svn commit: r335555 - stable/11/sys/x86/x86

[Andriy Gapon]

Author: avg
Date: Fri Jun 22 10:44:33 2018
New Revision: 335555
URL: https://svnweb.freebsd.org/changeset/base/335555

Log:
  MFC r333321,r333707: x86 cpususpend_handler: call wbinvd after setting suspend state bits
  
  Without a subsequent wbinvd the changes to suspended_cpus (and
  resuming_cpus) can be lost at least on AMD systems that use MOESI cache
  coherency protocol.  That can happen because one of APs ends up as an
  Owner of the corresponding cache line(s) and the changes may never reach
  the main memory before the AP is reset.
  
  This change fixed suspend to RAM a previously broken AMD-based system.

Modified:
  stable/11/sys/x86/x86/mp_x86.c
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/sys/x86/x86/mp_x86.c
==============================================================================
--- stable/11/sys/x86/x86/mp_x86.c	Fri Jun 22 10:39:22 2018	(r335554)
+++ stable/11/sys/x86/x86/mp_x86.c	Fri Jun 22 10:44:33 2018	(r335555)
@@ -1317,15 +1317,33 @@ cpususpend_handler(void)
 #else
 		npxsuspend(susppcbs[cpu]->sp_fpususpend);
 #endif
-		wbinvd();
-		CPU_SET_ATOMIC(cpu, &suspended_cpus);
 		/*
-		 * Hack for xen, which does not use resumectx() so never
-		 * uses the next clause: set resuming_cpus early so that
-		 * resume_cpus() can wait on the same bitmap for acpi and
-		 * xen.  resuming_cpus now means eventually_resumable_cpus.
+		 * suspended_cpus is cleared shortly after each AP is restarted
+		 * by a Startup IPI, so that the BSP can proceed to restarting
+		 * the next AP.
+		 *
+		 * resuming_cpus gets cleared when the AP completes
+		 * initialization after having been released by the BSP.
+		 * resuming_cpus is probably not the best name for the
+		 * variable, because it is actually a set of processors that
+		 * haven't resumed yet and haven't necessarily started resuming.
+		 *
+		 * Note that suspended_cpus is meaningful only for ACPI suspend
+		 * as it's not really used for Xen suspend since the APs are
+		 * automatically restored to the running state and the correct
+		 * context.  For the same reason resumectx is never called in
+		 * that case.
 		 */
+		CPU_SET_ATOMIC(cpu, &suspended_cpus);
 		CPU_SET_ATOMIC(cpu, &resuming_cpus);
+
+		/*
+		 * Invalidate the cache after setting the global status bits.
+		 * The last AP to set its bit may end up being an Owner of the
+		 * corresponding cache line in MOESI protocol.  The AP may be
+		 * stopped before the cache line is written to the main memory.
+		 */
+		wbinvd();
 	} else {
 #ifdef __amd64__
 		fpuresume(susppcbs[cpu]->sp_fpususpend);
@@ -1337,7 +1355,7 @@ cpususpend_handler(void)
 		PCPU_SET(switchtime, 0);
 		PCPU_SET(switchticks, ticks);
 
-		/* Indicate that we are resuming */
+		/* Indicate that we have restarted and restored the context. */
 		CPU_CLR_ATOMIC(cpu, &suspended_cpus);
 	}