<?xml version="1.0" encoding="utf-8"?>
<feed xml:lang="en" xmlns="http://www.w3.org/2005/Atom"><title>Recent changes to patches</title><link href="https://sourceforge.net/p/canfestival/patches/" rel="alternate"/><link href="https://sourceforge.net/p/canfestival/patches/feed.atom" rel="self"/><id>https://sourceforge.net/p/canfestival/patches/</id><updated>2019-04-12T11:22:20.324000Z</updated><subtitle>Recent changes to patches</subtitle><entry><title>#19 Add data pointer argument to scanIndexOD_t</title><link href="https://sourceforge.net/p/canfestival/patches/19/?limit=25#5bc9" rel="alternate"/><published>2019-04-12T11:22:20.324000Z</published><updated>2019-04-12T11:22:20.324000Z</updated><author><name>Siert Wieringa</name><uri>https://sourceforge.net/u/siert/</uri></author><id>https://sourceforge.netd36c3db441d2f631726b80aa45fdd41a4be3078d</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;I forgot to write that this patch was created on top of Mercurial changeset de1fc3261f21, which is the current newest changeset in &lt;a href="https://hg.beremiz.org/CanFestival-3" rel="nofollow"&gt;https://hg.beremiz.org/CanFestival-3&lt;/a&gt;&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Add data pointer argument to scanIndexOD_t</title><link href="https://sourceforge.net/p/canfestival/patches/19/" rel="alternate"/><published>2019-04-12T11:20:29.813000Z</published><updated>2019-04-12T11:20:29.813000Z</updated><author><name>Siert Wieringa</name><uri>https://sourceforge.net/u/siert/</uri></author><id>https://sourceforge.nete97d1918a17b0492c8348b8df4ca12a811c4edcb</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;Hi,&lt;/p&gt;
&lt;p&gt;I've created  C++ classes for creating an object dictionary and all the data necessary for the CO_Data struct. In this way I can create new independent instances of CO_Data dynamically at run time.&lt;/p&gt;
&lt;p&gt;I use this because in my setup I have a configurable number of CAN devices, each one attached to its own independent CAN bus. On each CAN bus I have an instance of CanFestival as CanOpen master.&lt;/p&gt;
&lt;p&gt;However, I have a problem with implementing the scanIndexOD function in this generic setup, since that function does not take any argument from which I can derive which open CAN handle called it. In my opinion the scanIndexOD_t function pointer type should be extended to include the CO_Data pointer as a first argument, since this makes it possible to implement one static function that can decided which instance of the object dictionary to index (d-&amp;gt;NODEPREFIX_objdict).&lt;/p&gt;
&lt;p&gt;I believe that even though the data pointer parameter is not necessary in the rather static usage of CanFestival that was apparently envisioned by its designers it is a good improvement also for consistency (all other callbacks do have the data pointer).&lt;/p&gt;
&lt;p&gt;The attached patch updates the scanIndexOD_t type, all examples, and the object dictionary generation python script.&lt;/p&gt;
&lt;p&gt;If anybody is interested I can make my C++ wrappers publically available, they do make managing object dictionaries much easier for C++ users.&lt;/p&gt;
&lt;p&gt;Best regards,&lt;/p&gt;
&lt;p&gt;Siert Wieringa&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>TimeDispatch</title><link href="https://sourceforge.net/p/canfestival/patches/18/" rel="alternate"/><published>2017-05-19T15:30:15.545000Z</published><updated>2017-05-19T15:30:15.545000Z</updated><author><name>Luca Girotti</name><uri>https://sourceforge.net/u/itlugi/</uri></author><id>https://sourceforge.net8dfa9a413dfde54a21c747e87085f321d63fd335</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;See in attachment my revision.&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>[AVR] timer overflow in timerscfg.h</title><link href="https://sourceforge.net/p/canfestival/patches/17/" rel="alternate"/><published>2013-08-20T15:57:32.625000Z</published><updated>2013-08-20T15:57:32.625000Z</updated><author><name>Joris Lambrecht</name><uri>https://sourceforge.net/u/yorisimo/</uri></author><id>https://sourceforge.net96015c9ae91e99b5fed28d396751b9f414f7e6d2</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;In timerscfg.h, change&lt;br /&gt;
 MS_TO_TIMEVAL(ms) ((ms) * 125) &lt;br /&gt;
to&lt;br /&gt;
 MS_TO_TIMEVAL(ms) ((ms) * 125UL)&lt;/p&gt;
&lt;p&gt;The overflow manifests itself as an inaccurate heartbeat period while sdo upload/downloads are being performed and an inaccurate sdo timeout.&lt;/p&gt;
&lt;p&gt;Another user found the issue as well and posted here:&lt;br /&gt;
&lt;a href="http://osdir.com/ml/embedded.canopen.canfestival/2008-07/msg00005.html" rel="nofollow"&gt;http://osdir.com/ml/embedded.canopen.canfestival/2008-07/msg00005.html&lt;/a&gt;&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>SDO abort code patch</title><link href="https://sourceforge.net/p/canfestival/patches/16/" rel="alternate"/><published>2013-06-05T06:46:38.659000Z</published><updated>2013-06-05T06:46:38.659000Z</updated><author><name>Uwe Kindler</name><uri>https://sourceforge.net/u/incredibleukee/</uri></author><id>https://sourceforge.netbb818d76c347c2dc3441c4f56426471cb45f1204</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;The following patch adds missing DS301 SDO abort codes and provides proper enumeration values for abort codes and object dictionary return codes. The patch also fixes the values of OD_VALUE_TOO_HIGH and OD_VALUE_TOO_LOW because they was reversed.&lt;/p&gt;
&lt;p&gt;Because the patch uses the same names it does not break existing code.&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Common DS301 object dictionary index identfiers</title><link href="https://sourceforge.net/p/canfestival/patches/15/" rel="alternate"/><published>2013-06-05T06:01:45.736000Z</published><updated>2013-06-05T06:01:45.736000Z</updated><author><name>Uwe Kindler</name><uri>https://sourceforge.net/u/incredibleukee/</uri></author><id>https://sourceforge.netc5523eba910eda73be7c8f38ea682d285ee01103</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;The following patch adds object dictionary identifiers to ease object dictionary access for applications that need to access certain object dictionary entries via readLocalDict() or setODentry()&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>LSS and compiler warning fixes</title><link href="https://sourceforge.net/p/canfestival/patches/14/" rel="alternate"/><published>2013-06-04T07:30:37.842000Z</published><updated>2013-06-04T07:30:37.842000Z</updated><author><name>Uwe Kindler</name><uri>https://sourceforge.net/u/incredibleukee/</uri></author><id>https://sourceforge.net74bebcf4de444c727ddba228483346effd092c3c</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;Hi,&lt;/p&gt;
&lt;p&gt;the attached patch fixes a lot compiler warnings when compiling with MinGW GCC4.7.2 and also fixes two invalid comparisons of zero terminated strings via != operator in lss.c&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>EMCY: forward Manufacturer Specific error field to callback</title><link href="https://sourceforge.net/p/canfestival/patches/13/" rel="alternate"/><published>2013-03-07T14:24:27Z</published><updated>2013-03-07T14:24:27Z</updated><author><name>Thomas Hemmer</name><uri>https://sourceforge.net/u/themmer/</uri></author><id>https://sourceforge.netdf3197bdd9e3718c2f8226c6eeec112c26caa920</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;Cited from Melchior Franz's posting to the developer mailing list:&lt;/p&gt;
&lt;p&gt;"Let proceedEMCY() forward a "Manufactuer Specific Error Field" to the&lt;br /&gt;
_post_emcy callback(). Let sendEMCY() always send 8 byte long EMCY&lt;br /&gt;
messages, where the optional Error Field is either taken from a UNS8[5]&lt;br /&gt;
argument or gets zeroed out if it is NULL. This is in accordance with&lt;br /&gt;
the 301 draft in the repository and with the existing error message in&lt;br /&gt;
proceedEMCY(), which insists that EMCY messages with other lengths&lt;br /&gt;
than 8 bytes are an error."&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Placing Object Directory into Flash Memory (Atmel AVR)</title><link href="https://sourceforge.net/p/canfestival/patches/12/" rel="alternate"/><published>2012-12-24T14:08:34Z</published><updated>2012-12-24T14:08:34Z</updated><author><name>Anonymous</name><uri>https://sourceforge.net/u/userid-None/</uri></author><id>https://sourceforge.net4739c54b11b4cb3dadd9c7ee5ae799b900c2246d</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;How it works:&lt;br /&gt;
* macros R8, R16, R32 and RPTR for reading 8Bit, 16Bit, 32Bit and&lt;br /&gt;
pointers from the objdict structure. For systems without separated&lt;br /&gt;
memory they are defined with "#define R8(value) value". For AVR they are&lt;br /&gt;
defined with "#define R8(value) pgm_read_byte(&amp;amp;value)".&lt;/p&gt;
&lt;p&gt;* macro CONST for defining const values in the flash memory.&lt;br /&gt;
objdictgen.py is patched to add this keyword. For systems without&lt;br /&gt;
separted memory: "#define CONST const", for AVR "#define CONST const&lt;br /&gt;
PROGMEM".&lt;/p&gt;
&lt;p&gt;After applying this patch to changeset 765 (90b738e4ede5) you'll have to&lt;br /&gt;
regenerate your object directory with objdictgen and your object&lt;br /&gt;
directory will be placed in the flash. All the memory accesses to the&lt;br /&gt;
flash memory are handled by the library. If you have a system without&lt;br /&gt;
AVR but separated memory just adjust the macros R8, R16, R32, RPTR and&lt;br /&gt;
CONST.&lt;/p&gt;
&lt;p&gt;Patch is tested with an AVR AT90CAN128 system with a quite big object&lt;br /&gt;
directory. WARNING: You will lose the ability to dynamically add objects&lt;br /&gt;
at runtime. Don't know if this really used in applications.&lt;/p&gt;
&lt;p&gt;HTH&lt;br /&gt;
Günther Jehle&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Useless local variable res in  CanFestival\drivers\win32.c</title><link href="https://sourceforge.net/p/canfestival/patches/11/" rel="alternate"/><published>2012-01-29T18:48:19Z</published><updated>2012-01-29T18:48:19Z</updated><author><name>Jaroslav Fojtik</name><uri>https://sourceforge.net/u/fojtik/</uri></author><id>https://sourceforge.netf55cd08133016c35d00427971258d73dbd0edfcc</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;it eliminates useless variable "res".&lt;/p&gt;
&lt;p&gt;--------------ORIGINAL CODE--------------------&lt;/p&gt;
&lt;p&gt;/***************************************************************************/&lt;br /&gt;
UNS8 canSend(CAN_PORT port, Message *m)&lt;br /&gt;
{&lt;br /&gt;
UNS8 res = 1; //NOT OK&lt;br /&gt;
if (port &amp;amp;&amp;amp; (m_canSend != NULL))&lt;br /&gt;
{&lt;br /&gt;
res = m_canSend(((CANPort*)port)-&amp;gt;fd, m);&lt;br /&gt;
}&lt;br /&gt;
return res;&lt;br /&gt;
}&lt;/p&gt;
&lt;p&gt;--------------FIXED CODE--------------------&lt;/p&gt;
&lt;p&gt;/***************************************************************************/&lt;br /&gt;
/* return 0 - OK; 1 - failure */&lt;br /&gt;
UNS8 canSend(CAN_PORT port, Message *m)&lt;br /&gt;
{&lt;br /&gt;
if (port &amp;amp;&amp;amp; (m_canSend != NULL))&lt;br /&gt;
{&lt;br /&gt;
return m_canSend(((CANPort*)port)-&amp;gt;fd, m);&lt;br /&gt;
}&lt;br /&gt;
return 1; // NOT OK&lt;br /&gt;
}&lt;/p&gt;&lt;/div&gt;</summary></entry></feed>