cmp[l][o] get an optional L field only when generating 32-bit code.
dcbf, tlbie and tlbiel keep their optional L field, ditto for R field
of tbegin. cmprb, tsr., wlcr[all] and mtsle all change to a
compulsory L field.
L field of dcbf and wclr is 2 bits.
PR 20641
include/
* opcode/ppc.h (PPC_OPERAND_OPTIONAL32): Define.
opcodes/
* ppc-opc.c (L): Make compulsory.
(LOPT): New, optional form of L.
(HTM_R): Define as LOPT.
(L0, L1): Delete.
(L32OPT): New, optional for 32-bit L.
(L2OPT): New, 2-bit L for dcbf.
(SVC_LEC): Update.
(L2): Define.
(insert_l0, extract_l0, insert_l1, extract_l2): Delete.
(powerpc_opcodes <cmpli, cmpi, cmpl, cmp>): Use L32OPT.
<dcbf>: Use L2OPT.
<tlbiel, tlbie>: Use LOPT.
<wclr, wclrall>: Use L2.
gas/
* config/tc-ppc.c (md_assemble): Handle PPC_OPERAND_OPTIONAL32.
* testsuite/gas/ppc/power8.s: Provide tbegin. operand.
* testsuite/gas/ppc/power9.d: Update cmprb disassembly.
This patch extends Tag_GNU_Power_ABI_FP to cover long double ABIs,
makes the assembler warn about undefined tag values, and removes
similar warnings from the linker. I think it is better to not
warn in the linker about undefined tag values as future extensions to
the tags then won't result in likely bogus warnings. This is
consistent with the fact that an older linker won't warn on an
entirely new tag.
include/
* elf/ppc.h (Tag_GNU_Power_ABI_FP): Comment.
bfd/
* elf-bfd.h (_bfd_elf_ppc_merge_fp_attributes): Declare.
* elf32-ppc.c (_bfd_elf_ppc_merge_fp_attributes): New function.
(ppc_elf_merge_obj_attributes): Use it. Don't copy first file
attributes, merge them. Don't warn about undefined tag bits,
or copy unknown values to output.
* elf64-ppc.c (ppc64_elf_merge_private_bfd_data): Call
_bfd_elf_ppc_merge_fp_attributes.
binutils/
* readelf.c (display_power_gnu_attribute): Catch truncated section
for all powerpc attributes. Display long double ABI. Don't
capitalize words, except for names. Show known bits of tag values
when some unknown bits are present. Whitespace fixes.
gas/
* config/tc-ppc.c (ppc_elf_gnu_attribute): New function.
(md_pseudo_table <ELF>): Handle "gnu_attribute".
ld/
* testsuite/ld-powerpc/attr-gnu-4-4.s: Delete.
* testsuite/ld-powerpc/attr-gnu-4-14.d: Delete.
* testsuite/ld-powerpc/attr-gnu-4-24.d: Delete.
* testsuite/ld-powerpc/attr-gnu-4-34.d: Delete.
* testsuite/ld-powerpc/attr-gnu-4-41.d: Delete.
* testsuite/ld-powerpc/attr-gnu-4-32.d: Adjust expected warning.
* testsuite/ld-powerpc/attr-gnu-8-23.d: Likewise.
* testsuite/ld-powerpc/attr-gnu-4-01.d: Adjust expected output.
* testsuite/ld-powerpc/attr-gnu-4-02.d: Likewise.
* testsuite/ld-powerpc/attr-gnu-4-03.d: Likewise.
* testsuite/ld-powerpc/attr-gnu-4-10.d: Likewise.
* testsuite/ld-powerpc/attr-gnu-4-11.d: Likewise.
* testsuite/ld-powerpc/attr-gnu-4-20.d: Likewise.
* testsuite/ld-powerpc/attr-gnu-4-22.d: Likewise.
* testsuite/ld-powerpc/attr-gnu-4-33.d: Likewise.
* testsuite/ld-powerpc/attr-gnu-8-11.d: Likewise.
* testsuite/ld-powerpc/powerpc.exp: Don't run deleted tests.
ELF section sh_flags SHF_PPC_VLE was being set based on arch/mach,
which meant all code sections in an object file has the flag or all
lacked it. We can do better than that. Only those code sections
where VLE is enabled ought to have the flag, allowing an object file
to contain both VLE and non-VLE code.
Also, ELF header p_flags PF_PPC_VLE wasn't being set, and segments
were being split unnecessarily.
bfd/
* elf32-ppc.c (ppc_elf_section_processing): Delete.
(elf_backend_section_processing): Don't define.
(ppc_elf_modify_segment_map): Set p_flags and mark valid. Don't
split on non-exec sections differing in SHF_PPC_VLE. When
splitting segments, mark size invalid.
gas/
* config/tc-ppc.c (md_assemble): Set sh_flags for VLE. Test
ppc_cpu rather than calling ppc_mach to determine VLE mode.
(ppc_frag_check, ppc_handle_align): Likewise use ppc_cpu.
VLE is an encoding, not a particular processor architecture, so it
isn't really proper to select insns based on PPC_OPCODE_VLE. For
example
{"evaddw", VX (4, 512), VX_MASK, PPCSPE|PPCVLE, PPCNONE, {RS, RA, RB}},
{"vaddubs", VX (4, 512), VX_MASK, PPCVEC|PPCVLE, PPCNONE, {VD, VA, VB}},
shows two insns that have the same encoding, both available with VLE.
Enabling both with VLE means we can't disassemble the second variant
even if -Maltivec is given rather than -Mspe. Also, we don't check
user assembly against the processor type as well as we could.
Another problem is that when using the VLE encoding, insns from the
main ppc opcode table are not available, except those using opcode 4
and 31. Correcting this revealed two errors in the ld testsuite,
use of "nop" and "rfmci" when -mvle.
This patch fixes those problems in the opcode table, and removes
PPCNONE. I find a plain 0 distracts less from other values.
In addition, I've implemented code to recognize some machine values
from the apuinfo note present in ppc32 objects. It's not a complete
disambiguation since we're lacking info to detect newer chips, but
what we have should help with disassembly.
include/
* elf/ppc.h (APUINFO_SECTION_NAME, APUINFO_LABEL, PPC_APUINFO_ISEL,
PPC_APUINFO_PMR, PPC_APUINFO_RFMCI, PPC_APUINFO_CACHELCK,
PPC_APUINFO_SPE, PPC_APUINFO_EFS, PPC_APUINFO_BRLOCK,
PPC_APUINFO_VLE: Define.
opcodes/
* ppc-dis.c (ppc_opts): Delete extraneous parentheses. Default
cpu for "vle" to e500.
* ppc-opc.c (ALLOW8_SPRG): Remove PPC_OPCODE_VLE.
(NO371, PPCSPE, PPCISEL, PPCEFS, MULHW, DCBT_EO): Likewise.
(PPCNONE): Delete, substitute throughout.
(powerpc_opcodes): Remove PPCVLE from "flags". Add to "deprecated"
except for major opcode 4 and 31.
(vle_opcodes <se_rfmci>): Add PPCRFMCI to flags.
bfd/
* cpu-powerpc.c (powerpc_compatible): Allow bfd_mach_ppc_vle entry
to match other 32-bit archs.
* elf32-ppc.c (_bfd_elf_ppc_set_arch): New function.
(ppc_elf_object_p): Call it.
(ppc_elf_special_sections): Use APUINFO_SECTION_NAME. Fix
overlong line.
(APUINFO_SECTION_NAME, APUINFO_LABEL): Don't define here.
* elf64-ppc.c (ppc64_elf_object_p): Call _bfd_elf_ppc_set_arch.
* bfd-in.h (_bfd_elf_ppc_at_tls_transform,
_bfd_elf_ppc_at_tprel_transform): Move to..
* elf-bfd.h: ..here.
(_bfd_elf_ppc_set_arch): Declare.
* bfd-in2.h: Regenerate.
gas/
* config/tc-ppc.c (PPC_APUINFO_ISEL, PPC_APUINFO_PMR,
PPC_APUINFO_RFMCI, PPC_APUINFO_CACHELCK, PPC_APUINFO_SPE,
PPC_APUINFO_EFS, PPC_APUINFO_BRLOCK, PPC_APUINFO_VLE): Don't define.
(ppc_setup_opcodes): Check vle disables powerpc_opcodes overridden
by vle_opcodes, and that vle flag doesn't enable opcodes. Don't
add vle_opcodes twice.
(ppc_cleanup): Use APUINFO_SECTION_NAME and APUINFO_LABEL.
ld/
* testsuite/ld-powerpc/apuinfo1.s: Delete nop.
* testsuite/ld-powerpc/apuinfo-vle2.s: New.
* testsuite/ld-powerpc/powerpc.exp: Use apuinfo-vle2.s.
Commit b84bf58a accidentally extended the range of allowed negative
numbers.
* config/tc-ppc.c (ppc_insert_operand): Trim PPC_OPERAND_SIGNOPT
allowed negative range.
* testsuite/gas/ppc/power9.s: Test xxspltib of -128, not -256.
* testsuite/gas/ppc/power9.d: Update.
They are only used in one file, so we might as well restrict there scope to
that file, and theoretically this might slightly improve compilers ability to
optimize usage of these variables.
gas/ChangeLog:
2016-04-14 Trevor Saunders <tbsaunde+binutils@tbsaunde.org>
* config/tc-nios2.c (nios2_as_options): Make file static.
* config/tc-ppc.c (toc_reloc_ypes): Likewise.
* config/tc-sparc.c (native_op_table): Likewise.
Back in the day support for these processors was added, we probably
didn't want to waste PPC_OPCODE bits on minor variations. I've had a
complaint that disassembly of mfspr/mtspr was wrong for power8. This
patch fixes that problem.
Note that since -m860/-m850/-m821 are new gas options enabling the
mpc8xx specific mfspr/mtspr variants it is possible that this change
will break some mpc8xx assembly code. ie. you might need to modify
makefiles to pass -m860 to gas.
include/opcode/
* ppc.h (PPC_OPCODE_750, PPC_OPCODE_7450, PPC_OPCODE_860): Define.
opcodes/
* ppc-opc.c (PPC750, PPC7450, PPC860): Define using PPC_OPCODE_*.
* ppc-dis.c (ppc_opts): Add 821, 850 and 860 entries. Add
PPC_OPCODE_7450 to 7450 entry. Add PPC_OPCODE_750 to 750cl entry.
gas/
* config/tc-ppc.c (md_show_usage): Add -m821, -m850, -m860.
* doc/c-ppc.texi (PowerPC-Opts): Likewise.
gas/testsuite/
* gas/ppc/titan.d: Correct mfmcsrr0 disassembly.
ISA 2.07 (ie, POWER8) added the rfebb instruction which takes one operand
with the value of either a 0 or 1. It also defines an extended mnemonic
with no operands (ie, "rfebb") that is supposed to be equivalent to "rfebb 1".
I implemented rfebb's lone operand with PPC_OPERAND_OPTIONAL, but the
problem is, optional operands that are ommitted always default to the
value 0, which is wrong in this case. I have added support for allowing
non-zero default values by adding an additional flag PPC_OPERAND_OPTIONAL_VALUE
that specifies that the default operand value to be used is stored in the
SHIFT field of the operand field immediately following this one.
This fixes the rfebb issue. I also fixed the mftb and mfcr instructions
so they use the same mechanism. This allows us to flag invalid uses of
mfcr where we explicitly pass in a zero FXM value, like the use in a2.[sd].
include/opcode/
* ppc.h (PPC_OPERAND_OPTIONAL_VALUE): New.
(ppc_optional_operand_value): New inline function.
opcodes/
* ppc-dis.h (skip_optional_operands): Use ppc_optional_operand_value.
* ppc-opc.c (FXM4): Add non-zero optional value.
(TBR): Likewise.
(SXL): Likewise.
(insert_fxm): Handle new default operand value.
(extract_fxm): Likewise.
(insert_tbr): Likewise.
(extract_tbr): Likewise.
gas/
* config/tc-ppc.c (md_assemble): Use ppc_optional_operand_value.
Allow for optional operands without insert functions.
gas/testsuite/
* gas/ppc/power8.d: Fixup rfebb test results.
* gas/ppc/a2.s: Fix invalid mfcr test.
* gas/ppc/a2.d: Likewise.
Only set the VLE flag if the instruction has been pulled via the VLE
instruction set. This way the flag is guaranteed to be set for VLE-only
instructions or for VLE-only processors, however it'll remain clear for
dual-mode instructions on dual-mode and, more importantly, standard-mode
processors.
gas/
* config/tc-ppc.c (md_assemble): Only set the PPC_APUINFO_VLE
flag if both the processor and opcode flags match.
ld/testsuite/
* ld-powerpc/apuinfo-vle.rd: New test.
* ld-powerpc/apuinfo-vle.s: New test source.
* ld-powerpc/apuinfo.rd: Adjust according to GAS PPC_APUINFO_VLE
handling change.
* ld-powerpc/powerpc.exp: Run the new test.
There are legitimate reasons to allow a signed value in a cmpli insn
field, for example to test for a "stw r1,lock@sdarel(r13)" instruction
in user code, a kernel might use
subis r3,r3,STW_R1_0R13@ha # subtract off high part
cmplwi r3,lock@sdarel # is low part accessing lock?
Since the lock@sdarel may take a range of -32768 to 32767,
the allowed range of cmpli immediate must be at least [-32768,65535].
bfd/
* elf32-ppc.c (ppc_elf_relocate_section): Treat field of cmpli
insn as a bitfield; Use complain_overflow_bitfield.
* elf64-ppc.c (ppc64_elf_relocate_section): Likewise.
opcodes/
* ppc-opc.c (UISIGNOPT): Define and use with cmpli.
gas/
* config/tc-ppc.c (ppc_insert_operand): Handle PPC_OPERAND_SIGNOPT
on unsigned fields. Comment on PPC_OPERAND_SIGNOPT signed fields
in 64-bit mode.
gold/
* powerpc.cc (relocate): Treat field of cmpli insn as a bitfield.
It turns out that glibc's sysdeps/powerpc/powerpc64/start.S uses this
feature. :-(
* config/tc-ppc.c (ppc_is_toc_sym): Revert 2014-03-05.
(md_assemble): Likewise. Warn.
This adds support for "func@localentry", an expression that returns the
ELFv2 local entry point address of function "func". I've excluded
dynamic relocation support because that obviously would require glibc
changes.
include/elf/
* ppc64.h (R_PPC64_REL24_NOTOC, R_PPC64_ADDR64_LOCAL): Define.
bfd/
* elf64-ppc.c (ppc64_elf_howto_raw): Add R_PPC64_ADDR64_LOCAL entry.
(ppc64_elf_reloc_type_lookup): Support R_PPC64_ADDR64_LOCAL.
(ppc64_elf_check_relocs): Likewise.
(ppc64_elf_relocate_section): Likewise.
* Add BFD_RELOC_PPC64_ADDR64_LOCAL.
* bfd-in2.h: Regenerate.
* libbfd.h: Regenerate.
gas/
* config/tc-ppc.c (ppc_elf_suffix): Support @localentry.
(md_apply_fix): Support R_PPC64_ADDR64_LOCAL.
ld/testsuite/
* ld-powerpc/elfv2-2a.s, ld-powerpc/elfv2-2b.s: New files.
* ld-powerpc/elfv2-2exe.d, ld-powerpc/elfv2-2so.d: New files.
* ld-powerpc/powerpc.exp: Run new test.
elfcpp/
* powerpc.h (R_PPC64_REL24_NOTOC, R_PPC64_ADDR64_LOCAL): Define.
gold/
* powerpc.cc (Target_powerpc::Scan::local, global): Support
R_PPC64_ADDR64_LOCAL.
(Target_powerpc::Relocate::relocate): Likewise.
This patch allows gas to assemble a testcase like
li 3,ext_sym
which oddly was not accepted while the following is OK:
li 3,ext_sym@l
* config/tc-ppc.c (md_assemble): Move code adjusting reloc types
later. Merge absolute and relative branch reloc selection.
Generate 16-bit relocs for most 16-bit insn fields given a
non-constant expression.
The XCOFF assembler does some wierd things with instructions like
`lwz 9,sym(30'. See the comment in md_apply_fix. From an ELF
perspective, it's weird even to magically select a TOC16 reloc
when a symbol is in the TOC/GOT. ELF assemblers generally use
modifiers like @toc to select relocs, so remove this "feature"
for ELF. I believe this was to support gcc -m32 -mcall-aixdesc
but that combination of gcc options has been broken for a long
time.
* config/tc-ppc.c (ppc_is_toc_sym): Remove OBJ_ELF support.
(md_assemble): Don't call ppc_is_toc_sym for ELF.
This defines the ELF symbol st_other field used to encode the number
of instructions between a function "global entry" and its "local entry",
and adds support related to the local entry offset.
include/elf/
* ppc64.h (STO_PPC64_LOCAL_BIT, STO_PPC64_LOCAL_MASK): Define.
(ppc64_decode_local_entry, ppc64_encode_local_entry): New functions.
(PPC64_LOCAL_ENTRY_OFFSET, PPC64_SET_LOCAL_ENTRY_OFFSET): Define.
bfd/
* elf64-ppc.c (struct ppc_stub_hash_entry): Add "other".
(stub_hash_newfunc): Init new ppc_stub_hash_entry field, and one
we forgot, "plt_ent".
(ppc64_elf_add_symbol_hook): Check ELFv1 objects don't have
st_other bits only valid in ELFv2.
(ppc64_elf_merge_symbol_attribute): New function.
(ppc_type_of_stub): Add local_off param to test branch range.
(ppc_build_one_stub): Adjust destinations for ELFv2 locals.
(ppc_size_one_stub, toc_adjusting_stub_needed): Similarly.
(ppc64_elf_size_stubs): Pass local_off to ppc_type_of_stub.
Set "other" field.
(ppc64_elf_relocate_section): Adjust destination for ELFv2 local
calls.
gas/
* config/tc-ppc.c (md_pseudo_table): Add .localentry.
(ppc_elf_localentry): New function.
(ppc_force_relocation): Force relocs on all branches to localenty
symbols.
(ppc_fix_adjustable): Don't reduce such symbols to section+offset.
binutils/
* readelf.c (get_ppc64_symbol_other): New function.
(get_symbol_other): Use it for EM_PPC64.
Defines bits in ELF e_flags to differentiate ELFv2 objects from ELFv2,
adds .abiversion directive to explicitly choose the ABI, and code to
check and automatically select ABI.
include/elf/
* ppc64.h (EF_PPC64_ABI): Define.
bfd/
* elf64-ppc.c (abiversion, set_abiversion): New functions.
(ppc64_elf_get_synthetic_symtab): Handle ELFv2 objects without .opd.
(struct ppc_link_hash_table): Add opd_abi.
(ppc64_elf_check_relocs): Check no .opd with ELFv2.
(ppc64_elf_merge_private_bfd_data): New function.
(ppc64_elf_print_private_bfd_data): New function.
(ppc64_elf_tls_setup): Set htab->opd_abi.
(ppc64_elf_size_dynamic_sections): Don't emit OPD related dynamic
tags for ELFv2.
(ppc_build_one_stub): Use R_PPC64_IRELATIVE for ELFv2 ifunc.
(ppc64_elf_finish_dynamic_symbol): Likewise
binutils/
* readelf.c (get_machine_flags): Display ABI version for EM_PPC64.
gas/
* config/tc-ppc.c: Include elf/ppc64.h.
(ppc_abiversion): New variable.
(md_pseudo_table): Add .abiversion.
(ppc_elf_abiversion, ppc_elf_end): New functions.
* config/tc-ppc.h (md_end): Define.
This changes the behaviour of @h and @ha on PowerPC64 to report errors
on 32-bit overflow. The motivation for this change is that on
PowerPC64, most uses of @h and @ha modifiers and their corresponding
relocations are to build up 32-bit offsets. We'd like to know when
such offsets overflow. Only rarely do people use @h or @ha with the
high 32-bit modifiers to build a 64-bit constant. Those uses will now
need to use two new modifiers, @high and @higha, if the constant isn't
known at assembly time. For now, we won't report overflow at assembly
time..
This also fixes an error when applying some of the HIGHER and HIGHEST
relocations.
include/elf/
* ppc64.h (R_PPC64_ADDR16_HIGH, R_PPC64_ADDR16_HIGHA,
R_PPC64_TPREL16_HIGH, R_PPC64_TPREL16_HIGHA,
R_PPC64_DTPREL16_HIGH, R_PPC64_DTPREL16_HIGHA): New.
(IS_PPC64_TLS_RELOC): Match new tls relocs.
bfd/
* reloc.c (BFD_RELOC_PPC64_ADDR16_HIGH, BFD_RELOC_PPC64_ADDR16_HIGHA,
BFD_RELOC_PPC64_TPREL16_HIGH, BFD_RELOC_PPC64_TPREL16_HIGHA,
BFD_RELOC_PPC64_DTPREL16_HIGH, BFD_RELOC_PPC64_DTPREL16_HIGHA): New.
* elf64-ppc.c (ppc64_elf_howto_raw): Add entries for new relocs.
Make all _HA and _HI relocs report signed overflow.
(ppc64_elf_reloc_type_lookup): Handle new relocs.
(must_be_dyn_reloc, ppc64_elf_check_relocs): Likewise.
(dec_dynrel_count, ppc64_elf_relocate_section): Likewise.
(ppc64_elf_relocate_section): Don't apply 0x8000 adjust to
R_PPC64_TPREL16_HIGHER, R_PPC64_TPREL16_HIGHEST,
R_PPC64_DTPREL16_HIGHER, and R_PPC64_DTPREL16_HIGHEST.
* libbfd.h: Regenerate.
* bfd-in2.h: Regenerate.
gas/
* config/tc-ppc.c (SEX16): Don't mask.
(REPORT_OVERFLOW_HI): Define as zero.
(ppc_elf_suffix): Support @high, @higha, @dtprel@high, @dtprel@higha,
@tprel@high, and @tprel@higha modifiers.
(md_assemble): Ignore X_unsigned when applying 16-bit insn fields.
Add (disabled) code to check @h and @ha reloc overflow for powerpc64.
Handle new relocs.
(md_apply_fix): Similarly.
elfcpp/
* powerpc.h (R_PPC64_ADDR16_HIGH, R_PPC64_ADDR16_HIGHA,
R_PPC64_TPREL16_HIGH, R_PPC64_TPREL16_HIGHA,
R_PPC64_DTPREL16_HIGH, R_PPC64_DTPREL16_HIGHA): Define.
gold/
* powerpc.cc (Target_powerpc::Scan::check_non_pic): Handle new relocs.
(Target_powerpc::Scan::global, local): Likewise.
(Target_powerpc::Relocate::relocate): Likewise. Check for overflow
on all ppc64 @h and @ha relocs.
* config/tc-ppc.c (md_apply_fix): Handle defined after use toc
symbols.
gas/testsuite/
* gas/ppc/aix.exp: Run xcoff-toc-1 test.
* gas/ppc/xcoff-toc-1.s, gas/ppc/xcoff-toc-1.d: New test.
* elf64-ppc.c (ppc64_elf_relocate_section): Set "relocation" for
.TOC. after relocatable check.
gas/
* config/tc-ppc.c (ppc_elf_adjust_symtab): Don't make .TOC. weak.
and data fixups performing shift/high adjust/sign extension on
fieldval. Sink fx_pcrel handling and checks. Use fixP->fx_size
when writing data fixups rather than recalculating size.
