Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Interpreter] Signed fix #1208

Merged
merged 2 commits into from
Oct 17, 2022
Merged

[Interpreter] Signed fix #1208

merged 2 commits into from
Oct 17, 2022

Conversation

EclecticGriffin
Copy link
Collaborator

A cute little bug that would erroneously cause the python deparse process to present values as negative when they should not be.

@EclecticGriffin EclecticGriffin enabled auto-merge (squash) October 17, 2022 03:27
@EclecticGriffin EclecticGriffin merged commit ff4a03c into master Oct 17, 2022
@EclecticGriffin EclecticGriffin deleted the signed-fix branch October 17, 2022 03:49
janpaulpl added a commit that referenced this pull request Nov 8, 2022
* data gen recognizes seq_mem_d1 too now (#1186)

* Update to Dahlia means we Change some of the .expect files (#1187)

* update to dahlia means we change .expect files

* update tvm generation tests

* modified relay frontend + updated relay tests

* tried to update tvm version (might not work)

* undo workflow change

* tried altering workflow

* probably won't work

* pls work

* altered test cases to fit new tvm version. Also caught a slight bug in conv2d

Co-authored-by: Rachit Nigam <rachit.nigam12@gmail.com>

* I think this might fix the polybench test? (#1188)

* enable integration tests on every PR (#1189)

* enable integration tests on every PR

* disable interpreter integration tests

* Update Live Range Analysis  (#1181)

* updated live range analysis

* added test case

* updated LRA to be more aggressive

* fixed some bugs and added test cases

* Use Docker Image for CI (#1190)

* attempt to use docker image in the CI run

* disable polybench for a bit

* remove working-dir

* y

* y

* y

* y

* y

* y

* sudo

* y

* y

* y

* y

* y

* y

* add PYTHONPATH to ENV in dockerfile

* y

* y

* y

* y

* y

* y

* y

* y

* y

* y

* y

* y

* y

* compress dockerfile

* y

* y

* y

* y

* y

* final config

* Remove integration.yml

* Multi-Dimensional Sequential Memories  (#1195)

* first (incorrect) try at 2d seq-mems

* incorrect seq-mem-d2

* added multi-dim-seq mems

* data-gen handles multi-dim-seqs

* improved function documentation

* [FUD] Interpreter-data utility (#1198)

* add interpreter-data target

* get flake off my case

* AXI generation docs (#1171)

* Initial commit

* Added axi-gen.md to SUMMARY.md

* Implemented suggestions from the repo's PR.

Mainly typos, clarificaitons, and adding links

Co-authored-by: Nathaniel Navarro <nrn25@cornell.edu>
Co-authored-by: Rachit Nigam <rachit.nigam12@gmail.com>

* sync doc (#1178)

* sync doc

* fixed test case

* Update docs/lang/sync.md

Co-authored-by: Rachit Nigam <rachit.nigam12@gmail.com>

* Update docs/lang/sync.md

Co-authored-by: Rachit Nigam <rachit.nigam12@gmail.com>

* addressed comments

Co-authored-by: Rachit Nigam <rachit.nigam12@gmail.com>

* Implement `@data` port optimization in the backend (#1196)

* Implement data port optimization for verilog backend

* use rev from vast repo while waiting for new version to be published

* add docs

* upd tests

* vast package

* Extract period and frequency information from timing reports

* Some Things to Fix with New TVM Version  (#1197)

* start on new tvm version

* added back accidentally removed lines

* relay frontend does seq-mems now

* tvm generates data json correctly

* modified frontend tests

* got rid of unnecesary code

* disabled (not sure if this is right) seq-mem tests

* Add cocotb and iverilog to Dockerfile (#1202)

* Add cocotb and iverilog to Dockerfile

* better practices for cache busting

* Hardcoded iverilog v11 installation

Co-authored-by: Nathaniel Navarro <nrn25@cornell.edu>
Co-authored-by: Rachit Nigam <rachit.nigam12@gmail.com>

* Sync dot product (#1185)

* added the compile-sync pass

* fixed tdcc

* simplified code

* format change

* added comments

* addressed comments and added if tests

* addressed more comments

* deleted unnecessary file

* changed test cases to have alternating roles of producer and consumer.
also added comments to tests

* minor change

* the sync-dot-product test

* sync dot product tests

* Added doc

* added comments

* added 0's in middle of vector representation

* deleted zero's in the middle

* add reference

* concocted the link to sync-dot-product example.

Co-authored-by: Rachit Nigam <rachit.nigam12@gmail.com>

* Add first-class combinational components (#1205)

* Missing Dockerfile changes

* [Interpreter] Add sequential mems and std mem refactor (#1204)

* macro shenanigans

* checkpoint

* Sketch out generic version of the memory for 1d

* membinder impl

* handle serialization

* checkpoint

* clippy

* checkpoint

* new mem defs finished

* switch out the primitive defs

* add seq mems

* make message better

* switch to older clippy

* fix seq mem behavior

* seq mem tests are back

Co-authored-by: Rachit Nigam <rachit.nigam12@gmail.com>

* add CompileInvoke to no-opt pipeline (#1206)

* allow passing tcl file (#1207)

* [Interpreter] Signed fix (#1208)

* fix the parse bug

* fix black invocation

* Reorder logical and expressions. (#1209)

* Fix Papercut For Multiple Go-Done Pairs (#1210)

* Do not group2invoke components with multiple go/done pairs (#1211)

* Update Contributor List (#1212)

* fixes LeNet (#1213)

* Fix Dockerfile not working with runt cocotb tests (#1216)

* Cocotb-bus is now built from github repo

LANG env variable is also set explicitly to be utf-8

* Fix calyx branch and reduce apt-get calls

Co-authored-by: Nathaniel Navarro <nrn25@cornell.edu>

* small changes to bounded sharing (#1214)

* Vivado Maps seq-mems to Urams (#1219)

* vivado maps to urams

* small change

Co-authored-by: Caleb Min Sup Kim <cmk265@havarti.cs.cornell.edu>

* facilitate ref declaration (#1218)

* facilitate ref declaration

* check that a cell is not both external and passed in by reference

* fix types

* Any Base Exponentiation  (#1226)

* added any base

* got rid of muck

* got rid of code garbage

* rewrote tests

* Adding reserved words (#1236)

Discussed in #1233

* [Interpreter] turn off group assignments the same cycle done becomes high (#1237)

* add alias to avoid confusion

* using boxed error instead

* bunch of clippy lints

* bug fixing

Co-authored-by: calebmkim <55243755+calebmkim@users.noreply.github.com>
Co-authored-by: Rachit Nigam <rachit.nigam12@gmail.com>
Co-authored-by: Griffin Berlstein <griffin@berlste.in>
Co-authored-by: Nathaniel Navarro <nathanielnrn99@gmail.com>
Co-authored-by: Nathaniel Navarro <nrn25@cornell.edu>
Co-authored-by: paili0628 <92661158+paili0628@users.noreply.github.com>
Co-authored-by: Andrew Butt <andrewb1999@gmail.com>
Co-authored-by: Chris Gyurgyik <37983775+cgyurgyik@users.noreply.github.com>
Co-authored-by: Caleb Min Sup Kim <cmk265@havarti.cs.cornell.edu>
Co-authored-by: Susan Garry <sgarry406@gmail.com>
janpaulpl added a commit that referenced this pull request Nov 29, 2022
* add CompileInvoke to no-opt pipeline (#1206)

* allow passing tcl file (#1207)

* [Interpreter] Signed fix (#1208)

* fix the parse bug

* fix black invocation

* Reorder logical and expressions. (#1209)

* Fix Papercut For Multiple Go-Done Pairs (#1210)

* Do not group2invoke components with multiple go/done pairs (#1211)

* Update Contributor List (#1212)

* fixes LeNet (#1213)

* Fix Dockerfile not working with runt cocotb tests (#1216)

* Cocotb-bus is now built from github repo

LANG env variable is also set explicitly to be utf-8

* Fix calyx branch and reduce apt-get calls

Co-authored-by: Nathaniel Navarro <nrn25@cornell.edu>

* small changes to bounded sharing (#1214)

* Vivado Maps seq-mems to Urams (#1219)

* vivado maps to urams

* small change

Co-authored-by: Caleb Min Sup Kim <cmk265@havarti.cs.cornell.edu>

* facilitate ref declaration (#1218)

* facilitate ref declaration

* check that a cell is not both external and passed in by reference

* fix types

* Any Base Exponentiation  (#1226)

* added any base

* got rid of muck

* got rid of code garbage

* rewrote tests

* Adding reserved words (#1236)

Discussed in #1233

* [Interpreter] turn off group assignments the same cycle done becomes high (#1237)

* add alias to avoid confusion

* using boxed error instead

* bunch of clippy lints

* bug fixing

* Print out X-values in dat output (#1242)

* Generate X-values in the final DAT file

* format frontend code

* High-level "builder" (kind of an embedded DSL?) in the Python library for easy program generation (#1229)

* PoC Python AST builder library

* Operator overloading for guard expressions

* Absurd contextual assignment syntax

* Bind group names in context statements

* Ridiculous avoidance of Python builtins

* Support Python integer literals

* Support width inference for `done` holes

* Naming & docs

* Slight context access reorg

* Simplify the example a tiny bit

* Flip assignment ordering

For the modern AST library, which has sensibly reversed these things.

* Mild Python cleanup

* New attempt at conditional assignments

* Type cleanliness

* More docs suggested by @cgyurgyik

* Avoid pyramid. (#1243)

* Small LiveRangeAnalysis Change (#1222)

* small lra change

* documentation change

* clarified code + docs

* added clarifying comments

* save_mem flag to allow `dense` operation to save memory  (#1230)

* allow dense to save memory

* addded clarifying comment

* made relay_visitor.py runnable as script

* simplified save_mem flag and added documentation

* added lrn, cleaned up gen_exp (#1231)

* added lrn, cleaned up gen_exp

* updated docs

* small doc changes

* super small change (#1247)

* Check that there are no ref cells before compiling invokes (#1248)

* Disable static timing tests (#1252)

* `group2invoke` should not convert groups with guarded assignments to go/done ports (#1251)

* `group2invoke` should not convert groups with guarded assignments to go/done ports

* comment about wait group

* Comment about the wait restore group

* synchronize before starting new iteration

* remove redundant sync test

* skip sync-while test

* Preserve attributes in group2invoke (#1256)

* `group2invoke` should not convert groups with guarded assignments to go/done ports

* comment about wait group

* Comment about the wait restore group

* synchronize before starting new iteration

* Fix #1253

* fmt

* Fix std_fp_mult_pipe Primitive (#1246)

* Fix std_fp_mult_pipe primitive

* Maybe enable real pipelined multiply

* update exp tests

* systolic and ntt test updates

Co-authored-by: Rachit Nigam <rachit.nigam12@gmail.com>

* separate out compiler tests (#1258)

* Check read and write specs in combinational groups (#1259)

* clippy warnings

* Read write specs checked in comb groups

* add test

* Don't use `then_some` which is not available in older version

* Bump minimatch from 3.0.4 to 3.1.2 in /web (#1263)

Bumps [minimatch](https://github.com/isaacs/minimatch) from 3.0.4 to 3.1.2.
- [Release notes](https://github.com/isaacs/minimatch/releases)
- [Commits](isaacs/minimatch@v3.0.4...v3.1.2)

---
updated-dependencies:
- dependency-name: minimatch
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <support@github.com>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* Fix #1239 (#1267)

* Add ability to parse empty control with attributes (#1266)

* Add autoupdate workflow (#1257)

* Add autoupdate workflow

* disable requirement on eval

* autoupdate separate out workflow

* Revert "Add autoupdate workflow (#1257)"

This reverts commit e3c1412.

* limit concurrency in actions (#1271)

* Reduce struct sizes (#1269)

* Reduce size of Error

* Box span in ir::Id

* Make Span unclonable and behind Rc

* Add ability to parse empty control with attributes (#1266)

* Add autoupdate workflow (#1257)

* Add autoupdate workflow

* disable requirement on eval

* autoupdate separate out workflow

* Fix #1244 (#1268)

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* Fix #1277 (#1279)

* Add timing specified timing information to latency_data

* Fix #1277

* Generate program without buffering (#1280)

* vscode nonsense

* print to output stream directly instead of buffering

* Generate body without buffering

* flip

* update tests

* Upgrade dependencies

* Revert "Upgrade dependencies"

This reverts commit b807bf5.

* Upgrade dependencies (#1281)

* Upgrade dependencies

* rust upd

* Log parsing time

* CDF Generator  (#1275)

* started on cdf-gen

* more progress

* cdf generator

* formatting

* updated cdf formulation to be more accurate

* cleaned code changed way we count sharing

* clippy

* Make builder work with python 3.9

* add test for builder

* Update Verilator v5.002 (#1289)

* vscode nonsense

* Fix #1260

* update tests

* Use builder for systolic frontend (#1288)

* vscode nonsense

* Port systolic frontend to use builder

* update test

* remove pe def

* cleanup

* upd

* [Interpreter] Misc reorg (#1290)

* commit regression test from earlier

* Do some light reorganization

* fix the tests

* use usize instead of u64

* Ref problem (#1255)

* fixed the issue with ref

* added comments

* made changes to insert port in the start function

* formatting issue

* modified compile-ref and test cases

* format

* change compile-sync to original version

* formatting

* canonicalize

* fix borrow error

* fixed small error

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Rachit Nigam <rachit.nigam12@gmail.com>

* Update documentation (#1291)

* Fill TK in lang-ref

* inlining trick

* document data format

* minor fixes

* Modify sync (#1285)

* change compile-sync

* modified compile-sync so it adds barriers to single group enables and cell invokes

* changed compile-sync and modified passes

* modify tests

* changed example test cases

* added error message, changed name for `wait`, changed markdown

* format

* fix format

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: Rachit Nigam <rachit.nigam12@gmail.com>
Co-authored-by: Griffin Berlstein <griffin@berlste.in>
Co-authored-by: Chris Gyurgyik <37983775+cgyurgyik@users.noreply.github.com>
Co-authored-by: Andrew Butt <andrewb1999@gmail.com>
Co-authored-by: calebmkim <55243755+calebmkim@users.noreply.github.com>
Co-authored-by: Nathaniel Navarro <nathanielnrn99@gmail.com>
Co-authored-by: Nathaniel Navarro <nrn25@cornell.edu>
Co-authored-by: Caleb Min Sup Kim <cmk265@havarti.cs.cornell.edu>
Co-authored-by: Susan Garry <sgarry406@gmail.com>
Co-authored-by: Adrian Sampson <adrian@radbox.org>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: paili0628 <92661158+paili0628@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant