Linux gfortran-13 code missing library #4473

acombescure · 2023-05-02T09:32:42Z

acombescure
May 2, 2023

Additional Info

I am using Windows Subsystem for Linux

Output of `brew config`

HOMEBREW_VERSION: 4.0.16
ORIGIN: https://github.com/Homebrew/brew
HEAD: 01e0c20d01c6aab28d844280d130f0e745e2ddd9
Last commit: 2 days ago
Core tap JSON: 01 May 20:52 UTC
HOMEBREW_PREFIX: /opt/homebrew
HOMEBREW_CASK_OPTS: []
HOMEBREW_DISPLAY: /private/tmp/com.apple.launchd.QfXfe7D7TZ/org.xquartz:0
HOMEBREW_MAKE_JOBS: 8
Homebrew Ruby: 2.6.10 => /System/Library/Frameworks/Ruby.framework/Versions/2.6/usr/bin/ruby
CPU: octa-core 64-bit arm_blizzard_avalanche
Clang: 14.0.3 build 1403
Git: 2.39.2 => /Applications/Xcode.app/Contents/Developer/usr/bin/git
Curl: 7.87.0 => /usr/bin/curl
macOS: 13.3.1-arm64
CLT: 14.3.0.0.1.1679647830
Xcode: 14.3
Rosetta 2: false

Output of `brew doctor`

Your system is ready to brew.

Description of issue

i have compiled and linked with gfortran-13 uploaded with the brew command brew install gfortran
a fortran program of about 1000 subroutines that works on all machines since 40 Years.
It was nicely working on my previous mac (OS 10)
the compilation and link stage have no problem

but when i try to launch the program i get a library problem. described in the following lines...

dyld[34032]: dyld cache '(null)' not loaded: syscall to map cache into shared region failed
dyld[34032]: Library not loaded: /usr/lib/libSystem.B.dylib
Referenced from: <281461F9-AB34-36A4-9311-FF47FC96240A> /Users/alaincombescure/Documents/Stanlax/dinca/coq/incaessai
Reason: tried: '/usr/lib/libSystem.B.dylib' (no such file), '/System/Volumes/Preboot/Cryptexes/OS/usr/lib/libSystem.B.dylib'

what can i do to overcome this apparently missing library problem ?(libSystem.B.dylib not found)

thanks for the help..

SMillerDev · 2023-05-02T10:02:52Z

SMillerDev
May 2, 2023

Afaik GCC on ARM macs is still experimental and officially unsupported so you'd have to report this to https://github.com/iains/gcc-13-branch who created the patch we're using.

0 replies

fxcoudert · 2023-05-02T10:49:00Z

fxcoudert
May 2, 2023

Can you compile and run this short test program?

meau /tmp $ cat a.f90 
  real :: x = 2.
  print *, sqrt(x)
end
meau /tmp $ gfortran a.f90 && ./a.out
   1.41421354    
meau /tmp $ otool -L ./a.out         
./a.out:
	/opt/homebrew/opt/gcc/lib/gcc/current/libgfortran.5.dylib (compatibility version 6.0.0, current version 6.0.0)
	/opt/homebrew/opt/gcc/lib/gcc/current/libquadmath.0.dylib (compatibility version 1.0.0, current version 1.0.0)
	/usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 1319.100.3)

0 replies

acombescure · 2023-05-02T11:49:31Z

acombescure
May 2, 2023
Author

Thanks … I get the following parse error…. Sorry but I do not know what to do… ***@***.*** coq % meau /tmp $ cat a.f90 real :: x = 2. print *, sqrt(x) end meau /tmp $ gfortran a.f90 && ./a.out 1.41421354 meau /tmp $ otool -L ./a.out ./a.out: /opt/homebrew/opt/gcc/lib/gcc/current/libgfortran.5.dylib (compatibility version 6.0.0, current version 6.0.0) /opt/homebrew/opt/gcc/lib/gcc/current/libquadmath.0.dylib (compatibility version 1.0.0, current version 1.0.0) /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 1319.100.3) zsh: parse error near `end’ Regards

…

Le 2 mai 2023 à 12:49, FX Coudert ***@***.***> a écrit : Can you compile and run this short test program? meau /tmp $ cat a.f90 real :: x = 2. print *, sqrt(x) end meau /tmp $ gfortran a.f90 && ./a.out 1.41421354 meau /tmp $ otool -L ./a.out ./a.out: /opt/homebrew/opt/gcc/lib/gcc/current/libgfortran.5.dylib (compatibility version 6.0.0, current version 6.0.0) /opt/homebrew/opt/gcc/lib/gcc/current/libquadmath.0.dylib (compatibility version 1.0.0, current version 1.0.0) /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 1319.100.3) — Reply to this email directly, view it on GitHub <#4473 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A7MEWFUDIYHSAMEUINALK23XEDRCNANCNFSM6AAAAAAXSZG6DU>. You are receiving this because you authored the thread.

1 reply

fxcoudert May 2, 2023

Create a file named a.f90 with the following content:

  real :: x = 2.
  print *, sqrt(x)
end

Then run gfortran a.f90 && ./a.out
Then run otool -L ./a.out

acombescure · 2023-05-02T15:28:35Z

acombescure
May 2, 2023
Author

Thanks, It works Here is the result The program seems to find the library ***@***.*** coq % gfortran a.f90 && ./a.out 1.41421354 ***@***.*** coq % otool -L ./a.out ./a.out: /opt/homebrew/opt/gcc/lib/gcc/current/libgfortran.5.dylib (compatibility version 6.0.0, current version 6.0.0) /opt/homebrew/opt/gcc/lib/gcc/current/libquadmath.0.dylib (compatibility version 1.0.0, current version 1.0.0) /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 1319.100.3) It also works with gfortran-13 Regards Alain

…

Le 2 mai 2023 à 14:20, FX Coudert ***@***.***> a écrit : Create a file named a.f90 with the following content: real :: x = 2. print *, sqrt(x) end Then run gfortran a.f90 && ./a.out Then run otool -L ./a.out — Reply to this email directly, view it on GitHub <#4473 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A7MEWFSC4BOZQ3F25FBZLWDXED32RANCNFSM6AAAAAAXSZG6DU>. You are receiving this because you authored the thread.

1 reply

fxcoudert May 2, 2023

So it works on a simple code, meaning the compiler is functional. Why it fails on your real example, I cannot tell, you would have to reduce it to a simpler example that we can reproduce. The real error is this line:

dyld[34032]: dyld cache '(null)' not loaded: syscall to map cache into shared region failed

dyld should be able to load its cache, and the cache should not be null. Why that occurs, I have no idea.

acombescure · 2023-05-02T15:59:44Z

acombescure
May 2, 2023
Author

thanks Here is the linking phase shell #! /bin/bash echo ">>>>> edition de liens de INCA <<<<" #################### gfortran time gfortran-13 -O -o./incaessai /Users/alaincombescure/Documents/Stanlax/castem/inca/main.o /Users/alaincombescure/Documents/Stanlax/castem/inca/inca.a chmod +x ./incaessai exit Do you see something wrong Regards Alain

…

Le 2 mai 2023 à 17:49, FX Coudert ***@***.***> a écrit : So it works on a simple code, meaning the compiler is functional. Why it fails on your real example, I cannot tell, you would have to reduce it to a simpler example that we can reproduce. The real error is this line: dyld[34032]: dyld cache '(null)' not loaded: syscall to map cache into shared region failed dyld should be able to load its cache, and the cache should not be null. Why that occurs, I have no idea. — Reply to this email directly, view it on GitHub <#4473 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A7MEWFSZY4E2BP5QLDZOYG3XEEUG7ANCNFSM6AAAAAAXSZG6DU>. You are receiving this because you authored the thread.

1 reply

fxcoudert May 2, 2023

No, I do not seen anything wrong with that. In 90% of errors I can get a feeling of what went wrong, or broadly where the issue is. But in your case it's a new one, I've never seen it. It's either something weird, or something whose cause is deep.

Given it's a dyld cache issue, among the “weird” category: are you running macOS in a weird state? (recovery mode, or with SIP protection disabled)

acombescure · 2023-05-02T19:05:56Z

acombescure
May 2, 2023
Author

Thanks No I am not on a recovery mode…nor did disabled SIP protection .. (at least I have not done this on purpose) Regards

…

Le 2 mai 2023 à 18:31, FX Coudert ***@***.***> a écrit : No, I do not seen anything wrong with that. In 90% of errors I can get a feeling of what went wrong, or broadly where the issue is. But in your case it's a new one, I've never seen it. It's either something weird, or something whose cause is deep. Given it's a dyld cache issue, among the “weird” category: are you running macOS in a weird state? (recovery mode, or with SIP protection disabled) — Reply to this email directly, view it on GitHub <#4473 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A7MEWFTQ4E2IAZ7QZK4HL7TXEEZGLANCNFSM6AAAAAAXSZG6DU>. You are receiving this because you authored the thread.

0 replies

acombescure · 2023-05-02T20:29:44Z

acombescure
May 2, 2023
Author

I have found the reason for dylib cache .. Rather simple I had chosen to reserve a place for memory of 500 000 000 R*8 When I diminishes this value to 250 000 000 no more library problem This is annoying.. on Mac OS X 500 000 000 was OK Regards Alain

…

Le 2 mai 2023 à 18:31, FX Coudert ***@***.***> a écrit : No, I do not seen anything wrong with that. In 90% of errors I can get a feeling of what went wrong, or broadly where the issue is. But in your case it's a new one, I've never seen it. It's either something weird, or something whose cause is deep. Given it's a dyld cache issue, among the “weird” category: are you running macOS in a weird state? (recovery mode, or with SIP protection disabled) — Reply to this email directly, view it on GitHub <#4473 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A7MEWFTQ4E2IAZ7QZK4HL7TXEEZGLANCNFSM6AAAAAAXSZG6DU>. You are receiving this because you authored the thread.

0 replies

fxcoudert · 2023-05-02T21:38:59Z

fxcoudert
May 2, 2023

You can allocate that amount dynamically, it works:

meau /tmp $ cat a.f90 
  real(kind=8), allocatable :: x(:)
  allocate(x(500000000))
  x(1) = 2
  print *, sqrt(x(1))
  print *, sizeof(x) / 1024 / 1024, 'MB'
end
meau /tmp $ gfortran a.f90 && ./a.out 
   1.4142135623730951     
                 3814 MB

0 replies

acombescure · 2023-05-03T04:46:51Z

acombescure
May 3, 2023
Author

Thanks a lot I will do this Regards Alain

…

Le 2 mai 2023 à 23:39, FX Coudert ***@***.***> a écrit : You can allocate that amount dynamically, it works: meau /tmp $ cat a.f90 real(kind=8), allocatable :: x(:) allocate(x(500000000)) x(1) = 2 print *, sqrt(x(1)) print *, sizeof(x) / 1024 / 1024, 'MB' end meau /tmp $ gfortran a.f90 && ./a.out 1.4142135623730951 3814 MB — Reply to this email directly, view it on GitHub <#4473 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A7MEWFSMCN3J6WNP5BV23UTXEF5H3ANCNFSM6AAAAAAXSZG6DU>. You are receiving this because you authored the thread.

0 replies

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Homebrew

Linux gfortran-13 code missing library #4473

{{title}}

Replies: 9 comments 3 replies

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

Select a reply

Linux gfortran-13 code missing library #4473

Additional Info

Output of brew config

Output of brew doctor

Description of issue

Replies: 9 comments · 3 replies

acombescure May 2, 2023 Author

acombescure May 2, 2023 Author

acombescure May 2, 2023 Author

acombescure May 2, 2023 Author

acombescure May 2, 2023 Author

acombescure May 3, 2023 Author

Output of `brew config`

Output of `brew doctor`

Replies: 9 comments 3 replies

acombescure
May 2, 2023
Author

acombescure
May 2, 2023
Author

acombescure
May 2, 2023
Author

acombescure
May 2, 2023
Author

acombescure
May 2, 2023
Author

acombescure
May 3, 2023
Author