On Tue, 22 Aug 2017, Richard Biener wrote:
> On August 22, 2017 6:38:55 PM GMT+02:00, "Richard Earnshaw (lists)"
> wrote:
> >On 22/08/17 13:55, Kyrill Tkachov wrote:
> >> Hi Richard,
> >> [roping in more aarch64 maintainers]
> >>
> >> On 22/08/17 13:27, Richard Biener wrote:
> >>> On Tue, 22 Au
Mike Stump writes:
> On Aug 21, 2017, at 4:12 AM, Richard Sandiford
> wrote:
>>
>> Later patches will add support for frame sizes that are a run-time
>> invariant but not a compile-time constant.
>
> I'm assuming dwarf and C++ exceptions will remain working. :-)
Sure :-) The vector length is
On Wed, 23 Aug 2017, Richard Biener wrote:
> On Tue, 22 Aug 2017, Richard Biener wrote:
>
> > On August 22, 2017 6:38:55 PM GMT+02:00, "Richard Earnshaw (lists)"
> > wrote:
> > >On 22/08/17 13:55, Kyrill Tkachov wrote:
> > >> Hi Richard,
> > >> [roping in more aarch64 maintainers]
> > >>
> > >
On Wed, Aug 23, 2017 at 10:04 AM, Richard Biener wrote:
> On Wed, 23 Aug 2017, Richard Biener wrote:
>
>> On Tue, 22 Aug 2017, Richard Biener wrote:
>>
>> > On August 22, 2017 6:38:55 PM GMT+02:00, "Richard Earnshaw (lists)"
>> > wrote:
>> > >On 22/08/17 13:55, Kyrill Tkachov wrote:
>> > >> Hi R
On Tue, Aug 22, 2017 at 5:52 PM, Jeff Law wrote:
> On 08/18/2017 08:01 AM, Richard Biener wrote:
>> On Mon, Jul 31, 2017 at 7:35 AM, Jeff Law wrote:
>>> This patch introduces the stack clash protection options
>>>
>>> Changes since V2:
>>>
>>> Adds two new params. The first controls the size of
On 08/23/2017 05:17 AM, Sergio Durigan Junior wrote:
> The GCC patch improves the libcc1::compiler_triplet_regexp::find and
> libcp1::compiler_triplet_regexp::find methods by first trying to match
> the triplet in the compiler name and correctly discarding the triplet
> part of the regexp if the ma
On Wed, 23 Aug 2017, Uros Bizjak wrote:
> On Wed, Aug 23, 2017 at 10:04 AM, Richard Biener wrote:
> > On Wed, 23 Aug 2017, Richard Biener wrote:
> >
> >> On Tue, 22 Aug 2017, Richard Biener wrote:
> >>
> >> > On August 22, 2017 6:38:55 PM GMT+02:00, "Richard Earnshaw (lists)"
> >> > wrote:
> >>
Hi,
Commit r251104 introduced uses of SIGKILL and SIGQUIT which broke MinGW
build due to those not being defined. This proposed patch check for the
macros availability.
ChangeLog entry is as follows:
*** gcc/ChangeLog ***
2017-08-23 Thomas Preud'homme
* gcc.c (execute): Only test f
On Wed, Aug 23, 2017 at 11:22 AM, Richard Biener wrote:
>> > /* If the machine does not have a case insn that compares the bounds,
>> > Index: gcc/config/i386/i386.c
>> > ===
>> > --- gcc/config/i386/i386.c (revision 251275)
>>
On Wed, 23 Aug 2017, Uros Bizjak wrote:
> On Wed, Aug 23, 2017 at 11:22 AM, Richard Biener wrote:
> >> > /* If the machine does not have a case insn that compares the bounds,
> >> > Index: gcc/config/i386/i386.c
> >> > ===
> >> > --
On Wed, Aug 23, 2017 at 11:40 AM, Uros Bizjak wrote:
> On Wed, Aug 23, 2017 at 11:22 AM, Richard Biener wrote:
>>> > /* If the machine does not have a case insn that compares the bounds,
>>> > Index: gcc/config/i386/i386.c
>>> > ===
On Wed, 23 Aug 2017, Uros Bizjak wrote:
> On Wed, Aug 23, 2017 at 11:40 AM, Uros Bizjak wrote:
> > On Wed, Aug 23, 2017 at 11:22 AM, Richard Biener wrote:
> >>> > /* If the machine does not have a case insn that compares the bounds,
> >>> > Index: gcc/config/i386/i386.c
> >>> >
On Wed, Aug 23, 2017 at 11:55 AM, Richard Biener wrote:
> On Wed, 23 Aug 2017, Uros Bizjak wrote:
>
>> On Wed, Aug 23, 2017 at 11:40 AM, Uros Bizjak wrote:
>> > On Wed, Aug 23, 2017 at 11:22 AM, Richard Biener wrote:
>> >>> > /* If the machine does not have a case insn that compares the bounds,
On Wed, 23 Aug 2017, Uros Bizjak wrote:
> On Wed, Aug 23, 2017 at 11:55 AM, Richard Biener wrote:
> > On Wed, 23 Aug 2017, Uros Bizjak wrote:
> >
> >> On Wed, Aug 23, 2017 at 11:40 AM, Uros Bizjak wrote:
> >> > On Wed, Aug 23, 2017 at 11:22 AM, Richard Biener
> >> > wrote:
> >> >>> > /* If th
On Wed, 23 Aug 2017, Thomas Preudhomme wrote:
> Hi,
>
> Commit r251104 introduced uses of SIGKILL and SIGQUIT which broke MinGW
> build due to those not being defined. This proposed patch check for the
> macros availability.
>
> ChangeLog entry is as follows:
>
> *** gcc/ChangeLog ***
>
> 2017
On Tue, Aug 22, 2017 at 5:13 PM, Jeff Law wrote:
> This patch addresses some issues with conditional equivalences.
>
> First, it stops recording blindly recording conditional equivalences.
> Which leads to regressions...
>
> So for certain binary expressions (for example x - y), if we lookup the
>
On Tue, 22 Aug 2017, Tamar Christina wrote:
> Sorry, forgot to add some maintainers :)
>
> From: gcc-patches-ow...@gcc.gnu.org on behalf
> of Tamar Christina
> Sent: Tuesday, August 22, 2017 5:27:23 PM
> To: gcc-patches@gcc.gnu.org
> Cc: nd; l...@redhat.
On Tue, 22 Aug 2017, Tamar Christina wrote:
> Hi All,
>
> Since XORSIGN is a mid-end transformation I'm adding a generic test
> to check for targets that support it if the transformation has been
> carried out or not in tree.
>
> Regtested on aarch64-none-linux-gnu, arm-none-eabi
> and x86_64-pc
This patch uses HWI_COMPUTABLE_MODE_P (X) instead of
GET_MODE_PRECISION (X) <= HOST_BITS_PER_WIDE_INT in cases
where X also needs to be a scalar integer.
Tested on aarch64-linux-gnu and x86_64-linux-gnu, and by making sure
that there were no differences in testsuite assembly output for one
target
On Wed, Aug 23, 2017 at 12:18 PM, Richard Biener wrote:
> On Wed, 23 Aug 2017, Uros Bizjak wrote:
>
>> On Wed, Aug 23, 2017 at 11:55 AM, Richard Biener wrote:
>> > On Wed, 23 Aug 2017, Uros Bizjak wrote:
>> >
>> >> On Wed, Aug 23, 2017 at 11:40 AM, Uros Bizjak wrote:
>> >> > On Wed, Aug 23, 2017
This patch uses df_read_modify_subreg_p to check whether writing
to a subreg would preserve some of the existing contents.
This has the effect of putting more emphasis on the
REGMODE_NATURAL_SIZE-based definition of whether something can be
partially modified, instead of using UNITS_PER_WORD uncon
This patch makes use of the subreg_offset_from_lsb function
added by an earlier patch in the SVE series.
Tested on aarch64-linux-gnu and x86_64-linux-gnu, and by making sure
that there were no differences in testsuite assembly output for one
target per CPU. OK to install?
Richard
2017-08-23 R
This patch uses subreg_lowpart_offset in places that open-coded
the calculation. It also uses it in regcprop.c to test whether,
after a mode change, the first register in a multi-register group
is still the right one.
Tested on aarch64-linux-gnu and x86_64-linux-gnu, and by making sure
that there
This patch uses subreg_size_lowpart_offset in places that open-coded
the calculation. The reload use (and the LRA one that was based on it)
seemed to ignore the BYTES_BIG_ENDIAN != WORDS_BIG_ENDIAN case; it's not
obvious whether that was deliberate or an oversight.
Tested on aarch64-linux-gnu and
This patch uses byte_lowpart_offset in places that open-coded the
calculation.
Tested on aarch64-linux-gnu and x86_64-linux-gnu, and by making sure
that there were no differences in testsuite assembly output for one
target per CPU. OK to install?
Richard
2017-08-23 Richard Sandiford
This patch uses GET_MODE_UNIT_SIZE instead of GET_MODE_SIZE in
cases where, for compound modes, the mode of the scalar elements
is what matters. E.g. the choice between truncation and extension
is really based on the modes of the consistuent scalars rather
than the mode as a whole.
None of the ex
This patch is like the previous GET_MODE_UNIT_SIZE one,
but for bit rather than byte sizes.
Tested on aarch64-linux-gnu and x86_64-linux-gnu, and by making sure
that there were no differences in testsuite assembly output for one
target per CPU. OK to install?
Richard
2017-08-23 Richard Sandif
Ping?
Best regards,
Thomas
On 17/07/17 17:25, Thomas Preudhomme wrote:
My bad, found an off-by-one error in the sizing of bitmaps. Please find fixed
patch in attachment.
ChangeLog entry is unchanged:
*** gcc/ChangeLog ***
2017-06-13 Thomas Preud'homme
* config/arm/arm.c (arm_optio
This patch is like the earlier GET_MODE_UNIT_SIZE one,
but for precisions rather than sizes. There is one behavioural
change in expand_debug_expr: we shouldn't use lowpart subregs
for non-scalar truncations, since that would just reinterpret
some of the scalars and drop the rest. (This probably d
Ping?
Best regards,
Thomas
On 17/07/17 09:51, Thomas Preudhomme wrote:
Ping?
Best regards,
Thomas
On 12/07/17 14:31, Thomas Preudhomme wrote:
Coprocessor intrinsic tests in gcc.target/arm/acle test whether
__ARM_FEATURE_COPROC has the right bit defined before calling the
intrinsic. This al
On Tue, Aug 22, 2017 at 10:43 PM, Alexandre Oliva wrote:
> On Aug 21, 2017, Richard Biener wrote:
>
>> On Fri, Aug 18, 2017 at 11:20 PM, Alexandre Oliva wrote:
>
>>> Besides implementing these new features, the patch contains multiple
>>> fixes for -fcompare-debug errors detected at various opti
On Wed, Aug 23, 2017 at 12:29 AM, Alexandre Oliva wrote:
> On Aug 21, 2017, Richard Biener wrote:
>
>> +gno-statement-frontiers
>> +Common Driver RejectNegative Var(debug_nonbind_markers_p, 0) Init(2)
>> +Don't enforce progressive recommended breakpoint locations.
>> +
>> +gstatement-frontiers
>>
Hi,
it seems that there are many undocumented IA-32 builtins. This patch starts
small and documents the builtins corresponding to the basic instrinsics which
are declared in ia32intrin.h.
Tested with 'make doc', OK for all active branches?
2017-08-23 Eric Botcazou
* doc/extend.te
The following fixes PR81940 by allowing the case of missing early
debug due to compiling with -g0 and linking with -g when using LTO.
Committed as obvious.
If more intricate issues like this keep popping up another
solution is to record whether -g0 was in effect at compile-time
and override -g a
Ping
From: Tamar Christina
Sent: Thursday, August 17, 2017 11:49:51 AM
To: Richard Biener; Joseph Myers
Cc: Christophe Lyon; Markus Trippelsdorf; Jeff Law; GCC Patches; Wilco
Dijkstra; Michael Meissner; nd
Subject: RE: [PATCH][GCC][PATCHv3] Improve fpclassi
On Wed, 23 Aug 2017, Tamar Christina wrote:
> Ping
>
> From: Tamar Christina
> Sent: Thursday, August 17, 2017 11:49:51 AM
> To: Richard Biener; Joseph Myers
> Cc: Christophe Lyon; Markus Trippelsdorf; Jeff Law; GCC Patches; Wilco
> Dijkstra; Michael Meiss
On Wednesday, August 23 2017, Pedro Alves wrote:
> On 08/23/2017 05:17 AM, Sergio Durigan Junior wrote:
>> The GCC patch improves the libcc1::compiler_triplet_regexp::find and
>> libcp1::compiler_triplet_regexp::find methods by first trying to match
>> the triplet in the compiler name and correctl
On 08/23/2017 02:07 PM, Sergio Durigan Junior wrote:
> On Wednesday, August 23 2017, Pedro Alves wrote:
>
>> On 08/23/2017 05:17 AM, Sergio Durigan Junior wrote:
>>> The GCC patch improves the libcc1::compiler_triplet_regexp::find and
>>> libcp1::compiler_triplet_regexp::find methods by first try
On Wednesday, August 23 2017, Pedro Alves wrote:
> On 08/23/2017 02:07 PM, Sergio Durigan Junior wrote:
>> On Wednesday, August 23 2017, Pedro Alves wrote:
>>
>>> On 08/23/2017 05:17 AM, Sergio Durigan Junior wrote:
The GCC patch improves the libcc1::compiler_triplet_regexp::find and
li
On Wed, Aug 23, 2017 at 12:50 AM, Daniel Santos wrote:
> Changes to lib/target-supports.exp and documentation:
> * Add effective-targets avx512f and avx512f_runtime (needed for new
> tests).
> * Corrects bug in check_avx2_hw_available.
> * Adds documentation for effective-targets avx2, avx2_runt
On Wed, 23 Aug 2017, Uros Bizjak wrote:
> On Wed, Aug 23, 2017 at 12:18 PM, Richard Biener wrote:
> > On Wed, 23 Aug 2017, Uros Bizjak wrote:
> >
> >> On Wed, Aug 23, 2017 at 11:55 AM, Richard Biener wrote:
> >> > On Wed, 23 Aug 2017, Uros Bizjak wrote:
> >> >
> >> >> On Wed, Aug 23, 2017 at 11:
On Wed, Aug 23, 2017 at 12:34 AM, Daniel Santos wrote:
> I had to fix a few things for x32 compatibility and I this is ready
> now. H.J. tested on machine with avx512 (including x32) and I've tested
> both native x32 and normal x86_64 with m64, m32 and mx32 and all is
> well. I've made more chan
When enabling free-lang-data the lang_hooks.decl_printable_name hook
gets reset. The following patch reduces the noise from that by
using verbosity 1 (do not print arguments) from the places where
we dump the function name and then follow by arguments manually.
This makes dumping consistent with
Richard,
> When enabling free-lang-data the lang_hooks.decl_printable_name hook
> gets reset. The following patch reduces the noise from that by
> using verbosity 1 (do not print arguments) from the places where
> we dump the function name and then follow by arguments manually.
>
> This makes du
On Wed, 23 Aug 2017, Arnaud Charlet wrote:
> Richard,
>
> > When enabling free-lang-data the lang_hooks.decl_printable_name hook
> > gets reset. The following patch reduces the noise from that by
> > using verbosity 1 (do not print arguments) from the places where
> > we dump the function name a
> > You have typos in the a bove changelog BTW, the "b" and "s" should be
> > part of the filenames (e.g. gnat.dg/noinline2.adb: Likewise.)
>
> Ooops. Will fix.
>
> > I do not understand the changes in the ada testsuite: cana you explain why
> > you renames some variables?
>
> Because the scan
On Wed, 23 Aug 2017, Richard Biener wrote:
> On Wed, 23 Aug 2017, Arnaud Charlet wrote:
>
> > Richard,
> >
> > > When enabling free-lang-data the lang_hooks.decl_printable_name hook
> > > gets reset. The following patch reduces the noise from that by
> > > using verbosity 1 (do not print argume
Jackson Woodruff writes:
> diff --git a/gcc/config/aarch64/aarch64-simd.md
> b/gcc/config/aarch64/aarch64-simd.md
> index
> 74de9b8c89dd5e4e3d87504594c969de0e0128ce..ce1b981fc005edf48a401a456def2a37cf9d9022
> 100644
> --- a/gcc/config/aarch64/aarch64-simd.md
> +++ b/gcc/config/aarch64/aarch64-s
On 08/23/2017 04:37 AM, Richard Biener wrote:
> On Tue, Aug 22, 2017 at 5:13 PM, Jeff Law wrote:
>> However, there are still cases where we have a conditional equivalence
>> and the names have the same cost to compute. A greatly simplified
>> example can be found in gcc.dg/tree-ssa/20030922-2.c.
Richard Sandiford wrote:
>
> Sorry for only noticing now, but the call to aarch64_legitimate_address_p
> is asking whether the MEM itself is a legitimate LDP/STP address. Also,
> it might be better to pass false for strict_p, since this can be called
> before RA. So maybe:
>
>if (GET_CODE (op
On 08/22/2017 09:51 PM, Jason Merrill wrote:
The C and C++ front ends already have a diagnose_mismatched_attributes
function, which seems like the natural place to add more conflict checking.
There are a few major problems with handling attribute conflicts
in diagnose_mismatched_attributes.
Th
2017-08-23 Uros Bizjak
* config/i386/i386.opt: Remove unneeded Init(0) initializations.
(mstackrealign): Do not init to -1.
* config/i386/i386.c (ix86_option_override_internal):
Check opts_set, not opts when setting default value of
opts->x_ix86_force_align_arg_pointer.
Boo
Hi Richard,
Thanks for the feedback,
>
> I think you should split up your work a bit. The pieces from
> fold_builtin_* you remove and move to lowering that require no control flow
> like __builtin_isnan (x) -> x UNORD x should move to match.pd patterns (aka
> foldings that will then be applied
On 8/18/17 9:19 PM, Peter Bergner wrote:
> On 8/18/17 6:27 PM, Segher Boessenkool wrote:
>> On Thu, Aug 17, 2017 at 07:02:14PM -0500, Peter Bergner wrote:
>>> This is also broken in GCC 7, GCC 6 and GCC 5. Ok for those after this
>>> has been on trunk for a little while and assuming testing passes
On Tue, 2017-08-22 at 13:01 +0200, Andreas Schwab wrote:
> On Aug 18 2017, David Malcolm wrote:
>
> > gcc/c/ChangeLog:
> > * c-parser.c (c_parser_expr_list): Use c_expr::get_location ()
> > rather than peeking the location of the first token.
> > * c-tree.h (c_expr::get_location): New
Hello!
Recent changes caused bootstrap on alpha with:
/space/homedirs/uros/gcc-svn/trunk/gcc/dwarf2out.c: In function ‘void
init_sections_and_labels(bool)’:
/space/homedirs/uros/gcc-svn/trunk/gcc/dwarf2out.c:27181:1: error:
‘%ld’ directive writing between 1 and 10 bytes into a region of size 8
[-
On Wed, Aug 23, 2017 at 11:49:03AM +0100, Richard Sandiford wrote:
> This patch uses df_read_modify_subreg_p to check whether writing
> to a subreg would preserve some of the existing contents.
combine does not keep the DF info up-to-date -- but that is no
problem here, df_read_modify_subreg_p use
I have various patch kits under development that improve
how the C and C++ frontends handle mismatched types of
params at callsites.
For example, given:
extern int callee_1 (int one, const char *two, float three);
int test_1 (int first, int second, float third)
{
return callee_1 (first
On Wed, Aug 23, 2017 at 11:50:13AM +0100, Richard Sandiford wrote:
> This patch makes use of the subreg_offset_from_lsb function
> added by an earlier patch in the SVE series.
>
> Tested on aarch64-linux-gnu and x86_64-linux-gnu, and by making sure
> that there were no differences in testsuite ass
On August 23, 2017 5:49:56 PM GMT+02:00, Uros Bizjak wrote:
>Hello!
>
>Recent changes caused bootstrap on alpha with:
>
>/space/homedirs/uros/gcc-svn/trunk/gcc/dwarf2out.c: In function ‘void
>init_sections_and_labels(bool)’:
>/space/homedirs/uros/gcc-svn/trunk/gcc/dwarf2out.c:27181:1: error:
>‘%ld
On 08/23/2017 03:07 AM, Richard Biener wrote:
>>> Both -fstack-clash-protection and -fstack-check cannot be turned
>>> off per function. This means they would need merging in lto-wrapper.
>>> The alternative is to mark them with 'Optimization' and allow per-function
>>> specification (like we do
On August 23, 2017 6:36:32 PM GMT+02:00, Jeff Law wrote:
>On 08/23/2017 03:07 AM, Richard Biener wrote:
>
Both -fstack-clash-protection and -fstack-check cannot be turned
off per function. This means they would need merging in
>lto-wrapper.
The alternative is to mark them with 'Opt
Ping.
On 13 Aug 2017 Boris Kolpackov writes:
Hi,
I've been instructed to resend this patch from the gcc mailing list:
Currently GCC does not write extracted header dependency information
if there are errors. However, this can be useful when dealing with
outdated generated headers that trigger
Hi!
On Tue, Aug 22, 2017 at 10:48:17AM +, Wilco Dijkstra wrote:
> The register allocator inserts move preferences when an instruction has
> one or more dead sources in add_insn_allocno_copies. If an instruction
> doesn't have a matching constraint (eg. "0"), then any dead source is treated
> a
Segher Boessenkool wrote:
> On Tue, Aug 22, 2017 at 10:48:17AM +, Wilco Dijkstra wrote:
> > The register allocator inserts move preferences when an instruction has
> > one or more dead sources in add_insn_allocno_copies. If an instruction
> > doesn't have a matching constraint (eg. "0"), then a
Ping:
2017-08-15 Michael Meissner
PR libquadmath/81848
* configure.ac (powerpc*-linux*): Use attribute mode KC to create
complex __float128 on PowerPC instead of attribute mode TC.
* quadmth.h (__complex128): Likewise.
* configure: Regenerate.
*
In r251239 I added a c_expr::get_location method for use by
c_parser_expr_list for building the vec for
an expression list, rather than using the location of the first token.
When determining whether to use the location within the tree node,
or fall back to the range in the c_expr, I used EXPR_CAN
This patch eliminates the undocumented debugging options -mpower9-dform,
-mpower9-dform-scalar, and -mpower9-dform-vector. These switches were added
when I added the support for the ISA 3.0 (power9) d-form (register+offset)
vector addressing.
I have built and bootstrapped compilers on a little en
On 08/23/2017 05:17 AM, Sergio Durigan Junior wrote:
> Hi there,
>
> This is a series of two patches, one for GDB and one for GCC, which aims
> to improve the detection and handling of triplets present on compiler
> names. The motivation for this series was mostly the fact that GDB's
> "compile"
On Wednesday, August 23 2017, Pedro Alves wrote:
> On 08/23/2017 05:17 AM, Sergio Durigan Junior wrote:
>> Hi there,
>>
>> This is a series of two patches, one for GDB and one for GCC, which aims
>> to improve the detection and handling of triplets present on compiler
>> names. The motivation fo
Here's a patch from the name-lookup branch. Search stats had somewhat
bit rotted, with several of the counters never being incremented. Most
of the remaining counters won't make sense once I've completed
class-member reorganization.
This patch simply excises the search stats. We can always
This patch extends various call-handling locations within the C++
frontend so that they can accept a possibly NULL vec *,
using the location information if available in various diagnostics.
It updates the C++ frontend's parser for this production:
postfix-expression:
postfix-expression (
Here's another cleanup, applied to trunk.
Both decls_match and add_method had similar code to check for function
version declarations. The add_method one was particularly confused as
it checked for extern "C" methods -- which are not a thing.
Anyway, this does the obvious breaking out of a c
Bug 81908 is about a -Wstringop-overflow warning for a Fortran
test triggered by a recent VRP improvement. A simple test case
that approximates the warning is:
void f (char *d, const char *s, size_t n)
{
if (n > 0 && n <= SIZE_MAX / 2)
n = 0;
memcpy (d, s, n); // n in ~[1, S
Ping: https://gcc.gnu.org/ml/gcc-patches/2017-08/msg00912.html
Jeff, is this version good to commit or are there any other
changes you'd like to see?
Martin
On 08/14/2017 04:40 PM, Martin Sebor wrote:
On 08/10/2017 01:29 PM, Martin Sebor wrote:
diff --git a/gcc/builtins.c b/gcc/builtins.c
ind
And here's one for finish-member-declaration.
I move inhibiting of a USING_DECL's debug info out of a non-fn
member-specific block, and unconditionally set C++ linkage. Now that
Java linkage is no longer a thing, we were effectively doing:
if (bool_thing == false)
bool_thing = true;
Af
Hi Richard,
> On Fri, 12 May 2017, Richard Biener wrote:
>
>>
>> This is a heads-up that I am in the process of implementing the last
>> of Jasons review comments on the dwarf2out parts of early LTO debug
>> support. I hope to post final patches early next week after thoroughly
>> re-testing eve
On 08/23/2017 08:26 AM, Uros Bizjak wrote:
>> @@ -1822,6 +1845,7 @@ proc check_avx2_hw_available { } {
>> expr 0
>> } else {
>> check_runtime_nocache avx2_hw_available {
>> + #include
> Why is the above include needed?
It is only needed to #define N
this patch moves the direct class member lookup functions from search.c
to name-lookup.c. I'll be morphing them into a saner data representation.
These are the functions that look only in a specified class. The
hierarchy searching functions remain in search.c
This is a direct transplant wit
On Wed, Aug 23, 2017 at 05:15:03PM +, Wilco Dijkstra wrote:
> Segher Boessenkool wrote:
> > On Tue, Aug 22, 2017 at 10:48:17AM +, Wilco Dijkstra wrote:
> > > The register allocator inserts move preferences when an instruction has
> > > one or more dead sources in add_insn_allocno_copies. If
On 08/23/2017 01:12 AM, Uros Bizjak wrote:
> On Wed, Aug 23, 2017 at 7:23 AM, Daniel Santos
> wrote:
>> On 08/22/2017 03:00 PM, Uros Bizjak wrote:
>>> On Tue, Aug 22, 2017 at 9:47 PM, Daniel Santos
>>> wrote:
> Please add UNKNOWN_ABI to the enum and initialize -mabi in i386.opt to
> UNK
On 08/22/2017 06:48 AM, Wilco Dijkstra wrote:
Hi,
The register allocator inserts move preferences when an instruction has
one or more dead sources in add_insn_allocno_copies. If an instruction
doesn't have a matching constraint (eg. "0"), then any dead source is treated
as a copy with all dest
On 8/23/17 5:22 PM, Segher Boessenkool wrote:
> On Wed, Aug 23, 2017 at 05:15:03PM +, Wilco Dijkstra wrote:
>> What is a preferred alternative? The current register allocator simply
>> ignores
>> any combination of "0r", "r0", ("r", "0") and ("0", "r") and just picks the
>> most
>> generic al
On 07/13/2017 02:53 AM, Richard Sandiford wrote:
> The main loop of simplify_comparison starts with:
>
> if (GET_MODE_CLASS (mode) != MODE_INT
> && ! (mode == VOIDmode
> && (GET_CODE (op0) == COMPARE || COMPARISON_P (op0
> break;
>
> So VOIDmode is only
On 07/13/2017 02:54 AM, Richard Sandiford wrote:
> After previous patches, top-level calls to simplify_and_const_int
> always pass a scalar_int_mode. The only call site that needs to
> be updated is the recursive one in simplify_and_const_int_1,
> at which point we know "varop" has a scalar intege
On 07/13/2017 02:54 AM, Richard Sandiford wrote:
> This patch passes the mode of the shifted value to extract_left_shift
> and updates the only caller so that the mode is a scalar_int_mode.
>
> 2017-07-13 Richard Sandiford
> Alan Hayward
> David Sherwood
>
> gcc/
>
On 07/13/2017 02:54 AM, Richard Sandiford wrote:
> The comment for widest_int_mode said that it returns "the widest integer
> mode no wider than SIZE", but it actually returns the widest integer
> mode that is narrower than SIZE. In practice SIZE is always greater
> than 1, so it can always pick Q
On 07/13/2017 02:56 AM, Richard Sandiford wrote:
> When subregs contain promoted values, as indicated by
> SUBREG_PROMOTED_VAR_P, both the unpromoted (outer) and
> promoted (inner) values are known to be scalar integers.
> This patch adds helper routines that get the modes as
> scalar_int_modes.
>
On 07/13/2017 02:57 AM, Richard Sandiford wrote:
> The caller of const_scalar_mask_from_tree has proven that
> the mode is a MODE_INT, so this patch passes it down as a
> scalar_int_mode. It also expands the comment a little.
>
> 2017-07-13 Richard Sandiford
> Alan Hayward
>
On 07/13/2017 02:58 AM, Richard Sandiford wrote:
> This patch adjusts a couple of cases in which we had established
> that two modes were equal and happened to be using the one with the
> more general type instead of the one with the more specific type.
>
> 2017-07-13 Richard Sandiford
>
On 08/10/2017 05:24 AM, Alexander Monakov wrote:
> On Wed, 9 Aug 2017, Jeff Law wrote:
The _5th macro isn't that bad either, appart from using reserved namespace
identifiers (it really should be something like qsort_5th and the arguments
shouldn't start with underscores).
>>>
>>> I d
On 07/13/2017 02:55 AM, Richard Sandiford wrote:
> nonzero_bits1 assumed that all bits of a floating-point or vector mode
> were needed. It seems likely that fixed-point modes should have been
> handled in the same way. After excluding those, the only remaining
> modes that are likely to be usefu
On 07/13/2017 02:55 AM, Richard Sandiford wrote:
> Similarly to the nonzero_bits patch, this one moves the mode
> class check and VOIDmode handling from num_sign_bit_copies1
> to num_sign_bit_copies itself, then changes the subroutines
> to operate on scalar_int_modes.
>
> gcc/
> 2017-07-13 Richa
On 07/13/2017 02:55 AM, Richard Sandiford wrote:
> The two implementations of the reg_nonzero_bits and reg_num_sign_bits
> hooks ignored the "known_x", "known_mode" and "known_ret" arguments,
> so this patch removes them. It adds a new scalar_int_mode parameter
> that specifies the mode of "x". (
As discussed in the PR...
CCP is passing a precision of 0 to get_nonzero_bits for complex
numbers. With my last wide_int sign extension patch, the sign
extension of -1 with a precision of 0 returns 0, whereas it previously
was preturning all ones. The attached patch calls element_precision,
to a
95 matches
Mail list logo
