-
Notifications
You must be signed in to change notification settings - Fork 553
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
make -j6 minitest failing #19829
Comments
If you use either a different version of |
|
There does indeed appear to be a compiler-specific aspect to this problem. On FreeBSD-12, I configured with a somewhat stripped-down version of your configuration invocation, using first First, some build-time warnings: Then: With However, I get through to the start of I don't often run |
|
With |
|
With |
|
Compiling with |
|
|
However, notwithstanding the failure in |
|
There appears to be a "test jobs" aspect to this problem in addition to the C-compiler problem. With the following configuration on the same (FreeBSD-12) machine, ... if I then run |
|
This problem emerged somewhere in the 5.35 development cycle. I compiled at the However, I'm currently compiling and building at the |
@hvds, in principle this should be bisectable, but I haven't been able to figure out the syntax for an invocation of |
I'll be busy with family stuff for a day or two, but I'll try a manual bisect when I have time - I don't have a clear failure mode for bisect.pl to detect. I do recall that there were recent changes to the build process that I thought risked impacting parallel builds and appeared to gave me occasional failures on |
|
This is potentially a symptom of the same underlying problem.
This is irrelevant, feel free to invoke with |
I've no idea what this is, but I'm pretty sure it's unrelated. |
Yes, I observed this in my OP (" |
minitest depends on $(MINIPERL_EXE) and minitest_prep, but minitest_prep calls back into make to build lib/Config.pm which indirectly depends on $(MINIPERL_EXE). This can result in a race between the parallel work of the parent make and the work of the child make, as they both try to build $(MINIPERL_EXE) and its dependencies, causing some of the errors described in the ticket. Note that trying to `make -j6 minitest all` could lead to similar conflicts, but I think that's an unlikely and not worth supporting anyway. Fixes Perl#19829 (I think)
minitest depends on $(MINIPERL_EXE) and minitest_prep, but minitest_prep calls back into make to build lib/Config.pm which indirectly depends on $(MINIPERL_EXE). This can result in a race between the parallel work of the parent make and the work of the child make, as they both try to build $(MINIPERL_EXE) and its dependencies, causing some of the errors described in the ticket. Note that trying to `make -j6 minitest all` could lead to similar conflicts, but I think that's an unlikely and not worth supporting anyway. Fixes Perl#19829 (I think)
minitest depends on $(MINIPERL_EXE) and minitest_prep, but minitest_prep calls back into make to build lib/Config.pm which indirectly depends on $(MINIPERL_EXE). This can result in a race between the parallel work of the parent make and the work of the child make, as they both try to build $(MINIPERL_EXE) and its dependencies, causing some of the errors described in the ticket. Note that trying to `make -j6 minitest all` could lead to similar conflicts, but I think that's an unlikely and not worth supporting anyway. Fixes #19829 (I think)
minitest depends on $(MINIPERL_EXE) and minitest_prep, but minitest_prep calls back into make to build lib/Config.pm which indirectly depends on $(MINIPERL_EXE). This can result in a race between the parallel work of the parent make and the work of the child make, as they both try to build $(MINIPERL_EXE) and its dependencies, causing some of the errors described in the ticket. Note that trying to `make -j6 minitest all` could lead to similar conflicts, but I think that's an unlikely and not worth supporting anyway. Fixes Perl#19829 (I think)
This failed three times in a row on blead@8fb00d81c6. The specific directory was different each time, but the general pattern was the same - shortly before the first "Permission denied" error it tries to make miniperl a second time for no obvious reason.
This was my default debug build:
./Configure -des -Dcc=gcc -Dprefix=/opt/scratch -Doptimize='-g -O6' -DDEBUGGING -Dusedevel -Uversiononly.make -j6 minitest_prepdid not fail, which is all the wierder (maybe I just got lucky).make minitestalso did not fail, which is less surprising.Any clues how to debug this?
make output (click to expand)
The text was updated successfully, but these errors were encountered: