RISC-V: Clear load reservations while restoring hart contexts

This is almost entirely a comment.  The bug is unlikely to manifest on
existing hardware because there is a timeout on load reservations, but
manifests on QEMU because there is no timeout.

Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Paul Walmsley <paul.walmsley@sifive.com>

arch/riscv/include/asm/asm.h

@@ -22,6 +22,7 @@
 
 #define REG_L		__REG_SEL(ld, lw)
 #define REG_S		__REG_SEL(sd, sw)
+#define REG_SC		__REG_SEL(sc.d, sc.w)
 #define SZREG		__REG_SEL(8, 4)
 #define LGREG		__REG_SEL(3, 2)
 

arch/riscv/kernel/entry.S

@@ -98,7 +98,26 @@ _save_context:
  */
 	.macro RESTORE_ALL
 	REG_L a0, PT_SSTATUS(sp)
-	REG_L a2, PT_SEPC(sp)
+	/*
+	 * The current load reservation is effectively part of the processor's
+	 * state, in the sense that load reservations cannot be shared between
+	 * different hart contexts.  We can't actually save and restore a load
+	 * reservation, so instead here we clear any existing reservation --
+	 * it's always legal for implementations to clear load reservations at
+	 * any point (as long as the forward progress guarantee is kept, but
+	 * we'll ignore that here).
+	 *
+	 * Dangling load reservations can be the result of taking a trap in the
+	 * middle of an LR/SC sequence, but can also be the result of a taken
+	 * forward branch around an SC -- which is how we implement CAS.  As a
+	 * result we need to clear reservations between the last CAS and the
+	 * jump back to the new context.  While it is unlikely the store
+	 * completes, implementations are allowed to expand reservations to be
+	 * arbitrarily large.
+	 */
+	REG_L  a2, PT_SEPC(sp)
+	REG_SC x0, a2, PT_SEPC(sp)
+
 	csrw CSR_SSTATUS, a0
 	csrw CSR_SEPC, a2