When the VEX.L=1 decode matches that of both EVEX.L'L=1 and EVEX.L'L=2
(typically when all three are invalid) the (smaller) VEX table entry can
be reused by EVEX, instead of duplicating data. (Note that XM and XMM as
well as EXxmm_md and EXd are equivalent at least for the purposes here.)
The order of decodes influences the overall number of table entries.
Reduce table size quite a bit by first decoding few-alternatives
attributes common to all valid leaves.
This also adds a PREFIX_DATA 7531c61332 ("x86: simplify decode of
opcodes valid with (embedded) 66 prefix only") missed to apply to
vbroadcastf64x4.
The order of decodes influences the overall number of table entries.
Reduce table size quite a bit by first decoding few-alternatives
attributes common to all valid leaves.
The order of decodes influences the overall number of table entries.
Reduce table size quite a bit by first decoding few-alternatives
attributes common to all valid leaves.
The order of decodes influences the overall number of table entries.
Reduce table size quite a bit by first decoding few-alternatives
attributes common to all valid leaves.
All encodings not used in this range are (reserved) NOPs. Hence their
decoding should be fully consistent. For this to work the PREFIX_IGNORED
logic needs slightly extending, such that the attribute will also
- have an effect when used inside prefix_table[], yet without always
falling back to using slot 0,
- cause prefixes marked as ignored while decoding through prefix_table[]
to no longer be considered decoded, when encountered in a subsequent
decoding step.
In adjacent code also drop meaningless PREFIX_OPCODE.
The offset-within-section field in the symbol table entry is only 32
bits wide, so rather than emitting bogus entries omit them, and issue
a diagnostic identifying the issue.
This requires adjusting the PR/22267 test to no longer produce symbols
with out of range values on 64-bit BFD. This also depends on
adjustments to testsuite/ld-scripts/map-address.* made by an earlier
patch. The purpose of the test can very well be achieved nevertheless.
The internal AdaCore test suite revealed a bug in the new evaluator.
A hunk of evaluate_funcall was not correctly transcribed. This was
not caught in my original testing because the feature in question was
apparently not tested in gdb.
This patch fixes the oversight. The idea here is that ordinary
function calls should use the function's formal parameter types as the
expected types of subexpressions.
Regression tested on x86-64 Fedora 32.
gdb/ChangeLog
2021-03-09 Tom Tromey <tromey@adacore.com>
* eval.c (operation::evaluate_funcall): Use function formal
parameter types when evaluating.
gdb/testsuite/ChangeLog
2021-03-09 Tom Tromey <tromey@adacore.com>
* gdb.base/cast-call.exp: New file.
* gdb.base/cast-call.c: New file.
mcore-pe and sh-pe pointed out that the testcase added by 6fa7408d72
("ld: don't generate base relocations in PE output for absolute
symbols") wasn't quite generic enough yet. Copy entry point stuff from
another test and increase alignment to also cover targets with a page
size lower than 4k.
Despite all of this the test needs to be xfail-ed for mcore, as this
target isn't included in the set getting DLL_SUPPORT enabled in
emultempl/pe.em, and hence no base relocations get generated there
(yet).
In commit:
commit 5b7d941b90
Date: Fri Jan 22 12:21:09 2021 -0500
gdb: add owner-related methods to struct type
two fields of struct maint_type were renamed. 'flag_objfile_owned'
became 'm_flag_objfile_owned' and 'owner' became 'm_owner'. Update
our python helper script to take this into account.
I've added a basic test that uses the self-test framework to load the
pretty printers, and print a type.
The test relies on stopping in GDB's `value_print` function.
gdb/ChangeLog:
* gdb-gdb.py.in (StructMainTypePrettyPrinter) <owner_to_string>:
Updated fields names flag_objfile_owned to m_flag_objfile_owned,
and owner to m_owner.
gdb/testsuite/ChangeLog:
* gdb.gdb/python-helper.exp: New file.
LOC(X) returns the address of X as an integer:
https://gcc.gnu.org/onlinedocs/gfortran/LOC.html
Before:
(gdb) p LOC(r)
No symbol "LOC" in current context.
After:
(gdb) p LOC(r)
$1 = 0xffffdf48
gdb/ChangeLog:
2021-03-09 Felix Willgerodt <felix.willgerodt@intel.com>
* f-exp.h (eval_op_f_loc): Declare.
(expr::fortran_loc_operation): New typedef.
* f-exp.y (exp): Handle UNOP_FORTRAN_LOC after parsing an
UNOP_INTRINSIC.
(f77_keywords): Add LOC keyword.
* f-lang.c (eval_op_f_loc): New function.
* std-operator.def (UNOP_FORTRAN_LOC): New operator.
gdb/testsuite/ChangeLog:
2020-03-09 Felix Willgerodt <felix.willgerodt@intel.com>
* gdb.fortran/intrinsics.exp: Add LOC tests.
Add support for the SHAPE keyword to GDB's Fortran expression parser.
gdb/ChangeLog:
* f-exp.h (eval_op_f_array_shape): Declare.
(fortran_array_shape_operation): New type.
* f-exp.y (exp): Handle UNOP_FORTRAN_SHAPE after parsing
UNOP_INTRINSIC.
(f77_keywords): Add "shape" keyword.
* f-lang.c (fortran_array_shape): New function.
(eval_op_f_array_shape): New function.
* std-operator.def (UNOP_FORTRAN_SHAPE): New operator.
gdb/testsuite/ChangeLog:
* gdb.fortran/shape.exp: New file.
* gdb.fortran/shape.f90: New file.
Add support for the 'SIZE' keyword to the Fortran expression parser.
This returns the number of elements either in an entire array (passing
a single argument to SIZE), or in a particular dimension of an
array (passing two arguments to SIZE).
At this point I have not added support for the optional third argument
to SIZE, which controls the exact integer type of the result.
gdb/ChangeLog:
* f-exp.y (eval_op_f_array_size): Declare 1 and 2 argument forms
of this function.
(expr::fortran_array_size_1arg): New type.
(expr::fortran_array_size_2arg): Likewise.
* f-exp.y (exp): Handle FORTRAN_ARRAY_SIZE after parsing
UNOP_OR_BINOP_INTRINSIC.
(f77_keywords): Add "size" keyword.
* f-lang.c (fortran_array_size): New function.
(eval_op_f_array_size): New function, has a 1 arg and 2 arg form.
* std-operator.def (FORTRAN_ARRAY_SIZE): New operator.
gdb/testsuite/ChangeLog:
* gdb.fortran/size.exp: New file.
* gdb.fortran/size.f90: New file.
gfortran supports the RANK keyword, see:
https://gcc.gnu.org/onlinedocs/gfortran/RANK.html#RANK
this commit adds support for this keyword to GDB's Fortran expression
parser.
gdb/ChangeLog:
* f-exp.h (eval_op_f_rank): Declare.
(expr::fortran_rank_operation): New typedef.
* f-exp.y (exp): Handle UNOP_FORTRAN_RANK after parsing an
UNOP_INTRINSIC.
(f77_keywords): Add "rank" keyword.
* f-lang.c (eval_op_f_rank): New function.
* std-operator.def (UNOP_FORTRAN_RANK): New operator.
gdb/testsuite/ChangeLog:
* gdb.fortran/rank.exp: New file.
* gdb.fortran/rank.f90: New file.
RepPrefixOk, HLEPrefixOk, and NoTrackPrefixOk can't be specified
together, so can share an enum-like field. IsLockable can be inferred
from HLE setting and hence only needs specifying when neither of them
is present.
Despite SYSEXIT being an Intel-only insn in long mode, its behavior
there is similar to SYSRET's: Depending on REX.W execution continues in
either 64-bit or compatibility mode. Hence distinguishing by suffix is
as necessary here as it is there.
In PE images section addresses get expressed as addresses relative to
the image base. Therefore the VA of a section must be no less than the
image base, and after subtraction of the image base the resulting value
should fit in 32 bits. (The issue is particularly obvious to notice when
sections, perhaps because of ELF assumptions, get placed at VA 0 by
default. Debugging info sections as well as .comment, when input files
are ELF, are a good example. All such sections need proper mentioning in
the linker script to avoid this warning.)
There are a number of test cases which previously produced bogus images,
yet still declared the test a success. Like done for other tests
already, force a zero image base for these. This then also allows (and
requires) dropping again xfail-s which 39a7b38fac ("Fix linker tests
to work with 16-bit targets") had added to ld-scripts/default-script*.d
(originally as skip-s). This also depends on similar adjustments to
testsuite/ld-scripts/map-address.* made by an earlier patch.
For ld-scripts/print-memory-usage.* I suppose xcoff could be dropped
from the exclusion list by suppressing garbage collection, just like
already done in e.g. (as seen in the diff here) ld-scripts/data.*, but I
didn't want to make unrelated adjustments.
The sim's recently switch to using Automake caused a build failure for
me, because I didn't have the correct auto* tools in my path.
However, the rule in the tree is that this is not needed in general.
This patch adds a call to AM_MAINTAINER_MODE, to align the sim with
the way the rest of the tree works here.
sim/ChangeLog
2021-03-08 Tom Tromey <tromey@adacore.com>
* aclocal.m4, configure, Makefile.in: Rebuild.
* configure.ac: Use AM_MAINTAINER_MODE.
When not parsing for completion, parse_expression ensures that the
resulting expression has operations. This patch removes a couple of
unnecessary checks for this situation.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* printcmd.c (set_command): Remove null check.
* value.c (init_if_undefined_command): Remove null check.
This changes var_msym_value_operation to use a bound_minimal_symbol
rather than separate minsym and objfile parameters. The main benefit
of this is removing the possibly-confusing check_objfile overload for
a plain minimal symbol.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* parse.c (parser_state::push_symbol, parser_state::push_dollar):
Update.
* p-exp.y (variable): Update.
* go-exp.y (variable): Update.
* expprint.c (dump_for_expression): Use bound_minimal_symbol.
Remove overload for objfile.
* expop.h (eval_op_var_msym_value): Use bound_minimal_symbol
parameter.
(check_objfile): Likewise.
(dump_for_expression): Likewise. Remove overload for objfile.
(class var_msym_value_operation): Use bound_minimal_symbol.
* eval.c (eval_op_var_msym_value): Use bound_minimal_symbol
parameter.
(var_msym_value_operation::evaluate_for_address)
(var_msym_value_operation::evaluate_for_sizeof)
(var_msym_value_operation::evaluate_for_cast): Update.
* d-exp.y (PrimaryExpression): Update.
* c-exp.y (variable): Update.
* ax-gdb.c (var_msym_value_operation::do_generate_ax): Update.
* ada-lang.c (ada_var_msym_value_operation::evaluate_for_cast):
Update.
* ada-exp.y (write_var_or_type): Update.
The big deletion patch left some identical namespace scopes sitting
next to one another. This patch removes the redundant "namespace
expr"s.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* eval.c: Merge "namespace" scopes.
This inlines the expr_builder constructor and release method. These
are straightforward, so this seemed simpler.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* parser-defs.h (struct expr_builder) <expr_builder>: Inline.
<release>: Inline.
* parse.c (expr_builder::expr_builder, expr_builder::release):
Remove.
The struct expression constructor no longer does any real work, so
this inlines it. The default destructor can also be used now as well.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* parse.c (expression::expression, expression::~expression):
Remove.
* expression.h (struct expression): Inline constructor. Remove
destructor.
BINOP_END is used only as a "meaningless" value in various tables.
This patch changes these to use OP_NULL instead, and removes
BINOP_END.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* std-operator.def (BINOP_END): Remove.
* p-exp.y (tokentab3, tokentab2): Use OP_NULL, not BINOP_END.
* go-exp.y (tokentab2): Use OP_NULL, not BINOP_END.
* f-exp.y (dot_ops, f77_keywords): Use OP_NULL, not BINOP_END.
* d-exp.y (tokentab2, ident_tokens): Use OP_NULL, not BINOP_END.
* c-exp.y (tokentab3, tokentab2, ident_tokens): Use OP_NULL, not
BINOP_END.
OP_UNUSED_LAST is no longer needed with C++ -- the trailing comma is
fine.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* expression.h (enum exp_opcode) <OP_UNUSED_LAST>: Remove.
OP_EXTENDED0 was only used for an assertion in the code to rewrite an
expression into prefix form. That code is gone, so this patch removes
the constant.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* std-operator.def (OP_EXTENDED0): Remove.
Several Ada expression opcodes are now unused, and can be removed.
Most of these are handled in a different way by the code.
OP_ATR_IMAGE, however, was never implemented.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* std-operator.def (OP_NAME, OP_ATR_IMAGE, OP_ATR_MODULUS)
(OP_OTHERS, OP_CHOICES, OP_POSITIONAL, OP_DISCRETE_RANGE):
Remove.
As noted in an earlier patch, Modula-2 defined some opcodes but then
never implemented them. This patch removes the unnecessary constants.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* std-operator.def (UNOP_CAP, UNOP_CHR, UNOP_ORD, UNOP_FLOAT)
(UNOP_MAX, UNOP_MIN, UNOP_ODD, UNOP_TRUNC, OP_M2_STRING): Remove.
The OP_ATR_MIN and OP_ATR_MAX constants aren't truly needed.
Internally, they are converted to BINOP_MIN and BINOP_MAX. This patch
removes them in favor of simple reuse.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* std-operator.def (OP_ATR_MIN, OP_ATR_MAX): Remove.
* ada-lang.c (ada_binop_minmax): Update.
* ada-exp.h (ada_binop_min_operation, ada_binop_max_operation):
Use BINOP_MIN and BINOP_MAX.
Now that the C parser has switched to the new style, there is no need
for the old C evaluation code. This affects some other languages that
were relying on the C code.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* go-lang.c (go_language::expression_ops): Don't declare.
* go-lang.h (class go_language) <expression_ops>: Remove.
* opencl-lang.c (evaluate_subexp_opencl, exp_descriptor_opencl):
Remove.
(class opencl_language) <expression_ops>: Remove.
* d-lang.c (class d_language) <expression_ops>: Remove.
* c-lang.h (evaluate_subexp_c, exp_descriptor_c): Don't declare.
* c-lang.c (evaluate_subexp_c, exp_descriptor_c): Remove.
(class c_language, class cplus_language, class asm_language)
(class minimal_language) <expression_ops>: Remove.
Now that the Ada parser has switched to the new style, there is no
need for the old Ada evaluation code.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* ada-lang.c (resolve_subexp, replace_operator_with_call)
(evaluate_subexp_type, assign_aggregate)
(aggregate_assign_positional, aggregate_assign_from_choices)
(aggregate_assign_others, ada_evaluate_subexp_for_cast)
(ada_evaluate_subexp, ADA_OPERATORS, ada_operator_length)
(ada_operator_check, ada_forward_operator_length)
(ada_dump_subexp_body, ada_print_subexp, ada_exp_descriptor):
Remove.
(post_parser): Update.
(class ada_language) <expresssion_ops>: Remove.
Now that the Modula-2 parser has switched to the new style, there is
no need for the old Modula-2 evaluation code.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* m2-lang.h (class m2_language) <expresssion_ops,
exp_descriptor_modula2>: Remove.
* m2-lang.c (evaluate_subexp_modula2)
(m2_language::exp_descriptor_modula2): Remove.
Now that the Fortran parser has switched to the new style, there is no
need for the old Fortran evaluation code.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* f-lang.h (class f_language) <expresssion_ops>: Remove.
<exp_descriptor_tab>: Remove.
* f-lang.c (fortran_value_subarray, evaluate_subexp_f)
(operator_length_f, print_unop_subexp_f, print_binop_subexp_f)
(print_subexp_f, dump_subexp_body_f, operator_check_f)
(f_language::exp_descriptor_tab, fortran_prepare_argument):
Remove.
Now that the Rust parser has switched to the new style, there is no
need for the old Rust evaluation code.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* rust-lang.h (class rust_language) <expression_ops,
exp_descriptor_tab>: Remove.
* rust-lang.c (rust_evaluate_funcall): Remove.
(rust_range, rust_subscript, eval_op_rust_complement): Don't use
EVAL_SKIP.
(rust_evaluate_subexp): Remove.
(rust_aggregate_operation::evaluate): Don't use EVAL_SKIP.
(rust_operator_length, rust_dump_subexp_body, rust_print_subexp)
(rust_operator_check, rust_language::exp_descriptor_tab): Remove.
This converts the Ada parser to generate operations rather than
exp_elements.
This was the most difficult of the parser conversions, partly due to
the decision to integrate Ada expression resolution into the parse,
and partly due to Ada aggregregate assignment. A couple of new
per-parse globals are introduced, along with a number of helper
functions. Resolution is done in 'ada_pop', yielding the unfortunate
rule that ada-exp.y should generally not use parser_state::pop
(exceptions are marked).
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* ada-exp.y: Create operations.
(empty_stoken): Remove.
(ada_pop, ada_wrap, ada_addrof, ada_un_wrap2, ada_wrap2)
(ada_wrap_op, ada_wrap3, ada_funcall): New functions.
(components): New global.
(push_component, choice_component, pop_component, pop_components):
New functions.
(associations): New global
(push_association, pop_association, pop_associations): New
functions.
(ada_parse): Update.
(write_var_from_sym, write_int): Create operations.
(write_exp_op_with_string): Remove.
(write_object_renaming, write_selectors, write_ambiguous_var)
(write_var_or_type, write_name_assoc): Create operations.
* ada-lang.h (ada_index_type): Declare.
* ada-lang.c (ada_index_type): No longer static.
This converts the Fortran parser to generate operations rather than
exp_elements. A couple of tests of expression debug dumping are
updated to follow the new output.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* f-exp.y: Create operations.
(f_language::parser): Update.
gdb/testsuite/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* gdb.fortran/debug-expr.exp: Update tests.
This converts the Modula-2 parser to generate operations rather than
exp_elements.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* m2-exp.y: Create operations.
(m2_language::parser): Update.
This converts the Pascal parser to generate operations rather than
exp_elements.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* p-exp.y: Create operations.
(pascal_language::parser): Update.
This converts the D parser to generate operations rather than
exp_elements.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* d-exp.y: Create operations.
(d_parse): Update.
This converts the Go parser to generate operations rather than
exp_elements.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* go-exp.y: Create operations.
(go_language::parser): Update.
This converts the C parser to generate operations rather than
exp_elements.
One test needed a tweak in order to handle the different debugging
output.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* objc-lang.c (end_msglist): Create operations.
* c-exp.y: Change parser to create operations.
(write_destructor_name): Remove.
(c_parse): Update.
gdb/testsuite/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* gdb.base/debug-expr.exp: Update expected dump output.
This converts the Rust parser to generate operations rather than
exp_elements.
The Rust parser already made its own AST, that it then lowered to GDB
expressions. Ironically, this made conversion trickier, rather than
simpler, than the other parsers, primarily due to the way that binary
operations were lowered. Perhaps in the future, converting the Rust
parser to directly create operations while parsing would be
worthwhile.
gdb/ChangeLog
2021-03-08 Tom Tromey <tom@tromey.com>
* rust-exp.y: Create operations.
(rust_parser::convert_params_to_expression): Change return type.
(binop_maker_ftype): New typedef.
(maker_map): New global.
(rust_parser::convert_ast_to_expression): Change return type.
(rust_language::parser): Update.
(_initialize_rust_exp): Initialize maker_map.
