git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [PATCH v3 0/8] Fix GIT_CEILING_DIRECTORIES that contain symlinks

From
Michael Haggerty <mhagger@alum.mit.edu>
Date
Oct 22, 2012, 08:26 UTC
Message-ID
<5085033E.2070607@alum.mit.edu>
In-Reply-To
<7v7gqkgvxe.fsf@alter.siamese.dyndns.org>
On 10/21/2012 08:51 AM, Junio C Hamano wrote:
Show 15 quoted lines
> Michael Haggerty <mhagger@alum.mit.edu> writes:
> 
>> This patch series has the side effect that all of the directories
>> listed in GIT_CEILING_DIRECTORIES are accessed *unconditionally* to
>> resolve any symlinks that are present in their paths.  It is
>> admittedly odd that a feature intended to avoid accessing expensive
>> directories would now *intentionally* access directories near the
>> expensive ones.  In the above scenario this shouldn't be a problem,
>> because /home would be the directory listed in
>> GIT_CEILING_DIRECTORIES, and accessing /home itself shouldn't be
>> expensive.
> 
> Interesting observation.  In the last sentence, "accessing /home"
> does not exactly mean accessing /home, but accessing / to learn
> about "home" in it, no?

This is the extra overhead on my system for using GIT_CEILING_DIRECTORIES=/home:

stat("/home", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0 getcwd("/home/mhagger", 1024) = 14 chdir("/home") = 0 getcwd("/home", 4096) = 6 lstat("/home", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0 chdir("/home/mhagger") = 0

If I use GIT_CEILING_DIRECTORIES=/dev/shm, which is a symlink to /run/shm on my system, the overhead is comparable:

stat("/dev/shm", {st_mode=S_IFDIR|S_ISVTX|0777, st_size=200, ...}) = 0 getcwd("/home/mhagger", 1024) = 14 chdir("/dev/shm") = 0 getcwd("/run/shm", 4096) = 9 lstat("/run/shm", {st_mode=S_IFDIR|S_ISVTX|0777, st_size=200, ...}) = 0 chdir("/home/mhagger") = 0

Michael
-- 
Michael Haggerty
mhagger@alum.mit.edu
http://softwareswirl.blogspot.com/
Previous: Junio C HamanoNext: David Aguilar
Message 12 of 26 in “Fix GIT_CEILING_DIRECTORIES that contain symlinks”
  1. 0/8 Fix GIT_CEILING_DIRECTORIES that contain symlinksMichael Haggerty, Oct 21, 2012
  2. 1/8 Introduce new static function real_path_internal()Michael Haggerty, Oct 21, 2012
  3. 2/8 real_path_internal(): add comment explaining use of cwdMichael Haggerty, Oct 21, 2012
  4. 3/8 Introduce new function real_path_if_valid()Michael Haggerty, Oct 21, 2012
  5. 4/8 longest_ancestor_length(): use string_list_split()Michael Haggerty, Oct 21, 2012
  6. 5/8 longest_ancestor_length(): take a string_list argument for prefixesMichael Haggerty, Oct 21, 2012
  7. 6/8 longest_ancestor_length(): require prefix list entries to be normalizedMichael Haggerty, Oct 21, 2012
  8. Johannes SixtOct 22, 2012
  9. 7/8 normalize_ceiling_entry(): resolve symlinksMichael Haggerty, Oct 21, 2012
  10. 8/8 string_list_longest_prefix(): remove functionMichael Haggerty, Oct 21, 2012
  11. Junio C HamanoOct 21, 2012
  12. Michael HaggertyOct 22, 2012
  13. David AguilarOct 29, 2012
  14. Junio C HamanoOct 29, 2012
  15. Michael HaggertyOct 29, 2012
  16. Junio C HamanoNov 12, 2012
  17. David AguilarNov 13, 2012
  18. Michael HaggertyNov 15, 2012
  19. Anders KaseorgFeb 20, 2013
  20. Junio C HamanoFeb 20, 2013
  21. [RFC] Provide a mechanism to turn off symlink resolution in ceiling pathsMichael Haggerty, Feb 20, 2013
  22. Junio C HamanoFeb 20, 2013
  23. Junio C HamanoFeb 21, 2013
  24. Michael HaggertyFeb 22, 2013
  25. Anders KaseorgFeb 20, 2013
  26. Lars DamerowOct 29, 2012

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.