Aren/haiku
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

kernel/vfs: Fix missing lock in fs_mount().

Browse Source 
At least a read lock of the sVnodeLock must be held when calling
lookup_vnode, but we held none at all. This rectifies that problem.

This bug appears to have been around for many years, but no-one
noticed since ASSERT_READ_LOCKED_RW_LOCK only works with more
debug options turned on than the kernel is built with. I discovered
this while working on a new version of those additional options.
This commit is contained in:
Augustin Cavalier 2023-06-19 14:27:48 -04:00
parent 5e8058566c
commit f96456d863
1 changed files with 2 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
src/system/kernel/fs/vfs.cpp
View File

@ -7677,15 +7677,16 @@ fs_mount(char* path, const char* device, const char* fsName, uint32 flags,
// the root node is supposed to be owned by the file system - it must
// exist at this point
rw_lock_write_lock(&sVnodeLock);
mount->root_vnode = lookup_vnode(mount->id, rootID);
if (mount->root_vnode == NULL || mount->root_vnode->ref_count != 1) {
panic("fs_mount: file system does not own its root node!\n");
status = B_ERROR;
rw_lock_write_unlock(&sVnodeLock);
goto err4;
}
// set up the links between the root vnode and the vnode it covers
rw_lock_write_lock(&sVnodeLock);
if (coveredNode != NULL) {
if (coveredNode->IsCovered()) {
// the vnode is covered now