This test runs, not just compiles. Applied as obvious.
* gcc.target/powerpc/pr47755-2.c: Require vsx_hw.
Index: gcc/testsuite/gcc.target/powerpc/pr47755-2.c
===
--- gcc/testsuite/gcc.target/powerpc/pr47755-2.c(revisi
On Fri, Apr 15, 2011 at 1:20 PM, Michael Matz wrote:
> Hi,
>
> On Thu, 14 Apr 2011, Michael Matz wrote:
>
>> > Btw, I don't remember why I chose ALLOCA_FOR_VAR_P over
>> > CALL_ALLOCA_FOR_VAR_P but, given the name of the GIMPLE flag and
>> > predicate, it's probably time to change it.
>>
>> Good i
On Tue, May 10, 2011 at 8:12 PM, H.J. Lu wrote:
> On Sat, Apr 23, 2011 at 6:33 PM, Jan Hubicka wrote:
>> Hi,
>> the problems with LTO build reported by Toon is caused by a hack disabling
>> jump function in ipa-prop on wpa. The hack is no longer needed and it is
>> wrong,
>> since jump function
On Sat, Apr 23, 2011 at 6:33 PM, Jan Hubicka wrote:
> Hi,
> the problems with LTO build reported by Toon is caused by a hack disabling
> jump function in ipa-prop on wpa. The hack is no longer needed and it is
> wrong,
> since jump function makes no sense when they are not updated. Consequentely
We watched the level 2 this morning and believe that shorters are trying to
keep this company down! We wouldn't be surprise to see the excitement continues
throughout the week since this company is up to something massive from a
fundamental perspective! IDVC has released big news today and this
The following patch temporarily disable some of the checking which is
not fully 'lipo' aware. It will be checked into google/main and
further cleanups will follow.
David
2011-05-10 David Li
* cgraphunit.c (revision 173635) (verify_cgraph_node):
(cgraph_mark_functions_to_output):
* tree-cfg.
> OK for mainline?
Yes, this is pretty obvious. (You don't have to change to
prev_nondebug_insn btw).
Ayal.
From: Revital Eres
To: Ayal Zaks/Haifa/IBM@IBMIL
Cc: gcc-patches@gcc.gnu.org, Patch Tracking
Date: 08/05/2011 07:37 AM
Subject:[PATCH, SMS 2/3] Skip DEBUG_INSNs wh
Thanks Revital; it may actually be better to include the scheduling of the
branch even where the doloop pattern is simple, as it rightfully accounts
for its resources.
> OK for mainline?
Yes. I have only the following minor comments:
+/* Bump the SCHED_TIMEs of all nodes by AMOUNT. Set the valu
> -Original Message-
> From: Eric Botcazou [mailto:ebotca...@adacore.com]
> Sent: Tuesday, May 10, 2011 2:43 PM
> To: Denis Chertykov
> Cc: Weddington, Eric; gcc-patches@gcc.gnu.org; Anatoly Sokolov; Georg-
> Johann Lay
> Subject: Re: [AVR] Couple of tweaks
>
> > The port don't have to d
Hi,
On Tue, 10 May 2011, Nathan Froyd wrote:
> > > + /* Do not stream TYPE_POINTER_TO or TYPE_REFERENCE_TO. */
> >
> > Add some wording as to why not? This was copied from existing
> > comments, but I do not remember why we were doing this. Not too
> > critical, anyway.
>
> I'm not entirely
> The port don't have to disable subregs in legitimate_address_p.
They don't have not to disable them either, they do what's better for them.
> So, the current avr_legitimate_address_p is correct.
It's just making impossible to compile some code.
> May be it's GCC core bug ?
It's more of a rel
> Which compiler ?
The C compiler, see the first message in this thread...
--
Eric Botcazou
On 03/10/2011 11:23 PM, Nathan Froyd wrote:
After all that, we can finally make tree_exp inherit from typed_tree.
Quite anticlimatic.
Ping. http://gcc.gnu.org/ml/gcc-patches/2011-03/msg00559.html
-Nathan
On 05/09/2011 06:49 PM, Nathan Froyd wrote:
The patch below is an updated version of:
http://gcc.gnu.org/ml/libstdc++/2011-02/msg9.html
Sorry I didn't respond to that message.
In general, my preference is to have diagnostics collocated with the
tests that lead to them, and just run throu
On Tue, 10 May 2011, Bernd Schmidt wrote:
> * doc/invoke.texi (C6X Options): New section.
> * doc/md.texi (TI C6X family): New section.
> * config.gcc: Handle tic6x, in particular tic6x-*-elf and
> tic6x-*-uclinux.
> config/c6x/uclinux-elf.h: New file.
> config/
On Mon, May 9, 2011 at 17:58, Sandeep Soni wrote:
> This patch is just a slight modification to an earlier patch that I
> submitted which was non-working. This is working. It just recognizes
> variable declarations and might be helpful for ketaki to start with
> things. It is now more or less up
Honza,
This patch seems to have introduced approximately 150 new G++
testsuite failures on AIX. The errors all are verify_cgraph_node ICEs
related to thunks.
The all have a similar pattern:
/src/gcc/testsuite/g++.dg/abi/covariant2.C:31:20: error:
non-DECL_ONE_ONLY node in a same_comdat_group li
> "Nathan" == Nathan Froyd writes:
Nathan> gcc/java/
Nathan> * java-tree.h (TYPE_ARGUMENT_SIGNATURE): Use TYPE_MINVAL.
This is ok.
Tom
2011/5/10 Kai Tietz :
> 2011/5/10 Richard Guenther :
>> On Tue, May 10, 2011 at 5:30 PM, Paolo Bonzini wrote:
>>> On 05/10/2011 05:23 PM, Richard Guenther wrote:
I suppose you have testcases for all the cases you looked at, please
add some that cover these corner cases.
>>>
>>> Also
On Tue, May 3, 2011 at 9:40 AM, Easwaran Raman wrote:
> On Mon, May 2, 2011 at 8:37 PM, Jeff Law wrote:
>> -BEGIN PGP SIGNED MESSAGE-
>> Hash: SHA1
>>
>> On 04/26/11 16:06, Easwaran Raman wrote:
>>
>>>
You're right. The patch has correctness issues. It is not possible to
simply
On Fri, Mar 11, 2011 at 10:19, Richard Guenther
wrote:
> On Fri, Mar 11, 2011 at 5:23 AM, Nathan Froyd
> wrote:
>> Move CASE_CHAIN into a local operand for CASE_LABEL_EXPR. Nothing to
>> see here.
>
> I wonder if there isn't a better way to do this ... like always requiring
> operand 2 of SWITC
On Fri, Mar 11, 2011 at 02:19:14PM +0100, Richard Guenther wrote:
> On Fri, Mar 11, 2011 at 5:23 AM, Nathan Froyd
> wrote:
> > Move CASE_CHAIN into a local operand for CASE_LABEL_EXPR. Nothing to
> > see here.
>
> I wonder if there isn't a better way to do this ... like always requiring
> opera
Applied.
Jason
The introduction of = default means that now we need to separate
TYPE_NEEDS_CONSTRUCTING from the notion that we need to call a
constructor for non-aggregate initialization of a class. This patch
does that by introducing a new function type_build_ctor_call for the
latter case and adjusting cal
tsubst_copy_and_build was assuming that substituting into a pack
expansion would always produce a vector of elements, which is not true
if the template arguments are still dependent.
Tested x86_64-pc-linux-gnu, applied to trunk.
commit 3bc57a5e66cd7a4e6e9a77900249db0bf193456b
Author: Jason Merr
On Tue, May 10, 2011 at 02:28:06PM -0300, Diego Novillo wrote:
> On Tue, May 10, 2011 at 13:15, Nathan Froyd wrote:
> > Other types can of course be shrunk, but the memory savings from doing
> > so will be negligible
>
> Have you done any measurements on the potential savings?
Only back-of-the-e
On Tue, May 10, 2011 at 13:15, Nathan Froyd wrote:
> Other types can of course be shrunk, but the memory savings from doing
> so will be negligible
Have you done any measurements on the potential savings?
> +static void
> +lto_input_ts_type_common_tree_pointers (struct lto_input_block *ib,
> +
On Tue, May 10, 2011 at 1:25 PM, Richard Guenther wrote:
> The generated code is
> the same though, and still contains too many masking operations.
The lowering you're planning -- does that mean lowering all bitfield
ops for all targets? You've noticed "expand" makes some very
conservative assump
On 05/08/2011 03:39 PM, Uros Bizjak wrote:
Hello!
Attached patch fixes changed register allocation where "enabled"
attribute is used. The core of the problem was with IRA, where IRA
does not look at "enabled" attribute when scanning through
alternatives string to perform various tasks (including
On Tue, 10 May 2011, Bernd Schmidt wrote:
> On C6X, we have PARM_BOUNDARY == 8 (one byte), but some function
> argument slots still must be rounded to a larger value. As far as I
> could tell there's currently no way of telling gcc about this, hence a
> new target macro which controls this behavio
On Tue, 10 May 2011, Bernd Schmidt wrote:
> Index: doc/tm.texi.in
> ===
> --- doc/tm.texi.in.orig
> +++ doc/tm.texi.in
> @@ -6976,6 +6976,12 @@ if function is in @code{.text.name}, and
> otherwise.
> @end deftypefn
>
> +@hook TARG
On May 10, 2011, at 9:15 AM, Nathan Froyd wrote:
> The patch below implements the first step towards "tuplifying types":
> OK to commit?
Ok for Objective parts...
>
On Tue, 10 May 2011, Bernd Schmidt wrote:
> Binutils and uClibc already contain C6X patches; unfortunately there
> isn't an open-source sim.
The newlib support is also upstream.
> I've also run c6x tests, which look fairly clean on the whole.
Does this testing cover all six architecture variant
On May 10, 2011, at 8:26 AM, "Joseph S. Myers" wrote:
> On rerunning my target macro enumeration now it's no longer necessary
> to consider libgcc/config/ macros in host-side files
> OK to commit?
Ok.
>
2011/5/10 Richard Guenther :
> On Tue, May 10, 2011 at 5:30 PM, Paolo Bonzini wrote:
>> On 05/10/2011 05:23 PM, Richard Guenther wrote:
>>>
>>> I suppose you have testcases for all the cases you looked at, please
>>> add some that cover these corner cases.
>>
>> Also, there is quite some tree-vrp.
tree_type is a fairly big structure, and several of its fields are
overloaded. This means we waste some space, depending on the type, and
it's also somewhat confusing as to who uses what.
The patch below implements the first step towards "tuplifying types":
those pieces which all types use have b
On Tue, May 10, 2011 at 7:14 AM, H.J. Lu wrote:
> On Tue, May 10, 2011 at 2:52 AM, Jakub Jelinek wrote:
>> On Mon, May 09, 2011 at 01:35:39PM -0700, H.J. Lu wrote:
>>> (gdb) call debug_rtx (rtl)
>>> (var_location d (mem/s/j:SI (plus:DI (subreg:DI (ashift:SI
>>> (entry_value:SI (reg:SI 5 di [ i ])
This contains the testsuite changes for the C6X port.
Bernd
* gcc.target/tic6x/weak-call.c: New test.
* gcc.target/tic6x/fpcmp.c: New test.
* gcc.target/tic6x/fpdiv.c: New test.
* gcc.target/tic6x/rotdi16-scan.c: New test.
* gcc.target/tic6x/ffssi.c: New t
I've found it useful to use a construct such as the following:
(define_attr "units64"
"unknown,d,d_addr,l,m,s,dl,ds,dls,ls"
(const_string "unknown"))
(define_attr "units64p"
"unknown,d,d_addr,l,m,s,dl,ds,dls,ls"
(attr "units64"))
to define one attribute in terms of another by default, al
On C6X, we have PARM_BOUNDARY == 8 (one byte), but some function
argument slots still must be rounded to a larger value. As far as I
could tell there's currently no way of telling gcc about this, hence a
new target macro which controls this behaviour.
Bernd
* doc/tm.texi.in (FUNCTION_ARG
The C6X ELF ABI chooses to use ".const" rather than ".rodata" for
far-readonly data. This patch adds a target hook to deal with the problem.
Bernd
* doc/tm.texi.in (TARGET_ASM_MERGEABLE_RODATA_PREFIX): Document.
* doc/tm.texi: Regenerate.
* target.def (mergeable_rodata_pre
This is the second half of splitting TYPE_CANONICAL handling away
from the rest of the machinery. The following patch introduces
a simplified, non-SCC based hashing for TYPE_CANONICAL - it
still hashes things it shouldn't, but the patch is supposed to
be functional equivalent apart from the point
C6X has little- and big-endian modes, but in both modes, registers are
in little-endian order. This patch adds code to support this model in gcc.
Bernd
* reload.c (operands_match_p): Take it into account.
(reload_adjust_reg_for_mode): Likewise.
* rtlanal.c (subreg_get_info
I noticed that contrib/config-list.mk was missing from the new-port
checklist in doc/sourcebuild.texi. I've applied this patch to add it;
target triplets covering all config.gcc stanzas for a new port should be
added to config-list.mk so the port gets covered in all-targets build
tests (and, o
The scheduler knows that insns with different COND_EXEC conditions don't
conflict and can be scheduled independently. Unfortunately, sched-deps.c
does not try to keep the conditions valid as it progresses. For example,
[B0] A0 = [A1]
B0 = something
[!B0] [A2] = A0
The first and third insns h
On Tue, May 10, 2011 at 5:30 PM, Paolo Bonzini wrote:
> On 05/10/2011 05:23 PM, Richard Guenther wrote:
>>
>> I suppose you have testcases for all the cases you looked at, please
>> add some that cover these corner cases.
>
> Also, there is quite some tree-vrp.c dead code with these changes. Remov
On C6X, every jump instruction has 5 delay slots which can be filled
with normally scheduled instructions. With an issue width of 8
insns/cycle, this means that up to 40 insns can be issued after the jump
insn before the jump's side-effect takes place. I didn't particularaly
feel like using reorg.c
On 05/10/2011 05:23 PM, Richard Guenther wrote:
I suppose you have testcases for all the cases you looked at, please
add some that cover these corner cases.
Also, there is quite some tree-vrp.c dead code with these changes.
Removing the TRUTH_*_CODE handling in VRP will help finding more place
On rerunning my target macro enumeration now it's no longer necessary
to consider libgcc/config/ macros in host-side files, I saw that I'd
missed one remaining DECLARE_LIBRARY_RENAMES definition when poisoning
that macro host-side because of the move to libgcc/config/. Nathan
Froyd had moved all d
Jeff already approved this one a while ago. I haven't applied it yet
because so far it hasn't been necessary, but it will help for the
backtracking scheduler.
Bernd
* sched-int.h (struct _dep): Add member cost.
(DEP_COST, UNKNOWN_DEP_COST): New macros.
* sched-deps.c (init
This moves some local variables into a structure, in preparation for a
later patch which adds the capability to save/restore scheduler state.
Bernd
* haifa-sched.c (struct sched_block_state): New.
(schedule_block): Move some local variables into such a structure.
Index: gcc/haif
On Tue, May 10, 2011 at 5:08 PM, Kai Tietz wrote:
> 2011/5/10 Richard Guenther :
>> On Tue, May 10, 2011 at 3:56 PM, Kai Tietz wrote:
>>> Hello,
>>>
>>> this patch converts TRUTH_AND_EXPR, TRUTH_OR_EXPR, and TRUTH_XOR_EXPR
>>> expressions
>>> on gimplification to their binary form. Additionally
There is still some code duplication in schedule_block, and the inner
scheduling loop is somewhat convoluted. It would be easier to follow if
the sequence of operations was
1. sort the ready list
2. choose an instruction
3. schedule it
The following patch changes the loop to use that structure. I
I will post a series of patches as replies to this message which add a
port for TI's C6X processors to gcc. This involves a number of scheduler
changes and a few minor new target hooks and macros.
Some pieces are still missing in this submission. Paul Brook has
implemented exception handling suppo
2011/5/10 Richard Guenther :
> On Tue, May 10, 2011 at 3:56 PM, Kai Tietz wrote:
>> Hello,
>>
>> this patch converts TRUTH_AND_EXPR, TRUTH_OR_EXPR, and TRUTH_XOR_EXPR
>> expressions
>> on gimplification to their binary form. Additionally it takes care
>> that conditions
>> are getting boolified f
2011/5/10 Weddington, Eric :
>
>
>> -Original Message-
>> From: Denis Chertykov [mailto:cherty...@gmail.com]
>> Sent: Tuesday, May 10, 2011 5:42 AM
>> To: Eric Botcazou
>> Cc: Weddington, Eric; gcc-patches@gcc.gnu.org; Anatoly Sokolov; Georg-
>> Johann Lay
>> Subject: Re: [AVR] Couple of tw
On Tue, May 10, 2011 at 4:24 PM, Paul Pluzhnikov wrote:
> On Tue, May 10, 2011 at 7:02 AM, Ian Lance Taylor wrote:
>> Richard Guenther writes:
>
>>> I wonder why this is a GCC specific patch and not a linker patch. Why
>>> not change the linker(s) to accept such configure option that changes it
On Tue, 10 May 2011, Nathan Froyd wrote:
> gcc/c-family/
> * c-common.c (def_fn_type): Don't call build_function_type, call
> build_function_type_vec or build_varargs_function_type_vec instead.
> (c_common_nodes_and_builtins): Likewise.
The c-family changes are OK.
--
Joseph S
On Tue, May 10, 2011 at 3:56 PM, Kai Tietz wrote:
> Hello,
>
> this patch converts TRUTH_AND_EXPR, TRUTH_OR_EXPR, and TRUTH_XOR_EXPR
> expressions
> on gimplification to their binary form. Additionally it takes care
> that conditions
> are getting boolified for operation.
>
> ChangeLog
>
> 2011-0
On Tue, May 10, 2011 at 7:02 AM, Ian Lance Taylor wrote:
> Richard Guenther writes:
>> I wonder why this is a GCC specific patch and not a linker patch. Why
>> not change the linker(s) to accept such configure option that changes its
>> default behavior?
>
> It is traditionally gcc which tells
On Tue, 10 May 2011, Richard Guenther wrote:
> On Tue, 10 May 2011, Michael Matz wrote:
>
> > Hi,
> >
> > On Tue, 10 May 2011, Richard Guenther wrote:
> >
> > > struct B {
> > > unsigned bit0 : 1;
> > > unsigned bit1 : 1;
> > > };
> > >
> > > void
> > > foo (struct B *b)
> > > {
> > > b-
On Tue, May 10, 2011 at 10:14, Nathan Froyd wrote:
> gcc/c-family/
> * c-common.c (def_fn_type): Don't call build_function_type, call
> build_function_type_vec or build_varargs_function_type_vec instead.
> (c_common_nodes_and_builtins): Likewise.
>
> gcc/lto/
> * lto-l
On Tue, May 10, 2011 at 2:52 AM, Jakub Jelinek wrote:
> On Mon, May 09, 2011 at 01:35:39PM -0700, H.J. Lu wrote:
>> (gdb) call debug_rtx (rtl)
>> (var_location d (mem/s/j:SI (plus:DI (subreg:DI (ashift:SI
>> (entry_value:SI (reg:SI 5 di [ i ]))
>> (const_int 2 [0x2])) 0)
>>
On Tue, May 10, 2011 at 07:02:44AM -0700, Ian Lance Taylor wrote:
> > Otherwise if people link with ld they suddenly get different hash-style.
> > That looks wrong to me.
>
> That turns out not to be the case. Both gold and GNU ld accept the same
> set of --hash-style options.
And we keep tellin
On Tue, 10 May 2011, Michael Matz wrote:
> Hi,
>
> On Tue, 10 May 2011, Richard Guenther wrote:
>
> > struct B {
> > unsigned bit0 : 1;
> > unsigned bit1 : 1;
> > };
> >
> > void
> > foo (struct B *b)
> > {
> > b->bit0 = b->bit0 | b->bit1;
> > }
> >
> > we with this patch generate
> >
>
Richard Guenther writes:
> On Mon, May 9, 2011 at 7:11 PM, Paul Pluzhnikov
> wrote:
>> On Mon, May 9, 2011 at 9:51 AM, Joseph S. Myers
>> wrote:
>>> On Mon, 9 May 2011, Paolo Bonzini wrote:
>>>
Uhm, so we deadlocked, I thought the other way. I cannot really
express any opinion abou
On Mon, May 9, 2011 at 6:17 PM, Joseph S. Myers wrote:
> This patch makes the handling of the rs6000 -mcpu= and -mtune= options
> use Enum, completing the conversion of rs6000_handle_option - and so
> of all target option handlers - to avoid using modifiable global
> state.
>
> As with various oth
Hello,
this patch converts TRUTH_AND_EXPR, TRUTH_OR_EXPR, and TRUTH_XOR_EXPR
expressions
on gimplification to their binary form. Additionally it takes care
that conditions
are getting boolified for operation.
ChangeLog
2011-05-10 Kai Tietz
* gimplify.c (gimplify_exit_expr): Boolify co
Ping for:
http://gcc.gnu.org/ml/gcc-patches/2011-04/msg01327.html
It fixes the expansion of multiword operations in cases where the
suggested target is a hard register and where CANNOT_CHANGE_MODE_CLASS
forbids word-mode subparts.
Richard
"Joseph S. Myers" writes:
> 2011-05-07 Joseph Myers
>
> * config.gcc (libgcc_tm_file): Define instead of including files
> from ../../libgcc/config/ in tm_file.
> * configure.ac (libgcc_tm_file_list, libgcc_tm_include_list):
> Define.
> * configure: Regenerate.
>
As $SUBJECT suggests. Nothing much to see here.
Tested on x86_64-unknown-linux-gnu. OK to commit?
-Nathan
gcc/java/
* expr.c (build_jni_stub): Don't call build_function_type; call
build_function_type_vec instead.
* typeck.c (parse_signature_string): Likewise.
diff --gi
As $SUBJECT suggests. Rather pleasant how easy this was.
Tested on x86_64-unknown-linux-gnu. OK to commit?
-Nathan
gcc/ada/
* gcc-interface/utils.c (def_fn_type): Don't call build_function_type;
call build_function_type_vec or build_varargs_function_type_vec
instead.
As $SUBJECT suggests. Eradicating build_function_type from the C FE
entirely will take a bit more work.
The lto change is included because def_fn_type there is clearly a copy
of the C version; I am going to assume that given other approvals of
copy-paste code, this one is obvious.
Tested on x86_
On Mon, 9 May 2011 18:01:12 +0100
Julian Brown wrote:
> How does this look now? (Re-testing in progress.)
Results from re-testing are fine, btw.
Cheers,
Julian
> -Original Message-
> From: Denis Chertykov [mailto:cherty...@gmail.com]
> Sent: Tuesday, May 10, 2011 5:42 AM
> To: Eric Botcazou
> Cc: Weddington, Eric; gcc-patches@gcc.gnu.org; Anatoly Sokolov; Georg-
> Johann Lay
> Subject: Re: [AVR] Couple of tweaks
>
> 2011/5/8 Eric Botcazou :
> >
Hi,
On Tue, 10 May 2011, Richard Guenther wrote:
> struct B {
> unsigned bit0 : 1;
> unsigned bit1 : 1;
> };
>
> void
> foo (struct B *b)
> {
> b->bit0 = b->bit0 | b->bit1;
> }
>
> we with this patch generate
>
> D.2686_2 = b_1(D)->bit0;
> D.2688_4 = b_1(D)->bit1;
> D.2693_10 = D.2
This splits the TYPE_CANONICAL merging machinery from the regular
merging machinery. It just gets too confusing to share code-paths
as both cases require opposing conservatism (merge everything for
TYPE_CANONICAL, merge nothing for regular merging).
Bootstrapped and tested on x86_64-unknown-linu
2011/5/6 Georg-Johann Lay :
> Nathan Froyd schrieb:
>> On Mon, May 02, 2011 at 05:23:48PM +0200, Georg-Johann Lay wrote:
>>> PR45099 is an extension that gives an error when a fixed register is
>>> needed to pass a parameter to a function.
>>>
>>> Because the program will show malfunction when such
2011/5/8 Eric Botcazou :
>> Have you run the regression test suite for the AVR for this patch?
>
> The compiler doesn't even build without the print_operand_address hunk...
Which compiler ?
Denis.
2011/5/9 Georg-Johann Lay :
> Fixed the build warnings mentioned in PR48896.
>
> Johann
>
> --
>
> 2011-05-09 Georg-Johann Lay
>
> PR target/48896
> * config/avr/avr.c (avr_ret_register): Return unsigned int
> instead of int.
> (avr_function_value): Mark fn_decl_or_ty
On Tue, 10 May 2011, Nathan Sidwell wrote:
> +@item -mtls-dialect=@var{dialect}
> +@opindex mtls-dialect
> +Specify the dialect to use for accessing thread local storage. Two
> +dialects are supported - @option{arm} and @option{gnu}. The
Use "---" (em dash, unspaced) instead of " - ".
> +@opti
This patch makes us apply operand conversions of bitwise binary
operations to the operation result instead, which in turn allows
us to combine those conversions which helps bitfield related
operations. For
struct B {
unsigned bit0 : 1;
unsigned bit1 : 1;
};
void
foo (struct B *b)
{
b->bit
Hi Guys,
I am applying the patch below to improve the rx_memory_move_cost
function in the RX backend. This patch is already in the 4.5 branch
and mainline, but was omitted from the 4.6 branch.
Cheers
Nick
gcc/ChangeLog
2011-05-10 Nick Clifton
* config/rx/rx.c (rx_memory_move
Hi Guys,
I am applying the patch below to the mainline and 4.6 branch. It was
created by DJ Delorie to fix a performance regression with the RX port
of GCC, compared to the 4.5 branch. The problem was that jumps and
loops were not being aligned to an 8 byte boundary which can result in
This moves fold-consts conversion folding to forwprop, our tree
pattern matcher & optimizer. In addition to what fold-const
does this makes us canonicalize unsigned truncations to use
bitwise and of a mask, this form is more readily handled by
passes like VRP and this also makes it easier to comb
On Mon, May 09, 2011 at 01:35:39PM -0700, H.J. Lu wrote:
> (gdb) call debug_rtx (rtl)
> (var_location d (mem/s/j:SI (plus:DI (subreg:DI (ashift:SI
> (entry_value:SI (reg:SI 5 di [ i ]))
> (const_int 2 [0x2])) 0)
> (symbol_ref:DI ("a") )) [0 a S4 A32]))
> (gdb)
Ugh, paradoxi
On Mon, May 9, 2011 at 8:54 PM, Jakub Jelinek wrote:
> On Mon, May 09, 2011 at 01:41:13PM -0500, Aldy Hernandez wrote:
>> Jakub also gave me a testcase which triggered a buglet in
>> max_field_size. I have now added a parameter INNERDECL which is the
>> inner reference, so we can properly determi
On Mon, May 9, 2011 at 8:54 PM, Joseph S. Myers wrote:
> On Mon, 9 May 2011, H.J. Lu wrote:
>
>> > Index: lib/target-supports.exp
>> > ===
>> > --- lib/target-supports.exp (revision 173569)
>> > +++ lib/target-supports.exp (wo
On Mon, May 9, 2011 at 7:11 PM, Paul Pluzhnikov wrote:
> On Mon, May 9, 2011 at 9:51 AM, Joseph S. Myers
> wrote:
>> On Mon, 9 May 2011, Paolo Bonzini wrote:
>>
>>> Uhm, so we deadlocked, I thought the other way. I cannot really
>>> express any opinion about the desirability of the feature, but
Seems this didn't go to the list, so for the record...
On 8 May 2011 16:36, Richard Sandiford wrote:
> Eric Christopher writes:
>> No objection here.
>
> Same here. This looks great, thanks.
>
> Richard
At Tue, 10 May 2011 08:46:02 +0200,
Jakub Jelinek wrote:
> Hi!
> Just small style nits:
> s/int/bool/;s/0/false/;s/1/true/
> And the above condition is short enough that it could be on one line.
Ok. Tests still pass.
diff --git a/gcc/cp/class.c b/gcc/cp/class.c
index 12db2bc..94a339d 100644
--- a
91 matches
Mail list logo
