re PR debug/83917 (with -mcall-ms2sysv-xlogues, stepping into x86 tail-call restore...
[gcc.git] / libgcc / ChangeLog
index b7956211440229e63ac910abe924107a08bf1387..e93967799eade8ab0d7a698275452f2365bfdf29 100644 (file)
@@ -1,4 +1,169 @@
-2017-12-12  Kito Cheng <kito.cheng@gmail.com>
+2018-02-28  Jakub Jelinek  <jakub@redhat.com>
+
+       PR debug/83917
+       * configure.ac (AS_HIDDEN_DIRECTIVE): AC_DEFINE_UNQUOTED this to
+       $asm_hidden_op if visibility ("hidden") attribute works.
+       (HAVE_AS_CFI_SECTIONS): New AC_DEFINE.
+       * config/i386/i386-asm.h: Don't include auto-host.h.
+       (PACKAGE_VERSION, PACKAGE_NAME, PACKAGE_STRING, PACKAGE_TARNAME,
+       PACKAGE_URL): Don't undefine.
+       (USE_GAS_CFI_DIRECTIVES): Don't use nor define this macro, instead
+       guard cfi_startproc only on ifdef __GCC_HAVE_DWARF2_CFI_ASM.
+       (FN_HIDDEN): Change guard from #ifdef HAVE_GAS_HIDDEN to
+       #ifdef AS_HIDDEN_DIRECTIVE, use AS_HIDDEN_DIRECTIVE macro in the
+       definition instead of hardcoded .hidden.
+       * config/i386/cygwin.S: Include i386-asm.h first before .cfi_sections
+       directive.  Use #ifdef HAVE_AS_CFI_SECTIONS rather than
+       #ifdef HAVE_GAS_CFI_SECTIONS_DIRECTIVE to guard .cfi_sections.
+       (USE_GAS_CFI_DIRECTIVES): Don't define.
+       * configure: Regenerated.
+       * config.in: Likewise.
+
+2018-02-26  Jakub Jelinek  <jakub@redhat.com>
+
+       PR debug/83917
+       * config/i386/i386-asm.h (PACKAGE_VERSION, PACKAGE_NAME,
+       PACKAGE_STRING, PACKAGE_TARNAME, PACKAGE_URL): Undefine between
+       inclusion of auto-target.h and auto-host.h.
+       (USE_GAS_CFI_DIRECTIVES): Define if not defined already based on
+       __GCC_HAVE_DWARF2_CFI_ASM.
+       (cfi_startproc, cfi_endproc, cfi_adjust_cfa_offset,
+       cfi_def_cfa_register, cfi_def_cfa, cfi_register, cfi_offset, cfi_push,
+       cfi_pop): Define.
+       * config/i386/cygwin.S: Don't include auto-host.h here, just
+       define USE_GAS_CFI_DIRECTIVES to 1 or 0 and include i386-asm.h.
+       (cfi_startproc, cfi_endproc, cfi_adjust_cfa_offset,
+       cfi_def_cfa_register, cfi_register, cfi_push, cfi_pop): Remove.
+       * config/i386/resms64fx.h: Add cfi_* directives.
+       * config/i386/resms64x.h: Likewise.
+
+2018-02-20  Max Filippov  <jcmvbkbc@gmail.com>
+
+       * config/xtensa/ieee754-df.S (__adddf3_aux): Add
+       .literal_position directive.
+       * config/xtensa/ieee754-sf.S (__addsf3_aux): Likewise.
+
+2018-02-14  Igor Tsimbalist  <igor.v.tsimbalist@intel.com>
+
+       PR target/84148
+       * configure: Regenerate.
+
+2018-02-16  Igor Tsimbalist  <igor.v.tsimbalist@intel.com>
+
+       PR target/84239
+       * config/i386/shadow-stack-unwind.h (_Unwind_Frames_Extra):
+       Include cetintrin.h not x86intrin.h.
+
+2018-02-08  Igor Tsimbalist  <igor.v.tsimbalist@intel.com>
+
+       PR target/84239
+       * config/i386/shadow-stack-unwind.h (_Unwind_Frames_Extra):
+       Use new _get_ssp and _inc_ssp intrinsics.
+
+2018-02-02  Julia Koval  <julia.koval@intel.com>
+
+       * config/i386/cpuinfo.h (processor_subtypes): Add INTEL_COREI7_ICELAKE.
+
+2018-01-26  Claudiu Zissulescu  <claziss@synopsys.com>
+
+       * config/arc/lib1funcs.S (__udivmodsi4): Use safe version for RF16
+       option.
+       (__divsi3): Use RF16 safe registers.
+       (__modsi3): Likewise.
+
+2018-01-23  Max Filippov  <jcmvbkbc@gmail.com>
+
+       * config/xtensa/ieee754-df.S (__addsf3, __subsf3, __mulsf3)
+       (__divsf3): Make NaN return value quiet.
+       * config/xtensa/ieee754-sf.S (__adddf3, __subdf3, __muldf3)
+       (__divdf3): Make NaN return value quiet.
+
+2018-01-22  Sebastian Perta  <sebastian.perta@renesas.com>
+
+       * config/rl78/anddi3.S: New assembly file.
+       * config/rl78/t-rl78: Added anddi3.S to LIB2ADD.
+
+2018-01-22  Sebastian Perta  <sebastian.perta@renesas.com>
+
+       * config/rl78/umindi3.S: New assembly file.
+       * config/rl78/t-rl78: Added umindi3.S to LIB2ADD.
+
+2018-01-22  Sebastian Perta  <sebastian.perta@renesas.com>
+
+       * config/rl78/smindi3.S: New assembly file.
+       * config/rl78/t-rl78: Added smindi3.S to LIB2ADD.
+
+2018-01-22  Sebastian Perta  <sebastian.perta@renesas.com>
+       * config/rl78/smaxdi3.S: New assembly file.
+       * config/rl78/t-rl78: Added smaxdi3.S to LIB2ADD.
+
+2018-01-22  Sebastian Perta  <sebastian.perta@renesas.com>
+       * config/rl78/umaxdi3.S: New assembly file.
+       * config/rl78/t-rl78: Added umaxdi3.S to LIB2ADD.
+       
+2018-01-21  John David Anglin  <danglin@gcc.gnu.org>
+
+       PR lto/83452
+       * config/pa/stublib.c (L_gnu_lto_v1): New stub definition.
+       * config/pa/t-stublib (gnu_lto_v1-stub.o): Add make fragment.
+       
+2018-01-13  Richard Sandiford  <richard.sandiford@linaro.org>
+
+       * config/aarch64/value-unwind.h (aarch64_vg): New function.
+       (DWARF_LAZY_REGISTER_VALUE): Define.
+       * unwind-dw2.c (_Unwind_GetGR): Use DWARF_LAZY_REGISTER_VALUE
+       to provide a fallback register value.
+
+2018-01-08  Michael Meissner  <meissner@linux.vnet.ibm.com>
+
+       * config/rs6000/quad-float128.h (IBM128_TYPE): Explicitly use
+       __ibm128, instead of trying to use long double.
+       (CVT_FLOAT128_TO_IBM128): Use TFtype instead of __float128 to
+       accomidate -mabi=ieeelongdouble multilibs.
+       (CVT_IBM128_TO_FLOAT128): Likewise.
+       * config/rs6000/ibm-ldouble.c (IBM128_TYPE): New macro to define
+       the appropriate IBM extended double type.
+       (__gcc_qadd): Change all occurances of long double to IBM128_TYPE.
+       (__gcc_qsub): Likewise.
+       (__gcc_qmul): Likewise.
+       (__gcc_qdiv): Likewise.
+       (pack_ldouble): Likewise.
+       (__gcc_qneg): Likewise.
+       (__gcc_qeq): Likewise.
+       (__gcc_qne): Likewise.
+       (__gcc_qge): Likewise.
+       (__gcc_qle): Likewise.
+       (__gcc_stoq): Likewise.
+       (__gcc_dtoq): Likewise.
+       (__gcc_itoq): Likewise.
+       (__gcc_utoq): Likewise.
+       (__gcc_qunord): Likewise.
+       * config/rs6000/_mulkc3.c (toplevel): Include soft-fp.h and
+       quad-float128.h for the definitions.
+       (COPYSIGN): Use the f128 version instead of the q version.
+       (INFINITY): Likewise.
+       (__mulkc3): Use TFmode/TCmode for float128 scalar/complex types.
+       * config/rs6000/_divkc3.c (toplevel): Include soft-fp.h and
+       quad-float128.h for the definitions.
+       (COPYSIGN): Use the f128 version instead of the q version.
+       (INFINITY): Likewise.
+       (FABS): Likewise.
+       (__divkc3): Use TFmode/TCmode for float128 scalar/complex types.
+       * config/rs6000/extendkftf2-sw.c (__extendkftf2_sw): Likewise.
+       * config/rs6000/trunctfkf2-sw.c (__trunctfkf2_sw): Likewise.
+
+2018-01-05  Sebastian Huber  <sebastian.huber@embedded-brains.de>
+
+       * config.host (epiphany-*-elf*): Add (epiphany-*-rtems*)
+       configuration.
+
+2018-01-03  Jakub Jelinek  <jakub@redhat.com>
+
+       Update copyright years.
+
+2017-12-12  Kito Cheng  <kito.cheng@gmail.com>
 
        * config/riscv/t-elf: Use multi3.c instead of multi3.S.
        * config/riscv/multi3.c: New file.
        shared-object.mk, siditi-object.mk, static-object.mk: New files.
        * configure: Generated.
 \f
-Copyright (C) 2007-2017 Free Software Foundation, Inc.
+Copyright (C) 2007-2018 Free Software Foundation, Inc.
 
 Copying and distribution of this file, with or without modification,
 are permitted in any medium without royalty provided the copyright