ieee754fpu.git
4 years agoRemove unneeded and gate in gt_combiner
Michael Nolan [Fri, 7 Feb 2020 14:58:32 +0000 (09:58 -0500)]
Remove unneeded and gate in gt_combiner

The and gate that forces the lanes that are not the MSB in a partiton
to 0 is no longer needed because reorder_bits ignores those bits.

4 years agofix syntax errors for test_partsig
Luke Kenneth Casson Leighton [Fri, 7 Feb 2020 14:47:07 +0000 (14:47 +0000)]
fix syntax errors for test_partsig

4 years agogrr correct syntax for assert
Luke Kenneth Casson Leighton [Fri, 7 Feb 2020 14:46:49 +0000 (14:46 +0000)]
grr correct syntax for assert

4 years agoFix != implementation in partsig.py
Michael Nolan [Fri, 7 Feb 2020 14:38:47 +0000 (09:38 -0500)]
Fix != implementation in partsig.py

4 years agoModify test_partsig to handle the new changes to reorder_bits, add new testcases
Michael Nolan [Fri, 7 Feb 2020 14:30:30 +0000 (09:30 -0500)]
Modify test_partsig to handle the new changes to reorder_bits, add new testcases

Removes != test because it's not actually working yet

4 years agomodify reorder_bits to copy the MSB of the partition to each bit
Michael Nolan [Fri, 7 Feb 2020 14:29:32 +0000 (09:29 -0500)]
modify reorder_bits to copy the MSB of the partition to each bit

Previously, it would move the MSB to the LSB, and set the rest of the
bits to 0.

4 years agoadd unit test for part_mux
Luke Kenneth Casson Leighton [Fri, 7 Feb 2020 14:38:08 +0000 (14:38 +0000)]
add unit test for part_mux

4 years agowhoops syntax error
Luke Kenneth Casson Leighton [Fri, 7 Feb 2020 14:38:00 +0000 (14:38 +0000)]
whoops syntax error

4 years agoadd first version of part_mux
Luke Kenneth Casson Leighton [Fri, 7 Feb 2020 14:26:40 +0000 (14:26 +0000)]
add first version of part_mux

4 years agouse assert not raise ValueError
Luke Kenneth Casson Leighton [Fri, 7 Feb 2020 14:18:21 +0000 (14:18 +0000)]
use assert not raise ValueError

4 years agoripple of LSB from compare-operations, successful
Luke Kenneth Casson Leighton [Fri, 7 Feb 2020 14:07:45 +0000 (14:07 +0000)]
ripple of LSB from compare-operations, successful

4 years agoadd first (untested) version of ripple-LSB
Luke Kenneth Casson Leighton [Fri, 7 Feb 2020 14:06:09 +0000 (14:06 +0000)]
add first (untested) version of ripple-LSB

4 years agoadd extra operators to be implemented to partsig
Luke Kenneth Casson Leighton [Fri, 7 Feb 2020 13:56:01 +0000 (13:56 +0000)]
add extra operators to be implemented to partsig

4 years agoadd stub operators to partsig to help make it clear what is needed
Luke Kenneth Casson Leighton [Thu, 6 Feb 2020 18:07:20 +0000 (18:07 +0000)]
add stub operators to partsig to help make it clear what is needed

4 years agoTODO on NE partsig op
Luke Kenneth Casson Leighton [Thu, 6 Feb 2020 16:28:34 +0000 (16:28 +0000)]
TODO on NE partsig op

4 years agomake common function for testing comparators
Luke Kenneth Casson Leighton [Thu, 6 Feb 2020 14:57:57 +0000 (14:57 +0000)]
make common function for testing comparators

4 years agorestructure partsig, use common function for PartitionedEqGtGe
Luke Kenneth Casson Leighton [Thu, 6 Feb 2020 14:41:04 +0000 (14:41 +0000)]
restructure partsig, use common function for PartitionedEqGtGe

4 years agofix partsig gt test, add ge test
Luke Kenneth Casson Leighton [Thu, 6 Feb 2020 14:34:41 +0000 (14:34 +0000)]
fix partsig gt test, add ge test

4 years agoadd __ge__ partsig operator
Luke Kenneth Casson Leighton [Thu, 6 Feb 2020 14:31:42 +0000 (14:31 +0000)]
add __ge__ partsig operator

4 years agoadd gt part_sig operator
Luke Kenneth Casson Leighton [Thu, 6 Feb 2020 14:27:40 +0000 (14:27 +0000)]
add gt part_sig operator

4 years agoFix test_partsig's invalid result for the equals test
Michael Nolan [Thu, 6 Feb 2020 02:11:25 +0000 (21:11 -0500)]
Fix test_partsig's invalid result for the equals test

4 years agoAdd 0 assertions to the proof for eq_gt_ge
Michael Nolan [Thu, 6 Feb 2020 02:06:49 +0000 (21:06 -0500)]
Add 0 assertions to the proof for eq_gt_ge

4 years agoFix order of results from PartitionedEqGtGe
Michael Nolan [Wed, 5 Feb 2020 23:44:00 +0000 (18:44 -0500)]
Fix order of results from PartitionedEqGtGe

4 years agomove unused partitioned comparison modules to experiments/
Michael Nolan [Wed, 5 Feb 2020 15:54:24 +0000 (10:54 -0500)]
move unused partitioned comparison modules to experiments/

4 years agomove gt_combiner out of experiments/
Michael Nolan [Wed, 5 Feb 2020 15:52:47 +0000 (10:52 -0500)]
move gt_combiner out of experiments/

4 years agoadd extra tests, find out whats going on with eq_gt_ge
Luke Kenneth Casson Leighton [Wed, 5 Feb 2020 20:43:46 +0000 (20:43 +0000)]
add extra tests, find out whats going on with eq_gt_ge

4 years agoquick test shows eq_gt_ge.py returning output in reverse order
Luke Kenneth Casson Leighton [Wed, 5 Feb 2020 17:38:08 +0000 (17:38 +0000)]
quick test shows eq_gt_ge.py returning output in reverse order

4 years agodebug string for part-eq is an eq not +
Luke Kenneth Casson Leighton [Wed, 5 Feb 2020 17:06:06 +0000 (17:06 +0000)]
debug string for part-eq is an eq not +

4 years agouse PartitionedEqGtGe for eq function in partitioned signal
Luke Kenneth Casson Leighton [Wed, 5 Feb 2020 17:00:55 +0000 (17:00 +0000)]
use PartitionedEqGtGe for eq function in partitioned signal

4 years agoadd constants for opcode, EQ/GT/GE, probably better way to do this
Luke Kenneth Casson Leighton [Wed, 5 Feb 2020 16:58:52 +0000 (16:58 +0000)]
add constants for opcode, EQ/GT/GE, probably better way to do this

4 years agowhitespace
Luke Kenneth Casson Leighton [Wed, 5 Feb 2020 16:54:16 +0000 (16:54 +0000)]
whitespace

4 years ago80 char limit
Luke Kenneth Casson Leighton [Wed, 5 Feb 2020 16:53:48 +0000 (16:53 +0000)]
80 char limit

4 years agowhitespace
Luke Kenneth Casson Leighton [Wed, 5 Feb 2020 16:51:58 +0000 (16:51 +0000)]
whitespace

4 years agomore whitespace
Luke Kenneth Casson Leighton [Wed, 5 Feb 2020 16:49:26 +0000 (16:49 +0000)]
more whitespace

4 years agowhitespace
Luke Kenneth Casson Leighton [Wed, 5 Feb 2020 16:48:28 +0000 (16:48 +0000)]
whitespace

4 years agoAdd module to handle partitioned eq, gt, and ge comparisons
Michael Nolan [Wed, 5 Feb 2020 15:37:14 +0000 (10:37 -0500)]
Add module to handle partitioned eq, gt, and ge comparisons

4 years agoReverse order of gt combiner so it works MSB first
Michael Nolan [Wed, 5 Feb 2020 15:06:40 +0000 (10:06 -0500)]
Reverse order of gt combiner so it works MSB first

4 years agoAdd a signal to disable the gt inputs to gt_combiner
Michael Nolan [Wed, 5 Feb 2020 14:40:56 +0000 (09:40 -0500)]
Add a signal to disable the gt inputs to gt_combiner

4 years agoAdd proof for equals module
Michael Nolan [Wed, 5 Feb 2020 14:27:12 +0000 (09:27 -0500)]
Add proof for equals module

4 years agouse EQCombiner in PartitionedEq experiment
Luke Kenneth Casson Leighton [Wed, 5 Feb 2020 13:34:05 +0000 (13:34 +0000)]
use EQCombiner in PartitionedEq experiment

4 years agostart partitioned mux class
Luke Kenneth Casson Leighton [Tue, 4 Feb 2020 18:50:49 +0000 (18:50 +0000)]
start partitioned mux class

4 years agoadd experiments __init__.py
Luke Kenneth Casson Leighton [Tue, 4 Feb 2020 18:12:04 +0000 (18:12 +0000)]
add experiments __init__.py

4 years agoremove comments, sort out string name, add __init__.pys
Luke Kenneth Casson Leighton [Tue, 4 Feb 2020 18:11:25 +0000 (18:11 +0000)]
remove comments, sort out string name, add __init__.pys

4 years agobit of whitespace
Luke Kenneth Casson Leighton [Tue, 4 Feb 2020 15:12:02 +0000 (15:12 +0000)]
bit of whitespace

4 years agousing format is a bit long for 1 argument, just use %d
Luke Kenneth Casson Leighton [Tue, 4 Feb 2020 15:09:11 +0000 (15:09 +0000)]
using format is a bit long for 1 argument, just use %d

4 years agoModify gt experiment to handle eq as well
Michael Nolan [Tue, 4 Feb 2020 03:28:50 +0000 (22:28 -0500)]
Modify gt experiment to handle eq as well

4 years agoAdd proof for using the greater than combiner to do equals as well
Michael Nolan [Mon, 3 Feb 2020 20:16:05 +0000 (15:16 -0500)]
Add proof for using the greater than combiner to do equals as well

It works for the active outputs, however some of the active outputs
get set to 1 instead of 0

4 years agoAdd tree-based greater than experiment
Michael Nolan [Mon, 3 Feb 2020 20:00:53 +0000 (15:00 -0500)]
Add tree-based greater than experiment

4 years agoMove experiments with partition methods to a separate folder
Michael Nolan [Mon, 3 Feb 2020 19:41:36 +0000 (14:41 -0500)]
Move experiments with partition methods to a separate folder

4 years agoAdd tree based partitioned equals module
Michael Nolan [Mon, 3 Feb 2020 19:34:29 +0000 (14:34 -0500)]
Add tree based partitioned equals module

4 years agoRename partition_combiner to eq_combiner
Michael Nolan [Mon, 3 Feb 2020 17:13:20 +0000 (12:13 -0500)]
Rename partition_combiner to eq_combiner

4 years agoAdd an alternative partitioned equals combiner module
Michael Nolan [Mon, 3 Feb 2020 15:41:00 +0000 (10:41 -0500)]
Add an alternative partitioned equals combiner module

4 years agoadd first untested ge.py for partitioned signnal
Luke Kenneth Casson Leighton [Mon, 3 Feb 2020 11:51:51 +0000 (11:51 +0000)]
add first untested ge.py for partitioned signnal

4 years agoHandle -0 and +0 equals and <
Michael Nolan [Sun, 2 Feb 2020 19:41:56 +0000 (14:41 -0500)]
Handle -0 and +0 equals and <

4 years agoAdd test case for NaN and -0 handling to fpmax
Michael Nolan [Sun, 2 Feb 2020 18:37:10 +0000 (13:37 -0500)]
Add test case for NaN and -0 handling to fpmax

4 years agoReplace If and Switch statements with muxes in fpcmp.py
Michael Nolan [Sun, 2 Feb 2020 17:47:50 +0000 (12:47 -0500)]
Replace If and Switch statements with muxes in fpcmp.py

4 years agoAdd formal proof for FLT and FLE for FPCMP
Michael Nolan [Sun, 2 Feb 2020 17:31:17 +0000 (12:31 -0500)]
Add formal proof for FLT and FLE for FPCMP

4 years agoAdd FLT and FLE functionality to FPCMP
Michael Nolan [Sun, 2 Feb 2020 16:59:20 +0000 (11:59 -0500)]
Add FLT and FLE functionality to FPCMP

4 years agoHandle NaNs for FPCMP
Michael Nolan [Sun, 2 Feb 2020 16:42:17 +0000 (11:42 -0500)]
Handle NaNs for FPCMP

4 years agoAdd feq functionality to fpcmp
Michael Nolan [Sun, 2 Feb 2020 16:29:48 +0000 (11:29 -0500)]
Add feq functionality to fpcmp

4 years agoBegin adding formal proof for fpcmp
Michael Nolan [Sun, 2 Feb 2020 16:29:20 +0000 (11:29 -0500)]
Begin adding formal proof for fpcmp

4 years agoBegin adding FPCMP module (FEQ, FLT, FLE)
Michael Nolan [Sun, 2 Feb 2020 16:21:01 +0000 (11:21 -0500)]
Begin adding FPCMP module (FEQ, FLT, FLE)

4 years agoFix nmigen "domain" error when trying to run unit tests
Michael Nolan [Fri, 31 Jan 2020 21:54:25 +0000 (16:54 -0500)]
Fix nmigen "domain" error when trying to run unit tests

When I try to run any unit test with nmigen c42c3a0, I get the
following error:

Traceback (most recent call last):
  File "/usr/lib/python3.8/runpy.py", line 193, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/usr/lib/python3.8/runpy.py", line 86, in _run_code
    exec(code, run_globals)
  File "/home/mnolan/git/ieee754fpu/src/ieee754/fpmul/test/test_fpmul_pipe.py", line 24, in <module>
    test_pipe_fp16()
  File "/home/mnolan/git/ieee754fpu/src/ieee754/fpmul/test/test_fpmul_pipe.py", line 12, in test_pipe_fp16
    runfp(dut, 16, "test_fpmul_pipe_fp16", Float16, mul, n_vals=50)
  File "/home/mnolan/git/ieee754fpu/src/ieee754/fpcommon/test/fpmux.py", line 316, in runfp
    run_simulation(dut, fns, vcd_name="%s.vcd" % name)
  File "/home/mnolan/git/nmigen/nmigen/compat/sim/__init__.py", line 22, in run_simulation
    fragment.domains += ClockDomain("sync")
AttributeError: 'FPMULMuxInOut' object has no attribute 'domains'

This seems to fix this error by giving the functions in the testbench
a default clock domain, rather than leaving it up to nmigen to figure
out.

4 years agotidyup
Luke Kenneth Casson Leighton [Thu, 30 Jan 2020 18:52:57 +0000 (18:52 +0000)]
tidyup

4 years agotidy up comments
Luke Kenneth Casson Leighton [Thu, 30 Jan 2020 18:40:13 +0000 (18:40 +0000)]
tidy up comments

4 years agoworking partitioned eqs function
Luke Kenneth Casson Leighton [Thu, 30 Jan 2020 18:24:27 +0000 (18:24 +0000)]
working partitioned eqs function

4 years agowhoops overwrote input variable
Luke Kenneth Casson Leighton [Thu, 30 Jan 2020 16:04:53 +0000 (16:04 +0000)]
whoops overwrote input variable

4 years agoadd comments
Luke Kenneth Casson Leighton [Thu, 30 Jan 2020 13:57:46 +0000 (13:57 +0000)]
add comments

4 years agoexperimental eq (non-optimal)
Luke Kenneth Casson Leighton [Thu, 30 Jan 2020 13:49:22 +0000 (13:49 +0000)]
experimental eq (non-optimal)

4 years agoadd "as_sig" function to PartitionPoints to get the mask as a straight sequence
Luke Kenneth Casson Leighton [Thu, 30 Jan 2020 13:48:58 +0000 (13:48 +0000)]
add "as_sig" function to PartitionPoints to get the mask as a straight sequence

4 years agoUse nmigen's built in formal runner instead of mine
Michael Nolan [Tue, 28 Jan 2020 23:19:19 +0000 (18:19 -0500)]
Use nmigen's built in formal runner instead of mine

4 years agoFix handling of the .sby file directory when running in the current dir
Michael Nolan [Tue, 28 Jan 2020 22:05:25 +0000 (17:05 -0500)]
Fix handling of the .sby file directory when running in the current dir

4 years agoUse anyconst for the inputs of the dut in FMax formal proof
Michael Nolan [Tue, 28 Jan 2020 21:58:58 +0000 (16:58 -0500)]
Use anyconst for the inputs of the dut in FMax formal proof

4 years agoline-up, whitespace
Luke Kenneth Casson Leighton [Tue, 28 Jan 2020 21:32:40 +0000 (21:32 +0000)]
line-up, whitespace

4 years agoread proof file from any location
Luke Kenneth Casson Leighton [Tue, 28 Jan 2020 21:27:58 +0000 (21:27 +0000)]
read proof file from any location

4 years agoremove usage of m.If() in fpmax and replace with Mux
Michael Nolan [Tue, 28 Jan 2020 20:58:12 +0000 (15:58 -0500)]
remove usage of m.If() in fpmax and replace with Mux

4 years agoFlesh out the formal proof for fmax
Michael Nolan [Tue, 28 Jan 2020 17:38:35 +0000 (12:38 -0500)]
Flesh out the formal proof for fmax

4 years agoUpdate gitignore in formal directories to ignore proof folders
Michael Nolan [Tue, 28 Jan 2020 16:56:34 +0000 (11:56 -0500)]
Update gitignore in formal directories to ignore proof folders

4 years agoAdd rudimentary proof to fpmax
Michael Nolan [Tue, 28 Jan 2020 16:43:21 +0000 (11:43 -0500)]
Add rudimentary proof to fpmax

4 years agoAdd fpmin handling to fpmax module
Michael Nolan [Tue, 28 Jan 2020 16:42:58 +0000 (11:42 -0500)]
Add fpmin handling to fpmax module

4 years agoAdd FPMAX module
Michael Nolan [Tue, 28 Jan 2020 16:32:40 +0000 (11:32 -0500)]
Add FPMAX module

4 years agoallow fsgnj proof to be run from any location
Luke Kenneth Casson Leighton [Tue, 28 Jan 2020 19:34:45 +0000 (19:34 +0000)]
allow fsgnj proof to be run from any location

4 years agoFSGNJ: expandd formal proof to 16 and 64 bits
Michael Nolan [Tue, 28 Jan 2020 00:27:12 +0000 (19:27 -0500)]
FSGNJ: expandd formal proof to 16 and 64 bits

4 years agoFSGNJ: Convert proof to use FPNumDecode
Michael Nolan [Tue, 28 Jan 2020 00:23:12 +0000 (19:23 -0500)]
FSGNJ: Convert proof to use FPNumDecode

This uses FPNumDecode in the formal proof of the FSGNJPipeMod to
ensure that the bit indexing method is correct by going about the
proof in a different way

4 years agoAdd formal proof for FSGNJPipeMod module
Michael Nolan [Tue, 28 Jan 2020 00:01:41 +0000 (19:01 -0500)]
Add formal proof for FSGNJPipeMod module

4 years agoremove unneeded comment
Luke Kenneth Casson Leighton [Mon, 27 Jan 2020 19:47:16 +0000 (19:47 +0000)]
remove unneeded comment

4 years agoFSGNJ: Fix typo in abs
Michael Nolan [Mon, 27 Jan 2020 17:33:27 +0000 (12:33 -0500)]
FSGNJ: Fix typo in abs

4 years agoFSGNJ: Remove usage of FPNumDecode
Michael Nolan [Mon, 27 Jan 2020 17:11:24 +0000 (12:11 -0500)]
FSGNJ: Remove usage of FPNumDecode

FPNumDecode was introducing extra logic to subtract the floating point bias
then immediately re-add it when decoding/encoding operand a. This is
unneeded since this module doesn't modify the exponent or mantissa

4 years agoadd comments, remove one of the Muxes
Luke Kenneth Casson Leighton [Mon, 27 Jan 2020 15:54:26 +0000 (15:54 +0000)]
add comments, remove one of the Muxes

4 years agoreduce linelength <80chars
Luke Kenneth Casson Leighton [Mon, 27 Jan 2020 15:46:46 +0000 (15:46 +0000)]
reduce linelength <80chars

4 years agoAdd more/better comments to fsignj.py
Michael Nolan [Mon, 27 Jan 2020 14:51:37 +0000 (09:51 -0500)]
Add more/better comments to fsignj.py

4 years agoFSGNJ: Replace use of Switch() with explicit muxes
Michael Nolan [Mon, 27 Jan 2020 14:34:40 +0000 (09:34 -0500)]
FSGNJ: Replace use of Switch() with explicit muxes

From http://bugs.libre-riscv.org/show_bug.cgi?id=120

"If(), Switch() and friends are fine for modules that are strictly
scalar, but will not work if the module is converted to SIMD."

4 years agoFSGNJ: Use Floating point decode module instead of raw bits
Michael Nolan [Mon, 27 Jan 2020 14:29:52 +0000 (09:29 -0500)]
FSGNJ: Use Floating point decode module instead of raw bits

4 years agoMove patches in README to separate files
Michael Nolan [Mon, 27 Jan 2020 13:29:07 +0000 (08:29 -0500)]
Move patches in README to separate files

This is to facilitate writing better instructions on how to build sfpy
on the wiki

4 years agoadd (untested) partitioned xor/or/and operators
Luke Kenneth Casson Leighton [Mon, 27 Jan 2020 13:26:21 +0000 (13:26 +0000)]
add (untested) partitioned xor/or/and operators
these all work regardless of partitioning

4 years agoadd copyright notices
Luke Kenneth Casson Leighton [Mon, 27 Jan 2020 12:24:05 +0000 (12:24 +0000)]
add copyright notices

4 years agoreverse indices
Luke Kenneth Casson Leighton [Mon, 27 Jan 2020 11:27:13 +0000 (11:27 +0000)]
reverse indices

4 years agoExpand FSGNJ module to 16 and 64 bit floats
Michael Nolan [Mon, 27 Jan 2020 02:11:23 +0000 (21:11 -0500)]
Expand FSGNJ module to 16 and 64 bit floats

This modifies fsignj.py to be compliant with other Float widths
besides 32 bit. It also modifies the unit tests to include tests for
16 and 64 bit floats

4 years agoPep8 cleanup of FSGNJ module
Michael Nolan [Mon, 27 Jan 2020 02:10:31 +0000 (21:10 -0500)]
Pep8 cleanup of FSGNJ module

4 years agoFSGNJ working now in all three modes
Michael Nolan [Sun, 26 Jan 2020 20:59:15 +0000 (15:59 -0500)]
FSGNJ working now in all three modes

This adds the functionality of the FSIGNJ RISCV instruction, namely:

When the opcode is 0x00 - it combines the sign bit of operand B with
the rest of the value in operand A
When the opcode is 0x01 - it combines the inverted sign bit of operand
B with the rest of the value in operand A
When the opcode is 0x02 - it combines the sign bit composed of the
exclusive OR of the sign bits of A and B with the rest of the value in
operand A

This also modifies the unit test for the module to test each of these
behaviors for 32 bit floats