Priority: P3
Component: inline-asm
Assignee: unassigned at gcc dot gnu.org
Reporter: amodra at gmail dot com
Target Milestone: ---
gcc doesn't have a simple way to say that a pointer passed to an inline asm is
used to address an array. "m" (*p) unfortuna
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81890
Alan Modra changed:
What|Removed |Added
Keywords||documentation
Status|UNCONFIRME
|ASSIGNED
URL||https://gcc.gnu.org/ml/gcc-
||patches/2017-08/msg01174.ht
||ml
Assignee|unassigned at gcc dot gnu.org |amodra at gmail
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81747
Alan Modra changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81170
Alan Modra changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
||amodra at gmail dot com
Resolution|--- |FIXED
Target Milestone|7.3 |7.2
--- Comment #16 from Alan Modra ---
Fixed.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81298
Bug 81298 depends on bug 81295, which changed state.
Bug 81295 Summary: bootstrap broken on powerpc-linux-gnu with
--enable-default-pie
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81295
What|Removed |Added
---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63281
Alan Modra changed:
What|Removed |Added
Attachment #33503|0 |1
is obsolete|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63281
--- Comment #8 from Alan Modra ---
Created attachment 42187
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=42187&action=edit
[RS6000] Address cost
Somewhat related, costing constants properly also needs a proper cost to
loading from memory
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81996
Alan Modra changed:
What|Removed |Added
Summary|__builtin_return_address(0) |powerpc
|does not work on
gcc dot gnu.org |amodra at gmail dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81996
Alan Modra changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80479
Alan Modra changed:
What|Removed |Added
CC||amodra at gmail dot com
--- Comment #9
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79286
Alan Modra changed:
What|Removed |Added
URL|https://gcc.gnu.org/ml/gcc- |https://gcc.gnu.org/ml/gcc-
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79801
Alan Modra changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|---
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: amodra at gmail dot com
Target Milestone: ---
Configuring any powerpc linux target with --enable-default-pie gives a broken
compiler that uses crt1.o, crtbegin.o and crtend.o to
||2017-06-22
CC||doko at ubuntu dot com,
||wschmidt at gcc dot gnu.org
Assignee|unassigned at gcc dot gnu.org |amodra at gmail dot com
Ever confirmed|0
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67328
Alan Modra changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81170
Alan Modra changed:
What|Removed |Added
Keywords||patch
Priority|P3
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81193
Alan Modra changed:
What|Removed |Added
CC||amodra at gmail dot com
--- Comment #3
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81193
--- Comment #6 from Alan Modra ---
> Alternatively, we would need to do two passes. On the first pass resolve all
> of the normal functions and data locations, setting the location of ifunc
> functions in the TOC to be an error function. The se
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78614
Alan Modra changed:
What|Removed |Added
CC||amodra at gmail dot com
--- Comment #12
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78614
--- Comment #14 from Alan Modra ---
I meant the last one attached, which I see is the same as the first one. ie.
just use copy_rtx and don't worry about extra rtl.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78614
--- Comment #16 from Alan Modra ---
The patch referenced in comment #15 alone, ie. without
https://gcc.gnu.org/ml/gcc-patches/2016-11/msg03117.html, results in an
--enable-checking=yes,rtl bootstrap failure on powerpc64le-linux.
/home/amodra/src
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78614
--- Comment #17 from Alan Modra ---
I see the same sort of invalid rtl sharing errors on powerpc64le with both
https://gcc.gnu.org/ml/gcc-patches/2016-11/msg03117.html and
https://gcc.gnu.org/ml/gcc-patches/2016-11/msg03116.html applied.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78614
--- Comment #18 from Alan Modra ---
Ditto for just https://gcc.gnu.org/ml/gcc-patches/2016-11/msg03117.html
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78614
--- Comment #19 from Alan Modra ---
Cancel those last three comments. I wasn't actually testing with Jakub's
rs6000.c patch.
Everything now looks good with Jakub's rs6000.c and simplify-rtx.c patch, plus
the rs6000.c insn_is_swappable_p patch I
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=72749
Alan Modra changed:
What|Removed |Added
CC||amodra at gmail dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=72749
--- Comment #6 from Alan Modra ---
OK, currently testing TARGET_LEGITIMATE_COMBINED_INSN solution. Thanks for the
analysis!
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79066
Alan Modra changed:
What|Removed |Added
CC||amodra at gmail dot com
--- Comment #4
gcc dot gnu.org |amodra at gmail dot com
--- Comment #6 from Alan Modra ---
The problem happens in lra-contraints.c:process_address_1. Search for
HAVE_lo_sum. A high/lo_sum address is correctly rejected by the rs6000 backend
as improper for V4 pic here:
*ad.inner
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=72749
Alan Modra changed:
What|Removed |Added
CC|amodra at gcc dot gnu.org |
--- Comment #12 from Alan Modra --
gcc dot gnu.org |amodra at gmail dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79098
Alan Modra changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79066
Alan Modra changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
Assignee: unassigned at gcc dot gnu.org
Reporter: amodra at gmail dot com
Target Milestone: ---
Created attachment 40541
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=40541&action=edit
strncmp testcase
glibc built with current gcc fails elf/check-localplt due to str
||2017-01-19
Assignee|unassigned at gcc dot gnu.org |amodra at gmail dot com
Ever confirmed|0 |1
--- Comment #1 from Alan Modra ---
Testing a fix
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79144
Alan Modra changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
||amodra at gmail dot com
Assignee|unassigned at gcc dot gnu.org |amodra at gcc dot
gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79286
Alan Modra changed:
What|Removed |Added
CC||amodra at gmail dot com
Summary
Assignee: unassigned at gcc dot gnu.org
Reporter: amodra at gmail dot com
Target Milestone: ---
Found when looking for a testcase that could trigger a similar problem to
pr72986 for equivalences created in ira using validate_equiv_mem. Instead I
found a tree optimization
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79334
--- Comment #1 from Alan Modra ---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79286 I meant.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79343
Alan Modra changed:
What|Removed |Added
CC||amodra at gmail dot com
--- Comment #3
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79334
--- Comment #4 from Alan Modra ---
When you have the tree optimization bug fixed, this becomes an rtl optimization
bug since rtl pre does the same as tree pre..
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79343
--- Comment #4 from Alan Modra ---
Followup here:
https://github.com/xianyi/OpenBLAS/issues/1078
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79354
Alan Modra changed:
What|Removed |Added
CC||amodra at gmail dot com
--- Comment #5
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79354
--- Comment #6 from Alan Modra ---
Regression tested powerpc64le-linux successfully.
||2017-02-21
CC||amodra at gmail dot com
Assignee|unassigned at gcc dot gnu.org |amodra at gmail dot com
Ever confirmed|0 |1
--- Comment #2 from Alan Modra ---
I don't see a -mno-lra ICE
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79584
Alan Modra changed:
What|Removed |Added
Status|ASSIGNED|NEW
Assignee|amodra at gmail dot
gcc dot gnu.org |amodra at gmail dot com
--- Comment #4 from Alan Modra ---
Created attachment 40813
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=40813&action=edit
lra fix
Patch currently under test
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79584
--- Comment #5 from Alan Modra ---
The patch passes ppc64le bootstrap but that isn't really the best target to
test.
I'd test powerpc64-linux bi-arch but the machine I normally use for BE testing
had a disk die, and instead of trying to set ever
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79584
Alan Modra changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
: rtl-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: amodra at gmail dot com
Target Milestone: ---
On powerpc, the equivalences created by add_store_equivs in ira.c seem like
they might not be very effective.
For reg1 not being assigned a hard reg, they allow lra
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79801
Alan Modra changed:
What|Removed |Added
Severity|normal |enhancement
Assignee: unassigned at gcc dot gnu.org
Reporter: amodra at gmail dot com
Target Milestone: ---
-static and -pie are incompatible, yet gcc neither warns nor errors on the
combination. What's more, gcc's behaviour depends on whether gcc was built
with --enable-defaul
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80044
--- Comment #1 from Alan Modra ---
Revising my original comments..
According to the gcc docs -static is supposed to just prevent linking with
dynamic libraries. I agree that this is traditional linker behaviour.
That means, for example, that -
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80044
Alan Modra changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=60818
--- Comment #19 from Alan Modra ---
Yes, r246294 powerpc64le-linux-gcc -O1 -misel ICEs on the last testcase. An
earlier compiler I had laying around, 7.0.0 20160616, does not.
de |
Assignee|unassigned at gcc dot gnu.org |amodra at gmail dot com
--- Comment #19 from Alan Modra ---
Removing the autoreply email from CCs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=45053
Alan Modra changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
Assignee: unassigned at gcc dot gnu.org
Reporter: amodra at gmail dot com
CC: marxin at gcc dot gnu.org
Target Milestone: ---
Lack of a .note.GNU-stack section results in ld testsuite failures:
arm-linuxeabi FAIL: PR ld/12982
hppa-linux FAIL: PR ld/12982
m68k-linux
: normal
Priority: P3
Component: lto
Assignee: unassigned at gcc dot gnu.org
Reporter: amodra at gmail dot com
CC: marxin at gcc dot gnu.org
Target Milestone: ---
Summary: [8 Regression] lto debugobj references __gnu_lto_slim
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82575
--- Comment #2 from Alan Modra ---
Created attachment 42397
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=42397&action=edit
proposed patch
I hadn't debugged past grepping for "debugobj" when I created the bugzilla.
Now that I've looked f
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82575
--- Comment #3 from Alan Modra ---
Patch now bootstrapped and regression tested powerpc64le-linux.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81890
Alan Modra changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82575
--- Comment #5 from Alan Modra ---
> --- Comment #4 from Richard Biener ---
> Hmm, but those symbols will prevail, enlarging the final symbol table?
> Or are weak + hidden symbols removed even for shared libaries?
Well, they are still undefined
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82575
--- Comment #7 from Alan Modra ---
> --- Comment #6 from rguenther at suse dot de ---
> OK. I suppose they are properly prevailed by any global symbol of the same
> name
> as well? Like a weak definition with default visibility? Or is there the
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82575
Alan Modra changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82598
Alan Modra changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
||2017-10-24
CC||amodra at gmail dot com,
||marxin at gcc dot gnu.org
Component|other |lto
Ever confirmed|0 |1
--- Comment #1 from Alan
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82687
Alan Modra changed:
What|Removed |Added
Status|NEW |RESOLVED
Blocks|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82575
Bug 82575 depends on bug 82687, which changed state.
Bug 82687 Summary: [8 regression] g++.dg/asan/default-options-1.C fails
starting with r253914
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82687
What|Removed
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82757
Alan Modra changed:
What|Removed |Added
CC||amodra at gmail dot com
--- Comment #2
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82757
--- Comment #5 from Alan Modra ---
Incidentally, the only reason ld.bfd gets past the "plugin needed" warning is
that it only warns on a common symbol called __gnu_lto_slim, not an undefined
one.. Bump st_name by one for __gnu_lto_ symbols, perh
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82757
--- Comment #7 from Alan Modra ---
> Bump st_name by one for __gnu_lto_
Better make that two. Some targets will have an extra '_' prefix.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82757
--- Comment #9 from Alan Modra ---
If you start off with "___gnu_lto_", there is code in simple_object_elf.c to
handle the extra '_', and now in binutils too. Skipping just one '_' will
leave you with a symbol that matches "__gnu_lto_", and will
Assignee: unassigned at gcc dot gnu.org
Reporter: amodra at gmail dot com
Target Milestone: ---
Created attachment 42571
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=42571&action=edit
somewhat reduced testcase
Seen with current glibc "make check", gcc
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82928
Alan Modra changed:
What|Removed |Added
Attachment #42571|0 |1
is obsolete|
Priority: P3
Component: tree-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: amodra at gmail dot com
Target Milestone: ---
Created attachment 42686
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=42686&action=edit
preprocessed
: target
Assignee: unassigned at gcc dot gnu.org
Reporter: amodra at gmail dot com
Target Milestone: ---
Created attachment 39668
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=39668&action=edit
testcase
The attached testcase taken from rs6000.c output_cbranch shows
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77676
--- Comment #3 from Alan Modra ---
Bisect points to git commit b9833bfd, svn rev 240298.
-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: amodra at gmail dot com
Target Milestone: ---
Found during build of current binutils.
bfd/ihex.c:222:19: error: '%03o' directive writing between 3 and 11 bytes into
a region of size 9 [-Werror=format-length=]
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77743
Alan Modra changed:
What|Removed |Added
Status|WAITING |NEW
--- Comment #2 from Alan Modra ---
I s
Severity: normal
Priority: P3
Component: c
Assignee: unassigned at gcc dot gnu.org
Reporter: amodra at gmail dot com
Target Milestone: ---
int
f (int x)
{
int ret = 0;
switch (x)
{
case 1:
ret++;
/* Fall through. */
/* Important
||2016-10-04
CC||amodra at gmail dot com
Ever confirmed|0 |1
--- Comment #3 from Alan Modra ---
Confirmed. See also pr77853
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77853
Alan Modra changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78056
Alan Modra changed:
What|Removed |Added
CC||amodra at gmail dot com
--- Comment #10
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78056
--- Comment #13 from Alan Modra ---
Thanks Martin, so it is HAVE_AS_POWER9 being undefined, which then results in
TARGET_P9_VECTOR and others being defined to zero. Kelvin's fix of not
defining the builtins when no power9 support is available so
||2016-11-14
CC||amodra at gmail dot com
Assignee|unassigned at gcc dot gnu.org |amodra at gmail dot com
Ever confirmed|0 |1
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78325
--- Comment #2 from Alan Modra ---
Created attachment 40040
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=40040&action=edit
proposed fix
This hasn't been bootstrapped yet anywhere, but ought to keep pr70890 fixed and
does result in mips c
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78325
Alan Modra changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
|4.10.0
Last reconfirmed||2014-05-01
Component|middle-end |target
CC||amodra at gmail dot com
Assignee|unassigned at gcc dot gnu.org |amodra at gmail dot com
: target
Assignee: unassigned at gcc dot gnu.org
Reporter: amodra at gmail dot com
-mcmodel=medium introduced a code model where addresses typically need two
instructions. As illustrated in the following, the fact that rs6000 lacks an
address_cost function to say that
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=61032
--- Comment #1 from Alan Modra ---
This is a code quality regression from gcc-4.7.2
: target
Assignee: unassigned at gcc dot gnu.org
Reporter: amodra at gmail dot com
Seen on a for (i = 0; i < 1; i++) loop
13bc: 3d 20 05 f5 lis r9,1525
13c0: 7d 29 03 a6 mtctr r9
13c4: 7d 3e 1a 14 add r9,r30
,
||powerpc64le-linux
Status|UNCONFIRMED |ASSIGNED
Last reconfirmed||2014-05-07
Assignee|unassigned at gcc dot gnu.org |amodra at gmail dot com
Ever confirmed|0 |1
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=61098
--- Comment #1 from Alan Modra ---
Created attachment 32754
--> http://gcc.gnu.org/bugzilla/attachment.cgi?id=32754&action=edit
testcase
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60737
Alan Modra changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
||2014-05-08
Assignee|unassigned at gcc dot gnu.org |amodra at gmail dot com
Ever confirmed|0 |1
||2014-05-21
CC||amodra at gmail dot com
Assignee|unassigned at gcc dot gnu.org |amodra at gmail dot com
Ever confirmed|0 |1
--- Comment #9 from Alan Modra ---
So, what's happening here o
401 - 500 of 872 matches
Mail list logo
