Luke Kenneth Casson Leighton [Fri, 1 Mar 2019 22:47:31 +0000 (22:47 +0000)]
remove variable overflow
Luke Kenneth Casson Leighton [Fri, 1 Mar 2019 22:31:40 +0000 (22:31 +0000)]
connect corrections to pack without global z
Luke Kenneth Casson Leighton [Fri, 1 Mar 2019 22:25:40 +0000 (22:25 +0000)]
split roundz from norm z
Luke Kenneth Casson Leighton [Fri, 1 Mar 2019 22:14:02 +0000 (22:14 +0000)]
pass add0 z through to add1 independently
Luke Kenneth Casson Leighton [Fri, 1 Mar 2019 21:34:49 +0000 (21:34 +0000)]
add new temporary z for result chain
Luke Kenneth Casson Leighton [Fri, 1 Mar 2019 13:23:34 +0000 (13:23 +0000)]
connect add1 to norm1 overflow without global store
Luke Kenneth Casson Leighton [Fri, 1 Mar 2019 13:05:24 +0000 (13:05 +0000)]
merge normalise_1 and normalise_2 stages
Luke Kenneth Casson Leighton [Fri, 1 Mar 2019 12:01:26 +0000 (12:01 +0000)]
commennt use of intermediates
Luke Kenneth Casson Leighton [Fri, 1 Mar 2019 11:26:39 +0000 (11:26 +0000)]
store zero-extended a and b in temp signals
Luke Kenneth Casson Leighton [Fri, 1 Mar 2019 11:20:18 +0000 (11:20 +0000)]
store tests in temp signals
Luke Kenneth Casson Leighton [Fri, 1 Mar 2019 09:30:18 +0000 (09:30 +0000)]
experimenting with chaining Overflow module
Luke Kenneth Casson Leighton [Thu, 28 Feb 2019 13:15:53 +0000 (13:15 +0000)]
use output from align as input to add0
Luke Kenneth Casson Leighton [Thu, 28 Feb 2019 13:03:02 +0000 (13:03 +0000)]
remove commented-out code
Luke Kenneth Casson Leighton [Thu, 28 Feb 2019 13:02:18 +0000 (13:02 +0000)]
use GetOpMod for b
Luke Kenneth Casson Leighton [Thu, 28 Feb 2019 12:58:33 +0000 (12:58 +0000)]
create and use GetOp module
Luke Kenneth Casson Leighton [Thu, 28 Feb 2019 04:42:13 +0000 (04:42 +0000)]
move fpnum_b to class FPGetB
Luke Kenneth Casson Leighton [Thu, 28 Feb 2019 03:43:05 +0000 (03:43 +0000)]
narrowing down rounding error to use of Norm1 module
Luke Kenneth Casson Leighton [Thu, 28 Feb 2019 03:10:46 +0000 (03:10 +0000)]
separate denormalisation module and use it
Luke Kenneth Casson Leighton [Thu, 28 Feb 2019 02:48:45 +0000 (02:48 +0000)]
use denorm exponent signal
Luke Kenneth Casson Leighton [Thu, 28 Feb 2019 00:50:10 +0000 (00:50 +0000)]
sorting out unit tests, comply with IEEE754 on RISCV
Luke Kenneth Casson Leighton [Thu, 28 Feb 2019 00:13:05 +0000 (00:13 +0000)]
recompiled sfpy, testing FP16 again
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 23:09:06 +0000 (23:09 +0000)]
add Makefile patches to README
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 17:18:23 +0000 (17:18 +0000)]
whoops, overflow not right, reverting
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 17:14:21 +0000 (17:14 +0000)]
add failed test
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 17:14:17 +0000 (17:14 +0000)]
assign tests to signals
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 17:02:50 +0000 (17:02 +0000)]
create single and multi shift cycle, single doesnt work, multi does
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 15:34:32 +0000 (15:34 +0000)]
more chains between inputs and outputs
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 15:24:18 +0000 (15:24 +0000)]
move of = Overflow() out of FPADD, use chain
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 15:04:00 +0000 (15:04 +0000)]
remove tot from FPADD, use chain
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 13:58:51 +0000 (13:58 +0000)]
connect add0 to add1
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 13:56:31 +0000 (13:56 +0000)]
create add1 stage module and use it
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 13:30:36 +0000 (13:30 +0000)]
try some more chaining of inputs to outputs
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 13:21:22 +0000 (13:21 +0000)]
pass output from normalise_2 to input of roundz
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 13:15:01 +0000 (13:15 +0000)]
create add0 stage module and use it
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 12:31:15 +0000 (12:31 +0000)]
name modules correctly
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 12:29:32 +0000 (12:29 +0000)]
whoops norm2 using norm1 mod
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 12:26:16 +0000 (12:26 +0000)]
create normalise_2 module and use it
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 12:20:51 +0000 (12:20 +0000)]
put exponent > 126 logic in FPNumBase, use it in norm module
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 12:15:34 +0000 (12:15 +0000)]
split out first stage normalisation to module and use it
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 12:10:31 +0000 (12:10 +0000)]
reduce random case test numbers as well
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 11:47:15 +0000 (11:47 +0000)]
reduce number of unit test runs to get quicker more comprehensive coverage
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 11:46:54 +0000 (11:46 +0000)]
split special cases into separate module and use it
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 10:30:34 +0000 (10:30 +0000)]
create and use FPPack module
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 10:13:50 +0000 (10:13 +0000)]
create and use corrections submodule
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 10:04:06 +0000 (10:04 +0000)]
rounding done in module
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 08:07:54 +0000 (08:07 +0000)]
get roundz state to put answer in explicit output, sync it to z afterwards
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 00:29:50 +0000 (00:29 +0000)]
clean up unit_test_single get_case based on how dual_add works
Luke Kenneth Casson Leighton [Wed, 27 Feb 2019 00:12:27 +0000 (00:12 +0000)]
clear STB immediately after setting, stops add1 repeating computation
Luke Kenneth Casson Leighton [Tue, 26 Feb 2019 22:28:56 +0000 (22:28 +0000)]
moving internal strobe test forward is ok
Luke Kenneth Casson Leighton [Mon, 25 Feb 2019 08:15:57 +0000 (08:15 +0000)]
invert stb/ack between add1 and add2
Luke Kenneth Casson Leighton [Sun, 24 Feb 2019 09:27:30 +0000 (09:27 +0000)]
experimenting with dual add
Luke Kenneth Casson Leighton [Sat, 23 Feb 2019 12:57:26 +0000 (12:57 +0000)]
trying different testing for 2nd round
Luke Kenneth Casson Leighton [Sat, 23 Feb 2019 12:40:41 +0000 (12:40 +0000)]
use function to get chain of v/ack/stb
Luke Kenneth Casson Leighton [Sat, 23 Feb 2019 12:33:31 +0000 (12:33 +0000)]
yippee got dual add chained together
Luke Kenneth Casson Leighton [Sat, 23 Feb 2019 12:22:10 +0000 (12:22 +0000)]
whoops revert decode inside module FPNumIn, causing problems
Luke Kenneth Casson Leighton [Sat, 23 Feb 2019 11:56:58 +0000 (11:56 +0000)]
add dual unit test
Luke Kenneth Casson Leighton [Sat, 23 Feb 2019 11:56:28 +0000 (11:56 +0000)]
move unit test order
Luke Kenneth Casson Leighton [Sat, 23 Feb 2019 11:56:14 +0000 (11:56 +0000)]
remove unneeded class declaration
Luke Kenneth Casson Leighton [Sat, 23 Feb 2019 11:30:44 +0000 (11:30 +0000)]
add a dual-chained add experiment
Luke Kenneth Casson Leighton [Sat, 23 Feb 2019 08:43:58 +0000 (08:43 +0000)]
store logic-test conditions in intermediates
Luke Kenneth Casson Leighton [Thu, 21 Feb 2019 10:55:44 +0000 (10:55 +0000)]
isolate inputs and outputs in FPGetA class
Luke Kenneth Casson Leighton [Thu, 21 Feb 2019 10:19:31 +0000 (10:19 +0000)]
FPADD need no longer be derived from FPBase
Luke Kenneth Casson Leighton [Thu, 21 Feb 2019 10:18:15 +0000 (10:18 +0000)]
remove explicit code-adding of states, use for-loop instead
Luke Kenneth Casson Leighton [Thu, 21 Feb 2019 10:13:14 +0000 (10:13 +0000)]
move putz to separate class
Luke Kenneth Casson Leighton [Thu, 21 Feb 2019 10:11:53 +0000 (10:11 +0000)]
move pack to separate class
Luke Kenneth Casson Leighton [Thu, 21 Feb 2019 10:10:41 +0000 (10:10 +0000)]
move corrections to separate class
Luke Kenneth Casson Leighton [Thu, 21 Feb 2019 10:08:37 +0000 (10:08 +0000)]
move rounding to separate class
Luke Kenneth Casson Leighton [Thu, 21 Feb 2019 09:38:14 +0000 (09:38 +0000)]
move normalisation stages to separate classes
Luke Kenneth Casson Leighton [Thu, 21 Feb 2019 09:27:49 +0000 (09:27 +0000)]
move add1 stage to separate class
Luke Kenneth Casson Leighton [Thu, 21 Feb 2019 09:23:12 +0000 (09:23 +0000)]
add comment
Luke Kenneth Casson Leighton [Thu, 21 Feb 2019 09:22:24 +0000 (09:22 +0000)]
split out add0 stage into separate class
Luke Kenneth Casson Leighton [Thu, 21 Feb 2019 09:14:22 +0000 (09:14 +0000)]
move align to separate class
Luke Kenneth Casson Leighton [Thu, 21 Feb 2019 09:10:07 +0000 (09:10 +0000)]
create separate denormalisation class
Luke Kenneth Casson Leighton [Thu, 21 Feb 2019 08:52:12 +0000 (08:52 +0000)]
move special cases to separate state class
Luke Kenneth Casson Leighton [Thu, 21 Feb 2019 08:45:33 +0000 (08:45 +0000)]
move get_a and get_b to their own classes
Aleksandar Kostovic [Wed, 20 Feb 2019 17:50:07 +0000 (18:50 +0100)]
Remove coments with verilog code
Luke Kenneth Casson Leighton [Wed, 20 Feb 2019 05:24:06 +0000 (05:24 +0000)]
split denormalisation to separate state
Luke Kenneth Casson Leighton [Wed, 20 Feb 2019 04:54:41 +0000 (04:54 +0000)]
latch into FPNumIn within module
Luke Kenneth Casson Leighton [Wed, 20 Feb 2019 04:26:31 +0000 (04:26 +0000)]
create separate modules for fpnum in and out
Luke Kenneth Casson Leighton [Wed, 20 Feb 2019 02:52:10 +0000 (02:52 +0000)]
make module out of overflow class
Luke Kenneth Casson Leighton [Wed, 20 Feb 2019 02:30:03 +0000 (02:30 +0000)]
create module for FPNum
Luke Kenneth Casson Leighton [Wed, 20 Feb 2019 02:17:33 +0000 (02:17 +0000)]
reset allowed on FPop, not on FPNum
Luke Kenneth Casson Leighton [Wed, 20 Feb 2019 00:45:56 +0000 (00:45 +0000)]
store roundz test in comb variable
Luke Kenneth Casson Leighton [Tue, 19 Feb 2019 15:52:13 +0000 (15:52 +0000)]
store testing of nan/inf/zero in comb Signals
Luke Kenneth Casson Leighton [Tue, 19 Feb 2019 15:51:20 +0000 (15:51 +0000)]
move setting of stb into else block
Luke Kenneth Casson Leighton [Tue, 19 Feb 2019 15:27:12 +0000 (15:27 +0000)]
reset_less on signals that do not need it
Luke Kenneth Casson Leighton [Tue, 19 Feb 2019 12:51:50 +0000 (12:51 +0000)]
reorganise unit test single to do much more comprehensive test cases.
specific edge cases on the exponent are covered, with random mantissas:
-126, -127, 127, 128
Luke Kenneth Casson Leighton [Tue, 19 Feb 2019 11:23:22 +0000 (11:23 +0000)]
take out FP16 non-canonical NaN weirdness for now
Luke Kenneth Casson Leighton [Tue, 19 Feb 2019 09:20:13 +0000 (09:20 +0000)]
add corner-cases +/-0 + NaN
Luke Kenneth Casson Leighton [Tue, 19 Feb 2019 08:24:20 +0000 (08:24 +0000)]
add FP16 add unit test
Luke Kenneth Casson Leighton [Tue, 19 Feb 2019 08:23:20 +0000 (08:23 +0000)]
INF + -INF bug
Luke Kenneth Casson Leighton [Tue, 19 Feb 2019 08:05:29 +0000 (08:05 +0000)]
whoops FP16 mantissa off-by-one
Luke Kenneth Casson Leighton [Tue, 19 Feb 2019 07:41:35 +0000 (07:41 +0000)]
remove hard-coded width
Luke Kenneth Casson Leighton [Tue, 19 Feb 2019 07:41:23 +0000 (07:41 +0000)]
add FP16 format
Luke Kenneth Casson Leighton [Tue, 19 Feb 2019 06:41:50 +0000 (06:41 +0000)]
add shift up multi function
Luke Kenneth Casson Leighton [Tue, 19 Feb 2019 05:41:28 +0000 (05:41 +0000)]
add extra regression tests (a + -a) for add
Luke Kenneth Casson Leighton [Tue, 19 Feb 2019 05:39:14 +0000 (05:39 +0000)]
comment for a + -a special case add
Luke Kenneth Casson Leighton [Mon, 18 Feb 2019 21:32:38 +0000 (21:32 +0000)]
add 64 bit mul unit test
Luke Kenneth Casson Leighton [Mon, 18 Feb 2019 21:32:20 +0000 (21:32 +0000)]
whoops, off-by-one in use of mw, in multiply_1 stage
Luke Kenneth Casson Leighton [Mon, 18 Feb 2019 21:26:52 +0000 (21:26 +0000)]
whoops, messing up on m_width *sigh*