[lkml]   [2003]   [Jan]   [6]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
Subject[BK SUMMARY] remove 32 group limit (re-send)

I've sent this a bunch of times. I don't know if it is just getting dropped
or if there is some reason you don't want it in. Could you please either
explain why it will never go in or what needs to be fixed for it to go in?

This patch removes the hard NGROUPS limit. It has been in use in a similar
form on our systems for some time. I've sent it several times, and it has
evolved a lot from the original form. I've had no complaints from anyone
about this version of the patch.

There is a small change needed for glibc, and I will send that patch to the
glibc people when this gets pulled.

Unlike some prior versions of this patch, I have changed qsort() to a simple
non-recursive sort. Several people indicated that this change would solve
their objections. I have also changed the behavior of sys_setgroups to use an
inline array of gid_t for small sets of groups. Larger sets of groups use
kmalloc, and very large sets use vmalloc. This has turned up a big speed
increase for the (admittedly stupid) test of setgroups() in a loop with random
data and sets of 1-32 items, repeated hundreds of thousands of times, as well
as the tests of setgroups with random set sizes between 1 and 10000.

Lastly, this does not fixup all the architectures. I have or will soon have
other patchsets for that, which need to be reviewed by arch maintainers.


Please do a

bk pull

This will update the following files:

fs/nfsd/auth.c | 11 +--
fs/proc/array.c | 2
include/asm-i386/param.h | 4 -
include/linux/init_task.h | 1
include/linux/kernel.h | 3
include/linux/limits.h | 3
include/linux/sched.h | 6 +
include/linux/sunrpc/svcauth.h | 3
kernel/exit.c | 7 +
kernel/fork.c | 7 +
kernel/sys.c | 145 ++++++++++++++++++++++++++++++++++-------
kernel/uid16.c | 64 +++++++++++++-----
lib/Makefile | 4 -
lib/bsearch.c | 49 +++++++++++++
net/sunrpc/svcauth_unix.c | 4 -
15 files changed, 257 insertions(+), 56 deletions(-)

through these ChangeSets (diffs in separate email):

<> (02/12/16 1.888)
Remove the limit of 32 groups. We now have a per-task, dynamic array of
groups, which is kept sorted and refcounted. If the task has less than 32
groups, we behave like older kernels and use an inline array.

This ChangeSet incorporates all the core functionality. but does not fixup
all the incorrect architecture usages of groups. That will follow.

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
More majordomo info at
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:32    [W:0.035 / U:2.856 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site