est_killed
? 0 : i0dest);
rtx elim_i0 = (local_elim_i0 == 0
|| (newi2pat && reg_set_p (i0dest, newi2pat))
? 0 : i0dest);
--
Eric Botcazou
may
> not be -Werror-clean when built with other versions.) Do this for both
> 32-bit and 64-bit hosts.
Done, no changes required.
--
Eric Botcazou
the local variavbles.
>
> gcc/testsuite/ChangeLog
> 2014-12-22 Bin Cheng
>
> PR rtl-optimization/62151
> * gcc.c-torture/execute/pr62151.c: New test.
OK, thanks.
--
Eric Botcazou
> Hmm, I further revised comment in the patch since in try_combine, i2
> is always after i0/i1. The original comment is inaccurate about that.
OK.
--
Eric Botcazou
pragma Atomic forces fields to be aligned in record types but we nevertheless
accept atomic misaligned standalone variables. This changes the latter.
Tested on x86_64-suse-linux, applied on the mainline.
2014-12-22 Eric Botcazou
* gcc-interface/decl.c (check_ok_for_atomic): Rename
pragma Volatile forces fields to be of nominal size in record types but we
nevertheless accept volatile oversized variables. This changes the former.
Tested on x86_64-suse-linux, applied on the mainline.
2014-12-22 Eric Botcazou
* gcc-interface/decl.c (gnat_to_gnu_field): Post the
djust comments.
2014-12-22 Eric Botcazou
* gnat.dg/nan_max.adb: New test.
--
Eric Botcazou
Revised version after Joseph's comments and latest libgcc changes.
gcc/ChangeLog
2014-12-22 Eric Botcazou
* config.gcc: Add Visium support.
* configure.ac: Likewise.
* configure: Regenerate.
* doc/extend.texi (interrupt attribute): Add Visium.
> PR ada/64349
> * env.c (__gnat_environ): Use the _NSGetEnviron() interface to access
> the environment for Darwin.
OK if Tristan has no objections to the patch:
https://gcc.gnu.org/ml/gcc-patches/2014-12/msg01753.html
--
Eric Botcazou
iner for this port?
Yes, I'm going to be the maintainer for now.
> I think this is good to go into the trunk. The blockage issue (if it's
> an issue) can be resolved as a follow-up.
OK, thanks again.
--
Eric Botcazou
inline.
>
>
> 2014-12-22 Eric Botcazou
>
> * gcc-interface/trans.c (Attribute_to_gnu) : If the
> type is a floating-point type, implement the semantics of the C99
> f{min,max} routines with regard to NaNs.
> (gnat_to_gnu): Call builtin_dec
Sanity checked only, but that's good enough so you can apply it, thanks.
--
Eric Botcazou
ng you as the maintainer for the port a formality.
> I'll start that process, but go ahead and list yourself as the
> maintainer in the MAINTAINERS file.
Will do.
--
Eric Botcazou
> Thank you for your notes, I'm just going to apply following patch
> with changed ChangeLog entries.
Thanks!
--
Eric Botcazou
Hi,
the attached patch removes obsolete ports (c4x, m68hc11 and ms1), toggles the
'p' letter and adjust accordingly (only avr, fr30, m68k, mcore, rs6000 and sh
still use define_peephole) and removes trailing spaces.
OK to commit?
Some ports are missing (lm32, moxie, nios2, nvptx, rl78, rx) so
I think that we don't want new define_peephole's at this point.
Applied on the mainline as obvious.
2015-01-05 Eric Botcazou
* config/nds32/nds32-peephole2.md: Do not mention define_peephole.
--
Eric BotcazouIndex: config/nds32/nds32-pe
OK to commit?
--
Eric BotcazouIndex: index.html
===
RCS file: /cvs/gcc/wwwdocs/htdocs/index.html,v
retrieving revision 1.946
diff -u -p -r1.946 index.html
--- index.html 23 Dec 2014 18:44:10 - 1.946
+++ index.html 6 Jan 2015 11:5
> I normally build with --disable-libsanitizer, because the sanitizers
> testresults are very unreproducable, so just annoying noise. This however
> makes most (all?) ubsan testcases fail, since they want to load a shared
> library that does not exist.
Same here, but I also have the problem with
> What errors do you get? Just curious.
Something along of "cannot load libtsan.so.0" and this comes from PR 58867.
> That said, I think your patch is fine.
Thanks, installed after testing on x86-64/Linux.
--
Eric Botcazou
ck_caller): New function.
> (ipa_sra_preliminary_function_checks): Use it.
--
Eric Botcazou
> I might claim updating this shouldn't ever require a review :-)
OK, it was mainly about toggling the 'p' letter...
I have installed the attached patch for lm32 (based on info privately provided
by Sébastien) and visium.
--
Eric BotcazouIndex: backends.html
===
> I think that for nios2 the correct entry is probably "SCpd", but the
> descriptions of some of the things in the table are too terse for me to
> be 100% sure that's correct. E.g., the nios2 architecture doesn't
> include the concept of "condition codes" at all, so how many registers
> they might
me the string
as Sandra did for the nios2 port and I'll update the document.
--
Eric Botcazou
> Xtensa is now supported by the QEMU, so probably question mark in its 'S'
> slot may be removed.
Done, thanks.
--
Eric Botcazou
> the attached patch removes obsolete ports (c4x, m68hc11 and ms1), toggles
> the 'p' letter and adjust accordingly (only avr, fr30, m68k, mcore, rs6000
> and sh still use define_peephole) and removes trailing spaces.
Same treatment for the 'd' letter, the ports that do not use DFA scheduler
desc
> the attached patch removes obsolete ports (c4x, m68hc11 and ms1), toggles
> the 'p' letter and adjust accordingly (only avr, fr30, m68k, mcore, rs6000
> and sh still use define_peephole) and removes trailing spaces.
This replaces 'a' with 'i' for multiple inheritance thunks and reuses 'a' for
L
s a nice translation of the new code. :-)
It seems to me that the patch wants to extend the support of generic subregs
to modes whose sizes are not multiple of each other, which is a requirement of
the existing code, but does that in a very specific case for the sake of the
ARM port without saying where all the above restrictions come from.
--
Eric Botcazou
matter of fact, there are no ports left with 'f'.
> a - Port uses neither LRA nor reload. Might be a new characteristic
> (along with several others).
>
> The closest string is probably
>
> SQCqfbde
Thanks, I have installed "SQCqgbde" based on the discussion above.
--
Eric Botcazou
FAICS every insn
(not expander) generates exactly 1 assembly instruction. Not very important
though since only iq2000 and mep also have this nice property.
--
Eric Botcazou
by Martin.
Bootstrapped/regtested on x86_64-suse-linux, OK for mainline and 4.9 branch?
2015-01-14 Eric Botcazou
* expr.c (expand_expr_real_1) : Use the expression to
set the memory attributes in all cases but clear MEM_EXPR if need be.
2015-01-14 Eric Botcazou
offsets produced by the Ada frontend.
> * omp-low.c (extract_omp_for_data): Convert the loop step to
> signed for pointer adjustments.
OK, thanks for digging. I'll have a look.
--
Eric Botcazou
eLog:
> 2012-10-31 Dehao Chen
>
> * emit-rtl.c (reorder_insns): Reset the source location for
> instructions moved out of its original residing basic block.
No, that isn't acceptable.
--
Eric Botcazou
. The debug
info describes the relationship between the generated code and the source code
and, at high optimization levels, this relationship is not isomorphic at all.
It's up to the source level debugger to filter out the non-isomorphic part of
the mapping if it deems desirable to do so.
--
Eric Botcazou
tions): Split complicated conditionals out into
> above new functions.
> * Makefile.in (CGRAPH_H): Add is-a.h as used by cgraph.h.
The installed patch touches the ada/, cp/ and lto/ subdirectories without
modifying their ChangeLog files. Please add the missing entries.
[Some people, like me, do use these ChangeLogs to synchronize trees]
--
Eric Botcazou
ck in rtl_verify_flow_info_1 is wrong as
well. But it contains a thinko that masks the bug. :-)
Tested on x86_64-suse-linux, applied on the mainline.
2012-11-02 Eric Botcazou
* cfg-flags.def (ABNORMAL_CALL): Fix comment.
(EH): Likewise.
(SIBCALL): Likewise.
* c
> Done and committed.
That's great, thanks!
--
Eric Botcazou
Yes, that can generate jumpiness in GDB, but this is far better that
breaking the coverage info by giving the same source location to instructions
that have different coverage status.
--
Eric Botcazou
happens if you go one step farther and always continue?
--
Eric Botcazou
ny case, the
piece of code which generates this PLUS needs also to be fixed (at least on
the mainline).
--
Eric Botcazou
compared to the other
architectures (I presume that the test passes on most of them).
> Ok to commit?
>
> gcc/testsuite:
> PR testsuite/55186
> * gcc/testsuite/gcc.dg/const-uniq-1.c (a): Increase length four times.
No objections by me though.
--
Eric Botcazou
rt][GET_MODE (ret_reg)]
&& copy_num >= nregs
&& OBJECT_P (SET_SRC (return_copy_pat)))
forced_late_switch = 1;
> 2012-11-04 Vladimir Yakovlev
> Uros Bizjak
>
> * mode-switching.c (create_pre_exit): Added code for
> maybe_builtin_apply case.
Present tense in ChangeLog.
--
Eric Botcazou
patch as-is for 4.8.
--
Eric Botcazou
final mode switch after it. */
--
Eric Botcazou
it so that an appropriate cast is
added back at the end if necessary.
Bootstrapped/regtested on x86_64-suse-linux, OK for mainline and 4.7 branch?
2012-11-05 Eric Botcazou
PR tree-optimization/54986
* gimple-fold.c (canonicalize_constructor_val): Strip again all no-op
c
if (mode != num_modes[e] && mode != MODE_EXIT (e))
break;
}
(with a few shortcuts to speed it up) be sufficient?
--
Eric Botcazou
it necessary?
--
Eric Botcazou
eason the field size would be different than
the size we got from the type? */
}
This would mean removing the first ??? comment. As for the second ???
comment, the answer is easy: because that's pretty much what defines a
bitfield! The size is DECL_SIZE_UNIT and not TYPE_SIZE_UNIT for them.
--
Eric Botcazou
> 2012-11-07 Kaz Kojima
>
> PR middle-end/49220
> * mode-switching.c (create_pre_exit): Set short_block if there
> are no copy insns.
OK, but clearly a rewrite of the function would be in order.
--
Eric Botcazou
more general
arguments to help Ada.
Fixed by disabling recursion entirely, as it was originally.
Bootstrapped/regtested on x86_64-suse-linux, applied on mainline and 4.7
branch as obvious.
2012-11-07 Eric Botcazou
PR middle-end/55219
* fold-const.c
e start to change the type of the
operation, why not always reassociate in the unsigned version of the type?
int
foo (int x, int y)
{
return (x + 1) + (int) (y + 1);
}
int
bar (int x, unsigned int y)
{
return (x + 1) + (int) (y + 1);
}
--
Eric Botcazou
ion
but doesn't update cfun->machine accordingly.
Tested on x86_64-suse-linux, applied on mainline and 4.7 branch as obvious.
2012-11-09 Eric Botcazou
* config/i386/i386.c (release_scratch_register_on_entry): Also adjust
sp_offset manually.
2012-11-09 Er
> Updated patch attached, tested on powerpc64-unknown-linux-gnu.
> This is exactly the same patch as before, except that it now uses
> "cleanup_cfg (0)" instead of delete_unreachable_blocks.
>
> OK for trunk?
OK if you adjust the ChangeLog as well, thanks.
--
Eric Botcazou
back-end, which wasn't updated for the new option machinery.
Tested on x86_64-apple-darwin10.8.0, OK for mainline and 4.7 branch?
2012-11-09 Eric Botcazou
* common.opt (fvar-tracking-uninit): Document.
* toplev.c (process_options): Fix handling of flag_var_tracking_u
This fixes a couple of cases where gnat_to_gnu would recurse on an empty node
and removes all traces of SCIL references from gigi.
Tested on x86_64-suse-linux, applied on mainline.
2012-11-09 Eric Botcazou
* gcc-interface/trans.c (gnat_to_gnu) : Deal
with a missing library
ut -fvar-tracking-uninit.
--
Eric Botcazou
This removes s-taspri-lynxos.ads, which is unused, and converts the 3 last
files under GPL V2+ (plus exception) to GPL V3+ (plus exception), as all the
other similar files. Applied on mainline and 4.7 branch.
2012-11-09 Eric Botcazou
PR other/52438
* s-taspri-lynxos.ads
re_constructor_field useless as well.
Bootstrapped/regtested on x86_64-suse-linux, sparc64-sun-solaris2.9, sparc-
sun-solaris2.10, powerpc-linux-gnu, ia64-linux-gnu and mips64el-linux-gnu
and applied on the mainline.
2012-11-10 Eric Botcazou
* expr.c (store_field): Remove T
> Tested as described in the covering note. OK to install?
>
> Richard
>
> gcc/
> * combine.c (make_extraction): Handle TRUNCATEd INNERs.
OK, thanks.
--
Eric Botcazou
ire the size to be known for adjust_object.
> (adjust_automodify_address_1, widen_memory_access): Update calls
> to adjust_address_1.
No objections by me.
--
Eric Botcazou
is looks better now, thanks.
--
Eric Botcazou
> This patch also handles SIGN_EXTEND. Tested on Linux/x32. OK to
> install?
I'd cautious here, that's uncharted territory and the SIGN_EXTEND case isn't
covered by your testing.
> 2012-11-10 H.J. Lu
>
> PR middle-end/55247
> PR middle-end/55259
> * emit-rtl.c (adjust_add
to remove unreachable
> blocks.
>
> testsuite/
> PR middle-end/55263
> * g++.dg/pr55263.C: New test.
OK, thanks.
--
Eric Botcazou
> MThis patch looks OK. i will re-test and commit it on the afernoon.
Ping. The ICE shows up when building Python.
--
Eric Botcazou
> It fixes the problem. Can you check it in?
Done.
--
Eric Botcazou
This fixes a pasto in the --with-cpu=niagara4 support on Solaris, as well as
applies a minor tweak to config/sparc/sparc.h.
Tested on SPARC/Solaris, applied on the mainline and 4.7 branch.
2012-11-11 Eric Botcazou
* config/sparc/sparc.h (AS_NIAGARA3_FLAG): Tweak.
* config
a recent 'as' supporting
the SPARC-T4? I'm mainly interested in the values of the -xarch= option.
Thanks in advance.
--
Eric Botcazou
> I comitted it this afternoon, hope everything is fine now.
Yes, thanks.
--
Eric Botcazou
cse.c (cse_insn): Where checking for blocking insns, treat
> UNSPEC_VOLATILE as blocking, besides volatile ASM.
That's fine on principle, but there is a predicate for this (volatile_insn_p)
so I think we should use it here. Moreover, cselib_process_insn has the same
check so we
had its own set
of -xarch values for a while and they haven't been compatible with 'as'.
--
Eric Botcazou
ignment. At this
> point, we are interested only in hard registers that are also used for
> function return value. Actually, even in pre-reload pass, there are no
> other assignments to hard registers.
Fine with me if this passes testing on x86-avx and SH4.
--
Eric Botcazou
ange our scheme, using 'e' instead of
'd' for SPARC4 instructions should work just fine with both GNU and Sun as.
--
Eric Botcazou
They show up when a lattice value changed is to CONSTANT.
Tested on x86_64-suse-linux, applied on mainline as obvious.
2012-11-12 Eric Botcazou
* tree-ssa-ccp.c (dump_lattice_value) : Fix duplication.
--
Eric BotcazouIndex: tree-ssa-ccp.c
ble. GCC supports far more architectures and OSes
than LLVM and thus requires additional and specific support code and we don't
want to have to go through LLVM to change it. As Andrew already suggested,
this should be handled like libffi.
--
Eric Botcazou
tpos,
bitregion_start, bitregion_end,
align, volatilep)
{
if (largest_mode != VOIDmode
&& GET_MODE_SIZE (mode) > GET_MODE_SIZE (largest_mode)
break;
if (iter.prefer_smaller_modes ())
return mode;
widest_mode = mode;
}
return widest_mode;
and the implementation entirely hidden.
--
Eric Botcazou
blity, reject modes that are wider than the
> + alignment. This has both advantages and disadvantages.
"For compatibility" is a bit vague (compatibility with what?). I'd write "For
historical reasons" or something along these lines.
--
Eric Botcazou
is still controversial so I'm not sure that we want
let it spread. Can't the client code just skip the problematic modes?
--
Eric Botcazou
ory not ending
in /usr/gnat/lib
make[4]: *** [install-toolexeclibLTLIBRARIES] Error 1
make[4]: Leaving directory
--
Eric Botcazou
> We really need to start using the newer names, as Sun is not going to
> provide single letter indicators for sparc4 or future xarch values.
>
> In fact, that's exactly what needed to be worked on from the beginning
> for the solaris side of this cbcond patch. We're talking in circles.
OK, sorr
> Thanks for finding this, that's definitely incorrect behavior. I bet there
> is some unintended override triggered by sparc4 selection, and I'll go and
> fix that soon.
You're welcome. That's the reason why I needed to go the ASM_ARCH way, the
straightforward approach would have put the -32/-
; i ; ++i)
> {
> enum machine_mode mode = *i;
> ...
> }
I pondered on that for half an hour. :-) But the amount of stuff you need to
write to make it work in this particular case will make the implementation
convoluted and bloated for no obvious gains IMO.
--
Eric Botcazou
well.
Thanks for persevering in getting this through despite all the roadblocks. ;-)
--
Eric Botcazou
efile syntax, cannot be evaluated in configure.ac
if_multiarch = $(if $(wildcard $(shell echo
$(SYSTEM_HEADER_DIR))/../../usr/lib/*/crti.o),$(1))
else
if_multiarch =
endif
--
Eric Botcazou
> ahh, this is with GNU make 3.80. Checked in the following patch as obvious.
Thanks!
--
Eric Botcazou
info. Ian has clearly outlined
the correct approach to addressing it.
Note that I didn't specifically reply to Dehao's patch here, which might be
acceptable in the end, only to David's seemingly general statement about the
"natural way" of putting a location on these statements.
--
Eric Botcazou
volatile bitfields on
top of bitregion_{start,end}, that's why I'm not sure we want to make it part
of the interface at all.
--
Eric Botcazou
problem: you cannot make that a general rule, since you don't
know the coverage status of the instruction just above the insertion point.
If a later optimization moves the new statements around, you may well end up
with wrong coverage info.
--
Eric Botcazou
-bitfields),
> or do I need to add new code to the expmed.c routines?
The former, I think.
--
Eric Botcazou
ainline?
2012-11-15 Eric Botcazou
PR middle-end/55321
* optabs.c (mark_nothrow_nononlocal): New function extracted from...
(emit_libcall_block_1): ...here. Call it.
(expand_mem_thread_fence): Call it for synchronize_libfunc.
2012-11-15
e COMPONENT_REF's.
Fine with me.
--
Eric Botcazou
Nobody has adapted (or submitted to the FSF) the
#ifdef __ARM_EABI_UNWINDER__ / #endif adjustments present in libsupc++ or
libjava.
--
Eric Botcazou
Hi,
that's not really a bugfix, but I don't see the point in keeping this flag,
which was already dead in 4.7, one more release.
Tested on x86_64-suse-linux, OK for the mainline?
2012-11-16 Eric Botcazou
* tree.h (DECL_RESTRICTED_P): Delete.
(DECL_READ_P): A
anitizer/sanitizer_common/sanitizer_linux.cc:42:1:
warning: control reaches end of non-void function [-Wreturn-type]
}
^
At global scope:
cc1plus: warning: unrecognized command line option "-Wno-c99-extensions"
[enabled by default]
make[5]: *** [sanitizer_linux.lo] Error 1
--
Eric Botcazou
ING: program timed out.
WARNING: program timed out.
--
Eric Botcazou
* gcc-interface/utils.c: Likewise.
> * gcc-interface/utils2.c: Likewise.
Thanks for the great work, everything seems to be working fine in Ada.
Minor nit: the patch has introduced a couple of trailing spaces. No big deal,
but we have a script that rejects the changes in this case. Fixed
han the limit (the outer loop has 78000 instructions)
>
> I will prepare patches to fix those issues.
Thanks for the analysis (and don't worry, I won't revert anything :-).
--
Eric Botcazou
> Unfortunately, it causes regressions; read on for a very brief
> analysis.
>
> For x86_64-linux (base multilib):
>
> Running
> /home/hp/gcctop/tmp/pr55030-2n/gcc/gcc/testsuite/gcc.dg/guality/guality.exp
> ... ...
> FAIL: gcc.dg/guality/pr36728-1.c -O1 line 12 arg7 == 30
> [...]
>
> I looked i
> Yes, I'll be looking into this soon.
We have a related regression on SPARC:
FAIL: gfortran.dg/minmaxloc_5.f90 -O3 -fomit-frame-pointer -funroll-loops
execution test
FAIL: gfortran.dg/minmaxloc_5.f90 -O3 -fomit-frame-pointer -funroll-all-loops
-finline-functions execution test
FAIL: gfortr
5 of them fail on Solaris 9 and 10 because they require __cxa_atexit to pass.
And 4 others fail on Solaris 9 only because of a TLS initialization failure.
Adjusted thus, tested on SPARC/Solaris 9, SPARC/Solaris 10 and x86-64/Linux,
applied on the mainline.
2012-11-19 Eric Botcazou
mbly file, which translates into (mem:SF (const_int 0 [0])) in the
RTL dumps. And the executable does segfault at runtime.
--
Eric Botcazou
27;t we arrange to clean up the REG_EQUAL/REG_EQUIV notes when we use them,
i.e. when DF_EQ_NOTES is set?
--
Eric Botcazou
2701 - 2800 of 4611 matches
Mail list logo
