vcswatch reports that
this package seems to have a new changelog entry (version
2026.03.11-1, distribution
unstable) and new commits
in its VCS. You should consider whether it's time to make
an upload.
Here are the relevant commit messages:
commit 77119492f91e2a3cce2e71dc5bdbcb62d397be83
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Sun Mar 15 13:01:35 2026 -0700
update 'watch' file version
Lintian says:
The version= line in the debian/watch file in this package declares an older
version. Please upgrade when you have a chance.
Reported-by: Lintian v2.131.0
Suggested-by: uscan 2.26.6
commit b1f6fe2f0d4b1bc402e1d987f5c3fc9e596489f3
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Sun Mar 15 13:01:35 2026 -0700
no longer set 'Rules-Requires-Root'
Lintian says:
This package currently sets the Rules-Requires-Root field in the
debian/control file to "no".
As of dpkg version 1.22.13, this field is set to "no" by default. As such, in
this case the Rules-Requires-Root field is redundant and should be removed.
Reported-by: Lintian v2.131.0
commit 404aa74520f786e7e44e054cc3f38bf23e1a2c45
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Sun Mar 15 13:01:35 2026 -0700
update Debian standards version 4.7.2 → 4.7.3
This lets us drop `priority` too as advised by the Upgrade Checklist.¹
¹ https://www.debian.org/doc/debian-policy/upgrading-checklist.html#version-4-7-3
Reported-by: Lintian v2.131.0
commit 1b426029a313af4fb037976c03a7919d939bb364
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Wed Mar 11 20:50:58 2026 -0700
update Debian changelog
commit 23ab3a25bf76cdaa558bdca4f8fdcf4380a75a35
Merge: eaa6169 f632c52
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Wed Mar 11 20:48:44 2026 -0700
Merge remote-tracking branch 'origin/main' into packaging/debian
commit f632c52d83a7507fb943e3f43ee3f8eb9d32fc9e
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Wed Mar 11 17:10:50 2026 -0700
release v2026.03.11
commit 2a07e1f65d85871665e77c7f144509c42cde6345
Merge: d080662 7cb037a
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Wed Mar 11 17:02:18 2026 -0700
Merge pull request #307 from Smattr/smattr/debian-1128916
fix sandboxed execution on Linux aarch64 with THP
commit d0806621565ba4e378329751bcdb094919d5fa69
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Tue Mar 10 21:57:08 2026 -0700
CI: upgrade Linux GCC 15.1 → 15.2
As per bd92714320d95317adf690826529a325f14d9269, we should be able to upgrade
these jobs by removing use of the Gold linker that is not installed in the newer
images.
commit 008369e4d01eb6717e440d377c867ceb7fc42f94
Merge: bd92714 960b6e5
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Tue Mar 10 21:55:21 2026 -0700
Merge pull request #308 from Smattr/smattr/72363b83-1767-45ab-81e9-267306f02e2f
fixes for '&', '|' inference
commit bd92714320d95317adf690826529a325f14d9269
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Tue Mar 10 17:23:32 2026 -0700
CI: upgrade Linux GCC 14.2 → 14.3
This seems obvious in retrospect, but the failures encountered in
cfed1801af86016b17e8db68ffbd2e10f7fc896e and similar attempts to upgrade are
presumably due to requesting the Gold linker and it not being available in the
Docker image. The Gold linker should not be required for the sanitizers on a
modern toolchain.
commit 7cb037ac5108b00a16f30cd9775e75b9cdac6d49
Author: Aurélien Jarno <aurel32@debian.org>
Date: Thu Feb 26 17:44:20 2026 -0800
fix sandboxed execution on Linux aarch64 with THP
To detect whether Transparent Huge Pages (THP) are enabled on 64-bit ARM, system
allocation routines read /sys/kernel/mm/transparent_hugepage/enabled using
`openat`. Description from Emanuele Rocca:
rumur fails to build from source on arm64 when using glibc 2.43, currently in
experimental…
The cause for the above failures is that the seccomp sandbox stops the
attempted open of /sys/kernel/mm/transparent_hugepage/enabled:
E AssertionError: model failed: …
[...]
E openat(AT_FDCWD, "/sys/kernel/mm/transparent_hugepage/enabled", O_RDONLY) = -1 ENETDOWN (Network is down)
E --- SIGSYS {si_signo=SIGSYS, si_code=SYS_SECCOMP, si_call_addr=0xe3ac16e98c60, si_syscall=__NR_openat, si_arch=AUDIT_ARCH_AARCH64} ---
E +++ killed by SIGSYS (core dumped) +++
One possible solution would be updating the seccomp filter to allow the
action above, but perhaps more discussion with glibc upstream is needed
to see if there are alternatives to opening a file under /sys, which can
cause problems in other scenarios too.
In the process of debugging this, Glibc made the decision to change to using
`madvise` instead of reading /sys/kernel/mm/transparent_hugepage/enabled to
detect THP. This resolves the `openat` failure within the sandbox, but now
incurs an `madvise` regardless if we are multithreaded or not. So this change
allows `madvise` unconditionally in the sandbox.
Commentary from Matthew Fernandez:
Unconditionally adding `madvise` as an allowable call does not meaningfully
expand the attack surface as I am not aware of anything security-relevant you
can do with `madvise`.
Particular praise to Aurélien’s thoroughness here. Rather than tweaking the
location where `madvise` was conditionally allowed, Aurélien identified that
semantically this now belongs earlier, with the syscalls we are enabling due
to malloc’s actions. This kind of careful attention to detail is what saves
churn and countless downstream merge conflicts.
¹ https://inbox.sourceware.org/libc-alpha/PAWPR08MB8982EEFF3D3C7A2B3DB803D68373A@PAWPR08MB8982.eurprd08.prod.outlook.com/
Debian: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1128916
Reported-by: Emanuele Rocca <ema@debian.org>
Tested-by: Aurélien Jarno <aurel32@debian.org>
commit 960b6e50524d81612bd3474ad35a507f7e25497e
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Tue Mar 10 17:01:23 2026 -0700
fix resolution of ambiguous '|' with typedefed LHS
One of the Rumur extensions to the Murphi language is to overload the `|`
operator to mean either logical-or (the original Murphi semantics) or
bitwise-or. To achieve this, the LHS of the expression is inspected to
disambiguate which of these operations the `|` is. Prior to this change, the `|`
would only be resolved to a bitwise-or when the LHS was of a bare range type.
This change makes this rule also apply when the type of the LHS is (recursively)
a type alias of a range type.
commit 15eb2c4569be44cee9d336d61b100abf4a119779
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Tue Mar 10 17:01:23 2026 -0700
add a test case for a failing '|' inference
commit 216567afdd447746d185255397faf9e85564e37a
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Tue Mar 10 17:01:23 2026 -0700
fix resolution of ambiguous '&' with typedefed LHS
One of the Rumur extensions to the Murphi language is to overload the `&`
operator to mean either logical-and (the original Murphi semantics) or
bitwise-and. To achieve this, the LHS of the expression is inspected to
disambiguate which of these operations the `&` is. Prior to this change, the `&`
would only be resolved to a bitwise-and when the LHS was of a bare range type.
This change makes this rule also apply when the type of the LHS is (recursively)
a type alias of a range type.
commit 3887c0e4f1ca7b94cac428607e16c2e7792550ca
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Tue Mar 10 17:01:23 2026 -0700
add a test case for a failing '&' inference
commit 096d3a4414cfd9ee1275565d667dff39b9e82830
Merge: e513577 9f89eae
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Tue Feb 24 18:49:10 2026 -0800
Merge pull request #306 from Smattr/smattr/c47e6e13-c83e-42a3-8673-ed1b94ef1a57
CI upgrades
commit 9f89eae3258b8d850d2994f8279d20752053df06
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Tue Feb 24 17:20:22 2026 -0800
CI: upgrade FreeBSD 14.2 → 14.3 and move off snaps
It seems the snaps no longer boot:
Failed to start an instance: INVALID_ARGUMENT: Snap images are not supported
due to boot stability.
I guess this is consistent with what we have previously observed in e.g.
65e1c89dee65c051a579e91eda8dea4f7d9ddd38. This picks the FreeBSD 14 image
suggested on https://cirrus-ci.org/guide/FreeBSD/.
commit 5fe4c6e6951e5d906428202af3931ba9b9986520
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Tue Feb 24 17:20:22 2026 -0800
CI: move FreeBSD 15 jobs off snaps
It seems the snaps no longer boot:
Failed to start an instance: INVALID_ARGUMENT: Snap images are not supported
due to boot stability.
I guess this is consistent with what we have previously observed in e.g.
65e1c89dee65c051a579e91eda8dea4f7d9ddd38. This arbitrarily picks one of the
FreeBSD 15 images suggested on https://cirrus-ci.org/guide/FreeBSD/.
commit f72ab811035cc7cffbd6a7702c1a085e18c28090
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Tue Feb 24 17:18:15 2026 -0800
CI: upgrade FreeBSD 13.4 → 13.5
FreeBSD 13.4 was EOLed on 2024-09-17¹ and its images are no longer available.
¹ https://www.freebsd.org/releases/
commit abf8b2f00f59c9335277aa45ef45da6fab885b45
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Tue Feb 24 17:02:12 2026 -0800
CI: upgrade Macports 2.11.5 → 2.12.1
commit e513577283900c3f8797ec5bcc1f89f3b62d1d71
Merge: 3eb7f81 cfed180
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Sat Sep 20 15:04:39 2025 -0700
Merge pull request #305 from Smattr/smattr/cb587e81-11d5-496c-9832-0c53d46ac56e
minor improvements
commit cfed1801af86016b17e8db68ffbd2e10f7fc896e
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Sat Sep 20 11:13:41 2025 -0700
CI: downgrade Linux GCC 14.3 → 14.2
The GCC 14.3 and 15.2 images seemingly contain a toolchain that does not work:
-- The CXX compiler identification is GNU 14.3.0
…
-- Check for working CXX compiler: /usr/local/bin/c++
-- Check for working CXX compiler: /usr/local/bin/c++ - broken
CMake Error at /usr/share/cmake-3.31/Modules/CMakeTestCXXCompiler.cmake:73 (message):
The C++ compiler
"/usr/local/bin/c++"
is not able to compile a simple test program.
It fails with the following output:
Change Dir: '/tmp/cirrus-ci-build/build/CMakeFiles/CMakeScratch/TryCompile-4iCyxS'
Run Build Command(s): /usr/bin/cmake -E env VERBOSE=1 /usr/bin/gmake -f Makefile cmTC_24040/fast
/usr/bin/gmake -f CMakeFiles/cmTC_24040.dir/build.make CMakeFiles/cmTC_24040.dir/build
gmake[1]: Entering directory '/tmp/cirrus-ci-build/build/CMakeFiles/CMakeScratch/TryCompile-4iCyxS'
Building CXX object CMakeFiles/cmTC_24040.dir/testCXXCompiler.cxx.o
/usr/local/bin/c++ -pedantic -Werror -Wno-error=overloaded-virtual -g -fno-omit-frame-pointer -fsanitize=address,undefined -fno-sanitize-recover=undefined -fuse-ld=gold -o CMakeFiles/cmTC_24040.dir/testCXXCompiler.cxx.o -c /tmp/cirrus-ci-build/build/CMakeFiles/CMakeScratch/TryCompile-4iCyxS/testCXXCompiler.cxx
Linking CXX executable cmTC_24040
/usr/bin/cmake -E cmake_link_script CMakeFiles/cmTC_24040.dir/link.txt --verbose=1
collect2: fatal error: cannot find 'ld'
compilation terminated.
/usr/local/bin/c++ -pedantic -Werror -Wno-error=overloaded-virtual -g -fno-omit-frame-pointer -fsanitize=address,undefined -fno-sanitize-recover=undefined -fuse-ld=gold CMakeFiles/cmTC_24040.dir/testCXXCompiler.cxx.o -o cmTC_24040
gmake[1]: *** [CMakeFiles/cmTC_24040.dir/build.make:102: cmTC_24040] Error 1
gmake[1]: Leaving directory '/tmp/cirrus-ci-build/build/CMakeFiles/CMakeScratch/TryCompile-4iCyxS'
gmake: *** [Makefile:134: cmTC_24040/fast] Error 2
Not too sure what is going on here, but lets see if downgrading GCC helps.
commit 26c8e97e142f8761d25b9e02579b0a6ae97aaeaf
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Fri Sep 19 18:35:06 2025 -0700
strength-reduce some single-character string printing
commit 1716046a592037031a8784cb0c248f315b98cda4
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Fri Sep 19 18:34:08 2025 -0700
use arrays instead of pointers for some constant strings
A slight reduction in complexity.
commit 9c68f077e97f49b99ed7201999b68253b2ceb785
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Fri Sep 19 18:33:37 2025 -0700
CI: upgrade Macports 2.10.5 → 2.11.5
commit 3eb7f81bfd7edd34216c2c46b26e526265817f47
Author: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Wed Sep 3 17:47:02 2025 -0700
update CMake build requirement note