riscv: Improve exception and system call latency

Many CPUs implement return address branch prediction as a stack. The
RISCV architecture refers to this as a return address stack (RAS). If
this gets corrupted then the CPU will mispredict at least one but
potentally many function returns.

There are two issues with the current RISCV exception code:

- We are using the alternate link stack (x5/t0) for the indirect branch
  which makes the hardware think this is a function return. This will
  corrupt the RAS.

- We modify the return address of handle_exception to point to
  ret_from_exception. This will also corrupt the RAS.

Testing the null system call latency before and after the patch:

Visionfive2 (StarFive JH7110 / U74)
baseline: 189.87 ns
patched:  176.76 ns

Lichee pi 4a (T-Head TH1520 / C910)
baseline: 666.58 ns
patched:  636.90 ns

Just over 7% on the U74 and just over 4% on the C910.

Signed-off-by: Anton Blanchard <antonb@tenstorrent.com>
Signed-off-by: Cyril Bur <cyrilbur@tenstorrent.com>
Tested-by: Jisheng Zhang <jszhang@kernel.org>
Reviewed-by: Jisheng Zhang <jszhang@kernel.org>
Link: https://lore.kernel.org/r/20240607061335.2197383-1-cyrilbur@tenstorrent.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>

diff --git a/arch/riscv/kernel/entry.S b/arch/riscv/kernel/entry.S
index 68a24cf9481afe0649e671bd764fa688307c23bf..c933460ed3e9fc3363ccae2b962ba32e6bf9a74e 100644 (file)
--- a/arch/riscv/kernel/entry.S
+++ b/arch/riscv/kernel/entry.S
@@ -88,7 +88,6 @@ SYM_CODE_START(handle_exception)
        call riscv_v_context_nesting_start
 #endif
        move a0, sp /* pt_regs */
-       la ra, ret_from_exception
 
        /*
         * MSB of cause differentiates between
@@ -97,7 +96,8 @@ SYM_CODE_START(handle_exception)
        bge s4, zero, 1f
 
        /* Handle interrupts */
-       tail do_irq
+       call do_irq
+       j ret_from_exception
 1:
        /* Handle other exceptions */
        slli t0, s4, RISCV_LGPTR
@@ -105,11 +105,14 @@ SYM_CODE_START(handle_exception)
        la t2, excp_vect_table_end
        add t0, t1, t0
        /* Check if exception code lies within bounds */
-       bgeu t0, t2, 1f
-       REG_L t0, 0(t0)
-       jr t0
-1:
-       tail do_trap_unknown
+       bgeu t0, t2, 3f
+       REG_L t1, 0(t0)
+2:     jalr t1
+       j ret_from_exception
+3:
+
+       la t1, do_trap_unknown
+       j 2b
 SYM_CODE_END(handle_exception)
 ASM_NOKPROBE(handle_exception)
 

diff --git a/arch/riscv/kernel/stacktrace.c b/arch/riscv/kernel/stacktrace.c
index 10e311b2759d39d4480f1c3274c313c01b7efcb4..c6d5de22463f9d54a85675f1fbb3b63ba4a47619 100644 (file)
--- a/arch/riscv/kernel/stacktrace.c
+++ b/arch/riscv/kernel/stacktrace.c
@@ -16,7 +16,7 @@
 
 #ifdef CONFIG_FRAME_POINTER
 
-extern asmlinkage void ret_from_exception(void);
+extern asmlinkage void handle_exception(void);
 
 static inline int fp_is_valid(unsigned long fp, unsigned long sp)
 {
@@ -71,7 +71,7 @@ void notrace walk_stackframe(struct task_struct *task, struct pt_regs *regs,
                        fp = frame->fp;
                        pc = ftrace_graph_ret_addr(current, &graph_idx, frame->ra,
                                                   &frame->ra);
-                       if (pc == (unsigned long)ret_from_exception) {
+                       if (pc == (unsigned long)handle_exception) {
                                if (unlikely(!__kernel_text_address(pc) || !fn(arg, pc)))
                                        break;