Commit Graph

40607 Commits

Author SHA1 Message Date
Matthew Flatt
fa2ef64adb schemify: adjust simple? for multiple returns
The check for whether an expression is "simple" at the schemify layer
didn't take into account the expected arity.
2019-11-29 15:09:32 -07:00
Paulo Matos
430760ddb4
Add testing for 3m and CS in push CI (#2943) 2019-11-29 16:59:01 +01:00
Matthew Flatt
2920fc3f64 enable unreliable tests via PLT_RUN_UNRELIABLE_TESTS
The attempt in 7a12d55e7d to handle some flaky tests didn't work.
This attempt effectively removes the tests, although they can
be enabled by setting `PLT_RUN_UNRELIABLE_TESTS` to "timing".

Also, replace the old hack of enabling some locale-specific tests
to also use `PLT_RUN_UNRELIABLE_TESTS`.
2019-11-29 07:46:26 -07:00
Paulo Matos
b8d2513b6d
Disabling sync flaky test on CGC
Due to:
https://github.com/racket/racket/runs/325146638#step:8:33
2019-11-29 08:01:07 +01:00
Matthew Flatt
b62de0e14d adjust openssl test for Racket CS error message 2019-11-28 16:52:17 -07:00
Matthew Flatt
338955046d configure: repair for --enable-embedfw
`-S` needed for `strip` with some(?) versions of the Mac build tools.
2019-11-28 16:32:22 -07:00
Paulo Matos
8a6a34674a
Add racket-test-core tests to CGC push CI (#2934)
Renamed from `Add per-section coretests` since we diverged from initial intention.
2019-11-28 19:52:19 +01:00
Brian Wignall
9474b1a1e8 Fix spelling typos 2019-11-28 07:51:39 -07:00
Matthew Flatt
197d6db634 openssl: adjust peer-verif test for renegotiation
TLS v1.3 does not support renegotiation.
2019-11-28 07:31:33 -07:00
Matthew Flatt
83fb6d2ca8 fix race conditions in some tests
Closes #2940
2019-11-28 06:28:42 -07:00
Matthew Flatt
da70c65f8d skip a GC test for CGC
Closes #2939
2019-11-28 06:26:22 -07:00
Matthew Flatt
a410905c99 cs: module in some error messages only when enabled
Check `(error-print-source-location)` before adding a module name in
an error message about use/set before definition.
2019-11-28 05:51:48 -07:00
Matthew Flatt
8b00249265 schemify: another variable-order repair
Continues 70459f4105.
2019-11-28 05:39:11 -07:00
Paulo Matos
bb77f736b6 Revert "Simplify build jobs with anchors (#2938)"
This reverts commit 13c9e2d27b.

I was sure this passed CI on my fork (it didn't) - however,
it is now giving syntax errors and they might not yet be supported:
https://github.community/t5/GitHub-Actions/Support-for-YAML-anchors/m-p/30336
2019-11-28 00:16:11 +01:00
Paulo Matos
13c9e2d27b
Simplify build jobs with anchors (#2938) 2019-11-27 23:27:16 +01:00
Matthew Flatt
2f4e4c5e6a configure: add --enable-embedfw for Mac OS
Building with `--enable-embedfw` creates executables that link
statically to the Racket runtime system, instead of linking to a
Racket framework, and that embed boot files in the case of Racket CS.

Executables built this way are meant to cooperate better with code
signing.

Related to #2910
2019-11-27 13:15:19 -07:00
Matthew Flatt
85459e5fef openssl: repairs related to listen on port number 0
Allow 0 as the port number in `ssl-listen`, and fix
`ssl-addresses` to work on a listener. Update a test
to use these features so that it won't conflict with
other tests that listen on TCP addresses.
2019-11-27 13:15:19 -07:00
lkh01
627d164a46 make syntax->string work with pairs and _
fixes #1628 and #1629
2019-11-27 13:57:45 -05:00
Paulo Matos
4d4e75983e
Load core-tests instead of mz-tests (#2932)
`mz-tests.rktl` became `core-tests.rktl` in 29a0c44c98.
2019-11-27 14:46:14 +01:00
Matthew Flatt
ad7511dd29 place: fix place-kill for Racket thread simulation
Where true places are not supported, `place` is simulated
using Racket threads, and `place-kill` did not kill off the
simulated place well enough.

Relevant to #2930
2019-11-27 06:21:56 -07:00
Matthew Flatt
82f4067b9a avoid a memory/performance test for CGC
Related to #2930
2019-11-27 06:19:33 -07:00
Matthew Flatt
7a12d55e7d mark some flaky synchronization tests
Try out `flaky-test` as a way of marking a test that is not guaranteed
to pass (e.g., because it depends on timing), but where many failures
in a row are unlikely unless something is broken.

Closes #2936
2019-11-27 06:14:02 -07:00
Matthew Flatt
dcc26b1b90 adjust GC will test for CGC
Closes #2935
2019-11-27 05:51:26 -07:00
Matthew Flatt
70459f4105 schemify: preserve variable assignment order at module end
A sequence of definitions (or `define-values`) at the end of
a module could get reordered. That only matters for the order
of checks for attempting to assign to a constant, though.
2019-11-26 16:16:34 -07:00
Matthew Flatt
80a29553c9 schmify: remove some unnecessary wrap tests
In many places, we know that we're dealing with plain pairs.
2019-11-26 16:16:34 -07:00
Paulo Matos
bdce2c8be7 Revert "Add job to test racket CGC on Linux and macOS (#2918)"
This reverts commit 568433c1a9.

There are some flaky tests that need to be fixed. See differences between:
    - https://github.com/racket/racket/runs/321618267
    - https://github.com/racket/racket/runs/321344052
2019-11-26 19:44:42 +01:00
Matthew Flatt
0e87b8a73f cs & io: fix ellipses printing in structs that are partially opaque 2019-11-26 09:17:02 -07:00
Matthew Flatt
98f0c16b8c cs & io: name for default error value conversion handler 2019-11-26 08:55:35 -07:00
Matthew Flatt
e15a6f99d4 cs: more consistent arity errors
Suppress the compiler's conversion of statically obvious arity errors
to an expression that reports a custom error with the an S-expression,
since that interferes with Rumble's normalization of error messages.
2019-11-26 08:44:39 -07:00
Matthew Flatt
79981af50b cs: fix build-time arity check
Bootstrap properly to avoid problems on updates and version changes.
2019-11-26 08:44:39 -07:00
Matthew Flatt
afff5ca7ac cs: adjust default prompt abort handler arity error message 2019-11-26 08:44:39 -07:00
Matthew Flatt
6389b18b01 cs: fix prompt handler for module/eval prompt and multiple values 2019-11-26 08:44:39 -07:00
Paulo Matos
568433c1a9
Add job to test racket CGC on Linux and macOS (#2918) 2019-11-26 15:18:16 +01:00
Matthew Flatt
256f57bfd8 openssl: adjust to make XonX mode work for Catalina
Same repair as #2859, but for XonX mode.
2019-11-26 04:47:01 -07:00
Matthew Flatt
1048341c62 cs & io: fix mode argument when print-as-expression is #f 2019-11-25 20:05:23 -07:00
Matthew Flatt
4c981258f6 cs: improve location info in backtrace
Provide paths instead of strings, and include position and span
information.
2019-11-25 19:00:33 -07:00
Matthew Flatt
254f6a4f14 cs: fix 'nonatomic allocation in 32-bit mode 2019-11-25 16:23:52 -07:00
Matthew Flatt
f3cfac8981 cs: fix popcount in data/bit-vector for 32-bit mode
Related to #2925
2019-11-25 15:06:09 -07:00
Paulo Matos
b1b8beece5
Use AC_CHECK_MEMBERS to check for dirent name length flds (#2923)
If `struct dirent` has `namlen` field, define `HAVE_DIRENT_NAMLEN`.
If the field is instead `namelen`, define `HAVE_DIRENT_NAMELEN` (case of QNX). 

Use this checks in `rktio_fs.c` and simplify `platform.h`.
2019-11-25 16:35:10 +01:00
Matthew Flatt
219bac7fe7 cs configure: propagate preprocessor flags to rktio 2019-11-24 08:10:47 -05:00
Matthew Flatt
cd4ce30ca6 avoid compiler warning 2019-11-24 06:58:02 -05:00
Matthew Flatt
ebbf578b6c cs: fix memset address handling 2019-11-23 20:19:50 -05:00
Matthew Flatt
e0851e6753 cs: range check on seconds->date 2019-11-23 20:19:39 -05:00
Matthew Flatt
c611f126fd JIT: fix 32-bit non-Mac stack alignment
Fix function-call setup to align the stack to a 16-byte boundary on
all platforms.
2019-11-23 17:33:08 -05:00
Matthew Flatt
0bb23c3998 cs: improve checking for "uuid.h" 2019-11-23 17:27:24 -05:00
Paulo Matos
aecee7ff30
Fix typos on windows build README (#2919) 2019-11-23 22:00:21 +01:00
Matthew Flatt
d1bfa6a203 cs: fix error reporting for struct selectors
Fix error reporting for non-inlined struct selectors to use Racket
style instead of Chez Scheme style.

Closes #2926
2019-11-23 15:39:27 -05:00
Matthew Flatt
7eb1ef9250 cs: fix printing on struct type with custom write
Don't try to use a struct type's custom-write procedure on the struct
type itself.

Relevant to #2926
2019-11-23 15:32:10 -05:00
Matthew Flatt
cd048cb1d0 repair arity problems
Fix many incorrect arity declarations and actual arities in Racket CS,
and fix several incorrect arities in traditional Racket. Building
Racket CS now checks the information in "racket/src/cs/primitives"
against both Racket variants to make sure that they're all consistent.

Closes #2924
2019-11-23 15:32:10 -05:00
Matthew Flatt
91190bee63 continuation-prompt-available?: repairs CS and non-CS Racket
Racket CS did not support the optional second argument for
`continuation-prompt-available?`. Traditional racket did not produce a
sensible result for the prompt tag that is used to delimit a
composable continuation or in some cases for the default continuation
prompt tag.
2019-11-23 14:56:17 -05:00