Rev |
Log message |
Author |
Age |
Path |
118 |
Lots of changes. The biggest are to the CPU: MPYxHI instructions are now
verified to be working, ILLegal instructions stop at the right location,
the STEP bit no longer self-clears, etc. Other changes cleaned up the
internal documentation and removed parameters that should only have local
scope from the global parameter list. The NEW_INSTRUCTION_SET was also
removed from the CPU, since ... it's been new for too long to really be new
anymore. |
dgisselq |
2775d 18h |
/xulalx25soc/trunk/rtl/ |
117 |
Updates, to include new README and bench/cpp/Makefile that doesnt depend upon
a static VERILATOR_ROOT location. |
dgisselq |
2800d 10h |
/xulalx25soc/trunk/rtl/ |
114 |
Added ZipBones to the list of dependencies, so this will (should) build
properly for the XULA9 as well as the XULA25. |
dgisselq |
2893d 11h |
/xulalx25soc/trunk/rtl/ |
113 |
Lots of updates, mostly focused on getting the XuLA board and CPU able to pass
a comprehensive test suite. At this point, everything passes and logic usage
is down even. Among the few changes includes the "break" bit in the uCC
register, used to indicate a switch to supervisor mode occurred as a result
of a user break, and the ability for the supervisor to clear the instruction
cache. |
dgisselq |
2893d 11h |
/xulalx25soc/trunk/rtl/ |
109 |
This continues the updates to the wishbone-uart conversion. It fixes several
bugs within wbuexec, and pipelines the compression scheme. Further, the
read codeword was adjusted so that a read of 8 can be requested with six-bits,
rather than requiring 12. Likewise, the dependence upon the read of 8 on
incrementing the address pointer has been removed. All told, the design
builds for a 200MHz Artix-7, and it has been tested with the CMod-S6. (Writing
flash seems to be one of the most comprehensive tests ...) |
dgisselq |
2901d 19h |
/xulalx25soc/trunk/rtl/ |
108 |
Minor documentation updates. |
dgisselq |
2902d 05h |
/xulalx25soc/trunk/rtl/ |
106 |
Minor, inconsequential changes. |
dgisselq |
2902d 05h |
/xulalx25soc/trunk/rtl/ |
102 |
Updated documentation. The documentation for these now also reflects that
these were drawn from an FPGA Library project, that is shared among many
FPGA builds. |
dgisselq |
2902d 06h |
/xulalx25soc/trunk/rtl/ |
101 |
Fixed the `defines at the top so that this can be built without any CPU.
This was then used to measure the impact of the CPU on the entire build, as
you could now build with no CPU, and then with a CPU to compare. |
dgisselq |
2902d 06h |
/xulalx25soc/trunk/rtl/ |
100 |
Includes updates so this can run at higher clocks speeds within an FPGA. |
dgisselq |
2902d 06h |
/xulalx25soc/trunk/rtl/ |
99 |
Includes high-speed updates. rxuart and txuart will now run at 200MHz on
an Artix-7, so should not impact timing (any more) on the XuLA platform. |
dgisselq |
2902d 06h |
/xulalx25soc/trunk/rtl/ |
98 |
Updated copyright notices for the new year, to reflect that changes have been
made in 2016. |
dgisselq |
2902d 06h |
/xulalx25soc/trunk/rtl/ |
90 |
Reads and writes to the SD over SPI port now work. The card appears, as of now,
to be fully functional. |
dgisselq |
2925d 10h |
/xulalx25soc/trunk/rtl/ |
89 |
Bug fixes, following the adf_ce logic combining ALU/DIVIDE/FPU pipeline logic
into one register, this fixes that logic so that instructions without their
condition fulfilled are still "executed" and marked as done. |
dgisselq |
2925d 10h |
/xulalx25soc/trunk/rtl/ |
88 |
Adjusted copyright date. |
dgisselq |
2925d 10h |
/xulalx25soc/trunk/rtl/ |
87 |
Placed the interrupt into the carry chain for less logic area. |
dgisselq |
2925d 10h |
/xulalx25soc/trunk/rtl/ |
86 |
Fine tuning the `defines, so that you can build pipelined without pipelined
bus access and so forth. |
dgisselq |
2925d 10h |
/xulalx25soc/trunk/rtl/ |
85 |
First version of the SD-SPI interface, with partial functionality. (No the
empty link that was here before.) |
dgisselq |
2929d 07h |
/xulalx25soc/trunk/rtl/ |
84 |
First part of switching to proper sdspi.v, and not just the link. |
dgisselq |
2929d 07h |
/xulalx25soc/trunk/rtl/ |
83 |
Fixes a bug in the LX9 build whereby the flash was never ever granted permission
to use the SPI port. |
dgisselq |
2930d 11h |
/xulalx25soc/trunk/rtl/ |
74 |
Adds the SD-card capability, and connects the debug wires to/from the uartdev
in case it needs to be debugged. |
dgisselq |
2931d 06h |
/xulalx25soc/trunk/rtl/ |
73 |
Simplified logic. |
dgisselq |
2931d 06h |
/xulalx25soc/trunk/rtl/ |
72 |
Sets XULA25 as the default. |
dgisselq |
2931d 06h |
/xulalx25soc/trunk/rtl/ |
71 |
Needed to play with subtle timing to get this to build. Expect me to play
with these two clock numbers more. |
dgisselq |
2931d 06h |
/xulalx25soc/trunk/rtl/ |
70 |
Cosmetic (minor) update. |
dgisselq |
2931d 06h |
/xulalx25soc/trunk/rtl/ |
69 |
Massive logic simplification. This is also the first (verified) working
version. |
dgisselq |
2931d 07h |
/xulalx25soc/trunk/rtl/ |
68 |
Fixes the debug ack line, so it no longer acks when there isn't a dbg_stb.
Fixed the multiply option parameter, so it sets a 3-clock multiply properly.
Adjusted the watchdog timer so that it now produces a timer that doesn't
reload--since this is pointless for a watchdog. Finally, connects a reset
line to the DMA, to make certain that resetting the CPU will also reset any
ongoing DMA operation. |
dgisselq |
2931d 07h |
/xulalx25soc/trunk/rtl/ |
67 |
Simplifies logic, and guarantees that the minimum set value will always
produce an int. For example, if the count was X before, setting X-1 wouldn't
produce an interrupt since it had passed. Now it produces an interrupt, and
keeps the next interrupt valid. |
dgisselq |
2931d 07h |
/xulalx25soc/trunk/rtl/ |
66 |
Simplified logic (barely). |
dgisselq |
2931d 07h |
/xulalx25soc/trunk/rtl/ |
65 |
Makes the auto-reload feature a parameterized (generic) feature, so the same
code will work for both auto-reloadable and non-autoreloadable (i.e. watchdog)
timers. |
dgisselq |
2931d 07h |
/xulalx25soc/trunk/rtl/ |