Date: Sat, 5 Sep 2020 00:26:04 +0000 (UTC) From: Chuck Silvers <chs@FreeBSD.org> To: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: svn commit: r365351 - head/sys/kern Message-ID: <202009050026.0850Q4Gk023051@repo.freebsd.org>
next in thread | raw e-mail | index | archive | help
Author: chs Date: Sat Sep 5 00:26:03 2020 New Revision: 365351 URL: https://svnweb.freebsd.org/changeset/base/365351 Log: vfs: avoid exposing partially constructed vnodes If multiple threads race calling vfs_hash_insert() while creating vnodes with the same identity, all of the vnodes which lose the race must be destroyed before any other thread can see them. Previously this was accomplished by the vput() in vfs_hash_insert() resulting in the vnode's VOP_INACTIVE() method calling vgone() before the vnode lock was unlocked, but at some point changes to the the vnode refcount/inactive logic have caused that to no longer work, leading to crashes, so instead vfs_hash_insert() must call vgone() itself before calling vput() on vnodes which lose the race. Reviewed by: mjg, kib Sponsored by: Netflix Differential Revision: https://reviews.freebsd.org/D26291 Modified: head/sys/kern/vfs_hash.c Modified: head/sys/kern/vfs_hash.c ============================================================================== --- head/sys/kern/vfs_hash.c Sat Sep 5 00:20:32 2020 (r365350) +++ head/sys/kern/vfs_hash.c Sat Sep 5 00:26:03 2020 (r365351) @@ -172,6 +172,7 @@ vfs_hash_insert(struct vnode *vp, u_int hash, int flag rw_wlock(&vfs_hash_lock); LIST_INSERT_HEAD(&vfs_hash_side, vp, v_hashlist); rw_wunlock(&vfs_hash_lock); + vgone(vp); vput(vp); if (!error) *vpp = vp2;
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?202009050026.0850Q4Gk023051>