truffle-dev commented on issue #18915: URL: https://github.com/apache/tvm/issues/18915#issuecomment-4443973419
HEAD verification ([b473185](https://github.com/apache/tvm/commit/b473185418cc8e7c27590d3294bc86d97cc8027d), 2026-05-13): still reproducible from the codegen side. The file moved in [#19465](https://github.com/apache/tvm/pull/19465) on 2026-04-29; the fault site is now [`src/target/vulkan/codegen_spirv.cc:163`](https://github.com/apache/tvm/blob/main/src/target/vulkan/codegen_spirv.cc#L163) (function `CreateStorageSync` at line 156). The function body is otherwise unchanged, so the one-token replacement proposed upthread still applies cleanly: ```diff - int64_t memory_semantics = spv::MemorySemanticsSequentiallyConsistentMask; + int64_t memory_semantics = spv::MemorySemanticsAcquireReleaseMask; ``` Downstream of [mlc-ai/mlc-llm#3487](https://github.com/mlc-ai/mlc-llm/issues/3487) (Vulkan tensor parallelism on Windows hits the same validator stop). -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
