Hi
Thanks to the recent changes made to stage 2 and 3 (now built with g++),
i noticed a little error in the configure script that tries the
system getopt declaration. Indeed, if your system defines it in a system
header file named "getopt.h" (for example /usr/include/getopt.h on a Red
Hat 4 co
On Tue, Aug 9, 2011 at 8:07 PM, Michael Meissner
wrote:
> This is an initial patch to work around the slow down of sphinx3 in power7 VSX
> that first shows up in GCC 4.6 and is still present in the current GCC 4.7
> trunk. http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50031
>
> The key part of the
PING...
BTW, in patch fix_cond_cmp_2.patch, the file mode of thumb2.md is carelessly
changed, so please check attached new patch file fix_cond_cmp_3.patch.
Thanks,
-Jiangning
> -Original Message-
> From: Jiangning Liu [mailto:jiangning@arm.com]
> Sent: Monday, August 08, 2011 2:01 PM
On Wednesday 10 August 2011 01:02:50, Steve Ellcey wrote:
> On Tue, 2011-08-09 at 16:50 -0700, Richard Henderson wrote:
> > >
> > > I think I like using a union to ensure the alignment of checksum better.
> > > In dwarf2out.c we are always using one md5_ctx structure and one
> > > checksum buffer
On Tue, 9 Aug 2011, Hans-Peter Nilsson wrote:
> On Tue, 9 Aug 2011, Richard Guenther wrote:
> >
> > This fixes PR49937 - callers of get_{pointer,object}_alignment
> > probably should not use BIGGEST_ALIGNMENT to limit what these
> > functions return (why do they do that? Maybe because formerly
>
On Wed, Aug 10, 2011 at 10:48 AM, Pedro Alves wrote:
> On Wednesday 10 August 2011 01:02:50, Steve Ellcey wrote:
>> On Tue, 2011-08-09 at 16:50 -0700, Richard Henderson wrote:
>> > >
>> > > I think I like using a union to ensure the alignment of checksum better.
>> > > In dwarf2out.c we are always
Thanks for looking at this.
Bernd Schmidt writes:
> On 07/21/11 17:42, Richard Sandiford wrote:
>> /* When optimizing for speed, don't introduce dependencies between
>> memory references in the chain and memory references outside of it,
>> since doing so would limit schedu
Ping
On Wed, 3 Aug 2011, Alexander Monakov wrote:
> Hello,
>
> This is a series of selective scheduler bug fixes. They fix problems that
> have been discovered during internal testing, and one patch is necessary as
> preparation to predication support in the selective scheduler (predication
> p
"Paulo J. Matos" writes:
> There is a typo in the internal documentation. This patch fixes this.
Thanks, applied to trunk under the "obviously correct" rule. Sorry for
the slow response.
Richard
For the record, the changelog was:
2011-08-10 Paulo J. Matos
* doc/tm.texi.in (CLASS_M
Nothing in the middle-end happens conditional on
can_trust_pointer_alignment anymore - we can always "trust" pointer
alignment, that function and its comment is somewhat gross.
In fact we can now track alignment properly via CCP and thus
the hack in cfgexpand.c:expand_call_stmt should not be necc
2011/8/2 Romain Geissler :
> Hi,
>
> For now, plugins can't compare types. This patch allows
> c-tree.h to be installed as a plugin header, allowing
> plugins to see "comptypes" (among other things).
>
> Romain Geissler
>
>
> 2011-08-02 Romain Geissler
>
> * Makefile.in (PLUGIN_HEADERS):
On 10 August 2011 09:20, Jiangning Liu wrote:
> PING...
>
> BTW, in patch fix_cond_cmp_2.patch, the file mode of thumb2.md is carelessly
> changed, so please check attached new patch file fix_cond_cmp_3.patch.
>
Please do not top post.
I've been away for the whole of last week and then been awa
On Tue, Aug 9, 2011 at 8:39 PM, Aldy Hernandez wrote:
>
>> ok, so now you do this only for the first field in a bitfield group. But
>> you
>> do it for _all_ bitfield groups in a struct, not only for the interesting
>> one.
>>
>> May I suggest to split the loop into two, first searching the first
As has been mentioned before, one impediment to complete the toplevel
libgcc move is libada's use of TARGET_LIBGCC2_CFLAGS. AFAICS, it is
primarily used for gnatlib-*shared targets, so I assume that this was
done mostly (exclusively?) to get the flags necessary for PIC code
generation. Since thos
Jason Merrill writes:
> On 08/09/2011 09:14 AM, Marc Glisse wrote:
>> I don't think we should define the C++ 2011 value yet. In my opinion, we
>> should wait until:
>> 1) the standard is official
>> 2) gcc implements most of it: people will want to use __cplusplus as a
>> test to know if they can
Those functions now return conservative alignments and all callers
that call them with an argument of BIGGEST_ALIGNMENT do so without
any good reason. If the functions return a larger alignment then
it is known to be larger.
Bootstrapped and tested on x86_64-unknown-linux-gnu, applied to trunk.
On Wed, 10 Aug 2011, Richard Guenther wrote:
>
> Nothing in the middle-end happens conditional on
> can_trust_pointer_alignment anymore - we can always "trust" pointer
> alignment, that function and its comment is somewhat gross.
> In fact we can now track alignment properly via CCP and thus
> t
This appears to have brought back PR 26478, build failure with readonly
source directory:
cp unwind.h ../../.././gcc/include/unwind.h
cp: cannot create regular file `../../.././gcc/include/unwind.h': Permission
denied
make[4]: *** [install-unwind_h] Error 1
--
Joseph S. Myers
jos...@codesource
On Sat, Mar 29, 2008 at 10:11 PM, H.J. Lu wrote:
> This patch restores proper checking the third argument on blendpd and
> and blendps. It also adds 2 tests, including pblendw. Tested on
> Linux/Intel64. OK to install?
The gcc.target/i386/sse4_1-blendps-2.c test randomly fails because
src3 is u
"Joseph S. Myers" writes:
> This appears to have brought back PR 26478, build failure with readonly
> source directory:
>
> cp unwind.h ../../.././gcc/include/unwind.h
> cp: cannot create regular file `../../.././gcc/include/unwind.h': Permission
> denied
> make[4]: *** [install-unwind_h] Error
On Wed, 10 Aug 2011, Rainer Orth wrote:
> "Joseph S. Myers" writes:
>
> > This appears to have brought back PR 26478, build failure with readonly
> > source directory:
> >
> > cp unwind.h ../../.././gcc/include/unwind.h
> > cp: cannot create regular file `../../.././gcc/include/unwind.h':
> >
This patch adds {u|s}mul{qi|hi}3_highpart patterns which
can help dividing by a constant when MUL is available, e.g.
int8_t sdiv1 (int8_t a)
{
return a / 3;
}
uint8_t udiv1 (uint8_t a)
{
return a / 3;
}
uint16_t udiv2 (uint16_t a)
{
return a / 10;
}
compiles with -O2 -mmcu=atmega8 t
"Joseph S. Myers" writes:
>> This is strange: they copy explicitly goes into $(gcc_objdir): from
>> libgcc/Makefile.in:
>>
>> install-unwind_h:
>> cp unwind.h $(gcc_objdir)/include/unwind.h
>> chmod a+r $(gcc_objdir)/include/unwind.h
>>
>> For an in-tree build, the source direct
On Sun, Jul 24, 2011 at 8:33 PM, Richard Henderson wrote:
> With the last two fixes, the only remaining i686 regression
> from Thursday is 1 fortran testcase, whose number I have
> misplaced. This highlights an accidental change I made while
> moving code around between functions, and an assertio
On Wed, Aug 10, 2011 at 6:46 AM, Richard Guenther
wrote:
> On Sat, Mar 29, 2008 at 10:11 PM, H.J. Lu wrote:
>> This patch restores proper checking the third argument on blendpd and
>> and blendps. It also adds 2 tests, including pblendw. Tested on
>> Linux/Intel64. OK to install?
>
> The gcc.ta
On Wed, Aug 10, 2011 at 4:04 PM, H.J. Lu wrote:
> On Wed, Aug 10, 2011 at 6:46 AM, Richard Guenther
> wrote:
>> On Sat, Mar 29, 2008 at 10:11 PM, H.J. Lu wrote:
>>> This patch restores proper checking the third argument on blendpd and
>>> and blendps. It also adds 2 tests, including pblendw. T
Richard Guenther wrote:
> On Fri, 5 Aug 2011, Martin Jambor wrote:
> > the patch below fixes PR 49923 by checking for misaligned accesses
> > before doing IPA-SRA (on strict alignment targets). I have checked it
> > fixes the issue on compile farm sparc64 and I also included this in a
> > bootstra
On Wed, 10 Aug 2011, Ulrich Weigand wrote:
> Richard Guenther wrote:
> > On Fri, 5 Aug 2011, Martin Jambor wrote:
> > > the patch below fixes PR 49923 by checking for misaligned accesses
> > > before doing IPA-SRA (on strict alignment targets). I have checked it
> > > fixes the issue on compile f
On Wed, Aug 10, 2011 at 7:12 AM, Richard Guenther
wrote:
> On Wed, Aug 10, 2011 at 4:04 PM, H.J. Lu wrote:
>> On Wed, Aug 10, 2011 at 6:46 AM, Richard Guenther
>> wrote:
>>> On Sat, Mar 29, 2008 at 10:11 PM, H.J. Lu wrote:
This patch restores proper checking the third argument on blendpd a
On Tue, Aug 9, 2011 at 10:23 PM, Artem Shinkarov
wrote:
> Sorry, I didn't attach the patch itself.
> Here we go, in the attachment.
I have committed the patch after re-bootstrapping and testing it
on x86_64-unknown-linux-gnu with {,-m32}.
Richard.
>
> Artem.
>
On Wed, Aug 10, 2011 at 4:42 PM, H.J. Lu wrote:
> On Wed, Aug 10, 2011 at 7:12 AM, Richard Guenther
> wrote:
>> On Wed, Aug 10, 2011 at 4:04 PM, H.J. Lu wrote:
>>> On Wed, Aug 10, 2011 at 6:46 AM, Richard Guenther
>>> wrote:
On Sat, Mar 29, 2008 at 10:11 PM, H.J. Lu wrote:
> This patc
Hello,
Here is a new version of the patch. Changes from the previous version
(http://gcc.gnu.org/ml/gcc-patches/2011-07/msg02240.html):
- updated to trunk
- TODO_remove_unused_locals flag was removed from todo_flags_finish
of reassoc pass
Bootstrapped and checked on x86_64-linux.
Thanks,
Ilya
On 08/10/2011 08:35 AM, Richard Guenther wrote:
* call.c (build_over_call): Call memcpy unconditionally.
OK. Have you tested the MEM_REF patch?
Jason
This seems like a problem with the Solaris headers that should be
handled by fixincludes.
Jason
On Wed, 10 Aug 2011, Jason Merrill wrote:
> On 08/10/2011 08:35 AM, Richard Guenther wrote:
> > * call.c (build_over_call): Call memcpy unconditionally.
>
> OK. Have you tested the MEM_REF patch?
No, not yet. I'll throw it to testing now.
Richard.
2011/8/10 Tom Tromey :
>> "Jie" == Jie Liu writes:
>
> Jie> RTEMS does not have virtual memory management, so there is no error
> Jie> when access the 0 address on rtems.
> Jie> So 'str->length()' donot throw NPE and just return an meaningless value.
>
> If you compile the Java parts of the li
On Tue, Aug 2, 2011 at 10:09, Romain Geissler wrote:
> 2011-08-02 Romain Geissler
>
> * Makefile.in (PLUGIN_HEADERS): Add C_TREE_H.
OK.
Diego.
On Mon, Aug 8, 2011 at 07:39, Romain Geissler wrote:
> 2011/7/20 Diego Novillo :
>> On Mon, Jul 18, 2011 at 03:06, Romain Geissler
>> I will commit this patch shortly.
>>
>>
>> Diego.
>>
>
> Ping !
Romain, please send me a current patch against today's trunk.
Thanks. Diego.
On Wed, 10 Aug 2011, Rainer Orth wrote:
> Could you try the obvious patch? It's probably quicker than me
> recreating the setup.
Actually I think it will be quicker for you to do this test.
--
Joseph S. Myers
jos...@codesourcery.com
On Tue, Aug 9, 2011 at 20:52, Gabriel Charette wrote:
> An ICE in lto_streamer_cache_get is not very intuitive to point out
> the problem to the user in this case imo
Right.
> I'm thinking we should at least warn when some flags like this one
> differ and when that difference has an impact
On Wed, 10 Aug 2011, Richard Guenther wrote:
> On Wed, 10 Aug 2011, Jason Merrill wrote:
>
> > On 08/10/2011 08:35 AM, Richard Guenther wrote:
> > > * call.c (build_over_call): Call memcpy unconditionally.
> >
> > OK. Have you tested the MEM_REF patch?
>
> No, not yet. I'll throw it to test
Hi Guys,
This is a resend of a previously submitted patch:
http://gcc.gnu.org/ml/gcc-patches/2009-03/msg00363.html
This version of the patch has been adjusted to apply to today's sources,
but otherwise it remains the same.
The point of the patch is that the cmpstrnsi machine pattern should no
On 08/10/2011 01:42 PM, Rainer Orth wrote:
* Centralize the determination of PICFLAG. Currently, three libraries
inside the gcc tree are built PIC without libtool: libgcc, libiberty,
and libgnat/libgnarl.
libiberty/configure.ac has a hardcoded list of PICFLAG that could be
moved to
On 08/10/2011 03:56 PM, Rainer Orth wrote:
"Joseph S. Myers" writes:
This is strange: they copy explicitly goes into $(gcc_objdir): from
libgcc/Makefile.in:
install-unwind_h:
cp unwind.h $(gcc_objdir)/include/unwind.h
chmod a+r $(gcc_objdir)/include/unwind.h
For an in-tree
Paolo Bonzini writes:
>> True: it is called once per multilib.
>
> Just to doublecheck, are we sure that unwind.h is always the same?
Yep: it's unwind-generic.h for almost all targets, just a few arm
targets use config/arm/unwind-arm.h for all multilibs.
Rainer
--
> "Jie" == Jie Liu writes:
Jie> + *-*-rtems*)
Jie> + can_unwind_signal=no
Jie> + CHECKREFSPEC=-fcheck-references
Jie> + DIVIDESPEC=-fuse-divide-subroutine
Jie> + ;;
This part is OK with a ChangeLog entry.
Jie> +
Spurious newline addition.
Jie> But it does not work as we want, is ther
On 08/10/2011 06:05 PM, Rainer Orth wrote:
>> True: it is called once per multilib.
>
> Just to doublecheck, are we sure that unwind.h is always the same?
Yep: it's unwind-generic.h for almost all targets, just a few arm
targets use config/arm/unwind-arm.h for all multilibs.
Patch doing rm
On Wed, 10 Aug 2011, Jason Merrill wrote:
This seems like a problem with the Solaris headers that should be handled by
fixincludes.
The goal of the set of patches, apart from defining __cplusplus=199711L,
was to start using the Solaris headers properly, without setting weird
macros to ignore
On Wednesday 10 August 2011 17:05:08, Rainer Orth wrote:
> Paolo Bonzini writes:
>
> >> True: it is called once per multilib.
> >
> > Just to doublecheck, are we sure that unwind.h is always the same?
>
> Yep: it's unwind-generic.h for almost all targets, just a few arm
> targets use config/arm/
On Wed, 2011-08-10 at 11:19 +0200, Richard Guenther wrote:
> On Wed, Aug 10, 2011 at 10:48 AM, Pedro Alves wrote:
> >
> > which makes me wonder if the right fix isn't to change
> > libiberty internally to not rely on the alignment.
>
> I think that would be the best fix. It hardly can be a perfo
On 08/10/2011 09:52 AM, Steve Ellcey wrote:
> * md5.c (md5_read_ctx): Handle mis-aligned resbuf pointer.
Ok.
r~
Hi,
For the previous analysis "libjava patches for RTEMS"[1], there are 6
cases need to pay more attention. PR18699, TLtest, Thread_Interrupt,
Thread_Sleep_2, Throw_2 and bclink.
After add patch to bdwgc for RTEMS pthread support[2]:
PR18699PASS after Modify [A bug in boehm-
Here is what it takes, on top of Lawrence's first patch, to clear new test
errors on my end.
Not sure what fixed the common removal of
// { dg-bogus "unistd.h:1144:34: error: declaration of .* ctermid.* has a
different exception specifier" "" { xfail *-*-* } 0 }
// { dg-bogus "stdio.h:858:14: er
It didn't fix everything.
Potentially another commit conflict with Diego's trunk merge from yesterday?
I just sent out another patch that fixes everything that was left on
my end. Let me know if it works on your end.
Cheers,
Gab
On Tue, Aug 9, 2011 at 7:26 PM, Lawrence Crowl wrote:
> Let me kn
On Wed, Aug 10, 2011 at 7:12 AM, Rainer Orth
wrote:
> Jason Merrill writes:
>
>> On 08/09/2011 09:14 AM, Marc Glisse wrote:
>>> I don't think we should define the C++ 2011 value yet. In my opinion, we
>>> should wait until:
>>> 1) the standard is official
>>> 2) gcc implements most of it: people
On Wed, Aug 10, 2011 at 13:17, Gabriel Charette wrote:
> 2011-08-10 Gabriel Charette
>
> * g++.dg/pph/x5dynarray7.h: Remove 2 bogus errors.
> * g++.dg/pph/x6dynarray6.h: Remove 2 bogus errors.
> Add 2 bogus errors.
> * g++.dg/pph/x7dynarray5.cc: Remove 2 bogus error
Hi,
This seems like a problem with the Solaris headers that should be
handled by fixincludes.
The goal of the set of patches, apart from defining
__cplusplus=199711L, was to start using the Solaris headers properly,
without setting weird macros to ignore half of them or fixincluding
half of
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 08/10/11 09:49, Nick Clifton wrote:
> Hi Guys,
>
> This is a resend of a previously submitted patch:
>
> http://gcc.gnu.org/ml/gcc-patches/2009-03/msg00363.html
>
> This version of the patch has been adjusted to apply to today's
> sources, but ot
There was a bug where c_finish_options would create some builtins and assign
them source_locations in the linemap other than BUILTINS_LOCATION == 1.
Thus, when calling DECL_IS_BUILTIN to know if a decl is a builtin, some of them
would return false as they had a source_location other than BUILTIN
Tested with bootstrap and full regression testing on x64.
On Wed, Aug 10, 2011 at 11:22 AM, Gabriel Charette wrote:
> There was a bug where c_finish_options would create some builtins and assign
> them source_locations in the linemap other than BUILTINS_LOCATION == 1.
>
> Thus, when calling DECL
On Tue, Aug 9, 2011 at 2:42 PM, Kirill Yukhin wrote:
> Here is second stage patch.
> It introduces AVX2 option, define etc.
>
> ChangeLog entry:
>
> 2011-08-09 Kirill Yukhin
>
> * common/config/i386/i386-common.c (OPTION_MASK_ISA_AVX2_SET): New.
> (OPTION_MASK_ISA_FMA_UNSET): Upd
On Wed, Aug 10, 2011 at 9:39 PM, Uros Bizjak wrote:
> diff --git a/gcc/common/config/i386/i386-common.c
> b/gcc/common/config/i386/i386-common.c
> index 1fd33bd..1e0ca5e 100644
> --- a/gcc/common/config/i386/i386-common.c
> +++ b/gcc/common/config/i386/i386-common.c
> @@ -52,6 +52,8 @@ along with
On 08/03/2011 11:30 AM, Alexander Monakov wrote:
From: Dmitry Melnik
EXPR_SPEC is an indicator of the speculativeness of an expression (an
instruction or just an rhs), as it is incremented each time the expression is
moved up across a conditional branch. When merging expr attributes for
similar
On 08/03/2011 11:30 AM, Alexander Monakov wrote:
From: Dmitry Melnik
This patch prevents duplicating (as bookkeeping code) instructions that are
either volatile or recognized by cannot_copy_insn_p target hook (in addition
to already present restrictions). This avoids generating incorrect assemb
On 08/03/2011 11:30 AM, Alexander Monakov wrote:
From: Sergey Grechanik
This fixes one place where hard_regno_nregs is incorrectly guarded by
reload_completed (as if before reload all regs are pseudos).
2011-08-04 Sergey Grechanik
* sel-sched.c (verify_target_availability): Fix usage
On 08/03/2011 11:30 AM, Alexander Monakov wrote:
From: Sergey Grechanik
There are several places where bitmap_bit_p function is used to test if some
register is in the regset. We need to take into account the fact that
depending on mode, we need to test multiple hard regs.
2011-08-04 Sergey G
On 08/03/2011 11:30 AM, Alexander Monakov wrote:
From: Dmitry Melnik
This fixes a bug caused by trying to initialize BB_AV_SET of a newly generated
jump. It assumes that jump is only generated in new bb, while it can be
replaced in same BB by try_redirect_by_replacing_jump, if jump was conditio
On 08/03/2011 11:30 AM, Alexander Monakov wrote:
From: Sergey Grechanik
This patch fixes a problem when new jumps created in
sel_redirect_edge_and_branch_force could not get correct seqnos.
get_seqno_of_a_pred is renamed to get_seqno_for_a_jump.
Implementation-wise, it supports looking at multi
On 08/03/2011 11:30 AM, Alexander Monakov wrote:
From: Sergey Grechanik
Sometimes we need to be able to call sched_get_condition_with_rev from
selective scheduler for an instruction with zero luid (i.e. before h_d_i_d
had been extended). On such occasion, we need to bypass the caching and use
"
On 08/03/2011 11:30 AM, Alexander Monakov wrote:
From: Sergey Grechanik
This patch avoids changing speculative bits of a register use when it is moved
up above a speculation check where that register is used as the address
register. We should only call ds_full_merge when the producer (speculati
Hi,
the following patch sets BINFOs of prevailing but non-main-variant
types to what the main variant has, so that for a type t with a binfo
the following is always true:
TYPE_BINFO (t) == TYPE_BINFO (TYPE_MAIN_VARIANT (t))
as I I believe in non-LTO world it always is. Moreover, the BINFO
trees
Hi,
the patch below changes devirtualization to use BINFO_VTABLE instead
of BINFO_VIRTUALS. This is in fact slightly more cumbersome to use in
almost any respect but has one huge advantage, it uses _much_ less
memory in LTO, because the virtual tables are unified just like all
other VAR_DECLs and
Hi,
this patch moves the graphite code to the documented cloog library
interface and removes the use of old/deprecated interfaces. It ensures
that graphite will also compile flawless with future cloog.org versions.
* graphite-clast-to-gimple.c (new_clast_name_index): Store a copy
of the s
On 08/02/2011 04:46 PM, Sebastian Pop wrote:
Ping.
Could one of the configure maintainers review these changes?
Ping II.
Would anyone be so kind to review this?
The missing documentation changes are here
http://gcc.gnu.org/ml/gcc-patches/2011-07/msg02309.html
Cheers
Tobi
Thanks,
Sebastia
On 08/02/2011 04:47 PM, Sebastian Pop wrote:
Ping.
Could one of the configure maintainers review these changes?
Ping II.
This is needed for move to the official cloog.org library interface [1].
Thanks a lot for your time
Tobi
[1] http://gcc.gnu.org/ml/gcc-patches/2011-08/msg00976.html
Tha
On 08/02/2011 04:47 PM, Sebastian Pop wrote:
Ping.
Could one of the configure maintainers review these changes?
Thanks,
Sebastian
Ping II.
This is needed for move graphite to the official cloog.org library
interface [1].
Thanks a lot for your time
Tobi
[1] http://gcc.gnu.org/ml/gcc-patch
Hi Tobi,
The patch looks good modulo some formatting changes:
+free(c->free_name);
+ int length = strlen(e->name);
[...]
Please follow the GNU style: there should be a space between the
function name and the open parenthesis.
- dom = new_Cloog_Domain_from_ppl_Pointset_Powerset (PBB_DO
On Wed, Aug 10, 2011 at 10:08:54AM +0200, Richard Guenther wrote:
> Are the arrays all well-aligned in practice? Thus, would versioning the loop
> for all-good-alignment help?
I suspect yes on 64-bit, but no on 32-bit, due to malloc not returning 128-bit
aligned memory in 32-bit. It only returns
This patch adds new well-factored PPH template tests. In the process,
some old template tests get removed.
Enable the e and z tests for PPH files with unsharable headers.
Add xdiff markers where appropriate.
Compile tests with -fno-dwarf2-cfi-asm to reduce environmental differences.
Tests run o
This patch applies on top of http://codereview.appspot.com/4863041/
It re-implements the PPH streamer to use the new tree streaming API.
This allows us to handle the pickle cache directly in PPH, on top of
which I am implementing the external referencing for decls.
The big changes are in pph_writ
Hi,
This patch adds a check in vect_operation_fits_smaller_type () that a
widening statement has a stmt_vec_info, i.e., that it is a loop
statement.
Bootstrapped and tested on powerpc64-suse-linux.
Committed.
Ira
ChangeLog:
PR tree-optimization/50039
* tree-vect-patterns.c (vect_opera
81 matches
Mail list logo
