summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
3 daysdependHEADmasterDamien Miller
3 dayscrank version in README and RPM spec filesDamien Miller
3 daysupstream: openssh-8.6djm@openbsd.org
OpenBSD-Commit-ID: b5f3e133c846127ec114812248bc17eff07c3e19
3 daysupstream: do not pass file/func to monitor; noted by Ilja van Sprundel;markus@openbsd.org
ok djm@ OpenBSD-Commit-ID: 85ae5c063845c410283cbdce685515dcd19479fa
5 dayssshd don't exit on transient read errorsDamien Miller
openssh-8.5 introduced a regression that would cause sshd to exit because of transient read errors on the network socket (e.g. EINTR, EAGAIN). Reported by balu.gajjala AT gmail.com via bz3297
8 daysperform report_failed_grab() inlineDamien Miller
8 daysdedicated gnome-ssk-askpass3 sourceDamien Miller
Compatibility with Wayland requires that we use the gdk_seat_grab() API for grabbing mouse/keyboard, however these API don't exist in Gtk+2. This branches gnome-ssk-askpass2.c => gnome-ssk-askpass3.c and makes the changes to use the gdk_seat_grab() instead of grabbing mouse/focus separately via GDK. In the future, we can also use the branched file to avoid some API that has been soft-deprecated in GTK+3, e.g. gtk_widget_modify_fg
10 daysEnsure valgrind-out exists.Darren Tucker
Normally the regress tests would create it, but running the unit tests on their own would fail because the directory did not exist.
10 daysPass OBJ to unit test make invocation.Darren Tucker
At least the Valgrind unit tests uses $OBJ.
11 daysAdd pattern for valgrind-unit.Darren Tucker
11 daysRun unit tests under valgrind.Darren Tucker
Run a separate build for the unit tests under Valgrind. They take long enough that running in parallel with the other Valgrind tests helps.
11 daysifdef out MIN and MAX.Darren Tucker
In -portable, defines.h ensures that these are defined, so redefining potentially causes a warning. We don't just delete it to make any future code syncs a little but easier. bz#3293.
11 daysRemove only use of warn().Darren Tucker
The warn() function is only used in one place in portable and does not exist upstream. Upgrade the only instance it's used to fail() (the privsep/sandbox+proxyconnect, from back when that was new) and remove the now-unused function.
11 daysMove make_tmpdir() into portable-specific area.Darren Tucker
Reduces diff vs OpenBSD and makes it more likely diffs will apply cleanly.
11 daysupstream: Add TEST_SSH_ELAPSED_TIMES environment variable to print thedtucker@openbsd.org
elapsed time in seconds of each test. This depends on "date +%s" which is not specified by POSIX but is commonly implemented. OpenBSD-Regress-ID: ec3c8c19ff49b2192116a0a646ee7c9b944e8a9c
11 daysMove the TEST_SSH_PORT section down a bit.Darren Tucker
This groups the portable-specific changes together and makes it a little more likely that patches will apply cleanly.
11 daysFurther split Valgrind tests.Darren Tucker
Even split in two, the Valgrind tests take by far the longest to run, so split them four ways to further increase parallelism.
12 daysupstream: include "ssherr.h" not <ssherr.h>; from Balu Gajjala viadjm@openbsd.org
bz#3292 OpenBSD-Commit-ID: e9535cd9966eb2e69e73d1ede1f44905c30310bd
12 dayswrap struct rlimit in HAVE_GETRLIMIT tooDamien Miller
12 dayswrap getrlimit call in HAVE_GETRLIMIT; bz3291Damien Miller
12 daysupstream: Don't check return value of unsetenv(). It's part of thedtucker@openbsd.org
environment setup and not part of the actual test, and some platforms -portable runs on declare it as returning void, which prevents the test from compiling. OpenBSD-Regress-ID: 24f08543ee3cdebc404f2951f3e388cc82b844a1
12 daysupstream: remove stray inserts; from matthias schmidtjmc@openbsd.org
OpenBSD-Commit-ID: 2c36ebdc54e14bbf1daad70c6a05479a073d5c63
12 daysupstream: missing comma; from kawashima jamesjmc@openbsd.org
OpenBSD-Commit-ID: 31cec6bf26c6db4ffefc8a070715ebef274e68ea
13 daysInstall libcbor with libfido2.Darren Tucker
2021-04-03enable authopt and misc unit testsDamien Miller
Neither were wired into the build, both required some build adaptations for -portable
2021-04-03upstream: typos in comments; GHPR#180 from Villdjm@openbsd.org
=?UTF-8?q?e=20Skytt=C3=A4?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit OpenBSD-Commit-ID: 93c732381ae0e2b680c79e67c40c1814b7ceed2c
2021-04-03upstream: sync CASignatureAlgorithms lists with reality. GHPR#174 fromdjm@openbsd.org
Matt Hazinski OpenBSD-Commit-ID: f05e4ca54d7e67b90fe58fe1bdb1d2a37e0e2696
2021-04-03polish whitespace for portable filesDamien Miller
2021-04-03upstream: highly polished whitespace, mostly fixing spaces-for-tabdjm@openbsd.org
and bad indentation on continuation lines. Prompted by GHPR#185 OpenBSD-Commit-ID: e5c81f0cbdcc6144df1ce468ec1bac366d8ad6e9
2021-04-03upstream: whitespace (tab after space)djm@openbsd.org
OpenBSD-Commit-ID: 0e2b3f7674e985d3f7c27ff5028e690ba1c2efd4
2021-04-03Save config.h and config.log on failure too.Darren Tucker
2021-04-03upstream: fix incorrect plural; from Ville Skytdjm@openbsd.org
=?UTF-8?q?t=C3=A4=20via=20GHPR#181?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit OpenBSD-Commit-ID: 92f31754c6296d8f403d7c293e09dc27292d22c9
2021-04-03upstream: ensure that pkcs11_del_provider() is called before exit -djm@openbsd.org
some PKCS#11 providers get upset if C_Initialize is not matched with C_Finalize. From Adithya Baglody via GHPR#234; ok markus OpenBSD-Commit-ID: f8e770e03b416ee9a58f9762e162add900f832b6
2021-04-03upstream: unused variabledjm@openbsd.org
OpenBSD-Commit-ID: 85f6a394c8e0f60d15ecddda75176f112007b205
2021-04-03upstream: Fix two problems in string->argv conversion: 1) multipledjm@openbsd.org
backslashes were not being dequoted correctly and 2) quoted space in the middle of a string was being incorrectly split. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit A unit test for these cases has already been committed prompted by and based on GHPR#223 by Eero Häkkinen; ok markus@ OpenBSD-Commit-ID: d7ef27abb4eeeaf6e167e9312e4abe9e89faf1e4
2021-04-03missing bits from 259d648eDamien Miller
2021-04-01upstream: cannot effectively test posix-rename extension afterdjm@openbsd.org
changes in feature advertisment. OpenBSD-Regress-ID: 5e390bf88d379162aaa81b60ed86b34cb0c54d29
2021-04-01upstream: add a test for misc.c:argv_split(), currently failsdjm@openbsd.org
OpenBSD-Regress-ID: ad6b96d6ebeb9643b698b3575bdd6f78bb144200
2021-04-01upstream: splitdjm@openbsd.org
OpenBSD-Regress-ID: f6c03c0e4c58b3b9e04b161757b8c10dc8378c34
2021-04-01upstream: Use new limits@openssh.com protocol extension to let thedjm@openbsd.org
client select good limits based on what the server supports. Split the download and upload buffer sizes to allow them to be chosen independently. In practice (and assuming upgraded sftp/sftp-server at each end), this increases the download buffer 32->64KiB and the upload buffer 32->255KiB. Patches from Mike Frysinger; ok dtucker@ OpenBSD-Commit-ID: ebd61c80d85b951b794164acc4b2f2fd8e88606c
2021-04-01upstream: do not advertise protocol extensions that have beendjm@openbsd.org
disallowed by the command-line options (e.g. -p/-P/-R); ok dtucker@ OpenBSD-Commit-ID: 3a8a76b3f5131741aca4b41bfab8d101c9926205
2021-03-29gnome-ssh-askpass3 is a valid target hereDamien Miller
2021-03-19upstream: return non-zero exit status when killed by signal; bz#3281 okdjm@openbsd.org
dtucker@ OpenBSD-Commit-ID: 117b31cf3c807993077b596bd730c24da9e9b816
2021-03-19upstream: increase maximum SSH2_FXP_READ to match the maximumdjm@openbsd.org
packet size. Also handle zero-length reads that are borderline nonsensical but not explicitly banned by the spec. Based on patch from Mike Frysinger, feedback deraadt@ ok dtucker@ OpenBSD-Commit-ID: 4e67d60d81bde7b84a742b4ee5a34001bdf80d9c
2021-03-19upstream: don't let logging clobber errno before usedjm@openbsd.org
OpenBSD-Commit-ID: ce6cca370005c270c277c51c111bb6911e1680ec
2021-03-13Only call dh_set_moduli_file if using OpenSSL.Darren Tucker
Fixes link failure when configuring --without-openssl since dh.c is not linked in.
2021-03-13Don't install moduli during tests.Darren Tucker
Now that we have TEST_SSH_MODULI_FILE pointing to the moduli in the soure directory we don't need to install the file to prevent warnings about it being missing.
2021-03-13Point TEST_SSH_MODULI_FILE at our own moduli.Darren Tucker
This will allow the test to run without requiring a moduli file installed at the configured default path.
2021-03-13upstream: spellingjsg@openbsd.org
OpenBSD-Commit-ID: 478bc3db04f62f1048ed6e1765400f3ab325e60f
2021-03-13upstream: Add ModuliFile keyword to sshd_config to specify thedtucker@openbsd.org
location of the "moduli" file containing the groups for DH-GEX. This will allow us to run tests against arbitrary moduli files without having to install them. ok djm@ OpenBSD-Commit-ID: 8df99d60b14ecaaa28f3469d01fc7f56bff49f66