Date: Fri, 3 Jan 2020 22:12:25 +0000 (UTC) From: Mateusz Guzik <mjg@FreeBSD.org> To: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: svn commit: r356336 - head/sys/fs/unionfs Message-ID: <202001032212.003MCP5L067726@repo.freebsd.org>
next in thread | raw e-mail | index | archive | help
Author: mjg Date: Fri Jan 3 22:12:25 2020 New Revision: 356336 URL: https://svnweb.freebsd.org/changeset/base/356336 Log: unionfs: fix up VOP_UNLOCK use after flags stopped being supported For the most part the code was passing the LK_RELEASE flag. The 2 cases which did not use the VOP_UNLOCK_FLAGS macro. This fixes a panic when stacking unionfs on top of e.g., tmpfs when debug is enabled. Note there are latent bugs which prevent unionfs from working with debug regardless of this change. PR: 243064 Reported by: Mason Loring Bliss Modified: head/sys/fs/unionfs/union_subr.c head/sys/fs/unionfs/union_vfsops.c head/sys/fs/unionfs/union_vnops.c Modified: head/sys/fs/unionfs/union_subr.c ============================================================================== --- head/sys/fs/unionfs/union_subr.c Fri Jan 3 22:10:11 2020 (r356335) +++ head/sys/fs/unionfs/union_subr.c Fri Jan 3 22:12:25 2020 (r356336) @@ -361,9 +361,9 @@ unionfs_noderem(struct vnode *vp, struct thread *td) VI_UNLOCK(vp); if (lvp != NULLVP) - VOP_UNLOCK(lvp, LK_RELEASE); + VOP_UNLOCK(lvp, 0); if (uvp != NULLVP) - VOP_UNLOCK(uvp, LK_RELEASE); + VOP_UNLOCK(uvp, 0); if (dvp != NULLVP && unp->un_hash.le_prev != NULL) unionfs_rem_cached_vnode(unp, dvp); @@ -551,7 +551,7 @@ unionfs_relookup(struct vnode *dvp, struct vnode **vpp cn->cn_flags |= NOCACHE; vref(dvp); - VOP_UNLOCK(dvp, LK_RELEASE); + VOP_UNLOCK(dvp, 0); if ((error = relookup(dvp, vpp, cn))) { uma_zfree(namei_zone, cn->cn_pnbuf); @@ -961,7 +961,7 @@ unionfs_vn_create_on_upper(struct vnode **vpp, struct } unionfs_vn_create_on_upper_free_out1: - VOP_UNLOCK(udvp, LK_RELEASE); + VOP_UNLOCK(udvp, 0); unionfs_vn_create_on_upper_free_out2: if (cn.cn_flags & HASBUF) { Modified: head/sys/fs/unionfs/union_vfsops.c ============================================================================== --- head/sys/fs/unionfs/union_vfsops.c Fri Jan 3 22:10:11 2020 (r356335) +++ head/sys/fs/unionfs/union_vfsops.c Fri Jan 3 22:12:25 2020 (r356336) @@ -167,7 +167,7 @@ unionfs_domount(struct mount *mp) uid = va.va_uid; gid = va.va_gid; } - VOP_UNLOCK(mp->mnt_vnodecovered, LK_RELEASE); + VOP_UNLOCK(mp->mnt_vnodecovered, 0); if (error) return (error); @@ -252,7 +252,7 @@ unionfs_domount(struct mount *mp) * Save reference */ if (below) { - VOP_UNLOCK(upperrootvp, LK_RELEASE); + VOP_UNLOCK(upperrootvp, 0); vn_lock(lowerrootvp, LK_EXCLUSIVE | LK_RETRY); ump->um_lowervp = upperrootvp; ump->um_uppervp = lowerrootvp; @@ -278,7 +278,7 @@ unionfs_domount(struct mount *mp) /* * Unlock the node */ - VOP_UNLOCK(ump->um_uppervp, LK_RELEASE); + VOP_UNLOCK(ump->um_uppervp, 0); /* * Get the unionfs root vnode. Modified: head/sys/fs/unionfs/union_vnops.c ============================================================================== --- head/sys/fs/unionfs/union_vnops.c Fri Jan 3 22:10:11 2020 (r356335) +++ head/sys/fs/unionfs/union_vnops.c Fri Jan 3 22:12:25 2020 (r356336) @@ -128,7 +128,7 @@ unionfs_lookup(struct vop_cachedlookup_args *ap) if (udvp != NULLVP) { dtmpvp = udvp; if (ldvp != NULLVP) - VOP_UNLOCK(ldvp, LK_RELEASE); + VOP_UNLOCK(ldvp, 0); } else dtmpvp = ldvp; @@ -136,7 +136,7 @@ unionfs_lookup(struct vop_cachedlookup_args *ap) error = VOP_LOOKUP(dtmpvp, &vp, cnp); if (dtmpvp == udvp && ldvp != NULLVP) { - VOP_UNLOCK(udvp, LK_RELEASE); + VOP_UNLOCK(udvp, 0); vn_lock(dvp, LK_EXCLUSIVE | LK_RETRY); } @@ -148,10 +148,10 @@ unionfs_lookup(struct vop_cachedlookup_args *ap) */ if (nameiop == DELETE || nameiop == RENAME || (cnp->cn_lkflags & LK_TYPE_MASK)) - VOP_UNLOCK(vp, LK_RELEASE); + VOP_UNLOCK(vp, 0); vrele(vp); - VOP_UNLOCK(dvp, LK_RELEASE); + VOP_UNLOCK(dvp, 0); *(ap->a_vpp) = dunp->un_dvp; vref(dunp->un_dvp); @@ -188,7 +188,7 @@ unionfs_lookup(struct vop_cachedlookup_args *ap) } if (nameiop == DELETE || nameiop == RENAME || (cnp->cn_lkflags & LK_TYPE_MASK)) - VOP_UNLOCK(uvp, LK_RELEASE); + VOP_UNLOCK(uvp, 0); } /* check whiteout */ @@ -232,7 +232,7 @@ unionfs_lookup(struct vop_cachedlookup_args *ap) return (lerror); } if (cnp->cn_lkflags & LK_TYPE_MASK) - VOP_UNLOCK(lvp, LK_RELEASE); + VOP_UNLOCK(lvp, 0); } } @@ -267,7 +267,7 @@ unionfs_lookup(struct vop_cachedlookup_args *ap) goto unionfs_lookup_out; if (LK_SHARED == (cnp->cn_lkflags & LK_TYPE_MASK)) - VOP_UNLOCK(vp, LK_RELEASE); + VOP_UNLOCK(vp, 0); if (LK_EXCLUSIVE != VOP_ISLOCKED(vp)) { vn_lock(vp, LK_EXCLUSIVE | LK_RETRY); lockflag = 1; @@ -275,7 +275,7 @@ unionfs_lookup(struct vop_cachedlookup_args *ap) error = unionfs_mkshadowdir(MOUNTTOUNIONFSMOUNT(dvp->v_mount), udvp, VTOUNIONFS(vp), cnp, td); if (lockflag != 0) - VOP_UNLOCK(vp, LK_RELEASE); + VOP_UNLOCK(vp, 0); if (error != 0) { UNIONFSDEBUG("unionfs_lookup: Unable to create shadow dir."); if ((cnp->cn_lkflags & LK_TYPE_MASK) == LK_EXCLUSIVE) @@ -372,7 +372,7 @@ unionfs_create(struct vop_create_args *ap) if (vp->v_type == VSOCK) *(ap->a_vpp) = vp; else { - VOP_UNLOCK(vp, LK_RELEASE); + VOP_UNLOCK(vp, 0); error = unionfs_nodeget(ap->a_dvp->v_mount, vp, NULLVP, ap->a_dvp, ap->a_vpp, cnp, curthread); vrele(vp); @@ -446,7 +446,7 @@ unionfs_mknod(struct vop_mknod_args *ap) if (vp->v_type == VSOCK) *(ap->a_vpp) = vp; else { - VOP_UNLOCK(vp, LK_RELEASE); + VOP_UNLOCK(vp, 0); error = unionfs_nodeget(ap->a_dvp->v_mount, vp, NULLVP, ap->a_dvp, ap->a_vpp, cnp, curthread); vrele(vp); @@ -903,7 +903,7 @@ unionfs_ioctl(struct vop_ioctl_args *ap) unionfs_get_node_status(unp, ap->a_td, &unsp); ovp = (unsp->uns_upper_opencnt ? unp->un_uppervp : unp->un_lowervp); unionfs_tryrem_node_status(unp, unsp); - VOP_UNLOCK(ap->a_vp, LK_RELEASE); + VOP_UNLOCK(ap->a_vp, 0); if (ovp == NULLVP) return (EBADF); @@ -930,7 +930,7 @@ unionfs_poll(struct vop_poll_args *ap) unionfs_get_node_status(unp, ap->a_td, &unsp); ovp = (unsp->uns_upper_opencnt ? unp->un_uppervp : unp->un_lowervp); unionfs_tryrem_node_status(unp, unsp); - VOP_UNLOCK(ap->a_vp, LK_RELEASE); + VOP_UNLOCK(ap->a_vp, 0); if (ovp == NULLVP) return (EBADF); @@ -990,7 +990,7 @@ unionfs_remove(struct vop_remove_args *ap) ump = NULL; vp = uvp = lvp = NULLVP; /* search vnode */ - VOP_UNLOCK(ap->a_vp, LK_RELEASE); + VOP_UNLOCK(ap->a_vp, 0); error = unionfs_relookup(udvp, &vp, cnp, &cn, td, cnp->cn_nameptr, strlen(cnp->cn_nameptr), DELETE); if (error != 0 && error != ENOENT) { @@ -1193,7 +1193,7 @@ unionfs_rename(struct vop_rename_args *ap) if ((error = vn_lock(fvp, LK_EXCLUSIVE)) != 0) goto unionfs_rename_abort; error = unionfs_copyfile(unp, 1, fcnp->cn_cred, td); - VOP_UNLOCK(fvp, LK_RELEASE); + VOP_UNLOCK(fvp, 0); if (error != 0) goto unionfs_rename_abort; break; @@ -1201,7 +1201,7 @@ unionfs_rename(struct vop_rename_args *ap) if ((error = vn_lock(fvp, LK_EXCLUSIVE)) != 0) goto unionfs_rename_abort; error = unionfs_mkshadowdir(ump, rfdvp, unp, fcnp, td); - VOP_UNLOCK(fvp, LK_RELEASE); + VOP_UNLOCK(fvp, 0); if (error != 0) goto unionfs_rename_abort; break; @@ -1258,13 +1258,13 @@ unionfs_rename(struct vop_rename_args *ap) if ((error = vn_lock(fdvp, LK_EXCLUSIVE)) != 0) goto unionfs_rename_abort; error = unionfs_relookup_for_delete(fdvp, fcnp, td); - VOP_UNLOCK(fdvp, LK_RELEASE); + VOP_UNLOCK(fdvp, 0); if (error != 0) goto unionfs_rename_abort; /* Locke of tvp is canceled in order to avoid recursive lock. */ if (tvp != NULLVP && tvp != tdvp) - VOP_UNLOCK(tvp, LK_RELEASE); + VOP_UNLOCK(tvp, 0); error = unionfs_relookup_for_rename(tdvp, tcnp, td); if (tvp != NULLVP && tvp != tdvp) vn_lock(tvp, LK_EXCLUSIVE | LK_RETRY); @@ -1282,11 +1282,11 @@ unionfs_rename(struct vop_rename_args *ap) } if (ltdvp != NULLVP) - VOP_UNLOCK(ltdvp, LK_RELEASE); + VOP_UNLOCK(ltdvp, 0); if (tdvp != rtdvp) vrele(tdvp); if (ltvp != NULLVP) - VOP_UNLOCK(ltvp, LK_RELEASE); + VOP_UNLOCK(ltvp, 0); if (tvp != rtvp && tvp != NULLVP) { if (rtvp == NULLVP) vput(tvp); @@ -1360,7 +1360,7 @@ unionfs_mkdir(struct vop_mkdir_args *ap) } if ((error = VOP_MKDIR(udvp, &uvp, cnp, ap->a_vap)) == 0) { - VOP_UNLOCK(uvp, LK_RELEASE); + VOP_UNLOCK(uvp, 0); cnp->cn_lkflags = LK_EXCLUSIVE; error = unionfs_nodeget(ap->a_dvp->v_mount, uvp, NULLVP, ap->a_dvp, ap->a_vpp, cnp, td); @@ -1458,7 +1458,7 @@ unionfs_symlink(struct vop_symlink_args *ap) if (udvp != NULLVP) { error = VOP_SYMLINK(udvp, &uvp, cnp, ap->a_vap, ap->a_target); if (error == 0) { - VOP_UNLOCK(uvp, LK_RELEASE); + VOP_UNLOCK(uvp, 0); cnp->cn_lkflags = LK_EXCLUSIVE; error = unionfs_nodeget(ap->a_dvp->v_mount, uvp, NULLVP, ap->a_dvp, ap->a_vpp, cnp, td); @@ -1817,7 +1817,7 @@ static void unionfs_revlock(struct vnode *vp, int flags) { if (flags & LK_RELEASE) - VOP_UNLOCK(vp, flags); + VOP_UNLOCK_FLAGS(vp, flags); else { /* UPGRADE */ if (vn_lock(vp, flags) != 0) @@ -1849,7 +1849,7 @@ unionfs_lock(struct vop_lock1_args *ap) vp = ap->a_vp; if (LK_RELEASE == (flags & LK_TYPE_MASK) || !(flags & LK_TYPE_MASK)) - return (VOP_UNLOCK(vp, flags | LK_RELEASE)); + return (VOP_UNLOCK_FLAGS(vp, flags | LK_RELEASE)); if ((flags & LK_INTERLOCK) == 0) VI_LOCK(vp); @@ -1884,12 +1884,12 @@ unionfs_lock(struct vop_lock1_args *ap) /* Share Lock is once released and a deadlock is avoided. */ vholdnz(uvp); uhold = 1; - VOP_UNLOCK(uvp, LK_RELEASE); + VOP_UNLOCK(uvp, 0); unp = VTOUNIONFS(vp); if (unp == NULL) { /* vnode is released. */ VI_UNLOCK(vp); - VOP_UNLOCK(lvp, LK_RELEASE); + VOP_UNLOCK(lvp, 0); vdrop(uvp); return (EBUSY); } @@ -1909,7 +1909,7 @@ unionfs_lock(struct vop_lock1_args *ap) /* vnode is released. */ VI_UNLOCK(vp); if (error == 0) - VOP_UNLOCK(lvp, LK_RELEASE); + VOP_UNLOCK(lvp, 0); vdrop(lvp); if (uhold != 0) vdrop(uvp); @@ -1940,10 +1940,10 @@ unionfs_lock(struct vop_lock1_args *ap) /* vnode is released. */ VI_UNLOCK(vp); if (error == 0) - VOP_UNLOCK(uvp, LK_RELEASE); + VOP_UNLOCK(uvp, 0); vdrop(uvp); if (lvp != NULLVP) { - VOP_UNLOCK(lvp, LK_RELEASE); + VOP_UNLOCK(lvp, 0); vdrop(lvp); } return (vop_stdlock(ap)); @@ -1974,7 +1974,6 @@ static int unionfs_unlock(struct vop_unlock_args *ap) { int error; - int flags; int uhold; struct vnode *vp; struct vnode *lvp; @@ -1985,7 +1984,6 @@ unionfs_unlock(struct vop_unlock_args *ap) error = 0; uhold = 0; - flags = ap->a_flags | LK_RELEASE; vp = ap->a_vp; unp = VTOUNIONFS(vp); @@ -1996,13 +1994,13 @@ unionfs_unlock(struct vop_unlock_args *ap) if (lvp != NULLVP) { vholdnz(lvp); - error = VOP_UNLOCK(lvp, flags); + error = VOP_UNLOCK(lvp, 0); } if (error == 0 && uvp != NULLVP) { vholdnz(uvp); uhold = 1; - error = VOP_UNLOCK(uvp, flags); + error = VOP_UNLOCK(uvp, 0); } if (lvp != NULLVP) @@ -2072,7 +2070,7 @@ unionfs_advlock(struct vop_advlock_args *ap) unionfs_tryrem_node_status(unp, unsp); } - VOP_UNLOCK(vp, LK_RELEASE); + VOP_UNLOCK(vp, 0); error = VOP_ADVLOCK(uvp, ap->a_id, ap->a_op, ap->a_fl, ap->a_flags); @@ -2081,7 +2079,7 @@ unionfs_advlock(struct vop_advlock_args *ap) return error; unionfs_advlock_abort: - VOP_UNLOCK(vp, LK_RELEASE); + VOP_UNLOCK(vp, 0); UNIONFS_INTERNAL_DEBUG("unionfs_advlock: leave (%d)\n", error);
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?202001032212.003MCP5L067726>