[binutils-gdb] MIPS/opcodes: Reorder legacy COP0, COP2, COP3 opcode instructions
Maciej W. Rozycki
macro@sourceware.org
Sat May 29 01:29:35 GMT 2021
https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=49149d595cfdfa32611b2abba564b9b5d7542c91
commit 49149d595cfdfa32611b2abba564b9b5d7542c91
Author: Maciej W. Rozycki <macro@orcam.me.uk>
Date: Sat May 29 03:26:33 2021 +0200
MIPS/opcodes: Reorder legacy COP0, COP2, COP3 opcode instructions
Group legacy instructions using the COP0, COP2, COP3 opcodes together
and by their coprocessor number, and move them towards the end of the
opcode table. No functional change.
With the addition of explicit ISA exclusions this is maybe not strictly
necessary anymore as the individual legacy instructions are not supposed
to match ISA levels or CPU implementations that have discarded them or
replaced with a new instruction each, but let's not have them scattered
randomly across blocks of unrelated instruction sets where someone chose
to put them previously. Perhaps they could be put back in alphabetical
order in the main instruction block, but let's leave it for another
occasion.
opcodes/
* mips-opc.c (mips_builtin_opcodes): Reorder legacy COP0, COP2,
COP3 opcode instructions.
Diff:
---
opcodes/ChangeLog | 5 ++
opcodes/mips-opc.c | 134 +++++++++++++++++++++++++++--------------------------
2 files changed, 73 insertions(+), 66 deletions(-)
diff --git a/opcodes/ChangeLog b/opcodes/ChangeLog
index 7f31e4e48d2..3d87487edf5 100644
--- a/opcodes/ChangeLog
+++ b/opcodes/ChangeLog
@@ -1,3 +1,8 @@
+2021-05-29 Maciej W. Rozycki <macro@orcam.me.uk>
+
+ * mips-opc.c (mips_builtin_opcodes): Reorder legacy COP0, COP2,
+ COP3 opcode instructions.
+
2021-05-29 Maciej W. Rozycki <macro@orcam.me.uk>
* mips-opc.c (mips_builtin_opcodes): Update exclusion list for
diff --git a/opcodes/mips-opc.c b/opcodes/mips-opc.c
index d3965965357..db72c039bfd 100644
--- a/opcodes/mips-opc.c
+++ b/opcodes/mips-opc.c
@@ -2103,67 +2103,6 @@ const struct mips_opcode mips_builtin_opcodes[] =
{"zcb", "(b)", 0x7000071f, 0xfc1fffff, RD_1|SM, 0, IOCT2, 0, 0 },
{"zcbt", "(b)", 0x7000075f, 0xfc1fffff, RD_1|SM, 0, IOCT2, 0, 0 },
-/* Coprocessor 0 move instructions cfc0 and ctc0 conflict with the
- mfhc0 and mthc0 XPA instructions, so they have been placed here
- to allow the XPA instructions to take precedence. */
-{"cfc0", "t,g", 0x40400000, 0xffe007ff, WR_1|RD_C0|LC, 0, I1, 0, I32 },
-{"ctc0", "t,g", 0x40c00000, 0xffe007ff, RD_1|WR_CC|CM, 0, I1, 0, I32 },
-
-/* Coprocessor 2 move/branch operations overlap with VR5400 .ob format
- instructions so they are here for the latters to take precedence. */
-{"bc2eqz", "E,p", 0x49200000, 0xffe00000, RD_C2|CBD, 0, I37, 0, 0 },
-{"bc2f", "p", 0x49000000, 0xffff0000, RD_CC|CBD, 0, I1, 0, N54|IOCT|IOCTP|IOCT2|I37 },
-{"bc2f", "N,p", 0x49000000, 0xffe30000, RD_CC|CBD, 0, I32, 0, IOCT|IOCTP|IOCT2|I37 },
-{"bc2fl", "p", 0x49020000, 0xffff0000, RD_CC|CBL, 0, I2|T3, 0, N54|IOCT|IOCTP|IOCT2|I37 },
-{"bc2fl", "N,p", 0x49020000, 0xffe30000, RD_CC|CBL, 0, I32, 0, IOCT|IOCTP|IOCT2|I37 },
-{"bc2nez", "E,p", 0x49a00000, 0xffe00000, RD_C2|CBD, 0, I37, 0, 0 },
-{"bc2t", "p", 0x49010000, 0xffff0000, RD_CC|CBD, 0, I1, 0, N54|IOCT|IOCTP|IOCT2|I37 },
-{"bc2t", "N,p", 0x49010000, 0xffe30000, RD_CC|CBD, 0, I32, 0, IOCT|IOCTP|IOCT2|I37 },
-{"bc2tl", "p", 0x49030000, 0xffff0000, RD_CC|CBL, 0, I2|T3, 0, N54|IOCT|IOCTP|IOCT2|I37 },
-{"bc2tl", "N,p", 0x49030000, 0xffe30000, RD_CC|CBL, 0, I32, 0, IOCT|IOCTP|IOCT2|I37 },
-{"cfc2", "t,g", 0x48400000, 0xffe007ff, WR_1|RD_C2|LC, 0, I1, 0, N54|IOCT|IOCTP|IOCT2|EE },
-{"cfc2", "t,+9", 0x48400000, 0xffe007ff, WR_1|RD_C2|LC, 0, EE, 0, 0 },
-{"cfc2.i", "t,+9", 0x48400001, 0xffe007ff, WR_1|RD_C2|LC, 0, EE, 0, 0 },
-{"cfc2.ni", "t,+9", 0x48400000, 0xffe007ff, WR_1|RD_C2|LC, 0, EE, 0, 0 },
-{"ctc2", "t,g", 0x48c00000, 0xffe007ff, RD_1|WR_CC|CM, 0, I1, 0, N54|IOCT|IOCTP|IOCT2|EE },
-{"ctc2", "t,+9", 0x48c00000, 0xffe007ff, RD_1|WR_CC|CM, 0, EE, 0, 0 },
-{"ctc2.i", "t,+9", 0x48c00001, 0xffe007ff, RD_1|WR_CC|CM, 0, EE, 0, 0 },
-{"ctc2.ni", "t,+9", 0x48c00000, 0xffe007ff, RD_1|WR_CC|CM, 0, EE, 0, 0 },
-{"dmfc2", "t,i", 0x48200000, 0xffe00000, WR_1|RD_C2|LC, 0, IOCT, 0, 0 },
-{"dmfc2", "t,G", 0x48200000, 0xffe007ff, WR_1|RD_C2|LC, 0, I3, 0, N54|IOCT|IOCTP|IOCT2|EE },
-{"dmfc2", "t,G,H", 0x48200000, 0xffe007f8, WR_1|RD_C2|LC, 0, I64, 0, IOCT|IOCTP|IOCT2 },
-{"dmtc2", "t,i", 0x48a00000, 0xffe00000, RD_1|WR_C2|WR_CC|CM, 0, IOCT, 0, 0 },
-{"dmtc2", "t,G", 0x48a00000, 0xffe007ff, RD_1|WR_C2|WR_CC|CM, 0, I3, 0, N54|IOCT|IOCTP|IOCT2|EE },
-{"dmtc2", "t,G,H", 0x48a00000, 0xffe007f8, RD_1|WR_C2|WR_CC|CM, 0, I64, 0, IOCT|IOCTP|IOCT2 },
-{"mfc2", "t,G", 0x48000000, 0xffe007ff, WR_1|RD_C2|LC, 0, I1, 0, N54|IOCT|IOCTP|IOCT2|EE },
-{"mfc2", "t,G,H", 0x48000000, 0xffe007f8, WR_1|RD_C2|LC, 0, I32, 0, IOCT|IOCTP|IOCT2 },
-{"mfhc2", "t,G", 0x48600000, 0xffe007ff, WR_1|RD_C2|LC, 0, I33, 0, IOCT|IOCTP|IOCT2 },
-{"mfhc2", "t,G,H", 0x48600000, 0xffe007f8, WR_1|RD_C2|LC, 0, I33, 0, IOCT|IOCTP|IOCT2 },
-{"mfhc2", "t,i", 0x48600000, 0xffe00000, WR_1|RD_C2|LC, 0, I33, 0, IOCT|IOCTP|IOCT2 },
-{"mtc2", "t,G", 0x48800000, 0xffe007ff, RD_1|WR_C2|WR_CC|CM, 0, I1, 0, N54|IOCT|IOCTP|IOCT2|EE },
-{"mtc2", "t,G,H", 0x48800000, 0xffe007f8, RD_1|WR_C2|WR_CC|CM, 0, I32, 0, IOCT|IOCTP|IOCT2 },
-{"mthc2", "t,G", 0x48e00000, 0xffe007ff, RD_1|WR_C2|WR_CC|CM, 0, I33, 0, IOCT|IOCTP|IOCT2 },
-{"mthc2", "t,G,H", 0x48e00000, 0xffe007f8, RD_1|WR_C2|WR_CC|CM, 0, I33, 0, IOCT|IOCTP|IOCT2 },
-{"mthc2", "t,i", 0x48e00000, 0xffe00000, RD_1|WR_C2|WR_CC|CM, 0, I33, 0, IOCT|IOCTP|IOCT2 },
-{"qmfc2", "t,+6", 0x48200000, 0xffe007ff, WR_1|RD_C2, 0, EE, 0, 0 },
-{"qmfc2.i", "t,+6", 0x48200001, 0xffe007ff, WR_1|RD_C2, 0, EE, 0, 0 },
-{"qmfc2.ni", "t,+6", 0x48200000, 0xffe007ff, WR_1|RD_C2, 0, EE, 0, 0 },
-{"qmtc2", "t,+6", 0x48a00000, 0xffe007ff, RD_1|WR_C2, 0, EE, 0, 0 },
-{"qmtc2.i", "t,+6", 0x48a00001, 0xffe007ff, RD_1|WR_C2, 0, EE, 0, 0 },
-{"qmtc2.ni", "t,+6", 0x48a00000, 0xffe007ff, RD_1|WR_C2, 0, EE, 0, 0 },
-/* Coprocessor 3 move/branch operations overlap with MIPS IV COP1X
- instructions, so they are here for the latters to take precedence. */
-{"bc3f", "p", 0x4d000000, 0xffff0000, RD_CC|CBD, 0, I1, 0, I3_33|EE },
-{"bc3fl", "p", 0x4d020000, 0xffff0000, RD_CC|CBL, 0, I2|T3, 0, I3_33|EE },
-{"bc3t", "p", 0x4d010000, 0xffff0000, RD_CC|CBD, 0, I1, 0, I3_33|EE },
-{"bc3tl", "p", 0x4d030000, 0xffff0000, RD_CC|CBL, 0, I2|T3, 0, I3_33|EE },
-{"cfc3", "t,g", 0x4c400000, 0xffe007ff, WR_1|RD_C3|LC, 0, I1, 0, I3_33|EE },
-{"ctc3", "t,g", 0x4cc00000, 0xffe007ff, RD_1|WR_CC|CM, 0, I1, 0, I3_33|EE },
-{"mfc3", "t,G", 0x4c000000, 0xffe007ff, WR_1|RD_C3|LC, 0, I1, 0, I3_33|EE },
-{"mfc3", "t,G,H", 0x4c000000, 0xffe007f8, WR_1|RD_C3|LC, 0, I32, 0, I3_33|EE },
-{"mtc3", "t,G", 0x4c800000, 0xffe007ff, RD_1|WR_C3|WR_CC|CM, 0, I1, 0, I3_33|EE },
-{"mtc3", "t,G,H", 0x4c800000, 0xffe007f8, RD_1|WR_C3|WR_CC|CM, 0, I32, 0, I3_33|EE },
-
/* Conflicts with the 4650's "mul" instruction. Nobody's using the
4010 any more, so move this insn out of the way. If the object
format gave us more info, we could do this right. */
@@ -2449,11 +2388,6 @@ const struct mips_opcode mips_builtin_opcodes[] =
{"dpaqx_sa.w.ph", "7,s,t", 0x7c0006b0, 0xfc00e7ff, RD_2|RD_3|MOD_a, 0, 0, D33, 0 },
{"dpsqx_s.w.ph", "7,s,t", 0x7c000670, 0xfc00e7ff, RD_2|RD_3|MOD_a, 0, 0, D33, 0 },
{"dpsqx_sa.w.ph", "7,s,t", 0x7c0006f0, 0xfc00e7ff, RD_2|RD_3|MOD_a, 0, 0, D33, 0 },
-/* Move bc0* after mftr and mttr to avoid opcode collision. */
-{"bc0f", "p", 0x41000000, 0xffff0000, RD_CC|CBD, 0, I1, 0, I4_32 },
-{"bc0fl", "p", 0x41020000, 0xffff0000, RD_CC|CBL, 0, I2|T3, 0, I4_32 },
-{"bc0t", "p", 0x41010000, 0xffff0000, RD_CC|CBD, 0, I1, 0, I4_32 },
-{"bc0tl", "p", 0x41030000, 0xffff0000, RD_CC|CBL, 0, I2|T3, 0, I4_32 },
/* ST Microelectronics Loongson-2E and -2F. */
{"mult.g", "d,s,t", 0x7c000018, 0xfc0007ff, WR_1|RD_2|RD_3, 0, IL2E, 0, 0 },
{"mult.g", "d,s,t", 0x70000010, 0xfc0007ff, WR_1|RD_2|RD_3, 0, IL2F, 0, 0 },
@@ -3395,9 +3329,77 @@ const struct mips_opcode mips_builtin_opcodes[] =
{"ginvi", "s", 0x7c00003d, 0xfc1fffff, RD_1, 0, 0, GINV, 0 },
{"ginvt", "s,+\\", 0x7c0000bd, 0xfc1ffcff, RD_1, 0, 0, GINV, 0 },
+/* Move bc0* after mftr and mttr to avoid opcode collision. */
+{"bc0f", "p", 0x41000000, 0xffff0000, RD_CC|CBD, 0, I1, 0, I4_32 },
+{"bc0fl", "p", 0x41020000, 0xffff0000, RD_CC|CBL, 0, I2|T3, 0, I4_32 },
+{"bc0t", "p", 0x41010000, 0xffff0000, RD_CC|CBD, 0, I1, 0, I4_32 },
+{"bc0tl", "p", 0x41030000, 0xffff0000, RD_CC|CBL, 0, I2|T3, 0, I4_32 },
+
+/* Coprocessor 0 move instructions cfc0 and ctc0 conflict with the
+ mfhc0 and mthc0 XPA instructions, so they have been placed here
+ to allow the XPA instructions to take precedence. */
+{"cfc0", "t,g", 0x40400000, 0xffe007ff, WR_1|RD_C0|LC, 0, I1, 0, I32 },
+{"ctc0", "t,g", 0x40c00000, 0xffe007ff, RD_1|WR_CC|CM, 0, I1, 0, I32 },
+
/* RFE conflicts with the new Virt spec instruction tlbgp. */
{"rfe", "", 0x42000010, 0xffffffff, 0, 0, I1|T3, 0, I3_32 },
+/* Coprocessor 2 move/branch operations overlap with VR5400 .ob format
+ instructions so they are here for the latters to take precedence. */
+{"bc2eqz", "E,p", 0x49200000, 0xffe00000, RD_C2|CBD, 0, I37, 0, 0 },
+{"bc2f", "p", 0x49000000, 0xffff0000, RD_CC|CBD, 0, I1, 0, N54|IOCT|IOCTP|IOCT2|I37 },
+{"bc2f", "N,p", 0x49000000, 0xffe30000, RD_CC|CBD, 0, I32, 0, IOCT|IOCTP|IOCT2|I37 },
+{"bc2fl", "p", 0x49020000, 0xffff0000, RD_CC|CBL, 0, I2|T3, 0, N54|IOCT|IOCTP|IOCT2|I37 },
+{"bc2fl", "N,p", 0x49020000, 0xffe30000, RD_CC|CBL, 0, I32, 0, IOCT|IOCTP|IOCT2|I37 },
+{"bc2nez", "E,p", 0x49a00000, 0xffe00000, RD_C2|CBD, 0, I37, 0, 0 },
+{"bc2t", "p", 0x49010000, 0xffff0000, RD_CC|CBD, 0, I1, 0, N54|IOCT|IOCTP|IOCT2|I37 },
+{"bc2t", "N,p", 0x49010000, 0xffe30000, RD_CC|CBD, 0, I32, 0, IOCT|IOCTP|IOCT2|I37 },
+{"bc2tl", "p", 0x49030000, 0xffff0000, RD_CC|CBL, 0, I2|T3, 0, N54|IOCT|IOCTP|IOCT2|I37 },
+{"bc2tl", "N,p", 0x49030000, 0xffe30000, RD_CC|CBL, 0, I32, 0, IOCT|IOCTP|IOCT2|I37 },
+{"cfc2", "t,g", 0x48400000, 0xffe007ff, WR_1|RD_C2|LC, 0, I1, 0, N54|IOCT|IOCTP|IOCT2|EE },
+{"cfc2", "t,+9", 0x48400000, 0xffe007ff, WR_1|RD_C2|LC, 0, EE, 0, 0 },
+{"cfc2.i", "t,+9", 0x48400001, 0xffe007ff, WR_1|RD_C2|LC, 0, EE, 0, 0 },
+{"cfc2.ni", "t,+9", 0x48400000, 0xffe007ff, WR_1|RD_C2|LC, 0, EE, 0, 0 },
+{"ctc2", "t,g", 0x48c00000, 0xffe007ff, RD_1|WR_CC|CM, 0, I1, 0, N54|IOCT|IOCTP|IOCT2|EE },
+{"ctc2", "t,+9", 0x48c00000, 0xffe007ff, RD_1|WR_CC|CM, 0, EE, 0, 0 },
+{"ctc2.i", "t,+9", 0x48c00001, 0xffe007ff, RD_1|WR_CC|CM, 0, EE, 0, 0 },
+{"ctc2.ni", "t,+9", 0x48c00000, 0xffe007ff, RD_1|WR_CC|CM, 0, EE, 0, 0 },
+{"dmfc2", "t,i", 0x48200000, 0xffe00000, WR_1|RD_C2|LC, 0, IOCT, 0, 0 },
+{"dmfc2", "t,G", 0x48200000, 0xffe007ff, WR_1|RD_C2|LC, 0, I3, 0, N54|IOCT|IOCTP|IOCT2|EE },
+{"dmfc2", "t,G,H", 0x48200000, 0xffe007f8, WR_1|RD_C2|LC, 0, I64, 0, IOCT|IOCTP|IOCT2 },
+{"dmtc2", "t,i", 0x48a00000, 0xffe00000, RD_1|WR_C2|WR_CC|CM, 0, IOCT, 0, 0 },
+{"dmtc2", "t,G", 0x48a00000, 0xffe007ff, RD_1|WR_C2|WR_CC|CM, 0, I3, 0, N54|IOCT|IOCTP|IOCT2|EE },
+{"dmtc2", "t,G,H", 0x48a00000, 0xffe007f8, RD_1|WR_C2|WR_CC|CM, 0, I64, 0, IOCT|IOCTP|IOCT2 },
+{"mfc2", "t,G", 0x48000000, 0xffe007ff, WR_1|RD_C2|LC, 0, I1, 0, N54|IOCT|IOCTP|IOCT2|EE },
+{"mfc2", "t,G,H", 0x48000000, 0xffe007f8, WR_1|RD_C2|LC, 0, I32, 0, IOCT|IOCTP|IOCT2 },
+{"mfhc2", "t,G", 0x48600000, 0xffe007ff, WR_1|RD_C2|LC, 0, I33, 0, IOCT|IOCTP|IOCT2 },
+{"mfhc2", "t,G,H", 0x48600000, 0xffe007f8, WR_1|RD_C2|LC, 0, I33, 0, IOCT|IOCTP|IOCT2 },
+{"mfhc2", "t,i", 0x48600000, 0xffe00000, WR_1|RD_C2|LC, 0, I33, 0, IOCT|IOCTP|IOCT2 },
+{"mtc2", "t,G", 0x48800000, 0xffe007ff, RD_1|WR_C2|WR_CC|CM, 0, I1, 0, N54|IOCT|IOCTP|IOCT2|EE },
+{"mtc2", "t,G,H", 0x48800000, 0xffe007f8, RD_1|WR_C2|WR_CC|CM, 0, I32, 0, IOCT|IOCTP|IOCT2 },
+{"mthc2", "t,G", 0x48e00000, 0xffe007ff, RD_1|WR_C2|WR_CC|CM, 0, I33, 0, IOCT|IOCTP|IOCT2 },
+{"mthc2", "t,G,H", 0x48e00000, 0xffe007f8, RD_1|WR_C2|WR_CC|CM, 0, I33, 0, IOCT|IOCTP|IOCT2 },
+{"mthc2", "t,i", 0x48e00000, 0xffe00000, RD_1|WR_C2|WR_CC|CM, 0, I33, 0, IOCT|IOCTP|IOCT2 },
+{"qmfc2", "t,+6", 0x48200000, 0xffe007ff, WR_1|RD_C2, 0, EE, 0, 0 },
+{"qmfc2.i", "t,+6", 0x48200001, 0xffe007ff, WR_1|RD_C2, 0, EE, 0, 0 },
+{"qmfc2.ni", "t,+6", 0x48200000, 0xffe007ff, WR_1|RD_C2, 0, EE, 0, 0 },
+{"qmtc2", "t,+6", 0x48a00000, 0xffe007ff, RD_1|WR_C2, 0, EE, 0, 0 },
+{"qmtc2.i", "t,+6", 0x48a00001, 0xffe007ff, RD_1|WR_C2, 0, EE, 0, 0 },
+{"qmtc2.ni", "t,+6", 0x48a00000, 0xffe007ff, RD_1|WR_C2, 0, EE, 0, 0 },
+
+/* Coprocessor 3 move/branch operations overlap with MIPS IV COP1X
+ instructions, so they are here for the latters to take precedence. */
+{"bc3f", "p", 0x4d000000, 0xffff0000, RD_CC|CBD, 0, I1, 0, I3_33|EE },
+{"bc3fl", "p", 0x4d020000, 0xffff0000, RD_CC|CBL, 0, I2|T3, 0, I3_33|EE },
+{"bc3t", "p", 0x4d010000, 0xffff0000, RD_CC|CBD, 0, I1, 0, I3_33|EE },
+{"bc3tl", "p", 0x4d030000, 0xffff0000, RD_CC|CBL, 0, I2|T3, 0, I3_33|EE },
+{"cfc3", "t,g", 0x4c400000, 0xffe007ff, WR_1|RD_C3|LC, 0, I1, 0, I3_33|EE },
+{"ctc3", "t,g", 0x4cc00000, 0xffe007ff, RD_1|WR_CC|CM, 0, I1, 0, I3_33|EE },
+{"mfc3", "t,G", 0x4c000000, 0xffe007ff, WR_1|RD_C3|LC, 0, I1, 0, I3_33|EE },
+{"mfc3", "t,G,H", 0x4c000000, 0xffe007f8, WR_1|RD_C3|LC, 0, I32, 0, I3_33|EE },
+{"mtc3", "t,G", 0x4c800000, 0xffe007ff, RD_1|WR_C3|WR_CC|CM, 0, I1, 0, I3_33|EE },
+{"mtc3", "t,G,H", 0x4c800000, 0xffe007f8, RD_1|WR_C3|WR_CC|CM, 0, I32, 0, I3_33|EE },
+
/* No hazard protection on coprocessor instructions--they shouldn't
change the state of the processor and if they do it's up to the
user to put in nops as necessary. These are at the end so that the
More information about the Binutils-cvs
mailing list