<html><head></head><body><div style="color:#000; background-color:#fff; font-family:HelveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif;font-size:12px"><div id="yui_3_16_0_ym19_1_1485439637963_14916"><span></span></div> <div id="yui_3_16_0_ym19_1_1485439637963_15100" class="qtdSeparateBR"><br><br></div><div style="display: block;" id="yui_3_16_0_ym19_1_1485439637963_15091" class="yahoo_quoted"> <div id="yui_3_16_0_ym19_1_1485439637963_15090" style="font-family: HelveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif; font-size: 12px;"> <div id="yui_3_16_0_ym19_1_1485439637963_15089" style="font-family: HelveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif; font-size: 16px;"> <div dir="ltr"><font face="Arial" size="2"> On Thursday, January 26, 2017 8:52 AM, Aleksander Morgado <aleksander@aleksander.es> wrote:<br></font></div> <br><br> <div id="yui_3_16_0_ym19_1_1485439637963_15088" class="y_msg_container">On Wed, Jan 25, 2017 at 11:02 PM, Bjørn Mork <<a shape="rect" ymailto="mailto:bjorn@mork.no" href="mailto:bjorn@mork.no">bjorn@mork.no</a>> wrote:<br clear="none">>> On Wed, 1/25/17, Aleksander Morgado <<a shape="rect" ymailto="mailto:aleksander@aleksander.es" href="mailto:aleksander@aleksander.es">aleksander@aleksander.es</a>> wrote:<br clear="none">>><br clear="none">>> Which<br clear="none">>> system is this, btw?<br clear="none">>><br clear="none">>> Tang: I am testing on PowerPC embedded system, running Debian Linux.<br clear="none">><br clear="none">> In big endian mode, I guess?<br clear="none">><br clear="none">> Ouch,... Well, it's really good to have that tested as early as this.<br clear="none">> Thanks for doing that.<br clear="none">><br clear="none">> But you know, the original code was developed by trial-and-error on a PC<br clear="none">> only. Which tends to leave some questions unanswered wrt endianness.<br clear="none">> So there was probably a few hundred endianness bugs in the code<br clear="none">> Aleksander got from me. Some of them could have survived his massive<br clear="none">> cleanup.<br clear="none"><br clear="none">Interestingly, the backtrace shows the bug happening deep within GUdev...<br clear="none"><div id="yui_3_16_0_ym19_1_1485439637963_15125"><br></div><div>Tang: You are right, Bjorn. Absolutely it is the endian problem because qmi-firmware-update works like a charm on ARM platform. I've just tested on ARM embedded system, and it works well.</div><div dir="ltr" id="yui_3_16_0_ym19_1_1485439637963_15314"><br></div><div dir="ltr"><br></div>Tang, would it be possible for you to debug this further, e.g.<br clear="none">installing GUdev and GLib/GIO debug symbols and retrying? Otherwise,<br clear="none">would you be able to give me ssh access to the system where you're<br clear="none">running this?<div class="yqt2961179685" id="yqtfd74002"><div><br></div><div>Tang: I'll try to install GUdev/Glib/GIO debug symbol for more debug info. Also I will try to setup a PowerPC system for ssh access.</div><div>Thanks Alek</div><div><br></div><br clear="none">-- <br clear="none">Aleksander</div><br clear="none"><a id="yui_3_16_0_ym19_1_1485439637963_15101" shape="rect" href="https://aleksander.es/" target="_blank">https://aleksander.es</a><br clear="none">_______________________________________________<br clear="none">libqmi-devel mailing list<br clear="none"><a shape="rect" ymailto="mailto:libqmi-devel@lists.freedesktop.org" href="mailto:libqmi-devel@lists.freedesktop.org">libqmi-devel@lists.freedesktop.org</a><br clear="none"><a shape="rect" href="https://lists.freedesktop.org/mailman/listinfo/libqmi-devel" target="_blank">https://lists.freedesktop.org/mailman/listinfo/libqmi-devel</a><div class="yqt2961179685" id="yqtfd88492"><br clear="none"></div><br><br></div> </div> </div> </div></div></body></html>