Date: Sat, 13 Jan 2024 18:44:06 +0000 From: bugzilla-noreply@freebsd.org To: bugs@FreeBSD.org Subject: [Bug 276057] crash in xa_destroy() while initializing the i915kms module Message-ID: <bug-276057-227-IUmVMeNmMO@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-276057-227@https.bugs.freebsd.org/bugzilla/>
index | next in thread | previous in thread | raw e-mail
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=276057 --- Comment #8 from Donn Seeley <donn@xmission.com> --- (In reply to Konstantin Belousov from comment #2) I had some time to look at this issue this morning. Some questions: If you can't reinitialize a mutex, then what is the point of MTX_NEW? The mutex(9) man page says that MTX_NEW prevents an assertion from firing if you call mtx_init() on a mutex without calling mtx_destroy() first. That language implies that you *can* safely call mtx_init() on a mutex that has been destroyed. If there's some reason why the witness code would have a problem with reinitializing a mutex, could we not just initialize the mutex with MTX_NOWITNESS? I noticed that the FreeBSD emulation of Linux spinlocks uses MTX_NOWITNESS, and the xarray code in Linux uses a spinlock for protection, so there's some precedent here. Comments, suggestions? -- You are receiving this mail because: You are the assignee for the bug.help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-276057-227-IUmVMeNmMO>