Hello!
Compiling expmed.c has been warning free for some time now.
2013-10-02 Uros Bizjak
* Makefile.in (expmed.o-warn): Remove.
Bootstrapped on x86_64-pc-linux-gnu.
OK for mainline?
Uros.
Index: Makefile.in
On Wed, Sep 26, 2012 at 5:16 PM, Marc Glisse wrote:
> Adding an x86 maintainer in Cc:
>>> this patch passes bootstrap+testsuite. It is probably wrong in many ways,
>>> but I don't know enough to do more without some advice.
>>>
>>> The goal is to recognize that v[0]+v[1] can be computed with hadd
On Thu, Sep 27, 2012 at 8:20 PM, Jakub Jelinek wrote:
> On Thu, Sep 27, 2012 at 08:04:58PM +0200, Uros Bizjak wrote:
>> After some off-line discussion with Richard, attached is v2 of the patch.
>>
>> 2012-09-27 Uros Bizjak
>>
>> PR rtl-optimization/
On Tue, Sep 18, 2012 at 1:31 PM, Uros Bizjak wrote:
>> This patch aims to fix all stability issues related to using the first
>> scheduler in gcc
>> for x86 target (there several reported issues related to this problem).
>>
>> Main idea of this activity is mostly to
On Thu, Sep 20, 2012 at 8:35 AM, Uros Bizjak wrote:
> On Thu, Sep 20, 2012 at 8:06 AM, Vladimir Yakovlev
> wrote:
>> The compiler with the patch and without post_reload.patch is built and works
>> successfully. It has the only failure with avx-vzeroupper-3 test because of
>
, %esi
je .L8
ret
2012-09-30 Uros Bizjak
* gcc.target/i386/pad-10.c (foo2): Return x - z.
Tested on x86_64-pc-linux-gnu, committed to mainline SVN.
Uros.
Index: gcc.target/i386/pad-10.c
===
--- gcc.target
this change, btest passes OK, failing log and
runtime/pprof from libgo testsuite also pass OK.
BTW: It would be enough to compile only backtrace.c and simple.c with
-fasynchronous-unwind-tables, since critical wrapper functions live
here.
2012-10-01 Uros Bizjak
PR other/54761
ables option.
I have simplified the check for -funwind-tables to just look if the
library is compiled with gcc. This option is supported by gcc-2.96
(and probably earlier versions too).
2012-10-02 Uros Bizjak
PR other/54761
* configure.ac (CFLAGS): Add -funwind-tables when co
Hello!
2012-10-02 Uros Bizjak
* config/x86/target.h (struct gtm_jmpbuf): Merge x86_64
and ia32 declarations some more.
* config/x86/sjlj.S (_ITM_beginTransaction): Move ret to common code.
Tested on x86_64-pc-linux-gnu, committed to mainline SVN.
Uros.
Index: config
to AM_CFLAGS. CFLAGS is reserved for the user to override.
Thanks, attached is a version that introduces EXTRA_FLAGS instead.
2012-10-02 Uros Bizjak
PR other/54761
* configure.ac (EXTRA_FLAGS): New.
* Makefile.am (AM_FLAGS): Add $(EXTRA_FLAGS).
* configure,
On Tue, Oct 2, 2012 at 11:35 AM, Vladimir Yakovlev wrote:
The compiler with the patch and without post_reload.patch is built and
works
successfully. It has the only failure with avx-vzeroupper-3 test because of
post reload problem.
>>>
>>> Ok, can you please elaborate a bit on
On Tue, Oct 2, 2012 at 12:08 PM, Vladimir Yakovlev wrote:
> Will we wait for LRA commit or is it possiple to commit to trank
> vzeroupper patch now?
Since we can emit vzeroupper now, we will wait for LRA.
Uros.
On Tue, Oct 2, 2012 at 3:08 PM, Ian Lance Taylor wrote:
>> 2012-10-02 Uros Bizjak
>>
>> PR other/54761
>> * configure.ac (EXTRA_FLAGS): New.
>> * Makefile.am (AM_FLAGS): Add $(EXTRA_FLAGS).
>> * configure, Makefile.in:
On Tue, Oct 2, 2012 at 7:44 PM, Gabriel Dos Reis
wrote:
>>> > On a related issue, it looks to me that the compiler itself should be
>>> > compiled with -funwind-tables, otherwise there are no backtraces
>>> > generated, even if libbacktrace is linked in and operational. Again,
>>> > x86_64-linux-
On Tue, Oct 2, 2012 at 9:35 PM, H.J. Lu wrote:
> This patch checks SSE and YMM state support for -march=native. Tested
> on Linux/x86-64. OK to install?
>
> 2012-10-02 H.J. Lu
>
> PR target/54741
> * config/i386/driver-i386.c (XCR_XFEATURE_ENABLED_MASK): New.
> (XSTA
On Tue, Oct 2, 2012 at 11:01 PM, H.J. Lu wrote:
>
> This patch documents -mprefer-avx128. OK for trunk and 4.7?
>
> 2012-10-02 H.J. Lu
>
> PR target/54785
> * doc/invoke.texi: Document -mprefer-avx128.
OK everywhere.
Thanks,
Uros.
STRTAB 05282212
0184 0 0 1
[40] .symtab SYMTAB 05282e18
000a1160 0018 41 19147 8
[41] .strtab STRTAB 000
Hello!
> With the strnlen-related Go failures now gone on Solaris, another issue
> crept up: several tests were failing like this:
>
> throw: /proc/self/exe
>
> goroutine 4 [running]:
> panic during panic
> FAIL: log
>
> This is due to the hardcoded use of /proc/self/exe in libbacktrace,
> which d
On Mon, Oct 8, 2012 at 4:40 PM, Marc Glisse wrote:
> On Fri, 28 Sep 2012, Uros Bizjak wrote:
>
>>>>> 2) {v[0]-v[1], v[0]-v[1]} is not recognized as a hsubpd because
>>>>> vec_duplicate doesn't match vec_concat. Do we really need to duplicate
>>>&g
On Mon, Oct 8, 2012 at 6:08 PM, Uros Bizjak wrote:
>> +(define_insn "*sse3_haddv2df3"
>>[(set (match_operand:V2DF 0 "register_operand" "=x,x")
>> (vec_concat:V2DF
>> - (plusminus:DF
>> + (plus:DF
>>
On Mon, Oct 8, 2012 at 9:36 PM, Marc Glisse wrote:
> On Mon, 8 Oct 2012, Uros Bizjak wrote:
>
>> You missed the most important sseadd1 addition, the one that prevents
>> checking of operand2 when calculating "memory" attribute:
>>
>> (and (eq_attr
Hello!
2012-10-08 Uros Bizjak
* config/i386/atom.md (atom_sse_4): Merge atom_sse_attr attibutes.
(atom_sse_5): Ditto.
Tested on x86_64-pc-linux-gnu {,-m32}, committed to mainline SVN.
Uros.
Index: config/i386/atom.md
Hello!
2012-10-09 Uros Bizjak
* testsuite/util/testsuite_abi.cc (check_version): Add CXXABI_1.3.7.
Tested on x86_64-pc-linux-gnu.
OK for mainline?
Uros.
Index: testsuite/util/testsuite_abi.cc
===
--- testsuite/util
Hello!
>> +2012-10-09 Venkataramanan Kumar
>> +
>> +PR testsuite/53397
>> +* gcc/testsuite/gcc.dg: Remove test cases pr53397-1.c and pr53397-2.c
>> +they are moved to gcc.target/i386 as they are specific to i386/x86_64.
>> +* gcc/testsuite/gcc.target/i386/pr53397-1.c: New test
>
Hello!
For 64bit targets, we can use mxcsr register to set soft-FP rounding mode.
2012-10-10 Uros Bizjak
* config/i386/sfp-machine.h (FP_RND_NEAREST, FP_RND_ZERO, FP_RND_PINF,
FP_RND_MINF, FP_RND_MASK, FP_INIT_ROUNDMODE, _FP_DECL_EX): Move to ...
* config/i386/32/sfp
Hello!
Recent commit changed various c++ TLS errors. Following patch tweaks
obj-c++ testcase accordingly.
2012-10-11 Uros Bizjak
* obj-c++.dg/tls/init-2.mm: Tweak errors.
Tested on alphaev68-pc-linux-gnu, committed to mainline SVN.
Uros.
Index: obj-c++.dg/tls/init-2.mm
Hello!
For 32bit targets, there is no guarantee that mxcsr register will be
set up correctly w.r.t. unmasked FP exceptions, so don't use SSE
instructions to generate INVALID and DIVZERO FP exceptions.
2012-10-10 Uros Bizjak
* config/i386/sfp-exceptions.c (__sfp_handle_excep
On Thu, Oct 11, 2012 at 9:19 AM, Gopalasubramanian, Ganesh
wrote:
> The attached patch (Patch.txt) enables the next version of AMD's bulldozer
> core.
Please handle new sseshuf type attribute in various attribute
calculations. You should at least add it to unit attribute
calculation, but please
Hello!
2012-10-11 Uros Bizjak
* config/alpha/alpha.md (IMODE): New mode iterator.
(I124MODE): Ditto.
(I248MODE): Ditto.
(modesuffix): Handle QI and HI modes.
(zero_extendqi2): Macroize insn from zero_extendqi{hi,si,di}2
using I248MODE mode
Hello!
2012-10-12 Uros Bizjak
* config/alpha/alpha.md (vecmodesuffix): New mode attribute.
(modesuffix): Handle V8QI and V4HI modes.
(any_maxmin): New code iterator.
(maxmin): New code attribute.
(3): Macroize insn from {smax,smin,umax,umin}{qi,hi}3
On Fri, Oct 12, 2012 at 7:57 PM, Ulrich Weigand wrote:
> Hello,
>
> I was running a couple of tests on various platforms in preparation
> of getting the find_reload_subreg_address patch needed by aarch64 upstream:
> http://gcc.gnu.org/ml/gcc-patches/2012-07/msg01421.html
>
> This unfortunately unc
Hello!
2012-10-13 Uros Bizjak
* config/alpha/alpha.md (I24MODE): New mode iterator.
(any_divmod): New code iterator.
(si3): Macroize expander from {div,mod,udiv,umod}si3 using
any_divmod code iterator.
(si3): Macroize expander from {div,mod,udiv,umod
On Sat, Oct 13, 2012 at 10:52 AM, Marc Glisse wrote:
> Hello,
>
> this patch provides an alternate pattern to let combine recognize scalar
> operations that preserve the high part of a vector. If the strategy is all
> right, I could do the same for more operations (mul, div, ...). Something
> simi
Hello!
2012-10-14 Uros Bizjak
* config/alpha/alpha.md: Remove empty predicates and/or constraints.
* config/alpha/sync.md: Ditto.
Tested on alphaev68-pc-linux-gnu, committed to mainline SVN.
Uros.
a.diff.txt.gz
Description: GNU Zip compressed data
On Mon, Oct 15, 2012 at 6:39 PM, Ulrich Weigand wrote:
>> On Fri, Oct 12, 2012 at 7:57 PM, Ulrich Weigand wrote:
>> > I was wondering if the i386 port maintainers could have a look at this
>> > pattern. Shouldn't we really have two patterns, one to *load* an unaligned
>> > value and one to *sto
FAIL: gcc.target/i386/avx256-unaligned-store-4.c scan-assembler
> avx_movups256/2
> But I guess these tests simply need to be updated for the new pattern names.)
2012-10-15 Uros Bizjak
* config/i386/sse.md (UNSPEC_MOVU): Remove.
(UNSPEC_LOADU): New.
(UNSPEC_STOR
On Thu, Oct 18, 2012 at 11:16 AM, Alexander Ivchenko wrote:
> Hi,
>
> this patch adds new intrinsics for fxsave, fxsave64, xsave, xsave64,
> xsaveopt and xsaveopt64 instructions
>
> Bootstrapped on x86-64
>
> Is it ok for trunk?
>
> Changelog entry:
> 2012-10-16 Alexander Ivchenko
>
> *
On Thu, Oct 18, 2012 at 5:45 PM, Uros Bizjak wrote:
> On Thu, Oct 18, 2012 at 11:16 AM, Alexander Ivchenko
> wrote:
>> Hi,
>>
>> this patch adds new intrinsics for fxsave, fxsave64, xsave, xsave64,
>> xsaveopt and xsaveopt64 instructions
>>
>> Boots
Hello!
Attached patch moves a pack of duplicated code from i386.md to i386.c.
No functional change.
2012-10-18 Uros Bizjak
* config/i386/i386.md (rdpmc): Remove expander.
(rdtsc): Ditto.
(rdtscp): Ditto.
(rdpmc): Rename from *rdpmc.
(rdpmc_rex64
length calculation. We have to look at
the mode of base or index _register_ of decomposed address, since
addr32 prefix is emitted only when registers are used in the address.
Note, that we never emit addr32 prefix for LEA insns.
2012-10-21 Uros Bizjak
* config/i386/i386-protos.h
Hello!
We don't need to check for REG_P on base and index, we are sure that
non-null RTXes are registers only. Also, we should determine the mode
of RTXes in addr32 calculation from original RTXes.
2012-10-22 Uros Bizjak
* config/i386/i386.c (memory_address_length): Assert tha
On Mon, Oct 22, 2012 at 5:25 PM, Alexander Ivchenko wrote:
> Please take a look at the updated patch. There is, thanks to Uros, changed
> expander and asm patterns.
>
> Considering H.J.'s comments:
>
> 1) Yes, I added new option -mxsaveopt
> 2) No.The FXSAVE and FXRSTOR instructions are not consid
On Tue, Oct 23, 2012 at 12:14 PM, Alexander Ivchenko wrote:
> Please take a look at the attached patch.
>
> I changed the asm-pattern implementation according to your recomendation.
> Changed the name of feature option from -mfxsave to -mfxsr, as it is in
> Intel SDM. Corrected the arguments name
On Tue, Oct 23, 2012 at 1:38 PM, Yuri Rumyantsev wrote:
> This fix is aimed to remove stability issues with using pre-reload
> scheduler for x86 targets caused by cross-block motion of function
> arguments passed in likely-spilled HW registers. We found one more
> issue in a process of more detai
Hello!
> I have committed a patch to update the mainline version of libgo to the
> current master Go library sources. At this point I will only be
> updating the gcc 4.7 branch for bug fixes.
>
> This is a substantial patch that brings in several months of work. As
> usual I am not posting the c
Hello!
> Hi, I was going to merge LRA into trunk last Sunday. It did not happen. LRA
> was actively changed last 4 weeks by implementing reviewer's proposals which
> resulted in a lot of new LRA regressions on GCC testsuite in comparison with
> reload. Finally, they were fixed and everything lo
On Tue, Oct 23, 2012 at 8:09 PM, Ian Lance Taylor wrote:
> On Tue, Oct 23, 2012 at 10:47 AM, Uros Bizjak wrote:
>>
>> Additional test fails on alphaev68-linux-gnu:
>>
>> --- FAIL: TestPassFD (0.15 seconds)
>> passfd_test.go:62: FileConn: dup: Bad file de
On Wed, Oct 24, 2012 at 10:01 AM, Uros Bizjak wrote:
> On Tue, Oct 23, 2012 at 8:09 PM, Ian Lance Taylor wrote:
>> On Tue, Oct 23, 2012 at 10:47 AM, Uros Bizjak wrote:
>>>
>>> Additional test fails on alphaev68-linux-gnu:
>>>
>>> --- FAIL: Te
On Wed, Oct 24, 2012 at 10:01 AM, Uros Bizjak wrote:
>>> Additional test fails on alphaev68-linux-gnu:
>>>
>>> --- FAIL: TestPassFD (0.15 seconds)
>>> passfd_test.go:62: FileConn: dup: Bad file descriptor
>>> FAIL
>>> FAIL: syscall
&g
On Wed, Oct 24, 2012 at 10:22 AM, Florian Weimer wrote:
>>> Is it OK to call dup on the same FD the second time?
>>
>>
>> To answer my own question:
>>
>> dup(4) = 9
>> ...
>> close(9)= 0
>> dup(4) =
On Wed, Oct 24, 2012 at 12:52 PM, H.J. Lu wrote:
>> Please take a look at the attached patch.
>>
>> I changed the asm-pattern implementation according to your recomendation.
>> Changed the name of feature option from -mfxsave to -mfxsr, as it is in
>> Intel SDM. Corrected the arguments name in the
On Wed, Oct 24, 2012 at 1:06 PM, Uros Bizjak wrote:
>>> Please take a look at the attached patch.
>>>
>>> I changed the asm-pattern implementation according to your recomendation.
>>> Changed the name of feature option from -mfxsave to -mfxsr, as it is in
&g
On Wed, Oct 24, 2012 at 2:18 PM, Andreas Schwab wrote:
> Uros Bizjak writes:
>
>> To answer my own question:
>>
>> dup(4) = 9
>> ...
>> close(9)= 0
>> dup(4)
On Wed, Oct 24, 2012 at 3:10 PM, Ian Lance Taylor wrote:
> On Wed, Oct 24, 2012 at 5:31 AM, Uros Bizjak wrote:
>> On Wed, Oct 24, 2012 at 2:18 PM, Andreas Schwab
>> wrote:
>>> Uros Bizjak writes:
>>>
>>>> To answer my own question:
>>
On Wed, Oct 24, 2012 at 3:01 PM, Andrey Turetskiy
wrote:
> On Tue, Oct 23, 2012 at 2:45 PM, Andrey Turetskiy
> wrote:
>> Hi,
>>
>> This patch replaces large const_vector constructions with
>> match_operand in sse.md to decrease its size.
>> Is it ok?
No, you don't have to touch generic expand m
On Wed, Oct 24, 2012 at 4:30 PM, Ian Lance Taylor wrote:
> On Wed, Oct 24, 2012 at 6:19 AM, Uros Bizjak wrote:
>> On Wed, Oct 24, 2012 at 3:10 PM, Ian Lance Taylor wrote:
>>> On Wed, Oct 24, 2012 at 5:31 AM, Uros Bizjak wrote:
>>>> On Wed, Oct 24, 2012 at 2:18 P
On Wed, Oct 24, 2012 at 7:46 PM, Ian Lance Taylor wrote:
> On Wed, Oct 24, 2012 at 9:34 AM, Uros Bizjak wrote:
>>
>> Continuing.
>> [New Thread 0x2000307b280 (LWP 8059)]
>>
>> Breakpoint 18, 0x020002e378c0 in socketpair () from /lib/libc.so.6.1
>>
>
On Wed, Oct 24, 2012 at 1:06 PM, Uros Bizjak wrote:
> On Wed, Oct 24, 2012 at 12:52 PM, H.J. Lu wrote:
>>> Please take a look at the attached patch.
>>>
>>> I changed the asm-pattern implementation according to your recomendation.
>>> Changed the name of
Hello!
Attached patch introduces __builtin_ia32_fxsave to crtfastmath.c. In
addition, it clears only fxsave.mxcsr_mask, as we will look at this
field only. Also, we don't have to execute additional stmxcsr, since
fxsave also saves this register.
2012-10-26 Uros Bizjak
* config
Hello!
Attached patch adds a testcase for RA problem that was fixed by LRA.
2012-10-27 Uros Bizjak
PR target/34283
* gcc.target/i386/pr34283.c: New test.
Tested on x86_64-pc-linux-gnu, committed to mainline SVN.
Uros.
Index: gcc.target/i386/pr34283.c
when to-be-removed pattern triggers, with the intention
that these patterns wil be removed entirely (An "invalid" pattern was
indeed generated elsewhere, see patch).
2012-10-18 Uros Bizjak
* config/i386/i386.md (ashift to lea splitter): Split to SImode mult.
(simple
;>> and MULT operators. I have attached RFC prototype patch that will
>>> trigger an ICE when to-be-removed pattern triggers, with the intention
>>> that these patterns wil be removed entirely (An "invalid" pattern was
>>> indeed generated elsewhere, se
On Sun, Oct 28, 2012 at 9:57 AM, Uros Bizjak wrote:
> On Sun, Oct 28, 2012 at 2:37 AM, H.J. Lu wrote:
>
>>>> As suggested by Richard S. [1], after the patch that converts subreg:M
>>>> (op:N (...)(...)) to op:M (subreg:M (...) subreg:M (...)), we can
>>>&
see patch).
I have committed following version that avoids all failures, reported by H.J.:
2012-10-29 Uros Bizjak
* config/i386/i386.c (ix86_decompose_address): Use simplify_gen_subreg
to generate SImode equivalent of address, zero-extended with AND RTX.
* config/i386/i386.
On Mon, Oct 29, 2012 at 9:07 AM, Uros Bizjak wrote:
>>>> As suggested by Richard S. [1], after the patch that converts subreg:M
>>>> (op:N (...)(...)) to op:M (subreg:M (...) subreg:M (...)), we can
>>>> remove several peephole2 patterns that handle subregs of
On Mon, Oct 29, 2012 at 10:06 AM, Uros Bizjak wrote:
>>>>> As suggested by Richard S. [1], after the patch that converts subreg:M
>>>>> (op:N (...)(...)) to op:M (subreg:M (...) subreg:M (...)), we can
>>>>> remove several peephole2 patterns that
Hello!
Since we use simplify_gen_subreg on address part, we can use it for
DImode subreg of SImode address as well. simplify_gen_subreg knows how
to strip subreg.
2012-10-29 Uros Bizjak
* config/i386/i386.c (ix86_decompose_address): Use simplify_gen_subreg
for all addresses
On 12-10-29 12:21 PM, Richard Sandiford wrote:
> Vladimir Makarov writes:
>>H.J. in
>>
>> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55116
>>
>>reported an interesting address
>>
>>> (and:DI (subreg:DI (plus:SI (ashift:SI (reg:SI 96 [ glob_vol_int.22 ])
>> (const_int 2 [0
On Tue, Oct 30, 2012 at 3:47 PM, Andrey Turetskiy
wrote:
> I changed the patch according Uros' remarks. Please, have a look.
>
> Changelog:
>
> 2012-10-30 Andrey Turetskiy
>
>* config/i386/i386.c (bdesc_args): Rename CODE_FOR_avx2_umulhrswv16hi3
> to
>CODE_FOR_avx2_pmulhrswv16h
On Tue, Oct 30, 2012 at 6:53 PM, Andrey Turetskiy
wrote:
> Thanks for explanation, I understand it.
> I fixed issue which you marked. Changelog is unchanged.
>
>>> I changed the patch according Uros' remarks. Please, have a look.
>>>
>>> Changelog:
>>>
>>> 2012-10-30 Andrey Turetskiy
>>>
>>>
Hello!
2013-10-02 Uros Bizjak
* config/i386/x-i386 (driver-i386.o): Remove header dependencies.
Use $(COMPILE) and $(POSTCOMPILE).
* config/alpha/x-alpha (driver-alpha.o): Ditto.
Bootstrapped on x86_64-pc-linux-gnu and alphaev68-pc-linux-gnu,
committed to mainline SVN.
Uros
On Wed, Oct 2, 2013 at 1:55 PM, Uros Bizjak wrote:
> Compiling expmed.c has been warning free for some time now.
As discussed briefly on IRC: let's try -Werror and see which target will break.
Committed to mainline.
Uros.
Hello!
2013-10-02 Uros Bizjak
* config/x-linux (host-linux.o): Remove header dependencies.
Use $(COMPILE) and $(POSTCOMPILE).
* config/t-linux-android (linux-android.o): Ditto.
Bootstrapped on x86_64-pc-linux-gnu and committed to mainline.
Uros.
Index: config/t-linux-android
Hello!
As shown in the attached testcase, arguments of various __atomic
builtins should be converted as signed, so the immediates get properly
extended.
2013-10-08 Uros Bizjak
* optabs.c (maybe_emit_atomic_exchange): Convert operands as signed.
(maybe_emit_sync_lock_test_and_set
On Wed, Oct 9, 2013 at 6:23 PM, Richard Henderson wrote:
> This doesn't seem right at all.
>
> The bug is that I gets set to UINT64_MAX, right? Where's the
> incorrect conversion from int to __int128_t? Surely you can
Please see Comment #5 of PR58542:
--cut here--
The problem actually starts
On Sat, Oct 12, 2013 at 2:16 AM, Sriraman Tallam wrote:
> Ping.
This looks nice. I suppose you grepped for effected targets and rs6000
was the only one besides i386.
This is ok given target maintainers do not object within 24h and the test
successfully bootstrapped and
On Fri, Oct 11, 2013 at 8:38 PM, H.J. Lu wrote:
> In x32, when a TLS address is in DImode and Pmode is SImode,
> copy_addr_to_reg will fail. This patch adds ix86_copy_addr_to_reg
> to first copy DImode address into a DImode register and then to generate
> SImode SUBREG in this case. Tested on x
Hello!
> 2013-10-11 Alexander Monakov
>
> * config/i386/i386.c (ix86_expand_sse_compare_and_jump): Use mode
> provided by ix86_fp_compare_mode instead of CCFPUmode.
>
> testsuite/:
> * gcc.target/i386/builtin-ucmp.c: New test.
OK for mainline.
Thanks,
Uros.
On Sat, Oct 12, 2013 at 4:57 PM, H.J. Lu wrote:
>>> In x32, when a TLS address is in DImode and Pmode is SImode,
>>> copy_addr_to_reg will fail. This patch adds ix86_copy_addr_to_reg
>>> to first copy DImode address into a DImode register and then to generate
>>> SImode SUBREG in this case. Tes
On Wed, Oct 16, 2013 at 8:28 AM, Gopalasubramanian, Ganesh
wrote:
> The below patch enables FMA4 for AMD bdver3 architectures.
>
> "make -k check" passes.
>
> +2013-10-16 Ganesh Gopalasubramanian
> +
> + * config/i386/i386.c (ix86_option_override_internal): Enable FMA4
> + for AMD b
Hello!
> This patch to libgo fixes a dumb bug in which catching a signal via the
> os/signal package could clobber the saved split-stack context, leading
> to disaster if the signal arrived at the wrong time. Bootstrapped and
> ran Go testsuite on x86_64-unknown-linux-gnu. Committed to mainline.
On Wed, Oct 16, 2013 at 6:06 PM, Kirill Yukhin wrote:
> It seems that gang of AVX* patterns were copy and
> pasted from SSE, however as far as they are NDD,
> we may remove corresponding expands which sort operands.
>
> ChangeLog:
> * config/i386/sse.md (vec_widen_umult_even_v8si): Remove
On Wed, Oct 16, 2013 at 6:06 PM, Kirill Yukhin wrote:
> It seems that gang of AVX* patterns were copy and
> pasted from SSE, however as far as they are NDD,
> we may remove corresponding expands which sort operands.
OTOH, I have some second thoughts on removing AVX2 expanders.
Please consider t
On Wed, Oct 16, 2013 at 7:01 PM, Richard Henderson wrote:
> On 10/16/2013 09:47 AM, Uros Bizjak wrote:
>> On Wed, Oct 16, 2013 at 6:06 PM, Kirill Yukhin
>> wrote:
>>
>>> It seems that gang of AVX* patterns were copy and
>>> pasted from SSE, however as
On Thu, Oct 17, 2013 at 12:47 PM, Kirill Yukhin wrote:
>
>> > I suspect gen_lowpart is bad turn when reload is completed, as
>> > far as it can create new pseudo. gen_lowpart () may call
>> > gen_reg_rtx (), which contain corresponging gcc_assert ().
>>
>> False. gen_lowpart is perfectly safe pos
approach requires less fixups in reload pass.
2013-10-17 Uros Bizjak
* config/i386/i386.c (ix86_fixup_binary_operands): When both source
operands are in memory, prefer to force non-matched operand 1 to
the register.
Patch was tested on x86_64-pc-linux-gnu {,-m32} and committed to
Hello!
The isa attribute is not needed since we already constrainted insn
with TARGET_AVX2.
2013-10-17 Uros Bizjak
* config/i386/sse.md (*vec_widen_smult_even_v8si): Remove
isa attribute.
Tested on x86_64-pc-linux-gnu {,-m32} and committed to mainline SVN.
Uros.
Index: config/i386
Hello!
>>> I would need the help of target maintainers to fix it this way since
>>> it touches every target and it would take time for me to build and
>>> test every target.
>>
>> For changes that only need a compile to ensure one didn't brake a port, a
>> configure and build of
> a target is 2 m
Hello!
> This patch to libgo adds some system calls to the syscall package.
> Bootstrapped and ran Go testsuite on x86_64-unknown-linux-gnu.
> Committed to mainline and 4.8 branch, since these calls were in Go
> 1.1.2.
Uros.
Index: runtime/go-nosys.c
=
o the patch clarifies that __builtin_apply
situation actually applies to all multiple return register exits. The
patch clarifies this situation.
No functional changes (modulo converting a couple of variables to boolean type).
2013-10-19 Uros Bizjak
* mode-switching.c (create_pre_exit): R
58792 by adding missing registers to
function_value_regno_p.
2013-10-19 Uros Bizjak
PR target/58792
* config/i386/i386.c (ix86_function_value_regno): Add DX_REG,
ST1_REG and XMM1_REG for 32bit and 64bit targets. Also add DI_REG
and SI_REG for 64bit SYSV ABI targets.
The patch
On Sat, Oct 19, 2013 at 2:55 PM, Uros Bizjak wrote:
>
> The list of possible return registers in ix86_function_value_regno_p
> is incomplete. 32bit and 64bit targets use %[er]dx register for
> DImode and TImode values respectively and %st(1) and %xmm1 for
> imaginary part of
Hello!
> I'm fighting with mode switching (to be more precise with
> create_pre_exit function) trying to make it work for MPX. I saw
> create_pre_exit had some stability issues before and now I'm facing
> similar issues trying to have it working when bound register is
> returned by function in ad
On Thu, Oct 17, 2013 at 4:16 PM, Kirill Yukhin wrote:
>> >> > I suspect gen_lowpart is bad turn when reload is completed, as
>> >> > far as it can create new pseudo. gen_lowpart () may call
>> >> > gen_reg_rtx (), which contain corresponging gcc_assert ().
>> >>
>> >> False. gen_lowpart is perfe
Hello!
We are splitting it to normal XOR that includes clobber, so better
start with the clobber from the beginning.
2013-10-20 Uros Bizjak
* config/i386/i386.md (kxnor): Add FLAGS_REG clobber.
Tested on x86_64-pc-linux-gnu and committed to mainline SVN.
Uros.
Index: i386.md
sequence gets the inverted condition.
The attached patch removes wrong definitions.
2013-10-22 Uros Bizjak
PR target/58779
* config/i386/i386.c (put_condition_code) :
Remove CCCmode handling.
: Return 'c' suffix for CCCmode.
: Return 'nc' suffix for CCCmod
On Tue, Oct 22, 2013 at 11:05 PM, Sriraman Tallam wrote:
>This simple patch fixes the -m32 -mno-sse bugs you reported. A few
> more places where I did not change references to global_options.
> Uros/Richard: Is this ok to commit?
>
> * config/i386/i386.c (ix86_option_override_internal):
> Cha
Hello!
> Currently GCC could not vectorize abs() operation for integers on x86
> with only SSE2 support. For int type, the reason is that the expand on
> abs() is not defined for vector type. This patch defines such an
> expand so that abs(int) will be vectorized with only SSE2.
+(define_expand "
On Thu, Oct 24, 2013 at 12:06 PM, Ilya Enkovich wrote:
> On 01 Oct 20:00, Uros Bizjak wrote:
>>
>> This is OK for mainline, on the condition that target independent part
>> is approved and committed first.
>>
>> Thanks,
>> Uros.
>
> Thanks for revi
On Thu, Oct 24, 2013 at 1:29 PM, Ilya Enkovich wrote:
This is OK for mainline, on the condition that target independent part
is approved and committed first.
Thanks,
Uros.
>>>
>>> Thanks for review!
>>>
>>> Attached is a version to be committed. The only difference from
401 - 500 of 6168 matches
Mail list logo
