user/sven/linux.git/fs/ceph/file.c, branch v3.0.76

ceph: fix sync and dio writes across stripe boundaries

2011-06-13T23:26:22Z

We were iterating across stripe boundaries properly, but not moving the write buffer pointer forward. This caused us to rewrite the same data after the break. Fix by adjusting the data pointer forward, and recalculating the io and buffer alignment after the break. Signed-off-by: Sage Weil

ceph: fix page alignment corrections

2011-06-13T23:26:10Z

dd if=/dev/urandom of=/mnt/fs_depot/dd10 bs=500 seek=8388 count=1 dd if=/mnt/fs_depot/dd10 of=/root/dd10out bs=500 skip=8388 count=1 Reported-by: Henry C Chang Signed-off-by: Sage Weil

ceph: fix ENOENT logic in striped_read

2011-06-08T04:34:16Z

Getting ENOENT is equivalent to reading 0 bytes. Make that correction before setting up the hit_stripe and was_short flags. Fixes the following case: dd if=/dev/zero of=/mnt/fs_depot/dd3 bs=1 seek=1048576 count=0 dd if=/mnt/fs_depot/dd3 of=/root/ddout1 skip=8 bs=500 count=2 iflag=direct Reported-by: Henry C Chang Signed-off-by: Sage Weil

ceph: fix short sync reads from the OSD

2011-06-08T04:34:14Z

If we get a short read from the OSD because the object is small, we need to zero the remainder of the buffer. For O_DIRECT reads, the attempted range is not trimmed to i_size by the VFS, so we were actually looping indefinitely. Fix by trimming by i_size, and the unconditionally zeroing the trailing range. Reported-by: Jeff Wu Signed-off-by: Sage Weil

ceph: use ihold when we already have an inode ref

2011-06-08T04:34:11Z

We should use ihold whenever we already have a stable inode ref, even when we aren't holding i_lock. This avoids adding new and unnecessary locking dependencies. Signed-off-by: Sage Weil

ceph: do not call __mark_dirty_inode under i_lock

2011-05-04T19:56:45Z

The __mark_dirty_inode helper now takes i_lock as of 250df6ed. Fix the one ceph callers that held i_lock (__ceph_mark_dirty_caps) to return the flags value so that the callers can do it outside of i_lock. Signed-off-by: Sage Weil

ceph: add request to the tail of unsafe write list

2011-03-21T19:24:25Z

In sync_write_wait(), we assume that the newest request is at the tail of unsafe write list. We should maintain the semantics here. Signed-off-by: Henry C Chang Signed-off-by: Sage Weil

ceph: remove request from unsafe list if it is canceled/timed out

2011-03-21T19:24:24Z

This fixes the list corruption warning like this: ------------[ cut here ]------------ WARNING: at lib/list_debug.c:30 __list_add+0x68/0x81() Hardware name: X8DTU list_add corruption. prev->next should be next (ffff880618931250), but was (null). (prev=ffff880c188b9130). Modules linked in: nfsd lockd nfs_acl auth_rpcgss exportfs ceph libceph libcrc32c sunrpc ipv6 fuse igb i2c_i801 ioatdma i2c_core iTCO_wdt iTCO_vendor_support joydev dca serio_raw usb_storage [last unloaded: scsi_wait_scan] Pid: 10977, comm: smbd Tainted: G W 2.6.32.23-170.Elaster.xendom0.fc12.x86_64 #1 Call Trace: [] warn_slowpath_common+0x7c/0x94 [] warn_slowpath_fmt+0x41/0x43 [] __list_add+0x68/0x81 [] ceph_aio_write+0x614/0x8a2 [ceph] [] do_sync_write+0xe8/0x125 [] ? autoremove_wake_function+0x0/0x39 [] ? selinux_file_permission+0x5c/0xb3 [] ? security_file_permission+0x16/0x18 [] vfs_write+0xae/0x10b [] sys_pwrite64+0x5a/0x76 [] system_call_fastpath+0x16/0x1b ---[ end trace 08573eb9f07ff6f4 ]--- Signed-off-by: Henry C Chang Signed-off-by: Sage Weil

ceph: mark user pages dirty on direct-io reads

2010-12-17T17:54:40Z

For read operation, we have to set the argument _write_ of get_user_pages to 1 since we will write data to pages. Also, we need to SetPageDirty before releasing these pages. Signed-off-by: Henry C Chang Signed-off-by: Sage Weil

ceph: fix direct-io on non-page-aligned buffers

2010-12-16T04:46:16Z

The user buffer may be 512-byte aligned, not page-aligned. We were assuming the buffer was page-aligned and only accounting for non-page-aligned io offsets. Signed-off-by: Henry C Chang Signed-off-by: Sage Weil