powerpc: Provide syscall wrapper

Implement syscall wrapper as per s390, x86, arm64. When enabled
cause handlers to accept parameters from a stack frame rather than
from user scratch register state. This allows for user registers to be
safely cleared in order to reduce caller influence on speculation
within syscall routine. The wrapper is a macro that emits syscall
handler symbols that call into the target handler, obtaining its
parameters from a struct pt_regs on the stack.

As registers are already saved to the stack prior to calling
system_call_exception, it appears that this function is executed more
efficiently with the new stack-pointer convention than with parameters
passed by registers, avoiding the allocation of a stack frame for this
method. On a 32-bit system, we see >20% performance increases on the
null_syscall microbenchmark, and on a Power 8 the performance gains
amortise the cost of clearing and restoring registers which is
implemented at the end of this series, seeing final result of ~5.6%
performance improvement on null_syscall.

Syscalls are wrapped in this fashion on all platforms except for the
Cell processor as this commit does not provide SPU support. This can be
quickly fixed in a successive patch, but requires spu_sys_callback to
allocate a pt_regs structure to satisfy the wrapped calling convention.
Co-developed-by: Andrew Donnellan <ajd@linux.ibm.com>
Signed-off-by: Andrew Donnellan <ajd@linux.ibm.com>
Signed-off-by: Rohan McLure <rmclure@linux.ibm.com>
Reviewed-by: Nicholas Piggin <npiggin@gmai.com>
[mpe: Make incompatible with COMPAT to retain clearing of high bits of args]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20220921065605.1051927-22-rmclure@linux.ibm.com

arch/powerpc/Kconfig

@@ -137,6 +137,7 @@ config PPC
 	select ARCH_HAS_STRICT_KERNEL_RWX	if (PPC_BOOK3S || PPC_8xx || 40x) && !HIBERNATION
 	select ARCH_HAS_STRICT_KERNEL_RWX	if PPC_85xx && !HIBERNATION && !RANDOMIZE_BASE
 	select ARCH_HAS_STRICT_MODULE_RWX	if ARCH_HAS_STRICT_KERNEL_RWX
+	select ARCH_HAS_SYSCALL_WRAPPER		if !SPU_BASE && !COMPAT
 	select ARCH_HAS_TICK_BROADCAST		if GENERIC_CLOCKEVENTS_BROADCAST
 	select ARCH_HAS_UACCESS_FLUSHCACHE
 	select ARCH_HAS_UBSAN_SANITIZE_ALL

arch/powerpc/include/asm/syscall.h

@@ -14,8 +14,12 @@
 #include <linux/sched.h>
 #include <linux/thread_info.h>
 
+#ifdef CONFIG_ARCH_HAS_SYSCALL_WRAPPER
+typedef long (*syscall_fn)(const struct pt_regs *);
+#else
 typedef long (*syscall_fn)(unsigned long, unsigned long, unsigned long,
 			   unsigned long, unsigned long, unsigned long);
+#endif
 
 /* ftrace syscalls requires exporting the sys_call_table */
 extern const syscall_fn sys_call_table[];

arch/powerpc/include/asm/syscall_wrapper.h

+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * syscall_wrapper.h - powerpc specific wrappers to syscall definitions
+ *
+ * Based on arch/{x86,arm64}/include/asm/syscall_wrapper.h
+ */
+
+#ifndef __ASM_POWERPC_SYSCALL_WRAPPER_H
+#define __ASM_POWERPC_SYSCALL_WRAPPER_H
+
+struct pt_regs;
+
+#define SC_POWERPC_REGS_TO_ARGS(x, ...)				\
+	__MAP(x,__SC_ARGS					\
+	      ,,regs->gpr[3],,regs->gpr[4],,regs->gpr[5]	\
+	      ,,regs->gpr[6],,regs->gpr[7],,regs->gpr[8])
+
+#define __SYSCALL_DEFINEx(x, name, ...)						\
+	long __powerpc_sys##name(const struct pt_regs *regs);			\
+	ALLOW_ERROR_INJECTION(__powerpc_sys##name, ERRNO);			\
+	static long __se_sys##name(__MAP(x,__SC_LONG,__VA_ARGS__));		\
+	static inline long __do_sys##name(__MAP(x,__SC_DECL,__VA_ARGS__));	\
+	long __powerpc_sys##name(const struct pt_regs *regs)			\
+	{									\
+		return __se_sys##name(SC_POWERPC_REGS_TO_ARGS(x,__VA_ARGS__));	\
+	}									\
+	static long __se_sys##name(__MAP(x,__SC_LONG,__VA_ARGS__))		\
+	{									\
+		long ret = __do_sys##name(__MAP(x,__SC_CAST,__VA_ARGS__));	\
+		__MAP(x,__SC_TEST,__VA_ARGS__);					\
+		__PROTECT(x, ret,__MAP(x,__SC_ARGS,__VA_ARGS__));		\
+		return ret;							\
+	}									\
+	static inline long __do_sys##name(__MAP(x,__SC_DECL,__VA_ARGS__))
+
+#define SYSCALL_DEFINE0(sname)							\
+	SYSCALL_METADATA(_##sname, 0);						\
+	long __powerpc_sys_##sname(const struct pt_regs *__unused);		\
+	ALLOW_ERROR_INJECTION(__powerpc_sys_##sname, ERRNO);			\
+	long __powerpc_sys_##sname(const struct pt_regs *__unused)
+
+#define COND_SYSCALL(name)							\
+	long __powerpc_sys_##name(const struct pt_regs *regs);			\
+	long __weak __powerpc_sys_##name(const struct pt_regs *regs)		\
+	{									\
+		return sys_ni_syscall();					\
+	}
+
+#define SYS_NI(name) SYSCALL_ALIAS(__powerpc_sys_##name, sys_ni_posix_timers);
+
+#endif // __ASM_POWERPC_SYSCALL_WRAPPER_H

arch/powerpc/include/asm/syscalls.h

@@ -15,6 +15,12 @@
 #include <asm/unistd.h>
 #include <asm/ucontext.h>
 
+#ifndef CONFIG_ARCH_HAS_SYSCALL_WRAPPER
+long sys_ni_syscall(void);
+#else
+long sys_ni_syscall(const struct pt_regs *regs);
+#endif
+
 struct rtas_args;
 
 /*
@@ -29,12 +35,12 @@ struct rtas_args;
 #define merge_64(high, low) (((u64)high << 32) | low)
 #endif
 
-long sys_ni_syscall(void);
-
 /*
  * PowerPC architecture-specific syscalls
  */
 
+#ifndef CONFIG_ARCH_HAS_SYSCALL_WRAPPER
+
 long sys_rtas(struct rtas_args __user *uargs);
 
 #ifdef CONFIG_PPC64
@@ -114,5 +120,19 @@ long sys_ppc_fadvise64_64(int fd, int advice,
 			  u32 len_high, u32 len_low);
 #endif
 
+#else
+
+#define __SYSCALL_WITH_COMPAT(nr, native, compat)	__SYSCALL(nr, native)
+#define __SYSCALL(nr, entry) \
+	long __powerpc_##entry(const struct pt_regs *regs);
+
+#ifdef CONFIG_PPC64
+#include <asm/syscall_table_64.h>
+#else
+#include <asm/syscall_table_32.h>
+#endif /* CONFIG_PPC64 */
+
+#endif /* CONFIG_ARCH_HAS_SYSCALL_WRAPPER */
+
 #endif /* __KERNEL__ */
 #endif /* __ASM_POWERPC_SYSCALLS_H */

arch/powerpc/kernel/syscall.c

@@ -15,7 +15,6 @@
 /* Has to run notrace because it is entered not completely "reconciled" */
 notrace long system_call_exception(struct pt_regs *regs, unsigned long r0)
 {
-	unsigned long r3, r4, r5, r6, r7, r8;
 	long ret;
 	syscall_fn f;
 
@@ -145,31 +144,34 @@ notrace long system_call_exception(struct pt_regs *regs, unsigned long r0)
 		return -ENOSYS;
 	}
 
-	r3 = regs->gpr[3];
-	r4 = regs->gpr[4];
-	r5 = regs->gpr[5];
-	r6 = regs->gpr[6];
-	r7 = regs->gpr[7];
-	r8 = regs->gpr[8];
-
 	/* May be faster to do array_index_nospec? */
 	barrier_nospec();
 
+#ifdef CONFIG_ARCH_HAS_SYSCALL_WRAPPER
+	// No COMPAT if we have SYSCALL_WRAPPER, see Kconfig
+	f = (void *)sys_call_table[r0];
+	ret = f(regs);
+#else
 	if (unlikely(is_compat_task())) {
+		unsigned long r3, r4, r5, r6, r7, r8;
+
 		f = (void *)compat_sys_call_table[r0];
 
-		r3 &= 0x00000000ffffffffULL;
-		r4 &= 0x00000000ffffffffULL;
-		r5 &= 0x00000000ffffffffULL;
-		r6 &= 0x00000000ffffffffULL;
-		r7 &= 0x00000000ffffffffULL;
-		r8 &= 0x00000000ffffffffULL;
+		r3 = regs->gpr[3] & 0x00000000ffffffffULL;
+		r4 = regs->gpr[4] & 0x00000000ffffffffULL;
+		r5 = regs->gpr[5] & 0x00000000ffffffffULL;
+		r6 = regs->gpr[6] & 0x00000000ffffffffULL;
+		r7 = regs->gpr[7] & 0x00000000ffffffffULL;
+		r8 = regs->gpr[8] & 0x00000000ffffffffULL;
 
+		ret = f(r3, r4, r5, r6, r7, r8);
 	} else {
 		f = (void *)sys_call_table[r0];
-	}
 
-	ret = f(r3, r4, r5, r6, r7, r8);
+		ret = f(regs->gpr[3], regs->gpr[4], regs->gpr[5],
+			regs->gpr[6], regs->gpr[7], regs->gpr[8]);
+	}
+#endif
 
 	/*
 	 * Ultimately, this value will get limited by KSTACK_OFFSET_MAX(),

arch/powerpc/kernel/systbl.c

@@ -15,13 +15,20 @@
 #include <asm/unistd.h>
 #include <asm/syscalls.h>
 
+#undef __SYSCALL_WITH_COMPAT
 #define __SYSCALL_WITH_COMPAT(nr, entry, compat) __SYSCALL(nr, entry)
 
+#undef __SYSCALL
+#ifdef CONFIG_ARCH_HAS_SYSCALL_WRAPPER
+#define __SYSCALL(nr, entry) [nr] = __powerpc_##entry,
+#define __powerpc_sys_ni_syscall	sys_ni_syscall
+#else
 /*
  * Coerce syscall handlers with arbitrary parameters to common type
  * requires cast to void* to avoid -Wcast-function-type.
  */
 #define __SYSCALL(nr, entry) [nr] = (void *) entry,
+#endif
 
 const syscall_fn sys_call_table[] = {
 #ifdef CONFIG_PPC64

arch/powerpc/kernel/vdso.c

@@ -39,6 +39,8 @@
 extern char vdso32_start, vdso32_end;
 extern char vdso64_start, vdso64_end;
 
+long sys_ni_syscall(void);
+
 /*
  * The vdso data page (aka. systemcfg for old ppc64 fans) is here.
  * Once the early boot kernel code no longer needs to muck around