Re: [patch] mlock-as-nonroot revisted

[Andrea Arcangeli]

On Thu, Jul 29, 2004 at 08:50:16PM -0400, Rik van Riel wrote:
> please read it first, especially the changes to shmem.c ;)

I read it, and it was obviously still broken.

I got CC direct from Andrew and I read only Andrew's comment and the
code, and I could stop reading the code pretty quick since hugetlbfs is
at the top and very well visibly broken.

chown on tmpfs and hugetlbfs sounds "interesting" with your approch,
I'm looking forward to the next fixed revision. I'm not against per-user
myself, but it's not like doing it for transient memory or transient
objects associated with the task itself. Furthemore I'm not convinced
rlimits should be used for such persistent things that have nothing to
do with running tasks but ok, I can live with it if it works.

I tend to believe if something we should use fs_quota in tmpfs and
hugetlbfs for such things. That is the place to catch chown etc..
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/