[PATCH] D49362: [ThinLTO] Internalize read only globals

Wed Nov 14 20:07:33 PST 2018

Sorry, let's try again. Two archives this time.

The first one is the exact snapshot after link failure, including input IR, intermediate IR and final object. For b.out, there is a cache hit for main.o, thus all the intermediate IR for main.o only generated once.
The second one is a snapshot for building b.out from empty cache which it succeeded. llvmcache-882BA8DC10FA68A3081AD002CE414F99AE583990 is not the same between two builds.

Steven

> On Nov 14, 2018, at 5:24 PM, Teresa Johnson <tejohnson at google.com> wrote:
> 
> Both sets of intermediate files have test1.c files (neither has test2.c files). 
> 
> On Wed, Nov 14, 2018 at 4:34 PM Steven Wu <stevenwu at apple.com <mailto:stevenwu at apple.com>> wrote:
> 
> 
>> On Nov 14, 2018, at 4:28 PM, Teresa Johnson <tejohnson at google.com <mailto:tejohnson at google.com>> wrote:
>> 
>> 
>> 
>> On Wed, Nov 14, 2018 at 1:47 PM Steven Wu <stevenwu at apple.com <mailto:stevenwu at apple.com>> wrote:
>> Here is a small reduced test case using the same main.c and foo.c from the review:
>> cat main.c
>>> int foo();
>>> int main() {
>>>   return foo();
>>> }
>> 
>> cat foo.c
>>> #include <stdlib.h>
>>> static int gFoo = 1;
>>> int foo() {
>>>   return gFoo;
>>> }
>>> void bar() {
>>>   gFoo = rand();
>>> }
>> 
>> cat test1.c
>>> int foo();
>>> int test() {
>>>   return foo();
>>> }
>> 
>> cat test2.c
>>> int foo();
>>> int bar();
>>> int test() {
>>>   return foo() + bar();
>>> }
>> 
>> To reproduce:
>> $ clang -O3 -flto=thin -c main.c test1.c test2.c foo.c
>> $ clang -O3 -flto=thin -Wl,-cache_path_lto,lto.cache main.o test2.o foo.o
>> $ clang -O3 -flto=thin -Wl,-cache_path_lto,lto.cache main.o test1.o foo.o
>> Undefined symbols for architecture x86_64:
>>   "_gFoo.llvm.17695746433417383459", referenced from:
>>       _main in lto.o
>> 
>> I can't reproduce this, although I am using gold (with  -Wl,-plugin-opt,cache-dir=lto.cache) since I don't have ld64. I notice that test() is dead, so with gold at least it is removed completely.
>> 
>> Do you have any temp files you can share (i.e. IR before and after importing)?
> 
> Attach all the intermediate file. I haven't had time to look in detail yet.
> 
> 
> Steven
> 
>> 
>> 
>> 
>> Steven
>> 
>> 
>>> On Nov 14, 2018, at 1:11 PM, Steven Wu <stevenwu at apple.com <mailto:stevenwu at apple.com>> wrote:
>>> 
>>> I am making some low progress on reproducing it on trunk. It is not very easy to get into this state but this is what I found right now:
>>> 
>>> ParseAST always has readonly reference to gCrashRecoveryEnabled and in both linkage unit, gCrashRecoveryEnabled is imported into ParseAST. The difference is that for libclang, there is a different module which modifies gCrashRecoveryEnabled. Maybe I am missing something but will there be any difference in hash value for ParseAST?
>>> 
>>> After importing, I see:
>>> 
>>> libclang ParseAST
>>> @_ZL21gCrashRecoveryEnabled.llvm.16761009941167954495 = available_externally hidden unnamed_addr global i1 false, align 1
>>> 
>>> clang-func-mapping ParseAST
>>> @_ZL21gCrashRecoveryEnabled.llvm.16761009941167954495 = internal unnamed_addr global i1 false, align 1 #0
>>> attributes #0 = { "thinlto-internalize" }
>>> 
>>> I am little suspecting about the libclang case. Is this semantically correct if it is inlined? It will not observe the changes to the value if modified.
>>> 
>>> Steven
>>> 
>>> 
>>>> On Nov 14, 2018, at 9:57 AM, Teresa Johnson <tejohnson at google.com <mailto:tejohnson at google.com>> wrote:
>>>> 
>>>> 
>>>> 
>>>> On Wed, Nov 14, 2018 at 9:42 AM Steven Wu <stevenwu at apple.com <mailto:stevenwu at apple.com>> wrote:
>>>> For the stage 2 build, just use clang/cmake/caches/Apple-stage2-ThinLTO.cmake. I also throw a profile data in my build just to match our builder but I suspect that might not be needed. I will double check to confirm that.
>>>> 
>>>> You also don't need to wait for the entire build to finish. Just run `ninja libclang.dylib && ninja clang-func-mapping` is consistently reproducing for me. Let me know if you need more help with this.
>>>> 
>>>> Another note, it will be good to add assembly support for readonly bits. It will be easier to debug.
>>>> 
>>>> Good point - I completely forgot about this when reviewing. Eugene - you will need to change AsmWriter.cpp, LLToken.h, LLParser.cpp and LLLexer.cpp (see the changes to those files in D53345 as an example, although that was adding a new function flag, but you would want to do something similar for the new gvar flag).
>>>> 
>>>> Teresa
>>>> 
>>>> 
>>>> Steven
>>>> 
>>>>> On Nov 14, 2018, at 9:27 AM, Evgeny Leviant <eleviant at accesssoftek.com <mailto:eleviant at accesssoftek.com>> wrote:
>>>>> 
>>>>> I couldn't reproduce this on Linux, now trying Mac OS X (will take the whole night as it is slow)
>>>>> Stewen, can you please send me CMake configuration string from your buildbot. It would be helpful 
>>>>> if pending OS X build also succeed.
>>>>> 
>>>>> What I'm doing is:
>>>>> - build LLVM/clang with thin LTO cache
>>>>> - When finished remove cache from disk, then remove and rebuild libclang.so(dylib)
>>>>> - Remove and rebuild clang-func-mapping
>>>>> 
>>>>> Are those steps correct?
>>>>> От: Teresa Johnson via Phabricator <reviews at reviews.llvm.org <mailto:reviews at reviews.llvm.org>>
>>>>> Отправлено: 14 ноября 2018 г. 19:00
>>>>> Кому: Evgeny Leviant; tejohnson at google.com <mailto:tejohnson at google.com>; joker.eph at gmail.com <mailto:joker.eph at gmail.com>
>>>>> Копия: trent.xin.tong at gmail.com <mailto:trent.xin.tong at gmail.com>; arphaman at gmail.com <mailto:arphaman at gmail.com>; dany.grumberg at gmail.com <mailto:dany.grumberg at gmail.com>; jfbastien at apple.com <mailto:jfbastien at apple.com>; anton at korobeynikov.info <mailto:anton at korobeynikov.info>; llvm at inglorion.net <mailto:llvm at inglorion.net>; eraman at google.com <mailto:eraman at google.com>; stevenwu at apple.com <mailto:stevenwu at apple.com>; dexonsmith at apple.com <mailto:dexonsmith at apple.com>; llvm-commits at lists.llvm.org <mailto:llvm-commits at lists.llvm.org>; George Rimar; Igor Kudrin; jun.l at samsung.com <mailto:jun.l at samsung.com>
>>>>> Тема: [PATCH] D49362: [ThinLTO] Internalize read only globals
>>>>>  
>>>>> CAUTION: This email originated from outside of the organization. Do not click links or open attachments unless you recognize the sender and know the content is safe.
>>>>> 
>>>>> tejohnson added a comment.
>>>>> 
>>>>> In https://reviews.llvm.org/D49362#1297188 <https://reviews.llvm.org/D49362#1297188>, @steven_wu wrote:
>>>>> 
>>>>> > I reverted this in r346768.
>>>>> >
>>>>> > I think this is indeed a caching problem. There are some dylib/binary in clang projects that toggles whether to enable the recovery context but some does not. I can reproduce the issue by thin link libclang.dylib first, then link clang-func-mapping on Darwin.
>>>>> >  libclang.dylib thinks the file scope variable gCrashRecoveryEnabled not read-only, so it promotes it.
>>>>> >  clang-func-mapping thinks gCrashRecoveryEnabled read-only, so it internalize and constant propagate the variable but ParseAST.o in the cache is still expecting gCrashRecoveryEnabled to be available.
>>>>> >
>>>>> > Let me know if you need more information.
>>>>> 
>>>>> 
>>>>> Thinking through this example, I'm not sure what is going on. This patch did change the cache key computation to include the read only bit from the global var summary of anything defined or imported. Since gCrashRecoveryEnabled is a static in CrashRecoveryContext.cpp, it must have been imported into ParseAST.o, which means that the read only bit on the associated gvar summary should have been hashed into ParseAST.o's cache key. So presumably we shouldn't have had a cache hit for ParseAST.o (built when the read only bit is not set on gCrashRecoveryEnabled) when thin linking clang-func-mapping which has this bit set for that variable.
>>>>> 
>>>>> Even if the variable was originally externally visible, in order for it to be marked read only by the thin link it would have had to have been imported at all use sites. Which means that any referencing module had to have it in the import set (and therefore would have hashed the read only bit) in the thin link where it was marked read only. So I am not sure offhand why we could ever share a referencing object between a link where it was read only and another link where it is not...   Hopefully Eugene can figure out what is going wrong here.
>>>>> 
>>>>> 
>>>>> Repository:
>>>>>   rL LLVM
>>>>> 
>>>>> https://reviews.llvm.org/D49362 <https://reviews.llvm.org/D49362>
>>>>> 
>>>>> 
>>>>> 
>>>> 
>>>> 
>>>> 
>>>> -- 
>>>> Teresa Johnson |	 Software Engineer |	 tejohnson at google.com <mailto:tejohnson at google.com> |
>>> 
>> 
>> 
>> 
>> -- 
>> Teresa Johnson |	 Software Engineer |	 tejohnson at google.com <mailto:tejohnson at google.com> |
> 
> 
> 
> -- 
> Teresa Johnson |	 Software Engineer |	 tejohnson at google.com <mailto:tejohnson at google.com> |	

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20181114/249ac5dc/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: b.out.tar.gz
Type: application/x-gzip
Size: 6791 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20181114/249ac5dc/attachment.bin>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20181114/249ac5dc/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: gv.tar.gz
Type: application/x-gzip
Size: 9811 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20181114/249ac5dc/attachment-0001.bin>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20181114/249ac5dc/attachment-0002.html>