Indeed, this doesn't make sense, as EA_REGNUM is greater than BA_REGNUM.
I'll assume that it was just a mistake and that these two should be
swapped.
The regnums for BA and EA are contiguous, so ultimately this particular
part of the condition is only true if regnum is == EA or == BA. These
registers are Exception Address and Breakpoint Address, so I guess it
makes sense for them to be in the system register group.
Simon Marchi [Thu, 21 May 2020 17:12:29 +0000 (13:12 -0400)]
gdb: remove unnecessary NULL checks before xfree
I was inspired by a series of patches merged by Alan Modra in the other
projects, so I did the same in GDB with a bit of Coccinelle and grep.
This patch removes the unnecessary NULL checks before calls to xfree.
They are unnecessary because xfree already does a NULL check. Since
free is supposed to handle NULL values correctly, the NULL check in
xfree itself is also questionable, but I've left it there for now.
Alan Modra [Thu, 21 May 2020 14:04:58 +0000 (23:34 +0930)]
Re: PR25993, read of freed memory
git commit 7b958a48e132 put the bfd filename in the bfd objalloc
memory. That means the filename is freed by _bfd_free_cached_info.
Which is called by _bfd_compute_and_write_armap to tidy up symbol
tables after they are done with.
Unfortunately, _bfd_write_archive_contents wants to seek and read from
archive elements after that point, and if the number of elements
exceeds max_open_files in cache.c then some of those elements will
have their files closed. To reopen, you need the filename.
PR 25993
* opncls.c (_bfd_free_cached_info): Keep a copy of the bfd
filename.
(_bfd_delete_bfd): Free the copy.
(_bfd_new_bfd): Free nbfd->memory on error.
Simon Marchi [Wed, 20 May 2020 19:44:24 +0000 (15:44 -0400)]
gdb: reset/recompute objfile section offsets in reread_symbols
This patch started as an investigation of this bug, where the program is
re-compiled between two "start" runs:
$ ./gdb -nx --data-directory=data-directory -q a.out
Reading symbols from a.out...
(gdb) start
Temporary breakpoint 1 at 0x1131: file test.c, line 1.
Starting program: /home/smarchi/build/wt/test/gdb/a.out
Temporary breakpoint 1, main () at test.c:1
1 int main() { return 0; }
*** re-compile a.out ***
(gdb) start
The program being debugged has been started already.
Start it from the beginning? (y or n) y
`/home/smarchi/build/wt/test/gdb/a.out' has changed; re-reading symbols.
Temporary breakpoint 2 at 0x555555555129: file test.c, line 1.
Starting program: /home/smarchi/build/wt/test/gdb/a.out
warning: Probes-based dynamic linker interface failed.
Reverting to original interface.
Temporary breakpoint 2, main () at test.c:1
1 int main() { return 0; }
(gdb)
To reproduce the bug, a.out needs to be a position-independent
executable (PIE).
Here's what happens:
1) We first read the symbols of a.out. The section offsets in the
objfile are all 0, so the symbols are created unrelocated.
2) The breakpoint on main is created, as you can see the breakpoint
address (derived from the `main` symbol with value 0x1129) is still
unrelocated (0x1131). Since the program is not yet started, we don't
know at which base address the executable is going to end at.
Everything good so far.
3) The execution starts, GDB finds out the executable's base address,
fills the objfile's section_offsets vector with a bunch of offsets,
and relocates the symbols with those offsets. The latter modifies
the symbol values (the `main` symbol is changed from 0x1129 to
0x555555555129).
4) We `start` again, we detect that `a.out` has changed, the
`reread_symbols` function kicks in. It tries to reset everything
in the `struct objfile` corresponding to `a.out`, except that it
leaves the `section_offsets` vector there.
5) `reread_symbols` reads the debug info (calls `read_symbols`). As the
DWARF info is read, symbols are created using the old offsets still
in `section_offsets`. For example, the `main` symbol is created with
the value 0x555555555129. Even though at this point there is no
process, so that address is bogus. There's probably more that
depends on section_offsets that is not done correctly.
6) Something in the SVR4 solib handling goes wrong, probably because
of something that went wrong in (5). I can't quite explain it (if
somebody would like to provide a more complete analysis, please go
ahead). But this is where it takes a wrong turn:
#0 elf_locate_base () at /home/smarchi/src/wt/test/gdb/solib-svr4.c:799
#1 0x000055f0a5bee6d5 in locate_base (info=<optimized out>) at /home/smarchi/src/wt/test/gdb/solib-svr4.c:848
#2 0x000055f0a5bf1771 in svr4_handle_solib_event () at /home/smarchi/src/wt/test/gdb/solib-svr4.c:1955
#3 0x000055f0a5c0ff92 in handle_solib_event () at /home/smarchi/src/wt/test/gdb/solib.c:1258
In the non-working case (without this patch), elf_locate_base returns
0, whereas in the working case (with this patch) it returns a valid
address, as we should expect.
This patch fixes this by making reread_symbols clear the
`section_offsets` vector, and re-create it by calling `sym_offsets`.
This is analogous to what syms_from_objfile_1 does. I didn't seem
absolutely necessary, but I also made it clear the various
`sect_index_*` fields, since their values no longer make sense (they
describe the old executable, and are indices in the now cleared
sections/section_offsets arrays).
I don't really like the approach taken by reread_symbols, trying to
reset everything manually on the objfile object, instead of, for
example, creating a new one from scratch. But I don't know enough yet
to propose a better solution.
One more reason I think this patch is needed is that the number of
sections of the new executable could be different from the number of
sections of the old executable. So if we don't re-create the
section_offsets array, not only we'll have wrong offsets, but we could
make accesses past the array.
Something else that silently fails (but doesn't seem to have
consequences) is the prologue analysis when we try to create the
breakpoint on `main`. Since the `main` symbol has the wrong value
0x555555555129, we try to access memory in that area, which fails. This
can be observed by debugging gdb and using `catch throw`. Before the
process is started, we need to access the memory at its unrelocated
address, 0x1129, which will read memory from the ELF file. This is now
what happens, with this patch applied.
It silently fails, probably because commit 46a62268b, "Catch exceptions
thrown from gdbarch_skip_prologue", papered over the problem and added
an empty catch clause. I'm quite sure that the root cause then was the
one fixed by this patch.
This fixes tests gdb.ada/exec_changed.exp and gdb.base/reread.exp for
me.
gdb/ChangeLog:
* symfile.c (reread_symbols): Clear objfile's section_offsets
vector and section indices, re-compute them by calling
sym_offsets.
Nelson Chu [Wed, 20 May 2020 16:22:48 +0000 (17:22 +0100)]
[PATCH v2 0/9] RISC-V: Support version controling for ISA standard extensions and CSR
1. Remove the -mriscv-isa-version and --with-riscv-isa-version options.
We can still use -march to choose the version for each extensions, so there is
no need to add these.
2. Change the arguments of options from [1p9|1p9p1|...] to [1.9|1.9.1|...].
Unlike the architecture string has specified by spec, ther is no need to do
the same thing for options.
3. Spilt the patches to reduce the burdens of review.
[PATCH 3/7] RISC-V: Support new GAS options and configure options to set ISA versions
to
[PATCH v2 3/9] RISC-V: Support GAS option -misa-spec to set ISA versions
[PATCH v2 4/9] RISC-V: Support configure options to set ISA versions by default.
[PATCH 4/7] RISC-V: Support version checking for CSR according to privilege version.
to
[PATCH v2 5/9] RISC-V: Support version checking for CSR according to privilege spec version.
[PATCH v2 6/9] RISC-V: Support configure option to choose the privilege spec version.
4. Use enum class rather than string to compare the choosen ISA spec in opcodes/riscv-opc.c.
The behavior is same as comparing the choosen privilege spec.
include * opcode/riscv.h: Include "bfd.h" to support bfd_boolean.
(enum riscv_isa_spec_class): New enum class. All supported ISA spec
belong to one of the class
(struct riscv_ext_version): New structure holds version information
for the specific ISA.
* opcode/riscv-opc.h (DECLARE_CSR): There are two version information,
define_version and abort_version. The define_version means which
privilege spec is started to define the CSR, and the abort_version
means which privilege spec is started to abort the CSR. If the CSR is
valid for the newest spec, then the abort_version should be
PRIV_SPEC_CLASS_DRAFT.
(DECLARE_CSR_ALIAS): Same as DECLARE_CSR, but only for the obselete CSR.
* opcode/riscv.h (enum riscv_priv_spec_class): New enum class. Define
the current supported privilege spec versions.
(struct riscv_csr_extra): Add new fields to store more information
about the CSR. We use these information to find the suitable CSR
address when user choosing a specific privilege spec.
binutils * dwarf.c: Updated since DECLARE_CSR is changed.
opcodes * riscv-opc.c (riscv_ext_version_table): The table used to store
all information about the supported spec and the corresponding ISA
versions. Currently, only Zicsr is supported to verify the
correctness of Z sub extension settings. Others will be supported
in the future patches.
(struct isa_spec_t, isa_specs): List for all supported ISA spec
classes and the corresponding strings.
(riscv_get_isa_spec_class): New function. Get the corresponding ISA
spec class by giving a ISA spec string.
* riscv-opc.c (struct priv_spec_t): New structure.
(struct priv_spec_t priv_specs): List for all supported privilege spec
classes and the corresponding strings.
(riscv_get_priv_spec_class): New function. Get the corresponding
privilege spec class by giving a spec string.
(riscv_get_priv_spec_name): New function. Get the corresponding
privilege spec string by giving a CSR version class.
* riscv-dis.c: Updated since DECLARE_CSR is changed.
* riscv-dis.c: Add new disassembler option -Mpriv-spec to dump the CSR
according to the chosen version. Build a hash table riscv_csr_hash to
store the valid CSR for the chosen pirv verison. Dump the direct
CSR address rather than it's name if it is invalid.
(parse_riscv_dis_option_without_args): New function. Parse the options
without arguments.
(parse_riscv_dis_option): Call parse_riscv_dis_option_without_args to
parse the options without arguments first, and then handle the options
with arguments. Add the new option -Mpriv-spec, which has argument.
* riscv-dis.c (print_riscv_disassembler_options): Add description
about the new OBJDUMP option.
bfd * elfxx-riscv.h (riscv_parse_subset_t): Add new callback function
get_default_version. It is used to find the default version for
the specific extension.
* elfxx-riscv.c (riscv_parsing_subset_version): Remove the parameters
default_major_version and default_minor_version. Add new bfd_boolean
parameter *use_default_version. Set it to TRUE if we need to call
the callback rps->get_default_version to find the default version.
(riscv_parse_std_ext): Call rps->get_default_version if we fail to find
the default version in riscv_parsing_subset_version, and then call
riscv_add_subset to add the subset into subset list.
(riscv_parse_prefixed_ext): Likewise.
(riscv_std_z_ext_strtab): Support Zicsr extensions.
* elfnn-riscv.c (riscv_merge_std_ext): Use strcasecmp to compare the
strings rather than characters.
riscv_merge_arch_attr_info): The callback function get_default_version
is only needed for assembler, so set it to NULL int the linker.
* elfxx-riscv.c (riscv_estimate_digit): Remove the static.
* elfxx-riscv.h: Updated.
gas * testsuite/gas/riscv/priv-reg-fail-read-only-01.s: Updated.
* config/tc-riscv.c (default_arch_with_ext, default_isa_spec):
Static variables which are used to set the ISA extensions. You can
use -march (or ELF build attributes) and -misa-spec to set them,
respectively.
(ext_version_hash): The hash table used to handle the extensions
with versions.
(init_ext_version_hash): Initialize the ext_version_hash according
to riscv_ext_version_table.
(riscv_get_default_ext_version): The callback function of
riscv_parse_subset_t. According to the choosed ISA spec,
get the default version for the specific extension.
(riscv_set_arch): Set the callback function.
(enum options, struct option md_longopts): Add new option -misa-spec.
(md_parse_option): Do not call riscv_set_arch for -march. We will
call it later in riscv_after_parse_args. Call riscv_get_isa_spec_class
to set default_isa_spec class.
(riscv_after_parse_args): Call init_ext_version_hash to initialize the
ext_version_hash, and then call riscv_set_arch to set the architecture
with versions according to default_arch_with_ext.
* testsuite/gas/riscv/attribute-02.d: Set 0p0 as default version for
x extensions.
* testsuite/gas/riscv/attribute-03.d: Likewise.
* testsuite/gas/riscv/attribute-09.d: New testcase. For i-ext, we
already set it's version to 2p1 by march, so no need to use the default
2p2 version. For m-ext, we do not set the version by -march and ELF arch
attribute, so set the default 2p0 to it. For zicsr, it is not defined in
ISA spec 2p2, so set 0p0 to it.
* testsuite/gas/riscv/attribute-10.d: New testcase. The version of
zicsr is 2p0 according to ISA spec 20191213.
* config/tc-riscv.c (DEFAULT_RISCV_ARCH_WITH_EXT)
(DEFAULT_RISCV_ISA_SPEC): Default configure option settings.
You can set them by configure options --with-arch and
--with-isa-spec, respectively.
(riscv_set_default_isa_spec): New function used to set the
default ISA spec.
(md_parse_option): Call riscv_set_default_isa_spec rather than
call riscv_get_isa_spec_class directly.
(riscv_after_parse_args): If the -isa-spec is not set, then we
set the default ISA spec according to DEFAULT_RISCV_ISA_SPEC by
calling riscv_set_default_isa_spec.
* testsuite/gas/riscv/attribute-01.d: Add -misa-spec=2.2, since
the --with-isa-spec may be set to different ISA spec.
* testsuite/gas/riscv/attribute-02.d: Likewise.
* testsuite/gas/riscv/attribute-03.d: Likewise.
* testsuite/gas/riscv/attribute-04.d: Likewise.
* testsuite/gas/riscv/attribute-05.d: Likewise.
* testsuite/gas/riscv/attribute-06.d: Likewise.
* testsuite/gas/riscv/attribute-07.d: Likewise.
* configure.ac: Add configure options, --with-arch and
--with-isa-spec.
* configure: Regenerated.
* config.in: Regenerated.
* config/tc-riscv.c (default_priv_spec): Static variable which is
used to check if the CSR is valid for the chosen privilege spec. You
can use -mpriv-spec to set it.
(enum reg_class): We now get the CSR address from csr_extra_hash rather
than reg_names_hash. Therefore, move RCLASS_CSR behind RCLASS_MAX.
(riscv_init_csr_hashes): Only need to initialize one hash table
csr_extra_hash.
(riscv_csr_class_check): Change the return type to void. Don't check
the ISA dependency if -mcsr-check isn't set.
(riscv_csr_version_check): New function. Check and find the CSR address
from csr_extra_hash, according to default_priv_spec. Report warning
for the invalid CSR if -mcsr-check is set.
(reg_csr_lookup_internal): Updated.
(reg_lookup_internal): Likewise.
(md_begin): Updated since DECLARE_CSR and DECLARE_CSR_ALIAS are changed.
(enum options, struct option md_longopts): Add new GAS option -mpriv-spec.
(md_parse_option): Call riscv_set_default_priv_version to set
default_priv_spec.
(riscv_after_parse_args): If -mpriv-spec isn't set, then set the default
privilege spec to the newest one.
(enum riscv_csr_class, struct riscv_csr_extra): Move them to
include/opcode/riscv.h.
* testsuite/gas/riscv/priv-reg-fail-fext.d: This test case just want
to check the ISA dependency for CSR, so fix the spec version by adding
-mpriv-spec=1.11.
* testsuite/gas/riscv/priv-reg-fail-fext.l: Likewise. There are some
version warnings for the test case.
* gas/testsuite/gas/riscv/priv-reg-fail-read-only-01.d: Likewise.
* gas/testsuite/gas/riscv/priv-reg-fail-read-only-01.l: Likewise.
* gas/testsuite/gas/riscv/priv-reg-fail-read-only-02.d: Likewise.
* gas/testsuite/gas/riscv/priv-reg-fail-rv32-only.d: Likewise.
* gas/testsuite/gas/riscv/priv-reg-fail-rv32-only.l: Likewise.
* gas/testsuite/gas/riscv/priv-reg-fail-version-1p9.d: New test case.
Check whether the CSR is valid when privilege version 1.9 is choosed.
* gas/testsuite/gas/riscv/priv-reg-fail-version-1p9.l: Likewise.
* gas/testsuite/gas/riscv/priv-reg-fail-version-1p9p1.d: New test case.
Check whether the CSR is valid when privilege version 1.9.1 is choosed.
* gas/testsuite/gas/riscv/priv-reg-fail-version-1p9p1.l: Likewise.
* gas/testsuite/gas/riscv/priv-reg-fail-version-1p10.d: New test case.
Check whether the CSR is valid when privilege version 1.10 is choosed.
* gas/testsuite/gas/riscv/priv-reg-fail-version-1p10.l: Likewise.
* gas/testsuite/gas/riscv/priv-reg-fail-version-1p11.d: New test case.
Check whether the CSR is valid when privilege version 1.11 is choosed.
* gas/testsuite/gas/riscv/priv-reg-fail-version-1p11.l: Likewise.
* config/tc-riscv.c (DEFAULT_RISCV_ISA_SPEC): Default configure option
setting. You can set it by configure option --with-priv-spec.
(riscv_set_default_priv_spec): New function used to set the default
privilege spec.
(md_parse_option): Call riscv_set_default_priv_spec rather than
call riscv_get_priv_spec_class directly.
(riscv_after_parse_args): If -mpriv-spec isn't set, then we set the
default privilege spec according to DEFAULT_RISCV_PRIV_SPEC by
calling riscv_set_default_priv_spec.
* testsuite/gas/riscv/csr-dw-regnums.d: Add -mpriv-spec=1.11, since
the --with-priv-spec may be set to different privilege spec.
* testsuite/gas/riscv/priv-reg.d: Likewise.
* configure.ac: Add configure option --with-priv-spec.
* configure: Regenerated.
* config.in: Regenerated.
* config/tc-riscv.c (explicit_attr): Rename explicit_arch_attr to
explicit_attr. Set it to TRUE if any ELF attribute is found.
(riscv_set_default_priv_spec): Try to set the default_priv_spec if
the priv attributes are set.
(md_assemble): Set the default_priv_spec according to the priv
attributes when we start to assemble instruction.
(riscv_write_out_attrs): Rename riscv_write_out_arch_attr to
riscv_write_out_attrs. Update the arch and priv attributes. If we
don't set the corresponding ELF attributes, then try to output the
default ones.
(riscv_set_public_attributes): If any ELF attribute or -march-attr
options is set (explicit_attr is TRUE), then call riscv_write_out_attrs
to update the arch and priv attributes.
(s_riscv_attribute): Make sure all arch and priv attributes are set
before any instruction.
* testsuite/gas/riscv/attribute-01.d: Update the priv attributes if any
ELF attribute or -march-attr is set. If the priv attributes are not
set, then try to update them by the default setting (-mpriv-spec or
--with-priv-spec).
* testsuite/gas/riscv/attribute-02.d: Likewise.
* testsuite/gas/riscv/attribute-03.d: Likewise.
* testsuite/gas/riscv/attribute-04.d: Likewise.
* testsuite/gas/riscv/attribute-06.d: Likewise.
* testsuite/gas/riscv/attribute-07.d: Likewise.
* testsuite/gas/riscv/attribute-08.d: Likewise.
* testsuite/gas/riscv/attribute-09.d: Likewise.
* testsuite/gas/riscv/attribute-10.d: Likewise.
* testsuite/gas/riscv/attribute-unknown.d: Likewise.
* testsuite/gas/riscv/attribute-05.d: Likewise. Also, the priv spec
set by priv attributes must be supported.
* testsuite/gas/riscv/attribute-05.s: Likewise.
* testsuite/gas/riscv/priv-reg-fail-version-1p9.d: Likewise. Updated
priv attributes according to the -mpriv-spec option.
* testsuite/gas/riscv/priv-reg-fail-version-1p9p1.d: Likewise.
* testsuite/gas/riscv/priv-reg-fail-version-1p10.d: Likewise.
* testsuite/gas/riscv/priv-reg-fail-version-1p11.d: Likewise.
* testsuite/gas/riscv/priv-reg.d: Removed.
* testsuite/gas/riscv/priv-reg-version-1p9.d: New test case. Dump the
CSR according to the priv spec 1.9.
* testsuite/gas/riscv/priv-reg-version-1p9p1.d: New test case. Dump the
CSR according to the priv spec 1.9.1.
* testsuite/gas/riscv/priv-reg-version-1p10.d: New test case. Dump the
CSR according to the priv spec 1.10.
* testsuite/gas/riscv/priv-reg-version-1p11.d: New test case. Dump the
CSR according to the priv spec 1.11.
* config/tc-riscv.c (md_show_usage): Add descriptions about
the new GAS options.
* doc/c-riscv.texi: Likewise.
Simon Marchi [Wed, 20 May 2020 14:50:39 +0000 (10:50 -0400)]
gdb/testsuite: check mmap ret val against MAP_FAILED
Fixup a few spots in the testsuite that use mmap to consistently check
the return value against MAP_FAILED.
One spot in gdb.base/coredump-filter.c checked against NULL, that is
wrong. The other spots either didn't check, or checked against -1.
MAP_FAILED has the value -1, at least on Linux, but it's better to check
against the documented define.
gdb/testsuite/ChangeLog:
PR gdb/26016
* gdb.base/coredump-filter.c (do_mmap): Check mmap ret val
against MAP_FAILED.
* gdb.base/coremaker.c (mmapdata): Likewise.
* gdb.base/jit-reader-host.c (main): Likewise.
* gdb.base/sym-file-loader.c (load): Likewise.
(load_shlib): Likewise.
Tom Tromey [Fri, 15 May 2020 16:25:56 +0000 (10:25 -0600)]
Fix array_char_idx.exp
Newer versions of GCC can statically initialize an array in the
array_char_idx.exp test case. This leads to a spurious failure. This
patch fixes the problem by having the test case recognize both
possible results.
I'm checking this in.
gdb/testsuite/ChangeLog
2020-05-20 Tom Tromey <tromey@adacore.com>
Tom Tromey [Tue, 5 May 2020 12:57:04 +0000 (06:57 -0600)]
Remove bound_name static from ada-lang.c
ada-lang.c has a "bound_name" static that is used when finding fields
in a structure in some cases. This patch removes it in favor of
computing the desired field name at runtime; this avoids an artificial
limit.
I'm checking this in. Tested on x86-64 Fedora 30, and also on the
internal AdaCore test suite.
gdb/ChangeLog
2020-05-20 Tom Tromey <tromey@adacore.com>
* ada-lang.c (bound_name, MAX_ADA_DIMENS): Remove.
(desc_one_bound, desc_index_type): Compute field name.
Tom de Vries [Wed, 20 May 2020 09:48:39 +0000 (11:48 +0200)]
[gdb/symtab] Handle .gdb_index in ada language mode
When running test-case gdb.base/with.exp with target board cc-with-gdb-index,
we have:
...
(gdb) PASS: gdb.base/with.exp: basics: show language
with language ada -- print g_s^M
'g_s' has unknown type; cast it to its declared type^M
(gdb) FAIL: gdb.base/with.exp: basics: with language ada -- print g_s
...
This is due to this bit in dw2_map_matching_symbols:
...
if (dwarf2_per_objfile->index_table != nullptr)
{
/* Ada currently doesn't support .gdb_index (see PR24713). We can get
here though if the current language is Ada for a non-Ada objfile
using GNU index. As Ada does not look for non-Ada symbols this
function should just return. */
return;
}
...
While the reasoning in the comment may be sound from language perspective, it
does introduce an inconsistency in gdb behaviour between:
- having a .gdb_index section, and
- having a .gdb_names section, or a partial symtab, or -readnow.
Fix the inconsistency by completing implementation of
dw2_map_matching_symbols.
Tested on x86_64-linux, both with native and target board
cc-with-debug-index.
Alan Modra [Tue, 19 May 2020 03:28:59 +0000 (12:58 +0930)]
PR25993, read of freed memory
ldmain.c:add_archive_element copies file name pointers from the bfd to
a lang_input_statement_type.
input->filename = abfd->filename;
input->local_sym_name = abfd->filename;
This results in stale pointers when twiddling the bfd filename in
places like the pe ld after_open. So don't free the bfd filename,
and make copies using bfd_alloc memory that won't result in small
memory leaks that annoy memory checkers.
PR 25993
bfd/
* archive.c (_bfd_get_elt_at_filepos): Don't strdup filename,
use bfd_set_filename.
* elfcode.h (_bfd_elf_bfd_from_remote_memory): Likewise.
* mach-o.c (bfd_mach_o_fat_member_init): Likewise.
* opncls.c (bfd_fopen, bfd_openstreamr, bfd_openr_iovec, bfd_openw),
(bfd_create): Likewise.
(_bfd_delete_bfd): Don't free filename.
(bfd_set_filename): Copy filename param to bfd_alloc'd memory,
return pointer to the copy or NULL on alloc fail.
* vms-lib.c (_bfd_vms_lib_get_module): Free newname and test
result of bfd_set_filename.
* bfd-in2.h: Regenerate.
gdb/
* solib-darwin.c (darwin_bfd_open): Don't strdup pathname for
bfd_set_filename.
* solib-aix.c (solib_aix_bfd_open): Use std::string for name
passed to bfd_set_filename.
* symfile-mem.c (add_vsyscall_page): Likewise for string
passed to symbol_file_add_from_memory.
(symbol_file_add_from_memory): Make name param a const char* and
don't strdup.
ld/
* emultempl/pe.em (gld_${EMULATION_NAME}_after_open): Don't copy
other_bfd_filename for bfd_set_filename, and test result of
bfd_set_filename call. Don't create a new is->filename, simply
copy from bfd filename. Free new_name after bfd_set_filename.
* emultempl/pep.em (gld_${EMULATION_NAME}_after_open): Likewise.
Peter Bergner [Tue, 19 May 2020 23:09:51 +0000 (18:09 -0500)]
Power10 dcbf, sync, and wait extensions.
opcodes/
* ppc-opc.c (insert_ls, extract_ls): Handle 3-bit L fields and new
WC values on POWER10 sync, dcbf and wait instructions.
(insert_pl, extract_pl): New functions.
(L2OPT, LS, WC): Use insert_ls and extract_ls.
(LS3): New , 3-bit L for sync.
(LS3, L3OPT): New, 3-bit L for sync and dcbf.
(SC2, PL): New, 2-bit SC and PL for sync and wait.
(XWCPL_MASK, XL3RT_MASK, XSYNCLS_MASK): New instruction masks.
(XOPL3, XWCPL, XSYNCLS): New opcode macros.
(powerpc_opcodes) <dcbflp, dcbfps, dcbstps pause_short, phwsync,
plwsync, stcisync, stncisync, stsync, waitrsv>: New extended mnemonics.
<wait>: Enable PL operand on POWER10.
<dcbf>: Enable L3OPT operand on POWER10.
<sync>: Enable SC2 operand on POWER10.
Pedro Alves [Tue, 19 May 2020 17:36:24 +0000 (18:36 +0100)]
Make exec-file-mismatch compare build IDs
The patch makes GDB do exec-file-mismatch validation by comparing
build IDs instead of the current method of comparing filenames.
Currently, the exec-file-mismatch feature simply compares filenames to
decide whether the exec file loaded in gdb and the exec file the
target reports is running match. This causes false positives when
remote debugging, because it'll often be the case that the paths in
the host and the target won't match. And of course misses the case of
the files having the same name but being actually different files
(e.g., different builds).
This also broke many testcases when running against gdbserver, causing
tests to be skipped like (here native-extended-gdbserver):
(gdb) run
Starting program: /home/pedro/gdb/binutils-gdb/build/gdb/testsuite/outputs/gdb.base/argv0-symlink/argv0-symlink-filelink
warning: Mismatch between current exec-file /home/pedro/gdb/binutils-gdb/build/gdb/testsuite/outputs/gdb.base/argv0-symlink/argv0-symlink-filelink
and automatically determined exec-file /home/pedro/gdb/binutils-gdb/build/gdb/testsuite/outputs/gdb.base/argv0-symlink/argv0-symlink
exec-file-mismatch handling is currently "ask"
Load new symbol table from "/home/pedro/gdb/binutils-gdb/build/gdb/testsuite/outputs/gdb.base/argv0-symlink/argv0-symlink"? (y or n) UNTESTED: gdb.base/argv0-symlink.exp: could not run to main
or to fail like (here native-gdbserver):
(gdb) spawn /home/pedro/gdb/binutils-gdb/build/gdb/testsuite/../../gdbserver/gdbserver --once localhost:2346 /home/pedro/gdb/binutils-gdb/build/gdb/te
stsuite/outputs/gdb.btrace/buffer-size/skip_btrace_tests-19968.x
Process /home/pedro/gdb/binutils-gdb/build/gdb/testsuite/outputs/gdb.btrace/buffer-size/skip_btrace_tests-19968.x created; pid = 20040
Listening on port 2346
target remote localhost:2346
Remote debugging using localhost:2346
warning: Mismatch between current exec-file /home/pedro/gdb/binutils-gdb/build/gdb/testsuite/temp/19968/skip_btrace_tests-19968.x
and automatically determined exec-file /home/pedro/gdb/binutils-gdb/build/gdb/testsuite/outputs/gdb.btrace/buffer-size/skip_btrace_tests-19968.x
exec-file-mismatch handling is currently "ask"
Load new symbol table from "/home/pedro/gdb/binutils-gdb/build/gdb/testsuite/outputs/gdb.btrace/buffer-size/skip_btrace_tests-19968.x"? (y or n) Quit
(gdb) UNSUPPORTED: gdb.btrace/buffer-size.exp: target does not support record-btrace
The former case is about GDB not realizing the two files are the same,
because one of the them is a symlink to the other. The latter case is
about GDB realizing that one file is a copy of the other.
Over the years, the toolchain has settled on build ID matching being
the canonical method to match core dumps to executables, and
executables with no debug info to their debug info.
This patch makes us use build IDs to match the running image of a
binary with its version loaded in gdb, which may or may not have debug
info. This is very much like the core dump/executable matching.
The change to gdb_bfd_open is necessary to get rid of the "transfers
from remote targets can be slow" warning when we open the remote file
to read its build ID:
(gdb) r
Starting program: /home/pedro/gdb/binutils-gdb/build/gdb/testsuite/outputs/gdb.base/break/break
Reading /home/pedro/gdb/binutils-gdb/build/gdb/testsuite/outputs/gdb.base/argv0-symlink/argv0-symlink from remote target...
warning: File transfers from remote targets can be slow. Use "set sysroot" to access files locally instead.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
warning: Mismatch between current exec-file /home/pedro/gdb/binutils-gdb/build/gdb/testsuite/outputs/gdb.base/break/break
and automatically determined exec-file /home/pedro/gdb/binutils-gdb/build/gdb/testsuite/outputs/gdb.base/argv0-symlink/argv0-symlink
exec-file-mismatch handling is currently "ask"
Load new symbol table from "/home/pedro/gdb/binutils-gdb/build/gdb/testsuite/outputs/gdb.base/argv0-symlink/argv0-symlink"? (y or n)
While trying this out, I was worried that bfd would read a lot of
stuff from the binary in order to extract the build ID, making it
potentially slow, but turns out we don't read all that much. Maybe a
couple hundred bytes, and most of it seemingly is the read-ahead
cache. So I'm not worried about that. Otherwise I'd consider whether
a new qXfer:buildid:read would be better. But I'm happy that we
seemingly don't need to worry about it.
gdb/ChangeLog:
2020-05-19 Pedro Alves <palves@redhat.com>
* NEWS (set exec-file-mismatch): Adjust entry.
* exec.c: Include "build-id.h".
(validate_exec_file): Try to match build IDs instead of filenames.
* gdb_bfd.c (struct gdb_bfd_open_closure): New.
(gdb_bfd_iovec_fileio_open): Adjust to use gdb_bfd_open_closure
and pass down 'warn_if_slow'.
(gdb_bfd_open): Add 'warn_if_slow' parameter. Use
gdb_bfd_open_closure to pass it down.
* gdb_bfd.h (gdb_bfd_open): Add 'warn_if_slow' parameter.
gdb/doc/ChangeLog:
2020-05-19 Pedro Alves <palves@redhat.com>
Pedro Alves [Tue, 19 May 2020 17:36:24 +0000 (18:36 +0100)]
Eliminate target_fileio_open_warn_if_slow
This basically makes target_fileio_open_1 extern, renamed to
target_fileio_open, and eliminates the current
target_fileio_open_warn_if_slow and target_fileio_open.
A following parameter will want to change gdb_bfd_iovec_fileio_open,
the only caller of target_fileio_open_warn_if_slow, to pass down
"warn_if_slow" true/false from the caller, instead of hardcoding
"warn_if_slow" true.
gdb/ChangeLog:
2020-05-19 Pedro Alves <palves@redhat.com>
* gdb_bfd.c (gdb_bfd_iovec_fileio_open): Adjust.
* target.c (target_fileio_open_1): Rename to target_fileio_open
and make extern. Use bool.
(target_fileio_open, target_fileio_open_warn_if_slow): Delete.
(target_fileio_read_alloc_1): Adjust.
* target.h (target_fileio_open): Add 'warn_if_slow' parameter.
(target_fileio_open_warn_if_slow): Delete declaration.
Simon Marchi [Tue, 19 May 2020 16:18:04 +0000 (12:18 -0400)]
gdb: use std::vector to store segments in symfile_segment_data
Instead of maintaining two vectors, I added a small `segment` class
which holds both the base address and size of one segment and replaced
the two `segment_bases` and `segment_sizes` arrays with a single vector.
The rest of the changes are straightforward, no behavior changes are
expected.
Simon Marchi [Tue, 19 May 2020 16:18:04 +0000 (12:18 -0400)]
gdb: allocate symfile_segment_data with new
- Allocate this structure with new instead of XNEW, use a unique pointer
to manage its lifetime.
- Change a few functions to return a unique pointer instead of a
plain pointer.
- Change free_symfile_segment_data to be symfile_segment_data's
destructor.
H.J. Lu [Tue, 19 May 2020 15:47:12 +0000 (16:47 +0100)]
Allow readelf to issue warnings if checking is enabled and there are discrepancies between the dynamic tags and the dynamic sections.
PR binutils/25809
* readelf.c (process_program_headers): Warn if the PT_DYNAMIC
segment doesn't match the .dynamic section and checks are
enabled.
(struct filedata): Add dynamic_symtab_section and
dynamic_strtab_section fields.
(process_section_headers): Set dynamic_symtab_section to the
.dynsym section. Set dynamic_strtab_section to the .dynstr
section.
(process_dynamic_section): Warn if the .dynsym section doesn't
match DT_SYMTAB and DT_SYMENT or the .dynstr section doesn't
DT_STRTAB and DT_STRSZ. But only if checks are enabled.
Stafford Horne [Tue, 19 May 2020 13:26:33 +0000 (14:26 +0100)]
OpenRISC BFD fixups for Glibc:
or1k: Fix static linking when with .rela.got relocations
or1k: Fix dynamic TLS symbol flag
or1k: Add TLS mask to handle multiple model access
or1k: Fix issue with multiple PCREL relocations
or1k: TLS offset to use tcb size and section alignment
or1k: refactor: Rename p to sec_relocs
or1k: refactor: Rename s to sgot and splt
or1k: Add dynamic flag to tpoff
bfd * elf32-or1k.c (or1k_elf_finish_dynamic_symbol): Rename srela
to relgot.
(or1k_elf_relocate_section): Access srelgot via
htab->root.srelgot. Add assertions for srelgot->contents.
Introduce local variable for srelgot to not reuse global
sreloc.
(or1k_elf_relocate_section): Fixup dynamic symbol detection.
(or1k_set_got_and_rela_sizes): New function.
(or1k_initial_exec_offset): New function.
(TLS_GD, TLS_IE, TLS_LD, TLS_LE): Redefine macros as masks.
(or1k_elf_relocate_section): Allow for TLS to handle multiple
model access.
(or1k_elf_check_relocs): Use OR to set TLS access.
(allocate_dynrelocs): Use or1k_set_got_and_rela_sizes to set
sizes.
(or1k_elf_size_dynamic_sections): Use
or1k_set_got_and_rela_sizes to set sizes.
(or1k_elf_relocate_section): Fixup PCREL relocation calculation.
(TCB_SIZE): New macro.
(tpoff): Use TCB_SIZE and alignment to calculate offset.
(allocate_dynrelocs, readonly_dynrelocs, or1k_elf_check_relocs)
(or1k_elf_size_dynamic_sections): Rename p to sec_relocs.
(allocate_dynrelocs): Rename s to splt or sgot based on usage.
(tpoff): Add dynamic boolean argument.
(or1k_elf_relocate_section): Pass dynamic flag to tpoff.
Stafford Horne [Tue, 19 May 2020 11:40:27 +0000 (20:40 +0900)]
or1k: Remove 64-bit support, it's not used and it breaks 32-bit hosts
Reported by Rich Felker when building on 32-bit hosts. Backwards jump
negative offsets were not calculated correctly due to improper 32-bit
to 64-bit zero-extension. The 64-bit fields are present because we
are mixing 32-bit and 64-bit architectures in our cpu descriptions.
Removing 64-bit fixes the issue. We don't use 64-bit, there is an architecture
spec for 64-bit but no implementations or simulators. My thought is if
we need them in the future we should do the proper work to support both
32-bit and 64-bit implementations co-existing then.
[PATCH v3] aarch64: Emit jump slot for conditional branch to undefined symbols
The linker silently writes out a conditional branch to 0 if the
target symbol in R_AARCH64_CONDBR19 or R_AARCH64_TSTBR14 relocations is
undefined. Emit a PLT instead so that behaviour is the same for these
relocations as the llvm linker.
The special behaviour for undefined weak symbols, where conditional
branches to such symbols result in a branch unto themselves, has been
retained. This is because the weak-undefined.s test explicitly checks
for that, leading me to conclude that it's expected behaviour.
bfd * elfnn-aarch64.c (elfNN_aarch64_final_link_relocate): Club
BFD_RELOC_AARCH64_BRANCH19 and BFD_RELOC_AARCH64_TSTBR14
cases with BFD_RELOC_AARCH64_JUMP26.
(elfNN_aarch64_check_relocs): Likewise.
ld * testsuite/ld-aarch64/aarch64-elf.exp: New test
emit-relocs-560.
* testsuite/ld-aarch64/emit-relocs-560.d: New file.
* testsuite/ld-aarch64/emit-relocs-560.s: New file.
Alan Modra [Tue, 19 May 2020 03:25:27 +0000 (12:55 +0930)]
Use bfd_get_filename throughout gas
* write.c (write_contents): Use bfd_get_filename rather than
accessing bfd->filename directly. Use bfd_section_name rather
than accessing section->name directly.
Rainer Orth [Mon, 18 May 2020 15:59:43 +0000 (17:59 +0200)]
Avoid short i386 register names on Solaris/x86 [PR25981]
This is the 32-bit companion to
Remove unused ps_lgetLDT etc. on Solaris/x86 [PR25981]
https://sourceware.org/pipermail/gdb-patches/2020-May/168713.html
A 32-bit-default gdb fails to compile with the updated <sys/regset.h>.
While it is also affected by the lack of a GS definition, which the
compantion patch above fixes, it also fails to compile i386-sol2-nat.c like
this
/vol/src/gnu/gdb/hg/master/git/gdb/i386-sol2-nat.c:181:3: error: 'EAX' was not declared in this scope
181 | EAX, ECX, EDX, EBX,
| ^~~
and several more.
While this could be fixed by either including <ucontext.h> here or
provding fallback definitions of the register macros, I chose to do what
the 64-bit-default code in the same file
(amd64_sol2_gregset32_reg_offset[]) does, namely just hardcode the
numeric values instead. They are part of the ABI and thus guaranteed
not to change.
With this patch, a i386-pc-solaris2.11 configuration on master compiles
again, however, it doesn't work. However, I could successfully test it
on the gdb-9 branch.
Compiling and testing proved to be messy, unfortunately:
* For one, Solaris <sys/procfs.h> and largefile support used to be
mutually exclusive (fixed in Solaris 11.4 and Illumos), which was
exacerbated by the fact that g++ predefines _FILE_OFFSET_BITS=64 since
GCC 9.1.0. For now I've worked around this by adding
-U_FILE_OFFSET_BITS to CXXFLAGS and configuring with
--disable-largefile. I hope to clean this up in a future patch.
* gdb still defaults to startup-with-shell on. However, /bin/bash is a
64-bit executable which cannot be debugged by a 32-bit gdb. I hacked
around that part by pointing $SHELL at a 32-bit bash before running
make check.
Rainer Orth [Mon, 18 May 2020 15:56:00 +0000 (17:56 +0200)]
Remove unused ps_lgetLDT etc. on Solaris/x86 [PR25981]
As reported in PR build/25981, a future Solaris 11.4 update will soon
remove the short i386 register names like SS etc. from <sys/regset.h>.
They could leak into user code (e.g. via <signal.h> -> <sys/signal.h> ->
<sys/ucontext.h>) and pollute the user namespace. Affected code would
have a hard time avoiding the issue: LLVM is one of those.
While the short names are required to be present by the i386 psABI, that
document only demands that they exist in <ucontext.h>, which is what the
upcoming update assures.
With this change, in a 64-bit-default configuration, procfs.c fails to
compile on Solaris/x86:
/vol/src/gnu/gdb/hg/master/git/gdb/procfs.c: In function 'ssd* procfs_find_LDT_entry(ptid_t)':
/vol/src/gnu/gdb/hg/master/git/gdb/procfs.c:1643:18: error: 'GS' was not declared in this scope
1643 | key = (*gregs)[GS] & 0xffff;
| ^~
make[2]: *** [Makefile:1607: procfs.o] Error 1
Initially I meant to provide a definition using the planned replacement
macro, but closer inspection revealed a better way. procfs_find_LDT_entry
and its helper proc_get_LDT_entry are only used to implement ps_lgetLDT,
one of the callback functions required by libthread_db.so.1
(cf. <proc_service.h>). While that function is still documented as being
required even in Solaris 11.4, I found that calls to it had been removed
long ago in Solaris 9, so just removing the three functions above is the
easiest fix.
The following patch does just that. It compiled successfully on
amd64-pc-solaris2.11, however, as reported in PR gdb/25939, master is
completely broken on Solaris since the multi-target patch. The patch
applies cleanly to the gdb-9 branch and there I could test it
successfully.
Alan Modra [Mon, 18 May 2020 11:14:30 +0000 (20:44 +0930)]
ECOFF slurp_relocs thinko
In git commit 806470a219 I swapped the order of internal vs. external
relocs memory allocation in ecoff_slurp_reloc_table, the idea being
that the external reloc size can be sanity checked against file size.
However, that fails badly with bfd_alloc memory where releasing any
block also releases all more recently allocated blocks.
* ecoff.c (ecoff_slurp_reloc_table): Malloc external_relocs so
they can be freed without also freeing internal_relocs.
Pedro Alves [Sun, 17 May 2020 18:17:56 +0000 (19:17 +0100)]
Stop considering hw and sw breakpoint locations duplicates (PR gdb/25741)
In the following conditions:
- A target with hardware breakpoints available, and
- A target that uses software single stepping,
- An instruction at ADDRESS loops back to itself,
Now consider the following steps:
1. The user places a hardware breakpoint at ADDRESS (an instruction
that loops to itself),
2. The inferior runs and hits the breakpoint at ADDRESS,
3. The user tells GDB to 'continue'.
In #3 when the user tells GDB to continue, GDB first disables the
hardware breakpoint at ADDRESS, and then inserts a software
single-step breakpoint at ADDRESS. The original user-created
breakpoint was a hardware breakpoint, while the single-step breakpoint
will be a software breakpoint.
GDB continues and immediately hits the software single-step
breakpoint.
GDB then deletes the software single-step breakpoint by calling
delete_single_step_breakpoints, which eventually calls
delete_breakpoint, which, once the breakpoint (and its locations) are
deleted, calls update_global_location_list.
During update_global_location_list GDB spots that we have an old
location (the software single step breakpoint location) that is
inserted, but being deleted, and a location (the original hardware
breakpoint) at the same address which we are keeping, but which is not
currently inserted, GDB then calls breakpoint_locations_match on these
two locations.
Currently the locations do match, and so GDB calls swap_insertion
which swaps the "inserted" state of the two locations. The user
created hardware breakpoint is marked as inserted, while the GDB
internal software single step breakpoint is now marked as not
inserted. After this GDB returns through the call stack and leaves
delete_single_step_breakpoints.
After this GDB continues with its normal "stopping" process, as part
of this stopping process GDB removes all the breakpoints from the
target. Due to the swap it is now the user-created hardware
breakpoint that is marked as inserted, so it is this breakpoint GDB
tries to remove.
The problem is that GDB inserted the software single-step breakpoint
as a software breakpoint, but is now trying to remove the hardware
breakpoint. The problem is removing a software breakpoint is very
different to removing a hardware breakpoint, this could result is some
undetected undefined behaviour, or as in the original bug report (PR
gdb/25741), could result in the target throwing an error.
With "set breakpoint always-inserted on", we can easily reproduce this
against GDBserver. E.g.:
(gdb) hbreak main
Sending packet: $m400700,40#28...Packet received: 89e58b....
Sending packet: $m400736,1#fe...Packet received: 48
Hardware assisted breakpoint 1 at 0x400736: file threads.c, line 57.
Sending packet: $Z1,400736,1#48...Packet received: OK
Packet Z1 (hardware-breakpoint) is supported
(gdb) b main
Note: breakpoint 1 also set at pc 0x400736.
Sending packet: $m400736,1#fe...Packet received: 48
Breakpoint 2 at 0x400736: file threads.c, line 57.
(gdb) del
Delete all breakpoints? (y or n) y
Sending packet: $z0,400736,1#67...Packet received: E01
warning: Error removing breakpoint 2
This patch adds a testcase that does exactly that.
Trying to enhance GDB to handle this scenario while continuing to
avoid inserting redundant software and hardware breakpoints at the
same address turns out futile, because, given non-stop and breakpoints
always-inserted, if the user:
#1 - inserts a hw breakpoint, then
#2 - inserts a sw breakpoint at the same address, and then
#3 - removes the original hw breakpoint,
GDB would have to make sure to insert the sw breakpoint before
removing the hw breakpoint, to avoid running threads missing the
breakpoint. I.e., there's always going to be a window where a target
needs to be able to handle both sw and a hw breakpoints installed at
the same address. You can see more detailed description of that issue
here:
https://sourceware.org/pipermail/gdb-patches/2020-April/167738.html
So the fix here is to just stop considering software breakpoints and
hw breakpoints duplicates, and let GDB insert sw and hw breakpoints at
the same address.
The central change is to make breakpoint_locations_match consider the
location's type too. There are several other changes necessary to
actually make that that work correctly, however:
- We need to handle the duplicates detection better. Take a look at
the loop at the tail end of update_global_location_list. Currently,
because breakpoint locations aren't sorted by type, we can end up
with, at the same address, a sw break, then a hw break, then a sw
break, etc. The result is that that second sw break won't be
considered a duplicate of the first sw break. Seems like we already
handle that incorrectly for range breakpoints.
- The "set breakpoint auto-hw on" handling is moved out of
insert_bp_location to update_global_location_list, before the
duplicates determination.
Moving "set breakpoint auto-hw off" handling as well and downgrading
it to a warning+'disabling the location' was considered too, but in
the end discarded, because we want to error out with internal and
momentary breakpoints, like software single-step breakpoints.
Disabling such locations at update_global_location_list time would
make GDB lose control of the inferior.
- In update_breakpoint_locations, the logic of matching old locations
with new locations, in the have_ambiguous_names case, is updated to
still consider sw vs hw locations the same.
- Review all ALL_BP_LOCATIONS_AT_ADDR uses, and update those that
might need to be updated, and update comments for those that don't.
Note that that macro walks all locations at a given address, and
doesn't call breakpoint_locations_match.
The result against GDBserver (with "set breakpoint
condition-evaluation host" to avoid seeing confusing reinsertions) is:
(gdb) hbreak main
Sending packet: $m400736,1#fe...Packet received: 48
Hardware assisted breakpoint 1 at 0x400736: file main.c, line 57.
Sending packet: $Z1,400736,1#48...Packet received: OK
(gdb) b main
Note: breakpoint 1 also set at pc 0x400736.
Sending packet: $m400736,1#fe...Packet received: 48
Breakpoint 4 at 0x400736: file main.c, line 57.
Sending packet: $Z0,400736,1#47...Packet received: OK
(gdb) del 3
Sending packet: $z1,400736,1#68...Packet received: OK
gdb/ChangeLog:
2020-05-17 Pedro Alves <palves@redhat.com>
Andrew Burgess <andrew.burgess@embecosm.com>
Keno Fischer <keno@juliacomputing.com>
PR gdb/25741
* breakpoint.c (build_target_condition_list): Update comments.
(build_target_command_list): Update comments and skip matching
locations.
(insert_bp_location): Move "set breakpoint auto-hw on" handling to
a separate function. Simplify "set breakpoint auto-hw off"
handling.
(insert_breakpoints): Update comment.
(tracepoint_locations_match): New parameter. For breakpoints,
compare location types too, if the caller wants to.
(handle_automatic_hardware_breakpoints): New functions.
(bp_location_is_less_than): Also sort by location type and
hardware breakpoint length.
(update_global_location_list): Handle "set breakpoint auto-hw on"
here.
(update_breakpoint_locations): Ask breakpoint_locations_match to
ignore location types.
gdb/testsuite/ChangeLog:
2020-05-17 Pedro Alves <palves@redhat.com>
PR gdb/25741
* gdb.base/hw-sw-break-same-address.exp: New file.
Pedro Alves [Sat, 16 May 2020 17:59:10 +0000 (18:59 +0100)]
Fix gdb.multi/multi-re-run.exp with native-gdbserver
The new exec-file-mismatch feature reveals that when running
gdb.multi/multi-re-run.exp against --target_board=native-gdbserver,
we've been starting gdbserver with the wrong program:
(gdb) spawn /home/pedro/brno/pedro/gdb/binutils-gdb/build/gdb/testsuite/../../gdbserver/gdbserver --once localhost:2347 /home/pedro/brno/pedro/gdb/binutils-gdb/build/gdb/te
stsuite/outputs/gdb.multi/multi-re-run/multi-re-run-2
Process /home/pedro/brno/pedro/gdb/binutils-gdb/build/gdb/testsuite/outputs/gdb.multi/multi-re-run/multi-re-run-2 created; pid = 6280
Listening on port 2347
target remote localhost:2347
Remote debugging using localhost:2347
warning: Mismatch between current exec-file /home/pedro/brno/pedro/gdb/binutils-gdb/build/gdb/testsuite/outputs/gdb.multi/multi-re-run/multi-re-run-1
and automatically determined exec-file /home/pedro/brno/pedro/gdb/binutils-gdb/build/gdb/testsuite/outputs/gdb.multi/multi-re-run/multi-re-run-2
exec-file-mismatch handling is currently "ask"
Load new symbol table from "/home/pedro/brno/pedro/gdb/binutils-gdb/build/gdb/testsuite/outputs/gdb.multi/multi-re-run/multi-re-run-2"? (y or n) Quit
(gdb) FAIL: gdb.multi/multi-re-run.exp: re_run_inf=1: iter=1: running to all_started in runto
The problem is that gdb_reload uses the last loaded file as binary to
spawn, but we load the program for inferior 2 and then switch to
inferior 1 and run it, so the last loaded file is the program for
inferior 2.
Fix this by tweaking last_loaded_file.
gdb/testsuite/ChangeLog:
2020-05-16 Pedro Alves <palves@redhat.com>
Simon Marchi [Sat, 16 May 2020 16:16:06 +0000 (12:16 -0400)]
gdb: remove TYPE_NAME macro
Remove `TYPE_NAME`, changing all the call sites to use `type::name`
directly. This is quite a big diff, but this was mostly done using sed
and coccinelle. A few call sites were done by hand.
gdb/ChangeLog:
* gdbtypes.h (TYPE_NAME): Remove. Change all cal sites to use
type::name instead.
Simon Marchi [Sat, 16 May 2020 16:15:54 +0000 (12:15 -0400)]
gdb: add type::name / type::set_name
Add the `name` and `set_name` methods on `struct type`, in order to
remove the `TYPE_NAME` macro. In this patch, the `TYPE_NAME` macro is
changed to use `type::name`, so all the call sites that are used to set
the type name are changed to use `type::set_name`. The next patch will
remove `TYPE_NAME` completely.
gdb/ChangeLog:
* gdbtypes.h (struct type) <name, set_name>: New methods.
(TYPE_CODE): Use type::name. Change all call sites used to set
the name to use type::set_name instead.
Pedro Alves [Sat, 16 May 2020 11:26:56 +0000 (12:26 +0100)]
Fix IA64 GNU/Linux build
This commit should fix:
../../gdb/ia64-linux-nat.c: In function ‘void enable_watchpoints_in_psr(ptid_t)’:
../../gdb/ia64-linux-nat.c:535:56: error: no matching function for call to ‘get_thread_regcache(ptid_t&)’
struct regcache *regcache = get_thread_regcache (ptid);
^
In file included from ../../gdb/ia64-linux-nat.c:25:0:
../../gdb/regcache.h:35:25: note: candidate: regcache* get_thread_regcache(process_stratum_target*, ptid_t)
extern struct regcache *get_thread_regcache (process_stratum_target *target,
^
../../gdb/regcache.h:35:25: note: candidate expects 2 arguments, 1 provided
../../gdb/regcache.h:39:25: note: candidate: regcache* get_thread_regcache(thread_info*)
extern struct regcache *get_thread_regcache (thread_info *thread);
^
../../gdb/regcache.h:39:25: note: no known conversion for argument 1 from ‘ptid_t’ to ‘thread_info*’
gdb/ChangeLog:
2020-05-16 Pedro Alves <palves@redhat.com>
* ia64-linux-nat.c
(ia64_linux_nat_target) <enable_watchpoints_in_psr(ptid_t)>:
Declare method.
(enable_watchpoints_in_psr): Now a method of ia64_linux_nat_target.
Simon Marchi [Sat, 16 May 2020 02:17:40 +0000 (22:17 -0400)]
gdb: remove unnecessary struct typedef in sparc64-tdep.c
When building with clang 11, I get:
CXX sparc64-tdep.o
/home/smarchi/src/binutils-gdb/gdb/sparc64-tdep.c:89:15: error: anonymous non-C-compatible type given name for linkage purposes by typedef declaration; add a tag name here [-Werror,-Wnon-c-typedef-for-linkage]
typedef struct
^
adi_stat_t
/home/smarchi/src/binutils-gdb/gdb/sparc64-tdep.c:103:16: note: type is not C-compatible due to this default member initializer
int tag_fd = 0;
^
/home/smarchi/src/binutils-gdb/gdb/sparc64-tdep.c:111:3: note: type is given name 'adi_stat_t' for linkage purposes by this typedef declaration
} adi_stat_t;
^
The typedef is not needed in C++ anyway, just remove them.
Tom Tromey [Fri, 15 May 2020 22:24:07 +0000 (16:24 -0600)]
Remove lookup_objfile_from_block
lookup_objfile_from_block mostly duplicates the functionality of
block_objfile, but in a less efficient way. This patch removes this
function and changes the callers to use block_objfile instead.
Tested by the buildbot.
gdb/ChangeLog
2020-05-15 Tom Tromey <tom@tromey.com>
* symtab.c (lookup_language_this, lookup_symbol_aux): Use
block_objfile.
(lookup_objfile_from_block): Remove.
(lookup_symbol_in_block, lookup_symbol_in_static_block)
(lookup_global_symbol): Use block_objfile.
* symtab.h (lookup_objfile_from_block): Don't declare.
* printcmd.c (clear_dangling_display_expressions): Use
block_objfile.
* parse.c (operator_check_standard): Use block_objfile.
Tom Tromey [Fri, 15 May 2020 22:11:33 +0000 (16:11 -0600)]
Remove allocate_symbol et al
This removes allocate_symbol, allocate_template_symbol, and
initialize_objfile_symbol in favor of changing the default values for
symbol members, and updating the one per-arch caller.
gdb/ChangeLog
2020-05-15 Tom Tromey <tom@tromey.com>
* language.c (language_alloc_type_symbol): Set
SYMBOL_SECTION.
* symtab.c (initialize_objfile_symbol): Remove.
(allocate_symbol): Remove.
(allocate_template_symbol): Remove.
* dwarf2/read.c (fixup_go_packaging): Use "new".
(new_symbol): Use "new".
(read_variable): Don't call initialize_objfile_symbol. Use
"new".
(read_func_scope): Use "new".
* xcoffread.c (process_xcoff_symbol): Don't call
initialize_objfile_symbol.
(SYMBOL_DUP): Remove.
* coffread.c (process_coff_symbol, coff_read_enum_type): Use
"new".
* symtab.h (allocate_symbol, initialize_objfile_symbol)
(allocate_template_symbol): Don't declare.
(struct symbol): Add copy constructor. Change defaults.
* jit.c (finalize_symtab): Use "new".
* ctfread.c (ctf_add_enum_member_cb, new_symbol, ctf_add_var_cb):
Use "new".
* stabsread.c (patch_block_stabs, define_symbol, read_enum_type)
(common_block_end): Use "new".
* mdebugread.c (parse_symbol): Use "new".
(new_symbol): Likewise.
Ensure class_alias is only used for user-defined aliases.
This commit finally does the (small) change that started this patch
series.
It ensures that the class_alias is only used for user-defined aliases.
So, the few GDB pre-defined aliases that were using the 'class_alias'
class are now using a real help class, typically the class of
the aliased command.
gdb/ChangeLog
2020-05-15 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* command.h (enum command_class): Improve comments, document
that class_alias is for user-defined aliases, give the class
name for each class, remove unused class_xdb.
* cli/cli-decode.c (add_com_alias): Document THECLASS intended usage.
* breakpoint.c (_initialize_breakpoint): Replace class_alias
by a precise class.
* infcmd.c (_initialize_infcmd): Likewise.
* reverse.c (_initialize_reverse): Likewise.
* stack.c (_initialize_stack): Likewise.
* symfile.c (_initialize_symfile): Likewise.
* tracepoint.c (_initialize_tracepoint): Likewise.
gdb/testsuite/ChangeLog
2020-05-15 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* gdb.base/alias.exp: Verify 'help aliases' shows user defined aliases.
Similarly to 'help CLASS', apropos possibly shows several
times the same help (for the command and for each of its aliases).
This patch changes 'apropos' so that the help for a command and
all its aliases is shown once.
So, apropos_cmd now skips all aliases/abbreviations, as these are printed
as part of the help of the aliased command.
When 'apropos' prints the help of a command, function 'help_cmd' now
unconditionally print the command name and its possible aliases (as we must
indicate to the user the command/aliases for which the help is printed).
When 'help somecommand' prints the help of a command, if the command is not
aliased, the command name is not printed (to avoid a useless first line), but if
it has aliases, then the command name and all its aliases are now printed.
In addition to provide to the user the choice of the best way to
type a command, it also avoids the strange behaviour that the output
of 'help somealias' does not mention somealias.
gdb/ChangeLog
2020-05-15 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* cli/cli-decode.c (apropos_cmd): Produce output for aliases
when their aliased command is traversed.
(help_cmd): Add fput_command_names_styled call to
output command name and aliases when command has an alias.
gdb/testsuite/ChangeLog
2020-05-15 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* gdb.base/help.exp: Test apropos and help for commands
having aliases. Fixed comments not starting with an
upper-case letter or not finishing with a dot.
Currently, help CLASS possibly shows several times the same help,
as it shows it once for the command, and once for each alias.
The final objective of this patch series is to have class_alias used only
for user defined aliases, not anymore for aliases predefined by GDB.
The command 'help aliases' will then only show the user defined aliases.
So, the idea is that GDB predefined aliases will be shown together
with their aliased command.
This commit changes 'help CLASS' so that a command is shown once in the output,
with all its aliases.
This ensures:
* that the user has only to read once the same help text
* and sees the command and all its aliases in a glance, a.o. allowing
the user to choose the preferred way (e.g. the shortest one,
or the most mnemonic one) to type the command.
For example, the old output:
(gdb) help stack
...
List of commands:
backtrace -- Print backtrace of all stack frames, or innermost COUNT frames.
bt -- Print backtrace of all stack frames, or innermost COUNT frames.
...
(note that 'where' is not shown in this output)
becomes
(gdb) help stack
...
List of commands:
backtrace, where, bt -- Print backtrace of all stack frames, or innermost COUNT frames.
...
The output layout chosen is to have the command first, followed by all its
aliases separated by a comma. Note that the command and alias names are
title-styled. For sure, other layouts could be discussed, but this one is IMO
readable and compact.
The function 'help_cmd_list' can be simplified by removing the prefix argument,
as the prefixname of a command can now be retrieved in the GDB command tree
structure.
This also fixes the fact that 'help aliases' wrongly shows a long
list of (non-alias) when defining an alias for a prefix command.
For example, after:
(gdb) alias montre = show
then
(gdb) help aliases
shows hundreds of sub-commands starting with the non aliased command,
such as:
montre -- Generic command for showing things about the debugger.
show ada -- Generic command for showing Ada-specific settings.
show ada print-signatures -- Show whether the output of formal ...
....
'help_cmd_list' is also made static, as it is only used inside cli-decode.c.
Note that the 'help CLASS' is somewhat broken, in the sense that it
sometimes shows too many commands (commands not belonging to CLASS)
and sometimes shows not enough commands (not showing some commands
belonging to CLASS).
For example, 'help breakpoints' shows the command
'disable pretty-printer' and 'disable unwinder', not related to breakpoints.
On the other end, 'help stack' does not show 'disable unwinder'
while 'disable unwinder' is defined in unwinders.py as belonging to class_stack.
Fixing the missing commands is easy to do,
but fixing the excess commands is not straightforward, as many
subcommands have a class 'no_class' or 'all_class'.
Possibly, some of this might be improved/fixed in another patch series.
With this patch series, the 'abbrev flag' has as only remaining purpose
to avoid having the abbreviation alias appearing in the completion list,
so change 'help alias' accordingly.
gdb/ChangeLog
2020-05-15 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* cli/cli-decode.h (help_cmd_list): Remove declaration.
* cli/cli-decode.c (help_cmd_list): Declare as static,
remove prefix argument, use bool for recurse arg, rework to show the aliases of
a command together with the command.
(fput_command_name_styled, fput_command_names_styled): New functions.
(print_help_for_command): Remove prefix arg, use bool for recurse arg, use
fput_command_name_styled.
(help_list, help_all): Update callers to remove prefix arg and use bool recurse.
* cli/cli-cmds.c (_initialize_cli_cmds): Update alias_command doc.
gdb/testsuite/ChangeLog
2020-05-15 Philippe Waroquiers <philippe.waroquiers@skynet.be>
Fix inconsistent output of prefix and bugs in 'show' command
cmd_show_list function implements the 'show' command.
cmd_show_list output is inconsistent: it sometimes shows a prefix
and sometimes does not.
For example, in the below, you see that there is a prefix before
each value, except for 'enabled'.
(gdb) show style
style address background: The "address" style background color is: none
style address foreground: The "address" style foreground color is: blue
style address intensity: The "address" style display intensity is: normal
enabled: CLI output styling is enabled.
style filename background: The "filename" style background color is: none
...
There are other inconsistencies or bugs e.g. in
the below we see twice insn-number-max, once with a prefix
and once without prefix : last line, just before the value of
instruction-history-size which is itself without prefix.
(gdb) show record
record btrace bts buffer-size: The record/replay bts buffer size is 65536.
record btrace cpu: btrace cpu is 'auto'.
record btrace pt buffer-size: The record/replay pt buffer size is 16384.
record btrace replay-memory-access: Replay memory access is read-only.
record full insn-number-max: Record/replay buffer limit is 200000.
record full memory-query: Whether query if PREC cannot record memory change of next instruction is off.
record full stop-at-limit: Whether record/replay stops when record/replay buffer becomes full is on.
function-call-history-size: Number of functions to print in "record function-call-history" is 10.
insn-number-max: instruction-history-size: Number of instructions to print in "record instruction-history" is 10.
(gdb)
Also, some values are output several times due to some aliases, so avoid outputting duplicated
values by skipping all aliases.
Now that the command structure has a correct 'back-pointer' from a command
to its prefix command, we can simplify cmd_show_list by removing its prefix argument
and at the same time fix the output inconsistencies and bugs.
gdb/ChangeLog
2020-05-15 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* cli/cli-setshow.h (cmd_show_list): Remove prefix argument.
* cli/cli-decode.c (do_show_prefix_cmd): Likewise.
* command.h (cmd_show_list): Likewise.
* dwarf2/index-cache.c (show_index_cache_command): Likewise.
* cli/cli-setshow.c (cmd_show_list): Use the prefix to produce the output. Skip aliases.
gdb/testsuite/ChangeLog
2020-05-15 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* gdb.base/default.exp: Update output following fixes.
command-def-selftests.c: detect missing or wrong prefix cmd in subcommands.
This test revealed a number of problems that are fixed in the previous commit.
2020-05-15 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* unittests/command-def-selftests.c (traverse_command_structure):
Verify all commands of a list have the same prefix command and
that only the top cmdlist commands have a null prefix.
Fix the problems reported by prefix check of command-def-selftests.c
The next commit updates command-def-selftests.c to detect missing
or wrong prefix commands in a list of subcommands.
This command structure selftest detects a series of problems
that are fixed by this commit.
Many commands have a null prefix command, e.g.
(gdb) maintenance selftest command_str
Running selftest command_structure_invariants.
list 0x560417949cb8 reachable via prefix 'append binary '. command 'memory' has null prefixcmd
list 0x560417949cb8 reachable via prefix 'append binary '. command 'value' has null prefixcmd
...
Most of these are fixed by the following changes:
* do_add_cmd searches the prefix command having the list
in which the command is added.
This ensures that a command defined after its prefix command
gets the correct prefix command.
* Due to the GDB initialization order, a GDB file can define
a subcommand before the prefix command is defined.
So, have add_prefix_cmd calling a new recursive function
'update_prefix_field_of_prefix_commands' to set the prefix
command of all sub-commands that are now reachable from
this newly defined prefix command. Note that this recursive
call replaces the function 'set_prefix_cmd' that was providing
a partial solution to this problem.
Following that, 2 python commands (defined after all the other GDB
commands) got a wrong prefix command, e.g. "info frame-filter" has
as prefix command the "i" alias of "info". This is fixed by having
lookup_cmd_for_prefixlist returning the aliased command rather than
the alias.
After that, one remaining problem:
(gdb) maintenance selftest command_str
Running selftest command_structure_invariants.
list 0x55f320272298 reachable via prefix 'set remote '. command 'system-call-allowed' has null prefixcmd
Self test failed: self-test failed at ../../classfix/gdb/unittests/command-def-selftests.c:196
Ran 1 unit tests, 1 failed
(gdb)
Caused by initialize_remote_fileio that was taking the address of
its arguments remote_set_cmdlist and remote_show_cmdlist instead
of receiving the correct values to use as list.
2020-05-15 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* cli/cli-decode.c (lookup_cmd_for_prefix): Return the aliased command
as prefix, not one of its aliases.
(set_cmd_prefix): Remove.
(do_add_cmd): Centralize the setting of the prefix of a command, when
command is defined after its full chain of prefix commands.
(add_alias_cmd): Remove call to set_cmd_prefix, as do_add_cmd does it.
(add_setshow_cmd_full): Likewise.
(update_prefix_field_of_prefixed_commands): New function.
(add_prefix_cmd): Replace non working call to set_cmd_prefix by
update_prefix_field_of_prefixed_commands.
* gdb/remote-fileio.c (initialize_remote_fileio): Use the real
addresses of remote_set_cmdlist and remote_show_cmdlist given
as argument, not the address of an argument.
* gdb/remote-fileio.h (initialize_remote_fileio): Likewise.
* gdb/remote.c (_initialize_remote): Likewise.
Fix problem that alias can be defined or not depending on the order.
When an alias name starts with the name of another alias,
GDB was accepting to define the aliases in one order (short first, long after),
but refused it the other way around.
So, fix the logic to recognise an already existing alias by using
lookup_cmd_composition.
Also, this revealed a bug in lookup_cmd_composition:
when the searched command is a prefix command, lookup_cmd_composition
was not returning the fact that a command was found even if the
TEXT to parse was fully consumed.
gdb/ChangeLog
YYYY-MM-DD Philippe Waroquiers <philippe.waroquiers@skynet.be>
* cli/cli-cmds.c (alias_command): Check for an existing alias
using lookup_cmd_composition, as valid_command_p is too strict
and forbids aliases that are the prefix of an existing alias
or command.
* cli/cli-decode.c (lookup_cmd_composition): Ensure a prefix
command is properly recognised as a valid command.
gdb/testsuite/ChangeLog
2020-05-15 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* gdb.base/alias.exp: Test aliases starting with a prefix of
another alias.
Add a selftest that detects a 'corrupted' command tree structure in GDB.
The GDB data structure that records the GDB commands is made of
'struct cmd_list_element' defined in cli-decode.h.
A cmd_list_element has various pointers to other cmd_list_element structures,
All these pointers are together building a graph of commands.
However, when following the 'next' and '*prefixlist' pointers of
cmd_list_element, the structure must better be a tree.
If such pointers do not form a tree, then some other elements of
cmd_list_element cannot get a correct semantic. In particular, the prefixname
has no correct meaning if the same prefix command can be reached via 2 different
paths.
This commit introduces a selftest that detects (at least some cases of) errors
leading to 'next' and '*prefixlist' not giving a tree structure.
The new 'command_structure_invariants' selftest detects one single case where
the command structure is not a tree:
(gdb) maintenance selftest command_structure_invariants
Running selftest command_structure_invariants.
list 0x56362e204b98 duplicated, reachable via prefix 'show ' and 'info set '. Duplicated list first command is 'ada'
Self test failed: self-test failed at ../../classfix/gdb/unittests/command-def-selftests.c:160
Ran 1 unit tests, 1 failed
(gdb)
This was fixed by the previous commit.
2020-05-15 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* unittests/help-doc-selftests.c: Rename to
unittests/command-def-selftests.c
* unittests/command-def-selftests.c (help_doc_tests): Update some
comments.
(command_structure_tests, traverse_command_structure): New namespace
and function.
(command_structure_invariants_tests): New function.
(_initialize_command_def_selftests) Renamed from
_initialize_help_doc_selftests, register command_structure_invariants
selftest.
Fix the only incorrect case found by command_structure_invariants selftest.
The next commit introduces a selftest that detects when the GDB
command structure does not define a tree when using the pointers
'next/*prefixlist'. This test detects one such case, fixed
by this commit.
The command 'info set' was defined as a specific prefix command,
but re-using the command list already used for the 'show' command.
This leads to the command tree 'next/*prefixlist' to not be a tree.
This change defines 'info set ' as an alias, thereby fixing the selftest.
2020-05-15 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* cli/cli-cmds.c (_initialize_cli_cmds): Define 'info set' as
an alias of 'show'.
Joel Brobecker [Fri, 15 May 2020 20:06:42 +0000 (15:06 -0500)]
update name of several Ada fixed-point type handling functions
The purpose of this patch is to prepare for the future where
fixed point types become described using standard DWARF info,
rather than GNAT encodings. For that, we rename a number of
routines manipulating Ada fixed point types to make it explicit
from their new names that they rely on the GNAT encodings to work.
This will allow us, when we introduce support for fixed point types
from standard DWARF to use names that are not ambiguous with
the functions that do similar work, but only for GNAT encodings.
gdb/ChangeLog:
* ada-lang.h: (ada_is_gnat_encoded_fixed_point_type): Renames
ada_is_fixed_point_type. Update all callers.
(gnat_encoded_fixed_point_delta): Renames ada_delta. Update
all callers.
* ada-lang.c (gnat_encoded_fixed_type_info): Renames fixed_type_info.
Update all callers.
* ada-typeprint.c (print_gnat_encoded_fixed_point_type): Renames
print_fixed_point_type. Update all callers.
* ada-valprint.c (ada_value_print_num): Replace call to
ada_is_fixed_point_type by ada_is_gnat_encoded_fixed_point_type.
Gary Benson [Fri, 15 May 2020 14:03:42 +0000 (15:03 +0100)]
Fix "control reaches end of non-void function" errors in testsuite
When running the testsuite with clang, a number of testcases fail to
build with the following errors:
warning: control reaches end of non-void function [-Wreturn-type]
warning: control may reach end of non-void function [-Wreturn-type]
This prevents a number of testcases from executing. This commit fixes.
Tom de Vries [Fri, 15 May 2020 12:49:48 +0000 (14:49 +0200)]
[gdb/testsuite] Rename *.exp.in to *.exp.tcl
Say we have some common tcl code that we want to include in test-cases
t1.exp and t1.exp.
We could put the common code into a file common.exp alongside the test-cases,
but that will make dejagnu treat that file as another test-case. To prevent
this, we use a suffix, currently .in, in other words we put the common code in
a file common.exp.in.
The .in suffix however is also used in autoconf, which might cause confusion.
This commit is for the benefit of GDB, but as the binutils-gdb
repository shares the contrib/ directory with gcc, this commit must
first be applied to gcc then copied back to binutils-gdb.
This commit extends the two scripts contrib/dg-extract-results.{py,sh}
to handle some new, GDB specific test result types. These test
results types should never appear in GCC, or any other tool that
shares the contrib/ directly, so this change should be harmless.
In this patch series:
https://sourceware.org/pipermail/gdb-patches/2020-April/167847.html
changes were made in GDB's use of Dejagnu so that two additional
conditions could be detected, these are:
1. Test names that contain either the build or source paths. Such
test names make it difficult to compare the results of two test runs
of GDB from two different directories, and
2. Duplicate test names. Duplicates make it difficult to track down
exactly which test has failed.
When running Dejagnu on GDB we can now (sometimes) see two additional
test result types matching the above conditions, these are '# of paths
in test names' and '# of duplicate test names'.
If the test is run in parallel mode (make -j...) then these extra test
results will appear in the individual test summary files, but are not
merged into the final summary file.
Additionally, within the summary file there are now two new types of
test summary line, these are 'PATH: ...' and 'DUPLICATE: ...', these
allow users to quickly search the test summary to track down where the
offending test names are. These lines are similarly not merged into
the unified gdb.sum file after a parallel test run.
This commit extends the dg-extract-results.* scripts to calculate the
totals for the two new result types, and to copy the new test summary
lines into the unified summary file.
contrib/ChangeLog:
* dg-extract-results.py: Update from gcc repo.
* dg-extract-results.sh: Likewise.
Pedro Alves [Fri, 15 May 2020 10:22:47 +0000 (11:22 +0100)]
Fix gdb.multi/multi-kill.exp
The previous patch misssed declaring the 'testpid' array as namespace
variable. While it at, might as well go back to having start_inferior
refer to the "global" testpid, using "variable" too.
gdb/testsuite/ChangeLog:
2020-05-15 Pedro Alves <palves@redhat.com>
* gdb.multi/multi-kill.exp (start_inferior): Remove
'testpid' parameter. Refer to namespace variable directly.
(testpid): Declare as namespace variable.
Pedro Alves [Fri, 15 May 2020 10:09:51 +0000 (11:09 +0100)]
Fix global variable collision in gdb.multi/multi-kill.exp
The new gdb.multi/multi-kill.exp testcase added an 'testpid' array,
which may conflict with other global 'testpid' variables used by other
testcases, resulting in:
...
ERROR: tcl error sourcing
/data/gdb_versions/devel/src/gdb/testsuite/gdb.multi/multi-kill.exp.
ERROR: can't set "testpid(1)": variable isn't array
while executing
"set testpid($num) [get_integer_valueof "pid" -1]"
or
$ runtest gdb.threads/check-libthread-db.exp gdb.multi/multi-kill.exp
...
Running /home/pedro/gdb/binutils-gdb/src/gdb/testsuite/gdb.multi/multi-kill.exp ...
Running /home/pedro/gdb/binutils-gdb/src/gdb/testsuite/gdb.threads/check-libthread-db.exp ...
ERROR: tcl error sourcing /home/pedro/gdb/binutils-gdb/src/gdb/testsuite/gdb.threads/check-libthread-db.exp.
ERROR: can't set "testpid": variable is array
while executing
"set testpid [spawn_id_get_pid $test_spawn_id]"
("uplevel" body line 8)
Fix this with a namespace, like gdb.linespec/explicit.exp does.
gdb/testsuite/ChangeLog:
2020-05-15 Pedro Alves <palves@redhat.com>
* gdb.multi/multi-kill.exp: Wrap in namespace.
(start_inferior): Add TESTPID parameter. Use it instead of the
testpid global.
(top level): Define empty TESTPID array, and pass it down to
start_inferior.
Hannes Domani [Mon, 11 May 2020 17:18:31 +0000 (19:18 +0200)]
Enable hardware breakpoints for gdbserver on Windows
When trying to use hardware breakpoints with gdbserver you get this error:
(gdb) hbreak main
Hardware assisted breakpoint 2 at 0x40162d: file gdb-9493.c, line 5.
(gdb) c
Continuing.
Warning:
Cannot insert hardware breakpoint 2.
Could not insert hardware breakpoints:
You may have requested too many hardware breakpoints/watchpoints.
It turns out the respective types just needed to be added to the
appropriate callback functions, because x86_dr_(insert|remove)_watchpoint
already handles them.
Alan Modra [Fri, 15 May 2020 08:36:05 +0000 (18:06 +0930)]
Fix tight loop on recursively-defined symbols
This patch fixes a bug in GAS where the assembler enters a tight loop
when attempting to resolve recursively-defined symbols, e.g. when
trying to assemble "a=a".
This is a regression introduced between binutils 2.32 and 2.33,
by commit 1903f1385bff9
* symbols.c (struct local_symbol): Update comment.
(resolve_symbol_value): For resolved symbols equated to other
symbols, verify that the referenced symbol is not a local_symbol
before accessing sy_value. Don't leave symbol loops during
finalize_syms resolution.
* testsuite/gas/all/assign-bad-recursive.d: New test.
* testsuite/gas/all/assign-bad-recursive.l: Error output for test.
* testsuite/gas/all/assign-bad-recursive.s: Assembly for test.
* testsuite/gas/all/gas.exp: Run it.
While it is the case that AMD processors have an LBR (last branch
record) bit in the DebugCtl MSR, it seems that it affects only four
MSRs when enabled. The names of these MSRs are LastBranchToIP,
LastBranchFromIP, LastIntToIP, and LastIntFromIP. I can find no
mention of anything more extensive. While looking at an Intel
architecture document, I noticed that Intel's P6 family from the
mid-90s had registers of the same name.
Therefore...
This commit disables "record btrace bts" support in GDB for AMD
processors.
Using the test case from gdb.base/break.exp, the sessions
below show the expected behavior (run on a machine with an
Intel processor) versus that on a machine with an AMD processor.
The AMD processor in question is reported as follows by "lscpu":
AMD Ryzen Threadripper 2950X 16-Core Processor . Finally, I'll
note that the AMD machine is actually a VM, but I see similar
behavior on both the virtualization host and the VM.
Intel machine - Desired behavior:
[kevinb@mohave gdb]$ ./gdb -q testsuite/outputs/gdb.base/break/break
Reading symbols from testsuite/outputs/gdb.base/break/break...
(gdb) start
Temporary breakpoint 1 at 0x401179: file /home/kevinb/sourceware-git/native-build/bld/../../binutils-gdb/gdb/testsuite/gdb.base/break.c, line 43.
Starting program: /home/kevinb/sourceware-git/native-build/bld/gdb/testsuite/outputs/gdb.base/break/break
Temporary breakpoint 1, main (argc=1, argv=0x7fffffffd748, envp=0x7fffffffd758)
at /home/kevinb/sourceware-git/native-build/bld/../../binutils-gdb/gdb/testsuite/gdb.base/break.c:43
43 if (argc == 12345) { /* an unlikely value < 2^16, in case uninited */ /* set breakpoint 6 here */
(gdb) record btrace
(gdb) b factorial
Breakpoint 2 at 0x40121b: file /home/kevinb/sourceware-git/native-build/bld/../../binutils-gdb/gdb/testsuite/gdb.base/break.c, line 63.
(gdb) c
Continuing.
Breakpoint 2, factorial (value=6)
at /home/kevinb/sourceware-git/native-build/bld/../../binutils-gdb/gdb/testsuite/gdb.base/break.c:63
63 if (value > 1) { /* set breakpoint 7 here */
(gdb) info record
Active record target: record-btrace
Recording format: Branch Trace Store.
Buffer size: 64kB.
Recorded 768 instructions in 22 functions (0 gaps) for thread 1 (process 19215).
(gdb) record function-call-history
13 do_lookup_x
14 _dl_lookup_symbol_x
15 _dl_fixup
16 _dl_runtime_resolve_xsavec
17 atoi
18 strtoq
19 ____strtoll_l_internal
20 atoi
21 main
22 factorial
(gdb) record instruction-history
759 0x00007ffff7ce0917 <____strtoll_l_internal+647>: pop %r15
760 0x00007ffff7ce0919 <____strtoll_l_internal+649>: retq
761 0x00007ffff7cdd064 <atoi+20>: add $0x8,%rsp
762 0x00007ffff7cdd068 <atoi+24>: retq
763 0x00000000004011b1 <main+75>: mov %eax,%edi
764 0x00000000004011b3 <main+77>: callq 0x401210 <factorial>
765 0x0000000000401210 <factorial+0>: push %rbp
766 0x0000000000401211 <factorial+1>: mov %rsp,%rbp
767 0x0000000000401214 <factorial+4>: sub $0x10,%rsp
768 0x0000000000401218 <factorial+8>: mov %edi,-0x4(%rbp)
AMD machine - Wrong behavior:
[kev@f32-1 gdb]$ ./gdb -q testsuite/outputs/gdb.base/break/break
Reading symbols from testsuite/outputs/gdb.base/break/break...
(gdb) start
Temporary breakpoint 1 at 0x401179: file /ironwood1/sourceware-git/f32-master/bld/../../worktree-master/gdb/testsuite/gdb.base/break.c, line 43.
Starting program: /mesquite2/sourceware-git/f32-master/bld/gdb/testsuite/outputs/gdb.base/break/break
Temporary breakpoint 1, main (argc=1, argv=0x7fffffffd5b8, envp=0x7fffffffd5c8)
at /ironwood1/sourceware-git/f32-master/bld/../../worktree-master/gdb/testsuite/gdb.base/break.c:43
43 if (argc == 12345) { /* an unlikely value < 2^16, in case uninited */ /* set breakpoint 6 here */
(gdb) record btrace
(gdb) b factorial
Breakpoint 2 at 0x40121b: file /ironwood1/sourceware-git/f32-master/bld/../../worktree-master/gdb/testsuite/gdb.base/break.c, line 63.
(gdb) c
Continuing.
Breakpoint 2, factorial (value=6)
at /ironwood1/sourceware-git/f32-master/bld/../../worktree-master/gdb/testsuite/gdb.base/break.c:63
63 if (value > 1) { /* set breakpoint 7 here */
(gdb) info record
Active record target: record-btrace
Recording format: Branch Trace Store.
Buffer size: 64kB.
warning: Recorded trace may be incomplete at instruction 7737 (pc = 0x405000).
warning: Recorded trace may be incomplete at instruction 7739 (pc = 0x0).
Recorded 7740 instructions in 46 functions (2 gaps) for thread 1 (process 1402911).
(gdb) record function-call-history
37 ??
38 values
39 some_enum_global
40 ??
41 some_union_global
42 some_variable
43 ??
44 [decode error (2): unknown instruction]
45 ??
46 [decode error (2): unknown instruction]
(gdb) record instruction-history
7730 0x0000000000404ff3: add %al,(%rax)
7731 0x0000000000404ff5: add %al,(%rax)
7732 0x0000000000404ff7: add %al,(%rax)
7733 0x0000000000404ff9: add %al,(%rax)
7734 0x0000000000404ffb: add %al,(%rax)
7735 0x0000000000404ffd: add %al,(%rax)
7736 0x0000000000404fff: .byte 0x0
7737 0x0000000000405000: Cannot access memory at address 0x405000
Lastly, I'll note that I see a lot of gdb.btrace failures without
this commit. Worse still, the results aren't always the same which
causes a lot of noise when comparing test results.
gdb: infrun: consume multiple events at each pass in stop_all_threads
[Simon: I send this patch on behalf of Laurent Morichetti, I added the
commit message and performance measurement stuff.
Also, this patch is better viewed with "git show -w".]
stop_all_threads, in infrun.c, is used to stop all running threads on
targets that are always non-stop. It's used, for example, when the
program hits a breakpoint while GDB is set to "non-stop off". It sends
a stop request for each running thread, then collects one wait event for
each.
Since new threads can spawn while we are stopping the threads, it's
written in a way where it makes multiple such "send stop requests to
running threads & collect wait events" passes. The function completes
when it has made two passes where it hasn't seen any running threads.
With the way it's written right now is, it iterates on the thread list,
sending a stop request for each running thread. It then waits for a
single event, after which it iterates through the thread list again. It
sends stop requests for any running threads that's been created since
the last iteration. It then consumes another single wait event.
This makes it so we iterate on O(n^2) threads in total, where n is the
number of threads. This patch changes the function to reduce it to
O(n). This starts to have an impact when dealing with multiple
thousands of threads (see numbers below). At each pass, we know the
number of outstanding stop requests we have sent, for which we need to
collect a stop event. We can therefore loop to collect this many stop
events before proceeding to the next pass and iterate on the thread list
again.
To check the performance improvements with this patch, I made an
x86/Linux program with a large number of idle threads (varying from 1000
to 10000). The program's main thread hits a breakpoint once all these
threads have started, which causes stop_all_threads to be called to stop
all these threads. I measured (by patching stop_all_threads):
- the execution time of stop_all_threads
- the total number of threads we iterate on during the complete
execution of the function (the total number of times we execute the
"for (thread_info *t : all_non_exited_threads ())" loop)
projects / thirdparty / binutils-gdb.git / log
