Repository navigation
Preserve device offsets in driver resource descriptors - #56
Draft
RicheyWorks wants to merge 1 commit into
Draft
RicheyWorks wants to merge 1 commit into
RicheyWorks wants to merge 1 commit into
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Driver resource descriptors currently marshal
linear_devPtrandpitch2D_devPtrwithgetNativePointerValue, dropping the offset carried byCUdeviceptr.withByteOffset. Creating a texture from a valid aligned suballocation consequently binds the allocation base instead of the requested region. The reverse conversion also leaves an old Java offset attached when writing a complete returned native address into a reused descriptor.Use the existing offset-aware
getPointerhelper for both device-pointer inputs, andsetPointerfor both returned pointers so stale offsets are cleared. Array/mipmapped-array handles and JNI signatures are unchanged. This is one descriptor-pointer round-trip fix covering linear and pitched memory.Add eight real driver cases: linear/pitched resources, fresh/reused output descriptors, zero/shifted pointers. Query the actual resource address via
cuTexObjectGetResourceDescbefore checking the reused descriptor, so offset cancellation cannot mask a wrong binding. Use device-reported alignment, tiny allocations, private CPU-thread contexts andfinallycleanup. No kernels, sampling operations or device reset.Verification:
jcuda-commoncommit9d41fb5): all eight pass under-Xcheck:jni, with no JNI warnings. JVM logging confirms the rebuilt DLL path. This does not require the separate common copy-back fix.CUDA contract: https://docs.nvidia.com/cuda/archive/12.6.0/cuda-driver-api/group__CUDA__TEXOBJECT.html
Focused test after rebuilding JNI: