MFC: r260457

The changes in r233781 attempted to make logging during a machine check
exception more readable.  In practice they prevented all logging during
a machine check exception on at least some systems.  Specifically, when
an uncorrected ECC error is detected in a DIMM on a Nehalem/Westmere
class machine, all CPUs receive a machine check exception, but only
CPUs on the same package as the memory controller for the erroring DIMM
log an error.  The CPUs on the other package would complete the scan of
their machine check banks and panic before the first set of CPUs could
log an error.  The end result was a clearer display during the panic
(no interleaved messages), but a crashdump without any useful info about
the error that occurred.

To handle this case, make all CPUs spin in the machine check handler
once they have completed their scan of their machine check banks until
at least one machine check error is logged.  I tried using a DELAY()
instead so that the CPUs would not potentially hang forever, but that
was not reliable in testing.

While here, don't clear MCIP from MSR_MCG_STATUS before invoking panic.
Only clear it if the machine check handler does not panic and returns
to the interrupted thread.

MFC: r263113

Correct type for malloc().

Submitted by:	"Conrad Meyer" <conrad.meyer@isilon.com>

MFC: r269052, r269239, r269242

Intel desktop Haswell CPUs may report benign corrected parity errors (see
HSD131 erratum in [1]) at a considerable rate. So filter these (default),
unless logging is enabled. Unfortunately, there really is no better way to
reasonably implement suppressing these errors than to just skipping them
in mca_log(). Given that they are reported for bank 0, they'd need to be
masked in MSR_MC0_CTL. However, P6 family processors require that register
to be set to either all 0s or all 1s, disabling way more than the one error
in question when using all 0s there. Alternatively, it could be masked for
the corresponding CMCI, but that still wouldn't keep the periodic scanner
from detecting these spurious errors. Apart from that, register contents of
MSR_MC0_CTL{,2} don't seem to be publicly documented, neither in the Intel
Architectures Developer's Manual nor in the Haswell datasheets.

Note that while HSD131 actually is only about C0-stepping as of revision
014 of the Intel desktop 4th generation processor family specification
update, these corrected errors also have been observed with D0-stepping
aka "Haswell Refresh".

1: http://www.intel.de/content/dam/www/public/us/en/documents/specification-updates/4th-gen-core-family-desktop-specification-update.pdf

Reviewed by:	jhb
Sponsored by:	Bally Wulff Games & Entertainment GmbH

1 files changed, 48 insertions, 7 deletions

diff --git a/sys/x86/x86/mca.c b/sys/x86/x86/mca.c
index f1369cd..b8098b3 100644
--- a/sys/x86/x86/mca.c
+++ b/sys/x86/x86/mca.c
@@ -53,6 +53,7 @@ __FBSDID("$FreeBSD$");
 #include <sys/taskqueue.h>
 #include <machine/intr_machdep.h>
 #include <machine/apicvar.h>
+#include <machine/cpu.h>
 #include <machine/cputypes.h>
 #include <x86/mca.h>
 #include <machine/md_var.h>
@@ -84,7 +85,7 @@ struct mca_internal {
 
 static MALLOC_DEFINE(M_MCA, "MCA", "Machine Check Architecture");
 
-static int mca_count;		/* Number of records stored. */
+static volatile int mca_count;	/* Number of records stored. */
 static int mca_banks;		/* Number of per-CPU register banks. */
 
 static SYSCTL_NODE(_hw, OID_AUTO, mca, CTLFLAG_RD, NULL,
@@ -100,6 +101,11 @@ TUNABLE_INT("hw.mca.amd10h_L1TP", &amd10h_L1TP);
 SYSCTL_INT(_hw_mca, OID_AUTO, amd10h_L1TP, CTLFLAG_RDTUN, &amd10h_L1TP, 0,
     "Administrative toggle for logging of level one TLB parity (L1TP) errors");
 
+static int intel6h_HSD131;
+TUNABLE_INT("hw.mca.intel6h_hsd131", &intel6h_HSD131);
+SYSCTL_INT(_hw_mca, OID_AUTO, intel6h_HSD131, CTLFLAG_RDTUN, &intel6h_HSD131, 0,
+    "Administrative toggle for logging of spurious corrected errors");
+
 int workaround_erratum383;
 SYSCTL_INT(_hw_mca, OID_AUTO, erratum383, CTLFLAG_RD, &workaround_erratum383, 0,
     "Is the workaround for Erratum 383 on AMD Family 10h processors enabled?");
@@ -243,12 +249,34 @@ mca_error_mmtype(uint16_t mca_error)
 	return ("???");
 }
 
+static int __nonnull(1)
+mca_mute(const struct mca_record *rec)
+{
+
+	/*
+	 * Skip spurious corrected parity errors generated by desktop Haswell
+	 * (see HSD131 erratum) unless reporting is enabled.
+	 * Note that these errors also have been observed with D0-stepping,
+	 * while the revision 014 desktop Haswell specification update only
+	 * talks about C0-stepping.
+	 */
+	if (rec->mr_cpu_vendor_id == CPU_VENDOR_INTEL &&
+	    rec->mr_cpu_id == 0x306c3 && rec->mr_bank == 0 &&
+	    rec->mr_status == 0x90000040000f0005 && !intel6h_HSD131)
+	    	return (1);
+
+	return (0);
+}
+
 /* Dump details about a single machine check. */
 static void __nonnull(1)
 mca_log(const struct mca_record *rec)
 {
 	uint16_t mca_error;
 
+	if (mca_mute(rec))
+	    	return;
+
 	printf("MCA: Bank %d, Status 0x%016llx\n", rec->mr_bank,
 	    (long long)rec->mr_status);
 	printf("MCA: Global Cap 0x%016llx, Status 0x%016llx\n",
@@ -699,8 +727,8 @@ cmci_setup(void)
 {
 	int i;
 
-	cmc_state = malloc((mp_maxid + 1) * sizeof(struct cmc_state **),
-	    M_MCA, M_WAITOK);
+	cmc_state = malloc((mp_maxid + 1) * sizeof(struct cmc_state *), M_MCA,
+	    M_WAITOK);
 	for (i = 0; i <= mp_maxid; i++)
 		cmc_state[i] = malloc(sizeof(struct cmc_state) * mca_banks,
 		    M_MCA, M_WAITOK | M_ZERO);
@@ -733,7 +761,8 @@ mca_setup(uint64_t mcg_cap)
 	TASK_INIT(&mca_refill_task, 0, mca_refill, NULL);
 	mca_fill_freelist();
 	SYSCTL_ADD_INT(NULL, SYSCTL_STATIC_CHILDREN(_hw_mca), OID_AUTO,
-	    "count", CTLFLAG_RD, &mca_count, 0, "Record count");
+	    "count", CTLFLAG_RD, (int *)(uintptr_t)&mca_count, 0,
+	    "Record count");
 	SYSCTL_ADD_PROC(NULL, SYSCTL_STATIC_CHILDREN(_hw_mca), OID_AUTO,
 	    "interval", CTLTYPE_INT | CTLFLAG_RW | CTLFLAG_MPSAFE, &mca_ticks,
 	    0, sysctl_positive_int, "I",
@@ -939,7 +968,7 @@ void
 mca_intr(void)
 {
 	uint64_t mcg_status;
-	int recoverable;
+	int old_count, recoverable;
 
 	if (!(cpu_feature & CPUID_MCA)) {
 		/*
@@ -953,15 +982,27 @@ mca_intr(void)
 	}
 
 	/* Scan the banks and check for any non-recoverable errors. */
+	old_count = mca_count;
 	recoverable = mca_scan(MCE);
 	mcg_status = rdmsr(MSR_MCG_STATUS);
 	if (!(mcg_status & MCG_STATUS_RIPV))
 		recoverable = 0;
 
+	if (!recoverable) {
+		/*
+		 * Wait for at least one error to be logged before
+		 * panic'ing.  Some errors will assert a machine check
+		 * on all CPUs, but only certain CPUs will find a valid
+		 * bank to log.
+		 */
+		while (mca_count == old_count)
+			cpu_spinwait();
+
+		panic("Unrecoverable machine check exception");
+	}
+
 	/* Clear MCIP. */
 	wrmsr(MSR_MCG_STATUS, mcg_status & ~MCG_STATUS_MCIP);
-	if (!recoverable)
-		panic("Unrecoverable machine check exception");
 }
 
 #ifdef DEV_APIC