Date: Tue, 19 May 2026 07:33:02 +0000 From: bugzilla-noreply@freebsd.org To: testing@FreeBSD.org Subject: [Bug 295380] Errors in tests/sys/geom/class/raid3 Message-ID: <bug-295380-32464-V5UkF7v5To@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-295380-32464@https.bugs.freebsd.org/bugzilla/>
index | next in thread | previous in thread | raw e-mail
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=295380 Dag-Erling Smørgrav <des@FreeBSD.org> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |des@FreeBSD.org --- Comment #2 from Dag-Erling Smørgrav <des@FreeBSD.org> --- The situation is actually a little more complicated. 1. In their current state, the graid3 tests don't actually test adding a disk to an array, because they pass the wrong device name to `graid3 insert` and fail to notice that the command fails. The patch in this PR partially fixes this: it corrects the device name, but the test will still not fail if `graid3 insert` (or any other command executed by the test) fails for some other reason. 2. When the patched tests are run on a debugging kernel, the `graid3 insert` command trips a locking assertion, resulting in a kernel panic. I recommend converting these tests to atf-sh(3), although adding `set -e` at the top of each script will do in a pinch. The tests also make unnecessary use of expr(1) for simple arithmetic which the shell is perfectly capable of performing, and use elaborate sequences of dd(1) and md5(1) to compare the array to its expected contents when a simple cmp(1) would do. I'll leave this open as a suitable junior hacker task, the locking issue will be dealt with in bug 295407. -- You are receiving this mail because: You are the assignee for the bug.home | help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-295380-32464-V5UkF7v5To>