1.. SPDX-License-Identifier: GPL-2.0 2 3========================== 4File Locking Release Notes 5========================== 6 7 Andy Walker <andy@lysaker.kvaerner.no> 8 9 12 May 1997 10 11 121. What's New? 13============== 14 151.1 Broken Flock Emulation 16-------------------------- 17 18The old flock(2) emulation in the kernel was swapped for proper BSD 19compatible flock(2) support in the 1.3.x series of kernels. With the 20release of the 2.1.x kernel series, support for the old emulation has 21been totally removed, so that we don't need to carry this baggage 22forever. 23 24This should not cause problems for anybody, since everybody using a 252.1.x kernel should have updated their C library to a suitable version 26anyway (see the file "Documentation/process/changes.rst".) 27 281.2 Allow Mixed Locks Again 29--------------------------- 30 311.2.1 Typical Problems - Sendmail 32^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 33Because sendmail was unable to use the old flock() emulation, many sendmail 34installations use fcntl() instead of flock(). This is true of Slackware 3.0 35for example. This gave rise to some other subtle problems if sendmail was 36configured to rebuild the alias file. Sendmail tried to lock the aliases.dir 37file with fcntl() at the same time as the GDBM routines tried to lock this 38file with flock(). With pre 1.3.96 kernels this could result in deadlocks that, 39over time, or under a very heavy mail load, would eventually cause the kernel 40to lock solid with deadlocked processes. 41 42 431.2.2 The Solution 44^^^^^^^^^^^^^^^^^^ 45The solution I have chosen, after much experimentation and discussion, 46is to make flock() and fcntl() locks oblivious to each other. Both can 47exists, and neither will have any effect on the other. 48 49I wanted the two lock styles to be cooperative, but there were so many 50race and deadlock conditions that the current solution was the only 51practical one. It puts us in the same position as, for example, SunOS 524.1.x and several other commercial Unices. The only OS's that support 53cooperative flock()/fcntl() are those that emulate flock() using 54fcntl(), with all the problems that implies. 55 56 571.3 Mandatory Locking As A Mount Option 58--------------------------------------- 59 60Mandatory locking, as described in 61'Documentation/filesystems/mandatory-locking.rst' was prior to this release a 62general configuration option that was valid for all mounted filesystems. This 63had a number of inherent dangers, not the least of which was the ability to 64freeze an NFS server by asking it to read a file for which a mandatory lock 65existed. 66 67From this release of the kernel, mandatory locking can be turned on and off 68on a per-filesystem basis, using the mount options 'mand' and 'nomand'. 69The default is to disallow mandatory locking. The intention is that 70mandatory locking only be enabled on a local filesystem as the specific need 71arises. 72 73