Simon Marchi [Fri, 25 Sep 2015 18:08:07 +0000 (14:08 -0400)]
Add some more casts (1/2)
Note: I needed to split this patch in two, otherwise it's too big for
the mailing list.
This patch adds explicit casts to situations where a void pointer is
assigned to a pointer to the "real" type. Building in C++ mode requires
those assignments to use an explicit cast. This includes, for example:
- callback arguments (cleanups, comparison functions, ...)
- data attached to some object (objfile, program space, etc) in the form
of a void pointer
- "user data" passed to some function
This patch comes from the commit "(mostly) auto-generated patch to insert
casts needed for C++", taken from Pedro's C++ branch.
Only files built on x86 with --enable-targets=all are modified, so the
native files for other arches will need to be dealt with separately.
I built-tested this with --enable-targets=all and reg-tested. To my
surprise, a test case (selftest.exp) had to be adjusted.
Here's the ChangeLog entry. Again, this was relatively quick to make
despite the length, thanks to David Malcom's script, although I don't
believe it's very useful information in that particular case...
Simon Marchi [Fri, 25 Sep 2015 18:08:06 +0000 (14:08 -0400)]
Add casts to memory allocation related calls
Most allocation functions (if not all) return a void* pointing to the
allocated memory. In C++, we need to add an explicit cast when
assigning the result to a pointer to another type (which is the case
more often than not).
The content of this patch is taken from Pedro's branch, from commit
"(mostly) auto-generated patch to insert casts needed for C++". I
validated that the changes make sense and manually reflowed the code to
make it respect the coding style. I also found multiple places where I
could use XNEW/XNEWVEC/XRESIZEVEC/etc.
Thanks a lot to whoever did that automated script to insert casts, doing
it completely by hand would have taken a ridiculous amount of time.
Only files built on x86 with --enable-targets=all are modified. This
means that all other -nat.c files are untouched and will have to be
dealt with later by using appropiate compilers. Or maybe we can try to
build them with a regular g++ just to know where to add casts, I don't
know.
I built-tested this with --enable-targets=all and reg-tested.
Here's the changelog entry, which was not too bad to make despite the
size, thanks to David Malcom's script. I fixed some bits by hand, but
there might be some wrong parts left (hopefully not).
Andrew Bennett [Fri, 25 Sep 2015 14:52:18 +0000 (15:52 +0100)]
[PATCH] Add micromips support to the MIPS simulator
2015-09-25 Andrew Bennett <andrew.bennett@imgtec.com>
Ali Lown <ali.lown@imgtec.com>
sim/common/
* sim-bits.h (EXTEND6): New macro.
(EXTEND12): New macro.
(EXTEND25): New macro.
sim/mips/
* Makefile.in (tmp-micromips): New rule.
(tmp-mach-multi): Add support for micromips.
* configure.ac (mips*-sde-elf* | mips*-mti-elf*): Made a multi sim
that works for both mips64 and micromips64.
(mipsisa32r2*-*-*): Made a multi sim that works for mips32 and
micromips32.
Add build support for micromips.
* dsp.igen (do_ph_s_absq, do_w_s_absq, do_qb_s_absq, do_addsc,
do_addwc, do_bitrev, do_extpv, do_extrv, do_extrv_s_h, do_insv,
do_lxx do_modsub, do_mthlip, do_mulsaq_s_w_ph, do_ph_packrl, do_qb_pick
do_ph_pick, do_qb_ph_precequ, do_qb_ph_preceu, do_w_preceq
do_w_ph_precrq, do_ph_qb_precrq, do_w_ph_rs_precrq do_qb_w_raddu,
do_rddsp, do_repl, do_shilov, do_ph_shl, do_qb_shl do_w_s_shllv,
do_ph_shrlv, do_w_r_shrav, do_wrdsp, do_qb_shrav, do_append,
do_balign, do_ph_w_mulsa, do_ph_qb_precr, do_prepend): New functions.
Refactored instruction code to use these functions.
* dsp2.igen: Refactored instruction code to use the new functions.
* interp.c (decode_coproc): Refactored to work with any instruction
encoding.
(isa_mode): New variable
(RSVD_INSTRUCTION): Changed to 0x00000039.
* m16.igen (BREAK16): Refactored instruction to use do_break16.
(JALX32): Add mips32, mips64, mips32r2 and mips64r2 models.
* micromips.dc: New file.
* micromips.igen: New file.
* micromips16.dc: New file.
* micromipsdsp.igen: New file.
* micromipsrun.c: New file.
* mips.igen (do_swc1): Changed to work with any instruction encoding.
(do_add do_addi do_andi do_dadd do_daddi do_dsll32 do_dsra32
do_dsrl32, do_dsub, do_break, do_break16, do_clo, do_clz, do_dclo
do_dclz, do_lb, do_lh, do_lwr, do_lwl, do_lwc, do_lw, do_lwu, do_lhu
do_ldc, do_lbu, do_ll, do_lld, do_lui, do_madd, do_dsp_madd, do_maddu
do_dsp_maddu, do_dsp_mfhi, do_dsp_mflo, do_movn, do_movz, do_msub
do_dsp_msub, do_msubu, do_dsp_msubu, do_mthi, do_dsp_mthi, do_mtlo
do_dsp_mtlo, do_mul, do_dsp_mult, do_dsp_multu, do_pref, do_sc, do_scd
do_sub, do_sw, do_teq, do_teqi, do_tge, do_tgei, do_tgeiu, do_tgeu, do_tlt
do_tlti, do_tltiu, do_tltu, do_tne, do_tnei, do_abs_fmt, do_add_fmt
do_alnv_ps, do_c_cond_fmt, do_ceil_fmt, do_cfc1, do_ctc1, do_cvt_d_fmt
do_cvt_l_fmt, do_cvt_ps_s, do_cvt_s_fmt, do_cvt_s_pl, do_cvt_s_pu
do_cvt_w_fmt, do_div_fmt, do_dmfc1b, do_dmtc1b, do_floor_fmt, do_luxc1_32
do_luxc1_64, do_lwc1, do_lwxc1, do_madd_fmt, do_mfc1b, do_mov_fmt, do_movtf
do_movtf_fmt, do_movn_fmt, do_movz_fmt, do_msub_fmt, do_mtc1b, do_mul_fmt
do_neg_fmt, do_nmadd_fmt, do_nmsub_fmt, do_pll_ps, do_plu_ps, do_pul_ps
do_puu_ps, do_recip_fmt, do_round_fmt, do_rsqrt_fmt, do_prefx, do_sdc1
do_suxc1_32, do_suxc1_64, do_sqrt_fmt, do_sub_fmt, do_swc1, do_swxc1
do_trunc_fmt): New functions, refactored from existing instructions.
Refactored instruction code to use these functions.
(RSVD): Changed to use new reserved instruction.
(loadstore_ea, not_word_value, unpredictable, check_mt_hilo, check_mf_hilo,
check_mult_hilo, check_div_hilo, check_u64, do_luxc1_32, do_sdc1, do_suxc1_32,
check_fmt_p, check_fpu, do_load_double, do_store_double): Added micromips32
and micromips64 models.
Added include for micromips.igen and micromipsdsp.igen
Add micromips32 and micromips64 models.
(DecodeCoproc): Updated to use new macro definition.
* mips3264r2.igen (do_dsbh, do_dshd, do_dext, do_dextm, do_dextu, do_di,
do_dins, do_dinsm, do_ei, do_ext, do_mfhc1, do_mthc1, do_ins, do_dinsu,
do_seb, do_seh do_rdhwr, do_wsbh): New functions.
Refactored instruction code to use these functions.
* sim-main.h (CP0_operation): New enum.
(DecodeCoproc): Updated macro.
(IMEM32_MICROMIPS, IMEM16_MICROMIPS, MICROMIPS_MINOR_OPCODE,
MICROMIPS_DELAYSLOT_SIZE_ANY, MICROMIPS_DELAYSLOT_SIZE_16, MICROMIPS_DELAYSLOT_SIZE_32,
ISA_MODE_MIPS32 and ISA_MODE_MICROMIPS): New defines.
(sim_state): Add isa_mode field.
sim/testsuite/sim/mips/
* basic.exp (run_micromips_test, run_sim_tests): New functions
Add support for micromips tests.
* hilo-hazard-4.s: New file.
* testutils.inc (_dowrite): Changed reserved instruction encoding.
(writemsg): Moved the la and li instructions before the data they are
assigned to, which prevents a bug where MIPS32 relocations are used instead
of micromips relocations when building for micromips.
[FT32] Implement pointer to address conversion method.
FT32 is a Harvard architecture with two address spaces -- RAM and flash.
The patch properly implements the pointer to address conversion method.
There are some other small fixes to handle address spaces.
gdb/
* ft32-tdep.c (ft32_register_type): Return gdbarch_tdep
(gdbarch)->pc_type instead of builtin_func_ptr.
(ft32_pointer_to_address): New function.
(ft32_address_class_type_flags): New function.
(ft32_address_class_type_flags_to_name): New function.
(ft32_address_class_name_to_type_flags): New function.
(ft32_gdbarch_init): Set tdep->pc_type. Call
set_gdbarch_pointer_to_address,
set_gdbarch_address_class_type_flags
set_gdbarch_address_class_name_to_type_flags,
and set_gdbarch_address_class_type_flags_to_name.
* ft32-tdep.h (struct gdbarch_tdep) <pc_type>: New field.
Alan Modra [Thu, 24 Sep 2015 03:43:57 +0000 (13:13 +0930)]
Relative DT_RPATH/DT_RUNPATH vs. sysroot capable ld.
This fixes two problems. First, the --sysroot option wasn't available
with a ld configured without --with-sysroot, a historical accident.
This led to people configuring binutils with --with-sysroot=/ in order
to enable sysroot support, which exposes a case where ld wrongly
prepends the sysroot to a relative path.
Compilers can materialize renamings of arrays (or of accesses to arrays)
in Ada into variables whose types are references to the actual array
types. Before this change, trying to use such an array renaming yielded
an error in GDB:
(gdb) print my_array(1)
cannot subscript or call a record
(gdb) print my_array_ptr(1)
cannot subscript or call something of type `(null)'
This behavior comes from bad handling for array renamings, in particular
the OP_FUNCALL expression operator handling from ada-lang.c
(ada_evaluate_subexp): in one place we turn the reference into a
pointer, but the code that follows expect the value to be an array.
This patch fixes how we handle references in call/subscript evaluation
so that we turn these references into the actual array values instead of
pointers to them.
gdb/ChangeLog:
* ada-lang.c (ada_evaluate_subexp) <OP_FUNCALL>: When the input
value is a reference, actually dereference it in order to get
the underlying value.
gdb/testsuite/ChangeLog:
* gdb.ada/array_ptr_renaming.exp: New testcase.
* gdb.ada/array_ptr_renaming/foo.adb: New file.
* gdb.ada/array_ptr_renaming/pack.ads: New file.
Nick Clifton [Wed, 23 Sep 2015 16:23:58 +0000 (17:23 +0100)]
Add support for files that contain multiple symbol index tables. Fixes PR 15835
binutils PR binutils/15835
* readelf.c (struct elf_section_list): New structure.
(symtab_shndx_hdr): Replace with symtab_shndx_list.
(get_32bit_elf_symbols): Scan for a symbol index table matching
the symbol table in use.
(get_64bit_elf_symbols): Likewise.
(process_section_headers): Handle multiple symbol index sections.
bfd * elf-bfd.h (struct elf_section_list): New structure.
(struct elf_obj_tdata): Replace symtab_shndx_hdr with
symtab_shndx_list. Delete symtab_shndx_section.
(elf_symtab_shndx): Replace macro with elf_symtab_shndx_list.
* elf.c (bfd_elf_get_syms): If symtab index sections are present,
scan them for the section that matches the provided symbol table.
(bfd_section_from_shdr): Record all SHT_SYMTAB_SHNDX sections.
(assign_section_numbers): Use the first symtab index table in the
list.
(_bfd_elf_compute_section_file_positions): Replace use of
symtab_shndx_hdr with use of symtab_shndx_list.
(find_section_in_list): New function.
(assign_file_postions_except_relocs): Use new function.
(_bfd_elf_copy_private_symbol_data): Likewise.
(swap_out_syms): Handle multiple symbol table index sections.
* elf32-m32c.c (m32c_elf_relax_section): Replace use of
symtab_shndx_hdr with use of symtab_shndx_list.
* elf32-rl78.c (rl78_elf_relax_section): Likewise.
* elf32-rx.c (rx_relax_section): Likewise.
* elf32-v850.c (v850_elf_relax_delete_bytes): Likewise.
* elflink.c (bfd_elf_final_link): Likewise.
James Bowman [Wed, 23 Sep 2015 02:43:56 +0000 (22:43 -0400)]
sim: ft32: add character input port
The FT32 simulator has character output, of course. This patch
adds character input, which lets the simulator run interactive
FT32 applications, e.g. language interpreters.
Simon Marchi [Tue, 22 Sep 2015 15:17:52 +0000 (11:17 -0400)]
stap-probe: Remove unnecessary cast
ret->args_u.text is const char *, probe_args is const char *, so no cast
is needed. Found while doing cxx-conversion stuff, since it wouldn't
build in C++.
H.J. Lu [Tue, 22 Sep 2015 13:08:55 +0000 (06:08 -0700)]
Set DF_1_PIE in gld${EMULATION_NAME}_after_parse
We can't add OPTION_PIE to gld${EMULATION_NAME}_handle_option since
it has been handled in parse_args in lexsup.c. This patch moves
setting DF_1_PIE to gld${EMULATION_NAME}_after_parse.
ld/
* emultempl/alphaelf.em (alpha_after_parse): Call
gld${EMULATION_NAME}_after_parse instead of
after_parse_default.
* emultempl/cr16elf.em (cr16elf_after_parse): Likewise.
* emultempl/crxelf.em (crxelf_after_parse); Likewise.
* emultempl/hppaelf.em (hppaelf_after_parse): Likewise.
* emultempl/mipself.em (mips_after_parse): Likewise.
* emultempl/nds32elf.em (nds32_elf_after_parse): Likewise.
* emultempl/elf32.em: Don't include ldlex.h.
(gld${EMULATION_NAME}_after_parse): New function.
(gld${EMULATION_NAME}_handle_option) [GENERATE_PIE_SCRIPT]
<OPTION_PIE>: Removed.
(ld_${EMULATION_NAME}_emulation): Replace after_parse_default
with gld${EMULATION_NAME}_after_parse.
* emultempl/ia64elf.em (gld${EMULATION_NAME}_after_parse):
Renamed to ...
(ia64elf_after_parse): This. Call
gld${EMULATION_NAME}_after_parse instead of after_parse_default.
(LDEMUL_AFTER_PARSE): Replace gld${EMULATION_NAME}_after_parse
with ia64elf_after_parse.
H.J. Lu [Tue, 22 Sep 2015 02:38:40 +0000 (19:38 -0700)]
Use bfd_get_32 and bfd_put_32 to access ch_type
The ch_type field in Elf64_External_Chdr is 4 bytes. We should use
bfd_get_32 and bfd_put_32 to access it.
* bfd.c (bfd_update_compression_header): Use bfd_put_32 on
ch_type.
(bfd_check_compression_header): Use bfd_get_32 on ch_type.
(bfd_convert_section_contents): Use bfd_get_32 and bfd_put_32
on ch_type.
Simon Marchi [Mon, 21 Sep 2015 22:47:33 +0000 (18:47 -0400)]
Add two missing consts
Two missing consts, found while doing cxx-conversion work. We end up
with a char*, even though we pass a const char* to strstr. I am pushing
this as obvious.
gdb/ChangeLog:
* cli/cli-setshow.c (cmd_show_list): Constify a variable.
* linespec.c (linespec_lexer_lex_string): Same.
H.J. Lu [Mon, 21 Sep 2015 17:19:25 +0000 (10:19 -0700)]
Change ch_type in Elf64_External_Chdr to 4 bytes
The ch_type field in Elf64_External_Chdr is 4 bytes, followed by a
4-byte padding. This change doesn't introduce any functional change
since only the lower 32 bits of the ch_type field are used.
* external.h (Elf64_External_Chdr): Change ch_type to 4 bytes
and add ch_reserved.
Pierre Langlois [Mon, 21 Sep 2015 14:01:05 +0000 (15:01 +0100)]
Add a test case for fast tracepoints' locking mechanism
When installing a fast tracepoint, we create a jump pad with a
spin-lock. This way, only one thread can collect a given tracepoint at
any time. This test case checks that this lock actually works as
expected.
This test works by creating a function which overrides the in-process
agent library's gdb_collect function. On start up, GDBserver will ask
GDB with the 'qSymbol' packet about symbols present in the inferior.
GDB will reply with the gdb_agent_gdb_collect function from the test
case instead of the one from the agent.
gdb/testsuite/ChangeLog:
* gdb.trace/ftrace-lock.c: New file.
* gdb.trace/ftrace-lock.exp: New file.
Pierre Langlois [Mon, 21 Sep 2015 14:01:05 +0000 (15:01 +0100)]
Add a gdb.trace test for instruction relocation
This test case makes sure that relocating PC relative instructions does
not change their behaviors. All PC relative AArch64 instructions are
covered. While call and jump (32 bit relative) instructions are covered
on x86.
The test case creates a static array of function pointers for each
supported architecture. Each function in this array tests a specific
instruction using inline assembly. They all need to contain a symbol in
the form of 'set_point\[0-9\]+' and finish by either calling pass or
fail. The number of 'set_pointN' needs to go from 0 to
(ARRAY_SIZE - 1).
The test will:
- look up the number of function pointers in the static array.
- set fast tracepoints on each 'set_point\[0-9\]+' symbol, one in each
functions from 0 to (ARRAY_SIZE - 1).
- run the trace experiment and make sure the pass function is called for
every function.
gdb/testsuite/ChangeLog:
* gdb.arch/insn-reloc.c: New file.
* gdb.arch/ftrace-insn-reloc.exp: New file.
We save the function's argument on the stack as well as the return
address and the frame pointer. We then set the current frame pointer to
point to the previous one.
The generated code for the expression will freely update the stack
pointer so we use the frame pointer to refer to `*value' and `*regs'.
`*value' needs to be accessed in the epilogue of the function, in order
to set it to whatever is on top of the stack. `*regs' needs to be passed
down to the `gdb_agent_get_raw_reg' function with the `reg' operation.
gdb/gdbserver/ChangeLog:
* linux-aarch64-low-.c: Include ax.h and tracepoint.h.
(enum aarch64_opcodes) <RET>, <SUBS>, <AND>, <ORR>, <ORN>,
<EOR>, <LSLV>, <LSRV>, <ASRV>, <SBFM>, <UBFM>, <CSINC>, <MUL>,
<NOP>: New.
(enum aarch64_condition_codes): New enum.
(w0): New static global.
(fp): Likewise.
(lr): Likewise.
(struct aarch64_memory_operand) <type>: New
MEMORY_OPERAND_POSTINDEX type.
(postindex_memory_operand): New helper function.
(emit_ret): New function.
(emit_load_store_pair): New function, factored out of emit_stp
with support for MEMORY_OPERAND_POSTINDEX.
(emit_stp): Rewrite using emit_load_store_pair.
(emit_ldp): New function.
(emit_load_store): Likewise.
(emit_ldr): Mention post-index instruction in comment.
(emit_ldrh): New function.
(emit_ldrb): New function.
(emit_ldrsw): Mention post-index instruction in comment.
(emit_str): Likewise.
(emit_subs): New function.
(emit_cmp): Likewise.
(emit_and): Likewise.
(emit_orr): Likewise.
(emit_orn): Likewise.
(emit_eor): Likewise.
(emit_mvn): Likewise.
(emit_lslv): Likewise.
(emit_lsrv): Likewise.
(emit_asrv): Likewise.
(emit_mul): Likewise.
(emit_sbfm): Likewise.
(emit_sbfx): Likewise.
(emit_ubfm): Likewise.
(emit_ubfx): Likewise.
(emit_csinc): Likewise.
(emit_cset): Likewise.
(emit_nop): Likewise.
(emit_ops_insns): New helper function.
(emit_pop): Likewise.
(emit_push): Likewise.
(aarch64_emit_prologue): New function.
(aarch64_emit_epilogue): Likewise.
(aarch64_emit_add): Likewise.
(aarch64_emit_sub): Likewise.
(aarch64_emit_mul): Likewise.
(aarch64_emit_lsh): Likewise.
(aarch64_emit_rsh_signed): Likewise.
(aarch64_emit_rsh_unsigned): Likewise.
(aarch64_emit_ext): Likewise.
(aarch64_emit_log_not): Likewise.
(aarch64_emit_bit_and): Likewise.
(aarch64_emit_bit_or): Likewise.
(aarch64_emit_bit_xor): Likewise.
(aarch64_emit_bit_not): Likewise.
(aarch64_emit_equal): Likewise.
(aarch64_emit_less_signed): Likewise.
(aarch64_emit_less_unsigned): Likewise.
(aarch64_emit_ref): Likewise.
(aarch64_emit_if_goto): Likewise.
(aarch64_emit_goto): Likewise.
(aarch64_write_goto_address): Likewise.
(aarch64_emit_const): Likewise.
(aarch64_emit_call): Likewise.
(aarch64_emit_reg): Likewise.
(aarch64_emit_pop): Likewise.
(aarch64_emit_stack_flush): Likewise.
(aarch64_emit_zero_ext): Likewise.
(aarch64_emit_swap): Likewise.
(aarch64_emit_stack_adjust): Likewise.
(aarch64_emit_int_call_1): Likewise.
(aarch64_emit_void_call_2): Likewise.
(aarch64_emit_eq_goto): Likewise.
(aarch64_emit_ne_goto): Likewise.
(aarch64_emit_lt_goto): Likewise.
(aarch64_emit_le_goto): Likewise.
(aarch64_emit_gt_goto): Likewise.
(aarch64_emit_ge_got): Likewise.
(aarch64_emit_ops_impl): New static global variable.
(aarch64_emit_ops): New target function, return
&aarch64_emit_ops_impl.
(struct linux_target_ops): Install it.
Pierre Langlois [Mon, 21 Sep 2015 14:01:04 +0000 (15:01 +0100)]
Add support for fast tracepoints
This patch adds support for fast tracepoints for aarch64-linux. With this
implementation, a tracepoint can only be placed in a +/- 128MB range of
the jump pad. This is due to the unconditional branch instruction
being limited to a (26 bit << 2) offset from the current PC.
Three target operations are implemented:
- target_install_fast_tracepoint_jump_pad
Building the jump pad the biggest change of this patch. We need to add
functions to emit all instructions needed to save and restore the
current state when the tracepoint is hit. As well as implementing a
lock and creating a collecting_t object identifying the current thread.
Steps performed by the jump pad:
* Save the current state on the stack.
* Push a collecting_t object on the stack. We read the special
tpidr_el0 system register to get the thread ID.
* Spin-lock on the shared memory location of all tracing threads. We
write the address of our collecting_t object there once we have the
lock.
* Call gdb_collect.
* Release the lock.
* Restore the state.
* Execute the replaced instruction which will have been relocated.
* Jump back to the program.
- target_get_thread_area
As implemented in ps_get_thread_area, target_get_thread_area uses ptrace
to fetch the NT_ARM_TLS register. At the architecture level, NT_ARM_TLS
represents the tpidr_el0 system register.
So this ptrace call (if lwpid is the current thread):
~~~
ptrace (PTRACE_GETREGSET, lwpid, NT_ARM_TLS, &iovec);
~~~
Is equivalent to the following instruction:
~~~
msr x0, tpidr_el0
~~~
This instruction is used when creating the collecting_t object that
GDBserver can read to know if a given thread is currently tracing.
So target_get_thread_area must get the same thread IDs as what the jump
pad writes into its collecting_t object.
- target_get_min_fast_tracepoint_insn_len
This just returns 4.
gdb/gdbserver/ChangeLog:
* Makefile.in (linux-aarch64-ipa.o, aarch64-ipa.o): New rules.
* configure.srv (aarch64*-*-linux*): Add linux-aarch64-ipa.o and
aarch64-ipa.o.
* linux-aarch64-ipa.c: New file.
* linux-aarch64-low.c: Include arch/aarch64-insn.h, inttypes.h
and endian.h.
(aarch64_get_thread_area): New target method.
(extract_signed_bitfield): New helper function.
(aarch64_decode_ldr_literal): New function.
(enum aarch64_opcodes): New enum.
(struct aarch64_register): New struct.
(struct aarch64_operand): New struct.
(x0): New static global.
(x1): Likewise.
(x2): Likewise.
(x3): Likewise.
(x4): Likewise.
(w2): Likewise.
(ip0): Likewise.
(sp): Likewise.
(xzr): Likewise.
(aarch64_register): New helper function.
(register_operand): Likewise.
(immediate_operand): Likewise.
(struct aarch64_memory_operand): New struct.
(offset_memory_operand): New helper function.
(preindex_memory_operand): Likewise.
(enum aarch64_system_control_registers): New enum.
(ENCODE): New macro.
(emit_insn): New helper function.
(emit_b): New function.
(emit_bcond): Likewise.
(emit_cb): Likewise.
(emit_tb): Likewise.
(emit_blr): Likewise.
(emit_stp): Likewise.
(emit_ldp_q_offset): Likewise.
(emit_stp_q_offset): Likewise.
(emit_load_store): Likewise.
(emit_ldr): Likewise.
(emit_ldrsw): Likewise.
(emit_str): Likewise.
(emit_ldaxr): Likewise.
(emit_stxr): Likewise.
(emit_stlr): Likewise.
(emit_data_processing_reg): Likewise.
(emit_data_processing): Likewise.
(emit_add): Likewise.
(emit_sub): Likewise.
(emit_mov): Likewise.
(emit_movk): Likewise.
(emit_mov_addr): Likewise.
(emit_mrs): Likewise.
(emit_msr): Likewise.
(emit_sevl): Likewise.
(emit_wfe): Likewise.
(append_insns): Likewise.
(can_encode_int32_in): New helper function.
(aarch64_relocate_instruction): New function.
(aarch64_install_fast_tracepoint_jump_pad): Likewise.
(aarch64_get_min_fast_tracepoint_insn_len): Likewise.
(struct linux_target_ops): Install aarch64_get_thread_area,
aarch64_install_fast_tracepoint_jump_pad and
aarch64_get_min_fast_tracepoint_insn_len.
Pierre Langlois [Mon, 21 Sep 2015 14:01:04 +0000 (15:01 +0100)]
Make aarch64_decode_adrp handle both ADR and ADRP instructions
We will need to decode both ADR and ADRP instructions in GDBserver.
This patch makes common code handle both cases, even if GDB only needs
to decode the ADRP instruction.
gdb/ChangeLog:
* aarch64-tdep.c (aarch64_analyze_prologue): New is_adrp
variable. Call aarch64_decode_adr instead of
aarch64_decode_adrp.
* arch/aarch64-insn.h (aarch64_decode_adrp): Delete.
(aarch64_decode_adr): New function declaration.
* arch/aarch64-insn.c (aarch64_decode_adrp): Delete.
(aarch64_decode_adr): New function, factored out from
aarch64_decode_adrp to decode both adr and adrp instructions.
Yao Qi [Mon, 21 Sep 2015 11:31:51 +0000 (12:31 +0100)]
Wrap gdb_agent_op_sizes by #ifndef IN_PROCESS_AGENT
Hi,
I see the following build warning with recent GCC built from mainline,
aarch64-none-linux-gnu-gcc -g -O2 -I. -I/home/yao/SourceCode/gnu/gdb/git/gdb/gdbserver -I/home/yao/SourceCode/gnu/gdb/git/gdb/gdbserver/../common -I/home/yao/SourceCode/gnu/gdb/git/gdb/gdbserver/../regformats -I/home/yao/SourceCode/gnu/gdb/git/gdb/gdbserver/.. -I/home/yao/SourceCode/gnu/gdb/git/gdb/gdbserver/../../include -I/home/yao/SourceCode/gnu/gdb/git/gdb/gdbserver/../gnulib/import -Ibuild-gnulib-gdbserver/import -Wall -Wpointer-arith -Wformat-nonliteral -Wno-char-subscripts -Wempty-body -Wdeclaration-after-statement -Werror -DGDBSERVER -DCONFIG_UST_GDB_INTEGRATION -fPIC -DIN_PROCESS_AGENT -fvisibility=hidden -c -o ax-ipa.o -MT ax-ipa.o -MMD -MP -MF .deps/ax-ipa.Tpo `echo " -Wall -Wpointer-arith -Wformat-nonliteral -Wno-char-subscripts -Wempty-body -Wdeclaration-after-statement " | sed "s/ -Wformat-nonliteral / -Wno-format-nonliteral /g"` /home/yao/SourceCode/gnu/gdb/git/gdb/gdbserver/ax.c
/home/yao/SourceCode/gnu/gdb/git/gdb/gdbserver/ax.c:73:28: error: 'gdb_agent_op_sizes' defined but not used [-Werror=unused-const-variable]
static const unsigned char gdb_agent_op_sizes [gdb_agent_op_last] =
^
cc1: all warnings being treated as errors
gdb_agent_op_sizes is only used in function is_goto_target, which is
defined inside #ifndef IN_PROCESS_AGENT. This warning is not arch
specific, so GCC mainline for other targets should produce this warning
too, although this warning is triggered by enabling aarch64 fast
tracepoint. The fix is to move gdb_agent_op_sizes to
gdb/gdbserver:
2015-09-21 Yao Qi <yao.qi@linaro.org>
* ax.c [!IN_PROCESS_AGENT] (gdb_agent_op_sizes): Define it.
Yao Qi [Mon, 21 Sep 2015 10:49:22 +0000 (11:49 +0100)]
[gdbserver] Remove unused max_jump_pad_size
This patch is to remove max_jump_pad_size which isn't used else where,
and it causes a recent gcc warning like this,
gdb/gdbserver/tracepoint.c:2920:18: error: 'max_jump_pad_size' defined but not used [-Werror=unused-const-variable]
static const int max_jump_pad_size = 0x100;
^
cc1: all warnings being treated as errors
This variable max_jump_pad_size wasn't used since it was added in 2010
by https://sourceware.org/ml/gdb-patches/2010-06/msg00002.html
Joel Brobecker [Tue, 1 Sep 2015 20:20:32 +0000 (22:20 +0200)]
crash printing non-local variable from nested subprogram
We have noticed that GDB would sometimes crash trying to print
from a nested function the value of a variable declared in an
enclosing scope. This appears to be target dependent, although
that correlation might only be fortuitious. We noticed the issue
on x86_64-darwin, x86-vxworks6 and x86-solaris. The investigation
was done on Darwin.
What happens is that, because the reference is non-local, we are trying
to follow the function's static link, which does...
/* If we don't know how to compute FRAME's base address, don't give up:
maybe the frame we are looking for is upper in the stace frame. */
if (framefunc != NULL
&& SYMBOL_BLOCK_OPS (framefunc)->get_frame_base != NULL
&& (SYMBOL_BLOCK_OPS (framefunc)->get_frame_base (framefunc, frame)
== upper_frame_base))
... or, in other words, calls the get_frame_base "method" of
framefunc's struct symbol_block_ops data. This resolves to
the block_op_get_frame_base function.
Printing dlbaton->per_cu gives a value that seems fairly bogus for
a memory address (0x60). Because of it, dwarf2_evaluate_loc_desc
then crashes trying to dereference it.
What's different on Darwin compared to Linux is that the function's
frame base is encoded using the following form:
... and so dwarf2_symbol_mark_computed ends up creating
a SYMBOL_LOCATION_BATON as a struct dwarf2_loclist_baton:
if (attr_form_is_section_offset (attr)
/* .debug_loc{,.dwo} may not exist at all, or the offset may be outside
the section. If so, fall through to the complaint in the
other branch. */
&& DW_UNSND (attr) < dwarf2_section_size (objfile, section))
{
struct dwarf2_loclist_baton *baton;
[...]
SYMBOL_LOCATION_BATON (sym) = baton;
However, if you look more closely at block_op_get_frame_base's
implementation, you'll notice that the function extracts the
symbol's SYMBOL_LOCATION_BATON as a dwarf2_locexpr_baton
(a DWARF _expression_ rather than a _location list_).
That's why we end up decoding the DLBATON improperly, and thus
pass a random dlbaton->per_cu when calling dwarf2_evaluate_loc_desc.
This works on x86_64-linux, because we indeed have the frame base
described using a different form:
This patch fixes the issue by doing what we do for most (if not all)
other such methods: providing one implementation each for loc-list,
and loc-expr. Both implementations are nearly identical, so perhaps
we might later want to improve this. But this patch first tries to
fix the crash first, leaving the design issue for later.
gdb/ChangeLog:
* dwarf2loc.c (locexpr_get_frame_base): Renames
block_op_get_frame_base.
(dwarf2_block_frame_base_locexpr_funcs): Replace reference to
block_op_get_frame_base by reference to locexpr_get_frame_base.
(loclist_get_frame_base): New function, near identical copy of
locexpr_get_frame_base.
(dwarf2_block_frame_base_loclist_funcs): Replace reference to
block_op_get_frame_base by reference to loclist_get_frame_base.
Tested on x86_64-darwin (AdaCore testsuite), and x86_64-linux
(official testsuite).
* findvar.c (default_read_var_value) <LOC_UNRESOLVED>: Include the
kind of minimal symbol in the error message.
* objfiles.c (objfile_flavour_name): New function.
* objfiles.h (objfile_flavour_name): Declare.
gdb/testsuite/ChangeLog:
* gdb.dwarf2/dw2-bad-unresolved.c: New file.
* gdb.dwarf2/dw2-bad-unresolved.exp: New file.
gdb/testsuite/
* gdb.mi/mi-pending.exp: Don't use directory prefix when setting
the pending breakpoint. Remove timeout override for "Run till
MI pending breakpoint on pendfunc3 on thread 2" test.
Yao Qi [Fri, 18 Sep 2015 12:59:42 +0000 (13:59 +0100)]
aarch64 multi-arch (part 3): get thread area
With the kernle fix <http://lists.infradead.org/pipermail/linux-arm-kernel/2015-July/356511.html>,
aarch64 GDB is able to read the base of thread area of 32-bit arm
program through NT_ARM_TLS.
This patch is to teach both GDB and GDBserver to read the base of
thread area correctly in the multi-arch case. A new function
aarch64_ps_get_thread_area is added, and is shared between GDB and
GDBserver.
With this patch applied, the following fails in multi-arch testing
(GDB is aarch64 but the test cases are arm) are fixed,
-FAIL: gdb.threads/tls-nodebug.exp: thread local storage
-FAIL: gdb.threads/tls-shared.exp: print thread local storage variable
-FAIL: gdb.threads/tls-so_extern.exp: print thread local storage variable
-FAIL: gdb.threads/tls-var.exp: print tls_var
-FAIL: gdb.threads/tls.exp: first thread local storage
-FAIL: gdb.threads/tls.exp: first another thread local storage
-FAIL: gdb.threads/tls.exp: p a_thread_local
-FAIL: gdb.threads/tls.exp: p file2_thread_local
-FAIL: gdb.threads/tls.exp: p a_thread_local second time
gdb:
2015-09-18 Yao Qi <yao.qi@linaro.org>
* nat/aarch64-linux.c: Include elf/common.h,
nat/gdb_ptrace.h, asm/ptrace.h and sys/uio.h.
(aarch64_ps_get_thread_area): New function.
* nat/aarch64-linux.h: Include gdb_proc_service.h.
(aarch64_ps_get_thread_area): Declare.
* aarch64-linux-nat.c (ps_get_thread_area): Call
aarch64_ps_get_thread_area.
gdb/gdbserver:
2015-09-18 Yao Qi <yao.qi@linaro.org>
* linux-aarch64-low.c: Don't include sys/uio.h.
(ps_get_thread_area): Call aarch64_ps_get_thread_area.
Markus Metzger [Mon, 7 Sep 2015 13:41:00 +0000 (15:41 +0200)]
infrun: scheduler-locking replay
Record targets behave as if scheduler-locking were on in replay mode. Add a
new scheduler-locking option "replay" to make this implicit behaviour explicit.
It behaves like "on" in replay mode and like "off" in record mode.
By making the current behaviour a scheduler-locking option, we allow the user
to change it. Since it is the current behaviour, this new option is also
the new default.
One caveat is that when resuming a thread that is at the end of its execution
history, record btrace implicitly stops replaying other threads and resumes
the entire process. This is a convenience feature to not require the user
to explicitly move all other threads to the end of their execution histories
before being able to resume the process.
We mimick this behaviour with scheduler-locking replay and move it from
record-btrace into infrun. With all-stop on top of non-stop, we can't do
this in record-btrace anymore.
Record full does not really support multi-threading and is therefore not
impacted. If it were extended to support multi-threading, it would 'benefit'
from this change. The good thing is that all record targets will behave the
same with respect to scheduler-locking.
I put the code for this into clear_proceed_status. It also sends the
about_to_proceed notification.
gdb/
* NEWS: Announce new scheduler-locking mode.
* infrun.c (schedlock_replay): New.
(scheduler_enums): Add schedlock_replay.
(scheduler_mode): Change default to schedlock_replay.
(user_visible_resume_ptid): Handle schedlock_replay.
(clear_proceed_status_thread): Stop replaying if resumed thread is
not replaying.
(schedlock_applies): Handle schedlock_replay.
(_initialize_infrun): Document new scheduler-locking mode.
* record-btrace.c (record_btrace_resume): Remove code to stop other
threads when not replaying the resumed thread.
doc/
* gdb.texinfo (All-Stop Mode): Describe new scheduler-locking mode.
Markus Metzger [Thu, 17 Sep 2015 09:14:55 +0000 (11:14 +0200)]
target: add to_record_will_replay target method
Add a new target method to_record_will_replay to query if there is a record
target that will replay at least one thread matching the argument PTID if it
were executed in the argument execution direction.
Markus Metzger [Tue, 8 Sep 2015 07:05:38 +0000 (09:05 +0200)]
btrace: allow full memory and register access for non-replaying threads
The record btrace target does not allow accessing memory and storing registers
while replaying. For multi-threaded applications, this prevents those
accesses also for threads that are at the end of their execution history as
long as at least one thread is replaying.
Change this to only check if the selected thread is replaying. This allows
threads that are at the end of their execution history to read and write
memory and to store registers.
Also change the error message to reflect this change.
gdb/
* record-btrace.c (record_btrace_xfer_partial)
(record_btrace_store_registers, record_btrace_prepare_to_store):
Call record_btrace_is_replaying with inferior_ptid instead of
minus_one_ptid.
(record_btrace_store_registers): Change error message.
Markus Metzger [Tue, 8 Sep 2015 06:26:16 +0000 (08:26 +0200)]
target, record: add PTID argument to to_record_is_replaying
The to_record_is_replaying target method is used to query record targets if
they are replaying. This is currently interpreted as "is any thread being
replayed".
Add a PTID argument and change the interpretation to "is any thread matching
PTID being replayed".
Change all users to pass minus_one_ptid to preserve the old meaning.
The record full target does not really support multi-threading and ignores
the PTID argument.
Markus Metzger [Mon, 7 Sep 2015 08:00:46 +0000 (10:00 +0200)]
infrun: switch to NO_HISTORY thread
A thread that runs out of its execution history is stopped. We already set
stop_pc and call stop_waiting. But we do not switch to the stopped thread.
In normal_stop, we call finish_thread_state_cleanup to set a thread's running
state. In all-stop mode, we call it with minus_one_ptid; in non-stop mode, we
only call it for inferior_ptid.
If in non-stop mode normal_stop is called on behalf of a thread that is not
inferior_ptid, that other thread will still be reported as running. If it is
actually stopped it can't be resumed again.
Record targets traditionally don't support non-stop and only resume
inferior_ptid. So this has not been a problem, so far.
Switch to the eventing thread for NO_HISTORY events as preparation to support
non-stop for the record btrace target.
gdb/
* infrun.c (handle_inferior_event_1): Switch to the eventing thread
in the TARKET_WAITKIND_NO_HISTORY case.
Markus Metzger [Thu, 27 Aug 2015 14:24:38 +0000 (16:24 +0200)]
btrace: async
The record btrace target runs synchronous with GDB. That is, GDB steps
resumed threads in record btrace's to_wait method. Without GDB calling
to_wait, nothing happens 'on the target'.
Check for further expected events in to_wait before reporting the current
event and mark record btrace's async event handler in async mode.
Markus Metzger [Wed, 26 Aug 2015 13:37:55 +0000 (15:37 +0200)]
btrace: temporarily set inferior_ptid in record_btrace_start_replaying
Get_current_frame uses inferior_ptid. In record_btrace_start_replaying,
we need to get the current frame of the argument thread. So far, this
has always been inferior_ptid. With non-stop, this is not guaranteed.
Temporarily set inferior_ptid to the ptid of the argument thread.
We already temporarily set the argument thread's executing flag to false.
Move both into a new function get_thread_current_frame that does the temporary
adjustments, calls get_current_frame, and restores the previous values.
Markus Metzger [Mon, 24 Aug 2015 14:28:06 +0000 (16:28 +0200)]
btrace: lock-step
Record btrace's to_wait method picks a single thread to step. When passed
minus_one_ptid, it picks the current thread. All other threads remain where
they are.
Change this to step all resumed threads together, one step at a time, until
the first thread reports an event.
We do delay reporting NO_HISTORY events until there are no other events to
report to prevent threads at the end of their execution history from starving
other threads.
We keep threads at the end of their execution history moving and replaying
until we announce their stop in to_wait. This shouldn't really be user-visible
but its a detail worth mentioning.
Since record btrace's to_resume method also picks only a single thread to
resume, there shouldn't be a difference with the current all-stop.
With non-stop or all-stop on top of non-stop, we will see differences. The
behaviour should be more natural as we're moving all threads.
gdb/
* record-btrace.c: Include vec.h.
(record_btrace_find_thread_to_move): Removed.
(btrace_step_no_resumed, btrace_step_again)
(record_btrace_stop_replaying_at_end): New.
(record_btrace_cancel_resume): Call record_btrace_stop_replaying_at_end.
(record_btrace_single_step_forward): Remove calls to
record_btrace_stop_replaying.
(record_btrace_step_thread): Do only one step for BTHR_CONT and
BTHR_RCONT. Keep threads at the end of their history moving.
(record_btrace_wait): Call record_btrace_step_thread for all threads
until one reports an event. Call record_btrace_stop_replaying_at_end
for the eventing thread.
Markus Metzger [Mon, 24 Aug 2015 13:55:02 +0000 (15:55 +0200)]
btrace: move breakpoint checking into stepping functions
Breakpoints are only checked for BTHR_CONT and BTHR_RCONT stepping requests.
A BTHR_STEP and BTHR_RSTEP request will always report stopped without reason.
Since breakpoints are reported correctly, I assume infrun is handling this.
Move the breakpoint check into the btrace single stepping functions. This
will cause us to report breakpoint hits now also for single-step requests.
One thing to notice is that
- when executing forwards, the breakpoint is checked before 'executing'
the instruction, i.e. before moving the PC to the next instruction.
- when executing backwards, the breakpoint is checked after 'executing'
the instruction, i.e. after moving the PC to the preceding instruction
in the recorded execution.
There is code in infrun (see, for example proceed and adjust_pc_after_break)
that handles this and also depends on this behaviour.
gdb/
* record-btrace.c (record_btrace_step_thread): Move breakpoint check
to ...
(record_btrace_single_step_forward): ... here and
(record_btrace_single_step_backward): ... here.
Markus Metzger [Mon, 24 Aug 2015 12:49:43 +0000 (14:49 +0200)]
btrace: split record_btrace_step_thread
The code for BTHR_STEP and BTHR_CONT is fairly similar. Extract the common
parts into a new function record_btrace_single_step_forward. The function
returns TARGET_WAITKIND_SPURIOUS to indicate that the single-step completed
without triggering a trap.
Markus Metzger [Tue, 8 Sep 2015 14:13:47 +0000 (16:13 +0200)]
btrace: fix non-stop check in to_wait
The record btrace target stops other threads in non-stop mode after stepping
the to-be-resumed thread.
The check is done on the non_stop variable. It should rather be done on
target_is_non_stop_p (). With all-stop on top of non-stop, infrun will
take care of stopping other threads.
gdb/
* record-btrace.c (record_btrace_wait): Replace non_stop check with
target_is_non_stop_p ().
Alan Modra [Fri, 18 Sep 2015 06:47:49 +0000 (16:17 +0930)]
Add PowerPC64 ld --tls-get-addr-optimize.
Sometimes it may be of benefit to force use of the __tls_get_addr_opt
call stub even when the glibc being used during linking does not
advertise __tls_get_addr_opt.
bfd/
* elf64-ppc.h (struct ppc64_elf_params <tls_get_addr_opt>): Rename
from no_tls_get_addr_opt.
* elf64-ppc.c: Update for rename and inversion of tls_get_addr_opt.
(ppc64_elf_tls_setup): Set tls_get_addr_opt to 0 only when at
default of -1.
ld/
* emultempl/ppc64elf.em (params): Init tls_get_addr_opt field to -1.
(OPTION_TLS_GET_ADDR_OPT): Define.
(PARSE_AND_LIST_LONGOPTS): Handle --tls-get-addr-opt.
(PARSE_AND_LIST_OPTIONS, PARSE_AND_LIST_ARGS_CASES): Likewise.
* ld.texinfo: Document --tls-get-addr-optimize and
--no-tls-get-addr-optimize.
Alan Modra [Thu, 17 Sep 2015 23:44:25 +0000 (09:14 +0930)]
Delay converting linker script defined symbols from absolute
Giving linker script symbols defined outside of output sections a
section-relative value early, leads to them being used in expressions
as if they were defined inside an output section. This can mean loss
of the section VMA, and wrong results.
ld/
PR ld/18963
* ldexp.h (struct ldexp_control): Add rel_from_abs.
(ldexp_finalize_syms): Declare.
* ldexp.c (new_rel_from_abs): Keep absolute for expressions
outside of output section statements. Set rel_from_abs.
(make_abs, exp_fold_tree, exp_fold_tree_no_dot): Clear rel_from_abs.
(struct definedness_hash_entry): Add final_sec, and comment.
(update_definedness): Set final_sec.
(set_sym_sections, ldexp_finalize_syms): New functions.
* ldlang.c (lang_process): Call ldexp_finalize_syms.
ld/testsuite
PR ld/18963
* ld-scripts/pr18963.d,
* ld-scripts/pr18963.t: New test.
* ld-scripts/expr.exp: Run it.
* ld-elf/provide-hidden-2.ld: Explicitly make "dot" absolute.
* ld-mips-elf/gp-hidden.sd: Don't care about _gp section.
* ld-mips-elf/no-shared-1-n32.d: Don't care about symbol shown at
start of .data section.
* ld-mips-elf/no-shared-1-n64.d: Likewise.
* ld-mips-elf/no-shared-1-o32.d: Likewise.
Pierre Langlois [Thu, 17 Sep 2015 10:39:10 +0000 (11:39 +0100)]
Add test case for tracepoints with conditions
This patch adds a test case for tracepoints with a condition expression.
Each case will test a condition against the number of frames that should
have been traced. Some of these tests fail on x86_64 and others on
i386, which have been marked as known failures for now, see PR/18955.
gdb/testsuite/ChangeLog:
2015-09-17 Pierre Langlois <pierre.langlois@arm.com>
Yao Qi <yao.qi@linaro.org>
* gdb.trace/trace-condition.c: New file.
* gdb.trace/trace-condition.exp: New file.
Wei-cheng Wang [Wed, 16 Sep 2015 15:20:51 +0000 (16:20 +0100)]
Fix argument to compiled_cond, and add cases for compiled-condition.
This patch fixes the argument passed to compiled_cond. It should be
regs buffer instead of tracepoint_hit_ctx. Test case is added as
well for testing compiled-cond.
gdb/gdbserver/ChangeLog
2015-09-16 Wei-cheng Wang <cole945@gmail.com>
* tracepoint.c (eval_result_type): Change prototype.
(condition_true_at_tracepoint): Fix argument to compiled_cond.
gdb/testsuite/ChangeLog
2015-09-16 Wei-cheng Wang <cole945@gmail.com>
* gdb.trace/ftrace.exp: (test_ftrace_condition) New function
for testing bytecode compilation.
Pedro Alves [Wed, 16 Sep 2015 14:51:36 +0000 (15:51 +0100)]
non-stop-fair-events.exp slower on software single-step && !displ-step targets
On software single-step targets that don't support displaced stepping,
threads keep hitting each other's single-step breakpoints, and then
GDB needs to pause all threads to step past those. The end result is
that progress in the main thread will be slower and it may take a bit
longer for the signal to be queued. This patch bumps the timeout on
such targets.
gdb/testsuite/ChangeLog:
2015-09-16 Pedro Alves <palves@redhat.com>
Sandra Loosemore <sandra@codesourcery.com>
* gdb.threads/non-stop-fair-events.c (timeout): New global.
(SECONDS): Redefine.
(main): Call pthread_kill and alarm early.
* gdb.threads/non-stop-fair-events.exp: Probe displaced stepping
support.
(test): If the target can't hardware step and doesn't support
displaced stepping, increase the timeout.
Pedro Alves [Wed, 16 Sep 2015 14:51:36 +0000 (15:51 +0100)]
Make it easier to debug non-stop-fair-events.exp
If we enable infrun debug running this test, it quickly fails with a
full expect buffer. That can be simply handled with a couple
exp_continues. As it's annoying to hack this every time we need to
debug the test, this patch adds bits to enable debugging support
easily, with a one-line change.
And then, if any iteration of the test fails, we end up with a long
cascade of time outs. Just bail out when we see the first fail.
gdb/testsuite/
2015-09-16 Pedro Alves <palves@redhat.com>
* gdb.threads/non-stop-fair-events.exp (gdb_test_no_anchor)
(enable_debug): New procedures.
(test): Use them. Bail out if waiting for threads fails.
(top level): Bail out if a test fails.
Alan Modra [Wed, 16 Sep 2015 08:37:03 +0000 (18:07 +0930)]
Fix slowdown in ld -r for most common case of out-of-order relocs
I chose insertion sort since relocs are mostly sorted, but there is a
common case we can handle better; A run of relocs put out of order
due to not linking input files in order.
PR 18867
* elflink.c (elf_link_adjust_relocs): Modify insertion sort to
insert a run. Return status in case of malloc failure.
Adjust callers.
[Ada] Enhance type printing for arrays with variable-sized elements
This change is relevant only for standard DWARF (as opposed to the GNAT
encodings extensions): at the time of writing it only makes a difference
with GCC patches that are to be integrated: see the patch series
submission at
<https://gcc.gnu.org/ml/gcc-patches/2015-07/msg01353.html>.
Given the following Ada declarations:
subtype Small_Int is Natural range 0 .. 100;
type R_Type (L : Small_Int := 0) is record
S : String (1 .. L);
end record;
type A_Type is array (Natural range <>) of R_Type;
A : A_Type := (1 => (L => 0, S => ""),
2 => (L => 2, S => "ab"));
Before this change, we would get the following GDB session:
(gdb) ptype a
type = array (1 .. 2) of foo.r_type <packed: 838-bit elements>
This is wrong: "a" is not a packed array. This output comes from the
fact that, because R_Type has a dynamic size (with a maximum), the
compiler has to describe in the debugging information the size allocated
for each array element (i.e. the stride, in DWARF parlance: see
DW_AT_byte_stride). Ada type printing currently assumes that arrays
with a stride are packed, hence the above output.
In practice, GNAT never performs bit-packing for arrays that contain
variable-sized elements. Leveraging this fact, this patch enhances type
printing so that ptype does not pretend that arrays are packed when they
have a stride and they contain dynamic elements. After this change, we
get the following expected output:
(gdb) ptype a
type = array (1 .. 2) of foo.r_type
gdb/ChangeLog:
* ada-typeprint.c (print_array_type): Do not describe arrays as
packed when they embed dynamic elements.
gdb/testsuite/ChangeLog:
* gdb.ada/array_of_variable_length.exp: New testcase.
* gdb.ada/array_of_variable_length/foo.adb: New file.
* gdb.ada/array_of_variable_length/pck.adb: New file.
* gdb.ada/array_of_variable_length/pck.ads: New file.
Max Filippov [Tue, 15 Sep 2015 03:37:14 +0000 (06:37 +0300)]
xtensa: generate PLT entries for call0 ABI
2015-09-15 Max Filippov <jcmvbkbc@gmail.com>
bfd/
* elf32-xtensa.c (elf_xtensa_be_plt_entry)
(elf_xtensa_le_plt_entry): Emit 'entry' instruction only for
windowed ABI.
(elf_xtensa_create_plt_entry): Generate 'l32r' offsets and fix
up instructions according to ABI.
This is already how the parent is reported in the vfork/fork events,
and is actually what the fix made gdbserver do. Following the
documentation change, the event would have been reported like this
instead:
projects / thirdparty / binutils-gdb.git / log
