13 Feb, 2007

1 commit

• 754661f14   [PATCH] mark struct inode_operations const 1 ... Browse Code »

  Many struct inode_operations in the kernel can be "const". Marking them const
  moves these to the .rodata section, which avoids false sharing with potential
  dirty data. In addition it'll catch accidental writes at compile time to
  these shared resources.

  Signed-off-by: Arjan van de Ven
  Signed-off-by: Andrew Morton
  Signed-off-by: Linus Torvalds

  Arjan van de Ven

  2007-02-13 01:48:46 +0800  

29 Jun, 2006

1 commit

• f5e54d6e5   [PATCH] mark address_space_operations const ... Browse Code »

  Same as with already do with the file operations: keep them in .rodata and
  prevents people from doing runtime patching.

  Signed-off-by: Christoph Hellwig
  Cc: Steven French
  Signed-off-by: Andrew Morton
  Signed-off-by: Linus Torvalds

  Christoph Hellwig

  2006-06-29 05:59:04 +0800  

29 Mar, 2006

1 commit

• 4b6f5d20b   [PATCH] Make most file operations structs in fs/ const ... Browse Code »

  This is a conversion to make the various file_operations structs in fs/
  const. Basically a regexp job, with a few manual fixups

  The goal is both to increase correctness (harder to accidentally write to
  shared datastructures) and reducing the false sharing of cachelines with
  things that get dirty in .data (while .rodata is nicely read only and thus
  cache clean)

  Signed-off-by: Arjan van de Ven
  Signed-off-by: Andrew Morton
  Signed-off-by: Linus Torvalds

  Arjan van de Ven

  2006-03-29 01:16:06 +0800  

10 Sep, 2005

1 commit

• fac92becd   [PATCH] bfs: fix endianness, signedness; add trivial bugfix ... Browse Code »

  * Makes BFS code endianness-clean.

  * Fixes some signedness warnings.

  * Fixes a problem in fs/bfs/inode.c:164 where inodes not synced to disk
  don't get fully marked as clean. Here's how to reproduce it:

  # mount -o loop -t bfs /bfs.img /mnt
  # df -i /mnt
  Filesystem Inodes IUsed IFree IUse% Mounted on
  /bfs.img 48 1 47 3% /mnt
  # df -k /mnt
  Filesystem 1K-blocks Used Available Use% Mounted on
  /bfs.img 512 5 508 1% /mnt
  # cp 60k-archive.zip /mnt/mt.zip
  # df -k /mnt
  Filesystem 1K-blocks Used Available Use% Mounted on
  /bfs.img 512 65 447 13% /mnt
  # df -i /mnt
  Filesystem Inodes IUsed IFree IUse% Mounted on
  /bfs.img 48 2 46 5% /mnt
  # rm /mnt/mt.zip
  # echo $?
  0

  [If the unlink happens before the buffers flush, the following happens:]

  # df -i /mnt
  Filesystem Inodes IUsed IFree IUse% Mounted on
  /bfs.img 48 2 46 5% /mnt
  # df -k /mnt
  Filesystem 1K-blocks Used Available Use% Mounted on
  /bfs.img 512 65 447 13% /mnt

  fs/bfs/bfs.h | 1

  Signed-off-by: Andrew Stribblehill
  Cc:
  Signed-off-by: Andrew Morton
  Signed-off-by: Linus Torvalds

  Andrew Stribblehill

  2005-09-10 04:57:32 +0800  

17 Apr, 2005

1 commit

• 1da177e4c   Linux-2.6.12-rc2 ... Browse Code »

  Initial git repository build. I'm not bothering with the full history,
  even though we have it. We can create a separate "historical" git
  archive of that later if we want to, and in the meantime it's about
  3.2GB when imported into git - space that would just make the early
  git days unnecessarily complicated, when we don't have a lot of good
  infrastructure for it.

  Let it rip!

  Linus Torvalds

  2005-04-17 06:20:36 +0800