elram added a comment.
Herald added subscribers: llvm-commits, delcypher.
This patch breaks cross-compiling the builtins for i686, with
COMPILER_RT_DEFAULT_TARGET_ONLY=YES.
CC is set to
clang -target i686-linux-musl -m32 -march=i686
But no supported target is detected by cmake and nothing
rnk added a comment.
In https://reviews.llvm.org/D26764#855791, @eugenis wrote:
> IMHO it was a very bad idea to include the "architecture name" (not even a
> target triple!) in the library path in the first place. No one else does
> that. GCC made the right choice and called theirs
eugenis added a comment.
I've landed the android special case in r312048
Repository:
rL LLVM
https://reviews.llvm.org/D26764
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
hans added a comment.
In https://reviews.llvm.org/D26764#855791, @eugenis wrote:
> IMHO it was a very bad idea to include the "architecture name" (not even a
> target triple!) in the library path in the first place. No one else does
> that. GCC made the right choice and called theirs
eugenis added a comment.
IMHO it was a very bad idea to include the "architecture name" (not even a
target triple!) in the library path in the first place. No one else does that.
GCC made the right choice and called theirs "libasan.so".
My plan is to tweak the code that sets the library name
mgorny added a comment.
I'm happy to help with whatever needs to be done to keep breakage to the
minimum, provided that we determine some clear path forward that doesn't
involve regressing to the half-broken state for non-Android Linux systems and
it's doable in the free time I can spare.
eugenis added a comment.
I may have underestimated the number of places that will need to be patched
because of this change. Perhaps we should restore the special case in the
android library name?
Repository:
rL LLVM
https://reviews.llvm.org/D26764
hans added inline comments.
Comment at: compiler-rt/trunk/lib/asan/scripts/asan_device_setup:98
if [[ $_ABI == x86* ]]; then
-_ARCH=i686
+_ARCH=i386
elif [[ $_ABI == armeabi* ]]; then
There are lots of copies of this script in various
This revision was automatically updated to reflect the committed changes.
Closed by commit rL311924: Reland r311842 - [cmake] Remove i686 target that is
duplicate to i386 (authored by mgorny).
Repository:
rL LLVM
https://reviews.llvm.org/D26764
Files:
mgorny reopened this revision.
mgorny added a comment.
This revision is now accepted and ready to land.
I have reverted the changes for now to fix Android.
Repository:
rL LLVM
https://reviews.llvm.org/D26764
___
cfe-commits mailing list
This revision was automatically updated to reflect the committed changes.
Closed by commit rL311842: [cmake] Remove i686 target that is duplicate to i386
(authored by mgorny).
Changed prior to commit:
https://reviews.llvm.org/D26764?vs=112807=112808#toc
Repository:
rL LLVM
mgorny updated this revision to Diff 112807.
mgorny added a comment.
I'm going to commit this now to see if it helps with the Android buildbot.
Updating to the rebased patch that's going to be committed.
https://reviews.llvm.org/D26764
Files:
cmake/Modules/CompilerRTUtils.cmake
eugenis accepted this revision.
eugenis added a comment.
This revision is now accepted and ready to land.
LGTM
https://reviews.llvm.org/D26764
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
mgorny added a comment.
Ping. @eugenis, do you consider it good to go then? Or do you want me to change
something specifically?
https://reviews.llvm.org/D26764
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
eugenis added a comment.
I think this is the right move together with https://reviews.llvm.org/D26796.
Android build system will need to support both names for some time, depending
on the toolchain version - looks doable. Technically, it can stick with the old
name forever.
mgorny added a comment.
In https://reviews.llvm.org/D26764#606638, @beanz wrote:
> Looping in @eugenis, who added i686 support in r218761.
>
> eugenis, since i686 is identical to i386 generating it as a separate target
> is undesirable. Can you help advise as to how we can better meet your
beanz added a reviewer: eugenis.
beanz added a subscriber: eugenis.
beanz added a comment.
Looping in @eugenis, who added i686 support in r218761.
eugenis, since i686 is identical to i386 generating it as a separate target is
undesirable. Can you help advise as to how we can better meet your
mgorny added a comment.
Ping.
https://reviews.llvm.org/D26764
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
mgorny created this revision.
mgorny added reviewers: samsonov, etienneb, beanz.
mgorny added a subscriber: cfe-commits.
Herald added subscribers: dberris, kubabrecka.
Remove the explicit i686 target that is completely duplicate to
the i386 target, with the latter being used more commonly.
1.
19 matches
Mail list logo