On 11/17/2017 08:53 PM, Jan Hubicka wrote:
Hi,
this patch makes static profile to be in range 0...2^30 rather than
0...1. This is safe now as profile-counts are taking care of
possible overflow when the profile ends up cummulating too high after
inlining.
MThere are two testcases that needs
Looks good to me (as in "pass all my tests").
Thanks,
Dominique
> Le 18 nov. 2017 à 14:02, Paul Richard Thomas
> a écrit :
>
> Dear Dominique,
>
> Please find attached a revised patch that I believe fixes the problem
> that you found. The changes are the additions in trans-decl.c.
>
> OK fo
On Mon, 13 Nov 2017, Richard Biener wrote:
On Sat, Nov 11, 2017 at 11:03 PM, Marc Glisse wrote:
Hello,
with undefined overflow, just because we know nothing about one of the
arguments of an addition doesn't mean we can't say something about the
result. We could constrain more the cases where
Hi,
this patch removes a semicolon after "do {} while (0)" in MCORE_EXPORT_NAME.
This allows the macro to be used in if-then-elses without curly braces.
Did mcore-unknown-elf build.
Committed as obvious.
Thanks,
- Tom
[mcore] Remove semicolon after do {} while (0) in MCORE_EXPORT_NAME
2017-1
Hi,
this patch removes a semicolon after "do {} while (0)" in
REGISTER_TARGET_PRAGMAS. This allows the macro to be used in
if-then-elses without curly braces.
Build for spu-unknown-elf.
Committed as obvious.
Thanks,
- Tom
[ft32, spu] Remove semicolon after do {} while (0) in REGISTER_TARGET
Hi,
this patch removes a semicolon after ASM_OUTPUT_CASE_END. This allows
the macro to be used in if-then-elses without curly braces.
Build visium-unknown-elf.
Committed as obvious.
Thanks,
- Tom
[visium] Remove semicolon after ASM_OUTPUT_CASE_END
2017-11-14 Tom de Vries
* config/visiu
Hi,
this patch removes a semicolon after "do {} while (0)" in
TARGET_OS_CPP_BUILTINS. This allows the macro to be used in
if-then-elses without curly braces.
Build arm-unknown-phoenix.
Committed as obvious.
Thanks,
- Tom
[phoenix] Remove semicolon after do {} while (0) in TARGET_OS_CPP_BUIL
Hi,
this patch removes a semicolon after "do {} while (0)" in
UPDATE_FS_FOR_CFA_GR. This allows the macro to be used in if-then-elses
without curly braces.
I don't manage to build alpha-dec-vms far enough to trigger this patch.
Can somebody with a proper setup build and test this patch, or a
On November 19, 2017 1:09:48 AM GMT+01:00, Tom de Vries
wrote:
>Hi,
>
>atm, any build for vms os will fail early in libgcc.
>
>dwarf2_debug_hooks.early_finish needs to be called before
>dwarf2_debug_hooks.finish.
>
>However, while vmsdbgout_finish calls dwarf2_debug_hooks.finish there's
>
>no vm
Hi,
this patch removes a semicolon after "do while (0)" in
FUNCTION_PROFILER. This allows the macro to be used in if-then-elses
without curly braces.
Build arc-unkown-elf as far as that goes atm (fails somewhere in libgcc).
Committed as trivial.
Thanks,
- Tom
[arc] Remove semicolon after do
Hi Nathan,
> The 82836 testcase fell out of creduce. In c++17 mode it fails horribly
> with missing return errors.
>
> Applying this fix, so it's valid in c++17. It still ICEs (in both 14 and
> 17 modes) with the 82836 fix removed.
unfortunately, the testcase has several more problems:
* It fa
Hi!
On Wed, 18 Oct 2017 14:48:13 +0200, Martin Liška wrote:
> This is second patch that addresses test-suite fallout. All these tests fail
> because -Wreturn-type is
> now on by default.
>
> Patch can bootstrap on ppc64le-redhat-linux and survives regression tests.
... but alters some LTO test
Hello,
Svante Signell, on lun. 06 nov. 2017 16:36:39 +0100, wrote:
> Another issue is that /proc/self/exe has to return an absolute path for the
> built program go-7 to execute properly. This is solved by a pending patch for
> glibc in Debian and will be available in the next upload of glibc-2.24.
Hi!
After r254437 "Instrument function exit with __builtin_unreachable in
C++" (assuming that I bisected that correctly), I'm seeing a number of
C++ "ifunc" test cases regress from PASS to UNSUPPORTED. This is because
of:
ifunc_available29518.c: In function 'void (* g())()':
ifunc_availa
Hi,
my understanding of the code is that trip_count is trying to estimate number
of iterations when counts are available and disable modulo scheduler for
loops with too few iterations.
We now have get_estimated_loop_iterations_int and get_max_loop_iterations_int
which keeps information detected ups
On 11/16/2017 05:15 PM, Jeff Law wrote:
On 11/13/2017 06:21 PM, Martin Sebor wrote:
Attached is an improved version that rather than hardcoding
the built-in functions to check uses the constness of each
built-in's char* argument as a trigger to do the checking.
This avoids the problem of the li
On 11/16/2017 05:15 PM, Jeff Law wrote:
On 11/13/2017 06:21 PM, Martin Sebor wrote:
Attached is an improved version that rather than hardcoding
the built-in functions to check uses the constness of each
built-in's char* argument as a trigger to do the checking.
This avoids the problem of the li
On 11/17/2017 12:47 AM, Jakub Jelinek wrote:
On Thu, Nov 16, 2017 at 05:15:20PM -0700, Jeff Law wrote:
+ tree_code code = gimple_assign_rhs_code (def);
+ if (code == ADDR_EXPR
+ || code == COMPONENT_REF
+ || code == VAR_DECL)
+
On Fri, Nov 17, 2017 at 09:45:35AM +, Thomas Preudhomme wrote:
> > Bootstrapped/regtested on {x86_64,i686,powerpc64le,powerpc64}-linux, ok for
> > trunk?
> >
> > The cases this patch can handle are less common than rhs_code INTEGER_CST
> > (stores of constants to adjacent memory) or MEM_REF (
The 11/17/2017 22:06, James Greenhalgh wrote:
> On Tue, Nov 14, 2017 at 03:54:56PM +, Tamar Christina wrote:
> > Hi All,
> >
> > Dot Product is intended to only be available for Armv8.2-a and newer.
> > While this restriction is reflected in the intrinsics, the patterns
> > themselves were mis
Hi,
this patch implements Richi's suggestion to not ICE when vectorization
costmodel is asked for cost of vector load/store that is not of vector
type.
Bootstrapped/regtsted x86_64-linux, comitted.
PR target/82713
* i386.c (ix86_builtin_vectorization_cost): Be ready for insane
Hi,
this patch avoids degeneration of inliner for function containing empty
volatile asm statement.
Bootstrapped/regtested x86_64-linux, comitted.
PR ipa/60243
* tree-inline.c (estimate_num_insns): Set to 1 at least.
Index: ../../gcc/tree-inline.c
==
Hi,
this patch fixes ICE with -fno-guess-branch-probability.
Bootstrapped/regtested x86_64-linux, comitted.
Honza
PR ipa/83001
* profile-count.c (profile_count::to_sreal_scale): Fix return value
for uninitialied counts.
Index: profile-count.c
Dear Dominique,
Committed to trunk as revision 254936.
Thanks for picking up the problem and for redoing the testing.
Regards
Paul
Hi,
this testcase triggers ICE because we try to inline into -O0 function.
Fixed thus.
Bootstrapped/regtested x86_64-linux, comitted.
Honza
* gcc.c-torture/compile/pr81360.c: New testcase.
* ipa-inline.c (can_inline_edge_p): Also check that caller is optimized
Index: testsuite/gc
On my local branch gcc.dg/torture/pr56349.c fails by sending GCC into an
infinite loop trying to simplify a self-referring statement. ie
something like
x_1 = x_1 + 10;
That, of course, shouldn't be happening in SSA form. After some digging
I've found the culprit.
Let's say we've got a PHI.
a_1
On 11/19/2017 03:41 AM, Marc Glisse wrote:
> On Mon, 13 Nov 2017, Richard Biener wrote:
>
>> On Sat, Nov 11, 2017 at 11:03 PM, Marc Glisse
>> wrote:
>>> Hello,
>>>
>>> with undefined overflow, just because we know nothing about one of the
>>> arguments of an addition doesn't mean we can't say som
Hello world,
the attached patch implements maxloc and minloc, a missing feature / bug
(now that we are shooting for f2003 compliance). I decided to do
everything on the library side, since I am more familiar with that
territory. I also suspect that any performance gain from inlining will
be less
On 11/18/2017 10:47 AM, Peter Bergner wrote:
> On 11/18/17 12:16 AM, Jeff Law wrote:
>> You might wander a bit and see if/how cse handles other similar
>> circumstances. For example (not (not (x)) (neg (neg x)) and (bswap
>> (bswap (x))
>
> I actually tried examples like that to see what CSE wou
On 11/19/2017 10:06 AM, Martin Sebor wrote:
>>> +/* If EXPR refers to a character array or pointer declared attribute
>>> + nonstring return a decl for that array or pointer and set *REF to
>>> + the referenced enclosing object or pointer. Otherwise returns
>>> + null. */
>> Generally we'd
On 11/18/2017 09:47 AM, Martin Sebor wrote:
> On 11/18/2017 12:53 AM, Jeff Law wrote:
>> On 11/17/2017 12:36 PM, Martin Sebor wrote:
>>> The attached patch enhances -Wstringop-overflow to detect more
>>> instances of buffer overflow at compile time by handling non-
>>> constant offsets into the des
I have seen it in the early IL but this late I couldn't come
up with a test case where such a loop would be necessary.
The COMPONENT_REF always refers to the member array. If you
can show me an example to test with I'll add the handling if
possible. There are cases where it isn't, such as in
This gets rid of the warning about the unrecognized option -nostdinc++ passed
to gnat1 during the build of the gnattools.
Tested on x86_64-suse-linux, applied on the mainline.
2017-11-20 Eric Botcazou
PR ada/83016
* gnatlink.adb (Process_Args): Accept multiple switches for -
On 10/23/2017 03:44 AM, Aldy Hernandez wrote:
>
>
>
> Forwarded Message
> Subject: [patch] implement generic debug() for vectors and hash sets
> Date: Mon, 16 Oct 2017 09:52:51 -0400
> From: Aldy Hernandez
> To: gcc-patches
>
> We have a generic mechanism for dumping types f
On 10/25/2017 11:54 PM, Michael Collison wrote:
> Ping. Original patch posted here:
>
> https://gcc.gnu.org/ml/gcc-patches/2017-10/msg01317.html
Is this still needed or was it addressed by the patch Jakub checked in here:
2017-11-01 Jakub Jelinek
PR rtl-optimization/82778
PR
Hi all,
After the usual considerable digging around I determined that for the test case
doing 100 writes to a string that we were allocating 100 newunit numbers
and the associated unit structures on the unit treap. Consequently, at the end
of the program, when all left open units are clean
On 10/24/2017 07:26 PM, co...@sdf.org wrote:
> On Fri, Oct 13, 2017 at 11:13:52AM -0600, Jeff Law wrote:
>> So we can't depend on patches that OpenBSD applies. What's important is
>> what is in the official GCC sources.
>>
>> I'd like to see some discussion about what these macros should look like
On 11/03/2017 05:51 AM, Eric Botcazou wrote:
> Hi,
>
> as explained in
> https://gcc.gnu.org/ml/gcc-patches/2017-11/msg00105.html
> we don't (necessarily) want to report non-executed blocks in Ada.
>
> The couple of attached patches prevent this from happening by detecting
> whether we are dea
On 11/18/2017 12:53 AM, Jeff Law wrote:
On 11/17/2017 12:36 PM, Martin Sebor wrote:
The attached patch enhances -Wstringop-overflow to detect more
instances of buffer overflow at compile time by handling non-
constant offsets into the destination object that are known to
be in some range. The s
On 11/15/2017 08:40 AM, Aaron Sawdey wrote:
> So, the story of this very small patch starts with me adding patterns
> for ppc instructions bdz[tf] and bdnz[tf] such as this:
>
> [(set (pc)
> (if_then_else
> (and
>(ne (match_operand:P 1 "register_operand" "c,*b,*b,*b")
>
On 11/09/2017 04:08 AM, Richard Sandiford wrote:
> Jeff Law writes:
>> On 10/25/2017 09:57 AM, Richard Sandiford wrote:
>>> We already allow the target to change the size and alignment of a mode.
>>> This patch does the same thing for the number of units, which is needed
>>> to give command-line c
On 11/03/2017 10:37 AM, Richard Sandiford wrote:
> extract_bit_field_1 tries to use vec_extract to extract part of a
> vector. However, if that pattern isn't defined or if the operands
> aren't suitable, another good approach is to try a direct subreg
> reference. This is particularly useful for
Hello,
new version, regtested on powerpc64le-unknown-linux-gnu. The front-end
parts are up for review.
2017-10-28 Marc Glisse
gcc/c/
* c-fold.c (c_fully_fold_internal): Handle POINTER_DIFF_EXPR.
* c-typeck.c (pointer_diff): Use POINTER_DIFF_EXPR.
gcc/c-family/
* c-
On 11/09/2017 06:24 AM, Richard Sandiford wrote:
> ...so that we can use them for variable-length vectors. For now
> constant-length vectors continue to use VEC_PERM_EXPR and the
> vec_perm_const optab even for cases that the new optabs could
> handle.
>
> The vector optabs are inconsistent about
On 11/17/2017 08:11 AM, Richard Sandiford wrote:
> Fully-masked loops can be profitable even if the iteration
> count is smaller than the vectorisation factor. In this case
> we're effectively doing a complete unroll followed by SLP.
>
> The documentation for min-vect-loop-bound says that the
> d
Sorry, it's taken so long to get back to this patch...
On 09/01/2017 02:51 AM, Tom de Vries wrote:
> On 08/31/2017 11:44 PM, Jeff Law wrote:
>> On 08/28/2017 12:26 PM, Tom de Vries wrote:
>>> Hi,
>>>
>>> I think I found a bug in r17465:
>>> ...
* cse.c (simplify_ternary_operation): H
On 09/06/2017 03:55 AM, Jackson Woodruff wrote:
> Hi all,
>
> A minor improvement came to mind while updating other parts of this patch.
>
> I've updated a testcase to make it more clear and a condition now uses a
> call to is_division_by rather than manually checking those conditions.
>
> Jacks
On 11/08/2017 08:14 AM, Richard Sandiford wrote:
> Supporting load/store lanes for variable-length vectors means that
> we use them instead of SLP (for which we can't yet handle external
> and constant definitions -- fixed by a later patch). Previously
> we'd fail to use load/store lanes too and f
>> 2017-11-17 Kito Cheng
>>
>> * longlong.h [__riscv] (__umulsidi3): Define.
>> [__riscv] (umul_ppmm) Likewise.
>> [__riscv] (__muluw3) Likewise.
>
>
> Apparently the point of this is that by defining __mulsi3/__muldi3 as an
> extended asm, we get better register allo
On Sun, Nov 19, 2017 at 7:34 PM, Kito Cheng wrote:
>
> I prefer write more comment in the code instead of ChangeLog, I add
> more comment in __muluw3.
> btw, long times ago, Vladimir was told me[1] it should contain what is
> done but not why it is done?
Yes, the ChangeLog entry only states what
Hi Jeff,
No longer needed. Jakub's patch does the job.
Michael Collison
> On Nov 19, 2017, at 3:50 PM, Jeff Law wrote:
>
>> On 10/25/2017 11:54 PM, Michael Collison wrote:
>> Ping. Original patch posted here:
>>
>> https://gcc.gnu.org/ml/gcc-patches/2017-10/msg01317.html
> Is this still need
Here is the latest version I plan to commit after validation.
I'd like to keep the change to the money_get/get/[char/wchar_t]/9.cc
tests. At the moment those tests passes only because there is no
operation on the is iterator (like a check for eof). If we add any,
internal _M_buf will be resete
On Sun, Nov 19, 2017 at 08:53:00PM -0800, Jim Wilson wrote:
> > 2017-11-24 Kito Cheng
> >
> > * longlong.h [__riscv] (__umulsidi3): Define.
> > [__riscv] (umul_ppmm) Likewise.
> > [__riscv] (__muluw3) Likewise.
BTW, two colons missing after )
Jakub
On Sun, Nov 19, 2017 at 09:06:19PM +0100, Jan Hubicka wrote:
> Hi,
> this testcase triggers ICE because we try to inline into -O0 function.
> Fixed thus.
>
> Bootstrapped/regtested x86_64-linux, comitted.
>
> Honza
>
> * gcc.c-torture/compile/pr81360.c: New testcase.
> * ipa-inline.c
Hi!
VEC_COND_EXPR is handled in constexpr code like COND_EXPR, but that is
wrong. VEC_COND_EXPR is more like an arbitrary arithmetics ternary
operation, we need to compute all 3 arguments and based on the elements of
the first argument pick up elements of 2nd and 3rd arguments.
The COND_EXPR hand
Hi!
I'd like to ping the following patches:
http://gcc.gnu.org/ml/gcc-patches/2017-10/msg01895.html
PR debug/82718
Fix DWARF5 .debug_loclist handling with hot/cold partitioning
http://gcc.gnu.org/ml/gcc-patches/2017-11/msg00851.html
C++2A P0428R2 - familiar template syntax for generic
56 matches
Mail list logo
