Author: mjg
Date: Mon May 25 04:57:57 2020
New Revision: 361441
URL: https://svnweb.freebsd.org/changeset/base/361441
Log:
vfs: use atomic_{store,load}_long to manage f_offset
... instead of depending on the compiler not to mess them up
Modified:
head/sys/kern/vfs_vnops.c
Modified: head/sys/kern/vfs_vnops.c
==============================================================================
--- head/sys/kern/vfs_vnops.c Mon May 25 04:56:41 2020 (r361440)
+++ head/sys/kern/vfs_vnops.c Mon May 25 04:57:57 2020 (r361441)
@@ -682,12 +682,8 @@ foffset_lock(struct file *fp, int flags)
KASSERT((flags & FOF_OFFSET) == 0, ("FOF_OFFSET passed"));
- /*
- * Caller only wants the current f_offset value. Assume that
- * the long and shorter integer types reads are atomic.
- */
if ((flags & FOF_NOLOCK) != 0)
- return (fp->f_offset);
+ return (atomic_load_long(&fp->f_offset));
/*
* According to McKusick the vn lock was protecting f_offset here.
@@ -695,7 +691,7 @@ foffset_lock(struct file *fp, int flags)
*/
flagsp = &fp->f_vnread_flags;
if (atomic_cmpset_acq_16(flagsp, 0, FOFFSET_LOCKED))
- return (fp->f_offset);
+ return (atomic_load_long(&fp->f_offset));
sleepq_lock(&fp->f_vnread_flags);
state = atomic_load_16(flagsp);
@@ -718,7 +714,7 @@ foffset_lock(struct file *fp, int flags)
sleepq_lock(&fp->f_vnread_flags);
state = atomic_load_16(flagsp);
}
- res = fp->f_offset;
+ res = atomic_load_long(&fp->f_offset);
sleepq_release(&fp->f_vnread_flags);
return (res);
}
@@ -732,7 +728,7 @@ foffset_unlock(struct file *fp, off_t val, int flags)
KASSERT((flags & FOF_OFFSET) == 0, ("FOF_OFFSET passed"));
if ((flags & FOF_NOUPDATE) == 0)
- fp->f_offset = val;
+ atomic_store_long(&fp->f_offset, val);
if ((flags & FOF_NEXTOFF) != 0)
fp->f_nextoff = val;
_______________________________________________
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
