arch/intel64: enable FPU and implement up_fpucmp to pass ostest

enable FPU and implement up_fpucmp to pass ostest

With aggressive optimization enabled (-O2/-O3), ostest FPU test will fail.
This is because the compiler will generate additional vector
instructions between subsequent up_fpucmp() calls (loop vectorization
somewhere in usleep() call), which will consequently overwrite
the expected FPU context (XMM registers).

Signed-off-by: p-szafonimateusz <p-szafonimateusz@xiaomi.com>

arch/Kconfig

@@ -130,6 +130,8 @@ config ARCH_X86
 config ARCH_X86_64
 	bool "x86_64"
 	select ARCH_HAVE_TCBINFO
+	select ARCH_HAVE_FPU
+	select ARCH_HAVE_DPFPU
 	select LIBC_ARCH_ELF_64BIT if LIBC_ARCH_ELF
 	---help---
 		x86-64 architectures.

arch/x86_64/include/intel64/irq.h

@@ -104,7 +104,7 @@
  */
 
 #define XCPTCONTEXT_XMM_AREA_SIZE 512
-#define XMMAREA_OFFSET  XCPTCONTEXT_XMM_AREA_SIZE / 8
+#define XMMAREA_OFFSET            (XCPTCONTEXT_XMM_AREA_SIZE / 8)
 
 /* Data segments */
 

arch/x86_64/src/intel64/CMakeLists.txt

@@ -55,4 +55,8 @@ else()
   list(APPEND SRCS intel64_timerisr.c)
 endif()
 
+if(CONFIG_ARCH_FPU)
+  list(APPEND SRCS intel64_fpucmp.c)
+endif()
+
 target_sources(arch PRIVATE ${SRCS})

arch/x86_64/src/intel64/Make.defs

@@ -52,3 +52,6 @@ ifeq ($(CONFIG_SCHED_TICKLESS),y)
 CHIP_CSRCS += intel64_tickless.c
 endif
 
+ifeq ($(CONFIG_ARCH_FPU),y)
+CHIP_CSRCS += intel64_fpucmp.c
+endif

arch/x86_64/src/intel64/intel64_fpucmp.c

@@ -0,0 +1,76 @@
+/****************************************************************************
+ * arch/x86_64/src/intel64/intel64_fpucmp.c
+ *
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.  The
+ * ASF licenses this file to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance with the
+ * License.  You may obtain a copy of the License at
+ *
+ *   http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.  See the
+ * License for the specific language governing permissions and limitations
+ * under the License.
+ *
+ ****************************************************************************/
+
+/****************************************************************************
+ * Included Files
+ ****************************************************************************/
+
+#include <nuttx/config.h>
+
+#include <stdint.h>
+#include <string.h>
+#include <nuttx/irq.h>
+
+#include "x86_64_internal.h"
+
+/****************************************************************************
+ * Pre-processor Definitions
+ ****************************************************************************/
+
+#ifdef CONFIG_ARCH_FPU
+
+/****************************************************************************
+ * Public Functions
+ ****************************************************************************/
+
+/****************************************************************************
+ * Name: up_fpucmp
+ *
+ * Description:
+ *   Compare FPU areas from thread context.
+ *
+ * Input Parameters:
+ *   saveregs1 - Pointer to the saved FPU registers.
+ *   saveregs2 - Pointer to the saved FPU registers.
+ *
+ * Returned Value:
+ *   True if FPU areas compare equal, False otherwise.
+ *
+ ****************************************************************************/
+
+bool up_fpucmp(const void *saveregs1, const void *saveregs2)
+{
+  const uint32_t *regs1 = saveregs1;
+  const uint32_t *regs2 = saveregs2;
+
+  /* IMPORTANT:
+   *
+   *   With aggresive optimization enabled (-O2/-O3), ostest FPU test will
+   *   fail. This is because the compiler will generate additional vector
+   *   instructions between subsequent up_fpucmp() calls (loop vectorization
+   *   somewhere in usleep() call), which will consequently overwrite
+   *   the expected FPU context (XMM registers).
+   */
+
+  /* XMM area starts from offset 0 */
+
+  return memcmp(&regs1[0], &regs2[0], XCPTCONTEXT_XMM_AREA_SIZE) == 0;
+}
+#endif /* CONFIG_ARCH_FPU */

arch/x86_64/src/intel64/intel64_handlers.c

@@ -86,12 +86,6 @@ static uint64_t *common_handler(int irq, uint64_t *regs)
 
   if (regs != g_current_regs)
     {
-#ifdef CONFIG_ARCH_FPU
-      /* Restore floating point registers */
-
-      up_restorefpu((uint64_t *)g_current_regs);
-#endif
-
 #ifdef CONFIG_ARCH_ADDRENV
       /* Make sure that the address environment for the previously
        * running task is closed down gracefully (data caches dump,