Corinna Vinschen [Mon, 26 Aug 2013 16:14:40 +0000 (16:14 +0000)]
* mtinfo.h (class mtinfo_part): Change type of block numbers to int64_t.
(mtinfo_part::initialize): Ditto for nblock parameter in declaration.
(class mtinfo_drive): Change type of block number to int64_t. Change
all parameters indicating a block number to int64_t in method
declarations.
* fhandler_tape.cc (mtinfo_part::initialize): Ditto in definition.
(mtinfo_drive::get_pos): Ditto. Replace low and high with a
ULARGE_INTEGER and use it's components in call to GetTapePosition.
Store full value in block.
(mtinfo_drive::_set_pos): Change type of count parameter to int64_t.
Change call to SetTapePosition accordingly.
(mtinfo_drive::set_pos): Change type of count parameter to int64_t.
Change local variables holding block numbers accordingly.
(mtinfo_drive::get_status): Don't bail out early if fetching media
parameters fails.
(mtinfo_drive::ioctl): Add explicit cast matching receiving type in
MTTELL and MTIOCPOS calls.
Corinna Vinschen [Wed, 21 Aug 2013 20:28:12 +0000 (20:28 +0000)]
* fhandler_tape.cc (mtinfo_drive::create_partitions): Allow partitioning
of drives supporting select partitions.
(mtinfo_drive::get_status): Fill in current partition and number of
partitions on tape into mt_resid.
* include/cygwin/mtio.h (struct mtget): Align mt_resid comment to
aforementioned change.
* include/cygwin/version.h (CYGWIN_VERSION_API_MINOR): Bump.
Corinna Vinschen [Tue, 20 Aug 2013 20:17:14 +0000 (20:17 +0000)]
* fhandler_tape.cc (mtinfo_drive::create_partitions): Fix long-standing
bug disabling creation of two partitions on drives supporting initiator
partitions.
(mtinfo_drive::set_blocksize): Update media information after setting
blocksize succeeded.
(mtinfo_drive::get_status): Fetch fresh media information.
opcodes/
* micromips-opc.c (micromips_opcodes): Replace "dext" and "dins"
macro entries with "dextm", "dextu", "dinsm" and "dinsu" aliases.
Use +H rather than +C for the real "dext".
* mips-opc.c (mips_builtin_opcodes): Likewise.
gas/
* config/tc-mips.c (report_bad_range, report_bad_field): Delete.
(macro): Remove M_DEXT and M_DINS handling.
gas/testsuite/
* gas/mips/ext-ill.l, gas/mips/mips64r2-ill.l: Expect DEXT and DINS
error messages to have the same form as the EXT and INS ones.
* gas/mips/micromips-insn32.d, gas/mips/micromips-noinsn32.d,
gas/mips/micromips-trap.d, gas/mips/micromips.d,
gas/mips/micromips@mips64r2.d, gas/mips/mips64r2.d: Expect
"dext" and "dins" instead of "dextm", "dextu", "dinsm" and "dinsu".
Corinna Vinschen [Mon, 19 Aug 2013 16:21:29 +0000 (16:21 +0000)]
* lc_msg.h: Regenerate.
* nlsfuncs.cc (__get_lcid_from_locale): Update list of Script-only
locales to Windows 8.
(__set_charset_from_locale): Take locales added with Windows 8 and 8.1
into account.
Joel Brobecker [Sat, 17 Aug 2013 01:07:52 +0000 (01:07 +0000)]
src-release: Strip "-cvs" from GDB source dir and tarball.
The nightly snapshots we have been creating in the past did not
include the "-cvs" suffix at the end of the version number. Snapshot
packaging started breaking ever since GDB switched to using BFD's
version number. Things got partially fixed with the previous change
to this file, but the change missed the fact that the "-cvs" suffix
in the tarball name (Eg: gdb-7.6.50-20130816-cvs.tar) is undesirable.
This patch removes it.
ChangeLog:
* src-release (VER): When using $(TOOL)/common/create-version.sh,
strip the "-cvs" suffix from the version number if present.
include/opcode/
2013-08-04 Jürgen Urban <JuergenUrban@gmx.de>
Richard Sandiford <rdsandiford@googlemail.com>
* mips.h: Document new VU0 operand characters.
(OP_VU0_SUFFIX, OP_VU0_MATCH_SUFFIX): New mips_operand_types.
(OP_REG_VF, OP_REG_VI, OP_REG_R5900_I, OP_REG_R5900_Q, OP_REG_R5900_R)
(OP_REG_R5900_ACC): New mips_reg_operand_types.
(INSN2_VU0_CHANNEL_SUFFIX): New macro.
(mips_vu0_channel_mask): Declare.
opcodes/
2013-08-04 Jürgen Urban <JuergenUrban@gmx.de>
Richard Sandiford <rdsandiford@googlemail.com>
* mips-dis.c (print_reg): Handle OP_REG_VI, OP_REG_VF, OP_REG_R5900_I,
OP_REG_R5900_Q, OP_REG_R5900_R and OP_REG_R5900_ACC.
(print_vu0_channel): New function.
(print_insn_arg): Handle OP_VU0_SUFFIX and OP_VU0_MATCH_SUFFIX.
(print_insn_args): Handle '#'.
(print_insn_mips): Handle INSN2_VU0_CHANNEL_SUFFIX.
* mips-opc.c (mips_vu0_channel_mask): New constant.
(decode_mips_operand): Handle new VU0 operand types.
(VU0, VU0CH): New macros.
(mips_builtin_opcodes): Add VU0 opcodes. Use "+7" rather than "E"
for LQC2 and SQC2. Use "+9" rather than "G" for EE CFC2 and CTC2.
Use "+6" rather than "G" for QMFC2 and QMTC2.
gas/
2013-08-04 Jürgen Urban <JuergenUrban@gmx.de>
Richard Sandiford <rdsandiford@googlemail.com>
* config/tc-mips.c (MAX_OPERANDS): Bump to 6.
(RWARN): Bump to 0x8000000.
(RTYPE_VI, RTYPE_VF, RTYPE_R5900_I, RTYPE_R5900_Q, RTYPE_R5900_R)
(RTYPE_R5900_ACC): New register types.
(RTYPE_MASK): Include them.
(R5900_I_NAMES, R5900_Q_NAMES, R5900_R_NAMES, R5900_ACC_NAMES): New
macros.
(reg_names): Include them.
(mips_parse_register_1): New function, split out from...
(mips_parse_register): ...here. Add a channels_ptr parameter.
Look for VU0 channel suffixes when nonnull.
(reg_lookup): Update the call to mips_parse_register.
(mips_parse_vu0_channels): New function.
(OT_CHANNELS, OT_DOUBLE_CHAR): New mips_operand_token_types.
(mips_operand_token): Add a "channels" field to the union.
Extend the comment above "ch" to OT_DOUBLE_CHAR.
(mips_parse_base_start): Match -- and ++. Handle channel suffixes.
(mips_parse_argument_token): Handle channel suffixes here too.
(validate_mips_insn): Handle INSN2_VU0_CHANNEL_SUFFIX.
Ignore OP_VU0_MATCH_SUFFIX when calculating the used bits.
Handle '#' formats.
(md_begin): Register $vfN and $vfI registers.
(operand_reg_mask): Handle OP_VU0_SUFFIX and OP_VU0_MATCH_SUFFIX.
(convert_reg_type): Handle OP_REG_VI, OP_REG_VF, OP_REG_R5900_I,
OP_REG_R5900_Q, OP_REG_R5900_R and OP_REG_R5900_ACC.
(match_vu0_suffix_operand): New function.
(match_operand): Handle OP_VU0_SUFFIX and OP_VU0_MATCH_SUFFIX.
(macro): Use "+7" rather than "E" for LDQ2 and STQ2.
(mips_lookup_insn): New function.
(mips_ip): Use it. Allow "+K" operands to be elided at the end
of an instruction. Handle '#' sequences.
gas/testsuite/
2013-08-04 Jürgen Urban <JuergenUrban@gmx.de>
* gas/mips/r5900-vu0.d: Expect $vfN and $viN instead of numeric
coprocessor registers.
* gas/mips/r5900-all-vu0.s, gas/mips/r5900-all-vu0.d,
gas/mips/r5900-full-vu0.s, gas/mips/r5900-full-vu0.d,
gas/mips/r5900-error-vu0.s, gas/mips/r5900-error-vu0.l: New tests.
* gas/mips/mips.exp: Run them.
include/opcode/
* mips.h (mips_pcrel_operand): Inherit from mips_int_operand.
(mips_int_operand_min, mips_int_operand_max): New functions.
(mips_decode_pcrel_operand): Use mips_decode_int_operand.
opcodes/
* mips-formats.h (PCREL): Reorder parameters and update the definition
to match new mips_pcrel_operand layout.
(JUMP, JALX, BRANCH): Update accordingly.
* mips16-opc.c (decode_mips16_operand): Likewise.
gas/
* config/tc-mips.c (match_int_operand): Use mips_int_operand_min
and mips_int_operand_max.
(mips16_immed_operand, mips16_immed_operands, MIPS16_NUM_IMMED):
Delete.
(mips16_immed_operand, mips16_immed_in_range_p): New functions.
(mips16_immed, mips16_extended_frag): Use them. Use mips_int_operand
instead of mips16_immed_operand.
opcodes/
* mips-opc.c (WR_1, WR_2, RD_1, RD_2, RD_3, RD_4, MOD_1, MOD_2, UDI):
New macros.
(WR_d, WR_t, WR_D, WR_T, WR_S, RD_s, RD_b, RD_t, RD_S, RD_T, RD_R)
(WR_z, WR_Z, RD_z, RD_Z, RD_d): Delete.
(mips_builtin_opcodes): Use the new position-based read-write flags
instead of field-based ones. Use UDI for "udi..." instructions.
* mips16-opc.c (WR_1, WR_2, RD_1, RD_2, RD_3, RD_4, MOD_1, MOD_2):
New macros.
(WR_x, WR_y, WR_z, WR_Y, RD_x, RD_y, RD_Z, RD_X): Delete.
(RD_T, WR_T, WR_31): Redefine using generic INSN_* flags.
(WR_SP, RD_16): New macros.
(RD_SP): Redefine as an INSN2_* flag.
(MOD_SP): Redefine in terms of RD_SP and WR_SP.
(mips16_opcodes): Use the new position-based read-write flags
instead of field-based ones. Use RD_16 for "nop". Move RD_SP to
pinfo2 field.
* micromips-opc.c (WR_1, WR_2, RD_1, RD_2, RD_3, RD_4, MOD_1, MOD_2):
New macros.
(WR_mb, RD_mc, RD_md, WR_md, RD_me, RD_mf, WR_mf, RD_mg, WR_mh, RD_mj)
(WR_mj, RD_ml, RD_mmn, RD_mp, WR_mp, RD_mq, RD_gp, WR_d, WR_t, WR_D)
(WR_T, WR_S, RD_s, RD_b, RD_t, RD_T, RD_S, RD_R, RD_D): Delete.
(RD_sp, WR_sp): Redefine to INSN2_READ_SP and INSN2_WRITE_SP.
(micromips_opcodes): Use the new position-based read-write flags
instead of field-based ones.
* mips-dis.c (print_insn_arg): Use mips_decode_reg_operand.
(print_insn_mips, print_insn_micromips): Use INSN_WRITE_1 instead
of field-based flags.
gas/
* config/tc-mips.c (MAX_OPERANDS): New macro.
(mips_operand_array): New structure.
(mips_operands, mips16_operands, micromips_operands): New arrays.
(micromips_to_32_reg_b_map, micromips_to_32_reg_c_map)
(micromips_to_32_reg_e_map, micromips_to_32_reg_f_map)
(micromips_to_32_reg_g_map, micromips_to_32_reg_l_map)
(micromips_to_32_reg_q_map): Delete.
(insn_operands, insn_opno, insn_extract_operand): New functions.
(validate_mips_insn): Take a mips_operand_array as argument and
use it to build up a list of operands. Extend to handle INSN_MACRO
and MIPS16.
(validate_mips16_insn): New function.
(validate_micromips_insn): Take a mips_operand_array as argument.
Handle INSN_MACRO.
(md_begin): Initialize mips_operands, mips16_operands and
micromips_operands. Call validate_mips_insn and
validate_micromips_insn for macro instructions too.
Call validate_mips16_insn for MIPS16 instructions.
(insn_read_mask, insn_write_mask, operand_reg_mask, insn_reg_mask):
New functions.
(gpr_read_mask, gpr_write_mask, fpr_read_mask, fpr_write_mask): Use
them. Handle INSN_UDI.
(get_append_method): Use gpr_read_mask.
gas/
* config/tc-mips.c (compact_branch_p, uncond_branch_p): Use the same
flags for MIPS16 and non-MIPS16 instructions.
(gpr_mod_mask): Move the INSN2_MOD_SP case outside the micromips block.
(gpr_read_mask): Use INSN2_READ_GPR_31 for MIPS16 instructions too.
(gpr_write_mask): Remove MIPS16_INSN_WRITE_SP handling.
(can_swap_branch_p, get_append_method): Use the same flags for MIPS16
and non-MIPS16 instructions. Fix formatting.
* cygheap.cc (cmalloc): Use size_t for size field.
(cmalloc_abort): Ditto.
(crealloc): Ditto.
(crealloc_abort): Ditto.
(ccalloc): Ditto.
(ccalloc_abort): Ditto.
* cygheap_malloc.h (HEAP_USER): Add.
(cmalloc): Use size_t for size field in declaration.
(cmalloc_abort): Ditto.
(crealloc): Ditto.
(crealloc_abort): Ditto.
(ccalloc): Ditto.
(ccalloc_abort): Ditto.
* path.cc (normalize_posix_path): Don't check existence of / or // dir in
parent dir check.
* Makefile.in (cyglsa.dll): Explicitely specify DLL entry point to
harden against gcc function and block reordering optimizations.
(cyglsa64.dll): Ditto.
* gcc.xml (gcc-default: Rename from gcc-cons. Change title.
(gcc-64): New section explaininig differences in programming for
64 bit Cygwin.
(gcc-gui): Simplify description and aim at UNIX/Linux developers only.
Note that X programming is preferred. Drop example.
* ntdll.h (struct _SEMAPHORE_BASIC_INFORMATION): Define.
(enum _SEMAPHORE_INFORMATION_CLASS): Define.
(NtQuerySemaphore): Declare.
* thread.h (class semaphore): Add member startvalue.
(semaphore::fixup_before_fork): New inline method.
(semaphore::_fixup_before_fork): Declare.
* thread.cc (MTinterface::fixup_before_fork): Additionally call
semaphore::fixup_before_fork.
(semaphore::semaphore): Set currentvalue to -1. Set startvalue to
incoming initializer value.
(semaphore::_getvalue): Just query semaphore using NtQuerySemaphore
rather then using WFSO/Release.
(semaphore::_post): Drop setting currentvalue. It's not thread-safe.
(semaphore::_trywait): Ditto.
(semaphore::_timedwait): Ditto.
(semaphore::_wait): Ditto.
(semaphore::_fixup_before_fork): New method, setting currentvalue from
actual windows semaphore right before fork.
(semaphore::_fixup_after_fork): Drop kludge from 2013-07-10. Drop
FIXME comment.
* libc/sys/linux/include/netdb.h, libc/sys/linux/net/getaddrinfo.c,
libc/sys/linux/net/getnameinfo.c: Add restrict keyword to getnameinfo()
and getaddrinfo() to increase standards compliance and match glibc.
Joel Brobecker [Mon, 22 Jul 2013 15:17:20 +0000 (15:17 +0000)]
src-release: Fix VER computation for TOOL=gdb
Without this patch, the DATE marker in gdb/version.in does not get
replaced by the source packaging date, causing the name of the
tarball being created to: have the DATE marker in the tarball
name and the name of the directory the sources unpack to (Eg:
gdb-7.6.50-DATE-cvs.tar instead of gdb-7.6.50-20130710-cvs.tar).
ChangeLog:
* src-release (VER): Use $(TOOL)/common/create-version.sh
if it exists.
* exceptions.cc (signal_exit): Only dump core when it's a "kernel" signal.
Only use RtlCaptureContext on x86_64. It doesn't seem to do what's expected on
x86.
* spawn.cc (child_info_spawn::worker): Reinstate using temp buffer for wide
character command-line storage. Use wcs method to convert command line.
* winf.h (lb_wcs): Delete.
(linebuf::wcs): Implement new single-argument method.
* common.din: Export GetCommandLine{A,W}.
* kernel32.cc: Add includes needed for GetCommandLine functions.
(ucmd): New function.
(cygwin_GetCommandLineW): Ditto.
(cygwin_GetCommandLineA): Ditto.
* spawn.cc (child_info_spawn::worker): Rename one_line -> cmd. Use lb_wcs
macro to generate a wide character version of the line buffer. Remove
duplicate printing of command line. Don't access members of linebuf directly.
* winf.h: Use pragma once.
(linebuf): Make storage private.
(linebuf::operator size_t): New operator. Return size of buf.
(linebuf::operator wchar_t): New operator.
(linebuf::wcs): New function.
(lb_wcs): New macro.
* include/cygwin/version.h: Bump API minor number to 268.
* strfuncs.cc: Clarify descriptive file comment.
* path.cc (normalize_posix_path): Start checking path before ".." at
dst, rather than at dst_start, otherwise suffer loss of one leading
slash in case of UNC paths.
* new-features.sgml (ov-new1.7.22): New section. Document dropped
support for /dev/mem, /dev/kmem, /dev/port.
* specialnames.xml (pathnames-posixdevices): Drop description of
/dev/mem, /dev/kmem, /dev/port.
include/opcode/
* mips.h (mips_operand_type, mips_reg_operand_type): New enums.
(mips_operand, mips_int_operand, mips_mapped_int_operand)
(mips_msb_operand, mips_reg_operand, mips_reg_pair_operand)
(mips_pcrel_operand): New structures.
(mips_insert_operand, mips_extract_operand, mips_signed_operand)
(mips_decode_int_operand, mips_decode_pcrel_operand): New functions.
(decode_mips_operand, decode_micromips_operand): Declare.
opcodes/
* mips-formats.h: New file.
* mips-opc.c: Include mips-formats.h.
(reg_0_map): New static array.
(decode_mips_operand): New function.
* micromips-opc.c: Remove <stdio.h> include. Include mips-formats.h.
(reg_0_map, reg_28_map, reg_29_map, reg_31_map, reg_m16_map)
(reg_mn_map, reg_q_map, reg_h_map1, reg_h_map2, int_b_map)
(int_c_map): New static arrays.
(decode_micromips_operand): New function.
* mips-dis.c (micromips_to_32_reg_b_map, micromips_to_32_reg_c_map)
(micromips_to_32_reg_d_map, micromips_to_32_reg_e_map)
(micromips_to_32_reg_f_map, micromips_to_32_reg_g_map)
(micromips_to_32_reg_h_map1, micromips_to_32_reg_h_map2)
(micromips_to_32_reg_l_map, micromips_to_32_reg_m_map)
(micromips_to_32_reg_n_map, micromips_to_32_reg_q_map)
(micromips_imm_b_map, micromips_imm_c_map): Delete.
(print_reg): New function.
(mips_print_arg_state): New structure.
(init_print_arg_state, print_insn_arg): New functions.
(print_insn_args): Change interface and use mips_operand structures.
Delete GET_OP_S. Move GET_OP definition to...
(print_insn_mips): ...here. Update the call to print_insn_args.
(print_insn_micromips): Use print_insn_args.
gas/
* config/tc-mips.c (validate_mips_insn): Move further up file.
Add insn_bits and decode_operand arguments. Use the mips_operand
fields to work out which bits an operand occupies. Detect double
definitions.
(validate_micromips_insn): Move further up file. Call into
validate_mips_insn.
gas/
* config/tc-mips.c (mips_flag_nan2008): New variable.
(options): Add OPTION_NAN enum value.
(md_longopts): Handle it.
(md_parse_option): Likewise.
(s_nan): New function.
(mips_elf_final_processing): Handle EF_MIPS_NAN2008.
(md_show_usage): Add -mnan.
* doc/as.texinfo (Overview): Add -mnan.
* doc/c-mips.texi (MIPS Opts): Document -mnan.
(MIPS NaN Encodings): New node. Document .nan directive.
(MIPS-Dependent): List the new node.
gas/testsuite/
* gas/mips/nan-2008-1.d: New test.
* gas/mips/nan-2008-2.d: New test.
* gas/mips/nan-2008-3.d: New test.
* gas/mips/nan-2008-4.d: New test.
* gas/mips/nan-legacy-1.d: New test.
* gas/mips/nan-legacy-2.d: New test.
* gas/mips/nan-legacy-3.d: New test.
* gas/mips/nan-legacy-4.d: New test.
* gas/mips/nan-legacy-5.d: New test.
* gas/mips/nan-error-1.l: New list test.
* gas/mips/nan-error-2.l: New list test.
* gas/mips/nan-2008-override.s: New test source.
* gas/mips/nan-2008.s: New test source.
* gas/mips/nan-legacy-override.s: New test source.
* gas/mips/nan-legacy.s: New test source.
* gas/mips/nan-error-1.s: New test source.
* gas/mips/nan-error-2.s: New test source.
* gas/mips/mips.exp: Run the new tests.
ld/testsuite/
* ld-mips-elf/nan-2008.d: New test.
* ld-mips-elf/nan-legacy.d: New test.
* ld-mips-elf/nan-mixed-1.d: New test.
* ld-mips-elf/nan-mixed-2.d: New test.
* ld-mips-elf/nan-2008.s: New test source.
* ld-mips-elf/nan-legacy.s: New test source.