wouter/d4r
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add accuracy mode and GLIBC 2.41 release builds

Browse Source 
Add default-off PreferAccuracy with conservative native variants for every RDNA3/RDNA4 target and a separate JIT cache. Preserve fast defaults and reject mixed accuracy sets.

Add a pinned Debian build environment, rebuild all release components against GLIBC 2.41, and include toolchain provenance and release notes for the accuracy changes and the preceding artifact/synchronization fixes. Refs #6; CPU-overhead confirmation for #7 remains pending.
This commit is contained in:
ayo committed 2026-10-01 01:18:01 -04:00
1 parent b2b3781aca
commit f71a483a00
25 files changed
+641 -105
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

kernels/tex/tex_common.h
+1 -1
View File
@@ -14,7 +14,7 @@
// Kernels that ZLUDA compiles in strict-FP mode (constrained intrinsics) cannot take inlined FP code from
// here (the AMDGPU backend fails on some constrained operations): build them with -DD4R_SUST_OUTLINE,
// which keeps the FP bodies of the surface stores in noinline leaf functions (no calls, no stack).
#ifdef D4R_SUST_OUTLINE
#if defined(D4R_SUST_OUTLINE) || defined(D4R_ACCURACY)
#define FP_BODY __attribute__((device, noinline)) static
#else
#define FP_BODY __attribute__((device, always_inline)) static inline