On Mon, Feb 6, 2023 at 6:49 PM Song Liu <song@xxxxxxxxxx> wrote: > > On Mon, Feb 6, 2023 at 8:29 AM Casey Schaufler <casey@xxxxxxxxxxxxxxxx> wrote: > > > > On 2/6/2023 5:04 AM, KP Singh wrote: > > > On Fri, Jan 20, 2023 at 5:36 AM Kees Cook <keescook@xxxxxxxxxxxx> wrote: > > >> On Fri, Jan 20, 2023 at 01:08:17AM +0100, KP Singh wrote: > > >>> The indirect calls are not really needed as one knows the addresses of > > > [...] > > > > > >>> +/* > > >>> + * Define static calls and static keys for each LSM hook. > > >>> + */ > > >>> + > > >>> +#define DEFINE_LSM_STATIC_CALL(NUM, NAME, RET, ...) \ > > >>> + DEFINE_STATIC_CALL_NULL(LSM_STATIC_CALL(NAME, NUM), \ > > >>> + *((RET(*)(__VA_ARGS__))NULL)); \ > > >>> + DEFINE_STATIC_KEY_FALSE(SECURITY_HOOK_ENABLED_KEY(NAME, NUM)); > > >> Hm, another place where we would benefit from having separated logic for > > >> "is it built?" and "is it enabled by default?" and we could use > > >> DEFINE_STATIC_KEY_MAYBE(). But, since we don't, I think we need to use > > >> DEFINE_STATIC_KEY_TRUE() here or else won't all the calls be > > >> out-of-line? (i.e. the default compiled state will be NOPs?) If we're > > >> trying to optimize for having LSMs, I think we should default to inline > > >> calls. (The machine code in the commit log seems to indicate that they > > >> are out of line -- it uses jumps.) > > >> > > > I should have added it in the commit description, actually we are > > > optimizing for "hot paths are less likely to have LSM hooks enabled" > > > (eg. socket_sendmsg). > > > > How did you come to that conclusion? Where is there a correlation between > > "hot path" and "less likely to be enabled"? > > I could echo KP's reasoning here. AFAICT, the correlation is that LSMs on > hot path will give more performance overhead. In our use cases (Meta), > we are very careful with "small" performance hits. 0.25% is significant +1 to everything Song said, I am not saying that one direction is better than the other and for distros that have LSMs (like SELinux and AppArmor enabled) it's okay to have this default to static_branch_likely. On systems that will have just the BPF LSM enabled, it's the opposite that is true, i.e. one would never add a hook on a hotpath as the overheads are unacceptable, and when one does add a hook, they are willing to add the extra overhead (this is already much less compared to the indirect calls). I am okay with the default being static_branch_likely if that's what the other LSM maintainers prefer. > overhead; 1% overhead will not fly without very good reasons (Do we > have to do this? Are there any other alternatives?). If it is possible to > achieve similar security on a different hook, we will not enable the hook on > the hot path. For example, we may not enable socket_sendmsg, but try > to disallow opening such sockets instead. > > > > > > But I do see that there are LSMs that have these > > > enabled. Maybe we can put this behind a config option, possibly > > > depending on CONFIG_EXPERT? > > > > Help me, as the maintainer of one of those LSMs, understand why that would > > be a good idea. > > IIUC, this is also from performance concerns. We would like to manage > the complexity at compile time for performance benefits. > > Thanks, > Song
