powerpc/signal: Don't manage floating point regs when no FPU

There is no point in copying floating point regs when there
is no FPU and MATH_EMULATION is not selected.

Create a new CONFIG_PPC_FPU_REGS bool that is selected by
CONFIG_MATH_EMULATION and CONFIG_PPC_FPU, and use it to
opt out everything related to fp_state in thread_struct.

The asm const used only by fpu.S are opted out with CONFIG_PPC_FPU
as fpu.S build is conditionnal to CONFIG_PPC_FPU.

The following app spends approx 8.1 seconds system time on an 8xx
without the patch, and 7.0 seconds with the patch (13.5% reduction).

On an 832x, it spends approx 2.6 seconds system time without
the patch and 2.1 seconds with the patch (19% reduction).

	void sigusr1(int sig) { }

	int main(int argc, char **argv)
	{
		int i = 100000;

		signal(SIGUSR1, sigusr1);
		for (;i--;)
			raise(SIGUSR1);
		exit(0);
	}
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/7569070083e6cd5b279bb5023da601aba3c06f3c.1597770847.git.christophe.leroy@csgroup.eu

arch/powerpc/Kconfig

@@ -422,6 +422,7 @@ config HUGETLB_PAGE_SIZE_VARIABLE
 config MATH_EMULATION
 	bool "Math emulation"
 	depends on 4xx || PPC_8xx || PPC_MPC832x || BOOKE
+	select PPC_FPU_REGS
 	help
 	  Some PowerPC chips designed for embedded applications do not have
 	  a floating-point unit and therefore do not implement the

arch/powerpc/include/asm/processor.h

@@ -164,8 +164,10 @@ struct thread_struct {
 #endif
 	/* Debug Registers */
 	struct debug_reg debug;
+#ifdef CONFIG_PPC_FPU_REGS
 	struct thread_fp_state	fp_state;
 	struct thread_fp_state	*fp_save_area;
+#endif
 	int		fpexc_mode;	/* floating-point exception mode */
 	unsigned int	align_ctl;	/* alignment handling control */
 #ifdef CONFIG_HAVE_HW_BREAKPOINT

arch/powerpc/kernel/asm-offsets.c

@@ -110,9 +110,11 @@ int main(void)
 #ifdef CONFIG_BOOKE
 	OFFSET(THREAD_NORMSAVES, thread_struct, normsave[0]);
 #endif
+#ifdef CONFIG_PPC_FPU
 	OFFSET(THREAD_FPEXC_MODE, thread_struct, fpexc_mode);
 	OFFSET(THREAD_FPSTATE, thread_struct, fp_state.fpr);
 	OFFSET(THREAD_FPSAVEAREA, thread_struct, fp_save_area);
+#endif
 	OFFSET(FPSTATE_FPSCR, thread_fp_state, fpscr);
 	OFFSET(THREAD_LOAD_FP, thread_struct, load_fp);
 #ifdef CONFIG_ALTIVEC

arch/powerpc/kernel/process.c

@@ -1730,7 +1730,9 @@ int copy_thread(unsigned long clone_flags, unsigned long usp,
 		p->thread.ptrace_bps[i] = NULL;
 #endif
 
+#ifdef CONFIG_PPC_FPU_REGS
 	p->thread.fp_save_area = NULL;
+#endif
 #ifdef CONFIG_ALTIVEC
 	p->thread.vr_save_area = NULL;
 #endif
@@ -1855,8 +1857,10 @@ void start_thread(struct pt_regs *regs, unsigned long start, unsigned long sp)
 #endif
 	current->thread.load_slb = 0;
 	current->thread.load_fp = 0;
+#ifdef CONFIG_PPC_FPU_REGS
 	memset(&current->thread.fp_state, 0, sizeof(current->thread.fp_state));
 	current->thread.fp_save_area = NULL;
+#endif
 #ifdef CONFIG_ALTIVEC
 	memset(&current->thread.vr_state, 0, sizeof(current->thread.vr_state));
 	current->thread.vr_state.vscr.u[3] = 0x00010000; /* Java mode disabled */

arch/powerpc/kernel/ptrace/Makefile

@@ -6,11 +6,11 @@
 CFLAGS_ptrace-view.o		+= -DUTS_MACHINE='"$(UTS_MACHINE)"'
 
 obj-y				+= ptrace.o ptrace-view.o
-obj-y				+= ptrace-fpu.o
+obj-$(CONFIG_PPC_FPU_REGS)	+= ptrace-fpu.o
 obj-$(CONFIG_COMPAT)		+= ptrace32.o
 obj-$(CONFIG_VSX)		+= ptrace-vsx.o
 ifneq ($(CONFIG_VSX),y)
-obj-y				+= ptrace-novsx.o
+obj-$(CONFIG_PPC_FPU_REGS)	+= ptrace-novsx.o
 endif
 obj-$(CONFIG_ALTIVEC)		+= ptrace-altivec.o
 obj-$(CONFIG_SPE)		+= ptrace-spe.o

arch/powerpc/kernel/ptrace/ptrace-decl.h

@@ -165,8 +165,22 @@ int ptrace_put_reg(struct task_struct *task, int regno, unsigned long data);
 extern const struct user_regset_view user_ppc_native_view;
 
 /* ptrace-fpu */
+#ifdef CONFIG_PPC_FPU_REGS
 int ptrace_get_fpr(struct task_struct *child, int index, unsigned long *data);
 int ptrace_put_fpr(struct task_struct *child, int index, unsigned long data);
+#else
+static inline int
+ptrace_get_fpr(struct task_struct *child, int index, unsigned long *data)
+{
+	return -EIO;
+}
+
+static inline int
+ptrace_put_fpr(struct task_struct *child, int index, unsigned long data)
+{
+	return -EIO;
+}
+#endif
 
 /* ptrace-(no)adv */
 void ppc_gethwdinfo(struct ppc_debug_info *dbginfo);

arch/powerpc/kernel/ptrace/ptrace-view.c

@@ -521,11 +521,13 @@ static const struct user_regset native_regsets[] = {
 		.size = sizeof(long), .align = sizeof(long),
 		.regset_get = gpr_get, .set = gpr_set
 	},
+#ifdef CONFIG_PPC_FPU_REGS
 	[REGSET_FPR] = {
 		.core_note_type = NT_PRFPREG, .n = ELF_NFPREG,
 		.size = sizeof(double), .align = sizeof(double),
 		.regset_get = fpr_get, .set = fpr_set
 	},
+#endif
 #ifdef CONFIG_ALTIVEC
 	[REGSET_VMX] = {
 		.core_note_type = NT_PPC_VMX, .n = 34,

arch/powerpc/kernel/signal.h

@@ -34,7 +34,7 @@ unsigned long copy_fpr_to_user(void __user *to, struct task_struct *task);
 unsigned long copy_ckfpr_to_user(void __user *to, struct task_struct *task);
 unsigned long copy_fpr_from_user(struct task_struct *task, void __user *from);
 unsigned long copy_ckfpr_from_user(struct task_struct *task, void __user *from);
-#else
+#elif defined(CONFIG_PPC_FPU_REGS)
 static inline unsigned long
 copy_fpr_to_user(void __user *to, struct task_struct *task)
 {
@@ -63,6 +63,18 @@ copy_ckfpr_from_user(struct task_struct *task, void __user *from)
 				ELF_NFPREG * sizeof(double));
 }
 #endif /* CONFIG_PPC_TRANSACTIONAL_MEM */
+#else
+static inline unsigned long
+copy_fpr_to_user(void __user *to, struct task_struct *task)
+{
+	return 0;
+}
+
+static inline unsigned long
+copy_fpr_from_user(struct task_struct *task, void __user *from)
+{
+	return 0;
+}
 #endif
 
 #ifdef CONFIG_PPC64

arch/powerpc/kernel/signal_32.c

@@ -814,7 +814,9 @@ int handle_rt_signal32(struct ksignal *ksig, sigset_t *oldset,
 	}
 	regs->link = tramp;
 
+#ifdef CONFIG_PPC_FPU_REGS
 	tsk->thread.fp_state.fpscr = 0;	/* turn off all fp exceptions */
+#endif
 
 	/* create a stack frame for the caller of the handler */
 	newsp = ((unsigned long)rt_sf) - (__SIGNAL_FRAMESIZE + 16);
@@ -1271,7 +1273,9 @@ int handle_signal32(struct ksignal *ksig, sigset_t *oldset,
 
 	regs->link = tramp;
 
+#ifdef CONFIG_PPC_FPU_REGS
 	tsk->thread.fp_state.fpscr = 0;	/* turn off all fp exceptions */
+#endif
 
 	/* create a stack frame for the caller of the handler */
 	newsp = ((unsigned long)frame) - __SIGNAL_FRAMESIZE;

arch/powerpc/kernel/traps.c

@@ -1190,7 +1190,9 @@ static void parse_fpe(struct pt_regs *regs)
 
 	flush_fp_to_thread(current);
 
+#ifdef CONFIG_PPC_FPU_REGS
 	code = __parse_fpscr(current->thread.fp_state.fpscr);
+#endif
 
 	_exception(SIGFPE, regs, code, regs->nip);
 }

arch/powerpc/platforms/Kconfig.cputype

@@ -218,9 +218,13 @@ config PPC_E500MC
 	  such as e5500/e6500), and must be disabled for running on
 	  e500v1 or e500v2.
 
+config PPC_FPU_REGS
+	bool
+
 config PPC_FPU
 	bool
 	default y if PPC64
+	select PPC_FPU_REGS
 
 config FSL_EMB_PERFMON
 	bool "Freescale Embedded Perfmon"