From: Mikko Ylinen <mikko.ylinen@linux.intel.com>
To: Haitao Huang <haitao.huang@linux.intel.com>
Cc: "Mehta, Sohil" <sohil.mehta@intel.com>,
"jarkko@kernel.org" <jarkko@kernel.org>,
"x86@kernel.org" <x86@kernel.org>,
"dave.hansen@linux.intel.com" <dave.hansen@linux.intel.com>,
"cgroups@vger.kernel.org" <cgroups@vger.kernel.org>,
"hpa@zytor.com" <hpa@zytor.com>,
"mingo@redhat.com" <mingo@redhat.com>,
"tj@kernel.org" <tj@kernel.org>,
"mkoutny@suse.com" <mkoutny@suse.com>,
"tglx@linutronix.de" <tglx@linutronix.de>,
"linux-sgx@vger.kernel.org" <linux-sgx@vger.kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"bp@alien8.de" <bp@alien8.de>, "Huang, Kai" <kai.huang@intel.com>,
Dave Hansen <dave.hansen@intel.com>,
"seanjc@google.com" <seanjc@google.com>,
"Zhang, Bo" <zhanb@microsoft.com>,
"kristen@linux.intel.com" <kristen@linux.intel.com>,
"anakrish@microsoft.com" <anakrish@microsoft.com>,
"sean.j.christopherson@intel.com"
<sean.j.christopherson@intel.com>,
"Li, Zhiquan1" <zhiquan1.li@intel.com>,
"yangjie@microsoft.com" <yangjie@microsoft.com>
Subject: Re: [PATCH v6 09/12] x86/sgx: Restructure top-level EPC reclaim function
Date: Fri, 5 Jan 2024 16:43:23 +0200 [thread overview]
Message-ID: <ZZgVi7rHAAzyzbz5@himmelriiki> (raw)
In-Reply-To: <op.2g1d81fqwjvjmi@hhuan26-mobl.amr.corp.intel.com>
On Thu, Jan 04, 2024 at 01:11:15PM -0600, Haitao Huang wrote:
> Hi Dave,
>
> On Wed, 03 Jan 2024 10:37:35 -0600, Dave Hansen <dave.hansen@intel.com>
> wrote:
>
> > On 12/18/23 13:24, Haitao Huang wrote:> @Dave and @Michal, Your
> > thoughts? Or could you confirm we should not
> > > do reclaim per cgroup at all?
> > What's the benefit of doing reclaim per cgroup? Is that worth the extra
> > complexity?
> >
>
> Without reclaiming per cgroup, then we have to always set the limit to
> enclave's peak usage. This may not be efficient utilization as in many cases
> each enclave can perform fine with EPC limit set less than peak. Basically
> each group can not give up some pages for greater good without dying :-)
+1. this is exactly my thinking too. The per cgroup reclaiming is
important for the containers use case we are working on. I also think
it makes the limit more meaningful: the per-container pool of EPC pages
to use (which is independent of the enclave size).
>
> Also with enclaves enabled with EDMM, the peak usage is not static so hard
> to determine upfront. Hence it might be an operation/deployment
> inconvenience.
>
> In case of over-committing (sum of limits > total capacity), one cgroup at
> peak usage may require swapping pages out in a different cgroup if system is
> overloaded at that time.
>
> > The key question here is whether we want the SGX VM to be complex and
> > more like the real VM or simple when a cgroup hits its limit. Right?
> >
>
> Although it's fair to say the majority of complexity of this series is in
> support for reclaiming per cgroup, I think it's manageable and much less
> than real VM after we removed the enclave killing parts: the only extra
> effort is to track pages in separate list and reclaim them in separately as
> opposed to track in on global list and reclaim together. The main reclaiming
> loop code is still pretty much the same as before.
>
>
> > If stopping at patch 5 and having less code is even remotely an option,
> > why not do _that_?
> >
> I hope I described limitations clear enough above.
> If those are OK with users and also make it acceptable for merge quickly,
You explained the gaps very well already. I don't think the simple
version without per-cgroup reclaiming is enough for the container case.
Mikko
next prev parent reply other threads:[~2024-01-05 14:44 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-30 18:20 [PATCH v6 00/12] Add Cgroup support for SGX EPC memory Haitao Huang
2023-10-30 18:20 ` [PATCH v6 01/12] cgroup/misc: Add per resource callbacks for CSS events Haitao Huang
2023-11-15 20:25 ` Jarkko Sakkinen
2024-01-09 3:37 ` Haitao Huang
2024-01-10 19:55 ` Jarkko Sakkinen
2024-01-05 9:45 ` Michal Koutný
2024-01-06 1:42 ` Haitao Huang
2023-10-30 18:20 ` [PATCH v6 02/12] cgroup/misc: Export APIs for SGX driver Haitao Huang
2023-10-30 18:20 ` [PATCH v6 03/12] cgroup/misc: Add SGX EPC resource type Haitao Huang
2023-10-30 18:20 ` [PATCH v6 04/12] x86/sgx: Implement basic EPC misc cgroup functionality Haitao Huang
2023-11-06 12:09 ` Huang, Kai
2023-11-06 18:59 ` Haitao Huang
2023-11-06 22:18 ` Huang, Kai
2023-11-07 1:16 ` Haitao Huang
2023-11-07 2:08 ` Haitao Huang
2023-11-07 19:07 ` Huang, Kai
2023-11-20 3:16 ` Huang, Kai
2023-11-26 16:01 ` Haitao Huang
2023-11-26 16:32 ` Haitao Huang
2023-11-06 22:23 ` Huang, Kai
2023-11-15 20:48 ` Jarkko Sakkinen
2023-10-30 18:20 ` [PATCH v6 05/12] x86/sgx: Add sgx_epc_lru_list to encapsulate LRU list Haitao Huang
2023-10-30 18:20 ` [PATCH v6 06/12] x86/sgx: Use sgx_epc_lru_list for existing active page list Haitao Huang
2023-10-30 18:20 ` [PATCH v6 07/12] x86/sgx: Introduce EPC page states Haitao Huang
2023-11-15 20:53 ` Jarkko Sakkinen
2024-01-05 17:57 ` Dave Hansen
2024-01-06 1:45 ` Haitao Huang
2023-10-30 18:20 ` [PATCH v6 08/12] x86/sgx: Use a list to track to-be-reclaimed pages Haitao Huang
2023-11-15 20:59 ` Jarkko Sakkinen
2023-10-30 18:20 ` [PATCH v6 09/12] x86/sgx: Restructure top-level EPC reclaim function Haitao Huang
2023-11-20 3:45 ` Huang, Kai
2023-11-26 16:27 ` Haitao Huang
2023-11-27 9:57 ` Huang, Kai
2023-12-12 4:04 ` Haitao Huang
2023-12-13 11:17 ` Huang, Kai
2023-12-15 19:49 ` Haitao Huang
2023-12-18 1:44 ` Huang, Kai
2023-12-18 17:32 ` Mikko Ylinen
2023-12-18 21:24 ` Haitao Huang
2024-01-03 16:37 ` Dave Hansen
2024-01-04 19:11 ` Haitao Huang
2024-01-04 19:19 ` Jarkko Sakkinen
2024-01-04 19:27 ` Dave Hansen
2024-01-04 21:01 ` Haitao Huang
2024-01-05 14:43 ` Mikko Ylinen [this message]
2024-01-04 12:38 ` Michal Koutný
2024-01-04 19:20 ` Haitao Huang
2024-01-12 17:07 ` Haitao Huang
2024-01-13 21:04 ` Jarkko Sakkinen
2024-01-13 21:08 ` Jarkko Sakkinen
2023-10-30 18:20 ` [PATCH v6 10/12] x86/sgx: Implement EPC reclamation for cgroup Haitao Huang
2023-11-06 15:58 ` [PATCH] x86/sgx: Charge proper mem_cgroup for usage due to EPC reclamation by cgroups Haitao Huang
2023-11-06 16:10 ` [PATCH v6 10/12] x86/sgx: Implement EPC reclamation for cgroup Haitao Huang
2023-10-30 18:20 ` [PATCH v6 11/12] Docs/x86/sgx: Add description for cgroup support Haitao Huang
2023-10-30 18:20 ` [PATCH v6 12/12] selftests/sgx: Add scripts for EPC cgroup testing Haitao Huang
2023-11-15 21:00 ` Jarkko Sakkinen
2023-11-15 21:22 ` Haitao Huang
2023-11-06 3:26 ` [PATCH v6 00/12] Add Cgroup support for SGX EPC memory Jarkko Sakkinen
2023-11-06 15:48 ` Haitao Huang
2023-11-08 1:00 ` Haitao Huang
2024-01-05 18:29 ` Dave Hansen
2024-01-05 20:13 ` Haitao Huang
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ZZgVi7rHAAzyzbz5@himmelriiki \
--to=mikko.ylinen@linux.intel.com \
--cc=anakrish@microsoft.com \
--cc=bp@alien8.de \
--cc=cgroups@vger.kernel.org \
--cc=dave.hansen@intel.com \
--cc=dave.hansen@linux.intel.com \
--cc=haitao.huang@linux.intel.com \
--cc=hpa@zytor.com \
--cc=jarkko@kernel.org \
--cc=kai.huang@intel.com \
--cc=kristen@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-sgx@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=mkoutny@suse.com \
--cc=sean.j.christopherson@intel.com \
--cc=seanjc@google.com \
--cc=sohil.mehta@intel.com \
--cc=tglx@linutronix.de \
--cc=tj@kernel.org \
--cc=x86@kernel.org \
--cc=yangjie@microsoft.com \
--cc=zhanb@microsoft.com \
--cc=zhiquan1.li@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).