RTOS/rtems
1
0
Fork 0
forked from Imagelibrary/rtems
Code Pull Requests Activity

Merger from CVS-trunk.

Browse Source
This commit is contained in:
Ralf Corsepius
2003-03-25 08:06:19 +00:00
parent 186d8b23ce
commit 4bab4aff51
2 changed files with 5 additions and 130 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
c/src/lib/libbsp/shared/ChangeLog
View File

@@ -1,3 +1,8 @@
2003-03-11 Ralf Corsepius <corsepiu@faw.uni-ulm.de>
* vmeUniverse/vmeUniverse.c: Remove CVS-Log (skrews up on
CVS-mergers).
2003-02-20 Till Straumann <strauman@slac.stanford.edu>
PR 349/bsps

130
c/src/lib/libbsp/shared/vmeUniverse/vmeUniverse.c
View File

@@ -5,136 +5,6 @@
* Nov 2000, Oct 2001, Jan 2002
*/
#if 0
/*
* $Log$
* Revision 1.34 2003/02/10 23:20:05 till
* - added some macro magic to make porting easier (ppcn_60x BSP in mind)
* - made mgrInstalled public (vmeUniverseIrqMgrInstalled) so BSPs can
* supply their own versions of the mgrInstall() routine.
* - added READMEs to CVS
*
* Revision 1.33.2.1 2003/02/10 23:01:40 till
* - added some macro magic to make porting easier (ppcn_60x BSP in mind)
* - made mgrInstalled public (vmeUniverseIrqMgrInstalled) so BSPs can
* supply their own versions of the mgrInstall() routine.
*
* Revision 1.32 2002/09/05 02:50:41 till
* - use k_vsprintf(), not vsprintf() during early boot (RTEMS)
*
* Revision 1.31 2002/08/16 23:35:15 strauman
* - fixed typos
*
* Revision 1.30 2002/08/16 22:53:37 till
* - made address translation more generic; allow to look for reverse mappings
* also.
* - removed vmeUniverseLocalToBus() and vmeUniverseBusToLocal() and made
* vmeUniverseXlateAddr() public instead.
*
* Revision 1.29 2002/08/16 22:16:25 till
* - tweak U2SPEC (fix Joerger vtr10012_8 problem) the same
* way the synergy BSP does (rev 2 chips only)
*
* Revision 1.28 2002/08/15 23:16:01 till
* - bugfix: vmeUniverseISRGet() dereferenced a possibly NULL pointer
*
* Revision 1.25 2002/07/19 05:18:40 till
* - CVS log: again a problem. Cannot embed /_* *_/ comments in the log
* because the log has to be commented as a whole. I had tried
* #if 0 #endif - doesn't work because cpp expands char constants???
* WHAT A NUISANCE
*
* Revision 1.24 2002/07/19 02:44:14 till
* - added a new parameter to the IRQ manager install routine:
* the main interrupt's line can now also be specified (reading
* from PCI config does not always work - some boards don't
* have correct information there - PMC's will have a similar
* problem, though!)
*
* Revision 1.21 2002/04/11 06:54:48 till
* - silenced message about 'successfully configured a port'
*
* Revision 1.20 2002/03/27 21:14:50 till
* - fix: handler table holds pointers, so hdlrTbl[vector]->usrData etc.
* not hdlrTbl[vector].usrData...
*
* Revision 1.19 2002/03/09 00:14:36 till
* - added vmeUniverseISRGet() to retrieve the currently installed
* ISR for a given vector
* - swapped the argument order for ISRs to (usrarg, vector)
*
* Revision 1.18 2002/02/07 19:53:48 till
* - reverted back to publish base_addr/irq_line as variables rather than
* through functions: the irq_line is read by the interrupt dispatcher...
*
* Revision 1.17 2002/01/24 08:28:10 till
* - initialize driver when reading base address or irq line.
* however, this requires the pci driver to be working already.
*
* Revision 1.16 2002/01/24 08:21:48 till
* - replaced public global vars for base address/irq line by routines.
*
* Revision 1.15 2002/01/23 06:15:30 till
* - changed master port data width to 64 bit.
* NOTE: reading the CY961 (Echotek ECDR814) with VDW32
* generated bus errors when reading 32-bit words
* - very weird, because the registers are 16-bit
* AFAIK.
* - 32-bit accesses worked fine on vxWorks which
* has the port set to 64-bit.
* ????????
*
* Revision 1.14 2002/01/11 19:30:54 till
* - added more register defines to header
* - completed vmeUniverseReset
*
* Revision 1.13 2002/01/11 05:06:18 till
* - fixed VMEISR failing to check (lint_stat & msk) when determining
* the highes level...
* - tested interrupt handling & nesting. Seems to work.
*
* Revision 1.12 2002/01/11 02:25:55 till
* - added interrupt manager
*
* Revision 1.11 2002/01/08 03:59:52 till
* - vxworks always defines _LITTLE_ENDIAN, fixed the conditionals
* so it should work on __vxworks and on __rtems__ now.
* - rtems uprintf wrapper reverts to printk if stdio is not yet
* initialized (uses _impure_ptr->__sdidinit)
* - tested bus address translation utility routines
*
* Revision 1.9 2002/01/05 02:36:32 till
* - added vmeUniverseBusToLocalAdrs / vmeUniverseLocalToBusAdrs for address
* space translations.
* - include bsp.h under rtems to hack around the libcpu/powerpc/shared/io.h
* #define _IO_BASE & friends problem.
*
* Revision 1.8 2002/01/04 04:12:51 till
* - changed some rtems/pci related names
*
* Revision 1.7 2002/01/04 03:06:30 till
* - added further register definitions
*
* Revision 1.6 2001/12/20 04:42:44 till
* - fixed endianness stuff; theoretically, PPC could be LITTLE_ENDIAN...
*
* Revision 1.4 2001/12/19 01:59:02 till
* - started adding interrupt stuff
* - private implementation of PCI scanning if necessary
*
* Revision 1.3 2001/07/27 22:22:51 till
* - added more DMA support routines and defines to include file
* - xxxPortsShow can now print to a given file descriptor argument
*
* Revision 1.2 2001/07/26 18:06:13 till
* - ported to RTEMS
* - fixed a couple of wrong pointer calculations.
*
* Revision 1.1.1.1 2001/07/12 23:15:19 till
* - cvs import
*/
#endif
#include <stdio.h>
#include <stdarg.h>
#include "vmeUniverse.h"