fix(build,hip): round-3 batch — HIP vif uninit-err + NVTX libdl + icx ssim fp-model (R3-6/9/10)

changelog.d/fixed/round3-build-gpu-batch.md

@@ -0,0 +1,19 @@
+### Fixed
+
+Round-3 bug-hunt — build/GPU error-path batch (golden-safe).
+
+- **R3-6** HIP `integer_vif` init returned an **uninitialized** error code
+  (`core/src/feature/hip/integer_vif_hip.c`): the `fail_stream`/`fail_submit`
+  goto paths reached `return err;` before `int err` was declared (it was
+  declared at the `vif_hip_module_load` call). Declare `int err = 0;` at the
+  top of `init_fex_hip` and drop the later redeclaration.
+- **R3-9** NVTX dependency linked an **empty** `shared_library('dl')` target
+  instead of the system libdl (`core/src/meson.build`): replaced with
+  `cc.find_library('dl', required: false)`, mirroring how `ze_loader`/`m`/etc.
+  are resolved.
+- **R3-10** icx strict-FP flag missing on the SSIM AVX2 carve-out
+  (`core/src/meson.build` `x86_ssim_avx2_lib`): appended
+  `_x86_simd_strict_fp_extra` (`-fp-model=precise` under intel-llvm) so SSIM
+  AVX2 matches the scalar reference under icx, matching its sibling carve-outs.
+  No-op on gcc/clang (`_x86_simd_strict_fp_extra` is empty) → golden gate
+  unchanged.

core/src/feature/hip/integer_vif_hip.c

@@ -380,6 +380,7 @@ static int init_fex_hip(VmafFeatureExtractor *fex, enum VmafPixelFormat pix_fmt,
     return -ENOSYS;
 #else
     VifStateHip *s = fex->priv;
+    int err = 0; /* R3-6: init early so fail_stream/fail_submit return a defined code */
 
     hipError_t rc = hipStreamCreate(&s->str);
     if (rc != hipSuccess)
@@ -391,7 +392,7 @@ static int init_fex_hip(VmafFeatureExtractor *fex, enum VmafPixelFormat pix_fmt,
     if (rc != hipSuccess)
         goto fail_submit;
 
-    int err = vif_hip_module_load(s);
+    err = vif_hip_module_load(s);
     if (err != 0)
         goto fail_finished;
 

core/src/meson.build

@@ -877,7 +877,7 @@ if is_asm_enabled
           'x86_ssim_avx2',
           feature_src_dir + 'x86/ssim_avx2.c',
           include_directories : vmaf_base_include,
-          c_args : ['-mavx', '-mavx2', '-mfma', '-ffp-contract=off'] + vmaf_cflags_common,
+          c_args : ['-mavx', '-mavx2', '-mfma', '-ffp-contract=off'] + _x86_simd_strict_fp_extra + vmaf_cflags_common,
       )
 
 
@@ -1011,7 +1011,7 @@ cuda_inc = []
 
 if is_nvtx_enabled
     cuda_dependency += declare_dependency(
-        link_with: shared_library('dl'),
+        dependencies: cc.find_library('dl', required: false),
         include_directories: include_directories('/usr/local/cuda/include'))
 endif
 

docs/rebase-notes.md

@@ -46639,3 +46639,6 @@ no ffmpeg-patch impact: edits the CLI (`core/tools/vmaf.cpp`, `cli_parse.cpp`) o
 
 ## chore/version-3.2.0 (2026-06-27)
 no rebase impact: bumps `core/meson.build` `version` (x-release-please-version) + `.release-please-manifest.json` `.` to `3.2.0` / `3.2.0-lusoris.0` to track upstream libvmaf 3.2.0 SONAME. On an upstream sync, keep the fork's libvmaf version aligned with Netflix's (`<upstream-X.Y.Z>-lusoris.N`).
+
+## fix/round3-build-gpu-batch (2026-06-27)
+no ffmpeg-patch impact. R3-6 HIP integer_vif uninit-err (init err=0). R3-9 NVTX libdl → cc.find_library('dl'). R3-10 ssim AVX2 carve-out + `_x86_simd_strict_fp_extra` (icx -fp-model=precise; no-op on gcc/clang). **Invariant:** every x86 SIMD carve-out lib that needs bit-exactness under icx must carry `_x86_simd_strict_fp_extra`; keep the ssim carve-out aligned with its psnr_hvs/ms_ssim/ssimulacra2 siblings.

docs/state.md

@@ -325,6 +325,8 @@ landed fix yet._
 
 ## Recently closed
 
+| **T-BUGHUNT-R3-BUILD-GPU-2026-06-27** | Round-3 build/GPU error-path batch. R3-6 HIP integer_vif init returned uninitialized err on fail_stream/fail_submit (declare int err=0 early). R3-9 NVTX linked empty shared_library('dl') → cc.find_library('dl'). R3-10 ssim AVX2 carve-out missing _x86_simd_strict_fp_extra (icx fp divergence; no-op gcc/clang). Deferred: R2-6 icpx-motion needs a scoped motion carve-out (golden-sensitive). | no ADR (bug-fix batch) | fix/round3-build-gpu-batch | meson reports libvmaf 3.2.0; CPU build 1280/1280; R3-6 HIP-only, R3-10 icx-only → golden gate unchanged. | (2026-06-27) |
+
 | **T-VERSION-TRACK-UPSTREAM-3.2.0-2026-06-27** | Fork libvmaf version drifted behind upstream (fork 3.0.0 vs Netflix 3.2.0 SONAME, upstream 3f9e02af25). Bumped `core/meson.build` version + `.release-please-manifest.json` base to 3.2.0 / 3.2.0-lusoris.0 per CLAUDE §11 (track upstream version + lusoris suffix). Source-only; no release performed. | no ADR (executes existing §11 versioning policy) | chore/version-3.2.0 | `meson setup` reports `libvmaf 3.2.0`; no build/golden impact (version string only). | (2026-06-27) |
 
 | **T-BUGHUNT-CLI-2026-06-27** | CLI bug-hunt (core/tools/, golden-safe). `--help` now prints to stdout + exits 0 (was stderr/exit 1). Output-write return is checked (diagnostic + nonzero exit on failure). Dead `vmaf.c` deleted (genuinely unreferenced; `cli_parse.c` KEPT — it's compiled into cli_parse/fuzz tests). Wall-time FPS (was clock()/CLOCKS_PER_SEC, over-counting ~n_threads). No-frames guard (picture_index==0 → exit 101, was UINT_MAX underflow). | no ADR (bug-fix batch) | fix/bughunt-cli | CPU build 1280/1280, fast suite 106/106; smoke: --help exit 0/stdout, bad-output exit 254, no-frames exit 101, golden 576x324 mean 94.32301. | (2026-06-27) |