[MIPS] R4000/R4400 errata workarounds
This is the gereric part of R4000/R4400 errata workarounds. They include
compiler and assembler support as well as some source code modifications
to address the problems with some combinations of multiply/divide+shift
instructions as well as the daddi and daddiu instructions.
Changes included are as follows:
1. New Kconfig options to select workarounds by platforms as necessary.
2. Arch top-level Makefile to pass necessary options to the compiler; also
incompatible configurations are detected (-mno-sym32 unsupported as
horribly intrusive for little gain).
3. Bug detection updated and shuffled -- the multiply/divide+shift problem
is lethal enough that if not worked around it makes the kernel crash in
time_init() because of a division by zero; the daddiu erratum might
also trigger early potentially, though I have not observed it. On the
other hand the daddi detection code requires the exception subsystem to
have been initialised (and is there mainly for information).
4. r4k_daddiu_bug() added so that the existence of the erratum can be
queried by code at the run time as necessary; useful for generated code
like TLB fault and copy/clear page handlers.
5. __udelay() updated as it uses multiplication in inline assembly.
Note that -mdaddi requires modified toolchain (which has been maintained
by myself and available from my site for ~4years now -- versions covered
are GCC 2.95.4 - 4.1.2 and binutils from 2.13 onwards). The -mfix-r4000
and -mfix-r4400 have been standard for a while though.
Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
diff --git a/include/asm-mips/war.h b/include/asm-mips/war.h
index d2808ed..22361d5 100644
--- a/include/asm-mips/war.h
+++ b/include/asm-mips/war.h
@@ -4,6 +4,7 @@
* for more details.
*
* Copyright (C) 2002, 2004, 2007 by Ralf Baechle
+ * Copyright (C) 2007 Maciej W. Rozycki
*/
#ifndef _ASM_WAR_H
#define _ASM_WAR_H
@@ -11,6 +12,67 @@
#include <war.h>
/*
+ * Work around certain R4000 CPU errata (as implemented by GCC):
+ *
+ * - A double-word or a variable shift may give an incorrect result
+ * if executed immediately after starting an integer division:
+ * "MIPS R4000PC/SC Errata, Processor Revision 2.2 and 3.0",
+ * erratum #28
+ * "MIPS R4000MC Errata, Processor Revision 2.2 and 3.0", erratum
+ * #19
+ *
+ * - A double-word or a variable shift may give an incorrect result
+ * if executed while an integer multiplication is in progress:
+ * "MIPS R4000PC/SC Errata, Processor Revision 2.2 and 3.0",
+ * errata #16 & #28
+ *
+ * - An integer division may give an incorrect result if started in
+ * a delay slot of a taken branch or a jump:
+ * "MIPS R4000PC/SC Errata, Processor Revision 2.2 and 3.0",
+ * erratum #52
+ */
+#ifdef CONFIG_CPU_R4000_WORKAROUNDS
+#define R4000_WAR 1
+#else
+#define R4000_WAR 0
+#endif
+
+/*
+ * Work around certain R4400 CPU errata (as implemented by GCC):
+ *
+ * - A double-word or a variable shift may give an incorrect result
+ * if executed immediately after starting an integer division:
+ * "MIPS R4400MC Errata, Processor Revision 1.0", erratum #10
+ * "MIPS R4400MC Errata, Processor Revision 2.0 & 3.0", erratum #4
+ */
+#ifdef CONFIG_CPU_R4400_WORKAROUNDS
+#define R4400_WAR 1
+#else
+#define R4400_WAR 0
+#endif
+
+/*
+ * Work around the "daddi" and "daddiu" CPU errata:
+ *
+ * - The `daddi' instruction fails to trap on overflow.
+ * "MIPS R4000PC/SC Errata, Processor Revision 2.2 and 3.0",
+ * erratum #23
+ *
+ * - The `daddiu' instruction can produce an incorrect result.
+ * "MIPS R4000PC/SC Errata, Processor Revision 2.2 and 3.0",
+ * erratum #41
+ * "MIPS R4000MC Errata, Processor Revision 2.2 and 3.0", erratum
+ * #15
+ * "MIPS R4400PC/SC Errata, Processor Revision 1.0", erratum #7
+ * "MIPS R4400MC Errata, Processor Revision 1.0", erratum #5
+ */
+#ifdef CONFIG_CPU_DADDI_WORKAROUNDS
+#define DADDI_WAR 1
+#else
+#define DADDI_WAR 0
+#endif
+
+/*
* Another R4600 erratum. Due to the lack of errata information the exact
* technical details aren't known. I've experimentally found that disabling
* interrupts during indexed I-cache flushes seems to be sufficient to deal
