; color:black">little endian. However the Bank headers will be swapped to be compatible with the rest of the CODA file.</span></p>
</div>
<div>
<p style="background:white"><span style="font-size:12.0pt; color:black"> </span></p>
</div>
<div>
<p style="background:white"><span style="font-size:12.0pt; color:black">An even more confusing possibility is that you might do a DMA from the VME bus into a CODA data Bank.</span></p>
</div>
<div>
<p style="background:white"><span style="font-size:12.0pt; color:black">The VME bus is Big endian. Therefore the data from the VME bus will stay Big endian in this bank.</span></p>
</div>
<div>
<p style="background:white"><span style="font-size:12.0pt; color:black"> </span></p>
</div>
<div>
<p style="background:white"><span style="font-size:12.0pt; color:black">Our general rule for CODA 3 is that for purposes of DAQ we will not touch (or modify) the User's data in any way.</span></p>
</div>
<p><span style="font-size:12.0pt; color:black; background:white">We will only modify the EVIO headers to match the endianess of whatever System writes the file.</span><span style="font-size:12.0pt; color:black"></span></p>
</div>
<div>
<div>
<p><span style="font-size:12.0pt; color:black"> </span></p>
</div>
<div id="x_m_-188557246818857469x_Signature">
<div>
<div id="x_m_-188557246818857469x_divtagdefaultwrapper">
<p><span style="font-size:12.0pt; color:black"> </span></p>
</div>
</div>
</div>
</div>
<div align="center" style="text-align:center">
<hr size="0" width="100%" align="center">
</div>
<div id="x_m_-188557246818857469x_divRplyFwdMsg">
<p><b><span style="color:black">From:</span></b><span style="color:black"> Sbs_daq <<a href="mailto:sbs_daq-bounces@jlab.org" target="_blank" rel="noreferrer">sbs_daq-bounces@jlab.org</a>> on behalf of Ole Hansen <<a href="mailto:ole@jlab.org" target="_blank" rel="noreferrer">ole@jlab.org</a>><br>
<b>Sent:</b> Sunday, October 3, 2021 1:06 PM<br>
<b>To:</b> Andrew Puckett <<a href="mailto:puckett@jlab.org" target="_blank" rel="noreferrer">puckett@jlab.org</a>>;
<a href="mailto:sbs_daq@jlab.org" target="_blank" rel="noreferrer">sbs_daq@jlab.org</a> <<a href="mailto:sbs_daq@jlab.org" target="_blank" rel="noreferrer">sbs_daq@jlab.org</a>><br>
<b>Subject:</b> Re: [Sbs_daq] Big endian raw data?</span> </p>
<div>
<p> </p>
</div>
</div>
<div>
<p style="margin-bottom:12.0pt">Maybe our various front-ends differ in endianness, so we write mixed-endian data?!? That would be disastrous since it is not supported by EVIO. A file can only be one or the other—a very binary view. (I guess EVIO was written
before we became diversity-aware ;) ).<br>
<br>
Ole</p>
<div>
<p>On 3.10.21 at 13:03, Andrew Puckett wrote:</p>
</div>
<blockquote style="margin-top:5.0pt; margin-bottom:5.0pt">
<div>
<p>Hi Ole, </p>
<p> </p>
<p>This is interesting. The GRINCH data are being read out by the new VETROC modules, I don’t know if they differ from the other modules in terms of “endian-ness”. Maybe a DAQ expert can weigh in here?</p>
<p> </p>
<p>Andrew </p>
<p> </p>
<div style="border:none; border-top:solid #b5c4df 1.0pt; padding:3.0pt 0in 0in 0in">
<p style="margin-bottom:12.0pt"><b><span style="font-size:12.0pt; color:black">From:
</span></b><span style="font-size:12.0pt; color:black">Sbs_daq <a href="mailto:sbs_daq-bounces@jlab.org" target="_blank" rel="noreferrer">
<sbs_daq-bounces@jlab.org></a> on behalf of Ole Hansen <a href="mailto:ole@jlab.org" target="_blank" rel="noreferrer">
<ole@jlab.org></a><br>
<b>Date: </b>Sunday, October 3, 2021 at 1:00 PM<br>
<b>To: </b><a href="mailto:sbs_daq@jlab.org" target="_blank" rel="noreferrer">sbs_daq@jlab.org</a>
<a href="mailto:sbs_daq@jlab.org" target="_blank" rel="noreferrer"><sbs_daq@jlab.org></a><br>
<b>Subject: </b>[Sbs_daq] Big endian raw data?</span></p>
</div>
<p style="margin-bottom:12.0pt">Hi guys,<br>
<br>
Bradley reported a crash of the replay (actually in EVIO) with /adaq1/data1/sbs/grinch_72.evio.0 (see
<a href="https://logbooks.jlab.org/entry/3916105" target="_blank" rel="noreferrer">
https://logbooks.jlab.org/entry/3916105</a>).<br>
<br>
When digging into the cause of this crash, I discovered that these raw data are written in big-endian format. How can this be? I thought the front-ends are Intel processors. Are we taking data with ARM chips that are configured for big-endian mode? Is this
a mistake, or is there some plan to it?<br>
<br>
These big-endian data have to be byte-swapped when processing them on x86, which is what all our compute nodes run. That's a LOT of work. It leads to significant and seemingly completely unnecessary overhead. I.e. we're burning CPU cycles for nothing good,
it seems.<br>
<br>
Please explain.<br>
<br>
Ole</p>
</div>
</blockquote>
<p> </p>
</div>
</div>
</div>
</div>
_______________________________________________<br>
Sbs_daq mailing list<br>
<a href="mailto:Sbs_daq@jlab.org" target="_blank" rel="noreferrer">Sbs_daq@jlab.org</a><br>
<a href="https://mailman.jlab.org/mailman/listinfo/sbs_daq" rel="noreferrer noreferrer" target="_blank">https://mailman.jlab.org/mailman/listinfo/sbs_daq</a><br>
</blockquote>
</div>
</div>
</div>
</div>
</body>
</html>