svn commit: r198978 - in releng/8.0/sys: amd64/amd64 i386/i386
Attilio Rao
attilio at FreeBSD.org
Fri Nov 6 10:17:08 UTC 2009
Author: attilio
Date: Fri Nov 6 10:17:08 2009
New Revision: 198978
URL: http://svn.freebsd.org/changeset/base/198978
Log:
MFC r198868, r198950:
Opteron rev E family of processor expose a bug where acq memory barriers
can be broken, resulting in random breakages.
Printout a warning message if affected family and model are found.
Approved by: re (kib)
Modified:
releng/8.0/sys/amd64/amd64/identcpu.c
releng/8.0/sys/i386/i386/identcpu.c
Directory Properties:
releng/8.0/sys/ (props changed)
releng/8.0/sys/amd64/include/xen/ (props changed)
releng/8.0/sys/cddl/contrib/opensolaris/ (props changed)
releng/8.0/sys/contrib/dev/acpica/ (props changed)
releng/8.0/sys/contrib/pf/ (props changed)
releng/8.0/sys/dev/xen/xenpci/ (props changed)
Modified: releng/8.0/sys/amd64/amd64/identcpu.c
==============================================================================
--- releng/8.0/sys/amd64/amd64/identcpu.c Fri Nov 6 10:15:15 2009 (r198977)
+++ releng/8.0/sys/amd64/amd64/identcpu.c Fri Nov 6 10:17:08 2009 (r198978)
@@ -607,6 +607,21 @@ print_AMD_info(void)
printf(", %d lines/tag", (regs[2] >> 8) & 0x0f);
print_AMD_l2_assoc((regs[2] >> 12) & 0x0f);
}
+
+ /*
+ * Opteron Rev E shows a bug as in very rare occasions a read memory
+ * barrier is not performed as expected if it is followed by a
+ * non-atomic read-modify-write instruction.
+ * As long as that bug pops up very rarely (intensive machine usage
+ * on other operating systems generally generates one unexplainable
+ * crash any 2 months) and as long as a model specific fix would be
+ * impratical at this stage, print out a warning string if the broken
+ * model and family are identified.
+ */
+ if (CPUID_TO_FAMILY(cpu_id) == 0xf && CPUID_TO_MODEL(cpu_id) >= 0x20 &&
+ CPUID_TO_MODEL(cpu_id) <= 0x3f)
+ printf("WARNING: This architecture revision has known SMP "
+ "hardware bugs which may cause random instability\n");
}
static void
Modified: releng/8.0/sys/i386/i386/identcpu.c
==============================================================================
--- releng/8.0/sys/i386/i386/identcpu.c Fri Nov 6 10:15:15 2009 (r198977)
+++ releng/8.0/sys/i386/i386/identcpu.c Fri Nov 6 10:17:08 2009 (r198978)
@@ -1303,6 +1303,21 @@ print_AMD_info(void)
(amd_whcr & 0x0100) ? "Enable" : "Disable");
}
}
+
+ /*
+ * Opteron Rev E shows a bug as in very rare occasions a read memory
+ * barrier is not performed as expected if it is followed by a
+ * non-atomic read-modify-write instruction.
+ * As long as that bug pops up very rarely (intensive machine usage
+ * on other operating systems generally generates one unexplainable
+ * crash any 2 months) and as long as a model specific fix would be
+ * impratical at this stage, print out a warning string if the broken
+ * model and family are identified.
+ */
+ if (CPUID_TO_FAMILY(cpu_id) == 0xf && CPUID_TO_MODEL(cpu_id) >= 0x20 &&
+ CPUID_TO_MODEL(cpu_id) <= 0x3f)
+ printf("WARNING: This architecture revision has known SMP "
+ "hardware bugs which may cause random instability\n");
}
static void
More information about the svn-src-all
mailing list