Page MenuHomePhabricator

dim (Dimitry Andric)
User

Projects

User does not belong to any projects.

User Details

User Since
Apr 15 2014, 12:19 PM (330 w, 3 d)

Recent Activity

Mon, Aug 10

dim added a comment to D85659: [PowerPC] in fixupIsDeadOrKill before RA, get def instructions by calling getVRegDef .

For me, this fixes the assertion for both the original test case (a fully preprocessed ASTReader.cpp) and the minimized test case from https://bugs.llvm.org/show_bug.cgi?id=47041#c1

Mon, Aug 10, 1:25 PM · Restricted Project

Sun, Aug 9

dim added inline comments to rG7ab05d9a60b0: [PowerPC] fold addi's imm operand to its imm form consumer's displacement.
Sun, Aug 9, 12:10 PM

Mon, Aug 3

dim added a comment to D85082: On FreeBSD, add -pthread to ASan dynamic compile flags for tests.

Can we instead add a dependency on libpthread.so/libthr.so to libasan.so?

Never mind, it seems like it's already there:

File: ./lib/clang/12.0.0/lib/freebsd/libclang_rt.asan-x86_64.so
Format: elf64-x86-64
Arch: x86_64
AddressSize: 64bit
LoadName: libclang_rt.asan-x86_64.so
NeededLibraries [
  libc++.so.1
  libc.so.7
  libcxxrt.so.1
  libdl.so.1
  libgcc_s.so.1
  libm.so.5
  librt.so.1
  libthr.so.3
]
Mon, Aug 3, 6:25 AM · Restricted Project

Sun, Aug 2

dim added a comment to D44536: Avoid segfault when destructor is not yet known.

Hm, this review's still open after two years, and even as of 2020-08-02 clang still crashes on the sample. :)

Sun, Aug 2, 3:19 AM

Sat, Aug 1

dim added a comment to D85082: On FreeBSD, add -pthread to ASan dynamic compile flags for tests.

Note that as of rG1db4318766256f25a03ef80af8dbb3f99743ebe9, applying this change results in the Passed testcases increasing from 49259 to 49666, so more than 400 Failed test cases are fixed by it.

Sat, Aug 1, 3:44 PM · Restricted Project
dim requested review of D85082: On FreeBSD, add -pthread to ASan dynamic compile flags for tests.
Sat, Aug 1, 3:40 PM · Restricted Project
dim closed D83411: Use md5 -q for HashProgramOutput.sh, to show only the checksum..
Sat, Aug 1, 6:36 AM

Fri, Jul 31

dim added a comment to D77776: [Driver] Default to libc++ on FreeBSD.
In D77776#2187273, @dim wrote:
+#ifdef __FreeBSD__
+	  return __FreeBSD_version / 100000;
+#else
+	  return 10;
+#endif

No, that would hardcode something at build time. We need to respect whatever triple the user is passing in.

Nobody has proposed not respecting that, this is all in the context of the if (Major == 0) fallback

Fri, Jul 31, 6:12 AM · Restricted Project
dim added inline comments to D83411: Use md5 -q for HashProgramOutput.sh, to show only the checksum..
Fri, Jul 31, 6:11 AM
dim added a comment to D77776: [Driver] Default to libc++ on FreeBSD.

I don't like the fact that this only changes one of the users of getTriple().getOSMajorVersion().

Fri, Jul 31, 5:43 AM · Restricted Project
dim reclaimed D83411: Use md5 -q for HashProgramOutput.sh, to show only the checksum..
Fri, Jul 31, 3:09 AM
dim abandoned D83411: Use md5 -q for HashProgramOutput.sh, to show only the checksum..
Fri, Jul 31, 2:54 AM

Sun, Jul 19

dim added a comment to D83645: Bump the default target CPU for i386-freebsd to i686.

@dim

Hi, your git commit contains extra Phabricator tags. You can drop Reviewers: Subscribers: Tags: and the text Summary: from the git commit with the following script:

arcfilter () {
        arc amend
        git log -1 --pretty=%B | awk '/Reviewers:|Subscribers:/{p=1} /Reviewed By:|Differential Revision:/{p=0} !p && !/^Summary:$/ {sub(/^Summary: /,"");print}' | git commit --amend --date=now -F -
}

Reviewed By: is considered important by some people. Please keep the tag. (--date=now is my personal preference (author dates are usually not useful. Using committer dates can make log almost monotonic in time))

llvm/utils/git/pre-push.py can validate the message does not include unneeded tags.

Sun, Jul 19, 2:46 AM · Restricted Project

Thu, Jul 16

dim accepted D83990: [test-release] fallback to py3's venv module.

LGTM.

Thu, Jul 16, 3:29 PM · Restricted Project

Jul 12 2020

dim committed rG02cfa7530d9e: Bump the default target CPU for i386-freebsd to i686 (authored by dim).
Bump the default target CPU for i386-freebsd to i686
Jul 12 2020, 2:45 PM
dim closed D83645: Bump the default target CPU for i386-freebsd to i686.
Jul 12 2020, 2:45 PM · Restricted Project
Herald added a project to D83645: Bump the default target CPU for i386-freebsd to i686: Restricted Project.
Jul 12 2020, 11:36 AM · Restricted Project

Jul 8 2020

dim created D83411: Use md5 -q for HashProgramOutput.sh, to show only the checksum..
Jul 8 2020, 10:25 AM

Jun 13 2020

dim updated subscribers of D69825: [Clang][Driver] Re-use the calling process instead of creating a new process for the cc1 invocation.

Okay, after a long while of attempting to make a reproduction scenario, I finally managed one that consistently worked. As BSD make puts in different redirections for stderr when running in -j mode, it turned out that I could simply run the compilation twice on a terminal, first normally, then with stderr redirected to /dev/null, e.g. my test script is:

Jun 13 2020, 4:01 PM · Restricted Project, Restricted Project

Jun 4 2020

dim added a comment to D69825: [Clang][Driver] Re-use the calling process instead of creating a new process for the cc1 invocation.

For completeness sake, here are the valgrind outputs of clang rGf7f1abdb889 (llvmorg-11-init-16778) built by gcc 9.3.0-10ubuntu2 (in RelWithDebInfo mode):

Jun 4 2020, 1:50 PM · Restricted Project, Restricted Project
dim added a comment to D69825: [Clang][Driver] Re-use the calling process instead of creating a new process for the cc1 invocation.
In D69825#2071687, @dim wrote:

Trying to reduce this now.

Jun 4 2020, 11:00 AM · Restricted Project, Restricted Project

Jun 3 2020

dim added a comment to D69825: [Clang][Driver] Re-use the calling process instead of creating a new process for the cc1 invocation.
In D69825#2063979, @dim wrote:

FWIW, this change causes https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=246630, see in particular https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=246630#c18.

For some reason, not spawning a new process for the cc1 stage can lead to non-reproducible output. The exact cause is not known yet.

Can you share the preprocessed source and compiler command-line for printf.c, either on the bug or here?

Jun 3 2020, 9:53 AM · Restricted Project, Restricted Project

May 29 2020

dim added a comment to D69825: [Clang][Driver] Re-use the calling process instead of creating a new process for the cc1 invocation.

FWIW, this change causes https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=246630, see in particular https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=246630#c18.

May 29 2020, 1:06 PM · Restricted Project, Restricted Project

May 15 2020

dim abandoned D56398: Add new EINTEGRITY errno.
May 15 2020, 8:06 AM
dim added a comment to D56398: Add new EINTEGRITY errno.

Since we should pick up FreeBSD's EINTEGRITY via include_next, I don't see why we need this patch. Can we abandon?

May 15 2020, 8:06 AM

May 12 2020

dim committed rGfc373522b044: [arm] Add big-endian version of pcrel fixups for adr instructions (authored by dim).
[arm] Add big-endian version of pcrel fixups for adr instructions
May 12 2020, 10:44 AM
dim closed D79774: [ARM] Add big-endian version of pcrel fixups for adr instructions.
May 12 2020, 10:44 AM · Restricted Project
dim added a comment to D79774: [ARM] Add big-endian version of pcrel fixups for adr instructions.
In D79774#2031755, @dim wrote:

Note I already merged this fix (minus the test case) into FreeBSD here: https://svnweb.freebsd.org/changeset/base/360697, since we needed it ASAP for being able to build big-endian ARM kernel code. So in that sense it's already been exercised for a bit. :)

MFC after: immediately, since this fix is meant for stable/11

I still think the offending FreeBSD assembly can be fixed to be more portable.

May 12 2020, 10:43 AM · Restricted Project
dim retitled D79774: [ARM] Add big-endian version of pcrel fixups for adr instructions from [arm] Add big-endian version of pcrel fixups for adr instructions to [ARM] Add big-endian version of pcrel fixups for adr instructions.
May 12 2020, 10:43 AM · Restricted Project
dim added a comment to D79774: [ARM] Add big-endian version of pcrel fixups for adr instructions.

Note I already merged this fix (minus the test case) into FreeBSD here: https://svnweb.freebsd.org/changeset/base/360697, since we needed it ASAP for being able to build big-endian ARM kernel code. So in that sense it's already been exercised for a bit. :)

May 12 2020, 9:39 AM · Restricted Project
dim created D79774: [ARM] Add big-endian version of pcrel fixups for adr instructions.
May 12 2020, 3:43 AM · Restricted Project

Apr 26 2020

dim accepted D78877: [ARM] Only produce qadd8b under hasV6Ops.

Verified that this fixes both the original test case (which is rather large so I didn't attach it to the PR), and the minimized test case.

Apr 26 2020, 8:29 AM · Restricted Project
dim added a comment to D78877: [ARM] Only produce qadd8b under hasV6Ops.

Trying this out now, thanks!

Apr 26 2020, 4:14 AM · Restricted Project

Apr 15 2020

dim committed rG7fb79105fe0c: Use maximum compression when packaging release tarballs. (authored by dim).
Use maximum compression when packaging release tarballs.
Apr 15 2020, 11:33 AM
dim committed rG9daadcec8137: Turn off core dumps before starting the main body of test-release.sh. (authored by dim).
Turn off core dumps before starting the main body of test-release.sh.
Apr 15 2020, 11:32 AM
dim closed D76192: Use maximum compression when packaging release tarballs..
Apr 15 2020, 11:32 AM · Restricted Project
dim closed D76191: Turn off core dumps before starting the main body of test-release.sh..
Apr 15 2020, 11:32 AM · Restricted Project

Apr 9 2020

dim added a comment to D77776: [Driver] Default to libc++ on FreeBSD.

Alternatively, the checks could be changed to also handle OSMajorVersion == 0 and translate that to 10.
This seems to be what NetBSD.cpp does. Darwin.cpp also infers the version from the host when running on macos.

Apr 9 2020, 11:53 AM · Restricted Project

Mar 26 2020

dim added a comment to D65034: [libc++] use constexpr for sizeof comparisons.

FYI FreeBSD is one of the few OSes that enables -Wsystem-headers by default, *and* uses libc++ as its default C++ library. So we've already submitted a bunch of warning fixes for libc++.

Mar 26 2020, 8:06 AM

Mar 20 2020

dim added a comment to D6379: Only warn about DWARF2 supporting one section per compilation unit for code sections.

Hello, we're hitting this issue when assembling the Linux kernel with Clang (https://github.com/ClangBuiltLinux/linux/issues/716). I'd be happy to dust this off/rebase and add tests. @dim would you prefer to do it or shall I?

Mar 20 2020, 4:50 PM

Mar 18 2020

dim accepted D75425: [docs] Added solutions to slow build under common problems.

As far as I'm concerned, most suggestions make sense, and it is good to have this information clearly stated.

Mar 18 2020, 2:07 PM · Restricted Project

Mar 17 2020

dim committed rG585a3cc31bb4: Fix -Wdeprecated-copy-dtor and -Wdeprecated-dynamic-exception-spec warnings. (authored by dim).
Fix -Wdeprecated-copy-dtor and -Wdeprecated-dynamic-exception-spec warnings.
Mar 17 2020, 11:18 AM
dim closed D76150: Fix -Wdeprecated-copy-dtor and -Wdeprecated-dynamic-exception-spec warnings..
Mar 17 2020, 11:17 AM · Restricted Project

Mar 16 2020

dim added a comment to D76192: Use maximum compression when packaging release tarballs..

This seems reasonable, but can we wait until after the release? I would prefer to not change these things right now.

Mar 16 2020, 3:09 AM · Restricted Project

Mar 15 2020

dim created D76192: Use maximum compression when packaging release tarballs..
Mar 15 2020, 4:48 AM · Restricted Project
dim created D76191: Turn off core dumps before starting the main body of test-release.sh..
Mar 15 2020, 3:43 AM · Restricted Project

Mar 13 2020

dim added a comment to D76150: Fix -Wdeprecated-copy-dtor and -Wdeprecated-dynamic-exception-spec warnings..

Note: one thing that I am unsure about is the _NOEXCEPT after some of the copy constructors. I only applied those to classes that have other _NOEXCEPT constructors, but copying may involve other things that could throw.

Mar 13 2020, 11:49 AM · Restricted Project
dim created D76150: Fix -Wdeprecated-copy-dtor and -Wdeprecated-dynamic-exception-spec warnings..
Mar 13 2020, 11:49 AM · Restricted Project

Feb 15 2020

dim committed rG62654cab7e65: Restore functionality of --sysroot on FreeBSD after b18cb9c47 (authored by dim).
Restore functionality of --sysroot on FreeBSD after b18cb9c47
Feb 15 2020, 4:55 PM
dim updated subscribers of rL211743: Land support for ARM EHABI unwinding for libunwind..
Feb 15 2020, 6:10 AM

Feb 10 2020

dim added a comment to D73990: [Sanitizers] Get link map on FreeBSD via documented API.
In D73990#1868085, @dim wrote:

This breaks in our downstream runtimes builds for armv7-none-linux-androideabi16. FYI

Can you point me to some build logs? (I didn't receive any failure mails from bots yet.)

Posted a fix at D74358

Feb 10 2020, 2:51 PM · Restricted Project, Restricted Project
dim added a comment to D74358: Marking internal_dlinfo as unsupported for Android sanitizers to fix runtimes builds..

Yeah sorry, I used a whitelist instead, here: rG52f2df1ecdd79cc550b694ab280f3b0396d7cf9a

Feb 10 2020, 2:51 PM · Restricted Project, Restricted Project
dim committed rG52f2df1ecdd7: [Sanitizers] Get link map on FreeBSD and NetBSD via documented API (authored by dim).
[Sanitizers] Get link map on FreeBSD and NetBSD via documented API
Feb 10 2020, 2:44 PM
dim committed rG480eea4e45bc: Revert "[Sanitizers] Get link map on FreeBSD via documented API" (authored by dim).
Revert "[Sanitizers] Get link map on FreeBSD via documented API"
Feb 10 2020, 2:33 PM
dim added a reverting change for rG92e267a94dc4: [Sanitizers] Get link map on FreeBSD via documented API: rG480eea4e45bc: Revert "[Sanitizers] Get link map on FreeBSD via documented API".
Feb 10 2020, 2:33 PM
dim added a comment to D73990: [Sanitizers] Get link map on FreeBSD via documented API.

This breaks in our downstream runtimes builds for armv7-none-linux-androideabi16. FYI

Feb 10 2020, 2:15 PM · Restricted Project, Restricted Project
dim committed rG92e267a94dc4: [Sanitizers] Get link map on FreeBSD via documented API (authored by dim).
[Sanitizers] Get link map on FreeBSD via documented API
Feb 10 2020, 10:23 AM
dim closed D73990: [Sanitizers] Get link map on FreeBSD via documented API.
Feb 10 2020, 10:23 AM · Restricted Project, Restricted Project

Feb 9 2020

dim added a comment to D73990: [Sanitizers] Get link map on FreeBSD via documented API.

Ping, can we get this committed then? :)

Feb 9 2020, 11:20 PM · Restricted Project, Restricted Project

Feb 7 2020

dim updated the diff for D73990: [Sanitizers] Get link map on FreeBSD via documented API.

Remove stray newline.

Feb 7 2020, 10:35 AM · Restricted Project, Restricted Project
dim updated the diff for D73990: [Sanitizers] Get link map on FreeBSD via documented API.

Add better NetBSD implementation of internal_dlinfo().

Feb 7 2020, 10:35 AM · Restricted Project, Restricted Project

Feb 6 2020

dim updated the diff for D73990: [Sanitizers] Get link map on FreeBSD via documented API.
  • Rebase onto reformatted files.
  • Add internal_dlinfo() for FreeBSD and NetBSD, with stubs for other platforms.
  • Add GET_LINK_MAP_BY_DLOPEN_HANDLE() macro for NetBSD.
Feb 6 2020, 2:25 PM · Restricted Project, Restricted Project
dim added a reviewer for D73990: [Sanitizers] Get link map on FreeBSD via documented API: krytarowski.
Feb 6 2020, 2:25 PM · Restricted Project, Restricted Project
dim committed rG924c9030a89d: [compiler-rt] clang-format FreeBSD-specific sanitizer sources (authored by dim).
[compiler-rt] clang-format FreeBSD-specific sanitizer sources
Feb 6 2020, 1:49 PM
dim added inline comments to D73990: [Sanitizers] Get link map on FreeBSD via documented API.
Feb 6 2020, 6:35 AM · Restricted Project, Restricted Project
dim added a comment to D73990: [Sanitizers] Get link map on FreeBSD via documented API.

Btw, note that according to the Linux dlinfo() manpage, it also has RTLD_DI_LINKMAP as a documented way to get at the link map. Maybe we should generalize this in a follow-up review?

Feb 6 2020, 6:17 AM · Restricted Project, Restricted Project
dim added inline comments to D73990: [Sanitizers] Get link map on FreeBSD via documented API.
Feb 6 2020, 6:08 AM · Restricted Project, Restricted Project

Feb 4 2020

dim added inline comments to D73990: [Sanitizers] Get link map on FreeBSD via documented API.
Feb 4 2020, 10:17 PM · Restricted Project, Restricted Project
dim committed rG8a1f4feb1b74: [compiler-rt] Fix sanitizer_common build for FreeBSD (authored by dim).
[compiler-rt] Fix sanitizer_common build for FreeBSD
Feb 4 2020, 1:56 PM
dim updated the diff for D73990: [Sanitizers] Get link map on FreeBSD via documented API.

Formatting.

Feb 4 2020, 1:55 PM · Restricted Project, Restricted Project
dim updated the diff for D73990: [Sanitizers] Get link map on FreeBSD via documented API.

Cast return value to link_map to avoid call site errors.

Feb 4 2020, 1:55 PM · Restricted Project, Restricted Project
dim created D73990: [Sanitizers] Get link map on FreeBSD via documented API.
Feb 4 2020, 1:09 PM · Restricted Project, Restricted Project

Jan 29 2020

dim committed rG5e416ba943b7: Define _LIBCPP_HAS_TIMESPEC_GET for FreeBSD when appropriate (authored by dim).
Define _LIBCPP_HAS_TIMESPEC_GET for FreeBSD when appropriate
Jan 29 2020, 11:03 PM
dim closed D71522: Define _LIBCPP_HAS_TIMESPEC_GET for FreeBSD when appropriate.
Jan 29 2020, 11:03 PM · Restricted Project

Jan 24 2020

dim added inline comments to D71600: PowerPC 32-bit - forces 8 byte lock/lock_free decisions at compiled time.
Jan 24 2020, 12:55 PM · Restricted Project, Restricted Project, Restricted Project
dim committed rG58592f6c4924: Include <cstdlib> for std::abort() in clangd (authored by dim).
Include <cstdlib> for std::abort() in clangd
Jan 24 2020, 11:58 AM

Jan 17 2020

dim added a comment to D72910: Fix a bug with clang with object destructor, while skipping object initialization - make clang crash.

Aha, which version of clang-cl are you using? With the released version of clang-cl 9.0.1, I get a warning instead of an error:

Jan 17 2020, 5:22 AM · Restricted Project
dim added a comment to D72910: Fix a bug with clang with object destructor, while skipping object initialization - make clang crash.

Eh, no it does not crash clang, at least not here? Instead it gives you a compile error, as it should:

Jan 17 2020, 3:02 AM · Restricted Project

Jan 9 2020

dim added inline comments to D71827: [clang] avoid strict aliasing violation in assert.
Jan 9 2020, 11:01 AM · Restricted Project

Jan 8 2020

dim accepted D72014: [PowerPC]: Add powerpcspe target triple subarch component.

LGTM.

Jan 8 2020, 9:43 AM · Restricted Project, Restricted Project

Dec 15 2019

dim added a comment to D68741: test-release.sh s/http/https/.

We should abandon this, for e.g. 9.0.1 the whole Base_url variable has become unused. :)

Dec 15 2019, 3:38 AM · Restricted Project
dim created D71522: Define _LIBCPP_HAS_TIMESPEC_GET for FreeBSD when appropriate.
Dec 15 2019, 3:29 AM · Restricted Project
dim updated the summary of D71522: Define _LIBCPP_HAS_TIMESPEC_GET for FreeBSD when appropriate.
Dec 15 2019, 3:29 AM · Restricted Project

Nov 18 2019

dim committed rGee31adb7fa42: Populate CUDA flags on FreeBSD too, as many other toolchains do. (authored by dim).
Populate CUDA flags on FreeBSD too, as many other toolchains do.
Nov 18 2019, 12:58 PM
dim closed D69990: Populate CUDA flags on FreeBSD too, as many other toolchains do..
Nov 18 2019, 12:58 PM · Restricted Project
dim added a comment to D69990: Populate CUDA flags on FreeBSD too, as many other toolchains do..
In D69990#1750348, @tra wrote:

LGTM, though I'm curious if it's particularly useful. Last time I checked NVIDIA didn't ship libcudart for FreeBSD and without it it's rather cumbersome to use CUDA in practice.

Nov 18 2019, 12:57 PM · Restricted Project
dim added reviewers for D69990: Populate CUDA flags on FreeBSD too, as many other toolchains do.: tra, yaxunl, ABataev.

Adding a few people who might know a bit more about CUDA specific things. Please take a look if this review makes any sense, thanks. :)

Nov 18 2019, 11:08 AM · Restricted Project
dim updated the diff for D69990: Populate CUDA flags on FreeBSD too, as many other toolchains do..
  • Add cuda options test, copied from cuda-options.cu.
Nov 18 2019, 11:08 AM · Restricted Project

Nov 17 2019

dim edited reviewers for D69990: Populate CUDA flags on FreeBSD too, as many other toolchains do., added: dim, emaste; removed: Restricted Project.

I'll work on a test.

Nov 17 2019, 11:09 PM · Restricted Project

Nov 13 2019

dim committed rG3db6783d8a7d: Check result of emitStrLen before passing it to CreateGEP (authored by dim).
Check result of emitStrLen before passing it to CreateGEP
Nov 13 2019, 11:11 PM
dim closed D70143: Check result of emitStrLen before passing it to CreateGEP.
Nov 13 2019, 11:11 PM · Restricted Project, Restricted Project
dim updated the diff for D70143: Check result of emitStrLen before passing it to CreateGEP.

Now opt supports -disable-builtin, move the test to llvm/test/Transforms/InstCombine.

Nov 13 2019, 1:19 PM · Restricted Project, Restricted Project
dim committed rG597b77fb7ff9: Add -disable-builtin option to opt (authored by dim).
Add -disable-builtin option to opt
Nov 13 2019, 12:33 PM
dim closed D70193: Add -disable-builtin option to opt.
Nov 13 2019, 12:33 PM · Restricted Project
dim updated the summary of D70193: Add -disable-builtin option to opt.
Nov 13 2019, 12:09 PM · Restricted Project
dim updated subscribers of D56398: Add new EINTEGRITY errno.

Sorry if this is a basic question, but why are we trying to provide EINTEGRITY in libc++'s errno.h header at all? I don't see this being part of the C or C++ Standard.

Nov 13 2019, 11:27 AM
dim added a comment to D70143: Check result of emitStrLen before passing it to CreateGEP.

I submitted D70193 for adding a -disable-builtin option to opt. Once that is committed, this review can continue.

Nov 13 2019, 11:18 AM · Restricted Project, Restricted Project
dim created D70193: Add -disable-builtin option to opt.
Nov 13 2019, 11:18 AM · Restricted Project
dim added inline comments to D70143: Check result of emitStrLen before passing it to CreateGEP.
Nov 13 2019, 6:53 AM · Restricted Project, Restricted Project

Nov 12 2019

dim added a comment to D56398: Add new EINTEGRITY errno.

cem == dim?

Nov 12 2019, 2:03 PM