Skip to content
This repository was archived by the owner on Mar 25, 2025. It is now read-only.

Fixed for watch statements codegen, global and NetSendBuffer_t variables update on CPU/GPU#710

Merged
pramodk merged 7 commits into
masterfrom
pramodk/watch-codegen-fixes
Jul 23, 2021
Merged

Fixed for watch statements codegen, global and NetSendBuffer_t variables update on CPU/GPU#710
pramodk merged 7 commits into
masterfrom
pramodk/watch-codegen-fixes

Conversation

@pramodk

@pramodk pramodk commented Jul 21, 2021 •
edited
Loading

Copy link
Copy Markdown
Contributor

Integration test from testcorenrn/hhwatch will be enabled in NEURON.

todos

Closes #679, closes #678, closes #675, closes #680

pramodk added 3 commits July 21, 2021 18:53
@pramodk
Fixed for watch statements, global variable on GPU
b7784ba 
  * fix watch statement code generation described in #679
  * fix global variable update to gpu descrbed in #678
  * atomic capture for net_send buffer

WIP fixes #679 #678 #675 #680
@pramodk
Fixes for watch test on GPU
1321b41 
 * Initial block was missing net_send_buffering related codegen
   when net_send was used
 * fix indentation with watch if-else statements
 * net_send_buffer updates from host & device missing
 * net_receieve_buffer code was out of place
@pramodk
Move code from C visitor to ACC visitor
7527197
pramodk added a commit to BlueBrain/CoreNeuron that referenced this pull request Jul 21, 2021
@pramodk
Update nmodl submodule: fixes for WATCH statements on GPU
aa238d8 
See BlueBrain/nmodl/pull/710
pramodk added a commit to neuronsimulator/nrn that referenced this pull request Jul 21, 2021
@pramodk
Update coreneuron submodule: fixes for WATCH statements on GPU with N…
f3f07f0 
…MODL

* update coreneuron submodule
* coreneuron updated nmodl submdule
* enable hhwatch test on GPU with NMODL

- [ ] Merge CoreNEURON PR
- [ ] Update CoreNEURON submodule

See BlueBrain/nmodl/pull/710
This was referenced Jul 21, 2021
Merged
Merged
@bbpbuildbot

bbpbuildbot commented Jul 21, 2021

Copy link
Copy Markdown
Collaborator

Logfiles from GitLab pipeline #11463 (:no_entry:) have been uploaded here!

Status and direct links:

@pramodk pramodk requested review from iomaganaris and olupton July 21, 2021 21:18
@pramodk pramodk marked this pull request as draft July 21, 2021 21:18
@bbpbuildbot

bbpbuildbot commented Jul 21, 2021

Copy link
Copy Markdown
Collaborator

Logfiles from GitLab pipeline #11474 (:no_entry:) have been uploaded here!

Status and direct links:

olupton
olupton reviewed Jul 23, 2021
View reviewed changes

@olupton olupton left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM so far

Comment thread src/codegen/codegen_acc_visitor.hpp Outdated
pramodk added 2 commits July 23, 2021 11:57
@pramodk
Incorporate BlueBrain/mod2c/pull/63
3f5e0a9
@pramodk
Fix CVF CI: INITIAL block shouldn't be executed on ISPC backend
2302bd2 
when mod file uses net_send/net_event method. This is because they
contain non-compatible code for event buffering used for GPU.
@pramodk pramodk marked this pull request as ready for review July 23, 2021 10:06
@bbpbuildbot

bbpbuildbot commented Jul 23, 2021

Copy link
Copy Markdown
Collaborator

Logfiles from GitLab pipeline #11634 (:white_check_mark:) have been uploaded here!

Status and direct links:

olupton
olupton approved these changes Jul 23, 2021
View reviewed changes

@olupton olupton left a comment •
edited
Loading

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM, except I think #680 is not addressed so we should change the description so merging this doesn't automatically close it.

We should wait for https://bbpgitlab.epfl.ch/hpc/coreneuron/-/pipelines/11640 to finish before merging this.

@pramodk

pramodk commented Jul 23, 2021

Copy link
Copy Markdown
Contributor Author

LGTM, except I think #680 is not addressed so we should change the description so merging this doesn't automatically close it.

Yeah, realised that after posting my previous comment. I am currently looking into that. If it's easy, I will fix it here. Otherwise, do that as a separate PR.

@pramodk
Fix the weight_index argument for net_send_buffering()
280887c 
generated for net_send() call in INITIAL block.

See analysis in #680

fixes #680
@pramodk

pramodk commented Jul 23, 2021

Copy link
Copy Markdown
Contributor Author

Fixed #680 based on analysis in #680 (comment)

@bbpbuildbot

bbpbuildbot commented Jul 23, 2021

Copy link
Copy Markdown
Collaborator

Logfiles from GitLab pipeline #11652 (:white_check_mark:) have been uploaded here!

Status and direct links:

@bbpbuildbot

bbpbuildbot commented Jul 23, 2021

Copy link
Copy Markdown
Collaborator

Logfiles from GitLab pipeline #11651 (:white_check_mark:) have been uploaded here!

Status and direct links:

iomaganaris
iomaganaris approved these changes Jul 23, 2021
View reviewed changes

@iomaganaris iomaganaris left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@pramodk pramodk merged commit 6678d8e into master Jul 23, 2021
@pramodk pramodk deleted the pramodk/watch-codegen-fixes branch July 23, 2021 16:31
pramodk added a commit to BlueBrain/CoreNeuron that referenced this pull request Jul 23, 2021
@pramodk
Update nmodl submodule: fixes for WATCH statements on GPU (#598)
96a22c9 
See BlueBrain/nmodl/pull/710
pramodk added a commit to neuronsimulator/nrn that referenced this pull request Nov 2, 2022
@pramodk
Update nmodl submodule: fixes for WATCH statements on GPU (BlueBrain/…
5006918 
…CoreNeuron#598)

See BlueBrain/nmodl/pull/710

CoreNEURON Repo SHA: BlueBrain/CoreNeuron@96a22c9
JCGoran pushed a commit to neuronsimulator/nrn that referenced this pull request Mar 12, 2025
@pramodk
Fixed for watch statements codegen, global and NetSendBuffer_t variab…
5fdcf09 
…les update on CPU/GPU (BlueBrain/nmodl#710)

* fixed for watch statements, global variable on GPU
* fixed watch statement code generation described in BlueBrain/nmodl#679
* fixed global variable update to gpu descrbed in BlueBrain/nmodl#678
* fixed atomic capture for net_send buffer
* fixed for watch test on GPU
     - Initial block was missing net_send_buffering related codegen
       when net_send was used
     - fix indentation with watch if-else statements
     - net_send_buffer updates from host & device missing
     - net_receieve_buffer code was out of place
* Incorporate BlueBrain/mod2c/pull/63
* Fix CVF CI: INITIAL block shouldn't be executed on ISPC backend
   when mod file uses net_send/net_event method. This is because they
   contain non-compatible code for event buffering used for GPU.
* Fix the weight_index argument for net_send_buffering()
   generated for net_send() call in INITIAL block.
   - See analysis in BlueBrain/nmodl#680

Closes BlueBrain/nmodl#679, closes BlueBrain/nmodl#678, closes BlueBrain/nmodl#675, closes BlueBrain/nmodl#680

NMODL Repo SHA: BlueBrain/nmodl@6678d8e
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Reviewers

2 more reviewers
@olupton olupton olupton approved these changes
@iomaganaris iomaganaris iomaganaris approved these changes
Reviewers whose approvals may not affect merge requirements

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

4 participants

@pramodk @bbpbuildbot @olupton @iomaganaris