Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
Commits on Apr 8, 2011
Commits on Feb 25, 2011
  1. Added call to unlock_new_inode in lzfs_xattr_set

    Prasad Joshi authored
Commits on Feb 21, 2011
  1. create, mknod, mkdir, symlink should call unlock_new_inode after dent…

    Prasad Joshi authored
    …ry is being instantiated. This, along with related zfs commit, fixes the iput BUG (inode->i_state == I_CLEAR).
    
    - Code working on 2.6.32 need to check with 2.6.37
Commits on Feb 18, 2011
  1. 2.6.37 API changes.

    Prasad Joshi authored
Commits on Feb 16, 2011
  1. @tushk2

    Merge branch 'master' of git@github.com:zfs-linux/lzfs into debian

    tushk2 authored
    Conflicts:
    
    	module/lzfs_vnops.c
    	module/lzfs_xattr.c
    	module/lzfs_xattr_security.c
    	module/lzfs_xattr_user.c
Commits on Feb 5, 2011
  1. Fixed a typo

    Prasad Joshi authored
Commits on Feb 4, 2011
  1. @tushk2
  2. Removing fsync_compat.h file inclusion.

    Prasad Joshi authored
  3. Changes from Massimo Maggi <massimo@mmmm.it>, fixing few pjd-fstest-u…

    Prasad Joshi authored
    …nstable ACL testcases.
  4. The file pointer passed to lzfs_vnop_fsync can be NULL on 2.6.32 and …

    Prasad Joshi authored
    …below. Use dentry pointer to get the inode.
Commits on Feb 3, 2011
  1. @tushk2
Commits on Jan 30, 2011
  1. Fixed a getfacl bug

    Prasad Joshi authored
  2. Use proper kernel versioning

    Prasad Joshi authored
  3. ACL support

    Prasad Joshi authored
  4. - Rearragned common code

    Prasad Joshi authored
    - Added lzfs_xattr_set a common function for extended attributes
    - Removed index parameter from lzfs_xattr_get(), caller is supposed to allocate memory for extended attributes name
Commits on Jan 20, 2011
  1. @deepenmehta
Commits on Jan 18, 2011
  1. @nkhare
Commits on Jan 10, 2011
  1. @invalid-email-address

    NFS uses linux aio to if it is available. We are not handling not han…

    invalid-email-address authored
    …dling
    
    these correctly leading to data verification failure when we read over NFS
    If we disable AIO support NFS falls back on standard read/write which works
    fine.
    
    Disabling AIO for now to make sure NFS works correctly. Will enable once we
    figure out what is going wrong within AIO.
Commits on Jan 7, 2011
  1. @nkhare

    SPL was doing wrong accounting of kmem_alloc used in lzfs_snap.c file.

    nkhare authored
    Replaced them with kzalloc.
  2. Snapshot inode cannot be spared from zfs_inactive.

    Anand Mitra authored
Commits on Jan 6, 2011
  1. MMAP_Verification: Do not mark the pages clean in lzfs_vnop_write and…

    Prasad Joshi authored
    … lzfs_writepages.
Commits on Jan 4, 2011
  1. @nkhare

    Traversing .zfs from the NFS is not supported yet.

    nkhare authored
    It will give "Remote I/O error" error now.
Commits on Dec 29, 2010
  1. @nkhare
  2. @nkhare

    Enabled logging of LZFS debug messages by putting it to SS_USER2 subs…

    nkhare authored
    …ystem of
    
    SPL.
    In all lzfs functions in which we have written SENTRY/SEXIT, will be listed in
    SPL debug messgaes.
    
    To check.
    * Insert the SPL module with "spl_debug_mask=0xffffffff
    spl_debug_subsys=0xffffffff" paramerts.
    * Insert all other modules
    * Mount the ZFS filesystem
    * Do any file/dir operations on the mount
    * Dump SPL logs
    $ echo 1 > /proc/sys/kernel/spl/debug/dump
    * Look the dumped log using spl utilit
    $ spl /tmp/spl-log.XXXXXX
    2000000:00000001:0:1293606802.256348:0:6133:0:(lzfs_vnops.c:173:lzfs_vnop_lookup())
    Process entered
    02000000:00000001:0:1293606802.258896:0:6133:0:(lzfs_vnops.c:183:lzfs_vnop_lookup())
    Process leaving
    02000000:00000001:0:1293606802.258904:0:6133:0:(lzfs_vnops.c:330:lzfs_vnop_mkdir())
    Process entered
    02000000:00000001:0:1293606802.260482:0:6133:0:(lzfs_super.c:136:lzfs_alloc_vnode())
    Process entered
    02000000:00000001:0:1293606802.260651:0:6133:0:(lzfs_super.c:142:lzfs_alloc_vnode())
    Process leaving
    02000000:00000001:0:1293606802.261288:0:6133:0:(lzfs_vnops.c:362:lzfs_vnop_mkdir())
    Process leaving
Commits on Dec 28, 2010
  1. @nkhare
  2. @nkhare
  3. @nkhare

    Backported TSD changes from Brian's git repo.

    nkhare authored
    Removed KQ's TSD implementation.
  4. @nkhare

    Using undeclared variable in lzfs_delete_vnode function when

    nkhare authored
    kernel version < 2.6.31. fixed it.
  5. @nkhare

    truncate_pagecache function was added in 2.6.32, so our code was not

    nkhare authored
    getting compiled on F12(2.6.31). To fix it added the kernel check in
    the lzfs_super.c file for now.
Commits on Dec 27, 2010
  1. - d_invalidate the directory over which the snapshot is mounted. We d…

    Anand Mitra authored
    …o this during snapshot unmount just so that we don't have bother with it during the snapshot destroy.
Commits on Dec 26, 2010
  1. Bug fix [zfs-linux/zfs GH-73] Fixes the problem of possible call to f…

    Prasad Joshi authored
    …unction that might sleep in the atomic context
Commits on Dec 19, 2010
  1. Bug fix: [zfs-linux/zfs GH-57] Writing with buffered IO and and readi…

    Prasad Joshi authored
    …ng with mmap corrupts the data.
Commits on Dec 11, 2010
  1. mv dir ( ENOENT ) bug fixed

    Rohan Puri authored
Commits on Dec 9, 2010
  1. @nkhare
Commits on Dec 8, 2010
  1. lzfs_*.h ( headers ) moved to lzfs/include dir from spl/include/sys dir

    Rohan Puri authored
    changes made to Makefile.in to set proper  variables during
    configuration
Something went wrong with that request. Please try again.