[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: [tor-bugs] #18934 [Core Tor/Tor]: test suite failures with expensive hardening.
#18934: test suite failures with expensive hardening.
----------------------------+------------------------------------
Reporter: weasel | Owner: nickm
Type: defect | Status: needs_review
Priority: Medium | Milestone: Tor: 0.2.9.x-final
Component: Core Tor/Tor | Version: Tor: unspecified
Severity: Normal | Resolution:
Keywords: review-group-1 | Actual Points:
Parent ID: | Points: very-small
Reviewer: | Sponsor:
----------------------------+------------------------------------
Changes (by cypherpunks):
* status: reopened => needs_review
Comment:
Turns out the solution was archived in
`contrib/clang/sanitize_blacklist.txt`.
{{{
# we need to allow the tor bt handler to catch SIGSEGV
# otherwise address sanitizer munges the expected output and the test
fails
# we can do this by setting an environmental variable
# See https://code.google.com/p/address-sanitizer/wiki/Flags
# ASAN_OPTIONS=allow_user_segv_handler=1
}}}
Unfortunately the `allow_user_segv_handler` is not supported by older GCC
versions but they do support `handle_segv=0` which prevents
AddressSanitizer from installing its own segfault handler. The patch
attachment:0001-Prevent-ASAN-from-registering-a-SIGSEGV-handler.patch
fixed the issue for me.
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/18934#comment:15>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
_______________________________________________
tor-bugs mailing list
tor-bugs@xxxxxxxxxxxxxxxxxxxx
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs