ICU-22787 Fix ClangCL compilation on Windows

icu4c/source/common/unicode/platform.h

@@ -235,7 +235,7 @@
 /**
  * \def U_PLATFORM_USES_ONLY_WIN32_API
  * Defines whether the platform uses only the Win32 API.
- * Set to 1 for Windows/MSVC and MinGW but not Cygwin.
+ * Set to 1 for Windows/MSVC, ClangCL and MinGW but not Cygwin.
  * @internal
  */
 #ifdef U_PLATFORM_USES_ONLY_WIN32_API
@@ -250,7 +250,7 @@
 /**
  * \def U_PLATFORM_HAS_WIN32_API
  * Defines whether the Win32 API is available on the platform.
- * Set to 1 for Windows/MSVC, MinGW and Cygwin.
+ * Set to 1 for Windows/MSVC, ClangCL, MinGW and Cygwin.
  * @internal
  */
 #ifdef U_PLATFORM_HAS_WIN32_API

icu4c/source/tools/genccode/genccode.c

@@ -70,6 +70,7 @@ enum {
 #ifdef CAN_GENERATE_OBJECTS
   kOptObject,
   kOptMatchArch,
+  kOptCpuArch,
   kOptSkipDllExport,
 #endif
   kOptFilename,
@@ -86,6 +87,7 @@ static UOption options[]={
 #ifdef CAN_GENERATE_OBJECTS
 /*6*/UOPTION_DEF("object", 'o', UOPT_NO_ARG),
      UOPTION_DEF("match-arch", 'm', UOPT_REQUIRES_ARG),
+     UOPTION_DEF("cpu-arch", 'c', UOPT_REQUIRES_ARG),
      UOPTION_DEF("skip-dll-export", '\0', UOPT_NO_ARG),
 #endif
      UOPTION_DEF("filename", 'f', UOPT_REQUIRES_ARG),
@@ -131,6 +133,8 @@ main(int argc, char* argv[]) {
             "\t-o or --object      write a .obj file instead of .c\n"
             "\t-m or --match-arch file.o  match the architecture (CPU, 32/64 bits) of the specified .o\n"
             "\t                    ELF format defaults to i386. Windows defaults to the native platform.\n"
+            "\t-c or --cpu-arch    Specify a CPU architecture for which to write a .obj file for ClangCL on Windows\n"
+            "\t                    Valid values for this opton are x64, x86 and arm64.\n"
             "\t--skip-dll-export   Don't export the ICU data entry point symbol (for use when statically linking)\n");
 #endif
         fprintf(stderr,
@@ -193,9 +197,17 @@ main(int argc, char* argv[]) {
                 break;
 #ifdef CAN_GENERATE_OBJECTS
             case CALL_WRITEOBJECT:
+                if(options[kOptCpuArch].doesOccur) {
+                    if (!checkCpuArchitecture(options[kOptCpuArch].value)) {
+                        fprintf(stderr,
+                            "CPU architecture \"%s\" is unknown.\n", options[kOptCpuArch].value);
+                        return -1;
+                    }
+                }
                 writeObjectCode(filename, options[kOptDestDir].value,
                                 options[kOptEntryPoint].doesOccur ? options[kOptEntryPoint].value : NULL,
                                 options[kOptMatchArch].doesOccur ? options[kOptMatchArch].value : NULL,
+                                options[kOptCpuArch].doesOccur ? options[kOptCpuArch].value : NULL,
                                 options[kOptFilename].doesOccur ? options[kOptFilename].value : NULL,
                                 NULL,
                                 0,

icu4c/source/tools/pkgdata/pkgdata.cpp

@@ -776,6 +776,7 @@ static int32_t pkg_executeOptions(UPKGOptions *o) {
                         o->entryName,
                         (optMatchArch[0] == 0 ? nullptr : optMatchArch),
                         nullptr,
+                        nullptr,
                         gencFilePath,
                         sizeof(gencFilePath),
                         true);

icu4c/source/tools/toolutil/pkg_genc.cpp

@@ -16,6 +16,9 @@
 #   define NOMCX
 #include <windows.h>
 #include <time.h>
+#   if defined(__clang__)
+#       include <exception>
+#   endif
 #   ifdef __GNUC__
 #       define WINDOWS_WITH_GNUC
 #   endif
@@ -294,6 +297,11 @@ checkAssemblyHeaderName(const char* optAssembly) {
     return false;
 }
 
+U_CAPI UBool U_EXPORT2
+checkCpuArchitecture(const char* optCpuArch) {
+    return strcmp(optCpuArch, "x64") == 0 || strcmp(optCpuArch, "x86") == 0 || strcmp(optCpuArch, "arm64") == 0;
+}
+
 
 U_CAPI void U_EXPORT2
 printAssemblyHeadersToStdErr() {
@@ -799,7 +807,12 @@ getOutFilename(
 
 #ifdef CAN_GENERATE_OBJECTS
 static void
-getArchitecture(uint16_t *pCPU, uint16_t *pBits, UBool *pIsBigEndian, const char *optMatchArch) {
+getArchitecture(
+    uint16_t *pCPU,
+    uint16_t *pBits,
+    UBool *pIsBigEndian,
+    const char *optMatchArch,
+    [[maybe_unused]] const char *optCpuArch) {
     union {
         char        bytes[2048];
 #ifdef U_ELF
@@ -847,7 +860,25 @@ getArchitecture(uint16_t *pCPU, uint16_t *pBits, UBool *pIsBigEndian, const char
 #   if defined(_M_IX86)
         *pCPU = IMAGE_FILE_MACHINE_I386;
 #   else
-        *pCPU = IMAGE_FILE_MACHINE_UNKNOWN;
+        // Linker for ClangCL doesn't handle IMAGE_FILE_MACHINE_UNKNOWN the same as
+        // linker for MSVC. Because of this optCpuArch is used to define the CPU
+        // architecture in that case. While _M_AMD64 and _M_ARM64 could be used,
+        // this would potentially be problematic when cross-compiling as this code
+        // would most likely be ran on host machine to generate the .obj file for
+        // the target architecture.
+#       if defined(__clang__)
+            if (strcmp(optCpuArch, "x64") == 0) {
+                *pCPU = IMAGE_FILE_MACHINE_AMD64;
+            } else if (strcmp(optCpuArch, "x86") == 0) {
+                *pCPU = IMAGE_FILE_MACHINE_I386;
+            } else if (strcmp(optCpuArch, "arm64") == 0) {
+                *pCPU = IMAGE_FILE_MACHINE_ARM64;
+            } else {
+                std::terminate(); // Unreachable.
+            }
+#       else
+            *pCPU = IMAGE_FILE_MACHINE_UNKNOWN;
+#       endif
 #   endif
 #   if defined(_M_IA64) || defined(_M_AMD64) || defined (_M_ARM64)
         *pBits = 64; // Doesn't seem to be used for anything interesting though?
@@ -934,6 +965,7 @@ writeObjectCode(
         const char *destdir,
         const char *optEntryPoint,
         const char *optMatchArch,
+        const char *optCpuArch,
         const char *optFilename,
         char *outFilePath,
         size_t outFilePathCapacity,
@@ -1201,7 +1233,7 @@ writeObjectCode(
 #endif
 
     /* deal with options, files and the entry point name */
-    getArchitecture(&cpu, &bits, &makeBigEndian, optMatchArch);
+    getArchitecture(&cpu, &bits, &makeBigEndian, optMatchArch, optCpuArch);
     if (optMatchArch)
     {
         printf("genccode: --match-arch cpu=%hu bits=%hu big-endian=%d\n", cpu, bits, makeBigEndian);

icu4c/source/tools/toolutil/pkg_genc.h

@@ -73,6 +73,9 @@ printAssemblyHeadersToStdErr(void);
 U_CAPI UBool U_EXPORT2
 checkAssemblyHeaderName(const char* optAssembly);
 
+U_CAPI UBool U_EXPORT2
+checkCpuArchitecture(const char* optCpuArch);
+
 U_CAPI void U_EXPORT2
 writeCCode(
     const char *filename,
@@ -98,6 +101,7 @@ writeObjectCode(
     const char *destdir,
     const char *optEntryPoint,
     const char *optMatchArch,
+    const char *optCpuArch,
     const char *optFilename,
     char *outFilePath,
     size_t outFilePathCapacity,