CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: t...@cvs.openbsd.org2023/01/29 00:41:20 Modified files: sysutils/borgbackup/2.0: Makefile Log message: borgbackup: USE_NOEXECONLY is needed due to static linking with OpenSSL. Move it to the top and annotate it.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: t...@cvs.openbsd.org2023/01/29 00:38:26 Modified files: security/openssl/1.1: Makefile Log message: openssl/1.1: also need to bump borgbackup/2.0 on changes.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: t...@cvs.openbsd.org2023/01/29 00:31:30 Modified files: security/libgcrypt: Makefile Log message: zap trailing blank
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: na...@cvs.openbsd.org 2023/01/28 17:07:55 Modified files: x11/xpostit: Makefile Added files: x11/xpostit/patches: patch-menu_c patch-note_c patch-util_c patch-xpostit_h Log message: x11/xpostit: explicitly cast integer to pointer to fix build with clang 15 The callback takes a pointer to be able to pass arbitrary data, but instead of allocating an int and passing a pointer to it, the code jams the int directly into the pointer for an int -> pointer -> int roundtrip. Problem reported by and discussed with jca@. Fix implicit function declarations.
games/dMagnetic 0.35 -> 0.36
Hello.
I know that it has only been a couple of days, but my project saw
another release.
My apologies that this causes more work than it should.
Anyways, the patch is attached to this email, hopefully it can become
part of your
ports tree once more. The changes are best summed up in the changelog
for the
Debian package:
* Minor bugfixes
* Backend engine can be encapsulated in a library
* HelloWorld example for interested frontend developers
Thomas Dettbarn
diff --git a/games/dmagnetic/Makefile b/games/dmagnetic/Makefile
index e347b5f6fa9..828a139962f 100644
--- a/games/dmagnetic/Makefile
+++ b/games/dmagnetic/Makefile
@@ -1,6 +1,6 @@
COMMENT= interpreter for Magnetic Scrolls games
-V= 0.35
+V= 0.36
DISTNAME= dMagnetic_${V}
PKGNAME= dmagnetic-${V}
EXTRACT_SUFX= .tar.bz2
diff --git a/games/dmagnetic/distinfo b/games/dmagnetic/distinfo
index b08dc0fda07..387524b23e8 100644
--- a/games/dmagnetic/distinfo
+++ b/games/dmagnetic/distinfo
@@ -1,2 +1,2 @@
-SHA256 (dMagnetic_0.35.tar.bz2) = bq4hQ5q6B97tG9scKL7OraFpx7zzLPnN2GDHUbO0p58=
-SIZE (dMagnetic_0.35.tar.bz2) = 92233
+SHA256 (dMagnetic_0.36.tar.bz2) = C3YU5H9nEc5Yz77nGI1W0XqSwQQAT1GrYXmLMY6JfY8=
+SIZE (dMagnetic_0.36.tar.bz2) = 93540
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: na...@cvs.openbsd.org 2023/01/28 15:03:46 Modified files: devel/cmake: cmake.port.mk net/libtorrent-rasterbar: Makefile security/keepassxc: Makefile Log message: revert "Handle ALL_TARGET and INSTALL_TARGET in cmake module" It breaks a number of ports at the fake step: Unknown argument -t Usage: cmake --install [options]
Re: NEW: databases/py-lmdb
On Sat, Jan 28, 2023 at 08:48:52PM +, Stuart Henderson wrote: > Python bindings for LMDB. OK to import? > Something's wrong. When trying to run tests, I got E ModuleNotFoundError: No module named 'cffi' So I added py3-cffi and then got this: tests/env_test.py .F...Fatal Python error: Segmentation fault Looks like a NULL deref: 46718 python3.10 CALL thrkill(0,SIGSEGV,0) 46718 python3.10 PSIG SIGSEGV SIG_DFL addr=0x0 trapno=0 46718 python3.10 NAMI "python3.10.core" I couldn't immediately spot what's wrong in the python code. Happens on both amd64 and arm64. Current thread 0x0575aea4aa28 (most recent call first): File "/usr/ports/pobj/py-lmdb-1.4.0-python3/lmdb-1.4.0/lmdb/cffi.py", line 552 in _mvstr File "/usr/ports/pobj/py-lmdb-1.4.0-python3/lmdb-1.4.0/lmdb/cffi.py", line 1783 in item File "/usr/ports/pobj/py-lmdb-1.4.0-python3/lmdb-1.4.0/lmdb/cffi.py", line 1799 in _iter File "/usr/ports/pobj/py-lmdb-1.4.0-python3/lmdb-1.4.0/tests/env_test.py", line 286 in File "/usr/local/lib/python3.10/unittest/case.py", line 201 in handle File "/usr/local/lib/python3.10/unittest/case.py", line 738 in assertRaises File "/usr/ports/pobj/py-lmdb-1.4.0-python3/lmdb-1.4.0/tests/env_test.py", line 286 in test_close File "/usr/local/lib/python3.10/unittest/case.py", line 549 in _callTestMethod ... #0 thrkill () at /tmp/-:3 #1 0xd193c8d5061dc530 in ?? () #2 #3 0x057609261a46 in PyBytes_FromStringAndSize () from /usr/local/lib/libpython3.10.so.0.0 #4 0x05762af91d53 in mb_slice (self=0x5760d076480, left=, right=) at c/minibuffer.h:37 #5 mb_subscript (self=0x5760d076480, item=) at c/minibuffer.h:267 #6 0x0576093c15f4 in _PyEval_EvalFrameDefault () from /usr/local/lib/libpython3.10.so.0.0 #7 0x0576093b7251 in _PyEval_Vector () from /usr/local/lib/libpython3.10.so.0.0 #8 0x0576093c80e8 in call_function () from /usr/local/lib/libpython3.10.so.0.0 #9 0x0576093bcb5d in _PyEval_EvalFrameDefault () from /usr/local/lib/libpython3.10.so.0.0 #10 0x0576093b7251 in _PyEval_Vector () from /usr/local/lib/libpython3.10.so.0.0 #11 0x057609277b95 in method_vectorcall () from /usr/local/lib/libpython3.10.so.0.0 #12 0x0576093c80e8 in call_function () from /usr/local/lib/libpython3.10.so.0.0
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: cas...@cvs.openbsd.org 2023/01/28 14:04:58 Modified files: meta/tor-browser: Tag: OPENBSD_7_2 Makefile www/tor-browser: Tag: OPENBSD_7_2 Makefile.inc www/tor-browser/browser: Tag: OPENBSD_7_2 Makefile distinfo www/tor-browser/browser/patches: Tag: OPENBSD_7_2 patch-toolkit_components_tor-launcher_TorProcess_jsm www/tor-browser/noscript: Tag: OPENBSD_7_2 Makefile distinfo Log message: Tor Browser: update to 12.0.2 OK tb@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: t...@cvs.openbsd.org2023/01/28 14:02:08 Modified files: security/libgcrypt: Makefile Added files: security/libgcrypt/patches: patch-cipher_sha1-avx2-bmi2-amd64_S patch-cipher_sha512-avx2-bmi2-amd64_S patch-cipher_sm3-avx-bmi2-amd64_S patch-cipher_sm4-aesni-avx2-amd64_S Log message: libgcrypt: more amd64 x-only assembly fixes. Remove USE_NOEXECONLY for amd64 now.
NEW: databases/py-lmdb
Python bindings for LMDB. OK to import? py-lmdb.tgz Description: application/tar-gz
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: t...@cvs.openbsd.org2023/01/28 12:19:47 Modified files: security/rust-ring: Makefile Log message: rust-ring: add a reminder to bump consumers when this port is changed.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: k...@cvs.openbsd.org2023/01/28 12:12:08 Modified files: net/qbittorrent: Makefile.inc net/qbittorrent/qbittorrent: Makefile net/qbittorrent/qbittorrent-nox: Makefile Log message: Switch to Qt6 At least under Xfce when using a custom (lower) DPI value, qbittorrent is unusable as buttons get cropped to the point where their text is no longer visible; menus have the wrong size. Use Qt6 to get proper rendering and a usable UI. (Other Qt5 apps like KeePassXC are fine, regardless of DPI and/or desktop environment.) maintainer timeout OK rsadowski
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: t...@cvs.openbsd.org2023/01/28 10:41:11 Modified files: x11/qt5/qtwebengine: Makefile Added files: x11/qt5/qtwebengine/patches: patch-src_3rdparty_chromium_third_party_boringssl_linux-x86_64_crypto_cipher_extra_chacha20_poly1305_x86_64_S patch-src_3rdparty_chromium_third_party_boringssl_linux-x86_64_crypto_fipsmodule_p256-x86_64-asm_S patch-src_3rdparty_chromium_third_party_boringssl_linux-x86_64_crypto_fipsmodule_rsaz-avx2_S patch-src_3rdparty_chromium_third_party_boringssl_linux-x86_64_crypto_fipsmodule_vpaes-x86_64_S Log message: qtwebengine: runtime testing showed the need for more assembly fixes.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: k...@cvs.openbsd.org2023/01/28 09:34:11 Modified files: security/keepassxc: Makefile distinfo net/libtorrent-rasterbar: Makefile distinfo devel/cmake: Makefile cmake.port.mk distinfo Log message: Oops, the previous commit used an old/unrelated message file. Correct commit message: --- Handle ALL_TARGET and INSTALL_TARGET in cmake module devel/cmake ignores explicitly set targets, mostly because the default "all", "install" do not always work like they do with autoconf. In CMake, the default target is only known after configure time. To allow passing explicit targets without hand-rolling do-* in ports Makefiles, default to empty *_TARGET values in devel/cmake so it can later recognise explicitly set values and pass them along. For tests ctest(1) is used which works differently, so no TEST_TARGET bits. Remove hand-rolled targets in the only two effected ports: libtorrent-rasterbar and keepassxc. OK rsadowski ---
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: k...@cvs.openbsd.org2023/01/28 09:25:13 Modified files: security/keepassxc: Makefile net/libtorrent-rasterbar: Makefile devel/cmake: cmake.port.mk Log message: Remove security/hlfl WWW is dead, the old hlfl.org is now another project, DESCR doesn't mention pf, long missing out on updates, unmaintained. OK benoit
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: o...@cvs.openbsd.org2023/01/28 09:15:36 Modified files: net/tut: Makefile distinfo modules.inc Log message: update net/tut to 2.0.1
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: k...@cvs.openbsd.org2023/01/28 09:08:19 Modified files: sysutils : Makefile Log message: +gitlab-cli
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: k...@cvs.openbsd.org2023/01/28 09:07:15 Log message: Import sysutils/gitlab-cli 1.24.1 OK sthen --- GLab is an open source GitLab CLI tool bringing GitLab to your terminal next to where you are already working with git and your code without switching between windows and browser tabs. Work with issues, merge requests, watch running pipelines directly from your CLI among other features. glab(1) is available for repositories hosted on GitLab.com and self-managed GitLab instances. glab supports multiple authenticated GitLab instances and automatically detects the authenticated hostname from the remotes available in the working Git directory. Status: Vendor Tag: kn Release Tags: kn_20230128 N ports/sysutils/gitlab-cli/Makefile N ports/sysutils/gitlab-cli/distinfo N ports/sysutils/gitlab-cli/modules.inc N ports/sysutils/gitlab-cli/pkg/DESCR N ports/sysutils/gitlab-cli/pkg/PLIST No conflicts created by this import
Re: new sysutils/gitlab-cli
On Sat, 28 Jan 2023 15:04:15 +, Klemens Nanni wrote: > 28.01.2023 14:52, Laurent Cheylus пишет: >> Could you set the real build date with MODGO_LDFLAGS ? > > I deliberately set it to a fixed string such that package contents don't > change between builds. > > Do you encounter any issues? No issue, build and tests OK. As said by sthen@, remove build date via MODGO_FLAGS rather than setting a fake date. Thanks for this new port. Laurent
Re: new sysutils/gitlab-cli
28.01.2023 15:14, Stuart Henderson пишет: > You could remove '-X main.build' completely which gives this which I > think everyone will be happy with: > > $ glab version > glab version 1.24.1 Even better, thanks!
Re: new sysutils/gitlab-cli
On 2023/01/28 15:04, Klemens Nanni wrote:
> 28.01.2023 14:52, Laurent Cheylus пишет:
> > Build and tests OK on amd64.
>
> Thanks.
>
> > Could you set the real build date with MODGO_LDFLAGS ?
> >
> > _BUILD_DATE = $$(date +%Y-%m-%d)
> > MODGO_LDFLAGS = -X main.build=${_BUILD_DATE} \
> > -X main.version=${V}
> >
> > With this modification, my build is OK :
> >
> > $ /usr/obj/ports/gitlab-cli-1.24.1/fake-amd64/usr/local/bin/glab version
> > glab version 1.24.1 (2023-01-28)
>
> I deliberately set it to a fixed string such that package contents don't
> change between builds.
>
> Do you encounter any issues?
>
You could remove '-X main.build' completely which gives this which I
think everyone will be happy with:
$ glab version
glab version 1.24.1
OK
Re: new sysutils/gitlab-cli
28.01.2023 14:52, Laurent Cheylus пишет:
> Build and tests OK on amd64.
Thanks.
> Could you set the real build date with MODGO_LDFLAGS ?
>
> _BUILD_DATE = $$(date +%Y-%m-%d)
> MODGO_LDFLAGS = -X main.build=${_BUILD_DATE} \
> -X main.version=${V}
>
> With this modification, my build is OK :
>
> $ /usr/obj/ports/gitlab-cli-1.24.1/fake-amd64/usr/local/bin/glab version
> glab version 1.24.1 (2023-01-28)
I deliberately set it to a fixed string such that package contents don't
change between builds.
Do you encounter any issues?
Re: new sysutils/gitlab-cli
On Sat, 28 Jan 2023 00:15:23 +, Klemens Nanni wrote:
> Same as devel/github-cli but different in that upstream sources extraact
> differently, so post-extract moves it into place.
>
> LDFLAGS and gen-docs flags are also slightly different, otherwise `glab`
> works and does the equivalent to `gh`.
>
> Feedback? OK?
> ---
> Information for inst:gitlab-cli-1.24.1
Build and tests OK on amd64.
Could you set the real build date with MODGO_LDFLAGS ?
_BUILD_DATE = $$(date +%Y-%m-%d)
MODGO_LDFLAGS = -X main.build=${_BUILD_DATE} \
-X main.version=${V}
With this modification, my build is OK :
$ /usr/obj/ports/gitlab-cli-1.24.1/fake-amd64/usr/local/bin/glab version
glab version 1.24.1 (2023-01-28)
Laurent
Re: new sysutils/gitlab-cli
On Sat, Jan 28, 2023 at 12:51:31PM +, Stuart Henderson wrote: > Build fails: Mixup from starting out with the .zip from gitlab.com and later switching to proper proxy.golang.org which our bsd.port.mk is unable to recognise as different file (same name) unless you remove distfiles manually. > Could you zap the Markdown formatting in pkg/DESCR too please. Done, thanks. gitlab-cli.tgz Description: Binary data
Re: cargo.port.mk suppprt for GH_*
On Sat, Jan 28, 2023 at 01:24:10PM +, Stuart Henderson wrote:
> Now that we have GH_DISTNAME, we can add support for GH_* in
> cargo.port.mk without too much mess, allowing those ports which use
> the module and GH_* distfiles to get rid of the "DISTFILES +=
> ${DISTNAME}${EXTRACT_SUFX}' boilerplate from their Makefiles.
>
> I've got this in my current i386 bulk (now most of the way through
> with no problems), assuming that succeeds, OK?
Yes please. ok semarie@
> After this goes in, the extra DISTFILES+=... can be removed from
> ports using MODULES=devel/cargo with GH_*; no need to do it beforehand.
> I'm happy to go through and clean those up.
>
> (I would have used !empty(GH_DISTFILE) instead of checking GH_ACCOUNT
> and GH_PROJECT, but due to the order that files get included, it's not
> defined until _after_ the .if is evaluated).
>
> Index: cargo.port.mk
> ===
> RCS file: /cvs/ports/devel/cargo/cargo.port.mk,v
> retrieving revision 1.34
> diff -u -p -r1.34 cargo.port.mk
> --- cargo.port.mk 18 Jan 2023 13:11:39 - 1.34
> +++ cargo.port.mk 28 Jan 2023 09:51:55 -
> @@ -60,6 +60,10 @@ MASTER_SITES${MODCARGO_MASTER_SITESN} ?=
> # allow override default configuration, and keep all files
> MODCARGO_CRATES_KEEP ?=
>
> +.if !empty(GH_ACCOUNT) && !empty(GH_PROJECT)
> +DISTFILES += ${GH_DISTFILE}
> +.endif
> +
> # Generated list of DISTFILES.
> .for _cratename _cratever in ${MODCARGO_CRATES}
> DISTFILES +=
> ${_MODCARGO_DIST_SUBDIR}${_cratename}-${_cratever}.tar.gz{${_cratename}/${_cratever}/download}:${MODCARGO_MASTER_SITESN}
--
Sebastien Marie
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: rob...@cvs.openbsd.org 2023/01/28 06:41:18 Removed files: x11/qt5/qtwebengine/patches: patch-src_3rdparty_chromium_v8_src_snapshot_embedded_platform-embedded-file-writer-generic_cc.orig Log message: remove .orig file
Re: audio/ncspot segfaults on -current
Hi Theo, * Theo Buehler wrote: > On Sat, Jan 28, 2023 at 01:19:49PM +0100, Matthias Schmidt wrote: > > Hi, > > > > I noticed today that audio/ncspot segfaults on a recent -current. I > > have > > > > OpenBSD 7.2-current (GENERIC.MP) #994: Thu Jan 26 20:09:20 MST 2023 > > dera...@amd64.openbsd.org:/usr/src/sys/arch/amd64/compile/GENERIC.MP > > > > with the latest ncspot-0.12.0p0. Upon startup it crashes with a > > segmentation fault: > > > > $ egdb /usr/local/bin/ncspot ./ncspot.core > > Please apply this in /usr/ports/security/rust-ring, 'make install' and > then rebuild and reinstall ncspot. Thanks for the quick response. That indeed did the trick, ncspot is working again! Cheers Matthias
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: rob...@cvs.openbsd.org 2023/01/28 06:39:20 Modified files: www/chromium : Makefile www/chromium/patches: patch-build_config_compiler_BUILD_gn www/iridium: Makefile www/iridium/patches: patch-build_config_compiler_BUILD_gn www/ungoogled-chromium: Makefile www/ungoogled-chromium/patches: patch-build_config_compiler_BUILD_gn x11/qt5/qtwebengine: Makefile Added files: www/chromium/patches: patch-v8_src_execution_isolate_cc patch-v8_src_snapshot_embedded_platform-embedded-file-writer-base_cc patch-v8_src_snapshot_embedded_platform-embedded-file-writer-base_h patch-v8_src_snapshot_embedded_platform-embedded-file-writer-generic_cc www/iridium/patches: patch-v8_src_execution_isolate_cc patch-v8_src_snapshot_embedded_platform-embedded-file-writer-base_cc patch-v8_src_snapshot_embedded_platform-embedded-file-writer-base_h patch-v8_src_snapshot_embedded_platform-embedded-file-writer-generic_cc www/ungoogled-chromium/patches: patch-v8_src_execution_isolate_cc patch-v8_src_snapshot_embedded_platform-embedded-file-writer-base_cc patch-v8_src_snapshot_embedded_platform-embedded-file-writer-base_h patch-v8_src_snapshot_embedded_platform-embedded-file-writer-generic_cc x11/qt5/qtwebengine/patches: patch-src_3rdparty_chromium_v8_src_execution_isolate_cc patch-src_3rdparty_chromium_v8_src_snapshot_embedded_embedded-file-writer_h patch-src_3rdparty_chromium_v8_src_snapshot_embedded_platform-embedded-file-writer-base_cc patch-src_3rdparty_chromium_v8_src_snapshot_embedded_platform-embedded-file-writer-base_h patch-src_3rdparty_chromium_v8_src_snapshot_embedded_platform-embedded-file-writer-generic_cc patch-src_3rdparty_chromium_v8_src_snapshot_embedded_platform-embedded-file-writer-generic_cc.orig patch-src_3rdparty_chromium_v8_src_snapshot_embedded_platform-embedded-file-writer-generic_h Log message: put the embedded blob code to the .openbsd.mutable section and mprotect it RX to support x-only idea from deraadt@, diff from kettenis@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: t...@cvs.openbsd.org2023/01/28 06:34:17 Modified files: audio/ncspot : Makefile devel/maturin : Makefile mail/meli : Makefile mail/stalwart/cli: Makefile mail/stalwart/imap: Makefile mail/stalwart/jmap: Makefile net/routinator : Makefile security/rbw : Makefile security/sn0int: Makefile security/vaultwarden: Makefile www/libreddit : Makefile www/nextcloud_notify_push: Makefile www/py-adblock : Makefile www/zola : Makefile Log message: bump consumers of rust-ring (static linking)
cargo.port.mk suppprt for GH_*
Now that we have GH_DISTNAME, we can add support for GH_* in
cargo.port.mk without too much mess, allowing those ports which use
the module and GH_* distfiles to get rid of the "DISTFILES +=
${DISTNAME}${EXTRACT_SUFX}' boilerplate from their Makefiles.
I've got this in my current i386 bulk (now most of the way through
with no problems), assuming that succeeds, OK?
After this goes in, the extra DISTFILES+=... can be removed from
ports using MODULES=devel/cargo with GH_*; no need to do it beforehand.
I'm happy to go through and clean those up.
(I would have used !empty(GH_DISTFILE) instead of checking GH_ACCOUNT
and GH_PROJECT, but due to the order that files get included, it's not
defined until _after_ the .if is evaluated).
Index: cargo.port.mk
===
RCS file: /cvs/ports/devel/cargo/cargo.port.mk,v
retrieving revision 1.34
diff -u -p -r1.34 cargo.port.mk
--- cargo.port.mk 18 Jan 2023 13:11:39 - 1.34
+++ cargo.port.mk 28 Jan 2023 09:51:55 -
@@ -60,6 +60,10 @@ MASTER_SITES${MODCARGO_MASTER_SITESN} ?=
# allow override default configuration, and keep all files
MODCARGO_CRATES_KEEP ?=
+.if !empty(GH_ACCOUNT) && !empty(GH_PROJECT)
+DISTFILES += ${GH_DISTFILE}
+.endif
+
# Generated list of DISTFILES.
.for _cratename _cratever in ${MODCARGO_CRATES}
DISTFILES +=
${_MODCARGO_DIST_SUBDIR}${_cratename}-${_cratever}.tar.gz{${_cratename}/${_cratever}/download}:${MODCARGO_MASTER_SITESN}
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: t...@cvs.openbsd.org2023/01/28 06:20:40 Modified files: security/rust-ring: Makefile Added files: security/rust-ring/patches: patch-pregenerated_sha256-x86_64-elf_S Log message: rust-ring: add missing .rodata annotation for SHA-256. SIGSEGV in ncspot reported by Matthias Schmidt.
Re: lang/sbcl will not work with x-only
On 2023/01/28 12:33:29 +, Stuart Henderson wrote: > (Also while on the subject of sbcl, we still have "map stack for pid XX > failed" in dmesg when it runs; I don't know how bad that is but sbcl is > the only program which gets run during an i386 ports build which does > that). it's not too bad but neither good. sbcl calls sigaltstack to have signals delivered on the stack of the "control thread". Now, that thread stack is mmap'ed as RWX (as all allocations done by sbcl via os_validates) and thus fails with sigaltstack. at runtime i'm not too sure what happens. I've tested some ncurses programs and guess SIGWINCH (at least) is handled correctly. A dummy program with an infinite loop that triggers a stack overflow dies with SIGILL now, where before was still dying but due to sbcl erroring. I had a wip diff to use a fresh RX page for the sigaltstack handling that seemed to work, but got stuck at trying to figuring out how the sbcl' backtrace interacts with the control thread and got nowhere.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: o...@cvs.openbsd.org2023/01/28 05:52:21 Modified files: net/lagrange : Makefile distinfo Log message: update net/lagrange to 1.15.1; from MAINTAINER Florian Viehweger, thanks! Changelog: - Fixed opening multiple Preferences windows at the same time. - Fixed the Preferences window missing a window icon. - Footer information moved out of the margin area and under the page content. - Updated UI translations.
Re: new sysutils/gitlab-cli
On 2023/01/28 00:15, Klemens Nanni wrote: > Same as devel/github-cli but different in that upstream sources extraact > differently, so post-extract moves it into place. > > LDFLAGS and gen-docs flags are also slightly different, otherwise `glab` > works and does the equivalent to `gh`. > > Feedback? OK? Build fails: $ make ===> gitlab-cli-1.24.1 depends on: go-=1.19.4p1 -> go-1.19.4p1 ===> gitlab-cli-1.24.1 depends on: unzip-* -> unzip-6.0p16 ===> Verifying specs: c pthread ===> found c.96.5 pthread.26.4 ===> Checking files for gitlab-cli-1.24.1 `/y/Download/ftp/pub/OpenBSD/distfiles/cli-v1.24.1.zip' is up to date. `/y/Download/ftp/pub/OpenBSD/distfiles/go_modules/cloud.google.com/go/@v/v0.81.0.zip' is up to date. <...snip...> >> (SHA256) go_modules/rsc.io/quote/v3/@v/v3.1.0.mod: OK >> (SHA256) go_modules/rsc.io/sampler/@v/v1.3.0.mod: OK ===> Extracting for gitlab-cli-1.24.1 # move WRKDIST where MODGO_SETUP_WORKSPACE expects it mkdir -p /pobj/gitlab-cli-1.24.1/gitlab.com/gitlab-org mv /pobj/gitlab-cli-1.24.1/cli-v1.24.1-* /pobj/gitlab-cli-1.24.1/gitlab.com/gitlab-org/cli@v1.24.1 mv: /pobj/gitlab-cli-1.24.1/cli-v1.24.1-*: No such file or directory *** Error 1 in . (Makefile:25 'post-extract') *** Error 2 in . (/usr/ports/infrastructure/mk/bsd.port.mk:2701 '/pobj/gitlab-cli-1.24.1/.extract_done': @cd /usr/ports/mystuff/sysutils/git...) *** Error 2 in /usr/ports/mystuff/sysutils/gitlab-cli (/usr/ports/infrastructure/mk/bsd.port.mk:2600 'all': @lock=gitlab-cli-1.24.1; export...) $ ls -l /pobj/gitlab-cli-1.24.1 total 16 drwxr-xr-x 2 _pbuild _pbuild 512 Jan 28 12:48 bin drwxr-xr-x 3 _pbuild _pbuild 512 Jan 28 12:48 gitlab.com > --- > Information for inst:gitlab-cli-1.24.1 > > Comment: > GitLab CLI tool > > Description: > GLab is an open source GitLab CLI tool bringing GitLab to your terminal > next to where you are already working with `git` and your code without > switching between windows and browser tabs. Work with issues, merge requests, > **watch running pipelines directly from your CLI** among other features. > > `glab` is available for repositories hosted on GitLab.com and self-managed > GitLab instances. `glab` supports multiple authenticated GitLab instances > and automatically detects the authenticated hostname from the remotes > available in the working Git directory. Could you zap the Markdown formatting in pkg/DESCR too please.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: rob...@cvs.openbsd.org 2023/01/28 05:46:46 Modified files: lang/node : Makefile Added files: lang/node/patches: patch-deps_v8_src_execution_isolate_cc patch-deps_v8_src_snapshot_embedded_platform-embedded-file-writer-base_cc patch-deps_v8_src_snapshot_embedded_platform-embedded-file-writer-base_h patch-deps_v8_src_snapshot_embedded_platform-embedded-file-writer-generic_cc patch-deps_v8_src_snapshot_embedded_platform-embedded-file-writer-generic_h Log message: put the embedded blob code to the .openbsd.mutable section and mprotect it RX to support x-only idea from deraadt@, diff from kettenis@
Re: [update] net/py-dpkt 1.9.1 -> 1.9.8
On 2023/01/28 07:34, dsp wrote:
> Hey list! :)
> sorry i let this port run a bit stale.
> Updating it to the (not so) new MODPY
> style.
>
>
> Index: Makefile
> ===
> RCS file: /cvs/ports/net/py-dpkt/Makefile,v
> retrieving revision 1.20
> diff -u -p -r1.20 Makefile
> --- Makefile 13 Nov 2022 15:28:50 - 1.20
> +++ Makefile 28 Jan 2023 06:20:13 -
> @@ -1,9 +1,9 @@
> COMMENT =python network packet manipulation library
>
> -MODPY_EGG_VERSION = 1.9.1
> +MODPY_EGG_VERSION = 1.9.8
> DISTNAME = dpkt-${MODPY_EGG_VERSION}
> -PKGNAME =py-${DISTNAME}
> -REVISION = 3
> +PKGNAME =${MODPY_PY_PREFIX}dpkt-${MODPY_EGG_VERSION}
py- is still OK here (and much easier to read)
> +REVISION = 0
drop REVISION rather than setting to 0
> CATEGORIES = net
>
> @@ -17,6 +17,9 @@ MAINTAINER =Thanasoulas Spiros
> # BSD
> PERMIT_PACKAGE = Yes
> +
> +MODPY_PI = Yes
here you've set MODPY_PI (to fetch from pypi) but still have GH_* (to
fetch from github); MODPY_PI takes priority so the GH_* should be
removed, also you must have fetched from github not pypi when
generating the distinfo file.
So I'll commit this diff instead.
Index: Makefile
===
RCS file: /cvs/ports/net/py-dpkt/Makefile,v
retrieving revision 1.20
diff -u -p -r1.20 Makefile
--- Makefile13 Nov 2022 15:28:50 - 1.20
+++ Makefile28 Jan 2023 12:40:41 -
@@ -1,22 +1,20 @@
-COMMENT = python network packet manipulation library
+COMMENT = Python network packet manipulation library
-MODPY_EGG_VERSION =1.9.1
+MODPY_EGG_VERSION =1.9.8
DISTNAME = dpkt-${MODPY_EGG_VERSION}
-PKGNAME = py-${DISTNAME}
-REVISION = 3
+PKGNAME = py-dpkt-${MODPY_EGG_VERSION}
CATEGORIES = net
HOMEPAGE = https://github.com/kbandla/dpkt
-GH_ACCOUNT = kbandla
-GH_PROJECT = dpkt
-GH_TAGNAME = v${MODPY_EGG_VERSION}
-
MAINTAINER = Thanasoulas Spiros
# BSD
PERMIT_PACKAGE = Yes
+
+MODPY_PI = Yes
+MODPY_PYBUILD =setuptools
MODULES = lang/python
Index: distinfo
===
RCS file: /cvs/ports/net/py-dpkt/distinfo,v
retrieving revision 1.5
diff -u -p -r1.5 distinfo
--- distinfo16 Jun 2018 06:12:50 - 1.5
+++ distinfo28 Jan 2023 12:40:41 -
@@ -1,2 +1,2 @@
-SHA256 (dpkt-1.9.1.tar.gz) = 4H0/F7H/mxk1LWqX2rLOSo6oLxekxXI7IZvZAq2XvG4=
-SIZE (dpkt-1.9.1.tar.gz) = 133711
+SHA256 (dpkt-1.9.8.tar.gz) = Q/hobkVdpQUoNf0e2iaJ1R3jZwqsl5mxsAz9IDkn7kU=
+SIZE (dpkt-1.9.8.tar.gz) = 180073
Index: pkg/PLIST
===
RCS file: /cvs/ports/net/py-dpkt/pkg/PLIST,v
retrieving revision 1.6
diff -u -p -r1.6 PLIST
--- pkg/PLIST 8 Sep 2022 10:10:38 - 1.6
+++ pkg/PLIST 28 Jan 2023 12:40:41 -
@@ -1,79 +1,163 @@
@pkgpath net/py-dpkt
lib/python${MODPY_VERSION}/site-packages/dpkt/
-lib/python${MODPY_VERSION}/site-packages/dpkt-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info
+lib/python${MODPY_VERSION}/site-packages/dpkt-${MODPY_EGG_VERSION}.dist-info/
+lib/python${MODPY_VERSION}/site-packages/dpkt-${MODPY_EGG_VERSION}.dist-info/AUTHORS
+lib/python${MODPY_VERSION}/site-packages/dpkt-${MODPY_EGG_VERSION}.dist-info/LICENSE
+lib/python${MODPY_VERSION}/site-packages/dpkt-${MODPY_EGG_VERSION}.dist-info/METADATA
+lib/python${MODPY_VERSION}/site-packages/dpkt-${MODPY_EGG_VERSION}.dist-info/RECORD
+lib/python${MODPY_VERSION}/site-packages/dpkt-${MODPY_EGG_VERSION}.dist-info/WHEEL
+lib/python${MODPY_VERSION}/site-packages/dpkt-${MODPY_EGG_VERSION}.dist-info/top_level.txt
lib/python${MODPY_VERSION}/site-packages/dpkt/__init__.py
-${MODPY_COMMENT}lib/python${MODPY_VERSION}/site-packages/dpkt/${MODPY_PYCACHE}
+${MODPY_COMMENT}lib/python${MODPY_VERSION}/site-packages/dpkt/${MODPY_PYCACHE}/
+lib/python${MODPY_VERSION}/site-packages/dpkt/${MODPY_PYCACHE}__init__.${MODPY_PYC_MAGIC_TAG}${MODPY_PYOEXTENSION}
lib/python${MODPY_VERSION}/site-packages/dpkt/${MODPY_PYCACHE}__init__.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/dpkt/${MODPY_PYCACHE}ah.${MODPY_PYC_MAGIC_TAG}${MODPY_PYOEXTENSION}
lib/python${MODPY_VERSION}/site-packages/dpkt/${MODPY_PYCACHE}ah.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/dpkt/${MODPY_PYCACHE}aim.${MODPY_PYC_MAGIC_TAG}${MODPY_PYOEXTENSION}
lib/python${MODPY_VERSION}/site-packages/dpkt/${MODPY_PYCACHE}aim.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/dpkt/${MODPY_PYCACHE}aoe.${MODPY_PYC_MAGIC_TAG}${MODPY_PYOEXTENSION}
lib/python${MODPY_VERSION}/site-packages/dpkt/${MODPY_PYCACHE}aoe.${MODPY_PYC_MAGIC_TAG}pyc
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2023/01/28 05:42:07 Modified files: net/py-dpkt: Makefile distinfo net/py-dpkt/pkg: PLIST Log message: update to py3-dpkt-1.9.8, modified from a diff from dsp (maintainer)
Re: lang/sbcl will not work with x-only
On 2023/01/27 16:02, Josh Elsasser wrote: > On Fri, Jan 27, 2023 at 05:02:33PM +0100, Christian Weisgerber wrote: > > lang/sbcl will not work with x-only. > > > > I took another look at the warnings about data directives in .text. > > In runtime/x86-64-assem.S we have snippets like these: > > > > GNAME(fun_end_breakpoint_trap): > > TRAP > > .byte trap_FunEndBreakpoint > > hlt # We should never return here. > > > > GNAME(do_pending_interrupt): > > TRAP > > .byte trap_PendingInterrupt > > ret > > > > GNAME(memory_fault_emulation_trap): > > TRAP > > .byte trap_MemoryFaultEmulation > > > > Where TRAP is defined to int3, and trap_* are consecutively numbered > > constants. The other half of this is sigtrap_handler() in > > runtime/x86-64-arch.c, which reads the byte after the int3 instruction > > > > trap = *(unsigned char *)OS_CONTEXT_PC(context); > > > > and then dispatches accordingly. > > > > ... And this brief analysis marks the end of my own involvement > > with this port. > > Unfortunately I don't have a new enough CPU to test xonly, so someone > should probably mark it USE_NOEXECONLY I don't know how sbcl works, but if it calls the OS linker at runtime when compiling software, it may need to use --no-execute-only on that call too. (Also while on the subject of sbcl, we still have "map stack for pid XX failed" in dmesg when it runs; I don't know how bad that is but sbcl is the only program which gets run during an i386 ports build which does that).
Re: audio/ncspot segfaults on -current
On Sat, Jan 28, 2023 at 01:19:49PM +0100, Matthias Schmidt wrote:
> Hi,
>
> I noticed today that audio/ncspot segfaults on a recent -current. I
> have
>
> OpenBSD 7.2-current (GENERIC.MP) #994: Thu Jan 26 20:09:20 MST 2023
> dera...@amd64.openbsd.org:/usr/src/sys/arch/amd64/compile/GENERIC.MP
>
> with the latest ncspot-0.12.0p0. Upon startup it crashes with a
> segmentation fault:
>
> $ egdb /usr/local/bin/ncspot ./ncspot.core
Please apply this in /usr/ports/security/rust-ring, 'make install' and
then rebuild and reinstall ncspot.
Index: Makefile
===
RCS file: /cvs/ports/security/rust-ring/Makefile,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 Makefile
--- Makefile18 Jan 2023 12:56:09 - 1.1.1.1
+++ Makefile28 Jan 2023 12:24:38 -
@@ -2,6 +2,7 @@ COMMENT = ring crate source patched for
VERSION = 0.16.20
DISTNAME = ring-${VERSION}
+REVISION = 0
MASTER_SITES = https://crates.io/api/v1/crates/
Index: patches/patch-pregenerated_sha256-x86_64-elf_S
===
RCS file: patches/patch-pregenerated_sha256-x86_64-elf_S
diff -N patches/patch-pregenerated_sha256-x86_64-elf_S
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-pregenerated_sha256-x86_64-elf_S 28 Jan 2023 12:24:26
-
@@ -0,0 +1,19 @@
+Index: pregenerated/sha256-x86_64-elf.S
+--- pregenerated/sha256-x86_64-elf.S.orig
pregenerated/sha256-x86_64-elf.S
+@@ -1735,6 +1735,7 @@ GFp_sha256_block_data_order:
+ .byte 0xf3,0xc3
+ .cfi_endproc
+ .size GFp_sha256_block_data_order,.-GFp_sha256_block_data_order
++.rodata
+ .align64
+ .type K256,@object
+ K256:
+@@ -1778,6 +1779,7 @@ K256:
+ .long 0x,0x,0x03020100,0x0b0a0908
+ .long 0x,0x,0x03020100,0x0b0a0908
+ .byte
83,72,65,50,53,54,32,98,108,111,99,107,32,116,114,97,110,115,102,111,114,109,32,102,111,114,32,120,56,54,95,54,52,44,32,67,82,89,80,84,79,71,65,77,83,32,98,121,32,60,97,112,112,114,111,64,111,112,101,110,115,115,108,46,111,114,103,62,0
++.previous
+ .type GFp_sha256_block_data_order_ssse3,@function
+ .align64
+ GFp_sha256_block_data_order_ssse3:
audio/ncspot segfaults on -current
Hi,
I noticed today that audio/ncspot segfaults on a recent -current. I
have
OpenBSD 7.2-current (GENERIC.MP) #994: Thu Jan 26 20:09:20 MST 2023
dera...@amd64.openbsd.org:/usr/src/sys/arch/amd64/compile/GENERIC.MP
with the latest ncspot-0.12.0p0. Upon startup it crashes with a
segmentation fault:
$ egdb /usr/local/bin/ncspot ./ncspot.core
Reading symbols from /usr/local/bin/ncspot...
[New process 489855]
[New process 615451]
[New process 160551]
[New process 329627]
[New process 441709]
[New process 542294]
[New process 176151]
[New process 245967]
Core was generated by `ncspot'.
Program terminated with signal SIGSEGV, Segmentation fault.
#0 GFp_sha256_block_data_order_avx () at
/usr/obj/ports/ncspot-0.12.0/ncspot-0.12.0/modcargo-crates/ring-0.16.20/pregenerated/sha256-x86_64-elf.S:2933
2933
/usr/obj/ports/ncspot-0.12.0/ncspot-0.12.0/modcargo-crates/ring-0.16.20/pregenerated/sha256-x86_64-elf.S:
No such file or directory.
[Current thread is 1 (process 489855)]
(gdb) bt
#0 GFp_sha256_block_data_order_avx () at
/usr/obj/ports/ncspot-0.12.0/ncspot-0.12.0/modcargo-crates/ring-0.16.20/pregenerated/sha256-x86_64-elf.S:2933
#1 0x0e40b5c1e8c8 in ring::digest::Context::update ()
#2 0x0e40b5c1d3bf in ring::rsa::verification::verify_rsa_ ()
#3 0x0e40b5c261c9 in ring::rsa::verificationverify ()
#4 0x0e40b5c1a586 in webpki::signed_data::verify_signature ()
#5 0x0e40b5c1826a in webpki::verify_cert::build_chain ()
#6 0x0e40b5c184c2 in webpki::verify_cert::build_chain ()
#7 0x0e40b5c1a200 in
webpki::end_entity::EndEntityCert::verify_is_valid_tls_server_cert ()
#8 0x0e40b5c060ce in ::verify_server_cert ()
#9 0x0e40b5c137c4 in >::handle
()
#10 0x0e40b5ba7f75 in rustls::conn::CommonState::process_main_protocol ()
#11 0x0e40b5ba889e in
rustls::conn::ConnectionCommon::process_new_packets ()
#12 0x0e40b5ba8238 in rustls::conn::ConnectionCommon::complete_io ()
#13 0x0e40b5bbaa53 in ureq::rtls::>::connect ()
#14 0x0e40b5bb89a2 in ureq::stream::connect_https ()
#15 0x0e40b5bd0e87 in ureq::unit::connect_socket ()
#16 0x0e40b5bcfa5a in ureq::unit::connect_inner ()
#17 0x0e40b5bceb6c in ureq::unit::connect ()
#18 0x0e40b5bb6e16 in ureq::request::Request::do_call::{{closure}} ()
#19 0x0e40b5bb6a8c in ureq::request::Request::do_call ()
#20 0x0e40b5bb61d3 in ureq::request::Request::call ()
#21 0x0e40b576f787 in rspotify_http::ureq::UreqClient::request ()
#22 0x0e40b58c0256 in rspotify::clients::base::BaseClient::endpoint_get ()
#23 0x0e40b58c4d6a in rspotify::clients::oauth::OAuthClient::current_user ()
#24 0x0e40b58133d5 in ncspot::spotify_api::WebApi::current_user ()
#25 0x0e40b5aa2a35 in ncspot::main ()
#26 0x0e40b59d6e4c in
std::sys_common::backtrace::__rust_begin_short_backtrace ()
#27 0x0e40b59e4206 in std::rt::lang_start::{{closure}} ()
#28 0x0e40b6003629 in std::rt::lang_start_internal ()
#29 0x0e40b5ab187d in main ()
(gdb)
My system has a recent CPU so is fully xonly compatible:
userland kernel
ld.so 0x1ac01d165b0 unreadable unreadable
mmap xz0x1abb51fb000 unreadable unreadable
mmap x 0x1ac3d953000 unreadable unreadable
mmap nrx 0x1abb2d19000 unreadable unreadable
mmap nwx 0x1abb9a1e000 unreadable unreadable
mmap xnwx 0x1ac5c715000 unreadable unreadable
main 0x1a99f76e490 unreadable unreadable
libc 0x1ac2ed949d0 unreadable unreadable
Might it be related to that?
Cheers
Matthias
OpenBSD 7.2-current (GENERIC.MP) #994: Thu Jan 26 20:09:20 MST 2023
dera...@amd64.openbsd.org:/usr/src/sys/arch/amd64/compile/GENERIC.MP
real mem = 34064420864 (32486MB)
avail mem = 33012658176 (31483MB)
random: good seed from bootblocks
mpath0 at root
scsibus0 at mpath0: 256 targets
mainbus0 at root
bios0 at mainbus0: SMBIOS rev. 3.3 @ 0x43ca1000 (119 entries)
bios0: vendor American Megatrends International, LLC. version "N.1.15A09" date
03/24/2022
bios0: TUXEDO TUXEDO InfinityBook Pro 14 Gen6
efi0 at bios0: UEFI 2.7
efi0: American Megatrends rev 0x50013
acpi0 at bios0: ACPI 6.2
acpi0: sleep states S0 S3 S4 S5
acpi0: tables DSDT FACP MCFG SSDT FIDT SSDT SSDT SSDT HPET APIC SSDT SSDT NHLT
UEFI LPIT SSDT SSDT DBGP DBG2 SSDT DMAR SSDT SSDT BGRT PTDT WSMT FPDT
acpi0: wakeup devices PEGP(S4) PEGP(S4) PEGP(S4) PEG0(S4) PEGP(S4) RP01(S4)
PXSX(S4) RP02(S4) PXSX(S4) RP03(S4) PXSX(S4) RP04(S4) PXSX(S4) RP05(S4)
PXSX(S4) RP06(S4) [...]
acpitimer0 at acpi0: 3579545 Hz, 24 bits
acpimcfg0 at acpi0
acpimcfg0: addr 0xc000, bus 0-255
acpihpet0
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2023/01/28 04:03:23 Modified files: x11/spice-gtk : Makefile distinfo x11/spice-gtk/pkg: PLIST Added files: x11/spice-gtk/patches: patch-subprojects_keycodemapdb_tools_keymap-gen Log message: Update to spice-gtk-0.42.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2023/01/28 03:55:35 Modified files: geo/geoclue2 : Makefile distinfo geo/geoclue2/pkg: PLIST Log message: Update to geoclue2-2.7.0.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2023/01/28 03:51:19 Modified files: sysutils/govmomi: Makefile distinfo Log message: Update to govc-0.30.1.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2023/01/28 03:34:03 Modified files: sysutils/awscli: Makefile distinfo sysutils/awscli/pkg: PLIST Log message: Update to awscli-1.27.59.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2023/01/28 03:33:43 Modified files: net/py-boto3 : Makefile distinfo Log message: Update to py3-boto3-1.26.59.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2023/01/28 03:33:32 Modified files: net/py-botocore: Makefile distinfo Log message: Update to py3-botocore-1.29.59.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2023/01/28 03:09:30 Modified files: devel/py-filetype: Makefile distinfo devel/py-filetype/pkg: PLIST Log message: update to py3-filetype-1.2.0
Re: firefox xonly fix
> mail/mozilla-thunderbird > www/firefox-esr > www/mozilla-firefox > www/seamonkey > www/tor-browser/browser I have added the patch to all of these. I have tested mozilla-firefox and firefox-esr thoroughly, for the others I only verified that the patch applies cleanly to the source. As mentioned in the commit, I was unable to trigger the issue with the tor browser, maybe it has audio recording disabled. I also have no idea if this code is reachable on thunderbird.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2023/01/28 02:48:30 Modified files: devel/py-simplejson: Makefile distinfo Log message: update to py-simplejson-3.18.1
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2023/01/28 02:48:27 Modified files: www/composer : Makefile distinfo Log message: update to composer-2.5.1
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2023/01/28 02:48:24 Modified files: multimedia/atomicparsley: Makefile distinfo Log message: update to atomicparsley-20221229
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2023/01/28 02:48:21 Modified files: print/ocrmypdf : Makefile distinfo Log message: update to ocrmypdf-14.0.2
Re: gzdoom update to 4.10.0
Thomas Frohwein writes:
> Hi,
>
> This diff updates gzdoom to latest version 4.10.0. I've been trying out
> a few gzdoom games and ended up updating this port in the process, even
> though not strictly necessary to run the one I mainly tested - the demo
> of Beyond Sunset [1]. Hands of Necromancy also works.
>
> The diff drops 2 patches that are no longer needed. Adds a new one to
> bypass get_current_dir_name() in i_system.cpp.
>
> CC'ing maintainer
>
> [1] https://metacorp.itch.io/beyond-sunset
>
> Index: Makefile
> ===
> RCS file: /cvs/ports/games/gzdoom/Makefile,v
> retrieving revision 1.17
> diff -u -p -r1.17 Makefile
> --- Makefile 18 Jul 2022 08:27:45 - 1.17
> +++ Makefile 24 Jan 2023 05:02:28 -
> @@ -6,7 +6,7 @@ ONLY_FOR_ARCHS = i386 amd64
>
> COMMENT =OpenGL engine for idTech 1 games like
> doom,hexen,heretic...
>
> -V = 4.8.2
> +V = 4.10.0
> GH_ACCOUNT = coelckers
> GH_PROJECT = gzdoom
> GH_TAGNAME = g${V}
> Index: distinfo
> ===
> RCS file: /cvs/ports/games/gzdoom/distinfo,v
> retrieving revision 1.7
> diff -u -p -r1.7 distinfo
> --- distinfo 18 Jul 2022 08:27:45 - 1.7
> +++ distinfo 24 Jan 2023 05:02:28 -
> @@ -1,2 +1,2 @@
> -SHA256 (gzdoom-4.8.2.tar.gz) = Tt1g+AlMncPft0AL28DrsKXldLCmdH7uTI2fH2r3cSc=
> -SIZE (gzdoom-4.8.2.tar.gz) = 24247668
> +SHA256 (gzdoom-4.10.0.tar.gz) = hwJSLAUEjf1KdlxqyConDYvQlC6BPVvG9LaXlaXSOiA=
> +SIZE (gzdoom-4.10.0.tar.gz) = 24319337
> Index: patches/patch-libraries_discordrpc_CMakeLists_txt
> ===
> RCS file: patches/patch-libraries_discordrpc_CMakeLists_txt
> diff -N patches/patch-libraries_discordrpc_CMakeLists_txt
> --- patches/patch-libraries_discordrpc_CMakeLists_txt 31 Jul 2022 20:50:40
> - 1.1
> +++ /dev/null 1 Jan 1970 00:00:00 -
> @@ -1,15 +0,0 @@
> -don't pick up clang-format is present
> -
> -Index: libraries/discordrpc/CMakeLists.txt
> libraries/discordrpc/CMakeLists.txt.orig
> -+++ libraries/discordrpc/CMakeLists.txt
> -@@ -10,7 +10,8 @@ file(GLOB_RECURSE ALL_SOURCE_FILES
> - )
> -
> - # Set CLANG_FORMAT_SUFFIX if you are using custom clang-format, e.g.
> clang-format-5.0
> --find_program(CLANG_FORMAT_CMD clang-format${CLANG_FORMAT_SUFFIX})
> -+#find_program(CLANG_FORMAT_CMD clang-format${CLANG_FORMAT_SUFFIX})
> -+set(CLANG_FORMAT_CMD, FALSE)
> -
> - if (CLANG_FORMAT_CMD)
> - add_custom_target(
> Index: patches/patch-src_common_platform_posix_sdl_i_system_cpp
> ===
> RCS file: patches/patch-src_common_platform_posix_sdl_i_system_cpp
> diff -N patches/patch-src_common_platform_posix_sdl_i_system_cpp
> --- /dev/null 1 Jan 1970 00:00:00 -
> +++ patches/patch-src_common_platform_posix_sdl_i_system_cpp 24 Jan 2023
> 05:02:28 -
> @@ -0,0 +1,52 @@
> +Index: src/common/platform/posix/sdl/i_system.cpp
> +--- src/common/platform/posix/sdl/i_system.cpp.orig
> src/common/platform/posix/sdl/i_system.cpp
> +@@ -54,6 +54,10 @@
> + #include "printf.h"
> + #endif
> +
> ++#ifdef __OpenBSD__
> ++#include "printf.h"
> ++#endif
> ++
> + #include
> +
> + #include "version.h"
> +@@ -408,14 +412,7 @@ FString I_GetFromClipboard (bool use_primary_selection
> +
> + FString I_GetCWD()
> + {
> +-char* curdir = get_current_dir_name();
> +-if (!curdir)
> +-{
> +-return "";
> +-}
> +-FString ret(curdir);
> +-free(curdir);
> +-return ret;
> ++return "";
> + }
> +
> + bool I_ChDir(const char* path)
> +@@ -447,18 +444,18 @@ unsigned int I_MakeRNGSeed()
> +
> + void I_OpenShellFolder(const char* infolder)
> + {
> +-char* curdir = get_current_dir_name();
> ++//char* curdir = get_current_dir_name();
> +
> + if (!chdir(infolder))
> + {
> + Printf("Opening folder: %s\n", infolder);
> + std::system("xdg-open .");
> +-chdir(curdir);
> ++//chdir(curdir);
> + }
> + else
> + {
> + Printf("Unable to open directory '%s\n", infolder);
> + }
> +-free(curdir);
> ++//free(curdir);
> + }
> +
> Index: patches/patch-src_d_main_cpp
> ===
> RCS file: patches/patch-src_d_main_cpp
> diff -N patches/patch-src_d_main_cpp
> --- patches/patch-src_d_main_cpp 18 Jul 2022 08:27:46 - 1.1
> +++ /dev/null 1 Jan 1970 00:00:00 -
> @@ -1,24 +0,0 @@
> -fix gzdoom.pk3 not found error, backport of
> -https://github.com/coelckers/gzdoom/pull/1665
> -
> -Index: src/d_main.cpp
> src/d_main.cpp.orig
> -+++ src/d_main.cpp
> -@@ -3534,6 +3534,8 @@ static int D_DoomMain_Internal (void)
> -
> - std::set_new_handler(NewFailure);
> - const char *batchout =
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: t...@cvs.openbsd.org2023/01/28 02:45:10 Modified files: mail/mozilla-thunderbird: Makefile Added files: mail/mozilla-thunderbird/patches: patch-media_ffvpx_libavcodec_x86_fft_asm Log message: thunderbird: apply kettenis's fft.asm fix also here.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: t...@cvs.openbsd.org2023/01/28 02:41:49 Modified files: www/seamonkey : Makefile Added files: www/seamonkey/patches: patch-media_ffvpx_libavcodec_x86_fft_asm Log message: seamonkey: apply kettenis's fft.asm fix. Also set USE_NOEXECONLY since the embedded skia does some embedded assembly stuff that is not easy neuter and makes seamonkey completely broken on PKU machines.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: t...@cvs.openbsd.org2023/01/28 02:36:35 Modified files: www/tor-browser/browser: Makefile Added files: www/tor-browser/browser/patches: patch-media_ffvpx_libavcodec_x86_fft_asm Log message: tor-browser: apply kettenis's fft.asm fix from ffmpeg also here. While the file is clearly compiled in, I was unable to trigger the crash with the reproducers known to me, so it may be neutered by whatever it is that the developers do to firefox.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: t...@cvs.openbsd.org2023/01/28 02:29:57 Modified files: www/firefox-esr: Makefile Added files: www/firefox-esr/patches: patch-media_ffvpx_libavcodec_x86_fft_asm Log message: firefox-esr: backport kettenis's fft.asm fix from ffmpeg. Fixes an x-only crash reported by ratchov.
Re: update plan9/plan9port
On 2023/01/26 18:12:49 +0100, Omar Polo wrote:
> I'm not using exactly this, I have my own build in ~/plan9 (was
> /usr/local/plan9 but had to move it to test the update) with a few
> patches, but I've been using it continuously at least for the last two
> years (rebasing on top of the latest commit every few months) without
> any issue. I'm mostly using Acme, grep (via `g') and the manpages.
>
> I've dropped the substitution for bin/wintext (was rewritten in rc)
> and while here i've simplified do-install: there are no more any
> .cvsignore, .hg, .hgignore or .hgtags file, and adjusted the find
> invocation to delete the other files in one go.
>
> The patch for libthread/daemonize.c was merged upstream and so can be
> dropped.
>
> ok?
Kyle Milz (offlist) noted that patch-src_mkhdr is not needed anymore.
it builds, links, package and works fine without. Thanks!
updated diff
Index: Makefile
===
RCS file: /home/cvs/ports/plan9/plan9port/Makefile,v
retrieving revision 1.40
diff -u -p -r1.40 Makefile
--- Makefile23 Sep 2022 02:28:27 - 1.40
+++ Makefile26 Jan 2023 16:37:16 -
@@ -2,11 +2,11 @@ ONLY_FOR_ARCHS = amd64 i386 powerpc powe
COMMENT = Plan 9 from user space
-DISTNAME = plan9port-20210323
+DISTNAME = plan9port-20230123
GH_ACCOUNT = 9fans
GH_PROJECT = plan9port
-GH_COMMIT =88a87fadae6629932d9c160f53ad5d79775f8f94
+GH_COMMIT =e2a141ae0952c2d28a9f500bb9da6dc72e57be3c
CATEGORIES = plan9
@@ -32,18 +32,17 @@ UPDATE_PLIST_ARGS += -i CC
pre-configure:
${SUBST_CMD} ${WRKSRC}/INSTALL ${WRKSRC}/bin/{9c,9l}
cd ${WRKSRC} && sed -i 's,#!/bin/bash,#!/bin/sh,' \
- bin/{macedit,wintext}
+ bin/macedit
do-build:
cd ${WRKSRC} && env CC9="${CC}" CCFLAGS="${CFLAGS}" ./INSTALL -b
do-install:
-.for l in .cvsignore .hg .hgignore .hgtags .gitignore .gitkeep *.orig
*.beforesubst .travis.yml
- find ${WRKSRC} -name "$l" -delete
-.endfor
+ rm -f ${WRKSRC}/.travis.yml
cd ${WRKSRC} && sed -i 's#/usr/local#${TRUEPREFIX}#' \
`cat lib/moveplan9.files`
-
+ find ${WRKSRC} \( -name .gitignore -o -name .gitkeep -o \
+ -name \*.orig -o -name \*.beforesubst \) -delete
${INSTALL_DATA_DIR} ${PLAN9}
cd ${WRKSRC} && pax -rw . ${PLAN9}
cd ${PREFIX} && ln -s ../plan9/bin/9 bin/9
Index: distinfo
===
RCS file: /home/cvs/ports/plan9/plan9port/distinfo,v
retrieving revision 1.17
diff -u -p -r1.17 distinfo
--- distinfo25 Mar 2021 21:44:06 - 1.17
+++ distinfo26 Jan 2023 11:12:14 -
@@ -1,2 +1,2 @@
-SHA256 (plan9port-20210323-88a87fad.tar.gz) =
1LhzoJQDUCT+VCgqf91lIlW3O1HE/XiudwF0bR4+jzg=
-SIZE (plan9port-20210323-88a87fad.tar.gz) = 23413373
+SHA256 (plan9port-20230123-e2a141ae.tar.gz) =
z7Tw88brcf/PDRH055Wbw2srEW/sH+MrXpLXzUSGnFw=
+SIZE (plan9port-20230123-e2a141ae.tar.gz) = 23411241
Index: patches/patch-INSTALL
===
RCS file: /home/cvs/ports/plan9/plan9port/patches/patch-INSTALL,v
retrieving revision 1.10
diff -u -p -r1.10 patch-INSTALL
--- patches/patch-INSTALL 11 Mar 2022 19:49:08 - 1.10
+++ patches/patch-INSTALL 26 Jan 2023 11:22:51 -
@@ -1,7 +1,7 @@
Index: INSTALL
--- INSTALL.orig
+++ INSTALL
-@@ -102,7 +102,7 @@ fi
+@@ -113,7 +113,7 @@ fi
if [ `uname` != Darwin ]; then
# Determine whether fontsrv X11 files are available.
rm -f a.out
@@ -10,7 +10,7 @@ Index: INSTALL
-I/usr/X11R7/include -I/usr/X11R7/include/freetype2 \
-I/usr/X11R6/include -I/usr/X11R6/include/freetype2
src/cmd/fontsrv/x11.c >/dev/null 2>&1
if [ -f a.out ]; then
-@@ -128,7 +128,7 @@ echo "* Compiler version:"
+@@ -139,7 +139,7 @@ echo "* Compiler version:"
cd src
if $dobuild; then
echo "* Building mk..."
@@ -19,7 +19,7 @@ Index: INSTALL
if [ ! -x ../bin/mk ]; then
echo "* Error: mk failed to build."
-@@ -136,7 +136,6 @@ if $dobuild; then
+@@ -147,7 +147,6 @@ if $dobuild; then
fi
echo "* Building everything (be patient)..."
@@ -27,7 +27,7 @@ Index: INSTALL
mk libs-nuke
mk all || exit 1
if [ ! -x $PLAN9/src/cmd/o.cleanname -o ! -x $PLAN9/src/cmd/acme/o.acme
]; then
-@@ -195,5 +194,5 @@ if $doinstall; then
+@@ -206,5 +205,5 @@ if $doinstall; then
echo " PLAN9=$PLAN9 export PLAN9"
echo ' PATH=$PATH:$PLAN9/bin export PATH'
fi
Index: patches/patch-bin_9l
===
RCS file: /home/cvs/ports/plan9/plan9port/patches/patch-bin_9l,v
retrieving revision 1.4
diff -u -p -r1.4 patch-bin_9l
--- patches/patch-bin_9l11 Mar 2022 19:49:08 - 1.4
+++
Re: protobuf port fix for mips64.
> It would be super nice if you could send a PR upstream, so we don't have > to carry this patch forever. Commented with the small diff on a protobuf github issue where exactly the same problem was "fixed" for x86-windows earlier. -- May the most significant bit of your life be positive.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2023/01/28 02:10:12 Modified files: net/librenms : Makefile distinfo Log message: update to librenms-23.1.1
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: t...@cvs.openbsd.org2023/01/28 01:42:03 Modified files: www/mozilla-firefox: Makefile Added files: www/mozilla-firefox/patches: patch-media_ffvpx_libavcodec_x86_fft_asm Log message: mozilla-firefox embeds a libavcodec that can be triggered by sites doing audio recording. Port over kettenis's fft.asm fix from ffmpeg. ok landry
