`RSHAPE_PARENT` is error prone because it returns a raw untagged
shape_id.
To check if a shape is a direct parent of another, tags should be
discarded. So providing a comparison function is better than exposing
untagged ids.
On the ruby side, this fixes a crash for methods with 39 or more
parameters. We used to miscomp those entry points due to Insn::Lea
picking ADDS which cannot reference SP:
# set method params: 40
mov x0, #0xfee8
movk x0, #0xffff, lsl #16
movk x0, #0xffff, lsl #32
movk x0, #0xffff, lsl #48
adds x0, xzr, x0
Have Lea work for all i32 displacements and avoid involving the split
pass. Previously, direct use of Insn::Lea directly from the user (as
opposed to generated by the split pass for some memory operations)
wasn't split, so being able to handle the whole range in arm64_emit()
was implicitly required. Also, not going through split reduces register
pressure.
This includes:
- Update test keys to the generic rsa-{1,2,3}.pem.
- Add omissions for enveloped-data tests so that the rest can be
tested in the FIPS mode.
- Add tests for PKCS7#error_string and #data.
- Check more error paths.
- Various style fixes.
58f0022de3
Set the error_string attribute to nil if PKCS7_verify() succeeds, since
the error queue should be empty in that case.
With AWS-LC, OpenSSL::PKCS#verify currently sets error_string to
"invalid library (0)" when the verification succeeds, whereas with
OpenSSL and LibreSSL, it becomes nil. ERR_reason_error_string() appears
to behave differently when an invalid error code is passed.
The branch to raise OpenSSL::PKCS7::PKCS7Error is removed because it
does not appear to be reachable.
c11c6631fa
ARGF supports compaction, but it pins all of its references, which means
that none of it can move. This commit changes it to actually support
compaction.
This reverts commit ec01cd9bbb.
This should no longer break the tests, now that the following changes
have been applied:
- RubyGems change: 32977f3869
- ruby/openssl change: e8261963c7
Commit ef277083ba overlooked a caller of ossl_x509_new() with NULL
argument. OpenSSL::X509::StoreContext#current_cert may not have a
certificate to return if StoreContext#verify has not been called.
4149b43890
During Ruby's shutdown, we no longer need to check the fstr of the symbol
because we don't use the fstr anymore for freeing the symbol. This can also
fix the following ASAN error:
==2721247==ERROR: AddressSanitizer: use-after-poison on address 0x75fa90a627b8 at pc 0x64a7b06fb4bc bp 0x7ffdf95ba9b0 sp 0x7ffdf95ba9a8
READ of size 8 at 0x75fa90a627b8 thread T0
#0 0x64a7b06fb4bb in RB_BUILTIN_TYPE include/ruby/internal/value_type.h:191:30
#1 0x64a7b06fb4bb in rb_gc_shutdown_call_finalizer_p gc.c:357:18
#2 0x64a7b06fb4bb in rb_gc_impl_shutdown_call_finalizer gc/default/default.c:3045:21
#3 0x64a7b06fb4bb in rb_objspace_call_finalizer gc.c:1739:5
#4 0x64a7b06ca1b2 in rb_ec_finalize eval.c:165:5
#5 0x64a7b06ca1b2 in rb_ec_cleanup eval.c:256:5
#6 0x64a7b06c98a3 in ruby_cleanup eval.c:179:12
The test was added in #5221 4 years ago but:
1. The insn it targets was removed in 2022 in #6187
2. The YJIT API `blocks_for` seems to be dropped in 2022 when it switched
to use Rust in #5826
So this test has not been run in more than 3 years and can't be run
anymore. I think we can remove it.
If a reference marked weak becomes a special const, it will crash because
it is not a GC handled object. We should skip special consts here.
870a79426b
* Port a Oniguruma patch: Integer overflow in forward_search_range()
db64ef3189
Co-Authored-By: K.Kosako <kkos@users.noreply.github.com>
* Port a Oniguruma patch: Integer overflow in backward_search_range() and onig_search_gpos()
bfc36d3d81
Co-Authored-By: K.Kosako <kkos@users.noreply.github.com>
* Port a Oniguruma patch: Integer overflow in onig_search_gpos()
778a43dd56
It differs from the Oniguruma patch in that it dosen't use `onigenc_get_prev_char_head()`
because this function's signature has been changed by Oniguruma and the change is not ported
in Onigmo for now. This patch respects the current Onigmo implementation.
Co-Authored-By: K.Kosako <kkos@users.noreply.github.com>
* Add castings to prevent warnings
* Correct castings to use OnigDistance
---------
Co-authored-by: K.Kosako <kkos@users.noreply.github.com>
The generic Ruby platform was getting unconditionally added in
truffleruby, preventing resolution in situations where there's no
generic ruby version (sorbet-static). Instead, the generic platform
should be considered per dependency, not globally.
a96afc5351
After digging into git history a bit, I figure this was about unifying
`bundle cache` and `bundle package`, which already happened a while ago.
So remove this TODO since it's now misleading.
5a0b06b84d
