Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Stafford Horne <shorne@gmail.com>
To:     LKML <linux-kernel@vger.kernel.org>
Cc:     Linux OpenRISC <linux-openrisc@vger.kernel.org>,
        Stafford Horne <shorne@gmail.com>,
        Oleg Nesterov <oleg@redhat.com>,
        Jonas Bonn <jonas@southpole.se>,
        Stefan Kristiansson <stefan.kristiansson@saunalahti.fi>
Subject: [PATCH 2/4] openrisc: Support storing and restoring fpu state
Date:   Tue, 18 Apr 2023 17:58:11 +0100
Message-Id: <20230418165813.1900991-3-shorne@gmail.com>
In-Reply-To: <20230418165813.1900991-1-shorne@gmail.com>
References: <20230418165813.1900991-1-shorne@gmail.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

OpenRISC floating point state is not so expensive to save as OpenRISC uses
general purpose registers for floating point instructions.  We need to save
only the floating point status and control register, FPCSR.

Add support to maintain the FPCSR unconditionally upon exceptions and
switches.  On machines that do not support FPU this will always just
store 0x0 and restore is a no-op.  On FPU systems this adds an
additional special purpose register read/write and read/write to memory
(already cached).

Signed-off-by: Stafford Horne <shorne@gmail.com>
---
 arch/openrisc/include/asm/ptrace.h |  4 ++--
 arch/openrisc/kernel/entry.S       | 14 ++++++++++++++
 arch/openrisc/kernel/traps.c       |  5 +++--
 3 files changed, 19 insertions(+), 4 deletions(-)

diff --git a/arch/openrisc/include/asm/ptrace.h b/arch/openrisc/include/asm/ptrace.h
index 01f81d4e97dc..375147ff71fc 100644
--- a/arch/openrisc/include/asm/ptrace.h
+++ b/arch/openrisc/include/asm/ptrace.h
@@ -59,7 +59,7 @@ struct pt_regs {
 	 * -1 for all other exceptions.
 	 */
 	long  orig_gpr11;	/* For restarting system calls */
-	long dummy;		/* Cheap alignment fix */
+	long fpcsr;		/* Floating point control status register. */
 	long dummy2;		/* Cheap alignment fix */
 };
 
@@ -115,6 +115,6 @@ static inline long regs_return_value(struct pt_regs *regs)
 #define PT_GPR31      124
 #define PT_PC	      128
 #define PT_ORIG_GPR11 132
-#define PT_SYSCALLNO  136
+#define PT_FPCSR      136
 
 #endif /* __ASM_OPENRISC_PTRACE_H */
diff --git a/arch/openrisc/kernel/entry.S b/arch/openrisc/kernel/entry.S
index a130c4dac48d..c7b47e571220 100644
--- a/arch/openrisc/kernel/entry.S
+++ b/arch/openrisc/kernel/entry.S
@@ -106,6 +106,8 @@
 	l.mtspr r0,r3,SPR_EPCR_BASE				;\
 	l.lwz   r3,PT_SR(r1)					;\
 	l.mtspr r0,r3,SPR_ESR_BASE				;\
+	l.lwz	r3,PT_FPCSR(r1)					;\
+	l.mtspr	r0,r3,SPR_FPCSR					;\
 	l.lwz   r2,PT_GPR2(r1)					;\
 	l.lwz   r3,PT_GPR3(r1)					;\
 	l.lwz   r4,PT_GPR4(r1)					;\
@@ -175,6 +177,8 @@ handler:							;\
 	/* r30 already save */					;\
 	l.sw    PT_GPR31(r1),r31					;\
 	TRACE_IRQS_OFF_ENTRY						;\
+	l.mfspr	r30,r0,SPR_FPCSR				;\
+	l.sw	PT_FPCSR(r1),r30				;\
 	/* Store -1 in orig_gpr11 for non-syscall exceptions */	;\
 	l.addi	r30,r0,-1					;\
 	l.sw	PT_ORIG_GPR11(r1),r30
@@ -215,6 +219,8 @@ handler:							;\
 	/* Store -1 in orig_gpr11 for non-syscall exceptions */	;\
 	l.addi	r30,r0,-1					;\
 	l.sw	PT_ORIG_GPR11(r1),r30				;\
+	l.mfspr	r30,r0,SPR_FPCSR				;\
+	l.sw	PT_FPCSR(r1),r30				;\
 	l.addi	r3,r1,0						;\
 	/* r4 is exception EA */				;\
 	l.addi	r5,r0,vector					;\
@@ -1087,6 +1093,10 @@ ENTRY(_switch)
 	l.sw    PT_GPR28(r1),r28
 	l.sw    PT_GPR30(r1),r30
 
+	/* Store the old FPU state to new pt_regs */
+	l.mfspr	r29,r0,SPR_FPCSR
+	l.sw	PT_FPCSR(r1),r29
+
 	l.addi	r11,r10,0			/* Save old 'current' to 'last' return value*/
 
 	/* We use thread_info->ksp for storing the address of the above
@@ -1109,6 +1119,10 @@ ENTRY(_switch)
 	l.lwz	r29,PT_SP(r1)
 	l.sw	TI_KSP(r10),r29
 
+	/* Restore the old value of FPCSR */
+	l.lwz	r29,PT_FPCSR(r1)
+	l.mtspr	r0,r29,SPR_FPCSR
+
 	/* ...and restore the registers, except r11 because the return value
 	 * has already been set above.
 	 */
diff --git a/arch/openrisc/kernel/traps.c b/arch/openrisc/kernel/traps.c
index fd9a0f2b66c4..f5bbe6b55849 100644
--- a/arch/openrisc/kernel/traps.c
+++ b/arch/openrisc/kernel/traps.c
@@ -75,8 +75,9 @@ void show_registers(struct pt_regs *regs)
 		in_kernel = 0;
 
 	printk("CPU #: %d\n"
-	       "   PC: %08lx    SR: %08lx    SP: %08lx\n",
-	       smp_processor_id(), regs->pc, regs->sr, regs->sp);
+	       "   PC: %08lx    SR: %08lx    SP: %08lx FPCSR: %08lx\n",
+	       smp_processor_id(), regs->pc, regs->sr, regs->sp,
+	       regs->fpcsr);
 	printk("GPR00: %08lx GPR01: %08lx GPR02: %08lx GPR03: %08lx\n",
 	       0L, regs->gpr[1], regs->gpr[2], regs->gpr[3]);
 	printk("GPR04: %08lx GPR05: %08lx GPR06: %08lx GPR07: %08lx\n",
-- 
2.39.1