2 Commits

Author	SHA1	Message	Date
Richard Yao	c38367c73f	Eliminate runtime function pointer mods in autotools checks ... PaX/GrSecurity patched kernels implement a dialect of C that relies on a GCC plugin for enforcement. A basic idea in this dialect is that function pointers in structures should not change during runtime. This causes code that modifies function pointers at runtime to fail to compile in many instances. The autotools checks rely on whether or not small test cases compile against a given kernel. Some autotools checks assume some default case if other cases fail. When one of these autotools checks tests a PaX/GrSecurity patched kernel by modifying a function pointer at runtime, the default case will be used. Early detection of such situations is possible by relying on compiler warnings, which are compiler errors when --enable-debug is used. Unfortunately, very few people build ZFS with --enable-debug. The more common situation is that these issues manifest themselves as runtime failures in the form of NULL pointer exceptions. Previous patches that addressed such issues with PaX/GrSecurity compatibility largely relied on rewriting autotools checks to avoid runtime function pointer modification or the addition of PaX/GrSecurity specific checks. This patch takes the previous work to its logical conclusion by eliminating the use of runtime function pointer modification. This permits the removal of PaX-specific autotools checks in favor of ones that work across all supported kernels. This should resolve issues that were reported to occur with PaX/GrSecurity-patched Linux 3.7.5 kernels on Gentoo Linux. https://bugs.gentoo.org/show_bug.cgi?id=457176 We should be able to prevent future regressions in PaX/GrSecurity compatibility by ensuring that all changes to ZFSOnLinux avoid runtime function pointer modification. At the same time, this does not solve the issue of silent failures triggering default cases in the autotools check, which is what permitted these regressions to become runtime failures in the first place. This will need to be addressed in a future patch. Reported-by: Marcin Mirosław <bug@mejor.pl> Signed-off-by: Richard Yao <ryao@cs.stonybrook.edu> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1300	2013-03-04 08:49:17 -08:00
Brian Behlendorf	ab26409db7	Linux 3.1 compat, super_block->s_shrink ... The Linux 3.1 kernel has introduced the concept of per-filesystem shrinkers which are directly assoicated with a super block. Prior to this change there was one shared global shrinker. The zfs code relied on being able to call the global shrinker when the arc_meta_limit was exceeded. This would cause the VFS to drop references on a fraction of the dentries in the dcache. The ARC could then safely reclaim the memory used by these entries and honor the arc_meta_limit. Unfortunately, when per-filesystem shrinkers were added the old interfaces were made unavailable. This change adds support to use the new per-filesystem shrinker interface so we can continue to honor the arc_meta_limit. The major benefit of the new interface is that we can now target only the zfs filesystem for dentry and inode pruning. Thus we can minimize any impact on the caching of other filesystems. In the context of making this change several other important issues related to managing the ARC were addressed, they include: * The dnlc_reduce_cache() function which was called by the ARC to drop dentries for the Posix layer was replaced with a generic zfs_prune_t callback. The ZPL layer now registers a callback to drop these dentries removing a layering violation which dates back to the Solaris code. This callback can also be used by other ARC consumers such as Lustre. arc_add_prune_callback() arc_remove_prune_callback() * The arc_reduce_dnlc_percent module option has been changed to arc_meta_prune for clarity. The dnlc functions are specific to Solaris's VFS and have already been largely eliminated already. The replacement tunable now represents the number of bytes the prune callback will request when invoked. * Less aggressively invoke the prune callback. We used to call this whenever we exceeded the arc_meta_limit however that's not strictly correct since it results in over zeleous reclaim of dentries and inodes. It is now only called once the arc_meta_limit is exceeded and every effort has been made to evict other data from the ARC cache. * More promptly manage exceeding the arc_meta_limit. When reading meta data in to the cache if a buffer was unable to be recycled notify the arc_reclaim thread to invoke the required prune. * Added arcstat_prune kstat which is incremented when the ARC is forced to request that a consumer prune its cache. Remember this will only occur when the ARC has no other choice. If it can evict buffers safely without invoking the prune callback it will. * This change is also expected to resolve the unexpect collapses of the ARC cache. This would occur because when exceeded just the arc_meta_limit reclaim presure would be excerted on the arc_c value via arc_shrink(). This effectively shrunk the entire cache when really we just needed to reclaim meta data. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #466 Closes #292	2012-01-11 11:46:02 -08:00