| Message ID | 20200518155727.10514-1-steve@sk2.org (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | sysctl: const-ify ngroups_max | expand |
On Mon, May 18, 2020 at 09:08:22AM -0700, Kees Cook wrote: > On Mon, May 18, 2020 at 05:57:27PM +0200, Stephen Kitt wrote: > > ngroups_max is a read-only sysctl entry, reflecting NGROUPS_MAX. Make > > it const, in the same way as cap_last_cap. > > > > Signed-off-by: Stephen Kitt <steve@sk2.org> > > Reviewed-by: Kees Cook <keescook@chromium.org> Kees, since there is quite a bit of sysctl cleanup stuff going on and I have a fs sysctl kitchen cleanup, are you alright if I carry this in a tree and send this to Andrew once done? This would hopefully avoid merge conflicts between these patches. I have to still re-spin my fs sysctl stuff, but will wait to do that once Xiaoming bases his series on linux-next. Luis
On Mon, May 18, 2020 at 05:25:09PM +0000, Luis Chamberlain wrote: > On Mon, May 18, 2020 at 09:08:22AM -0700, Kees Cook wrote: > > On Mon, May 18, 2020 at 05:57:27PM +0200, Stephen Kitt wrote: > > > ngroups_max is a read-only sysctl entry, reflecting NGROUPS_MAX. Make > > > it const, in the same way as cap_last_cap. > > > > > > Signed-off-by: Stephen Kitt <steve@sk2.org> > > > > Reviewed-by: Kees Cook <keescook@chromium.org> > > Kees, since there is quite a bit of sysctl cleanup stuff going on and I > have a fs sysctl kitchen cleanup, are you alright if I carry this in a > tree and send this to Andrew once done? This would hopefully avoid > merge conflicts between these patches. > > I have to still re-spin my fs sysctl stuff, but will wait to do that > once Xiaoming bases his series on linux-next. Yeah, totally. I don't technically have a sysctl tree (I've always just had akpm take stuff), so go for it. I'm just doing reviews. :)
On Mon, May 18, 2020 at 11:17:47AM -0700, Kees Cook wrote: > On Mon, May 18, 2020 at 05:25:09PM +0000, Luis Chamberlain wrote: > > On Mon, May 18, 2020 at 09:08:22AM -0700, Kees Cook wrote: > > > On Mon, May 18, 2020 at 05:57:27PM +0200, Stephen Kitt wrote: > > > > ngroups_max is a read-only sysctl entry, reflecting NGROUPS_MAX. Make > > > > it const, in the same way as cap_last_cap. > > > > > > > > Signed-off-by: Stephen Kitt <steve@sk2.org> > > > > > > Reviewed-by: Kees Cook <keescook@chromium.org> > > > > Kees, since there is quite a bit of sysctl cleanup stuff going on and I > > have a fs sysctl kitchen cleanup, are you alright if I carry this in a > > tree and send this to Andrew once done? This would hopefully avoid > > merge conflicts between these patches. > > > > I have to still re-spin my fs sysctl stuff, but will wait to do that > > once Xiaoming bases his series on linux-next. > > Yeah, totally. I don't technically have a sysctl tree (I've always just > had akpm take stuff), so go for it. I'm just doing reviews. :) Oh, I don't want a tree either, it was just that I can imagine these series can easily create conflcits, so I wanted to avoid that before passing them on to Andrew. Luis
On Mon, May 18, 2020 at 06:30:55PM +0000, Luis Chamberlain wrote: > On Mon, May 18, 2020 at 11:17:47AM -0700, Kees Cook wrote: > > On Mon, May 18, 2020 at 05:25:09PM +0000, Luis Chamberlain wrote: > > > On Mon, May 18, 2020 at 09:08:22AM -0700, Kees Cook wrote: > > > > On Mon, May 18, 2020 at 05:57:27PM +0200, Stephen Kitt wrote: > > > > > ngroups_max is a read-only sysctl entry, reflecting NGROUPS_MAX. Make > > > > > it const, in the same way as cap_last_cap. > > > > > > > > > > Signed-off-by: Stephen Kitt <steve@sk2.org> > > > > > > > > Reviewed-by: Kees Cook <keescook@chromium.org> > > > > > > Kees, since there is quite a bit of sysctl cleanup stuff going on and I > > > have a fs sysctl kitchen cleanup, are you alright if I carry this in a > > > tree and send this to Andrew once done? This would hopefully avoid > > > merge conflicts between these patches. > > > > > > I have to still re-spin my fs sysctl stuff, but will wait to do that > > > once Xiaoming bases his series on linux-next. > > > > Yeah, totally. I don't technically have a sysctl tree (I've always just > > had akpm take stuff), so go for it. I'm just doing reviews. :) > > Oh, I don't want a tree either, it was just that I can imagine these > series can easily create conflcits, so I wanted to avoid that before > passing them on to Andrew. Yup, that's cool. I happily defer to you on these cleanups! :)
diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 349cab382081..cc1fcba9d4d2 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -133,7 +133,7 @@ static unsigned long dirty_bytes_min = 2 * PAGE_SIZE; static int maxolduid = 65535; static int minolduid; -static int ngroups_max = NGROUPS_MAX; +static const int ngroups_max = NGROUPS_MAX; static const int cap_last_cap = CAP_LAST_CAP; /* @@ -2232,7 +2232,7 @@ static struct ctl_table kern_table[] = { #endif { .procname = "ngroups_max", - .data = &ngroups_max, + .data = (void *)&ngroups_max, .maxlen = sizeof (int), .mode = 0444, .proc_handler = proc_dointvec,
ngroups_max is a read-only sysctl entry, reflecting NGROUPS_MAX. Make it const, in the same way as cap_last_cap. Signed-off-by: Stephen Kitt <steve@sk2.org> --- This is split out from 2f4c33063ad7 ("docs: sysctl/kernel: document ngroups_max") which conflicted with f461d2dcd511 ("sysctl: avoid forward declarations"). kernel/sysctl.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) base-commit: bdecf38f228bcca73b31ada98b5b7ba1215eb9c9