Rev |
Log message |
Author |
Age |
Path |
100 |
Some changes to support early branching: branches are now ADD #x,PC instructions
instead of MOV #x(PC),PC--providing greater range to the CPU. When that range
is insufficient, ZPARSER now recognizes long jump instructions coded as
LOD (PC),PC followed by the jump address. (This change was made necessary by
the need to build an assembler/linker that could create instructions that would
jump to any address in the 32-bit address space. In short, a part of the
ongoing GCC upgrade and rework.) |
dgisselq |
3015d 09h |
/ |
99 |
Added big-endian versus little-endian functionality. You can now specify which
your input file is as a command line parameter, and zdump will properly
disassemble the file. |
dgisselq |
3015d 09h |
/ |
98 |
Added justed longjump instructions from the previous (not used, broken)
functionality to the new LOD (PC),PC functionality. |
dgisselq |
3015d 09h |
/ |
97 |
Added longjump instructions. |
dgisselq |
3015d 09h |
/ |
96 |
Added the longjump functionality, so that the assembler will properly assemble
instructions to arbitrary 32-bit addresses. |
dgisselq |
3015d 09h |
/ |
95 |
Fixed a bug whereby a mistaken code for CLR was masking a valid LDI of a large
integer value. |
dgisselq |
3017d 15h |
/ |
94 |
These changes make it possible to build binutils-2.25/ here in this
directory. "make binutils" should be all that is necessary to build the
entire binutils package for the Zip CPU.
The default configure script, run from gas-script.sh below, will build and
install these utilities in an install/ subdirectory made below sw/. |
dgisselq |
3042d 12h |
/ |
93 |
A BINUTILS BACKEND IS NOW AVAILABLE!!!! |
dgisselq |
3050d 10h |
/ |
92 |
Adjustments made to match the simplified early branching. |
dgisselq |
3050d 10h |
/ |
91 |
Minor updates. |
dgisselq |
3050d 10h |
/ |
90 |
Removed MOV x(PC),PC from the list of possible early branching instructions.
ADD X,PC and LDI X,PC are now the only recognized early branching instructions.
This was done to spare logic, although I don't think I spared more than a
LUT or two. |
dgisselq |
3050d 10h |
/ |
89 |
Minor changes, to include making default branching an ADD.[condition] X,PC
instruction, rather than allowing both MOV X(PC),PC and ADD X,PC instructions.
Further zopcodes.cpp contains several bug fixes. |
dgisselq |
3050d 10h |
/ |
88 |
Eliminated some warnings. The div fixes were to simplify the logic, even though
the result is less readable ... |
dgisselq |
3074d 09h |
/ |
87 |
Adjusted the operator input line to reflect actual logic inputs, rather
than the registered inputs which may have been out of date. (Indeed, they
were out of date for the bug I was chasing and fixed ...) |
dgisselq |
3076d 08h |
/ |
86 |
Removed the requirement to have the dev.scope.cpu hardware defined outside
of the Zip CPU (it was defined in another project). This was causing a bus
error in the simulator (which it should have), but taking it out fixes things
in the simulator (while removing capability from one special piece of H/W). |
dgisselq |
3076d 08h |
/ |
85 |
Minor update/correction to operand B definition. |
dgisselq |
3076d 08h |
/ |
84 |
Minor updates. |
dgisselq |
3076d 08h |
/ |
83 |
Added a flag to indicate whether an exception took place on the first
or second half of a VLIW instruction--will be zero in non-VLIW mode,
equivalent to the second half of the instruction having caused the
exception. (Expect these flags to be reordered some time in the future into
a less haphazard ordering ...)
Vastly simplified the pipeline logic, primarily for op_stall, but also touched
opA and opB. (Trying to fit within timing on Spartan 6 ...)
Changed division instruction to include a reset on clear_pipeline, to make
certain [BC $addr; DIV Rx,Ry ] works regardless of whether the condition is
true. |
dgisselq |
3076d 08h |
/ |
82 |
Found and (I hope) fixed a nasty bug that would send the prefetch into an
endless loop whenever you jumped to an instruction at the last location
in an unloaded cache line. |
dgisselq |
3076d 08h |
/ |
81 |
Trying to clean up ISE generated warnings. |
dgisselq |
3076d 08h |
/ |
80 |
Bug fix: declared the (combined) multiply to be signed again. Also
changed the name of the generate'd for block, to keep ISE from complaining. |
dgisselq |
3076d 08h |
/ |
79 |
Adjusted the opcodes for NOOP, BREAK, and LOCK. |
dgisselq |
3080d 12h |
/ |
78 |
Found/corrected annoying bug in floating point documentation of the opcode
table. |
dgisselq |
3080d 12h |
/ |
77 |
First check-in: the test bench for the divide instruction. |
dgisselq |
3081d 11h |
/ |
76 |
The biggest change here was to zippy_tb, to make it more similar to the debugger
and to make it work with VLIW-type instructions. |
dgisselq |
3081d 11h |
/ |
75 |
Modified for VLIW instructions. |
dgisselq |
3081d 11h |
/ |
74 |
Added a bunch of debugging code to the Dhrystone benchmark assembly file, as
well as two new testing assembly files. |
dgisselq |
3081d 11h |
/ |
73 |
Documentations updates. |
dgisselq |
3081d 11h |
/ |
72 |
Some updated graphics, now containing images of the CPU that include the
divide and (currently non-existant) floating point unit. |
dgisselq |
3081d 11h |
/ |
71 |
This contains a bunch of bug fixes. (A lot ...) For example, the pipeline
stall code has also seriously changed, to fixed the pipeline memory load/op
stage conflict, while maintaining no-stall operation for operands that don't
need an offset. This had a cascading effect, however, so that the multiply
could no longer complete in a single cycle. Therefore, the timing on the
multiplies was slowed down to two cycles from a single cycle. (It's the
only two-cycle ALU operation ...) The illegal instruction code has also been
fixed, so that illegal instructions no longer stalls the prefetch bus. |
dgisselq |
3081d 11h |
/ |