racket-cs build failure JFYI (fixed)

15 views
Skip to first unread message

John Clements

unread,
May 25, 2020, 9:52:47 PM5/25/20
to Racket Developers
Hmm… building racket cs is failing for me like this right now:

compile: /Users/clements/racket-cs/racket/src/io/port/custom-output-port.rkt
...during /Users/clements/racket-cs/racket/src/io/port/custom-output-port.rkt...
...during /Users/clements/racket-cs/racket/src/io/port/main.rkt...
...during /Users/clements/racket-cs/racket/src/io/main.rkt...
port/custom-output-port.rkt:2:18: only-in: identifier `unsafe-bytes->immutable-bytes!' not included in nested require spec
at: racket/unsafe/ops
in: (only-in racket/unsafe/ops unsafe-bytes->immutable-bytes!)
location...:
port/custom-output-port.rkt:2:18
context...:
raise-syntax-error
expand-import
try-next_223
require
/Users/clements/racket-cs/racket/src/expander/expand/main.rkt:400:0: apply-transformer-in-context
/Users/clements/racket-cs/racket/src/expander/expand/main.rkt:253:0: dispatch
/Users/clements/racket-cs/racket/src/expander/expand/local-expand.rkt:68:0: do-local-expand
do-wrapping-module-begin
/Users/clements/racket-cs/racket/src/expander/expand/main.rkt:400:0: apply-transformer-in-context
/Users/clements/racket-cs/racket/src/expander/expand/main.rkt:253:0: dispatch
/Users/clements/racket-cs/racket/src/expander/expand/module.rkt:713:2: loop
/Users/clements/racket-cs/racket/src/expander/expand/append.rkt:6:16: finish
/Users/clements/racket-cs/racket/src/expander/expand/module.rkt:300:7: pass-1-and-2-loop
/Users/clements/racket-cs/racket/src/expander/expand/module.rkt:221:3: module-begin-k
/Users/clements/racket-cs/racket/src/expander/expand/module.rkt:86:0: expand-module
/Users/clements/racket-cs/racket/src/expander/expand/module.rkt:47:1
...
make[11]: *** [io-src-generate] Error 1
make[10]: *** [compiled/io.rktl] Error 2
make[9]: *** [build-racket-so] Error 2
make[8]: *** [bounce] Error 2
make[7]: *** [racket-so] Error 2
make[6]: *** [cs] Error 2
make[5]: *** [cs-after-racket-with-abs-paths] Error 2
make[4]: *** [cs-after-racket-with-racket-and-scheme-src] Error 2
make[3]: *** [cs-after-racket-with-racket] Error 2
make[2]: *** [cs-only] Error 2
make[1]: *** [plain-cs] Error 2
make: *** [cs] Error 2

This is immediately after a git pull and a git clean -xdf. I was compiling with RACKET=racket, using a built-in installation of racket v7.7.0.1.

I rebuilt the racket I was using as the helper, and this error went away. I was suprised; it didn’t look like the kind of error that would be caused by an out-of-date helper. Ah well.

John



Matthew Flatt

unread,
May 25, 2020, 10:04:00 PM5/25/20
to John Clements, Racket Developers
At 25 May 2020 21:52:43 -0400, "'John Clements' via Racket Developers" wrote:
> This is immediately after a git pull and a git clean -xdf. I was compiling
> with RACKET=racket, using a built-in installation of racket v7.7.0.1.
>
> I rebuilt the racket I was using as the helper, and this error went
> away. I was suprised; it didn’t look like the kind of error that
> would be caused by an out-of-date helper. Ah well.

Yes, this is as intended: RACKET=racket requires the same version for
both the helper and the Racket being built.

John Clements

unread,
May 26, 2020, 1:56:38 AM5/26/20
to Matthew Flatt, Racket Developers
Okay, that makes sense. Can I suggest a friendly amendment to the following section of build.md? (the only change is in the first line):

Current:

If you have a sufficiently recent Racket installation already with at
least the `"compiler-lib"` package installed, you can supply
`RACKET=...` with `make cs` to skip that part of the bootstrap. And if
you have a Chez Scheme source directory already, you can supply that
with `SCHEME_SRC=<dir>` instead of downloading a new copy:

Proposed:

If you have already compiled the current version of Racket with at
least the `"compiler-lib"` package installed, you can supply
`RACKET=...` with `make cs` to skip that part of the bootstrap. And if
you have a Chez Scheme source directory already, you can supply that
with `SCHEME_SRC=<dir>` instead of downloading a new copy:

John
> --
> You received this message because you are subscribed to the Google Groups "Racket Developers" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to racket-dev+...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/racket-dev/20200525200357.154%40sirmail.smtp.cs.utah.edu.



Sam Tobin-Hochstadt

unread,
May 26, 2020, 6:24:09 AM5/26/20
to John Clements, Racket Developers
The list of identifiers exported by the kernel are compiled into the
expander -- the expander doesn't maintain its own list. So this is an
expected sort of error with an old bootstrap copy of Racket.

Sam
> This is immediately after a git pull and a git clean -xdf. I was compiling with RACKET=racket, using a built-in installation of racket v7.7.0.1.
>
> I rebuilt the racket I was using as the helper, and this error went away. I was suprised; it didn’t look like the kind of error that would be caused by an out-of-date helper. Ah well.
>
> John
>
>
>
> --
> You received this message because you are subscribed to the Google Groups "Racket Developers" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to racket-dev+...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/racket-dev/97ea467d-7a19-4708-8442-9fa20cb5e354%40mtasv.net.
Reply all
Reply to author
Forward
0 new messages