The AIX power alignment rules apply the natural alignment of the
"first member" if it is of a floating-point data type (or is an aggregate
whose recursively "first" member or element is such a type). The alignment
associated with these types for subsequent members use an alignment value
where the floating-point data type is considered to have 4-byte alignment.
GCC had been stripping array type but had not recursively looked
within structs and unions. This also applies to classes and
subclasses and, therefore, becomes more prominent with C++.
For example,
struct A {
double x[2];
int y;
};
struct B {
int i;
struct A a;
};
struct A has double-word alignment for the bare type, but
word alignment and offset within struct B despite the alignment of
struct A. If struct A were the first member of struct B, struct B
would have double-word alignment. One must search for the innermost
first member to increase the alignment if double and then search for
the innermost first member to reduce the alignment if the TYPE had
double-word alignment solely because the innermost first member was
double.
This patch recursively looks through the first member to apply the
double-word alignment to the struct / union as a whole and to apply
the word alignment to the struct or union as a member within a struct
or union.
This is an ABI change for GCC on AIX, but GCC on AIX had not correctly
implemented the AIX ABI and had not been compatible with the IBM XL
compiler.
Bootstrapped on powerpc-ibm-aix7.2.3.0.
gcc/ChangeLog:
* config/rs6000/aix.h (ADJUST_FIELD_ALIGN): Call function.
* config/rs6000/rs6000-protos.h (rs6000_special_adjust_field_align):
Declare.
* config/rs6000/rs6000.c (rs6000_special_adjust_field_align): New.
(rs6000_special_round_type_align): Recursively check innermost first
field.
gcc/testsuite/ChangeLog:
* gcc.target/powerpc/pr99557.c: New.
On the testcase in the PR with
-fno-tree-sink -O3 -fPIC -fomit-frame-pointer -fno-strict-aliasing -mstackrealign
we have prologue:
0000000000000000 <_func_with_dwarf_issue_>:
0: 4c 8d 54 24 08 lea 0x8(%rsp),%r10
5: 48 83 e4 f0 and $0xfffffffffffffff0,%rsp
9: 41 ff 72 f8 pushq -0x8(%r10)
d: 55 push %rbp
e: 48 89 e5 mov %rsp,%rbp
11: 41 57 push %r15
13: 41 56 push %r14
15: 41 55 push %r13
17: 41 54 push %r12
19: 41 52 push %r10
1b: 53 push %rbx
1c: 48 83 ec 20 sub $0x20,%rsp
and emit
00000000 0000000000000014 00000000 CIE
Version: 1
Augmentation: "zR"
Code alignment factor: 1
Data alignment factor: -8
Return address column: 16
Augmentation data: 1b
DW_CFA_def_cfa: r7 (rsp) ofs 8
DW_CFA_offset: r16 (rip) at cfa-8
DW_CFA_nop
DW_CFA_nop
00000018 0000000000000044 0000001c FDE cie=00000000 pc=0000000000000000..00000000000001d5
DW_CFA_advance_loc: 5 to 0000000000000005
DW_CFA_def_cfa: r10 (r10) ofs 0
DW_CFA_advance_loc: 9 to 000000000000000e
DW_CFA_expression: r6 (rbp) (DW_OP_breg6 (rbp): 0)
DW_CFA_advance_loc: 13 to 000000000000001b
DW_CFA_def_cfa_expression (DW_OP_breg6 (rbp): -40; DW_OP_deref)
DW_CFA_expression: r15 (r15) (DW_OP_breg6 (rbp): -8)
DW_CFA_expression: r14 (r14) (DW_OP_breg6 (rbp): -16)
DW_CFA_expression: r13 (r13) (DW_OP_breg6 (rbp): -24)
DW_CFA_expression: r12 (r12) (DW_OP_breg6 (rbp): -32)
...
unwind info for that. The problem is when async signal
(or stepping through in the debugger) stops after the pushq %rbp
instruction and before movq %rsp, %rbp, the unwind info says that
caller's %rbp is saved there at *%rbp, but that is not true, caller's
%rbp is either still available in the %rbp register, or in *%rsp,
only after executing the next instruction - movq %rsp, %rbp - the
location for %rbp is correct. So, either we'd need to temporarily
say:
DW_CFA_advance_loc: 9 to 000000000000000e
DW_CFA_expression: r6 (rbp) (DW_OP_breg7 (rsp): 0)
DW_CFA_advance_loc: 3 to 0000000000000011
DW_CFA_expression: r6 (rbp) (DW_OP_breg6 (rbp): 0)
DW_CFA_advance_loc: 10 to 000000000000001b
or to me it seems more compact to just say:
DW_CFA_advance_loc: 12 to 0000000000000011
DW_CFA_expression: r6 (rbp) (DW_OP_breg6 (rbp): 0)
DW_CFA_advance_loc: 10 to 000000000000001b
I've tried instead to deal with it through REG_FRAME_RELATED_EXPR
from the backend, but that failed miserably as explained in the PR,
dwarf2cfi.c has some rules (Rule 16 to Rule 19) that are specific to the
dynamic stack realignment using drap register that only the i386 backend
does right now, and by using REG_FRAME_RELATED_EXPR or REG_CFA* notes we
can't emulate those rules. The following patch instead does the deferring
of the hard frame pointer save rule in dwarf2cfi.c Rule 18 handling and
emits it on the (set hfp sp) assignment that must appear shortly after it
and adds assertion that it is the case.
The difference before/after the patch on the assembly is:
--- pr99334.s~ 2021-03-26 15:42:40.881749380 +0100
+++ pr99334.s 2021-03-26 17:38:05.729161910 +0100
@@ -11,8 +11,8 @@ _func_with_dwarf_issue_:
andq $-16, %rsp
pushq -8(%r10)
pushq %rbp
- .cfi_escape 0x10,0x6,0x2,0x76,0
movq %rsp, %rbp
+ .cfi_escape 0x10,0x6,0x2,0x76,0
pushq %r15
pushq %r14
pushq %r13
i.e. does just what we IMHO need, after pushq %rbp %rbp
still contains parent's frame value and so the save rule doesn't
need to be overridden there, ditto at the start of the next insn
before the side-effect took effect, and we override it only after
it when %rbp already has the right value.
If some other target adds dynamic stack realignment in the future and
the offset 0 case wouldn't be true there, the code can be adjusted so that
it works on all the drap architectures, I'm pretty sure the code would
need other adjustments too.
For the rule 18 and for the (set hfp sp) after it we already have asserts
for the drap cases that check whether the code looks the way i?86/x86_64
emit it currently.
2021-03-26 Jakub Jelinek <jakub@redhat.com>
PR debug/99334
* dwarf2out.h (struct dw_fde_node): Add rule18 member.
* dwarf2cfi.c (dwarf2out_frame_debug_expr): When handling (set hfp sp)
assignment with drap_reg active, queue reg save for hfp with offset 0
and flush queued reg saves. When handling a push with rule18,
defer queueing reg save for hfp and just assert the offset is 0.
(scan_trace): Assert that fde->rule18 is false.
NSDMIs are a C++11 thing, and here we ICE with them on the non-C++11
path. Fortunately all we need is a small tweak to my recent r11-7835
patch.
gcc/cp/ChangeLog:
PR c++/98352
* method.c (implicitly_declare_fn): Pass &raises to
synthesized_method_walk.
gcc/testsuite/ChangeLog:
PR c++/98352
* g++.dg/cpp0x/inh-ctor37.C: Remove dg-error.
* g++.dg/cpp0x/nsdmi17.C: New test.
This makes std::random_device usable on VxWorks when running on older
x86 hardware. Since the r10-728 fix for PR libstdc++/85494 the library
will use the new code unconditionally on x86, but the cpuid checks for
RDSEED and RDRAND can fail at runtime, depending on the hardware where
the code is executing. If the OS does not provide /dev/urandom then this
means the std::random_device constructor always fails. In previous
releases if /dev/urandom is unavailable then std::mt19937 was used
unconditionally.
This patch adds a fallback for the case where the runtime cpuid checks
for x86 hardware instructions fail, and no /dev/urandom is available.
When this happens a std::linear_congruential_engine object will be used,
with a seed based on hashing the engine's address and the current time.
Distinct std::random_device objects will use different seeds, unless an
object is created and destroyed and a new object created at the same
memory location within the clock tick. This is not great, but is better
than always throwing from the constructor, and better than always using
std::mt19937 with the same seed (as GCC 9 and earlier do).
libstdc++-v3/ChangeLog:
* src/c++11/random.cc (USE_LCG): Define when a pseudo-random
fallback is needed.
[USE_LCG] (bad_seed, construct_lcg_at, destroy_lcg_at, __lcg):
New helper functions and callback.
(random_device::_M_init): Add 'prng' and 'all' enumerators.
Replace switch with fallthrough with a series of 'if' statements.
[USE_LCG]: Construct an lcg_type engine and use __lcg when cpuid
checks fail.
(random_device::_M_init_pretr1) [USE_MT19937]: Accept "prng"
token.
(random_device::_M_getval): Check for callback unconditionally
and always pass _M_file pointer.
* testsuite/26_numerics/random/random_device/85494.cc: Remove
effective-target check. Use new random_device_available helper.
* testsuite/26_numerics/random/random_device/94087.cc: Likewise.
* testsuite/26_numerics/random/random_device/cons/default-cow.cc:
Remove effective-target check.
* testsuite/26_numerics/random/random_device/cons/default.cc:
Likewise.
* testsuite/26_numerics/random/random_device/cons/token.cc: Use
new random_device_available helper. Test "prng" token.
* testsuite/util/testsuite_random.h (random_device_available):
New helper function.
During development of modules, I had difficulty deciding whether the
module flags of a template should live on the decl_template_result,
the template_decl, or both. I chose the latter, and require them to
be consistent. This and a few other defects show how hard that
consistency is. Hence this patch move to holding the flags on the
template-decl-result decl. That's the entity various bits of the
parser have at the appropriate time. Once needs STRIP_TEMPLATE in a
bunch of places, which this patch adds. Also a check that we never
give a TEMPLATE_DECL to the module flag accessors.
This left a problem with how I was handling template aliases. These
were in two parts -- separating the TEMPLATE_DECL from the TYPE_DECL.
That seemed somewhat funky, but development showed it necessary. Of
course, that causes problems if the TEMPLATE_DECL cannot contain 'am
imported' information. Investigating now shows that we do not need to
treat them separately. By reverting a bit of template instantiation
machinery that caused the problem, we're back on course. I think what
has happened is that between then and now, other typedef fixes have
corrected the underlying problem this separation was working around.
It allows a bunch of cleanup in the decl streamer, as we no longer
have to handle a null TEMPLATE_DECL_RESULT.
PR c++/99283
gcc/cp/
* cp-tree.h (DECL_MODULE_CHECK): Ban TEMPLATE_DECL.
(SET_TYPE_TEMPLATE_INFO): Restore Alias template setting.
* decl.c (duplicate_decls): Remove template_decl module flag
propagation.
* module.cc (merge_kind_name): Add alias tmpl spec as a thing.
(dumper::impl::nested_name): Adjust for template-decl module flag
change.
(trees_in::assert_definition): Likewise.
(trees_in::install_entity): Likewise.
(trees_out::decl_value): Likewise. Remove alias template
separation of template and type_decl.
(trees_in::decl_value): Likewise.
(trees_out::key_mergeable): Likewise,
(trees_in::key_mergeable): Likewise.
(trees_out::decl_node): Adjust for template-decl module flag
change.
(depset:#️⃣:make_dependency): Likewise.
(get_originating_module, module_may_redeclare): Likewise.
(set_instantiating_module, set_defining_module): Likewise.
* name-lookup.c (name_lookup::search_adl): Likewise.
(do_pushdecl): Likewise.
* pt.c (build_template_decl): Likewise.
(lookup_template_class_1): Remove special alias_template handling
of DECL_TI_TEMPLATE.
(tsubst_template_decl): Likewise.
gcc/testsuite/
* g++.dg/modules/pr99283-2_a.H: New.
* g++.dg/modules/pr99283-2_b.H: New.
* g++.dg/modules/pr99283-2_c.H: New.
* g++.dg/modules/pr99283-3_a.H: New.
* g++.dg/modules/pr99283-3_b.H: New.
* g++.dg/modules/pr99283-4.H: New.
* g++.dg/modules/tpl-alias-1_a.H: Adjust scans.
* g++.dg/modules/tpl-alias-1_b.C: Adjust scans.
Relaxed memory should be considered more like memory then special memory.
gcc/ChangeLog:
PR target/99766
* ira-costs.c (record_reg_classes): Put case with
CT_RELAXED_MEMORY adjacent to one with CT_MEMORY.
* ira.c (ira_setup_alts): Ditto.
* lra-constraints.c (process_alt_operands): Ditto.
* recog.c (asm_operand_ok): Ditto.
* reload.c (find_reloads): Ditto.
gcc/testsuite/ChangeLog:
PR target/99766
* g++.target/aarch64/sve/pr99766.C: New.
Most postincrements are cheap on Neoverse V1, but it's
generally better to avoid them on LD[34] and ST[34] instructions.
This patch adds separate address costs fields for these cases.
Other CPUs continue to use the same costs for all postincrements.
gcc/
* config/aarch64/aarch64-protos.h
(cpu_addrcost_table::post_modify_ld3_st3): New member variable.
(cpu_addrcost_table::post_modify_ld4_st4): Likewise.
* config/aarch64/aarch64.c (generic_addrcost_table): Update
accordingly, using the same costs as for post_modify.
(exynosm1_addrcost_table, xgene1_addrcost_table): Likewise.
(thunderx2t99_addrcost_table, thunderx3t110_addrcost_table):
(tsv110_addrcost_table, qdf24xx_addrcost_table): Likewise.
(a64fx_addrcost_table): Likewise.
(neoversev1_addrcost_table): New.
(neoversev1_tunings): Use neoversev1_addrcost_table.
(aarch64_address_cost): Use the new post_modify costs for CImode
and XImode.
When SVE is enabled, GCC needs to do a three-way comparison
between scalar, Advanced SIMD and SVE code. The normal costs
tend to be latency-based, which is well-suited to SLP. However,
comparing sums of latency costs means that we effectively treat
the code as executing sequentially. This can hide the effect of
pipeline bubbles or resource contention that in practice are quite
important for loop vectorisation. This is particularly true for
loops that involve reductions.
This patch therefore tries to estimate how quickly each piece
of code could issue, using a very (very) simplistic model.
It then uses this to adjust the loop vector costs up or down as
appropriate. Part of the Advanced SIMD vs. SVE adjustment is
opt-in and is not enabled by default even for use_new_vector_costs.
Like with the previous patches, this one only becomes active if
a CPU selects use_new_vector_costs. It should therefore have
a very low impact on other CPUs. The code also mostly ignores
CPUs that have no issue information, even if use_new_vector_costs
is enabled for some reason.
gcc/
* config/aarch64/aarch64.opt
(-param=aarch64-loop-vect-issue-rate-niters=): New parameter.
* doc/invoke.texi: Document it.
* config/aarch64/aarch64-protos.h (aarch64_base_vec_issue_info)
(aarch64_scalar_vec_issue_info, aarch64_simd_vec_issue_info)
(aarch64_advsimd_vec_issue_info, aarch64_sve_vec_issue_info)
(aarch64_vec_issue_info): New structures.
(cpu_vector_cost): Write comments above the variables rather
than to the side.
(cpu_vector_cost::issue_info): New member variable.
* config/aarch64/aarch64.c: Include gimple-pretty-print.h
and tree-ssa-loop-niter.h.
(generic_vector_cost, a64fx_vector_cost, qdf24xx_vector_cost)
(thunderx_vector_cost, tsv110_vector_cost, cortexa57_vector_cost)
(exynosm1_vector_cost, xgene1_vector_cost, thunderx2t99_vector_cost)
(thunderx3t110_vector_cost): Initialize issue_info to null.
(neoversev1_scalar_issue_info, neoversev1_advsimd_issue_info)
(neoversev1_sve_issue_info, neoversev1_vec_issue_info): New structures.
(neoversev1_vector_cost): Use them.
(aarch64_vec_op_count, aarch64_sve_op_count): New structures.
(aarch64_vector_costs::saw_sve_only_op): New member variable.
(aarch64_vector_costs::num_vector_iterations): Likewise.
(aarch64_vector_costs::scalar_ops): Likewise.
(aarch64_vector_costs::advsimd_ops): Likewise.
(aarch64_vector_costs::sve_ops): Likewise.
(aarch64_vector_costs::seen_loads): Likewise.
(aarch64_simd_vec_costs_for_flags): New function.
(aarch64_analyze_loop_vinfo): Initialize num_vector_iterations.
Count the number of predicate operations required by SVE WHILE
instructions.
(aarch64_comparison_type, aarch64_multiply_add_p): New functions.
(aarch64_sve_only_stmt_p, aarch64_in_loop_reduction_latency): Likewise.
(aarch64_count_ops): Likewise.
(aarch64_add_stmt_cost): Record whether see an SVE operation
that cannot currently be implementing using Advanced SIMD.
Record issue information about the scalar, Advanced SIMD
and (where relevant) SVE versions of a loop.
(aarch64_vec_op_count::dump): New function.
(aarch64_sve_op_count::dump): Likewise.
(aarch64_estimate_min_cycles_per_iter): Likewise.
(aarch64_adjust_body_cost): If issue information is available,
try to compare the issue rates of the various loop implementations
and increase or decrease the vector body cost accordingly.
In practice it seems to be better not to cost a vector induction.
The scalar code generally needs the same induction but doesn't
cost it, making an apples-for-apples comparison harder. Most
inductions also have a low latency and their cost usually gets
hidden by other operations.
Like with the previous patches, this one only becomes active if
a CPU selects use_new_vector_costs. It should therefore have
a very low impact on other CPUs.
gcc/
* config/aarch64/aarch64.c (aarch64_detect_vector_stmt_subtype):
Assume a zero cost for induction phis.
So far the costing of COND_EXPRs hasn't distinguished between
cases in which the condition is calculated separately or is
built into the COND_EXPR itself. This patch adds the cost
of any embedded comparison.
Like with the previous patches, this one only becomes active if
a CPU selects use_new_vector_costs. It should therefore have
a very low impact on other CPUs.
gcc/
* config/aarch64/aarch64.c (aarch64_embedded_comparison_type): New
function.
(aarch64_adjust_stmt_cost): Add the costs of embedded scalar and
vector comparisons.
If the scalar code does an integer load followed by an integer
extension, we've tended to cost that as two separate operations,
even though the extension is probably going to be free in practice.
This patch treats the extension as having zero cost, like we already
do for extending SVE loads.
Like with previous patches, this one only becomes active if
a CPU selects use_new_vector_costs. It should therefore have
a very low impact on other CPUs.
gcc/
* config/aarch64/aarch64.c (aarch64_detect_scalar_stmt_subtype):
New function.
(aarch64_add_stmt_cost): Call it.
GCC usually costs the SVE and Advanced SIMD versions of a loop
and picks the one with the lowest cost. By default it will choose
SVE over Advanced SIMD in the event of tie.
This is normally the correct behaviour, not least because SVE can
handle every scalar iteration count whereas Advanced SIMD can only
handle full vectors. However, there is one important exception
that GCC failed to consider: we can completely unroll Advanced SIMD
code at compile time, but we can't do the same for SVE.
This patch therefore adds an opt-in heuristic to guess whether
the Advanced SIMD version of a loop is likely to be unrolled.
This will only be suitable for some CPUs, so it is not enabled
by default and is controlled separately from use_new_vector_costs.
Like with previous patches, this one only becomes active if a
CPU selects both of the new tuning parameters. It should therefore
have a very low impact on other CPUs.
gcc/
* config/aarch64/aarch64-tuning-flags.def (matched_vector_throughput):
New tuning parameter.
* config/aarch64/aarch64.c (neoversev1_tunings): Use it.
(aarch64_estimated_sve_vq): New function.
(aarch64_vector_costs::analyzed_vinfo): New member variable.
(aarch64_vector_costs::is_loop): Likewise.
(aarch64_vector_costs::unrolled_advsimd_niters): Likewise.
(aarch64_vector_costs::unrolled_advsimd_stmts): Likewise.
(aarch64_record_potential_advsimd_unrolling): New function.
(aarch64_analyze_loop_vinfo, aarch64_analyze_bb_vinfo): Likewise.
(aarch64_add_stmt_cost): Call aarch64_analyze_loop_vinfo or
aarch64_analyze_bb_vinfo on the first use of a costs structure.
Detect whether we're vectorizing a loop for SVE that might be
completely unrolled if it used Advanced SIMD instead.
(aarch64_adjust_body_cost_for_latency): New function.
(aarch64_finish_cost): Call it.
This patch makes the AArch64 vector code use its own vector
costs structure, rather than just using the default unsigned[3].
Unfortunately, it's not easy to make this change specific to
use_new_vector_costs, so this part is one that affects all CPUs.
The change is relatively mechanical though.
gcc/
* config/aarch64/aarch64.c (aarch64_vector_costs): New structure.
(aarch64_init_cost): New function.
(aarch64_add_stmt_cost): Use aarch64_vector_costs instead of
the default unsigned[3].
(aarch64_finish_cost, aarch64_destroy_cost_data): New functions.
(TARGET_VECTORIZE_INIT_COST): Override.
(TARGET_VECTORIZE_FINISH_COST): Likewise.
(TARGET_VECTORIZE_DESTROY_COST_DATA): Likewise.
This patch adds dedicated vector costs for Neoverse V1.
Previously we just used the Cortex-A57 costs, which isn't
ideal given that Cortex-A57 doesn't support SVE.
gcc/
* config/aarch64/aarch64.c (neoversev1_advsimd_vector_cost)
(neoversev1_sve_vector_cost): New cost structures.
(neoversev1_vector_cost): Likewise.
(neoversev1_tunings): Use them. Enable use_new_vector_costs.
Currently each element in a gather load is costed as a scalar_load
and each element in a scatter store is costed as a scalar_store.
The load side seems to work pretty well in practice, since many
CPU-specific costs give loads quite a high cost relative to
arithmetic operations. However, stores usually have a cost
of just 1, which means that scatters tend to appear too cheap.
This patch adds a separate cost for one element in a scatter store.
Like with the previous patches, this one only becomes active if
a CPU selects use_new_vector_costs. It should therefore have
a very low impact on other CPUs.
gcc/
* config/aarch64/aarch64-protos.h
(sve_vec_cost::scatter_store_elt_cost): New member variable.
* config/aarch64/aarch64.c (generic_sve_vector_cost): Update
accordingly, taking the cost from the cost of a scalar_store.
(a64fx_sve_vector_cost): Likewise.
(aarch64_detect_vector_stmt_subtype): Detect scatter stores.
Storing one element of a vector is costed as a vec_to_scalar
followed by a scalar_store. However, vec_to_scalar is also
used for reductions and for vector-to-GPR moves, which makes
it difficult to pick one cost for them all.
This patch therefore adds a cost for extracting one element
of a vector in preparation for storing it out. The store
itself is still costed separately.
Like with the previous patches, this one only becomes active if
a CPU selects use_new_vector_costs. It should therefore have
a very low impact on other CPUs.
gcc/
* config/aarch64/aarch64-protos.h
(simd_vec_cost::store_elt_extra_cost): New member variable.
* config/aarch64/aarch64.c (generic_advsimd_vector_cost): Update
accordingly, using the vec_to_scalar cost for the new field.
(generic_sve_vector_cost, a64fx_advsimd_vector_cost): Likewise.
(a64fx_sve_vector_cost, qdf24xx_advsimd_vector_cost): Likewise.
(thunderx_advsimd_vector_cost, tsv110_advsimd_vector_cost): Likewise.
(cortexa57_advsimd_vector_cost, exynosm1_advsimd_vector_cost)
(xgene1_advsimd_vector_cost, thunderx2t99_advsimd_vector_cost)
(thunderx3t110_advsimd_vector_cost): Likewise.
(aarch64_detect_vector_stmt_subtype): Detect single-element stores.
At the moment, we cost LD[234] and ST[234] as N vector loads
or stores, which effectively treats the implied permute as free.
This patch adds additional costs for the permutes, which apply on
top of the costs for the loads and stores.
Like with the previous patches, this one only becomes active if
a CPU selects use_new_vector_costs. It should therefore have
a very low impact on other CPUs.
gcc/
* config/aarch64/aarch64-protos.h (simd_vec_cost::ld2_st2_permute_cost)
(simd_vec_cost::ld3_st3_permute_cost): New member variables.
(simd_vec_cost::ld4_st4_permute_cost): Likewise.
* config/aarch64/aarch64.c (generic_advsimd_vector_cost): Update
accordingly, using zero for the new costs.
(generic_sve_vector_cost, a64fx_advsimd_vector_cost): Likewise.
(a64fx_sve_vector_cost, qdf24xx_advsimd_vector_cost): Likewise.
(thunderx_advsimd_vector_cost, tsv110_advsimd_vector_cost): Likewise.
(cortexa57_advsimd_vector_cost, exynosm1_advsimd_vector_cost)
(xgene1_advsimd_vector_cost, thunderx2t99_advsimd_vector_cost)
(thunderx3t110_advsimd_vector_cost): Likewise.
(aarch64_ld234_st234_vectors): New function.
(aarch64_adjust_stmt_cost): Likewise.
(aarch64_add_stmt_cost): Call aarch64_adjust_stmt_cost if using
the new vector costs.
Following on from the previous reduction costs patch, this one
adds costs for the SVE CLAST[AB] and FADDA instructions.
These instructions occur within the loop body, whereas the
reductions handled by the previous patch occur outside.
Like with the previous patch, this one only becomes active if
a CPU selects use_new_vector_costs. It should therefore have
a very low impact on other CPUs.
gcc/
* config/aarch64/aarch64-protos.h (sve_vec_cost): Turn into a
derived class of simd_vec_cost. Add information about CLAST[AB]
and FADDA instructions.
* config/aarch64/aarch64.c (generic_sve_vector_cost): Update
accordingly, using the vec_to_scalar costs for the new fields.
(a64fx_sve_vector_cost): Likewise.
(aarch64_reduc_type): New function.
(aarch64_sve_in_loop_reduction_latency): Likewise.
(aarch64_detect_vector_stmt_subtype): Take a vinfo parameter.
Use aarch64_sve_in_loop_reduction_latency to handle SVE reductions
that occur in the loop body.
(aarch64_add_stmt_cost): Update call accordingly.
This patch is part of a series that makes opt-in tweaks to the
AArch64 vector cost model.
At the moment, all reductions are costed as vec_to_scalar, which
also includes things like extracting a single element from a vector.
This is a bit too coarse in practice, since the cost of a reduction
depends very much on the type of value that it's processing.
This patch therefore adds separate costs for each case. To start with,
all the new costs are copied from the associated vec_to_scalar ones.
Due the extreme lateness of this patch in the GCC 11 cycle, I've added
a new tuning flag (use_new_vector_costs) that selects the new behaviour.
This should help to ensure that the risk of the new code is only borne
by the CPUs that need it. Generic tuning is not affected.
gcc/
* config/aarch64/aarch64-tuning-flags.def (use_new_vector_costs):
New tuning flag.
* config/aarch64/aarch64-protos.h (simd_vec_cost): Put comments
above the fields rather than to the right.
(simd_vec_cost::reduc_i8_cost): New member variable.
(simd_vec_cost::reduc_i16_cost): Likewise.
(simd_vec_cost::reduc_i32_cost): Likewise.
(simd_vec_cost::reduc_i64_cost): Likewise.
(simd_vec_cost::reduc_f16_cost): Likewise.
(simd_vec_cost::reduc_f32_cost): Likewise.
(simd_vec_cost::reduc_f64_cost): Likewise.
* config/aarch64/aarch64.c (generic_advsimd_vector_cost): Update
accordingly, using the vec_to_scalar_cost for the new fields.
(generic_sve_vector_cost, a64fx_advsimd_vector_cost): Likewise.
(a64fx_sve_vector_cost, qdf24xx_advsimd_vector_cost): Likewise.
(thunderx_advsimd_vector_cost, tsv110_advsimd_vector_cost): Likewise.
(cortexa57_advsimd_vector_cost, exynosm1_advsimd_vector_cost)
(xgene1_advsimd_vector_cost, thunderx2t99_advsimd_vector_cost)
(thunderx3t110_advsimd_vector_cost): Likewise.
(aarch64_use_new_vector_costs_p): New function.
(aarch64_simd_vec_costs): New function, split out from...
(aarch64_builtin_vectorization_cost): ...here.
(aarch64_is_reduction): New function.
(aarch64_detect_vector_stmt_subtype): Likewise.
(aarch64_add_stmt_cost): Call aarch64_detect_vector_stmt_subtype if
using the new vector costs.
The libgdruntime_convenience library was built with `-fversion=Shared',
but the libphobos part wasn't when creating the static library.
As there are no issues compiling in Shared code into the static library,
to avoid mismatches the flag is now always present when --enable-shared
is turned on. Libtool's compiler PIC D flag is now the combination of
compiler PIC and D Shared flags, and AM_DFLAGS passes `-prefer-pic' to
libtool unless --enable-shared is turned off.
libphobos/ChangeLog:
* Makefile.in: Regenerate.
* configure: Regenerate.
* configure.ac: Substitute enable_shared, enable_static, and
phobos_lt_pic_flag.
* libdruntime/Makefile.am (AM_DFLAGS): Replace
phobos_compiler_pic_flag with phobos_lt_pic_flags, and
phobos_compiler_shared_flag.
* libdruntime/Makefile.in: Regenerate.
* src/Makefile.am (AM_DFLAGS): Replace phobos_compiler_pic_flag
with phobos_lt_pic_flag, and phobos_compiler_shared_flag.
* src/Makefile.in: Regenerate.
* testsuite/Makefile.in: Regenerate.
* testsuite/libphobos.druntime_shared/druntime_shared.exp: Remove
-fversion=Shared and -fno-moduleinfo from default extra test flags.
* testsuite/libphobos.phobos_shared/phobos_shared.exp: Likewise.
* testsuite/testsuite_flags.in: Add phobos_compiler_shared_flag to
--gdcflags.
In get_emutls_init_templ_addr, only thread-local declarations that were
DECL_ONE_ONLY would have a public initializer symbol, ignoring variables
that were declared with __attribute__((weak)).
gcc/ChangeLog:
PR ipa/99466
* tree-emutls.c (get_emutls_init_templ_addr): Mark initializer of weak
TLS declarations as public.
gcc/testsuite/ChangeLog:
PR ipa/99466
* gcc.dg/tls/pr99466-1.c: New test.
* gcc.dg/tls/pr99466-2.c: New test.
All target platforms that could run on SPARC should include this header
in order to avoid errors from memmodel being used in sparc-protos.h.
gcc/ChangeLog:
* config/freebsd-d.c: Include memmodel.h.
When adding P0784R7 constexpr new support, we still didn't have
P1331R2 implemented and so I had to change also build_vec_delete_1
- instead of having uninitialized tbase temporary later initialized
by MODIFY_EXPR I've set the DECL_INITIAL for it - because otherwise
it would be rejected during constexpr evaluation which didn't like
uninitialized vars. Unfortunately, that change broke the following
testcase.
The problem is that these temporaries (not just tbase but tbase was
the only one with an initializer) are created during NSDMI parsing
and current_function_decl is NULL at that point. Later when we
clone body of constructors, auto_var_in_fn_p is false for those
(as they have NULL DECL_CONTEXT) and so they aren't duplicated,
and what is worse, the DECL_INITIAL isn't duplicated either nor processed,
and during expansion we ICE because the code from DECL_INITIAL of that
var refers to the abstract constructor's PARM_DECL (this) rather than
the actual constructor's one.
So, either we can just revert those build_vec_delete_1 changes (as done
in the second patch - in attachment), or, as the first patch does, we can
copy the temporaries during bot_manip like we copy the temporaries of
TARGET_EXPRs. To me that looks like a better fix because e.g. if
break_out_of_target_exprs is called for the same NSDMI multiple times,
sharing the temporaries looks just wrong to me. If the temporaries
are declared as BIND_EXPR_VARS of some BIND_EXPR (which is the case
of the tbase variable built by build_vec_delete_1 and is the only way
how the DECL_INITIAL can be walked by *walk_tree*), then we need to
copy it also in the BIND_EXPR BIND_EXPR_VARS chain, other temporaries
(those that don't need DECL_INITIAL) often have just DECL_EXPR and no
corresponding BIND_EXPR.
Note, ({ }) are rejected in nsdmis, so all we run into are temporaries
the FE creates artificially.
2021-03-26 Jakub Jelinek <jakub@redhat.com>
PR c++/99705
* tree.c (bot_manip): Remap artificial automatic temporaries mentioned
in DECL_EXPR or in BIND_EXPR_VARS.
* g++.dg/cpp0x/new5.C: New test.
The srcloc19.C testcase shows inconsistency in
std::source_location::current() locations between calls from
templates and non-templates. The location used by __builtin_source_location
comes in both cases from input_location which is set on it by bot_manip
when handling the default argument, called during finish_call_expr.
The problem is that in templates that input_location comes from the
CALL_EXPR we built earlier and that has the combined locus with
range between first character of the function name and closing paren
with caret on the opening paren, so something printed as caret as:
foobar ();
~~~~~~^~
But outside of templates, finish_call_expr is called when input_location
is just the closing paren token, i.e.
foobar ();
^
and only after that returns we create the combined location and set
the CALL_EXPR location to that. So, it means std::source_location::current()
reports in templates the column of opening (, while outside of templates
closing ).
The following patch makes it consistent by creating the combined location
already before calling finish_call_expr and temporarily overriding
input_location to that.
2021-03-25 Jakub Jelinek <jakub@redhat.com>
PR c++/99672
* parser.c (cp_parser_postfix_expression): For calls, create
combined_loc and temporarily set input_location to it before
calling finish_call_expr.
* g++.dg/concepts/diagnostic2.C: Adjust expected caret line.
* g++.dg/cpp1y/builtin_location.C (f4, n6): Move #line directives
to match locus changes.
* g++.dg/cpp2a/srcloc1.C: Adjust expected column numbers.
* g++.dg/cpp2a/srcloc2.C: Likewise.
* g++.dg/cpp2a/srcloc15.C: Likewise.
* g++.dg/cpp2a/srcloc16.C: Likewise.
* g++.dg/cpp2a/srcloc19.C: New test.
* g++.dg/modules/adhoc-1_b.C: Adjust expected column numbers
and caret line.
* g++.dg/modules/macloc-1_c.C: Adjust expected column numbers.
* g++.dg/modules/macloc-1_d.C: Likewise.
* g++.dg/plugin/diagnostic-test-expressions-1.C: Adjust expected
caret line.
* testsuite/18_support/source_location/consteval.cc (main): Adjust
expected column numbers.
* testsuite/18_support/source_location/1.cc (main): Likewise.
This is an ICE on invalid where we crash because since r269032 we
keep error_mark_node around instead of using noexcept_false_spec
when things go wrong; see the walk_field_subobs hunk.
We crash in deduce_inheriting_ctor which calls synthesized_method_walk
to deduce the exception-specification, but fails to do so in this case,
because the testcase is invalid so get_nsdmi returns error_mark_node for
the member 'c', and per r269032 the error_mark_node propagates back to
deduce_inheriting_ctor which subsequently calls build_exception_variant
whereon we crash. I think we should return early if the deduction fails
and I decided to call mark_used to get an error right away instead of
hoping that it would get called later. My worry is that we could forget
that there was an error and think that we just deduced noexcept(false).
And then I noticed that the test still crashes in C++98. Here again we
failed to deduce the exception-specification in implicitly_declare_fn,
but nothing reported an error between synthesized_method_walk and the
assert. Well, not much we can do except calling synthesized_method_walk
again, this time in the verbose mode and making sure that we did get an
error.
gcc/cp/ChangeLog:
PR c++/94751
* call.c (build_over_call): Maybe call mark_used in case
deduce_inheriting_ctor fails and return error_mark_node.
* cp-tree.h (deduce_inheriting_ctor): Adjust declaration.
* method.c (deduce_inheriting_ctor): Return bool if the deduction
fails.
(implicitly_declare_fn): If raises is error_mark_node, call
synthesized_method_walk with diag being true.
gcc/testsuite/ChangeLog:
PR c++/94751
* g++.dg/cpp0x/inh-ctor37.C: New test.
The following invalid tests ICE because we don't diagnose (and drop) bare
parameter packs in bitfield widths.
2021-03-25 Jakub Jelinek <jakub@redhat.com>
PR c++/99745
* decl2.c (grokbitfield): Diagnose bitfields containing bare parameter
packs and don't set DECL_BIT_FIELD_REPRESENTATIVE in that case.
* g++.dg/cpp0x/variadic181.C: New test.
This PR complains that we issue a -Wconversion warning in
template <int N> struct X {};
template <class T> X<sizeof(T)> foo();
saying "conversion from 'long unsigned int' to 'int' may change value".
While it's not technically wrong, I suspect -Wconversion warnings aren't
all that useful for value-dependent expressions. So this patch disables
them. This is a regression that started with r241425:
@@ -7278,7 +7306,7 @@ convert_template_argument (tree parm,
val = error_mark_node;
}
}
- else if (!dependent_template_arg_p (orig_arg)
+ else if (!type_dependent_expression_p (orig_arg)
&& !uses_template_parms (t))
/* We used to call digest_init here. However, digest_init
will report errors, which we don't want when complain
Here orig_arg is SIZEOF_EXPR<T>; dependent_template_arg_p (orig_arg) was
true, but type_dependent_expression_p (orig_arg) is false so we warn in
convert_nontype_argument.
gcc/cp/ChangeLog:
PR c++/99331
* call.c (build_converted_constant_expr_internal): Don't emit
-Wconversion warnings.
gcc/testsuite/ChangeLog:
PR c++/99331
* g++.dg/warn/Wconversion5.C: New test.
For a target with none of aligned_alloc, memalign etc. we defined our
own aligned_alloc using malloc, so we need a declaration of malloc. As
in libsupc++/new_op.cc we need to declare it ourselves for freestanding
environments.
libstdc++-v3/ChangeLog:
* libsupc++/new_opa.cc [!_GLIBCXX_HOSTED]: Declare malloc.
The testsuite utilities that use random numbers use a
default-constructed mersenne_twister_engine, meaning the values are
reproducable. This adds support for seeding them, controlledby an
environment variable. Defining GLIBCXX_SEED_TEST_RNG=val in the
environment will cause the engines to be seeded with atoi(val) if that
is non-zero, or with a value read from std::random_device otherwise.
Running with different seeds revealed some bugs in the tests, where a
randomly selected iterator was past-the-end (which can't be erased), or
where the randomly populated container was empty, and then we tried to
remove elements from it unconditionally.
libstdc++-v3/ChangeLog:
* testsuite/util/exception/safety.h (setup_base::generate):
Support seeding random engine.
(erase_point, erase_range): Adjust range of random numbers to
ensure dereferenceable iterators are used where required.
(generation_prohibited::run): Do not try to erase from empty
containers.
* testsuite/util/testsuite_containergen.h (test_containers):
Support seeding random engine.
Minor patch to add a graceful exit in the rare case where an invalid
combination of TYPE_VECTOR_SUBPARTS for nunits_vectype and
*stmt_vectype_out is reached in vect_get_vector_types_for_stmt.
This resolves the ICE seen in PR tree-optimization/96974, however the issue
of correctly handling this rare vectorization combination is left for a
later patch.
Bootstrapped and reg-tested on aarch64-linux-gnu.
2021-03-25 Stam Markianos-Wright <stam.markianos-wright@arm.com>
gcc/ChangeLog:
PR tree-optimization/96974
* tree-vect-stmts.c (vect_get_vector_types_for_stmt): Replace assert
with graceful exit.
gcc/testsuite/ChangeLog:
PR tree-optimization/96974
* g++.target/aarch64/sve/pr96974.C: New test.
For unknown reasons, this had gotten added for the libgomp HSA plugin in commit
b8d89b03db (r242749) "Remove build dependence on
HSA run-time", and later propagated into the GCN plugin.
libgomp/
* plugin/plugin-gcn.c (init_environment_variables): Don't prepend
the 'HSA_RUNTIME_LIB' path to 'libhsa-runtime64.so'.
* plugin/configfrag.ac (HSA_RUNTIME_LIB): Clean up.
* config.h.in: Regenerate.
* configure: Likewise.
This patch is to initialize the inside_cost as zero, can avoid
to use its uninitialized value when some path doesn't assign it.
gcc/ChangeLog:
* tree-vect-loop.c (vect_model_reduction_cost): Init inside_cost.
Honza has fairly recently changed operand_equal_p to compare
DECL_FIELD_OFFSET for COMPONENT_REFs when comparing addresses.
As the first testcase in this patch shows, while that is very nice
for optimizations, for the -Wduplicated-branches warning it causes
regressions. Pedantically a union in both C and C++ has only one
active member at a time, so using some other union member even if it has the
same type is UB, so I think the warning shouldn't warn when it sees access
to different fields that happen to have the same offset and should consider
them different.
In my first attempt to fix this I've keyed the old behavior on
OEP_LEXICOGRAPHIC, but unfortunately that has various problems, the warning
has a quick non-lexicographic compare in build_conditional_expr* and another
lexicographic more expensive one later during genericization and turning the
first one into lexicographic would mean wasting compile time on large
conditionals.
So, this patch instead introduces a new OEP_ flag and makes sure to pass it
to operand_equal_p in all -Wduplicated-branches cases.
The cvt.c changes are because on the other testcase we were warning with
UNKNOWN_LOCATION, so the user wouldn't really know where the questionable
code is.
2021-03-25 Jakub Jelinek <jakub@redhat.com>
PR c++/99565
* tree-core.h (enum operand_equal_flag): Add OEP_ADDRESS_OF_SAME_FIELD.
* fold-const.c (operand_compare::operand_equal_p): Don't compare
field offsets if OEP_ADDRESS_OF_SAME_FIELD.
* c-warn.c (do_warn_duplicated_branches): Pass also
OEP_ADDRESS_OF_SAME_FIELD to operand_equal_p.
* c-typeck.c (build_conditional_expr): Pass OEP_ADDRESS_OF_SAME_FIELD
to operand_equal_p.
* call.c (build_conditional_expr_1): Pass OEP_ADDRESS_OF_SAME_FIELD
to operand_equal_p.
* cvt.c (convert_to_void): Preserve location_t on COND_EXPR or
or COMPOUND_EXPR.
* g++.dg/warn/Wduplicated-branches6.C: New test.
* g++.dg/warn/Wduplicated-branches7.C: New test.
For always_inline in system headers, we don't know if caller's ISAs are
compatible with callee's ISAs until much later. Skip ISA check for
always_inline in system headers if caller has target attribute.
gcc/
PR target/98209
PR target/99744
* config/i386/i386.c (ix86_can_inline_p): Don't check ISA for
always_inline in system headers.
gcc/testsuite/
PR target/98209
PR target/99744
* gcc.target/i386/pr98209.c: New test.
* gcc.target/i386/pr99744-1.c: Likewise.
* gcc.target/i386/pr99744-2.c: Likewise.
