Skip to content
Draft
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 0 additions & 2 deletions .github/workflows/docs.yml
Original file line number Diff line number Diff line change
Expand Up @@ -2,14 +2,12 @@ name: Build and Deploy Documentation

on:
push:
branches: [ main ]
paths:
- 'docs/**'
- 'iris/**'
- 'examples/**'
- '.github/workflows/docs.yml'
pull_request:
branches: [ main ]
paths:
- 'docs/**'
- 'iris/**'
Expand Down
2 changes: 0 additions & 2 deletions .github/workflows/iris-external-validation-test.yml
Original file line number Diff line number Diff line change
Expand Up @@ -2,9 +2,7 @@ name: Iris External Validation Test

on:
push:
branches: [ main ]
pull_request:
branches: [ main ]
workflow_dispatch:

concurrency:
Expand Down
2 changes: 0 additions & 2 deletions .github/workflows/iris-performance-regression-test.yml
Original file line number Diff line number Diff line change
Expand Up @@ -2,9 +2,7 @@ name: Iris Performance Regression Test

on:
push:
branches: [ main ]
pull_request:
branches: [ main ]
workflow_dispatch:

concurrency:
Expand Down
2 changes: 0 additions & 2 deletions .github/workflows/iris-tests.yml
Original file line number Diff line number Diff line change
Expand Up @@ -2,9 +2,7 @@ name: Iris Tests

on:
push:
branches: [ main ]
pull_request:
branches: [ main ]
workflow_dispatch:

concurrency:
Expand Down
21 changes: 5 additions & 16 deletions iris/iris.py
Original file line number Diff line number Diff line change
Expand Up @@ -1858,22 +1858,11 @@ def copy(
>>> iris.copy(remote_ptr, local_ptr, from_rank, to_rank, to_rank, heap_bases)
"""

cur_base = tl.load(heap_bases + cur_rank)

from_base = tl.load(heap_bases + from_rank)
to_base = tl.load(heap_bases + to_rank)

src_ptr_int = tl.cast(src_ptr, tl.uint64)
src_offset = src_ptr_int - cur_base

dst_ptr_int = tl.cast(dst_ptr, tl.uint64)
dst_offset = dst_ptr_int - cur_base

from_base_byte = tl.cast(from_base, tl.pointer_type(tl.int8))
to_base_byte = tl.cast(to_base, tl.pointer_type(tl.int8))

translated_src = tl.cast(from_base_byte + src_offset, src_ptr.dtype)
translated_dst = tl.cast(to_base_byte + dst_offset, src_ptr.dtype)
# Translate src_ptr from cur_rank's address space to from_rank's address space
translated_src = __translate(src_ptr, cur_rank, from_rank, heap_bases)

# Translate dst_ptr from cur_rank's address space to to_rank's address space
translated_dst = __translate(dst_ptr, cur_rank, to_rank, heap_bases)

data = tl.load(translated_src, mask=mask, cache_modifier=load_cache_modifier)
tl.store(translated_dst, data, mask=mask, cache_modifier=store_cache_modifier)
Expand Down