]> sourceware.org Git - lvm2.git/log
lvm2.git
22 years agoo if any pattern rejects a device, and there were no accepts then reject !
Joe Thornber [Mon, 12 Nov 2001 17:06:33 +0000 (17:06 +0000)]
o  if any pattern rejects a device, and there were no accepts then reject !

22 years agoo check result of an allocation
Joe Thornber [Mon, 12 Nov 2001 16:00:52 +0000 (16:00 +0000)]
o  check result of an allocation

22 years agoiospace restructured
Alasdair Kergon [Mon, 12 Nov 2001 15:10:01 +0000 (15:10 +0000)]
iospace restructured

22 years agoo add fs.c to the Makefile
Joe Thornber [Mon, 12 Nov 2001 13:02:06 +0000 (13:02 +0000)]
o  add fs.c to the Makefile

22 years agoo add uplink from vg to cmd_context
Joe Thornber [Mon, 12 Nov 2001 12:23:10 +0000 (12:23 +0000)]
o  add uplink from vg to cmd_context

22 years agoo Plug in fs_(add|del)_lv
Joe Thornber [Mon, 12 Nov 2001 12:20:58 +0000 (12:20 +0000)]
o  Plug in fs_(add|del)_lv

22 years agoo Split struct io_space into:
Joe Thornber [Mon, 12 Nov 2001 12:16:57 +0000 (12:16 +0000)]
o  Split struct io_space into:

   struct format_handler - format methods
   struct format_instance - links instance data, methods, and cmd
   struct cmd_context - dev_dir, memory allocator, device filter

22 years agoo Compile errors
Joe Thornber [Mon, 12 Nov 2001 11:48:31 +0000 (11:48 +0000)]
o  Compile errors

22 years agoo Ready for testing
Joe Thornber [Mon, 12 Nov 2001 11:42:29 +0000 (11:42 +0000)]
o  Ready for testing

22 years ago o Fix module ref counts so that you can actually unload dm-mod
Steven Whitehouse [Sat, 10 Nov 2001 17:11:36 +0000 (17:11 +0000)]
 o Fix module ref counts so that you can actually unload dm-mod
   N.B. This means that you have to take very great care in the event that
   you want to access the dcache tree from in kernel

 o Added extra field to allow out of memory conditions to result in the
   correct error code. (This hasn't received a lot of testing...)

I've ditched the final project (which would have cleared my whole list)
since its got other complications which I don't have time to fix right
now. Still as Meatloaf says, two out of three ain't bad!

22 years agoo Added lvextend
Alasdair Kergon [Fri, 9 Nov 2001 22:01:04 +0000 (22:01 +0000)]
o Added lvextend
o Full signed arguments to lvreduce/lvextend
o Consistent lv_number/pe map use
o Populate pv->pe_allocated
o Fixes for allocation/writing of multiple LVs

22 years agosync only, not ready yet
Joe Thornber [Fri, 9 Nov 2001 08:48:22 +0000 (08:48 +0000)]
sync only, not ready yet

22 years agolvdisplay & lvreduce
Alasdair Kergon [Thu, 8 Nov 2001 16:15:58 +0000 (16:15 +0000)]
lvdisplay & lvreduce

22 years agomore todo
Heinz Mauelshagen [Thu, 8 Nov 2001 08:19:06 +0000 (08:19 +0000)]
more todo

22 years agoo Always set LVM_READ.
Alasdair Kergon [Wed, 7 Nov 2001 22:47:43 +0000 (22:47 +0000)]
o Always set LVM_READ.
o Avoid duplicate deallocation.

22 years agoOops. Forgot to check this in earlier. Changes as per previous check in
Steven Whitehouse [Wed, 7 Nov 2001 19:27:17 +0000 (19:27 +0000)]
Oops. Forgot to check this in earlier. Changes as per previous check in
comments.

22 years agoo more todo
Joe Thornber [Wed, 7 Nov 2001 17:38:25 +0000 (17:38 +0000)]
o  more todo

22 years agoo Lot's to do
Joe Thornber [Wed, 7 Nov 2001 17:25:17 +0000 (17:25 +0000)]
o  Lot's to do

22 years agoo Added lvs_in_vgs_opened
Joe Thornber [Wed, 7 Nov 2001 15:02:07 +0000 (15:02 +0000)]
o  Added lvs_in_vgs_opened

22 years agoo pool_empty was very wrong
Joe Thornber [Wed, 7 Nov 2001 14:11:20 +0000 (14:11 +0000)]
o  pool_empty was very wrong

22 years agoTidy and changes to make code smaller.
Steven Whitehouse [Wed, 7 Nov 2001 12:12:56 +0000 (12:12 +0000)]
Tidy and changes to make code smaller.

 o Created dmfs.h as a private header for the filesystem code
 o Using seq_file.[ch] written by Al Viro as a generic mechanism for /proc
   style files which have one record per line. We use a slight modification
   here, so if you are using a recent -ac kernel you'll need to replace the
   existing seq_file.[ch] with  the ones here and do a bit of editing to make
   it work. I'll submit the changes to Al Viro shortly as they are very
   small and I think make sense generally.
 o Using fail_writepage()
 o Init code for filesystem now all in dmfs-super.c
 o Some common code reduction amoung the dmfs-*.c files
 o Auto allocation of major device number (default). You can specify a
   particular major by using a module argument. If built in then you don't
   get this option at the moment but it could be added if required.
 o Hotplug support
 o General tidying
 o Updated projects.txt file
 o Patches updated to 2.4.14

22 years agoo Another pass at the activation code
Joe Thornber [Wed, 7 Nov 2001 11:51:42 +0000 (11:51 +0000)]
o  Another pass at the activation code

22 years agoAdded GPL disclaimer
Heinz Mauelshagen [Wed, 7 Nov 2001 08:50:07 +0000 (08:50 +0000)]
Added GPL disclaimer

22 years agolvcreate
Alasdair Kergon [Tue, 6 Nov 2001 19:02:26 +0000 (19:02 +0000)]
lvcreate

22 years agoo lv_extend
Joe Thornber [Tue, 6 Nov 2001 12:01:46 +0000 (12:01 +0000)]
o  lv_extend

22 years agoo Factor _allocate out for use by lv_extend
Joe Thornber [Tue, 6 Nov 2001 11:31:29 +0000 (11:31 +0000)]
o  Factor _allocate out for use by lv_extend

22 years agoo lv_reduce
Joe Thornber [Tue, 6 Nov 2001 11:19:33 +0000 (11:19 +0000)]
o  lv_reduce

o  pv_maps wasn't taking a list of acceptable pvs

22 years agoo Contiguous allocation
Joe Thornber [Tue, 6 Nov 2001 10:55:01 +0000 (10:55 +0000)]
o  Contiguous allocation

22 years agoo lv_manip.c will contain the code for lv_create, lv_extend and lv_reduce.
Joe Thornber [Tue, 6 Nov 2001 10:29:56 +0000 (10:29 +0000)]
o  lv_manip.c will contain the code for lv_create, lv_extend and lv_reduce.

22 years agoReinstate size of lv
Alasdair Kergon [Mon, 5 Nov 2001 18:07:44 +0000 (18:07 +0000)]
Reinstate size of lv

22 years agoo code for building free area lists on a pv. Compiles but not run.
Joe Thornber [Mon, 5 Nov 2001 16:41:38 +0000 (16:41 +0000)]
o  code for building free area lists on a pv.  Compiles but not run.

22 years agoo adjacent extents are now merged into a single target when activating.
Joe Thornber [Mon, 5 Nov 2001 13:37:13 +0000 (13:37 +0000)]
o  adjacent extents are now merged into a single target when activating.

22 years agolvcreate prototype
Alasdair Kergon [Mon, 5 Nov 2001 13:06:03 +0000 (13:06 +0000)]
lvcreate prototype

22 years agoo build lv name from <vg>_<lv>
Joe Thornber [Fri, 2 Nov 2001 16:45:44 +0000 (16:45 +0000)]
o build lv name from <vg>_<lv>

22 years agoLink in the activation library.
Alasdair Kergon [Fri, 2 Nov 2001 16:28:04 +0000 (16:28 +0000)]
Link in the activation library.

22 years agoo First pass at activation
Joe Thornber [Fri, 2 Nov 2001 13:45:05 +0000 (13:45 +0000)]
o  First pass at activation

22 years agoo Changes to activation interface
Alasdair Kergon [Wed, 31 Oct 2001 17:59:52 +0000 (17:59 +0000)]
o Changes to activation interface
o Add pointer lv->vg
o Some naming tweaks to improve clarity

22 years agoo LGPL list implementation
Joe Thornber [Wed, 31 Oct 2001 12:47:01 +0000 (12:47 +0000)]
o  LGPL list implementation

22 years agoo random little fixes
Joe Thornber [Tue, 30 Oct 2001 17:53:21 +0000 (17:53 +0000)]
o  random little fixes

22 years agolvchange
Alasdair Kergon [Tue, 30 Oct 2001 14:32:48 +0000 (14:32 +0000)]
lvchange

22 years agobasic lvscan
Alasdair Kergon [Mon, 29 Oct 2001 18:23:35 +0000 (18:23 +0000)]
basic lvscan

22 years agoo Add read_ahead and stripes to struct logical_volume
Joe Thornber [Mon, 29 Oct 2001 15:34:56 +0000 (15:34 +0000)]
o  Add read_ahead and stripes to struct logical_volume

22 years agoprefix & vgname in lvname
Alasdair Kergon [Mon, 29 Oct 2001 15:28:00 +0000 (15:28 +0000)]
prefix & vgname in lvname

22 years agolvremove
Alasdair Kergon [Mon, 29 Oct 2001 13:52:23 +0000 (13:52 +0000)]
lvremove

22 years ago o Added two items which ought to be done when we update to 2.4.14-pre3 or
Steven Whitehouse [Mon, 29 Oct 2001 11:06:46 +0000 (11:06 +0000)]
 o Added two items which ought to be done when we update to 2.4.14-pre3 or
   above.

22 years ago o Added a file containing a TODO list.
Steven Whitehouse [Mon, 29 Oct 2001 10:03:05 +0000 (10:03 +0000)]
 o Added a file containing a TODO list.

Please add to/edit this file as you think of new ideas or discover bugs. The
items in it are in no particular order. They are also only ideas and hence may
never get implemented depending on whether they turn out to be good ideas or
not.

22 years agoo rfilter was no longer accepting by default
Joe Thornber [Thu, 25 Oct 2001 18:12:44 +0000 (18:12 +0000)]
o  rfilter was no longer accepting by default

22 years agoo log/overwrite=1 in config file to overwrite instead of append to log
Alasdair Kergon [Thu, 25 Oct 2001 17:25:48 +0000 (17:25 +0000)]
o log/overwrite=1 in config file to overwrite instead of append to log

22 years agoo was freeing memory from the wrong pool
Joe Thornber [Thu, 25 Oct 2001 15:24:35 +0000 (15:24 +0000)]
o was freeing memory from the wrong pool

22 years agoo fopen error message
Alasdair Kergon [Thu, 25 Oct 2001 15:07:26 +0000 (15:07 +0000)]
o fopen error message
o debug options in makefile

22 years agoo pfilter stores results for all aliases.
Joe Thornber [Thu, 25 Oct 2001 14:51:51 +0000 (14:51 +0000)]
o  pfilter stores results for all aliases.

22 years agoo It's a bit of a hack, but the regex filter now makes sure a device path
Joe Thornber [Thu, 25 Oct 2001 14:41:28 +0000 (14:41 +0000)]
o  It's a bit of a hack, but the regex filter now makes sure a device path
   that passed the filter is at the front of the aliases list.

22 years agoo rfilter now checks all aliases for a match
Joe Thornber [Thu, 25 Oct 2001 14:19:39 +0000 (14:19 +0000)]
o  rfilter now checks all aliases for a match

22 years agoo use dev_name(dev) to get the name of a device, this operation is cheap
Joe Thornber [Thu, 25 Oct 2001 14:04:18 +0000 (14:04 +0000)]
o use dev_name(dev) to get the name of a device, this operation is cheap
  since it just get's the first alias.

22 years agoo Remove a couple of warnings, and one bug in ttree. Spotted by the optimiser
Joe Thornber [Thu, 25 Oct 2001 13:08:29 +0000 (13:08 +0000)]
o  Remove a couple of warnings, and one bug in ttree.  Spotted by the optimiser

22 years agoo Shuffle the keys to stop degeneracy.
Joe Thornber [Thu, 25 Oct 2001 12:38:18 +0000 (12:38 +0000)]
o  Shuffle the keys to stop degeneracy.

22 years agoo Trivial binary tree
Joe Thornber [Thu, 25 Oct 2001 11:38:19 +0000 (11:38 +0000)]
o  Trivial binary tree

22 years agoo added aliases list to struct device.
Joe Thornber [Thu, 25 Oct 2001 11:34:55 +0000 (11:34 +0000)]
o added aliases list to struct device.

22 years ago o Correction in logic for write access to tables
Steven Whitehouse [Thu, 25 Oct 2001 11:05:29 +0000 (11:05 +0000)]
 o Correction in logic for write access to tables

22 years ago o Fix typos from yesterday
Steven Whitehouse [Thu, 25 Oct 2001 10:37:05 +0000 (10:37 +0000)]
 o Fix typos from yesterday

22 years agoo Merged common code between hash_destroy and hash_wipe.
Joe Thornber [Thu, 25 Oct 2001 08:31:43 +0000 (08:31 +0000)]
o Merged common code between hash_destroy and hash_wipe.

22 years agopersistent cache fully incorporated. Goodbye to scanning /dev/cdrom :-)
Alasdair Kergon [Wed, 24 Oct 2001 17:53:50 +0000 (17:53 +0000)]
persistent cache fully incorporated.  Goodbye to scanning /dev/cdrom :-)

22 years agoo Updated 00_makefile
Joe Thornber [Wed, 24 Oct 2001 10:52:10 +0000 (10:52 +0000)]
o Updated 00_makefile

o 00_bh-async-3 has been merge with vanilla

22 years ago o Error list handling now part of fs rather than part of table.
Steven Whitehouse [Wed, 24 Oct 2001 08:26:10 +0000 (08:26 +0000)]
 o Error list handling now part of fs rather than part of table.

22 years ago o Fix bug in dmfs-error.c where it could return too many bytes under some
Steven Whitehouse [Wed, 24 Oct 2001 07:51:42 +0000 (07:51 +0000)]
 o Fix bug in dmfs-error.c where it could return too many bytes under some
   circumstances.
 o Use sscanf() in dmfs-table.c
 o Use do_generic_file_read() instead of original hand made loop in dmfs-table.c

22 years agopersistent filter & some log message changes
Alasdair Kergon [Tue, 23 Oct 2001 18:20:27 +0000 (18:20 +0000)]
persistent filter & some log message changes

22 years agoo removed old files
Joe Thornber [Tue, 23 Oct 2001 14:17:07 +0000 (14:17 +0000)]
o removed old files

22 years agoo forgot to use the path passed into _read_array.
Joe Thornber [Tue, 23 Oct 2001 13:12:05 +0000 (13:12 +0000)]
o forgot to use the path passed into _read_array.

22 years agoo test program for the new persistent filter.
Joe Thornber [Tue, 23 Oct 2001 13:11:28 +0000 (13:11 +0000)]
o test program for the new persistent filter.

22 years agodeallocations
Alasdair Kergon [Tue, 23 Oct 2001 12:33:57 +0000 (12:33 +0000)]
deallocations

22 years agoo rethink of the persistent filter
Joe Thornber [Tue, 23 Oct 2001 12:24:55 +0000 (12:24 +0000)]
o rethink of the persistent filter

22 years agofilter integration into tools
Alasdair Kergon [Tue, 23 Oct 2001 11:50:49 +0000 (11:50 +0000)]
filter integration into tools

22 years agoo forgot to retry on EINTR or EAGAIN, doh !
Joe Thornber [Tue, 23 Oct 2001 11:16:30 +0000 (11:16 +0000)]
o forgot to retry on EINTR or EAGAIN, doh !

22 years agoo forgot 'static'
Joe Thornber [Mon, 22 Oct 2001 14:40:31 +0000 (14:40 +0000)]
o forgot 'static'

22 years agoo composite filter that allows us to merge filters. Think of it as &&'ing
Joe Thornber [Mon, 22 Oct 2001 14:39:12 +0000 (14:39 +0000)]
o composite filter that allows us to merge filters.  Think of it as &&'ing
  filters in order.

eg,

f = composite_filter_create(2, regex_filter, persistent_filter);

  ownership of the filters passes, they will be destroyed when f's
  destroy method is called.

22 years agoo Filter which caches valid devices in a file. Pass in init == 1 to the
Joe Thornber [Mon, 22 Oct 2001 14:14:00 +0000 (14:14 +0000)]
o Filter which caches valid devices in a file.  Pass in init == 1 to the
  constructor if you want it to ignore the existing cache and check every
  device again (eg, vgscan, pvscan).

22 years agoreinstate a removed line
Alasdair Kergon [Mon, 22 Oct 2001 13:44:09 +0000 (13:44 +0000)]
reinstate a removed line

22 years agoo removed 00_latest since it never is.
Joe Thornber [Mon, 22 Oct 2001 10:03:05 +0000 (10:03 +0000)]
o removed 00_latest since it never is.

22 years agoo tidying
Joe Thornber [Sun, 21 Oct 2001 10:24:10 +0000 (10:24 +0000)]
o tidying

22 years agoo Filter for the dev cache that takes values from config file:
Joe Thornber [Fri, 19 Oct 2001 18:20:37 +0000 (18:20 +0000)]
o Filter for the dev cache that takes values from config file:

devices {

        # first match is final, eg.  /dev/ide/cdrom
        # get's rejected due to the first pattern

filter=["r/cdrom/",         # don't touch the music !
"a/hd[a-d][0-9]+/",
"a/ide/",
"a/sd/",
"a/md/",
"a|loop/[0-9]+|", # accept devfs style loop back
"r/loop/",        # and reject old style
"a/dasd/",
"a/dac960/",
"a/nbd/",
"a/ida/",
"a/cciss/",
"a/ubd/",
"r/.*/"] # reject all others
}

Alasdair this is ready to roll into the tools now.

22 years agoo First pass at the regex code. lib/regex/matcher takes an array of regex's
Joe Thornber [Fri, 19 Oct 2001 14:36:57 +0000 (14:36 +0000)]
o First pass at the regex code.  lib/regex/matcher takes an array of regex's
  and builds a *very* efficient engine that will tell you which regex a string
  matches with only a single pass through the string.  To be used in the config
  file when specifying devices.

o Anchor's aren't supported yet (^ and $) but that won't take long.

o Also when we get some realistic config files we may want to consider adding an
  extra level of indirection to the dfa state in order to compress the table.
  It all depends on how large typical tables get.

22 years agopvdisplay
Alasdair Kergon [Thu, 18 Oct 2001 16:55:19 +0000 (16:55 +0000)]
pvdisplay

22 years ago o Remove unused variable.
Steven Whitehouse [Thu, 18 Oct 2001 15:59:25 +0000 (15:59 +0000)]
 o Remove unused variable.

22 years agopvchange
Alasdair Kergon [Wed, 17 Oct 2001 15:29:31 +0000 (15:29 +0000)]
pvchange

22 years ago o Fix crash that Patrick reported
Steven Whitehouse [Wed, 17 Oct 2001 15:03:00 +0000 (15:03 +0000)]
 o Fix crash that Patrick reported

22 years ago o Fix a typo. This should fix devfs support.
Steven Whitehouse [Wed, 17 Oct 2001 14:34:53 +0000 (14:34 +0000)]
 o Fix a typo. This should fix devfs support.

22 years ago o Patches to go with earlier check in
Steven Whitehouse [Wed, 17 Oct 2001 13:13:25 +0000 (13:13 +0000)]
 o Patches to go with earlier check in

22 years agoOk. this is the big one.... the change to the new fs interface.
Steven Whitehouse [Wed, 17 Oct 2001 11:34:50 +0000 (11:34 +0000)]
Ok. this is the big one.... the change to the new fs interface.

Things to note:

 o Changes to the dm-*.c files have been kept as small as possible during
   the development of the new fs interface and there are a few places where
   the new code does odd things to give the original code what it wants. These
   places will gradually go away during the next few days once we are sure the
   new code is sound.
 o I've spent most of my testing time looking at the parser since thats where
   a lot of the changes are, I've not checked the actual I/O very much, but
   then that code hasn't changed at all.
 o The print operation in the target type operations is there to help in
   debugging and will go away eventually
 o There are some other printk's which will also go away once we are sure that
   things are working correctly.
 o I've tagged the old code with PRE_DMFS if you want to use that until this is
   stable.
 o There are no kernel patches for this yet (will fix after lunch... :-)
      o Makefile needs some changes
      o need to EXPORT_SYMBOL(deny_write_access); in ksyms.c

How to use the new interface ?

 mount -t dmfs dmfs /mnt/dm
 cd /mnt/dm
 mkdir fish fish/tank
 cd fish/tank
 cat ~/my.table > table
 cd ..
 ln -s tank ACTIVE

Creates a logical volume called fish and activates a table called tank, if
there is a problem doing the link, look in /mnt/dm/fish/tank/errors to see
what is wrong.

If you see any odd things happening, let me know right away as I'm sure there'll
be one or two things that slipped through my testing.

22 years ago o Update to parser
Steven Whitehouse [Wed, 17 Oct 2001 11:09:43 +0000 (11:09 +0000)]
 o Update to parser
 o Extra checks in symlink routines

22 years agopvscan
Alasdair Kergon [Tue, 16 Oct 2001 18:07:54 +0000 (18:07 +0000)]
pvscan

22 years ago o Fixed infinite loop in parser
Steven Whitehouse [Tue, 16 Oct 2001 17:09:27 +0000 (17:09 +0000)]
 o Fixed infinite loop in parser
 o Fixed error handling whilst creating volumes
 o General tidy up

22 years agovgchange
Alasdair Kergon [Tue, 16 Oct 2001 16:25:28 +0000 (16:25 +0000)]
vgchange

22 years ago o Get file size correct for table
Steven Whitehouse [Tue, 16 Oct 2001 12:17:54 +0000 (12:17 +0000)]
 o Get file size correct for table
 o Make parser look at the right object

22 years ago o Fixes to parsing code
Steven Whitehouse [Tue, 16 Oct 2001 11:56:55 +0000 (11:56 +0000)]
 o Fixes to parsing code

22 years ago o Tidy up, removing debugging printk's no longer needed
Steven Whitehouse [Tue, 16 Oct 2001 10:38:13 +0000 (10:38 +0000)]
 o Tidy up, removing debugging printk's no longer needed

22 years ago o More bug fixes
Steven Whitehouse [Mon, 15 Oct 2001 22:39:14 +0000 (22:39 +0000)]
 o More bug fixes

22 years agovgextend
Alasdair Kergon [Mon, 15 Oct 2001 22:04:27 +0000 (22:04 +0000)]
vgextend

22 years agovgcreate basic extent size validation
Alasdair Kergon [Mon, 15 Oct 2001 20:29:15 +0000 (20:29 +0000)]
vgcreate basic extent size validation

22 years agoMore vgcreate error trapping
Alasdair Kergon [Mon, 15 Oct 2001 18:39:40 +0000 (18:39 +0000)]
More vgcreate error trapping

22 years ago o More fixes
Steven Whitehouse [Mon, 15 Oct 2001 16:40:17 +0000 (16:40 +0000)]
 o More fixes

This page took 0.064165 seconds and 5 git commands to generate.