[core] Fix early return in AutoloadLibraryGenerator

[hahnjo]

Fixes commit e601bf8 and then removes the check from AutoloadLibraryMU (as I had requested in the original PR review).

[hahnjo]

Some numbers, following Vincenzo's methodology in #14287 (comment):

master

 $ time PYTHONPATH=lib/ python -c 'import ROOT;print(ROOT.gErrorIgnoreLevel);print(ROOT.kError)'
-1
3000

real    0m0.485s
user    0m0.384s
sys     0m0.099s

 $ PYTHONPATH=lib/ strace -e file python -c 'import ROOT;print(ROOT.gErrorIgnoreLevel);print(ROOT.kError)' 2>&1 | awk -v FS='"' '{ print $1 }' | sort | uniq -c
      1 -1
      1 3000
    163 access(
      2 chdir(
      1 execve(
      1 +++ exited with 0 +++
     94 getcwd(
    493 lstat(
    625 openat(AT_FDCWD, 
     92 readlink(
      1 --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=2429627, si_uid=1000, si_status=0, si_utime=0, si_stime=0} ---
      1 --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=2429629, si_uid=1000, si_status=0, si_utime=0, si_stime=0} ---
   8730 stat(

 $ PYTHONPATH=lib/ strace -z -f -o openat.log -e trace=open,openat python3 -c 'import ROOT;print(ROOT.gErrorIgnoreLevel);print(ROOT.kError)'
 $ grep openat openat.log | wc -l
453

cling-autoload

 $ time PYTHONPATH=lib/ python -c 'import ROOT;print(ROOT.gErrorIgnoreLevel);print(ROOT.kError)'
-1
3000

real    0m0.491s
user    0m0.387s
sys     0m0.103s

 $ PYTHONPATH=lib/ strace -e file python -c 'import ROOT;print(ROOT.gErrorIgnoreLevel);print(ROOT.kError)' 2>&1 | awk -v FS='"' '{ print $1 }' | sort | uniq -c
      1 -1
      1 3000
    163 access(
      2 chdir(
      1 execve(
      1 +++ exited with 0 +++
     94 getcwd(
    493 lstat(
    625 openat(AT_FDCWD, 
     92 readlink(
      1 --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=2441749, si_uid=1000, si_status=0, si_utime=0, si_stime=0} ---
      1 --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=2441751, si_uid=1000, si_status=0, si_utime=0, si_stime=0} ---
   8730 stat(

 $ PYTHONPATH=lib/ strace -z -f -o openat.log -e trace=open,openat python3 -c 'import ROOT;print(ROOT.gErrorIgnoreLevel);print(ROOT.kError)'
 $ grep openat openat.log | wc -l
453

[hahnjo]

I think this fix is relatively low-risk, but just to be sure: @aandvalenzuela @smuzaffar can you maybe run this through CMSSW testing?

[smuzaffar]

I think this fix is relatively low-risk, but just to be sure: @aandvalenzuela @smuzaffar can you maybe run this through CMSSW testing?

cmssw tests are running via cms-sw#213

[github-actions]

Test Results

    17 files      17 suites   4d 2h 15m 16s ⏱️
 2 678 tests  2 677 ✅ 0 💤 1 ❌
43 752 runs  43 751 ✅ 0 💤 1 ❌

For more details on these failures, see this check.

Results for commit a14c665.

♻️ This comment has been updated with latest results.

[smuzaffar]

@hahnjo , cmssw tests passed cms-sw#213 (comment)

[hahnjo]

@smuzaffar thanks for testing!

[hahnjo]

ping @vgvassilev

[vgvassilev]

Great work, @hahnjo! LGTM!