Rev |
Log message |
Author |
Age |
Path |
60 |
Selection of memory model or implementation memory is now made on minsoc_bench_defines.v. It is done by a definition instead of including different files for simulation.
minsoc_bench_defines.v definition of reset level was not correct. It based the level decision on defineds POSITIVE_RESET or NEGATIVE_RESET, which couldn't be defined by then, since minsoc_defines.v is not included in minsoc_bench_defines.v. The decision has been moved to minsoc_bench.v and made a localparam instead of a definition. |
rfajardo |
4868d 21h |
/minsoc/trunk/ |
59 |
undefinition of NEGATIVE_RESET on minsoc_bench_defines.v cannot affect other inclusions of minsoc_defines.v. Instead, the testbench now works with the right reset level to avoid the implementation ordering problem. |
rfajardo |
4868d 21h |
/minsoc/trunk/ |
58 |
Standard definitions depended on implementation order. Now, this should be solved.
minsoc_bench_defines.v: when setting reset to be positive (`define POSITIVE_RESET), NEGATIVE_RESET is undefined. This override the implementation order, so that independent of it, POSITIVE_RESET will be used.
minsoc_defines.v: when setting GENERIC_FPGA, FPGA_TAP and FPGA_CLOCK_DIVISION are undefined. This way, even if FPGA_TAP would come prior to GENERIC_TAP on the correspondent implementation, GENERIC_TAP would still be selected.
IMPORTANT: GENERIC_MEMORY must still be implemented first on minsoc_onchip_ram.v, because FPGA's memory is automatically selected from other definitions and cannot be undefined a priori. Since some other memory types can be selected, there is no trivial solution. However, this shouldn't be a big problem, since the beginning of this file will probably not be modified. |
rfajardo |
4868d 21h |
/minsoc/trunk/ |
57 |
If a FPGA manufacturer is selected, the FPGA families of other manufacturers are automatically ignored.
Some updated to comments.
CLOCK_DIVISOR back to 5. The number does not matter much, but 1 is a bad standard number, since it should never be selected. Comment says, use NO_CLOCK_DIVISION instead.
Changing standard FPGA back to Xilinx and Spartan3A. I'm only doing this because the synthesis examples page of wiki still assume this FPGA to be standardly selected. |
rfajardo |
4868d 22h |
/minsoc/trunk/ |
56 |
Macros for all Altera family devices and pll instantiation |
javieralso |
4875d 21h |
/minsoc/trunk/ |
55 |
Adjusting Makefiles to compile correctly with new firmware updates.
1) except.o not included into libsupport.a.
2) libsupport.a linked to except.o, now in correct order: ld libsupport.a except.o -o executable (not the other way around)
3) int.c was not being compiled because of a #ifdef which is not set anywhere |
rfajardo |
4877d 05h |
/minsoc/trunk/ |
54 |
Moving spr_defs.h to or1200.h |
ConX. |
4877d 07h |
/minsoc/trunk/ |
53 |
Indentation, deleting redundant files and adding externals |
ConX. |
4877d 08h |
/minsoc/trunk/ |
52 |
Altera ALTPLL Megafunction Instantiation |
javieralso |
4885d 22h |
/minsoc/trunk/ |
51 |
sw/support/uart.c: Changing the order of writes to the Divisor Latch of UART. (Thanks Ramkumar) |
rfajardo |
4892d 11h |
/minsoc/trunk/ |
50 |
Removing unused firmware files, respective to or1ksim actually.
Removing the inclusion of the removed file mc.h in reset.S, probably required by or1ksim at some point.
Reworked except.S to use a macro instead of repeating the same procedure 16 times or so. Explanation added to the macro as a leading comment. |
rfajardo |
4904d 07h |
/minsoc/trunk/ |
49 |
Language correction for README.txt. |
rfajardo |
4906d 05h |
/minsoc/trunk/ |
48 |
Clear some old docs that are already ported to MinSOC's Wiki |
ConX. |
4906d 06h |
/minsoc/trunk/ |
47 |
Firmware updated to work with gcc-4.5. It is actually working just fine and gcc-4.5 assembly code seems a lot cleaner.
I noticed that by increasing the free stack space for the interrupt handler by 130 made the resulting firmware not work. I assume it is because 130 is not aligned 130%4 = 2 and the stack simply does not work then. Instead I'm freeing the previous 116 plus 128 (for 32 registers as mentioned by Jeremy) and it works. |
rfajardo |
4907d 05h |
/minsoc/trunk/ |
46 |
Including an explanation of what has to be updated on gpio to port it smoothly to minsoc. |
rfajardo |
4907d 09h |
/minsoc/trunk/ |
45 |
A more stable version |
ConX. |
4908d 00h |
/minsoc/trunk/ |
44 |
Fixing some bugs. But it still works only in Debian/Ubuntu |
ConX. |
4908d 02h |
/minsoc/trunk/ |
43 |
Making some changes to MinSOC install script |
ConX. |
4908d 22h |
/minsoc/trunk/ |
41 |
Including setup scripts to install all required tools to work with minsoc and to download all required sources. Thanks for the contribution of Xanthopoulos Constantinos. |
rfajardo |
4913d 05h |
/minsoc/trunk/ |
40 |
Commiting a contributions directory, which has raw contributions of users. These contributions still have to be adapted to the system in order to work well. However, some users are certainly able to work with these versions.
FAQ: -added new idea to circumvent onboard DLC9 cable problem for adv_jtag_bridge
INSTALL: -gives a hint on how to install the system for Windows
HOWTO: -explains what a ucf file is and how to create it
backend: now we have the first contribution of a ucf file for the ML509 board (thanks to Matthew Hick)
Documentation: THESIS.txt explains to users, willing to tweak/understand the inner behavior of the OR1200 implementation of OpenRISC, what to read in order to get this information. |
rfajardo |
4919d 04h |
/minsoc/trunk/ |
39 |
FAQ:
-Adv_jtag_bridge self test fails?
-included now SoC flow mistakes
-forgetting to upload the bitfile
-leaving clock unconnected
-older information is now under "Known Issues" right below of SoC flow mistakes
Software:
-eth: included eth.h is from drivers and not from this directory. eth.h from this directory is old and has been removed.
-uart: included uart.h is from drivers and not from this directory. uart.h from this directory is old and has been removed. |
rfajardo |
4983d 20h |
/minsoc/trunk/ |
38 |
Small update to HOWTO: advices now to also include the ucf (pinout) file, for forgetful people :-).
FAQ: Added another option (hint) as a solution for people with on-board Xilinx USB cables, which do not reset.
uart.c: now uses the IRQ line definition instead of the line number directly. |
rfajardo |
5018d 20h |
/minsoc/trunk/ |
37 |
README.txt added, describing the installation and set-up processes. Also describing to use the FAQ or forum when you have problems and finally to read minsoc.pdf (the documentation) to understand the system after everything works, to know what to do next.
Clean-up of sw/utils, removing unused sources and files.
Update of sw/drivers/eth.c, direct casting to avoid compile warnings.
FAQ extended and with more links to the threads giving the solutions.
Synthesis examples inform that the firmware tweak for Spartan3E Starter Kit using Ethernet only works for system version up to 35.
minsoc.pdf and webpage now uses FPGA generic and specific codes instead of FPGA independent and dependent codes, because it suits it better. |
rfajardo |
5049d 07h |
/minsoc/trunk/ |
36 |
utils: -adding a 32 to 8 bit Wishbone bridge to connect 8 bit modules to the MinSoC interconnect.
-adding a Linux driver for parallel cables for the adv_jtag_bridge. It circumvents the necessity of privilege to run adv_jtag_bridge with cables xpc3 and xess.
sw/support: support library has been made completely standalone. That means, it is now independent from uClib and/or newlib. Thanks to Wojciech A. Koszek for his contribution on this.
sw/drivers: a driver library has been included, which supply drivers for the OpenRISC timer; and for UART, Ethernet, I2C, and CAN modules from OpenCores.
sw/uart and sw/eth: they don't provide code to control the UART and Ethernet controllers anymore. They use the drivers library instead.
Documentation:
-FAQ extended:
-table of contents
-explanation of how to use the 32 to 8 bit Wishbone bridge
-how to make adv_jtag_bridge work without privilege requirement for xess or xpc3
-explanation of locking problem of xpc_usb cable
-explanation of what to do if the cable is built on-board
-INSTALL:
-patching of gdb only required if version is 6.8, explanation on FAQ now.
-HOWTO:
-instruct to compile drivers library
-remove line numbers information on what to edit, better leave it to be found only by the informed context.
-adv_jtag_bridge execution now without 'sudo'. Note included that 'sudo' is required for Linux and xess or xpc3. Reference to FAQ to avoid this requirement. |
rfajardo |
5050d 02h |
/minsoc/trunk/ |
35 |
minsoc.pdf: -adjusting the date on the page header of MinSoC documentation. It was correct on the front and revision pages, now header adjusted accordingly. |
rfajardo |
5054d 20h |
/minsoc/trunk/ |
34 |
start_server changed: '-t' option of adv_jtag_bridge for vpi connection on simulation removed. or1200_v3 will not pass on CPU self test.
FAQ completed with asked questions since Februrary 2010.
INSTALL informs bsdl files only have to be copied to home directory for Xilinx devices.
synthesis_examples title includes Minimal OpenRISC System on Chip. |
rfajardo |
5064d 05h |
/minsoc/trunk/ |
33 |
Start-up Starter, included in the MinSoC top file, has been updated to three-phase instruction output; instruction assertion, acknowledge assertion, and next instruction with acknowledge deassertion.
Previously it was two-phase with next instruction, instruction assertion and acknowledge assertion together, and acknowledge deassertion.
That is required by the new Wishbone master interface used by OpenRISC release 3. |
rfajardo |
5072d 06h |
/minsoc/trunk/ |
32 |
Documentation revision 1.1, thanks to Wojciech A. Koszek for many comments on it.
Also updating howto, splitint it in INSTALL, HOWTO, FAQ and synthesis_examples, so it should be more clear now what to do when and not to try too much when you don't need. Like everyone was trying to debug the simulation but didn't even test the regular simulation before. Again thanks to Wojciech A. Koszek for his view on this matter. |
rfajardo |
5086d 04h |
/minsoc/trunk/ |
31 |
Adaption to or1200_r3. It is still important to change or1200_defines.v:
-`define OR1200_BOOT_ADR 32'hf0000100 to `define OR1200_BOOT_ADR 32'h00000100 |
rfajardo |
5141d 11h |
/minsoc/trunk/ |
30 |
minsoc SoC documentation had 2 small typo corrections. Performance penalty due to addition of register addresses was a wrong assumption. On project description "is composed by" -> "consists of". Thanks to Wojciech A. Koszek.
howto, at some places the howto did not tell the path from the files being talked about. I tried to always specify the path for every commented file.
Scripts for running the simulation called bash instead of sh. For compatibility reasons sh is now used, this should affect noone. Scripts do not use bash specific commands and generally every UNIX like computer has sh. Thanks again to Wojciech A. Koszek, who adapted that to port it to FreeBSD. |
rfajardo |
5184d 09h |
/minsoc/trunk/ |