Patchwork [0/6] replace genext2fs with populate-extfs.sh

login
register
mail settings
Submitter Robert Yang
Date Aug. 22, 2013, 1:13 p.m.
Message ID <cover.1377175027.git.liezhi.yang@windriver.com>
Download mbox
Permalink /patch/56247/
State New
Headers show

Pull-request

git://git.pokylinux.org/poky-contrib robert/extX

Comments

Robert Yang - Aug. 22, 2013, 1:13 p.m.
* The benefits:
  - Really support ext4
  - Support the sparse file (we lost the sparse file in the image in the
    past, the sparse file became into the common file)
  - Fix the error reported by fsck: (ext2/ext3)
     "Inode 1025, i_size is 16384, should be 17408."
  - Have a uniform code for ext2/3/4 generation

* Impact
  - Build time:
    a) If we build fresh core-image-sato, there is nearly no impact.

    b) If we do the image generation, which means:
       $ bitbake core-image-sato
       $ bitbake core-image-sato -ccleansstate
       $ bitbake core-image-sato
   About 40 extra seconds are needed, here is my test result:
   Before the patches: 4m53s
   After the patches: 5m35s
   This is because the genext2fs is much faster than the 
   populate-extfs.sh, we will replace this script by the mke2fs when it
   supports create the filesystem from a initial directory.

  - Disk space (take core-image-sato as an example)
    a) The image file size is the same as before (519M)

    b) The disk usage is a little different: (du -sh)
              before       now 
       ext2:  356M         379M
       ext3:  372M         395M
       ext4:  372M         380M

   I have done some simple runtime testing on core-image-sato based on
   ext2/3/4, they worked well.

* The "fsck.extX -fn <image>.extX" shows it is OK.

// Robert

The following changes since commit b2ff1add530b1fec2fb7f385227a03db47015c37:

  poky.conf: Don't force the addition of extra DISTRO_FEATURES (2013-08-20 22:58:04 +0100)

are available in the git repository at:

  git://git.pokylinux.org/poky-contrib robert/extX
  http://git.pokylinux.org/cgit.cgi/poky-contrib/log/?h=robert/extX

Robert Yang (6):
  e2fsprogs: the max length of debugfs argument is too short
  e2fsprogs: let debugfs do sparse copy
  e2fsprogs: only update the icache for ext2_inode
  e2fsprogs: properly set up extent header in do_write
  e2fsprogs: add populate-extfs.sh
  image_types.bbclass: replace genext2fs with populate-extfs.sh

 meta/classes/image_types.bbclass                   |   46 +++---
 .../e2fsprogs-1.42.8/debugfs-extent-header.patch   |   47 +++++++
 .../e2fsprogs-1.42.8/debugfs-sparse-copy.patch     |  147 ++++++++++++++++++++
 .../e2fsprogs-1.42.8/debugfs-too-short.patch       |   41 ++++++
 .../e2fsprogs/e2fsprogs-1.42.8/fix-icache.patch    |   69 +++++++++
 .../e2fsprogs/e2fsprogs-1.42.8/populate-extfs.sh   |   96 +++++++++++++
 .../recipes-devtools/e2fsprogs/e2fsprogs_1.42.8.bb |    6 +
 7 files changed, 424 insertions(+), 28 deletions(-)
 create mode 100644 meta/recipes-devtools/e2fsprogs/e2fsprogs-1.42.8/debugfs-extent-header.patch
 create mode 100644 meta/recipes-devtools/e2fsprogs/e2fsprogs-1.42.8/debugfs-sparse-copy.patch
 create mode 100644 meta/recipes-devtools/e2fsprogs/e2fsprogs-1.42.8/debugfs-too-short.patch
 create mode 100644 meta/recipes-devtools/e2fsprogs/e2fsprogs-1.42.8/fix-icache.patch
 create mode 100644 meta/recipes-devtools/e2fsprogs/e2fsprogs-1.42.8/populate-extfs.sh