Martin Liska [Tue, 9 Nov 2021 13:42:26 +0000 (14:42 +0100)]
Define m_rtx_reuse_manager only if not GENERATOR_FILE.
gcc/ChangeLog:
* print-rtl.c (rtx_writer::rtx_writer): Make the compilation
conditional based on
* print-rtl.h (class rtx_writer): Likewise.
Thomas Schwinge [Tue, 31 Aug 2021 21:30:25 +0000 (23:30 +0200)]
Generalize 'gcc/input.h:struct location_hash'
This is currently only used here ('gcc/input.h:class string_concat_db'), but is
actually generally useful, so advertize it as such.
Per the rationale given, we may use 'BUILTINS_LOCATION' as spare value for
'Deleted', in addition to the existing use of 'UNKNOWN_LOCATION' as spare value
for 'Empty'.
gcc/
* input.h (location_hash): Use 'BUILTINS_LOCATION' as spare value
for 'Deleted'. Turn into a '#define'.
Aldy Hernandez [Tue, 9 Nov 2021 09:43:25 +0000 (10:43 +0100)]
Cleanup path solver dumps.
This patch makes the path solver dumps a bit more consistent.
Tested on x86-64 Linux.
gcc/ChangeLog:
* gimple-range-path.cc (path_range_query::dump): Clean up.
(path_range_query::compute_ranges): Same.
* value-relation.cc (path_oracle::dump): Same.
Aldy Hernandez [Tue, 9 Nov 2021 09:14:25 +0000 (10:14 +0100)]
Remove TDF_THREADING flag in favor of param.
I am returning a TDF_* flag to the queue of available entries as I am
unconvinced that we need to burn an entire flag for internal debugging
constructs, especially since we seem to be running out of them.
I've added a --param=threader-debug entry similar to the one we use for
ranger debugging. Currently this only affects the backward threader,
but since the DOM threader is an outlier and on the chopping block, I
avoided using the "backward" name.
Tested on x86-64 Linux.
gcc/ChangeLog:
* dumpfile.c (dump_options): Remove TDF_THREADING entry.
* dumpfile.h (enum dump_flag): Remove TDF_THREADING and adjust
remaining entries.
* flag-types.h (enum threader_debug): New.
* gimple-range-path.cc (DEBUG_SOLVER): Use param_threader_debug.
* params.opt: Add entry for --param=threader-debug=.
Thomas Schwinge [Tue, 9 Nov 2021 09:55:15 +0000 (10:55 +0100)]
Get rid of infinite recursion for 'typedef' used with GTY-marked 'gcc/diagnostic-spec.h:nowarn_map' [PR101204, PR103157]
Reproduced with clang version 10.0.0-4ubuntu1:
gtype-desc.c:11333:1: warning: all paths through this function will call itself [-Winfinite-recursion]
... as well as some GCC's '-O2 -fdump-tree-optimized':
void gt_pch_nx(int_hash<unsigned int, 0u, 4294967295u>*, gt_pointer_operator, void*) ([...])
{
<bb 2>:
<bb 3>:
goto <bb 3>;
}
That three-arguments 'gt_pch_nx' function as well as two one-argument
'gt_ggc_mx', 'gt_pch_nx' functions now turn empty:
[...]
void
-gt_ggc_mx (int_hash<location_t,0,UINT_MAX>& x_r ATTRIBUTE_UNUSED)
+gt_ggc_mx (struct xint_hash_t& x_r ATTRIBUTE_UNUSED)
{
- int_hash<location_t,0,UINT_MAX> * ATTRIBUTE_UNUSED x = &x_r;
- gt_ggc_mx (&((*x)));
+ struct xint_hash_t * ATTRIBUTE_UNUSED x = &x_r;
}
[...]
void
-gt_pch_nx (int_hash<location_t,0,UINT_MAX>& x_r ATTRIBUTE_UNUSED)
+gt_pch_nx (struct xint_hash_t& x_r ATTRIBUTE_UNUSED)
{
- int_hash<location_t,0,UINT_MAX> * ATTRIBUTE_UNUSED x = &x_r;
- gt_pch_nx (&((*x)));
+ struct xint_hash_t * ATTRIBUTE_UNUSED x = &x_r;
}
[...]
void
-gt_pch_nx (int_hash<location_t,0,UINT_MAX>* x ATTRIBUTE_UNUSED,
+gt_pch_nx (struct xint_hash_t* x ATTRIBUTE_UNUSED,
ATTRIBUTE_UNUSED gt_pointer_operator op,
ATTRIBUTE_UNUSED void *cookie)
{
- gt_pch_nx (&((*x)), op, cookie);
}
[...]
gcc/
PR middle-end/101204
PR other/103157
* diagnostic-spec.h (typedef xint_hash_t): Turn into...
(struct xint_hash_t): ... this.
* doc/gty.texi: Update.
Przemyslaw Wirkus [Tue, 9 Nov 2021 09:40:05 +0000 (09:40 +0000)]
arm: add armv9-a architecture to -march
In this patch:
+ Add `armv9-a` to -march.
+ Update multilib with armv9-a and armv9-a+simd.
gcc/ChangeLog:
* config/arm/arm-cpus.in (armv9): New define.
(ARMv9a): New group.
(armv9-a): New arch definition.
* config/arm/arm-tables.opt: Regenerate.
* config/arm/arm.h (BASE_ARCH_9A): New arch enum value.
* config/arm/t-aprofile: Added armv9-a and armv9+simd.
* config/arm/t-arm-elf: Added arm9-a, v9_fps and all_v9_archs
to MULTILIB_MATCHES.
* config/arm/t-multilib: Added v9_a_nosimd_variants and
v9_a_simd_variants to MULTILIB_MATCHES.
* doc/invoke.texi: Update docs.
gcc/testsuite/ChangeLog:
* gcc.target/arm/multilib.exp: Update test with armv9-a entries.
* lib/target-supports.exp (v9a): Add new armflag.
(__ARM_ARCH_9A__): Add new armdef.
Martin Jambor [Tue, 9 Nov 2021 10:32:20 +0000 (11:32 +0100)]
ipa: Fix segfault when remapping debug_binds with expressions (PR 103132)
My initial implementation of the method
ipa_param_body_adjustments::remap_with_debug_expressions was based on
the assumption that if it was asked to remap an expression (as opposed
to a simple SSA_NAME), the expression would not contain an SSA_NAME
operand which is to be debug-reset. While that is true for when
called from ipa_param_body_adjustments::prepare_debug_expressions, it
turns out it is not true when invoked from remap_gimple_stmt in
tree-inline.c. This patch adds a simple logic to handle such cases
and simply map the entire value to NULL_TREE in those cases.
gcc/ChangeLog:
2021-11-08 Martin Jambor <mjambor@suse.cz>
PR ipa/103132
* ipa-param-manipulation.c (replace_with_mapped_expr): Early
return with error_mark_mode when part of expression is mapped to
NULL.
(ipa_param_body_adjustments::remap_with_debug_expressions): Set
mapped value to NULL if walk_tree returns error_mark_mode.
gcc/testsuite/ChangeLog:
2021-11-08 Martin Jambor <mjambor@suse.cz>
PR ipa/103132
* gcc.dg/ipa/pr103132.c: New test.
Ghjuvan Lacambre [Tue, 2 Nov 2021 10:32:24 +0000 (11:32 +0100)]
[Ada] Fix Codepeer warning
gcc/ada/
* freeze.adb (Check_Inherited_Conditions): Initialize
Ifaces_List.
Eric Botcazou [Wed, 27 Oct 2021 21:51:07 +0000 (23:51 +0200)]
[Ada] Tidy up implementation of Has_Compatible_Type
gcc/ada/
* sem_ch4.adb (Analyze_Membership_Op) <Find_Interpretation>: Handle
both overloaded and non-overloaded cases.
<Try_One_Interp>: Do a reversed call to Covers if the outcome of the
call to Has_Compatible_Type is false.
Simplify implementation after change to Find_Interpretation.
(Analyze_User_Defined_Binary_Op): Be prepared for previous errors.
(Find_Comparison_Types) <Try_One_Interp>: Do a reversed call to
Covers if the outcome of the call to Has_Compatible_Type is false.
(Find_Equality_Types) <Try_One_Interp>: Likewise.
* sem_type.adb (Has_Compatible_Type): Remove the reversed calls to
Covers. Add explicit return on all paths.
Justin Squirek [Fri, 29 Oct 2021 19:42:28 +0000 (15:42 -0400)]
[Ada] Invalid GPR_PROJECT_PATH_FILE confuses gnatls
gcc/ada/
* gnatls.adb (Initialize_Default_Project_Path): Remove early
abort on invalid path.
Eric Botcazou [Mon, 1 Nov 2021 09:30:51 +0000 (10:30 +0100)]
[Ada] Print Storage_Pool and Procedure_To_Call fields
gcc/ada/
* sprint.adb (Sprint_Node_Actual) <N_Allocator>: Also print the
Procedure_To_Call field if it is present.
<N_Extended_Return_Statement>: Also print the Storage_Pool and
Procedure_To_Call fields if they are present.
<N_Free_Statement>: Likewise.
<N_Simple_Return_Statement>: Likewise.
Patrick Bernardi [Thu, 28 Oct 2021 22:05:16 +0000 (18:05 -0400)]
[Ada] Do not indicate a pending abort if the task is already aborting
gcc/ada/
* libgnarl/s-tasini.adb (Check_Abort_Status): Don't return 1 if
the task is aborting.
* libgnarl/s-tasini.ads (Check_Abort_Status): Update comment.
Patrick Bernardi [Thu, 28 Oct 2021 19:05:17 +0000 (15:05 -0400)]
[Ada] RTEMS: use atomic primitive packages on aarch64
gcc/ada/
* Makefile.rtl: Add ATOMICS_TARGET_PAIRS and
ATOMICS_BUILTINS_TARGET_PAIRS to LIBGNAT_TARGET_PAIRS for
aarch64-rtems.
Justin Squirek [Sat, 30 Oct 2021 03:43:38 +0000 (23:43 -0400)]
[Ada] Hang on compilation of unit with type extension in body
gcc/ada/
* exp_util.adb (Ancestor_Primitive): Prevent return of an
subprogram alias when the ancestor primitive of the alias is the
same as Subp.
Justin Squirek [Fri, 29 Oct 2021 20:40:17 +0000 (16:40 -0400)]
[Ada] Crash on expansion of tagged membership test
gcc/ada/
* exp_ch4.adb (Tagged_Membership): Use corresponding record type
when expanding a tagged membership test on protected types.
Alexandre Oliva [Wed, 27 Oct 2021 21:26:27 +0000 (18:26 -0300)]
[Ada] Improve integration of strub with type systems
gcc/ada/
* strub.adb, strub.ads: New files.
* exp_attr.adb (Access_Cases): Copy strub mode to subprogram type.
* exp_disp.adb (Expand_Dispatching_Call): Likewise.
* freeze.adb (Check_Inherited_Conditions): Check that strub modes
match overridden subprograms and interfaces.
(Freeze_All): Renaming declarations too.
* sem_attr.adb (Resolve_Attribute): Reject 'Access to
strub-annotated data object.
* sem_ch3.adb (Derive_Subprogram): Copy strub mode to
inherited subprogram.
* sem_prag.adb (Analyze_Pragma): Propagate Strub Machine_Attribute
from access-to-subprogram to subprogram type when required,
but not from access-to-data to data type. Mark the entity that
got the pragma as having a gigi rep item.
* sem_res.adb (Resolve): Reject implicit conversions that
would change strub modes.
(Resolve_Type_Conversions): Reject checked conversions
between incompatible strub modes.
* doc/gnat_rm/security_hardening_features.rst: Update.
* gnat_rm.texi: Regenerate.
* libgnat/a-except.ads (Raise_Exception): Revert strub-callable
annotation in public subprogram.
* libgnat/s-arit128.ads (Multiply_With_Ovflo_Check128): Likewise.
* libgnat/s-arit64.ads (Multiply_With_Ovflo_Check64): Likewise.
* libgnat/s-secsta.ads (SS_Allocate): Likewise.
(SS_Mark, SS_Release): Likewise.
* gcc-interface/Make-lang.in (GNAT_ADA_OBJS): Add ada/strub.o.
Etienne Servais [Fri, 29 Oct 2021 10:32:11 +0000 (12:32 +0200)]
[Ada] Initialize Current_Source_Unit
gcc/ada/
* sinput.ads: Initialize Current_Source_Unit to No_Unit.
Piotr Trojanek [Wed, 27 Oct 2021 11:33:53 +0000 (13:33 +0200)]
[Ada] Use atomics in runtime on ARM and Aarch64 VxWorks
gcc/ada/
* Makefile.rtl (ARM and Aarch64 VxWorks): Use atomic variants of
runtime units.
* libgnat/a-strunb__shared.ads: Mention AARCH64 and ARM as
supported.
* libgnat/s-atocou.ads: Likewise.
Ghjuvan Lacambre [Wed, 27 Oct 2021 11:26:38 +0000 (13:26 +0200)]
[Ada] Accessibility fix
gcc/ada/
* sem_util.adb (Accessibility_Level): Use init_proc_level_formal
instead of computing a new accessibility level from Scope_Depth
(E).
Johannes Kliemann [Mon, 25 Oct 2021 13:53:59 +0000 (13:53 +0000)]
[Ada] Add gcc specs with vxworks7 base addresses for cert
gcc/ada/
* vxworks7-cert-rtp-link.spec: Replace the definition of
__wrs_rtp_base with the base_link spec.
* vxworks7-cert-rtp-base-link.spec: Add base_link spec with
__wrs_rtp_base definition for all architectures.
* vxworks7-cert-rtp-base-link__ppc64.spec: Add base_link spec
with __wrs_rtp_base definition for ppc64.
* vxworks7-cert-rtp-base-link__x86.spec: Add base_link spec with
__wrs_rtp_base definition for x86.
* vxworks7-cert-rtp-base-link__x86_64.spec: Add base_link spec
with __wrs_rtp_base definition for x86_64.
Piotr Trojanek [Wed, 27 Oct 2021 18:58:37 +0000 (20:58 +0200)]
[Ada] Cleanup copying of parameter lists
gcc/ada/
* sem_util.adb (Copy_Parameter_List): Refactor to remove
repeated calls to First_Formal.
Piotr Trojanek [Wed, 27 Oct 2021 18:43:24 +0000 (20:43 +0200)]
[Ada] Cleanup building of renamed equality
gcc/ada/
* exp_ch8.adb (Build_Body_For_Renaming): Remove unnecessary
calls to Sloc; set Handled_Statement_Sequence when building
subprogram body; whitespace cleanup.
Piotr Trojanek [Wed, 27 Oct 2021 15:32:52 +0000 (17:32 +0200)]
[Ada] Refine type in expansion of array equality
gcc/ada/
* exp_ch4.adb (Handle_One_Dimension): Parameter N must be always
positive, because it is translated into expression of 'First and
'Last.
Piotr Trojanek [Wed, 27 Oct 2021 14:34:20 +0000 (16:34 +0200)]
[Ada] Simplify building of entity occurrences
gcc/ada/
* tbuild.adb (New_Occurrence_Of): Simplify by reusing
Make_Identifier.
Etienne Servais [Tue, 26 Oct 2021 16:08:47 +0000 (18:08 +0200)]
[Ada] Return a valid value when handling Constraint_Error
gcc/ada/
* scng.adb (Set_Start_Column): Return Column_Number'Last when
handling Constraint_Error. Plus reformat a strangely 7 level if
loop.
Piotr Trojanek [Tue, 26 Oct 2021 16:05:43 +0000 (18:05 +0200)]
[Ada] Simplify expansion of array equality
gcc/ada/
* exp_ch4.adb (Test_Empty_Arrays): Simplify with Evolve_Or_Else;
remove unnecessary call to Relocate_Node.
(Test_Lengths_Correspond): Likewise.
Piotr Trojanek [Wed, 27 Oct 2021 08:33:32 +0000 (10:33 +0200)]
[Ada] Reference in Unbounded_String is almost never null
gcc/ada/
* libgnat/a-strunb.adb (Deallocate): Rename Reference_Copy to
Old, to make the code similar to other routines in this package.
(Realloc_For_Chunk): Use a temporary, deallocate the previous
string using a null-allowing copy of the string reference.
Ghjuvan Lacambre [Wed, 27 Oct 2021 08:26:00 +0000 (10:26 +0200)]
[Ada] sem_util.adb: fix typos in Accessibility_Level docs
gcc/ada/
* sem_util.adb (Accessibility_Level): Fix typos.
Tom Tromey [Wed, 6 Oct 2021 20:11:24 +0000 (14:11 -0600)]
[Ada] Document the SuSE kernel bug
gcc/ada/
* doc/gnat_ugn/platform_specific_information.rst: Document the
SuSE kernel bug.
* gnat_ugn.texi: Regenerate.
Gary Dismukes [Tue, 26 Oct 2021 00:45:50 +0000 (20:45 -0400)]
[Ada] Errors on globals in expressions of predicate aspects in generic bodies
gcc/ada/
* sem_ch13.adb (Freeze_Entity_Checks): Analyze the expression of
a pragma Predicate associated with an aspect at the freeze point
of the type, to ensure that references to globals get saved when
the aspect occurs within a generic body. Also, add
Aspect_Static_Predicate to the choices of the membership test of
the enclosing guard.
Piotr Trojanek [Tue, 26 Oct 2021 15:57:59 +0000 (17:57 +0200)]
[Ada] Tune comment about expansion of array equality
gcc/ada/
* exp_ch4.adb (Arr_Attr): Refine type of the parameter from Int
to Pos; refine name of the parameter from Num to Dim; fix
reference to "Expr" in comment.
Steve Baird [Thu, 21 Oct 2021 17:47:06 +0000 (10:47 -0700)]
[Ada] Add missing regular expression syntax error check
gcc/ada/
* libgnat/s-regexp.adb (Compile.Check_Well_Formed_Patern): When
a "|" operator is encountered in a pattern, check that it is not
the last character of the pattern.
Yannick Moy [Tue, 26 Oct 2021 10:21:40 +0000 (12:21 +0200)]
[Ada] Complete support for prefixed call on subtypes/derived types
gcc/ada/
* sem_ch3.adb (Analyze_Subtype_Declaration,
Derived_Type_Declaration): Initialize list of primitive
operations when Extensions_Allowed.
Piotr Trojanek [Mon, 25 Oct 2021 19:15:58 +0000 (21:15 +0200)]
[Ada] Fix detection of array aggregates with single others associations
gcc/ada/
* checks.adb (Apply_Constraint_Check): Guard against calling
Choices when the first association in an array aggregate is a
N_Iterated_Component_Association node.
Yannick Moy [Mon, 25 Oct 2021 10:48:22 +0000 (12:48 +0200)]
[Ada] Fix support for prefixed call with incomplete type declarations
gcc/ada/
* sem_ch3.adb (Analyze_Incomplete_Type_Decl): Add the missing
initialization.
Piotr Trojanek [Mon, 25 Oct 2021 14:33:24 +0000 (16:33 +0200)]
[Ada] Guard against illegal items in Global but not Depends
gcc/ada/
* sem_prag.adb (Check_Usage): Guard against calling Usage_Error
with illegal Item_Id. The intention to do this was already
described in the comment but not implemented.
Etienne Servais [Wed, 22 Sep 2021 15:20:27 +0000 (17:20 +0200)]
[Ada] Build activation chain for BIP only when needed
gcc/ada/
* exp_ch6.adb (Expand_Actuals): Add a condition to check for the
possibility of task.
Aldy Hernandez [Fri, 8 Oct 2021 13:54:23 +0000 (15:54 +0200)]
Convert strlen pass from evrp to ranger.
The following patch converts the strlen pass from evrp to ranger,
leaving DOM as the last remaining user.
No additional cleanups have been done. For example, the strlen pass
still has uses of VR_ANTI_RANGE, and the sprintf still passes around
pairs of integers instead of using a proper range. Fixing this
could further improve these passes.
Basically the entire patch is just adjusting the calls to range_of_expr
to include context. The previous context of si->stmt was mostly
empty, so not really useful ;-).
With ranger we are now able to remove the range calculation from
before_dom_children entirely. Just working with the ranger on-demand
catches all the strlen and sprintf testcases with the exception of
builtin-sprintf-warn-22.c which is due to a limitation of the sprintf
code. I have XFAILed the test and documented what the problem is.
On a positive note, these changes found two possible sprintf overflow
bugs in the C++ and Fortran front-ends which I have fixed below.
Tested on x86-64 Linux.
gcc/ChangeLog:
* tree-ssa-strlen.c (compare_nonzero_chars): Pass statement
context to ranger.
(get_addr_stridx): Same.
(get_stridx): Same.
(get_range_strlen_dynamic): Same.
(handle_builtin_strlen): Same.
(handle_builtin_strchr): Same.
(handle_builtin_strcpy): Same.
(maybe_diag_stxncpy_trunc): Same.
(handle_builtin_stxncpy_strncat): Same.
(handle_builtin_memcpy): Same.
(handle_builtin_strcat): Same.
(handle_alloc_call): Same.
(handle_builtin_memset): Same.
(handle_builtin_string_cmp): Same.
(handle_pointer_plus): Same.
(count_nonzero_bytes_addr): Same.
(count_nonzero_bytes): Same.
(handle_store): Same.
(fold_strstr_to_strncmp): Same.
(handle_integral_assign): Same.
(check_and_optimize_stmt): Same.
(class strlen_dom_walker): Replace evrp with ranger.
(strlen_dom_walker::before_dom_children): Remove evrp.
(strlen_dom_walker::after_dom_children): Remove evrp.
* gimple-ssa-warn-access.cc (maybe_check_access_sizes):
Restrict sprintf output.
gcc/cp/ChangeLog:
* ptree.c (cxx_print_xnode): Add more space to pfx array.
gcc/fortran/ChangeLog:
* misc.c (gfc_dummy_typename): Make sure ts->kind is
non-negative.
gcc/testsuite/ChangeLog:
* gcc.dg/tree-ssa/builtin-sprintf-warn-22.c: XFAIL.
Rasmus Villemoes [Fri, 5 Nov 2021 14:37:33 +0000 (15:37 +0100)]
libstdc++: only define _GLIBCXX_HAVE_TLS for VxWorks >= 6.6
According to
https://gcc.gnu.org/legacy-ml/gcc-patches/2008-03/msg01698.html, the
TLS support, including the __tls_lookup function, was added to VxWorks
in 6.6.
It certainly doesn't exist on our VxWorks 5 platform, but the fallback
code in eh_globals.cc using __gthread_key_create() etc. used to work
just fine.
libstdc++-v3/ChangeLog:
* config/os/vxworks/os_defines.h (_GLIBCXX_HAVE_TLS): Only
define for VxWorks >= 6.6.
Martin Liska [Tue, 9 Nov 2021 08:46:09 +0000 (09:46 +0100)]
Fix clang -Wcast-qual warning.
gcc/ChangeLog:
* genconditions.c (write_one_condition): Add const qualifier
to pointer casting.
Jeff Law [Tue, 9 Nov 2021 04:23:34 +0000 (23:23 -0500)]
Minor optimization of variable bit testing
gcc/
* match.pd: New pattern to simplify (1 << n) & M ==/!= 0 for M
being a power of 2.
gcc/testsuite
* gcc.dg/tree-ssa/bittest.c: New test
GCC Administrator [Tue, 9 Nov 2021 00:16:21 +0000 (00:16 +0000)]
Daily bump.
Bill Schmidt [Mon, 8 Nov 2021 21:08:59 +0000 (15:08 -0600)]
rs6000: Add escape-newline support for builtins files
2021-11-08 Bill Schmidt <wschmidt@linux.ibm.com>
gcc/
* config/rs6000/rs6000-builtin-new.def (VEC_INIT_V16QI): Use
escape-newline support.
(VEC_INIT_V4SI): Likewise.
(VEC_INIT_V8HI): Likewise.
(PACK_V1TI): Likewise.
(DIVDEU): Likewise.
(VFIRSTMISMATCHOREOSINDEX_V16QI): Likewise.
(VFIRSTMISMATCHOREOSINDEX_V8HI): Likewise.
(VFIRSTMISMATCHOREOSINDEX_V4SI): Likewise.
(CMPRB2): Likewise.
(VSTDCP): Likewise.
(VSIEDP): Likewise.
(FMAF128_ODD): Likewise.
(VSCEQPUO): Likewise.
(VSIEQP): Likewise.
(VSIEQPF): Likewise.
(VSTDCQP): Likewise.
(PACK_TD): Likewise.
(TABORTDC): Likewise.
(TABORTDCI): Likewise.
(SE_LXVRBX): Likewise.
(SE_LXVRHX): Likewise.
(SE_LXVRWX): Likewise.
(SE_LXVRDX): Likewise.
(VREPLACE_UN_UV2DI): Likewise.
(VREPLACE_UN_UV4SI): Likewise.
(VREPLACE_UN_V2DI): Likewise.
(VREPLACE_ELT_UV2DI): Likewise.
(VREPLACE_ELT_V2DI): Likewise.
(ZE_LXVRBX): Likewise.
(ZE_LXVRHX): Likewise.
(ZE_LXVRWX): Likewise.
(ZE_LXVRDX): Likewise.
(CFUGED): Likewise.
(CNTLZDM): Likewise.
(CNTTZDM): Likewise.
(PDEPD): Likewise.
(PEXTD): Likewise.
(PMXVBF16GER2): Likewise.
(PMXVBF16GER2_INTERNAL): Likewise.
(PMXVBF16GER2NN): Likewise.
(PMXVBF16GER2NN_INTERNAL): Likewise.
(PMXVBF16GER2NP): Likewise.
(PMXVBF16GER2NP_INTERNAL): Likewise.
(PMXVBF16GER2PN): Likewise.
(PMXVBF16GER2PN_INTERNAL): Likewise.
(PMXVBF16GER2PP): Likewise.
(PMXVBF16GER2PP_INTERNAL): Likewise.
(PMXVF16GER2): Likewise.
(PMXVF16GER2_INTERNAL): Likewise.
(PMXVF16GER2NN): Likewise.
(PMXVF16GER2NN_INTERNAL): Likewise.
(PMXVF16GER2NP): Likewise.
(PMXVF16GER2NP_INTERNAL): Likewise.
(PMXVF16GER2PN): Likewise.
(PMXVF16GER2PN_INTERNAL): Likewise.
(PMXVF16GER2PP): Likewise.
(PMXVF16GER2PP_INTERNAL): Likewise.
(PMXVF32GER_INTERNAL): Likewise.
(PMXVF32GERNN): Likewise.
(PMXVF32GERNN_INTERNAL): Likewise.
(PMXVF32GERNP): Likewise.
(PMXVF32GERNP_INTERNAL): Likewise.
(PMXVF32GERPN): Likewise.
(PMXVF32GERPN_INTERNAL): Likewise.
(PMXVF32GERPP): Likewise.
(PMXVF32GERPP_INTERNAL): Likewise.
(PMXVF64GER): Likewise.
(PMXVF64GER_INTERNAL): Likewise.
(PMXVF64GERNN): Likewise.
(PMXVF64GERNN_INTERNAL): Likewise.
(PMXVF64GERNP): Likewise.
(PMXVF64GERNP_INTERNAL): Likewise.
(PMXVF64GERPN): Likewise.
(PMXVF64GERPN_INTERNAL): Likewise.
(PMXVF64GERPP): Likewise.
(PMXVF64GERPP_INTERNAL): Likewise.
(PMXVI16GER2): Likewise.
(PMXVI16GER2_INTERNAL): Likewise.
(PMXVI16GER2PP): Likewise.
(PMXVI16GER2PP_INTERNAL): Likewise.
(PMXVI16GER2S): Likewise.
(PMXVI16GER2S_INTERNAL): Likewise.
(PMXVI16GER2SPP): Likewise.
(PMXVI16GER2SPP_INTERNAL): Likewise.
(PMXVI4GER8): Likewise.
(PMXVI4GER8_INTERNAL): Likewise.
(PMXVI4GER8PP): Likewise.
(PMXVI4GER8PP_INTERNAL): Likewise.
(PMXVI8GER4): Likewise.
(PMXVI8GER4_INTERNAL): Likewise.
(PMXVI8GER4PP): Likewise.
(PMXVI8GER4PP_INTERNAL): Likewise.
(PMXVI8GER4SPP): Likewise.
(PMXVI8GER4SPP_INTERNAL): Likewise.
* config/rs6000/rs6000-gen-builtins.c (MAXLINES): New macro.
(linebuf): Increase size.
(lines): New variable.
(lastline): Likewise.
(real_line_pos): New function.
(diag): Change signature.
(bif_diag): Change signature; support escape-newline handling.
(ovld_diag): Likewise.
(fatal): Move earlier.
(consume_whitespace): Adjust diag call.
(advance_line): Add escape-newline handling; call fatal.
(safe_inc_pos): Adjust diag call.
(match_identifier): Likewise.
(match_integer): Likewise.
(match_to_right_bracket): Call fatal instead of diag; adjust diag
call.
(match_basetype): Adjust diag calls.
(match_bracketed_pair): Likewise.
(match_const_restriction): Likewise.
(match_type): Likewise.
(parse_args): Likewise.
(parse_bif_attrs): Likewise.
(complete_vector_type): Likewise.
(complete_base_type): Likewise.
(parse_prototype): Likewise.
(parse_bif_entry): Likewise.
(parse_bif_stanza): Likewise.
(parse_ovld_entry): Likewise.
(parse_ovld_stanza): Likewise.
(main): Allocate buffers for lines[].
Eric Botcazou [Mon, 8 Nov 2021 21:09:16 +0000 (22:09 +0100)]
Fix couple of issues in large PIC model on x86-64/VxWorks
The first issue is that the !gotoff_operand path of legitimize_pic_address
in the large PIC model does not make use of REG when it is available, which
breaks for thunks because new pseudo-registers can no longer be created.
And the second issue is that the system compiler (LLVM) generates @GOTOFF
in large model even for RTP, so we do the same.
gcc/
* config/i386/i386.c (legitimize_pic_address): Adjust comment and
use the REG argument on the CM_LARGE_PIC code path as well.
* config/i386/predicates.md (gotoff_operand): Do not treat VxWorks
specially with the large code models.
Jan Hubicka [Mon, 8 Nov 2021 17:58:03 +0000 (18:58 +0100)]
Remove workaround allowing interposition of nested functions.
* gimple.c (gimple_call_static_chain_flags): Revert the workaround
allowing interposition since issues with binds_to_local_def were
hopefully solved.
Andrew MacLeod [Mon, 8 Nov 2021 14:32:42 +0000 (09:32 -0500)]
Don't calculate new values when using the private context callback.
When using rangers private callback mechanism to provide context
to fold_stmt calls, we are only suppose to be using the cache in read
only mode, never calculate new values.
gcc/
PR tree-optimization/103122
* gimple-range.cc (gimple_ranger::range_of_expr): Request the cache
entry with "calulate new values" set to false.
gcc/testsuite/
* g++.dg/pr103122.C: New.
Jan Hubicka [Mon, 8 Nov 2021 17:40:17 +0000 (18:40 +0100)]
Improve handling of some builtins.
For nested functions we output call to builtin_dwarf_cfa which
initializes frame entry used only for debugging. This however
prevents us from detecting functions containing nested functions
as const/pure or analyze side effects in modref.
builtin_dwarf_cfa is not documented and I wonder if it should be turned to
internal function. But I think we could consider functions using it const even
if in theory one can do things like test the return address and see the
difference between different frame addreses.
While doing so I also noticed that special_buitin_state handles quite few
builtins that are not special cased by ipa-modref. They do not make
user visible loads/stores and thus I think they shoul dbe annotated by
".c" to make this explicit for both modref and PTA.
Finally I aded dwarf_cfa and similar return_address to list of simple
bulitins since it compiles to simple stack frame load (and we consider
simple other builtins doing so).
* builtins.c (is_simple_builtin): Add builitin_dwarf_cfa
and builtin_return_address.
(builtin_fnspec): Annotate builtin_return,
bulitin_eh_pointer, builtin_eh_filter, builtin_unwind_resume,
builtin_cxa_end_cleanup, builtin_eh_copy_values,
builtin_frame_address, builtin_apply_args,
builtin_asan_before_dynamic_init, builtin_asan_after_dynamic_init,
builtin_prefetch, builtin_dwarf_cfa, builtin_return_addrss
as ".c"
* ipa-pure-const.c (special_builtin_state): Add builtin_dwarf_cfa
and builtin_return_address.
Jan Hubicka [Mon, 8 Nov 2021 17:38:09 +0000 (18:38 +0100)]
Move uncprop after modref
moveS uncprop after modref and pure/const pass and adds a comment that
this pass should alwasy be last since it is only supposed to help PHI lowering.
The pass replaces constant by SSA names that are known to be constant at the
place which hardly helps other passes.
gcc/ChangeLog:
PR tree-optimization/103177
* passes.def: Move uncprop after pure/const and modref.
Martin Jambor [Mon, 8 Nov 2021 16:49:54 +0000 (17:49 +0100)]
ipa: Unshare expresseions before putting them into debug statements (PR 103099, PR 103107)
My recent patch to improve debug experience when there are removed
parameters (by ipa-sra or ipa-split) was not careful to unshare the
expressions that were then put into debug statements, which manifests
itself as PR 103099. This patch adds unsharing them using
unshare_expr_without_location which is a bit more careful with stripping
locations than what we were doing manually and so also fixes PR 103107.
gcc/ChangeLog:
2021-11-08 Martin Jambor <mjambor@suse.cz>
PR ipa/103099
PR ipa/103107
* tree-inline.c (remap_gimple_stmt): Unshare the expression without
location before invoking remap_with_debug_expressions on it.
* ipa-param-manipulation.c
(ipa_param_body_adjustments::prepare_debug_expressions): Likewise.
gcc/testsuite/ChangeLog:
2021-11-08 Martin Jambor <mjambor@suse.cz>
PR ipa/103099
PR ipa/103107
* g++.dg/ipa/pr103099.C: New test.
* gcc.dg/ipa/pr103107.c: Likewise.
David Edelsohn [Mon, 8 Nov 2021 16:46:47 +0000 (11:46 -0500)]
powerpc: Fix vsx_splat_v4si_di breakage on Power8.
The vsx_splat_v4si_di pattern uses a Power8 and a Power9 instruction.
The final condition of TARGET_DIRECT_MODE_64BIT implicitly requires Power8.
The "we" constraint requires Power9, but also requires 64 bit. Because
the DImode pattern already requires 64 bit mode, this isn't horrible,
but it would be best to remove all uses of "we" constraint. The
mtvsrws instruction itself does not require 64 bit mode.
This patch reverts the previous change to fix the breakage.
gcc/ChangeLog:
* config/rs6000/vsx.md (vsx_splat_v4si_di): Revert "wa"
constraint to "we".
Richard Biener [Mon, 8 Nov 2021 14:21:08 +0000 (15:21 +0100)]
Fix spurious valgrind errors in irred loop verification
The sbitmap bitmap_{set,clear}_bit changes trigger spurious
uninit value use reportings from valgrind since we now
read the old value before setting/clearing a bit so
verify_loop_structures optimization to not clear the sbitmap is reported.
Fixed by using a temporary BB flag which should also be more
efficient in terms of cache re-use.
2021-11-08 Richard Biener <rguenther@suse.de>
* cfgloop.c (verify_loop_structure): Use a temporary BB flag
instead of an sbitmap to cache irreducible state.
Aldy Hernandez [Mon, 8 Nov 2021 11:13:33 +0000 (12:13 +0100)]
path solver: Avoid recalculating ranges already in the cache.
The problem here is an ordering issue with a path that starts
with 19->3:
<bb 3> [local count:
916928331]:
# value_20 = PHI <value_17(19), value_7(D)(17)>
# n_27 = PHI <n_16(19), 1(17)>
n_16 = n_27 + 4;
value_17 = value_20 / 10000;
if (value_20 >
42949672959999)
goto <bb 19>; [89.00%]
else
goto <bb 4>; [11.00%]
The problem here is that both value_17 and value_20 are in the set of
imports we must pre-calculate. The value_17 name occurs first in the
bitmap, so we try to resolve it first, which causes us to recursively
solve the value_20 range. We do so correctly and put them both in the
cache. However, when we try to solve value_20 from the bitmap, we
ignore that it already has a cached entry and try to resolve the PHI
with the wrong value of value_17:
# value_20 = PHI <value_17(19), value_7(D)(17)>
The right thing to do is to avoid recalculating definitions already
solved.
Regstrapped and checked for # threads before and after on x86-64 Linux.
gcc/ChangeLog:
PR tree-optimization/103120
* gimple-range-path.cc (path_range_query::range_defined_in_block):
Bail if there's a cache entry.
gcc/testsuite/ChangeLog:
* gcc.dg/pr103120.c: New test.
Bill Schmidt [Mon, 8 Nov 2021 14:34:03 +0000 (08:34 -0600)]
rs6000: Miscellaneous uses of rs6000_builtins_decl_x
There are a few leftover places where we use the old rs6000_builtins_decl
array, but we need to use rs6000_builtins_decl_x instead when the new
builtins infrastructure is in play.
2021-11-08 Bill Schmidt <wschmidt@linux.ibm.com>
gcc/
* config/rs6000/rs6000.c (rs6000_builtin_reciprocal): Use
rs6000_builtin_decls_x when appropriate.
(add_condition_to_bb): Likewise.
(rs6000_atomic_assign_expand_fenv): Likewise.
Bill Schmidt [Mon, 8 Nov 2021 14:17:53 +0000 (08:17 -0600)]
rs6000: Update rs6000_builtin_decl
Create a new version of this function that uses the new infrastructure,
and particularly checks for supported builtins the new way.
2021-11-08 Bill Schmidt <wschmidt@linux.ibm.com>
gcc/
* config/rs6000/rs6000-call.c (rs6000_new_builtin_decl): New function.
(rs6000_builtin_decl): Call it.
Martin Liska [Mon, 8 Nov 2021 11:57:08 +0000 (12:57 +0100)]
Fix static chain dump.
gcc/ChangeLog:
* cgraph.c (cgraph_node::dump): Dump it from decl.
Martin Liska [Fri, 5 Nov 2021 15:50:06 +0000 (16:50 +0100)]
gcov-profile: Fix -fcompare-debug with -fprofile-generate [PR100520]
PR gcov-profile/100520
gcc/ChangeLog:
* coverage.c (coverage_compute_profile_id): Strip .gk when
compare debug is used.
* system.h (endswith): New function.
gcc/testsuite/ChangeLog:
* gcc.dg/pr100520.c: New test.
Martin Liska [Mon, 8 Nov 2021 09:27:47 +0000 (10:27 +0100)]
Dump static chain for cgraph_node.
gcc/ChangeLog:
* cgraph.c (cgraph_node::dump): Dump static_chain_decl.
Thomas Schwinge [Fri, 22 Oct 2021 14:01:54 +0000 (16:01 +0200)]
Fix 'Copyright (C) 2020-21' into '2020-2021'
'contrib/update-copyright.py' currently complains:
gcc/config/rs6000/rs6000-gen-builtins.c: unrecognised year: 21
gcc/config/rs6000/rs6000-overload.def: unrecognised year: 21
gcc/config/rs6000/rbtree.h: unrecognised year: 21
gcc/config/rs6000/rbtree.c: unrecognised year: 21
gcc/config/rs6000/rs6000-builtin-new.def: unrecognised year: 21
Fix up files added in commit
fa5f8b49e55caf5bb341f5eb6b5ab828b9286425
"rs6000: Red-black tree implementation for balanced tree search",
commit
4a720a9547320699aceda7d2e0b08de5ab40132f
"rs6000: Add initial input files",
commit
bd5b625228d545d5ecb35df24f9f094edc95e3fa
"rs6000: Initial create of rs6000-gen-builtins.c".
gcc/
* config/rs6000/rbtree.c: Fix 'Copyright (C) 2020-21' into '2020-2021'
* config/rs6000/rbtree.h: Likewise.
* config/rs6000/rs6000-builtin-new.def: Likewise.
* config/rs6000/rs6000-gen-builtins.c: Likewise.
* config/rs6000/rs6000-overload.def: Likewise.
Thomas Schwinge [Fri, 22 Oct 2021 13:54:42 +0000 (15:54 +0200)]
Fix 'contrib/update-copyright.py': 'TypeError: exceptions must derive from BaseException'
Running 'contrib/update-copyright.py' currently fails:
[...]
Traceback (most recent call last):
File "contrib/update-copyright.py", line 365, in update_copyright
canon_form = self.canonicalise_years (dir, filename, filter, years)
File "contrib/update-copyright.py", line 270, in canonicalise_years
(min_year, max_year) = self.year_range (years)
File "contrib/update-copyright.py", line 253, in year_range
year_list = [self.parse_year (year)
File "contrib/update-copyright.py", line 253, in <listcomp>
year_list = [self.parse_year (year)
File "contrib/update-copyright.py", line 250, in parse_year
raise self.BadYear (string)
TypeError: exceptions must derive from BaseException
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "contrib/update-copyright.py", line 796, in <module>
GCCCmdLine().main()
File "contrib/update-copyright.py", line 527, in main
self.copyright.process_tree (dir, filter)
File "contrib/update-copyright.py", line 458, in process_tree
self.process_file (dir, filename, filter)
File "contrib/update-copyright.py", line 421, in process_file
res = self.update_copyright (dir, filename, filter,
File "contrib/update-copyright.py", line 366, in update_copyright
except self.BadYear as e:
TypeError: catching classes that do not inherit from BaseException is not allowed
Fix up for commit
3b25e83536bcd1b2977659a2c6d9f0f9bf2a3152
"Port update-copyright.py to Python3".
contrib/
* update-copyright.py (class BadYear): Derive from 'Exception'.
Richard Sandiford [Mon, 8 Nov 2021 10:36:25 +0000 (10:36 +0000)]
aarch64: LD3/LD4 post-modify costs for struct modes
The LD3/ST3 and LD4/ST4 address cost code had no test coverage (oops).
This patch fixes that and updates it for the new structure modes.
The test only covers Advanced SIMD because SVE doesn't have
post-increment forms.
gcc/
* config/aarch64/aarch64.c (aarch64_ldn_stn_vectors): New function.
(aarch64_address_cost): Use it instead of testing for CImode and
XImode directly.
gcc/testsuite/
* gcc.target/aarch64/neoverse_v1_1.c: New test.
Richard Sandiford [Mon, 8 Nov 2021 10:36:25 +0000 (10:36 +0000)]
genmodes: Define NUM_MODE_* macros
I was working on a patch that needed to calculate the number of
modes in a particular class. It seemed better to have genmodes
generate this directly rather than do the kind of dance that
expmed.h had.
gcc/
* genmodes.c (emit_insn_modes_h): Define NUM_MODE_* macros.
* expmed.h (NUM_MODE_INT): Delete in favor of genmodes definitions.
(NUM_MODE_PARTIAL_INT, NUM_MODE_VECTOR_INT): Likewise.
* real.h (real_format_for_mode): Use NUM_MODE_FLOAT and
NUM_MODE_DECIMAL_FLOAT.
(REAL_MODE_FORMAT): Likewise.
Andre Vieira [Mon, 8 Nov 2021 09:35:45 +0000 (09:35 +0000)]
[vect] Consider outside costs earlier for epilogue loops
gcc/ChangeLog:
* tree-vect-loop.c (vect_better_loop_vinfo_p): Change how epilogue loop
costs are compared.
Richard Biener [Mon, 8 Nov 2021 08:08:12 +0000 (09:08 +0100)]
tree-optimization/103102 - fix error in vectorizer refactoring
This fixes an oversight that caused vectorized epilogues to have
versioning for niters applied.
2021-11-08 Richard Biener <rguenther@suse.de>
* tree-vectorizer.h (vect_create_loop_vinfo): Add main_loop_info
parameter.
* tree-vect-loop.c (vect_create_loop_vinfo): Likewise. Set
LOOP_VINFO_ORIG_LOOP_INFO and conditionalize set of
LOOP_VINFO_NITERS_ASSUMPTIONS.
(vect_analyze_loop_1): Adjust.
(vect_analyze_loop): Move loop constraint setting and
SCEV/niter reset here from vect_create_loop_vinfo to perform
it only once.
(vect_analyze_loop_form): Move dumping of symbolic niters
here from vect_create_loop_vinfo.
Jan Hubicka [Mon, 8 Nov 2021 06:52:45 +0000 (07:52 +0100)]
Add loads/stores relative to static chain in ipa-modref
Adds tracking of accesses relative to static chain into modref
load/stores analysis. This helps some Fortran benchmarks however it is still
quite limited. One problem is that we never discover functions with nested
functions as const, pure or not accessing global memory because it contains
__builtin_dward_cfa call which we believe to be non-pure.
Bootstrapped/regtested x86_64-linux. Plan to commit it tomorrow if there are
no complains and once periodic testers picks today modref changes.
Honza
gcc/ChangeLog:
* ipa-modref-tree.h (enum modref_special_parms): New enum.
(struct modref_access_node): update for special parms.
(struct modref_ref_node): Likewise.
(struct modref_parm_map): Likewise.
(struct modref_tree): Likewise.
* ipa-modref.c (dump_access): Likewise.
(get_access): Detect static chain.
(parm_map_for_arg): Take tree as arg instead of
stmt and index.
(merge_call_side_effects): Compute map for static chain.
(process_fnspec): Update.
(struct escape_point): Remove retslot_arg and static_chain_arg.
(analyze_parms): Update.
(compute_parm_map): Update.
(propagate_unknown_call): Update.
(modref_propagate_in_scc): Update.
(modref_merge_call_site_flags): Update.
(ipa_merge_modref_summary_after_inlining): Update.
* tree-ssa-alias.c (modref_may_conflict): Handle static chain.
* ipa-modref-tree.c (test_merge): Update.
gcc/testsuite/ChangeLog:
* gcc.dg/tree-ssa/modref-12.c: New test.
Haochen Gui [Tue, 2 Nov 2021 06:09:32 +0000 (14:09 +0800)]
Disables gimple folding for VSX_BUILTIN_XVMINDP, VSX_BUILTIN_XVMAXDP,ALTIVEC_BUILTIN_VMINFP and ALTIVEC_BUILTIN_VMAXFP when fast-math is not set.
gcc/
* config/rs6000/rs6000-call.c (rs6000_gimple_fold_builtin): Disable
gimple fold for VSX_BUILTIN_XVMINDP, ALTIVEC_BUILTIN_VMINFP,
VSX_BUILTIN_XVMAXDP, ALTIVEC_BUILTIN_VMAXFP when fast-math is not
set.
gcc/testsuite/
* gcc.target/powerpc/vec-minmax-1.c: New test.
* gcc.target/powerpc/vec-minmax-2.c: Likewise.
liuhongt [Fri, 5 Nov 2021 02:41:22 +0000 (10:41 +0800)]
Update documentation for -ftree-loop-vectorize and -ftree-slp-vectorize which are enabled by default at -02.
gcc/ChangeLog:
PR tree-optimization/103077
* doc/invoke.texi (Options That Control Optimization):
Update documentation for -ftree-loop-vectorize and
-ftree-slp-vectorize which are enabled by default at -02.
liuhongt [Mon, 8 Nov 2021 01:32:17 +0000 (09:32 +0800)]
Add !HONOR_SNANS to simplifcation: (trunc)copysign((extend)a, (extend)b) to copysign (a, b).
> Note that this is not safe with -fsignaling-nans, so needs to be disabled
> for that option (if there isn't already logic somewhere with that effect),
> because the extend will convert a signaling NaN to quiet (raising
> "invalid"), but copysign won't, so this transformation could result in a
> signaling NaN being wrongly returned when the original code would never
> have returned a signaling NaN.
>
> --
> Joseph S. Myers
> joseph@codesourcery.com
gcc/ChangeLog
PR target/102464
* match.pd (Simplifcation (trunc)copysign((extend)a, (extend)b)
to .COPYSIGN (a, b)): Add !HONOR_SNANS.
liuhongt [Mon, 8 Nov 2021 01:19:29 +0000 (09:19 +0800)]
[Gimple] Simplify (trunc)fma ((extend)a, (extend)b, (extend)c) to IFN_FMA (a,b, c).
a, b, c are same type as truncation type and has less precision than
extend type, the optimization is guarded under
flag_unsafe_math_optimizations.
gcc/ChangeLog:
PR target/102464
* match.pd: Simplify
(trunc)fma ((extend)a, (extend)b, (extend)c) to IFN_FMA (a, b,
c) under flag_unsafe_math_optimizations.
gcc/testsuite/ChangeLog:
* gcc.target/i386/pr102464-fma.c: New test.
GCC Administrator [Mon, 8 Nov 2021 00:16:17 +0000 (00:16 +0000)]
Daily bump.
Thomas Koenig [Sun, 7 Nov 2021 14:38:35 +0000 (15:38 +0100)]
Fix keyword name for co_reduce.
gcc/fortran/ChangeLog:
* intrinsic.c (add_subroutines): Change keyword "operator"
to the correct one, "operation".
* check.c (gfc_check_co_reduce): Change OPERATOR to
OPERATION in error messages.
* intrinsic.texi: Change OPERATOR to OPERATION in
documentation.
gcc/testsuite/ChangeLog:
* gfortran.dg/co_reduce_2.f90: New test.
* gfortran.dg/coarray_collectives_14.f90: Change OPERATOR
to OPERATION.
* gfortran.dg/coarray_collectives_16.f90: Likewise.
* gfortran.dg/coarray_collectives_9.f90: Likewise.
Co-authored by: Steve Kargl <steve@gcc.gnu.org>
Jan Hubicka [Sun, 7 Nov 2021 21:58:19 +0000 (22:58 +0100)]
Limit range of modref-max-depth
gcc/ChangeLog:
PR ipa/103055
* params.opt (modref-max-depth): Add range.
(modref-max-adjustments): Fix range.
Aldy Hernandez [Mon, 1 Nov 2021 14:50:38 +0000 (15:50 +0100)]
Remove VRP threader.
Now that things have stabilized, we can remove the old code.
I have left the hybrid threader in tree-ssa-threadedge, even though the
VRP threader was the only user, because we may need it as an interim
step for DOM threading removal.
Tested on x86-64 Linux.
gcc/ChangeLog:
* tree-pass.h (make_pass_vrp_threader): Remove.
* tree-ssa-threadbackward.c
(back_threader_profitability::profitable_path_p): Remove
ASSERT_EXPR references.
* tree-ssa-threadedge.c (jt_state::register_equivs_stmt): Same.
* tree-vrp.c (vrp_folder::simplify_casted_conds): Same.
(execute_vrp): Same.
(class hybrid_threader): Remove.
(hybrid_threader::hybrid_threader): Remove.
(hybrid_threader::~hybrid_threader): Remove.
(hybrid_threader::before_dom_children): Remove.
(hybrid_threader::after_dom_children): Remove.
(execute_vrp_threader): Remove.
(class pass_vrp_threader): Remove.
(make_pass_vrp_threader): Remove.
Sandra Loosemore [Thu, 4 Nov 2021 22:43:29 +0000 (15:43 -0700)]
Fortran: Diagnose all operands/arguments with constraint violations
04-Nov-2021 Sandra Loosemore <sandra@codesourcery.com>
Bernhard Reutner-Fischer <aldot@gcc.gnu.org>
PR fortran/101337
gcc/fortran/ChangeLog:
* interface.c (gfc_compare_actual_formal): Continue checking
all arguments after encountering an error.
* intrinsic.c (do_ts29113_check): Likewise.
* resolve.c (resolve_operator): Continue resolving on op2 error.
gcc/testsuite/ChangeLog:
* gfortran.dg/bessel_3.f90: Expect additional diagnostics from
multiple bad arguments in the call.
* gfortran.dg/pr24823.f: Likewise.
* gfortran.dg/pr39937.f: Likewise.
* gfortran.dg/pr41011.f: Likewise.
* gfortran.dg/pr61318.f90: Likewise.
* gfortran.dg/c-interop/c407b-2.f90: Remove xfails.
* gfortran.dg/c-interop/c535b-2.f90: Likewise.
Jan Hubicka [Sun, 7 Nov 2021 17:20:45 +0000 (18:20 +0100)]
Fix inter-procedural EAF flags propagation with respect to !binds_to_current_def_p
While proofreading the code for handling EAF flags of !binds_to_current_def_p I
noticed that the interprocedural dataflow actually ignores the flag possibly
introducing wrong code on quite complex interposable functions in non-trivial
recursion cycles (or at ltrans partition boundary).
This patch unifies the flags changes to single place (remove_useless_eaf_flags)
and does extend modref_merge_call_site_flags to do the right thing.
lto-bootstrapped/regtested x86_64-linux. Plan to commit it today after bit
more testing (firefox/clang build).
gcc/ChangeLog:
* gimple.c (gimple_call_arg_flags): Use interposable_eaf_flags.
(gimple_call_retslot_flags): Likewise.
(gimple_call_static_chain_flags): Likewise.
* ipa-modref.c (remove_useless_eaf_flags): Do not remove everything for
NOVOPS.
(modref_summary::useful_p): Likewise.
(modref_summary_lto::useful_p): Likewise.
(analyze_parms): Do not give up on NOVOPS.
(analyze_function): When dumping report chnages in EAF flags
between IPA and local pass.
(modref_merge_call_site_flags): Compute implicit eaf flags
based on callee ecf_flags and fnspec; if the function does not
bind to current defs use interposable_eaf_flags.
(modref_propagate_flags_in_scc): Update.
* ipa-modref.h (interposable_eaf_flags): New function.
Bill Schmidt [Sun, 7 Nov 2021 13:56:07 +0000 (07:56 -0600)]
rs6000: Replace the builtin expansion machinery
This patch forms the meat of the improvements for this patch series.
We develop a replacement for rs6000_expand_builtin and its supporting
functions, which are inefficient and difficult to maintain.
Differences between the old and new support in this patch include:
- Make use of the new builtin data structures, directly looking up
a function's information rather than searching for the function
multiple times;
- Test for enablement of builtins at expand time, to support #pragma
target changes within a compilation unit;
- Use the builtin function attributes (e.g., bif_is_cpu) to control
special handling;
- Refactor common code into one place; and
- Provide common error handling in one place for operands that are
restricted to specific values or ranges.
2021-11-07 Bill Schmidt <wschmidt@linux.ibm.com>
gcc/
* config/rs6000/rs6000-call.c (rs6000_expand_new_builtin): New
forward decl.
(rs6000_invalid_new_builtin): New function.
(rs6000_expand_builtin): Call rs6000_expand_new_builtin.
(rs6000_expand_ldst_mask): New function.
(new_cpu_expand_builtin): Likewise.
(elemrev_icode): Likewise.
(ldv_expand_builtin): Likewise.
(lxvrse_expand_builtin): Likewise.
(lxvrze_expand_builtin): Likewise.
(stv_expand_builtin): Likewise.
(new_mma_expand_builtin): Likewise.
(new_htm_spr_num): Likewise.
(new_htm_expand_builtin): Likewise.
(rs6000_expand_new_builtin): Likewise.
(rs6000_init_builtins): Initialize altivec_builtin_mask_for_load.
Jan Hubicka [Sun, 7 Nov 2021 08:35:16 +0000 (09:35 +0100)]
Implement intra-procedural dataflow in ipa-modref flags propagation.
implement the (long promised) intraprocedural dataflow for
propagating eaf flags, so we can handle parameters that participate
in loops in SSA graphs. Typical example are acessors that walk linked
lists, for example.
I implemented dataflow using the standard iteration over BBs in RPO some time
ago, but did not like it becuase it had measurable compile time impact with
very small code quality effect. This is why I kept mainline to do the DFS walk
instead. The reason is that we care about flags of SSA names that corresponds
to parameters and those can be often determined from a small fraction of the
SSA graph so solving dataflow for all SSA names in a function is a waste.
This patch implements dataflow more carefully. The DFS walk is kept in place to
solve acyclic cases and discover the relevat part of SSA graph into new graph
(which is similar to one used for inter-procedrual dataflow - we only need to
know the edges and if the access is direct or derefernced). The RPO iterative
dataflow then works on this simplified graph.
This seems to be fast in practice. For GCC linktime we do dataflow for 4881
functions. Out of that 4726 finishes in one iteration, 144 in two and 10 in 3.
Overall 31979 functions are analysed, so we do dataflow only for bit over of
10% of cases. 131123 edges are visited by the solver. I measured no compile
time impact of this.
gcc/ChangeLog:
* ipa-modref.c (modref_lattice): Add do_dataflow,
changed and propagate_to fields.
(modref_lattice::release): Free propagate_to
(modref_lattice::merge): Do not give up early on unknown
lattice values.
(modref_lattice::merge_deref): Likewise.
(modref_eaf_analysis): Update toplevel comment.
(modref_eaf_analysis::analyze_ssa_name): Record postponned ssa names;
do optimistic dataflow initialization.
(modref_eaf_analysis::merge_with_ssa_name): Build dataflow graph.
(modref_eaf_analysis::propagate): New member function.
(analyze_parms): Update to new API of modref_eaf_analysis.
GCC Administrator [Sun, 7 Nov 2021 00:16:19 +0000 (00:16 +0000)]
Daily bump.
Jan Hubicka [Sat, 6 Nov 2021 22:36:08 +0000 (23:36 +0100)]
Fix can_be_discarded_p wrt partitioned functions.
gcc/ChangeLog:
* cgraph.h (cgraph_node::can_be_discarded_p): Do not
return true on functions from other partition.
gcc/lto/ChangeLog:
PR ipa/103070
PR ipa/103058
* lto-partition.c (must_not_rename): Update comment.
(promote_symbol): Set resolution to LDPR_PREVAILING_DEF_IRONLY.
Harald Anlauf [Sat, 6 Nov 2021 18:42:01 +0000 (19:42 +0100)]
Fortran: error recovery on rank mismatch of array and its initializer
gcc/fortran/ChangeLog:
PR fortran/102715
* decl.c (add_init_expr_to_sym): Reject rank mismatch between
array and its initializer.
gcc/testsuite/ChangeLog:
PR fortran/102715
* gfortran.dg/pr68019.f90: Adjust error message.
* gfortran.dg/pr102715.f90: New test.
David Edelsohn [Sat, 6 Nov 2021 00:33:45 +0000 (20:33 -0400)]
powerpc: Fix vsx_splat_v4si in 32 bit mode
Tamar's recent patch to teach CSE to perform vector extract exercises
VSX splat more frequently, which exposed a constraint error for the
vsx_splat patterns. The pattern could be created for Power9, but
the "we constraint only provided alternatives in 64 bit mode. The
instructions are valid in 32 bit mode and SImode is allowed in VSX
registers. This patch updates the constraints from "we" to "wa" to
allow the pattern and fix the failing testcases.
gcc/ChangeLog:
* config/rs6000/vsx.md (vsx_splat_v4si): Change constraints to "wa".
(vsx_splat_v4si_di): Change constraint to "wa".
Aldy Hernandez [Fri, 5 Nov 2021 21:25:19 +0000 (22:25 +0100)]
path oracle: Do not look at root oracle for killed defs.
The problem here is that we are incorrectly threading 41->20->21 here:
<bb 35> [local count:
56063504182]:
_134 = M.10_120 + 1;
if (_71 <= _134)
goto <bb 19>; [11.00%]
else
goto <bb 41>; [89.00%]
...
...
...
<bb 41> [local count:
49896518755]:
<bb 20> [local count:
56063503181]:
# lb_75 = PHI <_134(41), 1(18)>
_117 = mstep_49 + lb_75;
_118 = _117 + -1;
_119 = mstep_49 + _118;
M.10_120 = MIN_EXPR <_119, _71>;
if (lb_75 > M.10_120)
goto <bb 21>; [11.00%]
else
goto <bb 22>; [89.00%]
First, lb_17 == _134 because of the PHI.
Second, _134 > M.10_120 because of _134 = M.10_120 + 1.
We then assume that lb_75 > M.10_120, but this is incorrect because
M.10_120 was killed along the path.
This incorrect thread causes the miscompilation in 527.cam4_r.
Tested on x86-64 and ppc64le Linux.
gcc/ChangeLog:
PR tree-optimization/103061
* value-relation.cc (path_oracle::path_oracle): Initialize
m_killed_defs.
(path_oracle::killing_def): Set m_killed_defs.
(path_oracle::query_relation): Do not look at the root oracle for
killed defs.
* value-relation.h (class path_oracle): Add m_killed_defs.
David Edelsohn [Sat, 6 Nov 2021 15:24:08 +0000 (11:24 -0400)]
testsuite: Use posix_memalign on AIX for tsvc
AIX does not provide memalign, so the testcases much use
posix_memalign for portability on AIX.
gcc/testsuite/ChangeLog:
* gcc.dg/vect/tsvc/tsvc.h (init): Use posix_memalign on AIX.
Aldy Hernandez [Thu, 4 Nov 2021 18:44:15 +0000 (19:44 +0100)]
Cleanup back_threader::find_path_to_names.
The main path discovery function was due for a cleanup. First,
there's a nagging goto and second, my bitmap use was sloppy. Hopefully
this makes the code easier for others to read.
Regstrapped on x86-64 Linux. I also made sure there were no difference
in the number of threads with this patch.
No functional changes.
gcc/ChangeLog:
* tree-ssa-threadbackward.c (back_threader::find_paths_to_names):
Remove gotos and other cleanups.
GCC Administrator [Sat, 6 Nov 2021 00:16:24 +0000 (00:16 +0000)]
Daily bump.
Harald Anlauf [Fri, 5 Nov 2021 22:48:20 +0000 (23:48 +0100)]
Fortran: fix simplification of array-valued parameter expressions
gcc/fortran/ChangeLog:
PR fortran/102817
* expr.c (simplify_parameter_variable): Copy shape of referenced
subobject when simplifying.
gcc/testsuite/ChangeLog:
PR fortran/102817
* gfortran.dg/pr102817.f90: New test.
Jan Hubicka [Fri, 5 Nov 2021 22:32:55 +0000 (23:32 +0100)]
Fix ice in insert_access
gcc/ChangeLog:
PR ipa/103073
* ipa-modref-tree.h (modref_tree::insert): Do nothing for
paradoxical and zero sized accesses.
gcc/testsuite/ChangeLog:
PR ipa/103073
* g++.dg/torture/pr103073.C: New test.
* gcc.dg/tree-ssa/modref-11.c: New test.
Jan Hubicka [Fri, 5 Nov 2021 22:17:50 +0000 (23:17 +0100)]
Avoid left shift of negative value in ipa-modref-tree.h
gcc/ChangeLog:
PR ipa/103082
* ipa-modref-tree.h (struct modref_access_node): Avoid left shift
of negative value
Harald Anlauf [Fri, 5 Nov 2021 22:13:57 +0000 (23:13 +0100)]
Fortran: a symbol in a COMMON cannot be a coarray
gcc/fortran/ChangeLog:
PR fortran/69419
* match.c (gfc_match_common): Check array spec of a symbol in a
COMMON object list and reject it if it is a coarray.
gcc/testsuite/ChangeLog:
PR fortran/69419
* gfortran.dg/pr69419.f90: New test.
Jonathan Wakely [Fri, 5 Nov 2021 21:42:20 +0000 (21:42 +0000)]
libstdc++: Fix inconsistent noexcept-specific for valarray begin/end
These declarations should be noexcept after I added it to the
definitions in <valarray>.
libstdc++-v3/ChangeLog:
* include/bits/range_access.h (begin(valarray), end(valarray)):
Add noexcept.
Jonathan Wakely [Fri, 5 Nov 2021 21:40:58 +0000 (21:40 +0000)]
libstdc++: Fix pack expansions in tuple_size_v specializations
libstdc++-v3/ChangeLog:
* include/std/tuple (tuple_size_v): Fix pack expansion.
Bernhard Reutner-Fischer [Sun, 31 Oct 2021 16:44:45 +0000 (17:44 +0100)]
Fortran: Missing error with IMPLICIT none (external) [PR100972]
gcc/fortran/ChangeLog:
PR fortran/100972
* decl.c (gfc_match_implicit_none): Fix typo in warning.
* resolve.c (resolve_unknown_f): Reject external procedures
without explicit EXTERNAL attribute whe IMPLICIT none (external)
is in effect.
gcc/testsuite/ChangeLog:
PR fortran/100972
* gfortran.dg/implicit_14.f90: Adjust error.
* gfortran.dg/external_implicit_none_3.f08: New test.
Bernhard Reutner-Fischer [Wed, 27 Oct 2021 21:01:21 +0000 (23:01 +0200)]
Fortran: Delete unused decl in gfortran.h
gcc/fortran/ChangeLog:
* decl.c (gfc_insert_kind_parameter_exprs): Make static.
* expr.c (gfc_build_init_expr): Make static
(gfc_build_default_init_expr): Move below its static helper.
* gfortran.h (gfc_insert_kind_parameter_exprs, gfc_add_saved_common,
gfc_add_common, gfc_use_derived_tree, gfc_free_charlen,
gfc_get_ultimate_derived_super_type,
gfc_resolve_oacc_parallel_loop_blocks, gfc_build_init_expr,
gfc_iso_c_sub_interface): Delete.
* symbol.c (gfc_new_charlen, gfc_get_derived_super_type): Make
static.
Sandra Loosemore [Fri, 5 Nov 2021 21:01:03 +0000 (14:01 -0700)]
Fortran: Add more documentation for mixed-language programming [PR35276]
2021-11-05 Sandra Loosemore <sandra@codesourcery.com>
PR fortran/35276
gcc/fortran/
* gfortran.texi (Mixed-Language Programming): Talk about C++,
and how to link.
Iain Sandoe [Fri, 5 Nov 2021 19:19:53 +0000 (19:19 +0000)]
testsuite, Darwin : Fix tsvc test build on Darwin.
Currently all the tsvc tests fail to build on Darwin because
they assume that <malloc.h> and memalign() are available.
For Darwin, <stdlib.h> is sufficient to obtain the declarations
for malloc and the port has posix_memalign () but not memalign.
Fixed as below.
Signed-off-by: Iain Sandoe <iain@sandoe.co.uk>
gcc/testsuite/ChangeLog:
* gcc.dg/vect/tsvc/tsvc.h: Do not try to include malloc.h
on Darwin also use posix_memalign ().
Iain Sandoe [Wed, 2 Sep 2020 19:33:00 +0000 (20:33 +0100)]
Darwin : Make trampoline templates linker-visible.
For aarch64, the alignment of the LTRAMPn symbols matters.
Actually, the LTRAMPn symbols _are_ 8 byte aligned, but because
they are Local, the linker doesn't know that this guarantee can be met.
It assumes that they are not necessarily more aligned than the
containing section (ld64 atoms strike again).
The fix is to publish the trampoline symbol for the linker to access
directly - it can then see that the atom is suitably aligned.
Fixes issue #11 on the development branch.
Signed-off-by: Iain Sandoe <iain@sandoe.co.uk>
gcc/ChangeLog:
* config/darwin.h (ASM_GENERATE_INTERNAL_LABEL): Add LTRAMP
to the list of symbol prefixes that must be made linker-
visible.
Iain Sandoe [Fri, 28 Aug 2020 18:09:45 +0000 (19:09 +0100)]
Darwin, aarch64 : Ada fixes for hosted tools.
This will allow someone (with an existing Ada compiler on the
platform - which can be provided by the experimental aarch64-darwin
branch) - to build the host tools (gnatmake and friends) for a
non-native cross.
The existing provisions for iOS are OK for cross-compilation from
an x86-64-darwin platform, but we need some adjustments so that these
host tools can be built to run on aarch64-darwin.
Signed-off-by: Iain Sandoe <iain@sandoe.co.uk>
gcc/ada/
* gcc-interface/Make-lang.in: Use iOS signal trampoline code
for hosted Ada tools.
* sigtramp-ios.c: Wrap the declarations in extern "C" when
the code is built by a C++ compiler.
Iain Sandoe [Tue, 18 Aug 2020 21:29:51 +0000 (22:29 +0100)]
Darwin, aarch64 : Initial support for the self-host driver.
At present, there is no special action needed for aarch64-darwin
this just pulls in generic Darwin code.
Signed-off-by: Iain Sandoe <iain@sandoe.co.uk>
gcc/ChangeLog:
* config.host: Add support for aarch64-*-darwin.
* config/aarch64/host-aarch64-darwin.c: New file.
* config/aarch64/x-darwin: New file.
Iain Sandoe [Mon, 1 Nov 2021 16:58:20 +0000 (16:58 +0000)]
Darwin, crts: Fix a build warning.
We have a shim crt for Darwin10 that implements functionality
missing in libSystem. Provide this with a prototype to silence the
warning about this.
libgcc/ChangeLog:
* config/darwin10-unwind-find-enc-func.c: Include libgcc_tm.h.
* config/i386/darwin-lib.h: Declare Darwin10 crt function.
Signed-off-by: Iain Sandoe <iain@sandoe.co.uk>
Jonathan Wakely [Fri, 5 Nov 2021 12:18:52 +0000 (12:18 +0000)]
libstdc++: Add [[unlikely]] attributes to std::random_device routines
libstdc++-v3/ChangeLog:
* src/c++11/random.cc (__x86_rdrand, __x86_rdseed): Add
[[unlikely]] attribute.
Jonathan Wakely [Wed, 20 Oct 2021 08:25:24 +0000 (09:25 +0100)]
libstdc++: Add support for POWER9 DARN instruction to std::random_device
The ISA-3.0 instruction set includes DARN ("deliver a random number")
which can be used similarly to the existing support for RDRAND and RDSEED.
libstdc++-v3/ChangeLog:
* src/c++11/random.cc [__powerpc__] (USE_DARN): Define.
(__ppc_darn): New function to use POWER9 DARN instruction.
(Which): Add 'darn' enumerator.
(which_source): Check for __ppc_darn.
(random_device::_M_init): Support "darn" and "hw" tokens.
(random_device::_M_getentropy): Add darn to switch.
* testsuite/26_numerics/random/random_device/cons/token.cc:
Check "darn" token.
* testsuite/26_numerics/random/random_device/entropy.cc:
Likewise.