Merge branch 'blender-v4.1-release'

build_files/build_environment/patches/oidn.diff

@@ -52,3 +52,24 @@ diff -Naur oidn-2.2.0/devices/CMakeLists.txt external_openimagedenoise/devices/C
      BUILD_ALWAYS TRUE
      DEPENDS
        OpenImageDenoise_core
+diff --git a/devices/hip/hip_device.cpp b/devices/hip/hip_device.cpp
+index ae14ced..a49e131 100644
+--- a/devices/hip/hip_device.cpp
++++ b/devices/hip/hip_device.cpp
+@@ -93,10 +93,16 @@ OIDN_NAMESPACE_BEGIN
+   {
+     const std::string name = getArchName(prop);
+ 
++    // BLENDER: this comment is meant to generate a merge conflict if the code
++    // here changes, so we know that hipSupportsDeviceOIDN should be updated.
+     if (name == "gfx1030")
+       return HIPArch::DL;
++    // BLENDER: this comment is meant to generate a merge conflict if the code
++    // here changes, so we know that hipSupportsDeviceOIDN should be updated.
+     if (name == "gfx1100" || name == "gfx1101" || name == "gfx1102")
+       return HIPArch::WMMA;
++    // BLENDER: this comment is meant to generate a merge conflict if the code
++    // here changes, so we know that hipSupportsDeviceOIDN should be updated.
+     else
+       return HIPArch::Unknown;
+   }

intern/cycles/device/hip/device.cpp

@@ -184,7 +184,9 @@ void device_hip_info(vector<DeviceInfo> &devices)
 
     info.denoisers = 0;
 #  if defined(WITH_OPENIMAGEDENOISE)
-    if (OIDNDenoiserGPU::is_device_supported(info)) {
+    /* Check first if OIDN supports it, not doing so can crash the HIP driver with
+     * "hipErrorNoBinaryForGpu: Unable to find code object for all current devices". */
+    if (hipSupportsDeviceOIDN(num) && OIDNDenoiserGPU::is_device_supported(info)) {
       info.denoisers |= DENOISER_OPENIMAGEDENOISE;
     }
 #  endif

intern/cycles/device/hip/device_impl.cpp

@@ -227,19 +227,11 @@ string HIPDevice::compile_kernel(const uint kernel_features, const char *name, c
   int major, minor;
   hipDeviceGetAttribute(&major, hipDeviceAttributeComputeCapabilityMajor, hipDevId);
   hipDeviceGetAttribute(&minor, hipDeviceAttributeComputeCapabilityMinor, hipDevId);
-  hipDeviceProp_t props;
-  hipGetDeviceProperties(&props, hipDevId);
-
-  /* gcnArchName can contain tokens after the arch name with features, ie.
-   * `gfx1010:sramecc-:xnack-` so we tokenize it to get the first part. */
-  char *arch = strtok(props.gcnArchName, ":");
-  if (arch == NULL) {
-    arch = props.gcnArchName;
-  }
+  const std::string arch = hipDeviceArch(hipDevId);
 
   /* Attempt to use kernel provided with Blender. */
   if (!use_adaptive_compilation()) {
-    const string fatbin = path_get(string_printf("lib/%s_%s.fatbin", name, arch));
+    const string fatbin = path_get(string_printf("lib/%s_%s.fatbin", name, arch.c_str()));
     VLOG_INFO << "Testing for pre-compiled kernel " << fatbin << ".";
     if (path_exists(fatbin)) {
       VLOG_INFO << "Using precompiled kernel.";
@@ -267,10 +259,11 @@ string HIPDevice::compile_kernel(const uint kernel_features, const char *name, c
 #  ifndef NDEBUG
   options.append(" -save-temps");
 #  endif
-  options.append(" --amdgpu-target=").append(arch);
+  options.append(" --amdgpu-target=").append(arch.c_str());
 
   const string include_path = source_path;
-  const string fatbin_file = string_printf("cycles_%s_%s_%s", name, arch, kernel_md5.c_str());
+  const string fatbin_file = string_printf(
+      "cycles_%s_%s_%s", name, arch.c_str(), kernel_md5.c_str());
   const string fatbin = path_cache_get(path_join("kernels", fatbin_file));
   VLOG_INFO << "Testing for locally compiled kernel " << fatbin << ".";
   if (path_exists(fatbin)) {

intern/cycles/device/hip/util.h

@@ -4,6 +4,9 @@
 
 #pragma once
 
+#include <cstring>
+#include <string>
+
 #ifdef WITH_HIP
 
 #  ifdef WITH_HIP_DYNLOAD
@@ -46,6 +49,14 @@ const char *hipewCompilerPath();
 int hipewCompilerVersion();
 #  endif /* WITH_HIP_DYNLOAD */
 
+static std::string hipDeviceArch(const int hipDevId)
+{
+  hipDeviceProp_t props;
+  hipGetDeviceProperties(&props, hipDevId);
+  const char *arch = strtok(props.gcnArchName, ":");
+  return (arch == nullptr) ? props.gcnArchName : arch;
+}
+
 static inline bool hipSupportsDevice(const int hipDevId)
 {
   int major, minor;
@@ -55,6 +66,13 @@ static inline bool hipSupportsDevice(const int hipDevId)
   return (major >= 9);
 }
 
+static inline bool hipSupportsDeviceOIDN(const int hipDevId)
+{
+  /* Matches HIPDevice::getArch in HIP. */
+  const std::string arch = hipDeviceArch(hipDevId);
+  return (arch == "gfx1030" || arch == "gfx1100" || arch == "gfx1101" || arch == "gfx1102");
+}
+
 CCL_NAMESPACE_END
 
 #endif /* WITH_HIP */

intern/cycles/device/hiprt/device_impl.cpp

@@ -138,13 +138,7 @@ string HIPRTDevice::compile_kernel(const uint kernel_features, const char *name,
   int major, minor;
   hipDeviceGetAttribute(&major, hipDeviceAttributeComputeCapabilityMajor, hipDevId);
   hipDeviceGetAttribute(&minor, hipDeviceAttributeComputeCapabilityMinor, hipDevId);
-  hipDeviceProp_t props;
-  hipGetDeviceProperties(&props, hipDevId);
-
-  char *arch = strtok(props.gcnArchName, ":");
-  if (arch == NULL) {
-    arch = props.gcnArchName;
-  }
+  const std::string arch = hipDeviceArch(hipDevId);
 
   if (!use_adaptive_compilation()) {
     const string fatbin = path_get(string_printf("lib/%s_rt_gfx.hipfb", name));
@@ -162,10 +156,11 @@ string HIPRTDevice::compile_kernel(const uint kernel_features, const char *name,
   const string kernel_md5 = util_md5_string(source_md5 + common_cflags);
 
   const string include_path = source_path;
-  const string bitcode_file = string_printf("cycles_%s_%s_%s.bc", name, arch, kernel_md5.c_str());
+  const string bitcode_file = string_printf(
+      "cycles_%s_%s_%s.bc", name, arch.c_str(), kernel_md5.c_str());
   const string bitcode = path_cache_get(path_join("kernels", bitcode_file));
   const string fatbin_file = string_printf(
-      "cycles_%s_%s_%s.hipfb", name, arch, kernel_md5.c_str());
+      "cycles_%s_%s_%s.hipfb", name, arch.c_str(), kernel_md5.c_str());
   const string fatbin = path_cache_get(path_join("kernels", fatbin_file));
 
   VLOG(1) << "Testing for locally compiled kernel " << fatbin << ".";
@@ -229,7 +224,7 @@ string HIPRTDevice::compile_kernel(const uint kernel_features, const char *name,
 
     std::string rtc_options;
 
-    rtc_options.append(" --offload-arch=").append(arch);
+    rtc_options.append(" --offload-arch=").append(arch.c_str());
     rtc_options.append(" -D __HIPRT__");
     rtc_options.append(" -ffast-math -O3 -std=c++17");
     rtc_options.append(" -fgpu-rdc -c --gpu-bundle-output -c -emit-llvm");
@@ -260,7 +255,7 @@ string HIPRTDevice::compile_kernel(const uint kernel_features, const char *name,
   // After compilation, the bitcode produced is linked with HIP RT bitcode (containing
   // implementations of HIP RT functions, e.g. traversal, to produce the final executable code
   string linker_options;
-  linker_options.append(" --offload-arch=").append(arch);
+  linker_options.append(" --offload-arch=").append(arch.c_str());
   linker_options.append(" -fgpu-rdc --hip-link --cuda-device-only ");
   string hiprt_ver(HIPRT_VERSION_STR);
   string hiprt_bc = hiprt_path + "\\dist\\bin\\Release\\hiprt" + hiprt_ver + "_amd_lib_win.bc";