[clang] [llvm] [TargetParser] Add a list of Intel GPUs, and use it in offload-arch (PR #222072)

[Nikita Kornev via cfe-commits]

================
@@ -173,11 +194,28 @@ int printGPUsByLevelZero() {
     CALL_ZE_AND_CHECK(zeDeviceGet, Driver, &DeviceCount, Devices.data());
 
     for (auto Device : Devices) {
+      ze_device_ip_version_ext_t IPVersion = {};
+      IPVersion.stype = ZE_STRUCTURE_TYPE_DEVICE_IP_VERSION_EXT;
+      IPVersion.pNext = nullptr;
+
       ze_device_properties_t DeviceProperties = {};
       DeviceProperties.stype = ZE_STRUCTURE_TYPE_DEVICE_PROPERTIES;
-      DeviceProperties.pNext = nullptr;
+      DeviceProperties.pNext = &IPVersion;
       CALL_ZE_AND_CHECK(zeDeviceGetProperties, Device, &DeviceProperties);
-      llvm::outs() << DeviceProperties.name << '\n';
+
+      // A driver that does not support the extension leaves the chained
+      // structure untouched, in which case there is no architecture to name.
+      if (IPVersion.ipVersion == 0) {
+        if (Verbose)
+          llvm::errs() << "Unable to query the IP version of device '"
+                       << DeviceProperties.name << "'\n";
+        continue;
+      }
----------------
KornevNikita wrote:

It's a good idea, but it seems like clang silences the offload-arch tool warnings, so users won't get this warning unless they run offload-arch manually.
Probably to address that we need to update how `Driver::executeProgram` handles stderr, and I'm unsure that this way would be accepted by community.

https://github.com/llvm/llvm-project/pull/222072