ASoC: Merge v6.5-rc2

Get a similar baseline to my other branches, and fixes for people using
the branch.

27 files changed, 562 insertions, 386 deletions

diff --git a/tools/include/linux/coresight-pmu.h b/tools/include/linux/coresight-pmu.h
index cef3b1c25335..51ac441a37c3 100644
--- a/tools/include/linux/coresight-pmu.h
+++ b/tools/include/linux/coresight-pmu.h
@@ -21,19 +21,6 @@
  */
 #define CORESIGHT_LEGACY_CPU_TRACE_ID(cpu)  (0x10 + (cpu * 2))
 
-/* CoreSight trace ID is currently the bottom 7 bits of the value */
-#define CORESIGHT_TRACE_ID_VAL_MASK	GENMASK(6, 0)
-
-/*
- * perf record will set the legacy meta data values as unused initially.
- * This allows perf report to manage the decoders created when dynamic
- * allocation in operation.
- */
-#define CORESIGHT_TRACE_ID_UNUSED_FLAG	BIT(31)
-
-/* Value to set for unused trace ID values */
-#define CORESIGHT_TRACE_ID_UNUSED_VAL	0x7F
-
 /*
  * Below are the definition of bit offsets for perf option, and works as
  * arbitrary values for all ETM versions.
diff --git a/tools/include/nolibc/Makefile b/tools/include/nolibc/Makefile
index 9839feafd38a..64d67b080744 100644
--- a/tools/include/nolibc/Makefile
+++ b/tools/include/nolibc/Makefile
@@ -25,8 +25,23 @@ endif
 
 nolibc_arch := $(patsubst arm64,aarch64,$(ARCH))
 arch_file := arch-$(nolibc_arch).h
-all_files := ctype.h errno.h nolibc.h signal.h stackprotector.h std.h stdint.h \
-             stdio.h stdlib.h string.h sys.h time.h types.h unistd.h
+all_files := \
+		compiler.h \
+		ctype.h \
+		errno.h \
+		nolibc.h \
+		signal.h \
+		stackprotector.h \
+		std.h \
+		stdint.h \
+		stdlib.h \
+		string.h \
+		sys.h \
+		time.h \
+		types.h \
+		unistd.h \
+		stdio.h \
+
 
 # install all headers needed to support a bare-metal compiler
 all: headers
diff --git a/tools/include/nolibc/arch-aarch64.h b/tools/include/nolibc/arch-aarch64.h
index 383baddef701..11f294a406b7 100644
--- a/tools/include/nolibc/arch-aarch64.h
+++ b/tools/include/nolibc/arch-aarch64.h
@@ -7,6 +7,8 @@
 #ifndef _NOLIBC_ARCH_AARCH64_H
 #define _NOLIBC_ARCH_AARCH64_H
 
+#include "compiler.h"
+
 /* The struct returned by the newfstatat() syscall. Differs slightly from the
  * x86_64's stat one by field ordering, so be careful.
  */
@@ -173,27 +175,30 @@ char **environ __attribute__((weak));
 const unsigned long *_auxv __attribute__((weak));
 
 /* startup code */
-void __attribute__((weak,noreturn,optimize("omit-frame-pointer"))) _start(void)
+void __attribute__((weak,noreturn,optimize("omit-frame-pointer"))) __no_stack_protector _start(void)
 {
 	__asm__ volatile (
-		"ldr x0, [sp]\n"     // argc (x0) was in the stack
-		"add x1, sp, 8\n"    // argv (x1) = sp
-		"lsl x2, x0, 3\n"    // envp (x2) = 8*argc ...
-		"add x2, x2, 8\n"    //           + 8 (skip null)
-		"add x2, x2, x1\n"   //           + argv
-		"adrp x3, environ\n"          // x3 = &environ (high bits)
-		"str x2, [x3, #:lo12:environ]\n" // store envp into environ
-		"mov x4, x2\n"       // search for auxv (follows NULL after last env)
+#ifdef _NOLIBC_STACKPROTECTOR
+		"bl __stack_chk_init\n"   /* initialize stack protector                     */
+#endif
+		"ldr x0, [sp]\n"     /* argc (x0) was in the stack                          */
+		"add x1, sp, 8\n"    /* argv (x1) = sp                                      */
+		"lsl x2, x0, 3\n"    /* envp (x2) = 8*argc ...                              */
+		"add x2, x2, 8\n"    /*           + 8 (skip null)                           */
+		"add x2, x2, x1\n"   /*           + argv                                    */
+		"adrp x3, environ\n"          /* x3 = &environ (high bits)                  */
+		"str x2, [x3, #:lo12:environ]\n" /* store envp into environ                 */
+		"mov x4, x2\n"       /* search for auxv (follows NULL after last env)       */
 		"0:\n"
-		"ldr x5, [x4], 8\n"  // x5 = *x4; x4 += 8
-		"cbnz x5, 0b\n"      // and stop at NULL after last env
-		"adrp x3, _auxv\n"   // x3 = &_auxv (high bits)
-		"str x4, [x3, #:lo12:_auxv]\n" // store x4 into _auxv
-		"and sp, x1, -16\n"  // sp must be 16-byte aligned in the callee
-		"bl main\n"          // main() returns the status code, we'll exit with it.
-		"mov x8, 93\n"       // NR_exit == 93
+		"ldr x5, [x4], 8\n"  /* x5 = *x4; x4 += 8                                   */
+		"cbnz x5, 0b\n"      /* and stop at NULL after last env                     */
+		"adrp x3, _auxv\n"   /* x3 = &_auxv (high bits)                             */
+		"str x4, [x3, #:lo12:_auxv]\n" /* store x4 into _auxv                       */
+		"and sp, x1, -16\n"  /* sp must be 16-byte aligned in the callee            */
+		"bl main\n"          /* main() returns the status code, we'll exit with it. */
+		"mov x8, 93\n"       /* NR_exit == 93                                       */
 		"svc #0\n"
 	);
 	__builtin_unreachable();
 }
-#endif // _NOLIBC_ARCH_AARCH64_H
+#endif /* _NOLIBC_ARCH_AARCH64_H */
diff --git a/tools/include/nolibc/arch-arm.h b/tools/include/nolibc/arch-arm.h
index 42499f23e73c..ca4c66987497 100644
--- a/tools/include/nolibc/arch-arm.h
+++ b/tools/include/nolibc/arch-arm.h
@@ -7,6 +7,8 @@
 #ifndef _NOLIBC_ARCH_ARM_H
 #define _NOLIBC_ARCH_ARM_H
 
+#include "compiler.h"
+
 /* The struct returned by the stat() syscall, 32-bit only, the syscall returns
  * exactly 56 bytes (stops before the unused array). In big endian, the format
  * differs as devices are returned as short only.
@@ -196,41 +198,67 @@ struct sys_stat_struct {
 	_arg1;                                                                \
 })
 
+#define my_syscall6(num, arg1, arg2, arg3, arg4, arg5, arg6)                  \
+({                                                                            \
+	register long _num  __asm__(_NOLIBC_SYSCALL_REG) = (num);             \
+	register long _arg1 __asm__ ("r0") = (long)(arg1);                    \
+	register long _arg2 __asm__ ("r1") = (long)(arg2);                    \
+	register long _arg3 __asm__ ("r2") = (long)(arg3);                    \
+	register long _arg4 __asm__ ("r3") = (long)(arg4);                    \
+	register long _arg5 __asm__ ("r4") = (long)(arg5);                    \
+	register long _arg6 __asm__ ("r5") = (long)(arg6);                    \
+	                                                                      \
+	__asm__  volatile (                                                   \
+		_NOLIBC_THUMB_SET_R7                                          \
+		"svc #0\n"                                                    \
+		_NOLIBC_THUMB_RESTORE_R7                                      \
+		: "=r"(_arg1), "=r" (_num)                                    \
+		: "r"(_arg1), "r"(_arg2), "r"(_arg3), "r"(_arg4), "r"(_arg5), \
+		  "r"(_arg6), "r"(_num)                                       \
+		: "memory", "cc", "lr"                                        \
+	);                                                                    \
+	_arg1;                                                                \
+})
+
+
 char **environ __attribute__((weak));
 const unsigned long *_auxv __attribute__((weak));
 
 /* startup code */
-void __attribute__((weak,noreturn,optimize("omit-frame-pointer"))) _start(void)
+void __attribute__((weak,noreturn,optimize("omit-frame-pointer"))) __no_stack_protector _start(void)
 {
 	__asm__ volatile (
-		"pop {%r0}\n"                 // argc was in the stack
-		"mov %r1, %sp\n"              // argv = sp
+#ifdef _NOLIBC_STACKPROTECTOR
+		"bl __stack_chk_init\n"       /* initialize stack protector                          */
+#endif
+		"pop {%r0}\n"                 /* argc was in the stack                               */
+		"mov %r1, %sp\n"              /* argv = sp                                           */
 
-		"add %r2, %r0, $1\n"          // envp = (argc + 1) ...
-		"lsl %r2, %r2, $2\n"          //        * 4        ...
-		"add %r2, %r2, %r1\n"         //        + argv
-		"ldr %r3, 1f\n"               // r3 = &environ (see below)
-		"str %r2, [r3]\n"             // store envp into environ
+		"add %r2, %r0, $1\n"          /* envp = (argc + 1) ...                               */
+		"lsl %r2, %r2, $2\n"          /*        * 4        ...                               */
+		"add %r2, %r2, %r1\n"         /*        + argv                                       */
+		"ldr %r3, 1f\n"               /* r3 = &environ (see below)                           */
+		"str %r2, [r3]\n"             /* store envp into environ                             */
 
-		"mov r4, r2\n"                // search for auxv (follows NULL after last env)
+		"mov r4, r2\n"                /* search for auxv (follows NULL after last env)       */
 		"0:\n"
-		"mov r5, r4\n"                // r5 = r4
-		"add r4, r4, #4\n"            // r4 += 4
-		"ldr r5,[r5]\n"               // r5 = *r5 = *(r4-4)
-		"cmp r5, #0\n"                // and stop at NULL after last env
+		"mov r5, r4\n"                /* r5 = r4                                             */
+		"add r4, r4, #4\n"            /* r4 += 4                                             */
+		"ldr r5,[r5]\n"               /* r5 = *r5 = *(r4-4)                                  */
+		"cmp r5, #0\n"                /* and stop at NULL after last env                     */
 		"bne 0b\n"
-		"ldr %r3, 2f\n"               // r3 = &_auxv (low bits)
-		"str r4, [r3]\n"              // store r4 into _auxv
+		"ldr %r3, 2f\n"               /* r3 = &_auxv (low bits)                              */
+		"str r4, [r3]\n"              /* store r4 into _auxv                                 */
 
-		"mov %r3, $8\n"               // AAPCS : sp must be 8-byte aligned in the
-		"neg %r3, %r3\n"              //         callee, and bl doesn't push (lr=pc)
-		"and %r3, %r3, %r1\n"         // so we do sp = r1(=sp) & r3(=-8);
-		"mov %sp, %r3\n"              //
+		"mov %r3, $8\n"               /* AAPCS : sp must be 8-byte aligned in the            */
+		"neg %r3, %r3\n"              /*         callee, and bl doesn't push (lr=pc)         */
+		"and %r3, %r3, %r1\n"         /* so we do sp = r1(=sp) & r3(=-8);                    */
+		"mov %sp, %r3\n"
 
-		"bl main\n"                   // main() returns the status code, we'll exit with it.
-		"movs r7, $1\n"               // NR_exit == 1
+		"bl main\n"                   /* main() returns the status code, we'll exit with it. */
+		"movs r7, $1\n"               /* NR_exit == 1                                        */
 		"svc $0x00\n"
-		".align 2\n"                  // below are the pointers to a few variables
+		".align 2\n"                  /* below are the pointers to a few variables           */
 		"1:\n"
 		".word environ\n"
 		"2:\n"
@@ -239,4 +267,4 @@ void __attribute__((weak,noreturn,optimize("omit-frame-pointer"))) _start(void)
 	__builtin_unreachable();
 }
 
-#endif // _NOLIBC_ARCH_ARM_H
+#endif /* _NOLIBC_ARCH_ARM_H */
diff --git a/tools/include/nolibc/arch-i386.h b/tools/include/nolibc/arch-i386.h
index 2d98d78fd3f3..3d672d925e9e 100644
--- a/tools/include/nolibc/arch-i386.h
+++ b/tools/include/nolibc/arch-i386.h
@@ -7,6 +7,8 @@
 #ifndef _NOLIBC_ARCH_I386_H
 #define _NOLIBC_ARCH_I386_H
 
+#include "compiler.h"
+
 /* The struct returned by the stat() syscall, 32-bit only, the syscall returns
  * exactly 56 bytes (stops before the unused array).
  */
@@ -181,8 +183,6 @@ struct sys_stat_struct {
 char **environ __attribute__((weak));
 const unsigned long *_auxv __attribute__((weak));
 
-#define __ARCH_SUPPORTS_STACK_PROTECTOR
-
 /* startup code */
 /*
  * i386 System V ABI mandates:
@@ -190,35 +190,35 @@ const unsigned long *_auxv __attribute__((weak));
  * 2) The deepest stack frame should be set to zero
  *
  */
-void __attribute__((weak,noreturn,optimize("omit-frame-pointer"),no_stack_protector)) _start(void)
+void __attribute__((weak,noreturn,optimize("omit-frame-pointer"))) __no_stack_protector _start(void)
 {
 	__asm__ volatile (
-#ifdef NOLIBC_STACKPROTECTOR
-		"call __stack_chk_init\n"   // initialize stack protector
+#ifdef _NOLIBC_STACKPROTECTOR
+		"call __stack_chk_init\n"   /* initialize stack protector                    */
 #endif
-		"pop %eax\n"                // argc   (first arg, %eax)
-		"mov %esp, %ebx\n"          // argv[] (second arg, %ebx)
-		"lea 4(%ebx,%eax,4),%ecx\n" // then a NULL then envp (third arg, %ecx)
-		"mov %ecx, environ\n"       // save environ
-		"xor %ebp, %ebp\n"          // zero the stack frame
-		"mov %ecx, %edx\n"          // search for auxv (follows NULL after last env)
+		"pop %eax\n"                /* argc   (first arg, %eax)                      */
+		"mov %esp, %ebx\n"          /* argv[] (second arg, %ebx)                     */
+		"lea 4(%ebx,%eax,4),%ecx\n" /* then a NULL then envp (third arg, %ecx)       */
+		"mov %ecx, environ\n"       /* save environ                                  */
+		"xor %ebp, %ebp\n"          /* zero the stack frame                          */
+		"mov %ecx, %edx\n"          /* search for auxv (follows NULL after last env) */
 		"0:\n"
-		"add $4, %edx\n"            // search for auxv using edx, it follows the
-		"cmp -4(%edx), %ebp\n"      // ... NULL after last env (ebp is zero here)
+		"add $4, %edx\n"            /* search for auxv using edx, it follows the     */
+		"cmp -4(%edx), %ebp\n"      /* ... NULL after last env (ebp is zero here)    */
 		"jnz 0b\n"
-		"mov %edx, _auxv\n"         // save it into _auxv
-		"and $-16, %esp\n"          // x86 ABI : esp must be 16-byte aligned before
-		"sub $4, %esp\n"            // the call instruction (args are aligned)
-		"push %ecx\n"               // push all registers on the stack so that we
-		"push %ebx\n"               // support both regparm and plain stack modes
+		"mov %edx, _auxv\n"         /* save it into _auxv                            */
+		"and $-16, %esp\n"          /* x86 ABI : esp must be 16-byte aligned before  */
+		"sub $4, %esp\n"            /* the call instruction (args are aligned)       */
+		"push %ecx\n"               /* push all registers on the stack so that we    */
+		"push %ebx\n"               /* support both regparm and plain stack modes    */
 		"push %eax\n"
-		"call main\n"               // main() returns the status code in %eax
-		"mov %eax, %ebx\n"          // retrieve exit code (32-bit int)
-		"movl $1, %eax\n"           // NR_exit == 1
-		"int $0x80\n"               // exit now
-		"hlt\n"                     // ensure it does not
+		"call main\n"               /* main() returns the status code in %eax        */
+		"mov %eax, %ebx\n"          /* retrieve exit code (32-bit int)               */
+		"movl $1, %eax\n"           /* NR_exit == 1                                  */
+		"int $0x80\n"               /* exit now                                      */
+		"hlt\n"                     /* ensure it does not                            */
 	);
 	__builtin_unreachable();
 }
 
-#endif // _NOLIBC_ARCH_I386_H
+#endif /* _NOLIBC_ARCH_I386_H */
diff --git a/tools/include/nolibc/arch-loongarch.h b/tools/include/nolibc/arch-loongarch.h
index 029ee3cd6baf..ad3f266e7093 100644
--- a/tools/include/nolibc/arch-loongarch.h
+++ b/tools/include/nolibc/arch-loongarch.h
@@ -7,6 +7,8 @@
 #ifndef _NOLIBC_ARCH_LOONGARCH_H
 #define _NOLIBC_ARCH_LOONGARCH_H
 
+#include "compiler.h"
+
 /* Syscalls for LoongArch :
  *   - stack is 16-byte aligned
  *   - syscall number is passed in a7
@@ -158,7 +160,7 @@ const unsigned long *_auxv __attribute__((weak));
 #define LONG_ADDI    "addi.w"
 #define LONG_SLL     "slli.w"
 #define LONG_BSTRINS "bstrins.w"
-#else // __loongarch_grlen == 64
+#else /* __loongarch_grlen == 64 */
 #define LONGLOG      "3"
 #define SZREG        "8"
 #define REG_L        "ld.d"
@@ -170,31 +172,34 @@ const unsigned long *_auxv __attribute__((weak));
 #endif
 
 /* startup code */
-void __attribute__((weak,noreturn,optimize("omit-frame-pointer"))) _start(void)
+void __attribute__((weak,noreturn,optimize("omit-frame-pointer"))) __no_stack_protector _start(void)
 {
 	__asm__ volatile (
-		REG_L        " $a0, $sp, 0\n"         // argc (a0) was in the stack
-		LONG_ADDI    " $a1, $sp, "SZREG"\n"   // argv (a1) = sp + SZREG
-		LONG_SLL     " $a2, $a0, "LONGLOG"\n" // envp (a2) = SZREG*argc ...
-		LONG_ADDI    " $a2, $a2, "SZREG"\n"   //             + SZREG (skip null)
-		LONG_ADD     " $a2, $a2, $a1\n"       //             + argv
-
-		"move          $a3, $a2\n"            // iterate a3 over envp to find auxv (after NULL)
-		"0:\n"                                // do {
-		REG_L        " $a4, $a3, 0\n"         //   a4 = *a3;
-		LONG_ADDI    " $a3, $a3, "SZREG"\n"   //   a3 += sizeof(void*);
-		"bne           $a4, $zero, 0b\n"      // } while (a4);
-		"la.pcrel      $a4, _auxv\n"          // a4 = &_auxv
-		LONG_S       " $a3, $a4, 0\n"         // store a3 into _auxv
-
-		"la.pcrel      $a3, environ\n"        // a3 = &environ
-		LONG_S       " $a2, $a3, 0\n"         // store envp(a2) into environ
-		LONG_BSTRINS " $sp, $zero, 3, 0\n"    // sp must be 16-byte aligned
-		"bl            main\n"                // main() returns the status code, we'll exit with it.
-		"li.w          $a7, 93\n"             // NR_exit == 93
+#ifdef _NOLIBC_STACKPROTECTOR
+		"bl __stack_chk_init\n"               /* initialize stack protector                          */
+#endif
+		REG_L        " $a0, $sp, 0\n"         /* argc (a0) was in the stack                          */
+		LONG_ADDI    " $a1, $sp, "SZREG"\n"   /* argv (a1) = sp + SZREG                              */
+		LONG_SLL     " $a2, $a0, "LONGLOG"\n" /* envp (a2) = SZREG*argc ...                          */
+		LONG_ADDI    " $a2, $a2, "SZREG"\n"   /*             + SZREG (skip null)                     */
+		LONG_ADD     " $a2, $a2, $a1\n"       /*             + argv                                  */
+
+		"move          $a3, $a2\n"            /* iterate a3 over envp to find auxv (after NULL)      */
+		"0:\n"                                /* do {                                                */
+		REG_L        " $a4, $a3, 0\n"         /*   a4 = *a3;                                         */
+		LONG_ADDI    " $a3, $a3, "SZREG"\n"   /*   a3 += sizeof(void*);                              */
+		"bne           $a4, $zero, 0b\n"      /* } while (a4);                                       */
+		"la.pcrel      $a4, _auxv\n"          /* a4 = &_auxv                                         */
+		LONG_S       " $a3, $a4, 0\n"         /* store a3 into _auxv                                 */
+
+		"la.pcrel      $a3, environ\n"        /* a3 = &environ                                       */
+		LONG_S       " $a2, $a3, 0\n"         /* store envp(a2) into environ                         */
+		LONG_BSTRINS " $sp, $zero, 3, 0\n"    /* sp must be 16-byte aligned                          */
+		"bl            main\n"                /* main() returns the status code, we'll exit with it. */
+		"li.w          $a7, 93\n"             /* NR_exit == 93                                       */
 		"syscall       0\n"
 	);
 	__builtin_unreachable();
 }
 
-#endif // _NOLIBC_ARCH_LOONGARCH_H
+#endif /* _NOLIBC_ARCH_LOONGARCH_H */
diff --git a/tools/include/nolibc/arch-mips.h b/tools/include/nolibc/arch-mips.h
index bf83432d23ed..db24e0837a39 100644
--- a/tools/include/nolibc/arch-mips.h
+++ b/tools/include/nolibc/arch-mips.h
@@ -7,6 +7,8 @@
 #ifndef _NOLIBC_ARCH_MIPS_H
 #define _NOLIBC_ARCH_MIPS_H
 
+#include "compiler.h"
+
 /* The struct returned by the stat() syscall. 88 bytes are returned by the
  * syscall.
  */
@@ -180,45 +182,49 @@ char **environ __attribute__((weak));
 const unsigned long *_auxv __attribute__((weak));
 
 /* startup code, note that it's called __start on MIPS */
-void __attribute__((weak,noreturn,optimize("omit-frame-pointer"))) __start(void)
+void __attribute__((weak,noreturn,optimize("omit-frame-pointer"))) __no_stack_protector __start(void)
 {
 	__asm__ volatile (
-		//".set nomips16\n"
+		/*".set nomips16\n"*/
 		".set push\n"
 		".set    noreorder\n"
 		".option pic0\n"
-		//".ent __start\n"
-		//"__start:\n"
-		"lw $a0,($sp)\n"        // argc was in the stack
-		"addiu  $a1, $sp, 4\n"  // argv = sp + 4
-		"sll $a2, $a0, 2\n"     // a2 = argc * 4
-		"add   $a2, $a2, $a1\n" // envp = argv + 4*argc ...
-		"addiu $a2, $a2, 4\n"   //        ... + 4
-		"lui $a3, %hi(environ)\n"     // load environ into a3 (hi)
-		"addiu $a3, %lo(environ)\n"   // load environ into a3 (lo)
-		"sw $a2,($a3)\n"              // store envp(a2) into environ
-
-		"move $t0, $a2\n"             // iterate t0 over envp, look for NULL
-		"0:"                          // do {
-		"lw $a3, ($t0)\n"             //   a3=*(t0);
-		"bne $a3, $0, 0b\n"           // } while (a3);
-		"addiu $t0, $t0, 4\n"         // delayed slot: t0+=4;
-		"lui $a3, %hi(_auxv)\n"       // load _auxv into a3 (hi)
-		"addiu $a3, %lo(_auxv)\n"     // load _auxv into a3 (lo)
-		"sw $t0, ($a3)\n"             // store t0 into _auxv
+#ifdef _NOLIBC_STACKPROTECTOR
+		"jal __stack_chk_init\n" /* initialize stack protector                         */
+		"nop\n"                  /* delayed slot                                       */
+#endif
+		/*".ent __start\n"*/
+		/*"__start:\n"*/
+		"lw $a0,($sp)\n"        /* argc was in the stack                               */
+		"addiu  $a1, $sp, 4\n"  /* argv = sp + 4                                       */
+		"sll $a2, $a0, 2\n"     /* a2 = argc * 4                                       */
+		"add   $a2, $a2, $a1\n" /* envp = argv + 4*argc ...                            */
+		"addiu $a2, $a2, 4\n"   /*        ... + 4                                      */
+		"lui $a3, %hi(environ)\n"     /* load environ into a3 (hi)                     */
+		"addiu $a3, %lo(environ)\n"   /* load environ into a3 (lo)                     */
+		"sw $a2,($a3)\n"              /* store envp(a2) into environ                   */
+
+		"move $t0, $a2\n"             /* iterate t0 over envp, look for NULL           */
+		"0:"                          /* do {                                          */
+		"lw $a3, ($t0)\n"             /*   a3=*(t0);                                   */
+		"bne $a3, $0, 0b\n"           /* } while (a3);                                 */
+		"addiu $t0, $t0, 4\n"         /* delayed slot: t0+=4;                          */
+		"lui $a3, %hi(_auxv)\n"       /* load _auxv into a3 (hi)                       */
+		"addiu $a3, %lo(_auxv)\n"     /* load _auxv into a3 (lo)                       */
+		"sw $t0, ($a3)\n"             /* store t0 into _auxv                           */
 
 		"li $t0, -8\n"
-		"and $sp, $sp, $t0\n"   // sp must be 8-byte aligned
-		"addiu $sp,$sp,-16\n"   // the callee expects to save a0..a3 there!
-		"jal main\n"            // main() returns the status code, we'll exit with it.
-		"nop\n"                 // delayed slot
-		"move $a0, $v0\n"       // retrieve 32-bit exit code from v0
-		"li $v0, 4001\n"        // NR_exit == 4001
+		"and $sp, $sp, $t0\n"   /* sp must be 8-byte aligned                           */
+		"addiu $sp,$sp,-16\n"   /* the callee expects to save a0..a3 there!            */
+		"jal main\n"            /* main() returns the status code, we'll exit with it. */
+		"nop\n"                 /* delayed slot                                        */
+		"move $a0, $v0\n"       /* retrieve 32-bit exit code from v0                   */
+		"li $v0, 4001\n"        /* NR_exit == 4001                                     */
 		"syscall\n"
-		//".end __start\n"
+		/*".end __start\n"*/
 		".set pop\n"
 	);
 	__builtin_unreachable();
 }
 
-#endif // _NOLIBC_ARCH_MIPS_H
+#endif /* _NOLIBC_ARCH_MIPS_H */
diff --git a/tools/include/nolibc/arch-riscv.h b/tools/include/nolibc/arch-riscv.h
index e197fcb10ac0..a2e8564e66d6 100644
--- a/tools/include/nolibc/arch-riscv.h
+++ b/tools/include/nolibc/arch-riscv.h
@@ -7,6 +7,8 @@
 #ifndef _NOLIBC_ARCH_RISCV_H
 #define _NOLIBC_ARCH_RISCV_H
 
+#include "compiler.h"
+
 struct sys_stat_struct {
 	unsigned long	st_dev;		/* Device.  */
 	unsigned long	st_ino;		/* File serial number.  */
@@ -33,9 +35,13 @@ struct sys_stat_struct {
 #if   __riscv_xlen == 64
 #define PTRLOG "3"
 #define SZREG  "8"
+#define REG_L  "ld"
+#define REG_S  "sd"
 #elif __riscv_xlen == 32
 #define PTRLOG "2"
 #define SZREG  "4"
+#define REG_L  "lw"
+#define REG_S  "sw"
 #endif
 
 /* Syscalls for RISCV :
@@ -174,35 +180,38 @@ char **environ __attribute__((weak));
 const unsigned long *_auxv __attribute__((weak));
 
 /* startup code */
-void __attribute__((weak,noreturn,optimize("omit-frame-pointer"))) _start(void)
+void __attribute__((weak,noreturn,optimize("omit-frame-pointer"))) __no_stack_protector _start(void)
 {
 	__asm__ volatile (
 		".option push\n"
 		".option norelax\n"
 		"lla   gp, __global_pointer$\n"
 		".option pop\n"
-		"lw    a0, 0(sp)\n"          // argc (a0) was in the stack
-		"add   a1, sp, "SZREG"\n"    // argv (a1) = sp
-		"slli  a2, a0, "PTRLOG"\n"   // envp (a2) = SZREG*argc ...
-		"add   a2, a2, "SZREG"\n"    //             + SZREG (skip null)
-		"add   a2,a2,a1\n"           //             + argv
-
-		"add   a3, a2, zero\n"       // iterate a3 over envp to find auxv (after NULL)
-		"0:\n"                       // do {
-		"ld    a4, 0(a3)\n"          //   a4 = *a3;
-		"add   a3, a3, "SZREG"\n"    //   a3 += sizeof(void*);
-		"bne   a4, zero, 0b\n"       // } while (a4);
-		"lui   a4, %hi(_auxv)\n"     // a4 = &_auxv (high bits)
-		"sd    a3, %lo(_auxv)(a4)\n" // store a3 into _auxv
-
-		"lui a3, %hi(environ)\n"     // a3 = &environ (high bits)
-		"sd a2,%lo(environ)(a3)\n"   // store envp(a2) into environ
-		"andi  sp,a1,-16\n"          // sp must be 16-byte aligned
-		"call  main\n"               // main() returns the status code, we'll exit with it.
-		"li a7, 93\n"                // NR_exit == 93
+#ifdef _NOLIBC_STACKPROTECTOR
+		"call __stack_chk_init\n"    /* initialize stack protector                          */
+#endif
+		REG_L" a0, 0(sp)\n"          /* argc (a0) was in the stack                          */
+		"add   a1, sp, "SZREG"\n"    /* argv (a1) = sp                                      */
+		"slli  a2, a0, "PTRLOG"\n"   /* envp (a2) = SZREG*argc ...                          */
+		"add   a2, a2, "SZREG"\n"    /*             + SZREG (skip null)                     */
+		"add   a2,a2,a1\n"           /*             + argv                                  */
+
+		"add   a3, a2, zero\n"       /* iterate a3 over envp to find auxv (after NULL)      */
+		"0:\n"                       /* do {                                                */
+		REG_L" a4, 0(a3)\n"          /*   a4 = *a3;                                         */
+		"add   a3, a3, "SZREG"\n"    /*   a3 += sizeof(void*);                              */
+		"bne   a4, zero, 0b\n"       /* } while (a4);                                       */
+		"lui   a4, %hi(_auxv)\n"     /* a4 = &_auxv (high bits)                             */
+		REG_S" a3, %lo(_auxv)(a4)\n" /* store a3 into _auxv                                 */
+
+		"lui   a3, %hi(environ)\n"   /* a3 = &environ (high bits)                           */
+		REG_S" a2,%lo(environ)(a3)\n"/* store envp(a2) into environ                         */
+		"andi  sp,a1,-16\n"          /* sp must be 16-byte aligned                          */
+		"call  main\n"               /* main() returns the status code, we'll exit with it. */
+		"li a7, 93\n"                /* NR_exit == 93                                       */
 		"ecall\n"
 	);
 	__builtin_unreachable();
 }
 
-#endif // _NOLIBC_ARCH_RISCV_H
+#endif /* _NOLIBC_ARCH_RISCV_H */
diff --git a/tools/include/nolibc/arch-s390.h b/tools/include/nolibc/arch-s390.h
index 6b0e54ed543d..516dff5bff8b 100644
--- a/tools/include/nolibc/arch-s390.h
+++ b/tools/include/nolibc/arch-s390.h
@@ -5,8 +5,11 @@
 
 #ifndef _NOLIBC_ARCH_S390_H
 #define _NOLIBC_ARCH_S390_H
+#include <asm/signal.h>
 #include <asm/unistd.h>
 
+#include "compiler.h"
+
 /* The struct returned by the stat() syscall, equivalent to stat64(). The
  * syscall returns 116 bytes and stops in the middle of __unused.
  */
@@ -163,7 +166,7 @@ char **environ __attribute__((weak));
 const unsigned long *_auxv __attribute__((weak));
 
 /* startup code */
-void __attribute__((weak,noreturn,optimize("omit-frame-pointer"))) _start(void)
+void __attribute__((weak,noreturn,optimize("omit-frame-pointer"))) __no_stack_protector _start(void)
 {
 	__asm__ volatile (
 		"lg	%r2,0(%r15)\n"		/* argument count */
@@ -223,4 +226,12 @@ void *sys_mmap(void *addr, size_t length, int prot, int flags, int fd,
 	return (void *)my_syscall1(__NR_mmap, &args);
 }
 #define sys_mmap sys_mmap
-#endif // _NOLIBC_ARCH_S390_H
+
+static __attribute__((unused))
+pid_t sys_fork(void)
+{
+	return my_syscall5(__NR_clone, 0, SIGCHLD, 0, 0, 0);
+}
+#define sys_fork sys_fork
+
+#endif /* _NOLIBC_ARCH_S390_H */
diff --git a/tools/include/nolibc/arch-x86_64.h b/tools/include/nolibc/arch-x86_64.h
index f7f2a11d4c3b..6fc4d8392742 100644
--- a/tools/include/nolibc/arch-x86_64.h
+++ b/tools/include/nolibc/arch-x86_64.h
@@ -7,6 +7,8 @@
 #ifndef _NOLIBC_ARCH_X86_64_H
 #define _NOLIBC_ARCH_X86_64_H
 
+#include "compiler.h"
+
 /* The struct returned by the stat() syscall, equivalent to stat64(). The
  * syscall returns 116 bytes and stops in the middle of __unused.
  */
@@ -181,8 +183,6 @@ struct sys_stat_struct {
 char **environ __attribute__((weak));
 const unsigned long *_auxv __attribute__((weak));
 
-#define __ARCH_SUPPORTS_STACK_PROTECTOR
-
 /* startup code */
 /*
  * x86-64 System V ABI mandates:
@@ -190,31 +190,31 @@ const unsigned long *_auxv __attribute__((weak));
  * 2) The deepest stack frame should be zero (the %rbp).
  *
  */
-void __attribute__((weak,noreturn,optimize("omit-frame-pointer"))) _start(void)
+void __attribute__((weak,noreturn,optimize("omit-frame-pointer"))) __no_stack_protector _start(void)
 {
 	__asm__ volatile (
-#ifdef NOLIBC_STACKPROTECTOR
-		"call __stack_chk_init\n"   // initialize stack protector
+#ifdef _NOLIBC_STACKPROTECTOR
+		"call __stack_chk_init\n"   /* initialize stack protector                          */
 #endif
-		"pop %rdi\n"                // argc   (first arg, %rdi)
-		"mov %rsp, %rsi\n"          // argv[] (second arg, %rsi)
-		"lea 8(%rsi,%rdi,8),%rdx\n" // then a NULL then envp (third arg, %rdx)
-		"mov %rdx, environ\n"       // save environ
-		"xor %ebp, %ebp\n"          // zero the stack frame
-		"mov %rdx, %rax\n"          // search for auxv (follows NULL after last env)
+		"pop %rdi\n"                /* argc   (first arg, %rdi)                            */
+		"mov %rsp, %rsi\n"          /* argv[] (second arg, %rsi)                           */
+		"lea 8(%rsi,%rdi,8),%rdx\n" /* then a NULL then envp (third arg, %rdx)             */
+		"mov %rdx, environ\n"       /* save environ                                        */
+		"xor %ebp, %ebp\n"          /* zero the stack frame                                */
+		"mov %rdx, %rax\n"          /* search for auxv (follows NULL after last env)       */
 		"0:\n"
-		"add $8, %rax\n"            // search for auxv using rax, it follows the
-		"cmp -8(%rax), %rbp\n"      // ... NULL after last env (rbp is zero here)
+		"add $8, %rax\n"            /* search for auxv using rax, it follows the           */
+		"cmp -8(%rax), %rbp\n"      /* ... NULL after last env (rbp is zero here)          */
 		"jnz 0b\n"
-		"mov %rax, _auxv\n"         // save it into _auxv
-		"and $-16, %rsp\n"          // x86 ABI : esp must be 16-byte aligned before call
-		"call main\n"               // main() returns the status code, we'll exit with it.
-		"mov %eax, %edi\n"          // retrieve exit code (32 bit)
-		"mov $60, %eax\n"           // NR_exit == 60
-		"syscall\n"                 // really exit
-		"hlt\n"                     // ensure it does not return
+		"mov %rax, _auxv\n"         /* save it into _auxv                                  */
+		"and $-16, %rsp\n"          /* x86 ABI : esp must be 16-byte aligned before call   */
+		"call main\n"               /* main() returns the status code, we'll exit with it. */
+		"mov %eax, %edi\n"          /* retrieve exit code (32 bit)                         */
+		"mov $60, %eax\n"           /* NR_exit == 60                                       */
+		"syscall\n"                 /* really exit                                         */
+		"hlt\n"                     /* ensure it does not return                           */
 	);
 	__builtin_unreachable();
 }
 
-#endif // _NOLIBC_ARCH_X86_64_H
+#endif /* _NOLIBC_ARCH_X86_64_H */
diff --git a/tools/include/nolibc/arch.h b/tools/include/nolibc/arch.h
index 2d5386a8d6aa..82b43935650f 100644
--- a/tools/include/nolibc/arch.h
+++ b/tools/include/nolibc/arch.h
@@ -7,7 +7,7 @@
  * the syscall declarations and the _start code definition. This is the only
  * global part. On all architectures the kernel puts everything in the stack
  * before jumping to _start just above us, without any return address (_start
- * is not a function but an entry pint). So at the stack pointer we find argc.
+ * is not a function but an entry point). So at the stack pointer we find argc.
  * Then argv[] begins, and ends at the first NULL. Then we have envp which
  * starts and ends with a NULL as well. So envp=argv+argc+1.
  */
diff --git a/tools/include/nolibc/compiler.h b/tools/include/nolibc/compiler.h
new file mode 100644
index 000000000000..beddc3665d69
--- /dev/null
+++ b/tools/include/nolibc/compiler.h
@@ -0,0 +1,25 @@
+/* SPDX-License-Identifier: LGPL-2.1 OR MIT */
+/*
+ * NOLIBC compiler support header
+ * Copyright (C) 2023 Thomas Weißschuh <linux@weissschuh.net>
+ */
+#ifndef _NOLIBC_COMPILER_H
+#define _NOLIBC_COMPILER_H
+
+#if defined(__SSP__) || defined(__SSP_STRONG__) || defined(__SSP_ALL__) || defined(__SSP_EXPLICIT__)
+
+#define _NOLIBC_STACKPROTECTOR
+
+#endif /* defined(__SSP__) ... */
+
+#if defined(__has_attribute)
+#  if __has_attribute(no_stack_protector)
+#    define __no_stack_protector __attribute__((no_stack_protector))
+#  else
+#    define __no_stack_protector __attribute__((__optimize__("-fno-stack-protector")))
+#  endif
+#else
+#  define __no_stack_protector __attribute__((__optimize__("-fno-stack-protector")))
+#endif /* defined(__has_attribute) */
+
+#endif /* _NOLIBC_COMPILER_H */
diff --git a/tools/include/nolibc/nolibc.h b/tools/include/nolibc/nolibc.h
index 04739a6293c4..05a228a6ee78 100644
--- a/tools/include/nolibc/nolibc.h
+++ b/tools/include/nolibc/nolibc.h
@@ -99,11 +99,11 @@
 #include "sys.h"
 #include "ctype.h"
 #include "signal.h"
+#include "unistd.h"
 #include "stdio.h"
 #include "stdlib.h"
 #include "string.h"
 #include "time.h"
-#include "unistd.h"
 #include "stackprotector.h"
 
 /* Used by programs to avoid std includes */
diff --git a/tools/include/nolibc/stackprotector.h b/tools/include/nolibc/stackprotector.h
index d119cbbbc256..88f7b2d098ff 100644
--- a/tools/include/nolibc/stackprotector.h
+++ b/tools/include/nolibc/stackprotector.h
@@ -7,13 +7,9 @@
 #ifndef _NOLIBC_STACKPROTECTOR_H
 #define _NOLIBC_STACKPROTECTOR_H
 
-#include "arch.h"
+#include "compiler.h"
 
-#if defined(NOLIBC_STACKPROTECTOR)
-
-#if !defined(__ARCH_SUPPORTS_STACK_PROTECTOR)
-#error "nolibc does not support stack protectors on this arch"
-#endif
+#if defined(_NOLIBC_STACKPROTECTOR)
 
 #include "sys.h"
 #include "stdlib.h"
@@ -41,13 +37,14 @@ void __stack_chk_fail_local(void)
 __attribute__((weak,section(".data.nolibc_stack_chk")))
 uintptr_t __stack_chk_guard;
 
-__attribute__((weak,no_stack_protector,section(".text.nolibc_stack_chk")))
+__attribute__((weak,section(".text.nolibc_stack_chk"))) __no_stack_protector
 void __stack_chk_init(void)
 {
 	my_syscall3(__NR_getrandom, &__stack_chk_guard, sizeof(__stack_chk_guard), 0);
-	/* a bit more randomness in case getrandom() fails */
-	__stack_chk_guard ^= (uintptr_t) &__stack_chk_guard;
+	/* a bit more randomness in case getrandom() fails, ensure the guard is never 0 */
+	if (__stack_chk_guard != (uintptr_t) &__stack_chk_guard)
+		__stack_chk_guard ^= (uintptr_t) &__stack_chk_guard;
 }
-#endif // defined(NOLIBC_STACKPROTECTOR)
+#endif /* defined(_NOLIBC_STACKPROTECTOR) */
 
-#endif // _NOLIBC_STACKPROTECTOR_H
+#endif /* _NOLIBC_STACKPROTECTOR_H */
diff --git a/tools/include/nolibc/stdint.h b/tools/include/nolibc/stdint.h
index c1ce4f5e0603..4b282435a59a 100644
--- a/tools/include/nolibc/stdint.h
+++ b/tools/include/nolibc/stdint.h
@@ -36,8 +36,8 @@ typedef  ssize_t       int_fast16_t;
 typedef   size_t      uint_fast16_t;
 typedef  ssize_t       int_fast32_t;
 typedef   size_t      uint_fast32_t;
-typedef  ssize_t       int_fast64_t;
-typedef   size_t      uint_fast64_t;
+typedef  int64_t       int_fast64_t;
+typedef uint64_t      uint_fast64_t;
 
 typedef  int64_t           intmax_t;
 typedef uint64_t          uintmax_t;
@@ -84,16 +84,30 @@ typedef uint64_t          uintmax_t;
 #define  INT_FAST8_MIN   INT8_MIN
 #define INT_FAST16_MIN   INTPTR_MIN
 #define INT_FAST32_MIN   INTPTR_MIN
-#define INT_FAST64_MIN   INTPTR_MIN
+#define INT_FAST64_MIN   INT64_MIN
 
 #define  INT_FAST8_MAX   INT8_MAX
 #define INT_FAST16_MAX   INTPTR_MAX
 #define INT_FAST32_MAX   INTPTR_MAX
-#define INT_FAST64_MAX   INTPTR_MAX
+#define INT_FAST64_MAX   INT64_MAX
 
 #define  UINT_FAST8_MAX  UINT8_MAX
 #define UINT_FAST16_MAX  SIZE_MAX
 #define UINT_FAST32_MAX  SIZE_MAX
-#define UINT_FAST64_MAX  SIZE_MAX
+#define UINT_FAST64_MAX  UINT64_MAX
+
+#ifndef INT_MIN
+#define INT_MIN          (-__INT_MAX__ - 1)
+#endif
+#ifndef INT_MAX
+#define INT_MAX          __INT_MAX__
+#endif
+
+#ifndef LONG_MIN
+#define LONG_MIN         (-__LONG_MAX__ - 1)
+#endif
+#ifndef LONG_MAX
+#define LONG_MAX         __LONG_MAX__
+#endif
 
 #endif /* _NOLIBC_STDINT_H */
diff --git a/tools/include/nolibc/stdio.h b/tools/include/nolibc/stdio.h
index 6cbbb52836a0..0eef91daf289 100644
--- a/tools/include/nolibc/stdio.h
+++ b/tools/include/nolibc/stdio.h
@@ -21,17 +21,75 @@
 #define EOF (-1)
 #endif
 
-/* just define FILE as a non-empty type */
+/* just define FILE as a non-empty type. The value of the pointer gives
+ * the FD: FILE=~fd for fd>=0 or NULL for fd<0. This way positive FILE
+ * are immediately identified as abnormal entries (i.e. possible copies
+ * of valid pointers to something else).
+ */
 typedef struct FILE {
 	char dummy[1];
 } FILE;
 
-/* We define the 3 common stdio files as constant invalid pointers that
- * are easily recognized.
- */
-static __attribute__((unused)) FILE* const stdin  = (FILE*)-3;
-static __attribute__((unused)) FILE* const stdout = (FILE*)-2;
-static __attribute__((unused)) FILE* const stderr = (FILE*)-1;
+static __attribute__((unused)) FILE* const stdin  = (FILE*)(intptr_t)~STDIN_FILENO;
+static __attribute__((unused)) FILE* const stdout = (FILE*)(intptr_t)~STDOUT_FILENO;
+static __attribute__((unused)) FILE* const stderr = (FILE*)(intptr_t)~STDERR_FILENO;
+
+/* provides a FILE* equivalent of fd. The mode is ignored. */
+static __attribute__((unused))
+FILE *fdopen(int fd, const char *mode __attribute__((unused)))
+{
+	if (fd < 0) {
+		SET_ERRNO(EBADF);
+		return NULL;
+	}
+	return (FILE*)(intptr_t)~fd;
+}
+
+/* provides the fd of stream. */
+static __attribute__((unused))
+int fileno(FILE *stream)
+{
+	intptr_t i = (intptr_t)stream;
+
+	if (i >= 0) {
+		SET_ERRNO(EBADF);
+		return -1;
+	}
+	return ~i;
+}
+
+/* flush a stream. */
+static __attribute__((unused))
+int fflush(FILE *stream)
+{
+	intptr_t i = (intptr_t)stream;
+
+	/* NULL is valid here. */
+	if (i > 0) {
+		SET_ERRNO(EBADF);
+		return -1;
+	}
+
+	/* Don't do anything, nolibc does not support buffering. */
+	return 0;
+}
+
+/* flush a stream. */
+static __attribute__((unused))
+int fclose(FILE *stream)
+{
+	intptr_t i = (intptr_t)stream;
+
+	if (i >= 0) {
+		SET_ERRNO(EBADF);
+		return -1;
+	}
+
+	if (close(~i))
+		return EOF;
+
+	return 0;
+}
 
 /* getc(), fgetc(), getchar() */
 
@@ -41,14 +99,8 @@ static __attribute__((unused))
 int fgetc(FILE* stream)
 {
 	unsigned char ch;
-	int fd;
 
-	if (stream < stdin || stream > stderr)
-		return EOF;
-
-	fd = 3 + (long)stream;
-
-	if (read(fd, &ch, 1) <= 0)
+	if (read(fileno(stream), &ch, 1) <= 0)
 		return EOF;
 	return ch;
 }
@@ -68,14 +120,8 @@ static __attribute__((unused))
 int fputc(int c, FILE* stream)
 {
 	unsigned char ch = c;
-	int fd;
-
-	if (stream < stdin || stream > stderr)
-		return EOF;
-
-	fd = 3 + (long)stream;
 
-	if (write(fd, &ch, 1) <= 0)
+	if (write(fileno(stream), &ch, 1) <= 0)
 		return EOF;
 	return ch;
 }
@@ -96,12 +142,7 @@ static __attribute__((unused))
 int _fwrite(const void *buf, size_t size, FILE *stream)
 {
 	ssize_t ret;
-	int fd;
-
-	if (stream < stdin || stream > stderr)
-		return EOF;
-
-	fd = 3 + (long)stream;
+	int fd = fileno(stream);
 
 	while (size) {
 		ret = write(fd, buf, size);
diff --git a/tools/include/nolibc/stdlib.h b/tools/include/nolibc/stdlib.h
index 894c955d027e..902162f80337 100644
--- a/tools/include/nolibc/stdlib.h
+++ b/tools/include/nolibc/stdlib.h
@@ -102,7 +102,7 @@ char *_getenv(const char *name, char **environ)
 	return NULL;
 }
 
-static inline __attribute__((unused,always_inline))
+static __inline__ __attribute__((unused,always_inline))
 char *getenv(const char *name)
 {
 	extern char **environ;
@@ -231,7 +231,7 @@ int utoh_r(unsigned long in, char *buffer)
 /* converts unsigned long <in> to an hex string using the static itoa_buffer
  * and returns the pointer to that string.
  */
-static inline __attribute__((unused))
+static __inline__ __attribute__((unused))
 char *utoh(unsigned long in)
 {
 	utoh_r(in, itoa_buffer);
@@ -293,7 +293,7 @@ int itoa_r(long in, char *buffer)
 /* for historical compatibility, same as above but returns the pointer to the
  * buffer.
  */
-static inline __attribute__((unused))
+static __inline__ __attribute__((unused))
 char *ltoa_r(long in, char *buffer)
 {
 	itoa_r(in, buffer);
@@ -303,7 +303,7 @@ char *ltoa_r(long in, char *buffer)
 /* converts long integer <in> to a string using the static itoa_buffer and
  * returns the pointer to that string.
  */
-static inline __attribute__((unused))
+static __inline__ __attribute__((unused))
 char *itoa(long in)
 {
 	itoa_r(in, itoa_buffer);
@@ -313,7 +313,7 @@ char *itoa(long in)
 /* converts long integer <in> to a string using the static itoa_buffer and
  * returns the pointer to that string. Same as above, for compatibility.
  */
-static inline __attribute__((unused))
+static __inline__ __attribute__((unused))
 char *ltoa(long in)
 {
 	itoa_r(in, itoa_buffer);
@@ -323,7 +323,7 @@ char *ltoa(long in)
 /* converts unsigned long integer <in> to a string using the static itoa_buffer
  * and returns the pointer to that string.
  */
-static inline __attribute__((unused))
+static __inline__ __attribute__((unused))
 char *utoa(unsigned long in)
 {
 	utoa_r(in, itoa_buffer);
@@ -367,7 +367,7 @@ int u64toh_r(uint64_t in, char *buffer)
 /* converts uint64_t <in> to an hex string using the static itoa_buffer and
  * returns the pointer to that string.
  */
-static inline __attribute__((unused))
+static __inline__ __attribute__((unused))
 char *u64toh(uint64_t in)
 {
 	u64toh_r(in, itoa_buffer);
@@ -429,7 +429,7 @@ int i64toa_r(int64_t in, char *buffer)
 /* converts int64_t <in> to a string using the static itoa_buffer and returns
  * the pointer to that string.
  */
-static inline __attribute__((unused))
+static __inline__ __attribute__((unused))
 char *i64toa(int64_t in)
 {
 	i64toa_r(in, itoa_buffer);
@@ -439,7 +439,7 @@ char *i64toa(int64_t in)
 /* converts uint64_t <in> to a string using the static itoa_buffer and returns
  * the pointer to that string.
  */
-static inline __attribute__((unused))
+static __inline__ __attribute__((unused))
 char *u64toa(uint64_t in)
 {
 	u64toa_r(in, itoa_buffer);
diff --git a/tools/include/nolibc/string.h b/tools/include/nolibc/string.h
index fffdaf6ff467..0c2e06c7c477 100644
--- a/tools/include/nolibc/string.h
+++ b/tools/include/nolibc/string.h
@@ -90,7 +90,7 @@ void *memset(void *dst, int b, size_t len)
 
 	while (len--) {
 		/* prevent gcc from recognizing memset() here */
-		asm volatile("");
+		__asm__ volatile("");
 		*(p++) = b;
 	}
 	return dst;
@@ -139,7 +139,7 @@ size_t strlen(const char *str)
 	size_t len;
 
 	for (len = 0; str[len]; len++)
-		asm("");
+		__asm__("");
 	return len;
 }
 
diff --git a/tools/include/nolibc/sys.h b/tools/include/nolibc/sys.h
index 5d624dc63a42..856249a11890 100644
--- a/tools/include/nolibc/sys.h
+++ b/tools/include/nolibc/sys.h
@@ -12,15 +12,17 @@
 
 /* system includes */
 #include <asm/unistd.h>
-#include <asm/signal.h>  // for SIGCHLD
+#include <asm/signal.h>  /* for SIGCHLD */
 #include <asm/ioctls.h>
 #include <asm/mman.h>
 #include <linux/fs.h>
 #include <linux/loop.h>
 #include <linux/time.h>
 #include <linux/auxvec.h>
-#include <linux/fcntl.h> // for O_* and AT_*
-#include <linux/stat.h>  // for statx()
+#include <linux/fcntl.h> /* for O_* and AT_* */
+#include <linux/stat.h>  /* for statx() */
+#include <linux/reboot.h> /* for LINUX_REBOOT_* */
+#include <linux/prctl.h>
 
 #include "arch.h"
 #include "errno.h"
@@ -322,7 +324,7 @@ static __attribute__((noreturn,unused))
 void sys_exit(int status)
 {
 	my_syscall1(__NR_exit, status & 255);
-	while(1); // shut the "noreturn" warnings.
+	while(1); /* shut the "noreturn" warnings. */
 }
 
 static __attribute__((noreturn,unused))
@@ -336,6 +338,7 @@ void exit(int status)
  * pid_t fork(void);
  */
 
+#ifndef sys_fork
 static __attribute__((unused))
 pid_t sys_fork(void)
 {
@@ -351,6 +354,7 @@ pid_t sys_fork(void)
 #error Neither __NR_clone nor __NR_fork defined, cannot implement sys_fork()
 #endif
 }
+#endif
 
 static __attribute__((unused))
 pid_t fork(void)
@@ -858,7 +862,7 @@ int open(const char *path, int flags, ...)
 		va_list args;
 
 		va_start(args, flags);
-		mode = va_arg(args, mode_t);
+		mode = va_arg(args, int);
 		va_end(args);
 	}
 
@@ -873,6 +877,32 @@ int open(const char *path, int flags, ...)
 
 
 /*
+ * int prctl(int option, unsigned long arg2, unsigned long arg3,
+ *                       unsigned long arg4, unsigned long arg5);
+ */
+
+static __attribute__((unused))
+int sys_prctl(int option, unsigned long arg2, unsigned long arg3,
+		          unsigned long arg4, unsigned long arg5)
+{
+	return my_syscall5(__NR_prctl, option, arg2, arg3, arg4, arg5);
+}
+
+static __attribute__((unused))
+int prctl(int option, unsigned long arg2, unsigned long arg3,
+		      unsigned long arg4, unsigned long arg5)
+{
+	int ret = sys_prctl(option, arg2, arg3, arg4, arg5);
+
+	if (ret < 0) {
+		SET_ERRNO(-ret);
+		ret = -1;
+	}
+	return ret;
+}
+
+
+/*
  * int pivot_root(const char *new, const char *old);
  */
 
@@ -909,7 +939,7 @@ int sys_poll(struct pollfd *fds, int nfds, int timeout)
 		t.tv_sec  = timeout / 1000;
 		t.tv_nsec = (timeout % 1000) * 1000000;
 	}
-	return my_syscall4(__NR_ppoll, fds, nfds, (timeout >= 0) ? &t : NULL, NULL);
+	return my_syscall5(__NR_ppoll, fds, nfds, (timeout >= 0) ? &t : NULL, NULL, 0);
 #elif defined(__NR_poll)
 	return my_syscall3(__NR_poll, fds, nfds, timeout);
 #else
@@ -1131,23 +1161,26 @@ int sys_stat(const char *path, struct stat *buf)
 	long ret;
 
 	ret = sys_statx(AT_FDCWD, path, AT_NO_AUTOMOUNT, STATX_BASIC_STATS, &statx);
-	buf->st_dev     = ((statx.stx_dev_minor & 0xff)
-			  | (statx.stx_dev_major << 8)
-			  | ((statx.stx_dev_minor & ~0xff) << 12));
-	buf->st_ino     = statx.stx_ino;
-	buf->st_mode    = statx.stx_mode;
-	buf->st_nlink   = statx.stx_nlink;
-	buf->st_uid     = statx.stx_uid;
-	buf->st_gid     = statx.stx_gid;
-	buf->st_rdev    = ((statx.stx_rdev_minor & 0xff)
-			  | (statx.stx_rdev_major << 8)
-			  | ((statx.stx_rdev_minor & ~0xff) << 12));
-	buf->st_size    = statx.stx_size;
-	buf->st_blksize = statx.stx_blksize;
-	buf->st_blocks  = statx.stx_blocks;
-	buf->st_atime   = statx.stx_atime.tv_sec;
-	buf->st_mtime   = statx.stx_mtime.tv_sec;
-	buf->st_ctime   = statx.stx_ctime.tv_sec;
+	buf->st_dev          = ((statx.stx_dev_minor & 0xff)
+			       | (statx.stx_dev_major << 8)
+			       | ((statx.stx_dev_minor & ~0xff) << 12));
+	buf->st_ino          = statx.stx_ino;
+	buf->st_mode         = statx.stx_mode;
+	buf->st_nlink        = statx.stx_nlink;
+	buf->st_uid          = statx.stx_uid;
+	buf->st_gid          = statx.stx_gid;
+	buf->st_rdev         = ((statx.stx_rdev_minor & 0xff)
+			       | (statx.stx_rdev_major << 8)
+			       | ((statx.stx_rdev_minor & ~0xff) << 12));
+	buf->st_size         = statx.stx_size;
+	buf->st_blksize      = statx.stx_blksize;
+	buf->st_blocks       = statx.stx_blocks;
+	buf->st_atim.tv_sec  = statx.stx_atime.tv_sec;
+	buf->st_atim.tv_nsec = statx.stx_atime.tv_nsec;
+	buf->st_mtim.tv_sec  = statx.stx_mtime.tv_sec;
+	buf->st_mtim.tv_nsec = statx.stx_mtime.tv_nsec;
+	buf->st_ctim.tv_sec  = statx.stx_ctime.tv_sec;
+	buf->st_ctim.tv_nsec = statx.stx_ctime.tv_nsec;
 	return ret;
 }
 #else
@@ -1165,19 +1198,22 @@ int sys_stat(const char *path, struct stat *buf)
 #else
 #error Neither __NR_newfstatat nor __NR_stat defined, cannot implement sys_stat()
 #endif
-	buf->st_dev     = stat.st_dev;
-	buf->st_ino     = stat.st_ino;
-	buf->st_mode    = stat.st_mode;
-	buf->st_nlink   = stat.st_nlink;
-	buf->st_uid     = stat.st_uid;
-	buf->st_gid     = stat.st_gid;
-	buf->st_rdev    = stat.st_rdev;
-	buf->st_size    = stat.st_size;
-	buf->st_blksize = stat.st_blksize;
-	buf->st_blocks  = stat.st_blocks;
-	buf->st_atime   = stat.st_atime;
-	buf->st_mtime   = stat.st_mtime;
-	buf->st_ctime   = stat.st_ctime;
+	buf->st_dev          = stat.st_dev;
+	buf->st_ino          = stat.st_ino;
+	buf->st_mode         = stat.st_mode;
+	buf->st_nlink        = stat.st_nlink;
+	buf->st_uid          = stat.st_uid;
+	buf->st_gid          = stat.st_gid;
+	buf->st_rdev         = stat.st_rdev;
+	buf->st_size         = stat.st_size;
+	buf->st_blksize      = stat.st_blksize;
+	buf->st_blocks       = stat.st_blocks;
+	buf->st_atim.tv_sec  = stat.st_atime;
+	buf->st_atim.tv_nsec = stat.st_atime_nsec;
+	buf->st_mtim.tv_sec  = stat.st_mtime;
+	buf->st_mtim.tv_nsec = stat.st_mtime_nsec;
+	buf->st_ctim.tv_sec  = stat.st_ctime;
+	buf->st_ctim.tv_nsec = stat.st_ctime_nsec;
 	return ret;
 }
 #endif
@@ -1365,6 +1401,29 @@ ssize_t write(int fd, const void *buf, size_t count)
 	return ret;
 }
 
+
+/*
+ * int memfd_create(const char *name, unsigned int flags);
+ */
+
+static __attribute__((unused))
+int sys_memfd_create(const char *name, unsigned int flags)
+{
+	return my_syscall2(__NR_memfd_create, name, flags);
+}
+
+static __attribute__((unused))
+int memfd_create(const char *name, unsigned int flags)
+{
+	ssize_t ret = sys_memfd_create(name, flags);
+
+	if (ret < 0) {
+		SET_ERRNO(-ret);
+		ret = -1;
+	}
+	return ret;
+}
+
 /* make sure to include all global symbols */
 #include "nolibc.h"
 
diff --git a/tools/include/nolibc/types.h b/tools/include/nolibc/types.h
index aedd7d9e3f64..f96e28bff4ba 100644
--- a/tools/include/nolibc/types.h
+++ b/tools/include/nolibc/types.h
@@ -86,14 +86,6 @@
 #define SEEK_CUR       1
 #define SEEK_END       2
 
-/* cmd for reboot() */
-#define LINUX_REBOOT_MAGIC1         0xfee1dead
-#define LINUX_REBOOT_MAGIC2         0x28121969
-#define LINUX_REBOOT_CMD_HALT       0xcdef0123
-#define LINUX_REBOOT_CMD_POWER_OFF  0x4321fedc
-#define LINUX_REBOOT_CMD_RESTART    0x01234567
-#define LINUX_REBOOT_CMD_SW_SUSPEND 0xd000fce2
-
 /* Macros used on waitpid()'s return status */
 #define WEXITSTATUS(status) (((status) & 0xff00) >> 8)
 #define WIFEXITED(status)   (((status) & 0x7f) == 0)
@@ -206,9 +198,9 @@ struct stat {
 	off_t     st_size;    /* total size, in bytes */
 	blksize_t st_blksize; /* blocksize for file system I/O */
 	blkcnt_t  st_blocks;  /* number of 512B blocks allocated */
-	time_t    st_atime;   /* time of last access */
-	time_t    st_mtime;   /* time of last modification */
-	time_t    st_ctime;   /* time of last status change */
+	union { time_t st_atime; struct timespec st_atim; }; /* time of last access */
+	union { time_t st_mtime; struct timespec st_mtim; }; /* time of last modification */
+	union { time_t st_ctime; struct timespec st_ctim; }; /* time of last status change */
 };
 
 /* WARNING, it only deals with the 4096 first majors and 256 first minors */
diff --git a/tools/include/nolibc/unistd.h b/tools/include/nolibc/unistd.h
index ac7d53d986cd..0e832e10a0b2 100644
--- a/tools/include/nolibc/unistd.h
+++ b/tools/include/nolibc/unistd.h
@@ -56,6 +56,21 @@ int tcsetpgrp(int fd, pid_t pid)
 	return ioctl(fd, TIOCSPGRP, &pid);
 }
 
+#define _syscall(N, ...)                                                      \
+({                                                                            \
+	long _ret = my_syscall##N(__VA_ARGS__);                               \
+	if (_ret < 0) {                                                       \
+		SET_ERRNO(-_ret);                                             \
+		_ret = -1;                                                    \
+	}                                                                     \
+	_ret;                                                                 \
+})
+
+#define _syscall_narg(...) __syscall_narg(__VA_ARGS__, 6, 5, 4, 3, 2, 1, 0)
+#define __syscall_narg(_0, _1, _2, _3, _4, _5, _6, N, ...) N
+#define _syscall_n(N, ...) _syscall(N, __VA_ARGS__)
+#define syscall(...) _syscall_n(_syscall_narg(__VA_ARGS__), ##__VA_ARGS__)
+
 /* make sure to include all global symbols */
 #include "nolibc.h"
 
diff --git a/tools/include/uapi/asm-generic/bitsperlong.h b/tools/include/uapi/asm-generic/bitsperlong.h
index 23e6c416b85f..352cb81947b8 100644
--- a/tools/include/uapi/asm-generic/bitsperlong.h
+++ b/tools/include/uapi/asm-generic/bitsperlong.h
@@ -1,6 +1,18 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
 #ifndef _UAPI__ASM_GENERIC_BITS_PER_LONG
 #define _UAPI__ASM_GENERIC_BITS_PER_LONG
 
+#ifndef __BITS_PER_LONG
+/*
+ * In order to keep safe and avoid regression, only unify uapi
+ * bitsperlong.h for some archs which are using newer toolchains
+ * that have the definitions of __CHAR_BIT__ and __SIZEOF_LONG__.
+ * See the following link for more info:
+ * https://lore.kernel.org/linux-arch/b9624545-2c80-49a1-ac3c-39264a591f7b@app.fastmail.com/
+ */
+#if defined(__CHAR_BIT__) && defined(__SIZEOF_LONG__)
+#define __BITS_PER_LONG (__CHAR_BIT__ * __SIZEOF_LONG__)
+#else
 /*
  * There seems to be no way of detecting this automatically from user
  * space, so 64 bit architectures should override this in their
@@ -8,8 +20,8 @@
  * both 32 and 64 bit user space must not rely on CONFIG_64BIT
  * to decide it, but rather check a compiler provided macro.
  */
-#ifndef __BITS_PER_LONG
 #define __BITS_PER_LONG 32
 #endif
+#endif
 
 #endif /* _UAPI__ASM_GENERIC_BITS_PER_LONG */
diff --git a/tools/include/uapi/asm-generic/socket.h b/tools/include/uapi/asm-generic/socket.h
index 8756df13be50..54d9c8bf7c55 100644
--- a/tools/include/uapi/asm-generic/socket.h
+++ b/tools/include/uapi/asm-generic/socket.h
@@ -121,6 +121,9 @@
 
 #define SO_RCVMARK		75
 
+#define SO_PASSPIDFD		76
+#define SO_PEERPIDFD		77
+
 #if !defined(__KERNEL__)
 
 #if __BITS_PER_LONG == 64 || (defined(__x86_64__) && defined(__ILP32__))
diff --git a/tools/include/uapi/asm-generic/unistd.h b/tools/include/uapi/asm-generic/unistd.h
index 45fa180cc56a..dd7d8e10f16d 100644
--- a/tools/include/uapi/asm-generic/unistd.h
+++ b/tools/include/uapi/asm-generic/unistd.h
@@ -38,12 +38,12 @@ __SYSCALL(__NR_io_destroy, sys_io_destroy)
 __SC_COMP(__NR_io_submit, sys_io_submit, compat_sys_io_submit)
 #define __NR_io_cancel 3
 __SYSCALL(__NR_io_cancel, sys_io_cancel)
+
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_io_getevents 4
 __SC_3264(__NR_io_getevents, sys_io_getevents_time32, sys_io_getevents)
 #endif
 
-/* fs/xattr.c */
 #define __NR_setxattr 5
 __SYSCALL(__NR_setxattr, sys_setxattr)
 #define __NR_lsetxattr 6
@@ -68,58 +68,38 @@ __SYSCALL(__NR_removexattr, sys_removexattr)
 __SYSCALL(__NR_lremovexattr, sys_lremovexattr)
 #define __NR_fremovexattr 16
 __SYSCALL(__NR_fremovexattr, sys_fremovexattr)
-
-/* fs/dcache.c */
 #define __NR_getcwd 17
 __SYSCALL(__NR_getcwd, sys_getcwd)
-
-/* fs/cookies.c */
 #define __NR_lookup_dcookie 18
 __SC_COMP(__NR_lookup_dcookie, sys_lookup_dcookie, compat_sys_lookup_dcookie)
-
-/* fs/eventfd.c */
 #define __NR_eventfd2 19
 __SYSCALL(__NR_eventfd2, sys_eventfd2)
-
-/* fs/eventpoll.c */
 #define __NR_epoll_create1 20
 __SYSCALL(__NR_epoll_create1, sys_epoll_create1)
 #define __NR_epoll_ctl 21
 __SYSCALL(__NR_epoll_ctl, sys_epoll_ctl)
 #define __NR_epoll_pwait 22
 __SC_COMP(__NR_epoll_pwait, sys_epoll_pwait, compat_sys_epoll_pwait)
-
-/* fs/fcntl.c */
 #define __NR_dup 23
 __SYSCALL(__NR_dup, sys_dup)
 #define __NR_dup3 24
 __SYSCALL(__NR_dup3, sys_dup3)
 #define __NR3264_fcntl 25
 __SC_COMP_3264(__NR3264_fcntl, sys_fcntl64, sys_fcntl, compat_sys_fcntl64)
-
-/* fs/inotify_user.c */
 #define __NR_inotify_init1 26
 __SYSCALL(__NR_inotify_init1, sys_inotify_init1)
 #define __NR_inotify_add_watch 27
 __SYSCALL(__NR_inotify_add_watch, sys_inotify_add_watch)
 #define __NR_inotify_rm_watch 28
 __SYSCALL(__NR_inotify_rm_watch, sys_inotify_rm_watch)
-
-/* fs/ioctl.c */
 #define __NR_ioctl 29
 __SC_COMP(__NR_ioctl, sys_ioctl, compat_sys_ioctl)
-
-/* fs/ioprio.c */
 #define __NR_ioprio_set 30
 __SYSCALL(__NR_ioprio_set, sys_ioprio_set)
 #define __NR_ioprio_get 31
 __SYSCALL(__NR_ioprio_get, sys_ioprio_get)
-
-/* fs/locks.c */
 #define __NR_flock 32
 __SYSCALL(__NR_flock, sys_flock)
-
-/* fs/namei.c */
 #define __NR_mknodat 33
 __SYSCALL(__NR_mknodat, sys_mknodat)
 #define __NR_mkdirat 34
@@ -130,25 +110,21 @@ __SYSCALL(__NR_unlinkat, sys_unlinkat)
 __SYSCALL(__NR_symlinkat, sys_symlinkat)
 #define __NR_linkat 37
 __SYSCALL(__NR_linkat, sys_linkat)
+
 #ifdef __ARCH_WANT_RENAMEAT
 /* renameat is superseded with flags by renameat2 */
 #define __NR_renameat 38
 __SYSCALL(__NR_renameat, sys_renameat)
 #endif /* __ARCH_WANT_RENAMEAT */
 
-/* fs/namespace.c */
 #define __NR_umount2 39
 __SYSCALL(__NR_umount2, sys_umount)
 #define __NR_mount 40
 __SYSCALL(__NR_mount, sys_mount)
 #define __NR_pivot_root 41
 __SYSCALL(__NR_pivot_root, sys_pivot_root)
-
-/* fs/nfsctl.c */
 #define __NR_nfsservctl 42
 __SYSCALL(__NR_nfsservctl, sys_ni_syscall)
-
-/* fs/open.c */
 #define __NR3264_statfs 43
 __SC_COMP_3264(__NR3264_statfs, sys_statfs64, sys_statfs, \
 	       compat_sys_statfs64)
@@ -161,7 +137,6 @@ __SC_COMP_3264(__NR3264_truncate, sys_truncate64, sys_truncate, \
 #define __NR3264_ftruncate 46
 __SC_COMP_3264(__NR3264_ftruncate, sys_ftruncate64, sys_ftruncate, \
 	       compat_sys_ftruncate64)
-
 #define __NR_fallocate 47
 __SC_COMP(__NR_fallocate, sys_fallocate, compat_sys_fallocate)
 #define __NR_faccessat 48
@@ -186,20 +161,12 @@ __SYSCALL(__NR_openat, sys_openat)
 __SYSCALL(__NR_close, sys_close)
 #define __NR_vhangup 58
 __SYSCALL(__NR_vhangup, sys_vhangup)
-
-/* fs/pipe.c */
 #define __NR_pipe2 59
 __SYSCALL(__NR_pipe2, sys_pipe2)
-
-/* fs/quota.c */
 #define __NR_quotactl 60
 __SYSCALL(__NR_quotactl, sys_quotactl)
-
-/* fs/readdir.c */
 #define __NR_getdents64 61
 __SYSCALL(__NR_getdents64, sys_getdents64)
-
-/* fs/read_write.c */
 #define __NR3264_lseek 62
 __SC_3264(__NR3264_lseek, sys_llseek, sys_lseek)
 #define __NR_read 63
@@ -218,12 +185,9 @@ __SC_COMP(__NR_pwrite64, sys_pwrite64, compat_sys_pwrite64)
 __SC_COMP(__NR_preadv, sys_preadv, compat_sys_preadv)
 #define __NR_pwritev 70
 __SC_COMP(__NR_pwritev, sys_pwritev, compat_sys_pwritev)
-
-/* fs/sendfile.c */
 #define __NR3264_sendfile 71
 __SYSCALL(__NR3264_sendfile, sys_sendfile64)
 
-/* fs/select.c */
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_pselect6 72
 __SC_COMP_3264(__NR_pselect6, sys_pselect6_time32, sys_pselect6, compat_sys_pselect6_time32)
@@ -231,21 +195,17 @@ __SC_COMP_3264(__NR_pselect6, sys_pselect6_time32, sys_pselect6, compat_sys_psel
 __SC_COMP_3264(__NR_ppoll, sys_ppoll_time32, sys_ppoll, compat_sys_ppoll_time32)
 #endif
 
-/* fs/signalfd.c */
 #define __NR_signalfd4 74
 __SC_COMP(__NR_signalfd4, sys_signalfd4, compat_sys_signalfd4)
-
-/* fs/splice.c */
 #define __NR_vmsplice 75
 __SYSCALL(__NR_vmsplice, sys_vmsplice)
 #define __NR_splice 76
 __SYSCALL(__NR_splice, sys_splice)
 #define __NR_tee 77
 __SYSCALL(__NR_tee, sys_tee)
-
-/* fs/stat.c */
 #define __NR_readlinkat 78
 __SYSCALL(__NR_readlinkat, sys_readlinkat)
+
 #if defined(__ARCH_WANT_NEW_STAT) || defined(__ARCH_WANT_STAT64)
 #define __NR3264_fstatat 79
 __SC_3264(__NR3264_fstatat, sys_fstatat64, sys_newfstatat)
@@ -253,13 +213,13 @@ __SC_3264(__NR3264_fstatat, sys_fstatat64, sys_newfstatat)
 __SC_3264(__NR3264_fstat, sys_fstat64, sys_newfstat)
 #endif
 
-/* fs/sync.c */
 #define __NR_sync 81
 __SYSCALL(__NR_sync, sys_sync)
 #define __NR_fsync 82
 __SYSCALL(__NR_fsync, sys_fsync)
 #define __NR_fdatasync 83
 __SYSCALL(__NR_fdatasync, sys_fdatasync)
+
 #ifdef __ARCH_WANT_SYNC_FILE_RANGE2
 #define __NR_sync_file_range2 84
 __SC_COMP(__NR_sync_file_range2, sys_sync_file_range2, \
@@ -270,9 +230,9 @@ __SC_COMP(__NR_sync_file_range, sys_sync_file_range, \
 	  compat_sys_sync_file_range)
 #endif
 
-/* fs/timerfd.c */
 #define __NR_timerfd_create 85
 __SYSCALL(__NR_timerfd_create, sys_timerfd_create)
+
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_timerfd_settime 86
 __SC_3264(__NR_timerfd_settime, sys_timerfd_settime32, \
@@ -282,45 +242,35 @@ __SC_3264(__NR_timerfd_gettime, sys_timerfd_gettime32, \
 	  sys_timerfd_gettime)
 #endif
 
-/* fs/utimes.c */
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_utimensat 88
 __SC_3264(__NR_utimensat, sys_utimensat_time32, sys_utimensat)
 #endif
 
-/* kernel/acct.c */
 #define __NR_acct 89
 __SYSCALL(__NR_acct, sys_acct)
-
-/* kernel/capability.c */
 #define __NR_capget 90
 __SYSCALL(__NR_capget, sys_capget)
 #define __NR_capset 91
 __SYSCALL(__NR_capset, sys_capset)
-
-/* kernel/exec_domain.c */
 #define __NR_personality 92
 __SYSCALL(__NR_personality, sys_personality)
-
-/* kernel/exit.c */
 #define __NR_exit 93
 __SYSCALL(__NR_exit, sys_exit)
 #define __NR_exit_group 94
 __SYSCALL(__NR_exit_group, sys_exit_group)
 #define __NR_waitid 95
 __SC_COMP(__NR_waitid, sys_waitid, compat_sys_waitid)
-
-/* kernel/fork.c */
 #define __NR_set_tid_address 96
 __SYSCALL(__NR_set_tid_address, sys_set_tid_address)
 #define __NR_unshare 97
 __SYSCALL(__NR_unshare, sys_unshare)
 
-/* kernel/futex.c */
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_futex 98
 __SC_3264(__NR_futex, sys_futex_time32, sys_futex)
 #endif
+
 #define __NR_set_robust_list 99
 __SC_COMP(__NR_set_robust_list, sys_set_robust_list, \
 	  compat_sys_set_robust_list)
@@ -328,43 +278,40 @@ __SC_COMP(__NR_set_robust_list, sys_set_robust_list, \
 __SC_COMP(__NR_get_robust_list, sys_get_robust_list, \
 	  compat_sys_get_robust_list)
 
-/* kernel/hrtimer.c */
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_nanosleep 101
 __SC_3264(__NR_nanosleep, sys_nanosleep_time32, sys_nanosleep)
 #endif
 
-/* kernel/itimer.c */
 #define __NR_getitimer 102
 __SC_COMP(__NR_getitimer, sys_getitimer, compat_sys_getitimer)
 #define __NR_setitimer 103
 __SC_COMP(__NR_setitimer, sys_setitimer, compat_sys_setitimer)
-
-/* kernel/kexec.c */
 #define __NR_kexec_load 104
 __SC_COMP(__NR_kexec_load, sys_kexec_load, compat_sys_kexec_load)
-
-/* kernel/module.c */
 #define __NR_init_module 105
 __SYSCALL(__NR_init_module, sys_init_module)
 #define __NR_delete_module 106
 __SYSCALL(__NR_delete_module, sys_delete_module)
-
-/* kernel/posix-timers.c */
 #define __NR_timer_create 107
 __SC_COMP(__NR_timer_create, sys_timer_create, compat_sys_timer_create)
+
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_timer_gettime 108
 __SC_3264(__NR_timer_gettime, sys_timer_gettime32, sys_timer_gettime)
 #endif
+
 #define __NR_timer_getoverrun 109
 __SYSCALL(__NR_timer_getoverrun, sys_timer_getoverrun)
+
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_timer_settime 110
 __SC_3264(__NR_timer_settime, sys_timer_settime32, sys_timer_settime)
 #endif
+
 #define __NR_timer_delete 111
 __SYSCALL(__NR_timer_delete, sys_timer_delete)
+
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_clock_settime 112
 __SC_3264(__NR_clock_settime, sys_clock_settime32, sys_clock_settime)
@@ -377,15 +324,10 @@ __SC_3264(__NR_clock_nanosleep, sys_clock_nanosleep_time32, \
 	  sys_clock_nanosleep)
 #endif
 
-/* kernel/printk.c */
 #define __NR_syslog 116
 __SYSCALL(__NR_syslog, sys_syslog)
-
-/* kernel/ptrace.c */
 #define __NR_ptrace 117
 __SC_COMP(__NR_ptrace, sys_ptrace, compat_sys_ptrace)
-
-/* kernel/sched/core.c */
 #define __NR_sched_setparam 118
 __SYSCALL(__NR_sched_setparam, sys_sched_setparam)
 #define __NR_sched_setscheduler 119
@@ -406,13 +348,13 @@ __SYSCALL(__NR_sched_yield, sys_sched_yield)
 __SYSCALL(__NR_sched_get_priority_max, sys_sched_get_priority_max)
 #define __NR_sched_get_priority_min 126
 __SYSCALL(__NR_sched_get_priority_min, sys_sched_get_priority_min)
+
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_sched_rr_get_interval 127
 __SC_3264(__NR_sched_rr_get_interval, sys_sched_rr_get_interval_time32, \
 	  sys_sched_rr_get_interval)
 #endif
 
-/* kernel/signal.c */
 #define __NR_restart_syscall 128
 __SYSCALL(__NR_restart_syscall, sys_restart_syscall)
 #define __NR_kill 129
@@ -431,18 +373,18 @@ __SC_COMP(__NR_rt_sigaction, sys_rt_sigaction, compat_sys_rt_sigaction)
 __SC_COMP(__NR_rt_sigprocmask, sys_rt_sigprocmask, compat_sys_rt_sigprocmask)
 #define __NR_rt_sigpending 136
 __SC_COMP(__NR_rt_sigpending, sys_rt_sigpending, compat_sys_rt_sigpending)
+
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_rt_sigtimedwait 137
 __SC_COMP_3264(__NR_rt_sigtimedwait, sys_rt_sigtimedwait_time32, \
 	  sys_rt_sigtimedwait, compat_sys_rt_sigtimedwait_time32)
 #endif
+
 #define __NR_rt_sigqueueinfo 138
 __SC_COMP(__NR_rt_sigqueueinfo, sys_rt_sigqueueinfo, \
 	  compat_sys_rt_sigqueueinfo)
 #define __NR_rt_sigreturn 139
 __SC_COMP(__NR_rt_sigreturn, sys_rt_sigreturn, compat_sys_rt_sigreturn)
-
-/* kernel/sys.c */
 #define __NR_setpriority 140
 __SYSCALL(__NR_setpriority, sys_setpriority)
 #define __NR_getpriority 141
@@ -507,7 +449,6 @@ __SYSCALL(__NR_prctl, sys_prctl)
 #define __NR_getcpu 168
 __SYSCALL(__NR_getcpu, sys_getcpu)
 
-/* kernel/time.c */
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_gettimeofday 169
 __SC_COMP(__NR_gettimeofday, sys_gettimeofday, compat_sys_gettimeofday)
@@ -517,7 +458,6 @@ __SC_COMP(__NR_settimeofday, sys_settimeofday, compat_sys_settimeofday)
 __SC_3264(__NR_adjtimex, sys_adjtimex_time32, sys_adjtimex)
 #endif
 
-/* kernel/sys.c */
 #define __NR_getpid 172
 __SYSCALL(__NR_getpid, sys_getpid)
 #define __NR_getppid 173
@@ -534,12 +474,11 @@ __SYSCALL(__NR_getegid, sys_getegid)
 __SYSCALL(__NR_gettid, sys_gettid)
 #define __NR_sysinfo 179
 __SC_COMP(__NR_sysinfo, sys_sysinfo, compat_sys_sysinfo)
-
-/* ipc/mqueue.c */
 #define __NR_mq_open 180
 __SC_COMP(__NR_mq_open, sys_mq_open, compat_sys_mq_open)
 #define __NR_mq_unlink 181
 __SYSCALL(__NR_mq_unlink, sys_mq_unlink)
+
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_mq_timedsend 182
 __SC_3264(__NR_mq_timedsend, sys_mq_timedsend_time32, sys_mq_timedsend)
@@ -547,12 +486,11 @@ __SC_3264(__NR_mq_timedsend, sys_mq_timedsend_time32, sys_mq_timedsend)
 __SC_3264(__NR_mq_timedreceive, sys_mq_timedreceive_time32, \
 	  sys_mq_timedreceive)
 #endif
+
 #define __NR_mq_notify 184
 __SC_COMP(__NR_mq_notify, sys_mq_notify, compat_sys_mq_notify)
 #define __NR_mq_getsetattr 185
 __SC_COMP(__NR_mq_getsetattr, sys_mq_getsetattr, compat_sys_mq_getsetattr)
-
-/* ipc/msg.c */
 #define __NR_msgget 186
 __SYSCALL(__NR_msgget, sys_msgget)
 #define __NR_msgctl 187
@@ -561,20 +499,18 @@ __SC_COMP(__NR_msgctl, sys_msgctl, compat_sys_msgctl)
 __SC_COMP(__NR_msgrcv, sys_msgrcv, compat_sys_msgrcv)
 #define __NR_msgsnd 189
 __SC_COMP(__NR_msgsnd, sys_msgsnd, compat_sys_msgsnd)
-
-/* ipc/sem.c */
 #define __NR_semget 190
 __SYSCALL(__NR_semget, sys_semget)
 #define __NR_semctl 191
 __SC_COMP(__NR_semctl, sys_semctl, compat_sys_semctl)
+
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_semtimedop 192
 __SC_3264(__NR_semtimedop, sys_semtimedop_time32, sys_semtimedop)
 #endif
+
 #define __NR_semop 193
 __SYSCALL(__NR_semop, sys_semop)
-
-/* ipc/shm.c */
 #define __NR_shmget 194
 __SYSCALL(__NR_shmget, sys_shmget)
 #define __NR_shmctl 195
@@ -583,8 +519,6 @@ __SC_COMP(__NR_shmctl, sys_shmctl, compat_sys_shmctl)
 __SC_COMP(__NR_shmat, sys_shmat, compat_sys_shmat)
 #define __NR_shmdt 197
 __SYSCALL(__NR_shmdt, sys_shmdt)
-
-/* net/socket.c */
 #define __NR_socket 198
 __SYSCALL(__NR_socket, sys_socket)
 #define __NR_socketpair 199
@@ -615,40 +549,30 @@ __SYSCALL(__NR_shutdown, sys_shutdown)
 __SC_COMP(__NR_sendmsg, sys_sendmsg, compat_sys_sendmsg)
 #define __NR_recvmsg 212
 __SC_COMP(__NR_recvmsg, sys_recvmsg, compat_sys_recvmsg)
-
-/* mm/filemap.c */
 #define __NR_readahead 213
 __SC_COMP(__NR_readahead, sys_readahead, compat_sys_readahead)
-
-/* mm/nommu.c, also with MMU */
 #define __NR_brk 214
 __SYSCALL(__NR_brk, sys_brk)
 #define __NR_munmap 215
 __SYSCALL(__NR_munmap, sys_munmap)
 #define __NR_mremap 216
 __SYSCALL(__NR_mremap, sys_mremap)
-
-/* security/keys/keyctl.c */
 #define __NR_add_key 217
 __SYSCALL(__NR_add_key, sys_add_key)
 #define __NR_request_key 218
 __SYSCALL(__NR_request_key, sys_request_key)
 #define __NR_keyctl 219
 __SC_COMP(__NR_keyctl, sys_keyctl, compat_sys_keyctl)
-
-/* arch/example/kernel/sys_example.c */
 #define __NR_clone 220
 __SYSCALL(__NR_clone, sys_clone)
 #define __NR_execve 221
 __SC_COMP(__NR_execve, sys_execve, compat_sys_execve)
-
 #define __NR3264_mmap 222
 __SC_3264(__NR3264_mmap, sys_mmap2, sys_mmap)
-/* mm/fadvise.c */
 #define __NR3264_fadvise64 223
 __SC_COMP(__NR3264_fadvise64, sys_fadvise64_64, compat_sys_fadvise64_64)
 
-/* mm/, CONFIG_MMU only */
+/* CONFIG_MMU only */
 #ifndef __ARCH_NOMMU
 #define __NR_swapon 224
 __SYSCALL(__NR_swapon, sys_swapon)
@@ -691,6 +615,7 @@ __SC_COMP(__NR_rt_tgsigqueueinfo, sys_rt_tgsigqueueinfo, \
 __SYSCALL(__NR_perf_event_open, sys_perf_event_open)
 #define __NR_accept4 242
 __SYSCALL(__NR_accept4, sys_accept4)
+
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_recvmmsg 243
 __SC_COMP_3264(__NR_recvmmsg, sys_recvmmsg_time32, sys_recvmmsg, compat_sys_recvmmsg_time32)
@@ -706,6 +631,7 @@ __SC_COMP_3264(__NR_recvmmsg, sys_recvmmsg_time32, sys_recvmmsg, compat_sys_recv
 #define __NR_wait4 260
 __SC_COMP(__NR_wait4, sys_wait4, compat_sys_wait4)
 #endif
+
 #define __NR_prlimit64 261
 __SYSCALL(__NR_prlimit64, sys_prlimit64)
 #define __NR_fanotify_init 262
@@ -716,10 +642,12 @@ __SYSCALL(__NR_fanotify_mark, sys_fanotify_mark)
 __SYSCALL(__NR_name_to_handle_at, sys_name_to_handle_at)
 #define __NR_open_by_handle_at         265
 __SYSCALL(__NR_open_by_handle_at, sys_open_by_handle_at)
+
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_clock_adjtime 266
 __SC_3264(__NR_clock_adjtime, sys_clock_adjtime32, sys_clock_adjtime)
 #endif
+
 #define __NR_syncfs 267
 __SYSCALL(__NR_syncfs, sys_syncfs)
 #define __NR_setns 268
@@ -770,15 +698,19 @@ __SYSCALL(__NR_pkey_alloc,    sys_pkey_alloc)
 __SYSCALL(__NR_pkey_free,     sys_pkey_free)
 #define __NR_statx 291
 __SYSCALL(__NR_statx,     sys_statx)
+
 #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
 #define __NR_io_pgetevents 292
 __SC_COMP_3264(__NR_io_pgetevents, sys_io_pgetevents_time32, sys_io_pgetevents, compat_sys_io_pgetevents)
 #endif
+
 #define __NR_rseq 293
 __SYSCALL(__NR_rseq, sys_rseq)
 #define __NR_kexec_file_load 294
 __SYSCALL(__NR_kexec_file_load,     sys_kexec_file_load)
+
 /* 295 through 402 are unassigned to sync up with generic numbers, don't use */
+
 #if defined(__SYSCALL_COMPAT) || __BITS_PER_LONG == 32
 #define __NR_clock_gettime64 403
 __SYSCALL(__NR_clock_gettime64, sys_clock_gettime)
@@ -844,13 +776,14 @@ __SYSCALL(__NR_fsmount, sys_fsmount)
 __SYSCALL(__NR_fspick, sys_fspick)
 #define __NR_pidfd_open 434
 __SYSCALL(__NR_pidfd_open, sys_pidfd_open)
+
 #ifdef __ARCH_WANT_SYS_CLONE3
 #define __NR_clone3 435
 __SYSCALL(__NR_clone3, sys_clone3)
 #endif
+
 #define __NR_close_range 436
 __SYSCALL(__NR_close_range, sys_close_range)
-
 #define __NR_openat2 437
 __SYSCALL(__NR_openat2, sys_openat2)
 #define __NR_pidfd_getfd 438
@@ -865,7 +798,6 @@ __SC_COMP(__NR_epoll_pwait2, sys_epoll_pwait2, compat_sys_epoll_pwait2)
 __SYSCALL(__NR_mount_setattr, sys_mount_setattr)
 #define __NR_quotactl_fd 443
 __SYSCALL(__NR_quotactl_fd, sys_quotactl_fd)
-
 #define __NR_landlock_create_ruleset 444
 __SYSCALL(__NR_landlock_create_ruleset, sys_landlock_create_ruleset)
 #define __NR_landlock_add_rule 445
@@ -877,12 +809,11 @@ __SYSCALL(__NR_landlock_restrict_self, sys_landlock_restrict_self)
 #define __NR_memfd_secret 447
 __SYSCALL(__NR_memfd_secret, sys_memfd_secret)
 #endif
+
 #define __NR_process_mrelease 448
 __SYSCALL(__NR_process_mrelease, sys_process_mrelease)
-
 #define __NR_futex_waitv 449
 __SYSCALL(__NR_futex_waitv, sys_futex_waitv)
-
 #define __NR_set_mempolicy_home_node 450
 __SYSCALL(__NR_set_mempolicy_home_node, sys_set_mempolicy_home_node)
 
diff --git a/tools/include/uapi/asm/bitsperlong.h b/tools/include/uapi/asm/bitsperlong.h
index da5206517158..c65267afc341 100644
--- a/tools/include/uapi/asm/bitsperlong.h
+++ b/tools/include/uapi/asm/bitsperlong.h
@@ -1,8 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 */
 #if defined(__i386__) || defined(__x86_64__)
 #include "../../../arch/x86/include/uapi/asm/bitsperlong.h"
-#elif defined(__aarch64__)
-#include "../../../arch/arm64/include/uapi/asm/bitsperlong.h"
 #elif defined(__powerpc__)
 #include "../../../arch/powerpc/include/uapi/asm/bitsperlong.h"
 #elif defined(__s390__)
@@ -13,12 +11,8 @@
 #include "../../../arch/mips/include/uapi/asm/bitsperlong.h"
 #elif defined(__ia64__)
 #include "../../../arch/ia64/include/uapi/asm/bitsperlong.h"
-#elif defined(__riscv)
-#include "../../../arch/riscv/include/uapi/asm/bitsperlong.h"
 #elif defined(__alpha__)
 #include "../../../arch/alpha/include/uapi/asm/bitsperlong.h"
-#elif defined(__loongarch__)
-#include "../../../arch/loongarch/include/uapi/asm/bitsperlong.h"
 #else
 #include <asm-generic/bitsperlong.h>
 #endif
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 1bb11a6ee667..60a9d59beeab 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -1035,6 +1035,7 @@ enum bpf_attach_type {
 	BPF_TRACE_KPROBE_MULTI,
 	BPF_LSM_CGROUP,
 	BPF_STRUCT_OPS,
+	BPF_NETFILTER,
 	__MAX_BPF_ATTACH_TYPE
 };
 
@@ -1272,6 +1273,9 @@ enum {
 
 /* Create a map that will be registered/unregesitered by the backed bpf_link */
 	BPF_F_LINK		= (1U << 13),
+
+/* Get path from provided FD in BPF_OBJ_PIN/BPF_OBJ_GET commands */
+	BPF_F_PATH_FD		= (1U << 14),
 };
 
 /* Flags for BPF_PROG_QUERY. */
@@ -1420,6 +1424,13 @@ union bpf_attr {
 		__aligned_u64	pathname;
 		__u32		bpf_fd;
 		__u32		file_flags;
+		/* Same as dirfd in openat() syscall; see openat(2)
+		 * manpage for details of path FD and pathname semantics;
+		 * path_fd should accompanied by BPF_F_PATH_FD flag set in
+		 * file_flags field, otherwise it should be set to zero;
+		 * if BPF_F_PATH_FD flag is not set, AT_FDCWD is assumed.
+		 */
+		__s32		path_fd;
 	};
 
 	struct { /* anonymous struct used by BPF_PROG_ATTACH/DETACH commands */
@@ -3167,6 +3178,10 @@ union bpf_attr {
  *		**BPF_FIB_LOOKUP_DIRECT**
  *			Do a direct table lookup vs full lookup using FIB
  *			rules.
+ *		**BPF_FIB_LOOKUP_TBID**
+ *			Used with BPF_FIB_LOOKUP_DIRECT.
+ *			Use the routing table ID present in *params*->tbid
+ *			for the fib lookup.
  *		**BPF_FIB_LOOKUP_OUTPUT**
  *			Perform lookup from an egress perspective (default is
  *			ingress).
@@ -6821,6 +6836,7 @@ enum {
 	BPF_FIB_LOOKUP_DIRECT  = (1U << 0),
 	BPF_FIB_LOOKUP_OUTPUT  = (1U << 1),
 	BPF_FIB_LOOKUP_SKIP_NEIGH = (1U << 2),
+	BPF_FIB_LOOKUP_TBID    = (1U << 3),
 };
 
 enum {
@@ -6881,9 +6897,19 @@ struct bpf_fib_lookup {
 		__u32		ipv6_dst[4];  /* in6_addr; network order */
 	};
 
-	/* output */
-	__be16	h_vlan_proto;
-	__be16	h_vlan_TCI;
+	union {
+		struct {
+			/* output */
+			__be16	h_vlan_proto;
+			__be16	h_vlan_TCI;
+		};
+		/* input: when accompanied with the
+		 * 'BPF_FIB_LOOKUP_DIRECT | BPF_FIB_LOOKUP_TBID` flags, a
+		 * specific routing table to use for the fib lookup.
+		 */
+		__u32	tbid;
+	};
+
 	__u8	smac[6];     /* ETH_ALEN */
 	__u8	dmac[6];     /* ETH_ALEN */
 };
diff --git a/tools/include/uapi/linux/in.h b/tools/include/uapi/linux/in.h
index 4b7f2df66b99..e682ab628dfa 100644
--- a/tools/include/uapi/linux/in.h
+++ b/tools/include/uapi/linux/in.h
@@ -163,6 +163,7 @@ struct in_addr {
 #define IP_MULTICAST_ALL		49
 #define IP_UNICAST_IF			50
 #define IP_LOCAL_PORT_RANGE		51
+#define IP_PROTOCOL			52
 
 #define MCAST_EXCLUDE	0
 #define MCAST_INCLUDE	1