Merge patch series "RISC-V: Align the shadow stack"

Palmer Dabbelt <palmer@rivosinc.com> says:

This contains a pair of cleanups that depend on a fix that has already
landed upstream.

* b4-shazam-merge:
  RISC-V: Add some comments about the shadow and overflow stacks
  RISC-V: Align the shadow stack
  riscv: fix race when vmap stack overflow

Link: https://lore.kernel.org/r/20221130023515.20217-1-palmer@rivosinc.comSigned-off-by: Palmer Dabbelt <palmer@rivosinc.com>

arch/riscv/include/asm/asm.h

@@ -23,6 +23,7 @@
 #define REG_L		__REG_SEL(ld, lw)
 #define REG_S		__REG_SEL(sd, sw)
 #define REG_SC		__REG_SEL(sc.d, sc.w)
+#define REG_AMOSWAP_AQ	__REG_SEL(amoswap.d.aq, amoswap.w.aq)
 #define REG_ASM		__REG_SEL(.dword, .word)
 #define SZREG		__REG_SEL(8, 4)
 #define LGREG		__REG_SEL(3, 2)

arch/riscv/kernel/entry.S

@@ -397,6 +397,19 @@ handle_syscall_trace_exit:
 
 #ifdef CONFIG_VMAP_STACK
 handle_kernel_stack_overflow:
+	/*
+	 * Takes the psuedo-spinlock for the shadow stack, in case multiple
+	 * harts are concurrently overflowing their kernel stacks.  We could
+	 * store any value here, but since we're overflowing the kernel stack
+	 * already we only have SP to use as a scratch register.  So we just
+	 * swap in the address of the spinlock, as that's definately non-zero.
+	 *
+	 * Pairs with a store_release in handle_bad_stack().
+	 */
+1:	la sp, spin_shadow_stack
+	REG_AMOSWAP_AQ sp, sp, (sp)
+	bnez sp, 1b
+
 	la sp, shadow_stack
 	addi sp, sp, SHADOW_OVERFLOW_STACK_SIZE
 

arch/riscv/kernel/traps.c

@@ -208,13 +208,28 @@ int is_valid_bugaddr(unsigned long pc)
 #endif /* CONFIG_GENERIC_BUG */
 
 #ifdef CONFIG_VMAP_STACK
+/*
+ * Extra stack space that allows us to provide panic messages when the kernel
+ * has overflowed its stack.
+ */
 static DEFINE_PER_CPU(unsigned long [OVERFLOW_STACK_SIZE/sizeof(long)],
 		overflow_stack)__aligned(16);
 /*
- * shadow stack, handled_ kernel_ stack_ overflow(in kernel/entry.S) is used
- * to get per-cpu overflow stack(get_overflow_stack).
+ * A temporary stack for use by handle_kernel_stack_overflow.  This is used so
+ * we can call into C code to get the per-hart overflow stack.  Usage of this
+ * stack must be protected by spin_shadow_stack.
+ */
+long shadow_stack[SHADOW_OVERFLOW_STACK_SIZE/sizeof(long)] __aligned(16);
+
+/*
+ * A pseudo spinlock to protect the shadow stack from being used by multiple
+ * harts concurrently.  This isn't a real spinlock because the lock side must
+ * be taken without a valid stack and only a single register, it's only taken
+ * while in the process of panicing anyway so the performance and error
+ * checking a proper spinlock gives us doesn't matter.
  */
-long shadow_stack[SHADOW_OVERFLOW_STACK_SIZE/sizeof(long)];
+unsigned long spin_shadow_stack;
+
 asmlinkage unsigned long get_overflow_stack(void)
 {
 	return (unsigned long)this_cpu_ptr(overflow_stack) +
@@ -226,6 +241,15 @@ asmlinkage void handle_bad_stack(struct pt_regs *regs)
 	unsigned long tsk_stk = (unsigned long)current->stack;
 	unsigned long ovf_stk = (unsigned long)this_cpu_ptr(overflow_stack);
 
+	/*
+	 * We're done with the shadow stack by this point, as we're on the
+	 * overflow stack.  Tell any other concurrent overflowing harts that
+	 * they can proceed with panicing by releasing the pseudo-spinlock.
+	 *
+	 * This pairs with an amoswap.aq in handle_kernel_stack_overflow.
+	 */
+	smp_store_release(&spin_shadow_stack, 0);
+
 	console_verbose();
 
 	pr_emerg("Insufficient stack space to handle exception!\n");