Oh, a 32bit architecture. That's something new. Can you re-run the last command (the one that segfalts) with -ddump-cmm added to the command line? This will allow me to see offending Cmm - I should be able to diagnoze problem based on it. Janek Dnia niedziela, 8 września 2013, Austin Seipp napisał:

Hi Kazu,

I believe this is related to http://ghc.haskell.org/trac/ghc/ticket/8205 - can you try:

$ git revert d61c3ac186c94021c851f7a2a6d20631e35fc1ba

and see if your build finishes? Perhaps you can help Jan diagnose it if so...

On Sun, Sep 8, 2013 at 4:13 AM, Kazu Yamamoto wrote:

...

Hi,

I cannot build GHC head on Mac today:

"inplace/bin/ghc-stage1" -hisuf hi -osuf o -hcsuf hc -static -H32m -O -package-name ghc-7.7.20130907 -hide-all-packages -i -icompiler/basicTypes -icompiler/cmm -icompiler/codeGen -icompiler/coreSyn -icompiler/deSugar -icompiler/ghci -icompiler/hsSyn -icompiler/iface -icompiler/llvmGen -icompiler/main -icompiler/nativeGen -icompiler/parser -icompiler/prelude -icompiler/profiling -icompiler/rename -icompiler/simplCore -icompiler/simplStg -icompiler/specialise -icompiler/stgSyn -icompiler/stranal -icompiler/typecheck -icompiler/types -icompiler/utils -icompiler/vectorise -icompiler/stage2/build -icompiler/stage2/build/autogen -Icompiler/stage2/build -Icompiler/stage2/build/autogen -Icompiler/. -Icompiler/parser -Icompiler/utils -Icompiler/../rts/dist/build -Icompiler/stage2 -optP-DGHCI -optP-include -optPcompiler/stage2/build/autogen/cabal_macros.h -package Cabal-1.18.0 -package array-0.4.0.2 -package base-4.7.0.0 -package bin-package-db-0.0.0.0 -package bytestring-0.10

.3.0 -package containers-0.5.3.1 -package directory-1.2.0.1 -package filepath-1.3.0.2 -package hoopl-3.10.0.0 -package hpc-0.6.0.1 -package process-1.2.0.0 -package template-haskell-2.9.0.0 -package time-1.4.1 -package transformers-0.3.0.0 -package unix-2.7.0.0 -Wall -fno-warn-name-shadowing -XHaskell98 -XCPP -XMagicHash -XUnboxedTuples -XPatternGuards -XForeignFunctionInterface -XEmptyDataDecls -XTypeSynonymInstances -XMultiParamTypeClasses -XFlexibleInstances -XRankNTypes -XScopedTypeVariables -XDeriveDataTypeable -XBangPatterns -XNondecreasingIndentation -DGHCI_TABLES_NEXT_TO_CODE -DSTAGE=2 -O2 -no-user-package-db -rtsopts -odir compiler/stage2/build -hidir compiler/stage2/build -stubdir compiler/stage2/build -dynamic-too -c compiler/deSugar/Coverage.lhs -o compiler/stage2/build/Coverage.o -dyno compiler/stage2/build/Coverage.dyn_o

...

ghc-stage1: panic! (the 'impossible' happened) (GHC version 7.7.20130907 for i386-apple-darwin): expectJust block_order

Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug

make[1]: *** [compiler/stage2/build/Parser.o] Error 1 make[1]: *** Deleting file `compiler/stage2/build/Parser.o' make[1]: *** Waiting for unfinished jobs.... make: *** [all] Error 2

--Kazu _______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs

Good, we have a culprit then. Kazu, would you be willing to do a couple of things to provide me with more information? I'd need the Cmm dump that causes the panic. Also, tomorrow I'll add more information to this panic so that we get back name of offending code block - this should make debugging easier for me. Janek Dnia niedziela, 8 września 2013, Kazu Yamamoto napisał:

Hi,

...

I believe this is related to http://ghc.haskell.org/trac/ghc/ticket/8205 - can you try:

$ git revert d61c3ac186c94021c851f7a2a6d20631e35fc1ba

and see if your build finishes? Perhaps you can help Jan diagnose it if so...

Yes. After this command, I can build GHC head on Mac.

--Kazu

_______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs

Hello Kazu, thank you for the help! Here is the easiest way to get the output. First, run 'make' once by itself and make sure the build fails pretty quickly as a result of your error (always good to double check.) Next, when you do that, you will see the command that caused that error. It's the one you posted in the original email, in fact - the command invoking "ghc-stage1". You can copy and paste that command right back into the terminal and run it in order to reproduce the failure. Make sure you're at the root of the GHC source tree as well: $ cd ~/path/to/ghc-HEAD $ "inplace/bin/ghc-stage1" -hisuf hi -osuf ... -dyno compiler/stage2/build/Coverage.dyn_o This should fail with the exact same error. Finally, all you need to do is this: 1. Add -ddump-cmm to the command line. Anywhere will work. 2. Pipe the output somewhere. Alternatively, you can copy/paste it out of your terminal buffer. So: $ "inplace/bin/ghc-stage1" -ddump-cmm -hisuf hi -osuf ... -dyno compiler/stage2/build/Coverage.dyn_o 2>&1 > log.txt Double check the log.txt file - it should contain a lot of C-- output (thousands of lines, probably.) That's what Jan needs. On Sun, Sep 8, 2013 at 4:24 PM, Kazu Yamamoto wrote:

Hi,

...

Good, we have a culprit then. Kazu, would you be willing to do a couple of things to provide me with more information? I'd need the Cmm dump that causes the panic. Also, tomorrow I'll add more information to this panic so that we get back name of offending code block - this should make debugging easier for me.

OK. I understand more information is necessary for you. I can help. Please tell me exact operation sequences.

--Kazu

_______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs

-- Regards, Austin - PGP: 4096R/0x91384671

Hi,

Fortunately or unfortunately, I can build GHC head today without any reverting. I will reset the GHC tree to yesterday and try again.

Never mind. This is my misunderstanding. I cannot build GHC head today. I sent necessary information to Jan and Austin personally. --Kazu

Hi,

Thanks Kazu! This should get me going. Can you temporarily work with my commit reverted?

Yes. I got this log without reverting. And now I'm using GHC head with your commit reverted.

It's interesting that this time you got failure for a different file than in your original email to the list. Seems like the error was happening randomly, but according to my best knowledge ot shouldn't (I mean thing that I did that most likely breaks Mac build should be completely deterministic).

That's probably because I used different GHC trees yesterday and today. I did "git pull". Since this error happens at very late stage of building, I misunderstood that this problem was gone today. Yesterday, I got the error in very early stage as Austin described. --Kazu