Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Mon, 26 Dec 2022 20:43:58 +0000
From:   Jarkko Sakkinen <jarkko@kernel.org>
To:     Sean Christopherson <seanjc@google.com>
Cc:     Dave Hansen <dave.hansen@intel.com>,
        Kristen Carlson Accardi <kristen@linux.intel.com>,
        dave.hansen@linux.intel.com, tj@kernel.org,
        linux-kernel@vger.kernel.org, linux-sgx@vger.kernel.org,
        cgroups@vger.kernel.org, Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
        x86@kernel.org, "H. Peter Anvin" <hpa@zytor.com>,
        zhiquan1.li@intel.com
Subject: Re: [PATCH v2 14/18] x86/sgx: Add EPC OOM path to forcefully reclaim
 EPC
Message-ID: <Y6oHjrLCbDzWh7nE@kernel.org>
References: <20221202183655.3767674-1-kristen@linux.intel.com>
 <20221202183655.3767674-15-kristen@linux.intel.com>
 <Y5IBCOuF8X7jEK3+@kernel.org>
 <cb5abce531c1b14118de419ba68c2a501b016873.camel@linux.intel.com>
 <e5aff02b-713c-ccd8-7211-d07ff6d7adb2@intel.com>
 <Y5duYxIHtSpK1qkj@google.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <Y5duYxIHtSpK1qkj@google.com>
Precedence: bulk

On Mon, Dec 12, 2022 at 06:09:39PM +0000, Sean Christopherson wrote:
> On Fri, Dec 09, 2022, Dave Hansen wrote:
> > On 12/9/22 08:05, Kristen Carlson Accardi wrote:
> > > Aside from that though, I don't think that killing enclaves makes sense
> > > outside the context of cgroup limits. 
> > 
> > I think it makes a lot of sense in theory.  Whatever situation we get
> > into with a cgroup's EPC we can also get into with the whole system's EPC.
> > 
> > *But*, it's orders of magnitude harder to hit on the whole system.
> 
> ...
> 
> > If someone wants to extend this OOM support to system-wide EPC later, then go
> > ahead.  But, I don't think it makes a lot of sense to invert this series for
> > it.
> 
> +1 from the peanut gallery.  With VMM EPC oversubscription suport, no sane VMM
> will oversubscribe VEPC pages.  And for VA pages, supporting swap of VA pages is
> likely a more userspace-friendly approach if system-wide EPC OOM is a concern.

When swapping VA pages the topology of the VA page cache for swapped VA
pages is the main question. It is compromise between how long swap-in and
swap-out can take, and how generic solution it be, meaning how deep
hierarchies you want to build, or is just a flat list of parent VA pages
"good enough".

Also, there's the question whether it should be a global cache, per cgroup
and so forth.

Implementing any solution is not overly complicated. Locking in these
options is what puzzles me.

BR, Jarkko