From: "Haitao Huang" <haitao.huang@linux.intel.com>
To: "hpa@zytor.com" <hpa@zytor.com>,
"linux-sgx@vger.kernel.org" <linux-sgx@vger.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>,
"bp@alien8.de" <bp@alien8.de>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"jarkko@kernel.org" <jarkko@kernel.org>,
"tglx@linutronix.de" <tglx@linutronix.de>,
"Mehta, Sohil" <sohil.mehta@intel.com>,
"tj@kernel.org" <tj@kernel.org>,
"mingo@redhat.com" <mingo@redhat.com>,
"Huang, Kai" <kai.huang@intel.com>
Cc: "kristen@linux.intel.com" <kristen@linux.intel.com>,
"yangjie@microsoft.com" <yangjie@microsoft.com>,
"Li, Zhiquan1" <zhiquan1.li@intel.com>, "Christopherson,,
Sean" <seanjc@google.com>,
"mikko.ylinen@linux.intel.com" <mikko.ylinen@linux.intel.com>,
"Zhang, Bo" <zhanb@microsoft.com>,
"anakrish@microsoft.com" <anakrish@microsoft.com>
Subject: Re: [PATCH v5 12/18] x86/sgx: Add EPC OOM path to forcefully reclaim EPC
Date: Mon, 09 Oct 2023 20:04:54 -0500 [thread overview]
Message-ID: <op.2ckqmgs9wjvjmi@hhuan26-mobl.amr.corp.intel.com> (raw)
In-Reply-To: <1b265d0c9dfe17de2782962ed26a99cc9d330138.camel@intel.com>
On Mon, 09 Oct 2023 18:45:06 -0500, Huang, Kai <kai.huang@intel.com> wrote:
> On Fri, 2023-09-22 at 20:06 -0700, Haitao Huang wrote:
>> From: Sean Christopherson <sean.j.christopherson@intel.com>
>>
>> Introduce the OOM path for killing an enclave with a reclaimer that is
>> no
>> longer able to reclaim enough EPC pages. Find a victim enclave, which
>> will be an enclave with only "unreclaimable" EPC pages left in the
>> cgroup LRU lists. Once a victim is identified, mark the enclave as OOM
>> and zap the enclave's entire page range, and drain all mm references in
>> encl->mm_list. Block allocating any EPC pages in #PF handler, or
>> reloading any pages in all paths, or creating any new mappings.
>>
>> The OOM killing path may race with the reclaimers: in some cases, the
>> victim enclave is in the process of reclaiming the last EPC pages when
>> OOM happens, that is, all pages other than SECS and VA pages are in
>> RECLAIMING_IN_PROGRESS state. The reclaiming process requires access to
>> the enclave backing, VA pages as well as SECS. So the OOM killer does
>> not directly release those enclave resources, instead, it lets all
>> reclaiming in progress to finish, and relies (as currently done) on
>> kref_put on encl->refcount to trigger sgx_encl_release() to do the
>> final cleanup.
>>
>> Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
>> Co-developed-by: Kristen Carlson Accardi <kristen@linux.intel.com>
>> Signed-off-by: Kristen Carlson Accardi <kristen@linux.intel.com>
>> Co-developed-by: Haitao Huang <haitao.huang@linux.intel.com>
>> Signed-off-by: Haitao Huang <haitao.huang@linux.intel.com>
>> Cc: Sean Christopherson <seanjc@google.com>
>> ---
>> V5:
>> - Rename SGX_ENCL_OOM to SGX_ENCL_NO_MEMORY
>>
>> V4:
>> - Updates for patch reordering and typo fixes.
>>
>> V3:
>> - Rebased to use the new VMA_ITERATOR to zap VMAs.
>> - Fixed the racing cases by blocking new page allocation/mapping and
>> reloading when enclave is marked for OOM. And do not release any enclave
>> resources other than draining mm_list entries, and let pages in
>> RECLAIMING_IN_PROGRESS to be reaped by reclaimers.
>> - Due to above changes, also removed the no-longer needed encl->lock in
>> the OOM path which was causing deadlocks reported by the lock prover.
>>
>
> [...]
>
>> +
>> +/**
>> + * sgx_epc_oom() - invoke EPC out-of-memory handling on target LRU
>> + * @lru: LRU that is low
>> + *
>> + * Return: %true if a victim was found and kicked.
>> + */
>> +bool sgx_epc_oom(struct sgx_epc_lru_lists *lru)
>> +{
>> + struct sgx_epc_page *victim;
>> +
>> + spin_lock(&lru->lock);
>> + victim = sgx_oom_get_victim(lru);
>> + spin_unlock(&lru->lock);
>> +
>> + if (!victim)
>> + return false;
>> +
>> + if (victim->flags & SGX_EPC_OWNER_PAGE)
>> + return sgx_oom_encl_page(victim->encl_page);
>> +
>> + if (victim->flags & SGX_EPC_OWNER_ENCL)
>> + return sgx_oom_encl(victim->encl);
>
> I hate to bring this up, at least at this stage, but I am wondering why
> we need
> to put VA and SECS pages to the unreclaimable list, but cannot keep an
> "enclave_list" instead?
>
> So by looking the patch (" x86/sgx: Limit process EPC usage with misc
> cgroup
> controller"), if I am not missing anything, the whole "unreclaimable"
> list is
> just used to find the victim enclave when OOM needs to be done. Thus, I
> don't
> see why "enclave_list" cannot be used to achieve this.
>
> The reason that I am asking is because it seems using "enclave_list" we
> can
> simplify the code. At least the patches related to track VA/SECS pages,
> and the
> SGX_EPC_OWNER_PAGE/SGX_EPC_OWNER_ENCL thing can be eliminated
> completely.
> Using "enclave_list", I guess you just need to put the enclave to the
> current
> EPC cgroup when SECS page is allocated.
>
Later the hosting process could migrated/reassigned to another cgroup?
What to do when the new cgroup is OOM?
Thanks
Haitao
next prev parent reply other threads:[~2023-10-10 1:06 UTC|newest]
Thread overview: 119+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-23 3:06 [PATCH v5 00/18] Add Cgroup support for SGX EPC memory Haitao Huang
2023-09-23 3:06 ` [PATCH v5 01/18] cgroup/misc: Add per resource callbacks for CSS events Haitao Huang
2023-09-25 17:09 ` Jarkko Sakkinen
2023-09-26 3:04 ` Haitao Huang
2023-09-26 13:10 ` Jarkko Sakkinen
2023-09-26 13:13 ` Jarkko Sakkinen
2023-09-27 1:56 ` Haitao Huang
2023-10-02 22:47 ` Jarkko Sakkinen
2023-10-02 22:55 ` Jarkko Sakkinen
2023-10-04 15:45 ` Haitao Huang
2023-10-04 17:18 ` Tejun Heo
2023-09-27 9:20 ` Huang, Kai
2023-10-03 14:29 ` Haitao Huang
2023-10-17 18:55 ` Michal Koutný
2023-09-23 3:06 ` [PATCH v5 02/18] cgroup/misc: Add SGX EPC resource type and export APIs for SGX driver Haitao Huang
2023-09-25 18:50 ` Tejun Heo
2023-09-28 3:59 ` Huang, Kai
2023-10-03 7:00 ` Haitao Huang
2023-10-03 19:33 ` Huang, Kai
2023-09-23 3:06 ` [PATCH v5 03/18] x86/sgx: Add sgx_epc_lru_lists to encapsulate LRU lists Haitao Huang
2023-09-23 3:06 ` [PATCH v5 04/18] x86/sgx: Use sgx_epc_lru_lists for existing active page list Haitao Huang
2023-09-23 3:06 ` [PATCH v5 05/18] x86/sgx: Store reclaimable EPC pages in sgx_epc_lru_lists Haitao Huang
2023-09-27 10:14 ` Huang, Kai
2023-09-23 3:06 ` [PATCH v5 06/18] x86/sgx: Introduce EPC page states Haitao Huang
2023-09-25 17:11 ` Jarkko Sakkinen
2023-09-27 10:28 ` Huang, Kai
2023-10-03 4:49 ` Haitao Huang
2023-10-03 20:03 ` Huang, Kai
2023-10-04 15:24 ` Haitao Huang
2023-10-04 21:05 ` Huang, Kai
2023-09-23 3:06 ` [PATCH v5 07/18] x86/sgx: Introduce RECLAIM_IN_PROGRESS state Haitao Huang
2023-09-25 17:13 ` Jarkko Sakkinen
2023-09-27 10:42 ` Huang, Kai
2023-09-23 3:06 ` [PATCH v5 08/18] x86/sgx: Use a list to track to-be-reclaimed pages Haitao Huang
2023-09-28 9:28 ` Huang, Kai
2023-10-03 5:09 ` Haitao Huang
2023-09-23 3:06 ` [PATCH v5 09/18] x86/sgx: Store struct sgx_encl when allocating new VA pages Haitao Huang
2023-09-27 11:14 ` Huang, Kai
2023-09-27 15:35 ` Haitao Huang
2023-09-27 21:21 ` Huang, Kai
2023-09-29 15:06 ` Haitao Huang
2023-10-02 11:05 ` Huang, Kai
2023-09-27 11:35 ` Huang, Kai
2023-10-03 6:45 ` Haitao Huang
2023-10-03 20:07 ` Huang, Kai
2023-10-04 15:03 ` Haitao Huang
2023-10-04 21:13 ` Huang, Kai
2023-10-05 4:22 ` Haitao Huang
2023-10-05 6:49 ` Huang, Kai
2023-09-23 3:06 ` [PATCH v5 10/18] x86/sgx: Add EPC page flags to identify owner types Haitao Huang
2023-09-23 3:06 ` [PATCH v5 11/18] x86/sgx: store unreclaimable pages in LRU lists Haitao Huang
2023-09-27 11:57 ` Huang, Kai
2023-10-03 5:42 ` Haitao Huang
2023-09-28 9:41 ` Huang, Kai
2023-10-03 5:15 ` Haitao Huang
2023-10-03 20:12 ` Huang, Kai
2023-09-23 3:06 ` [PATCH v5 12/18] x86/sgx: Add EPC OOM path to forcefully reclaim EPC Haitao Huang
2023-10-09 23:45 ` Huang, Kai
2023-10-10 0:23 ` Sean Christopherson
2023-10-10 0:50 ` Huang, Kai
2023-10-10 1:34 ` Huang, Kai
2023-10-10 16:49 ` Haitao Huang
2023-10-11 0:51 ` Huang, Kai
2023-10-12 13:27 ` Haitao Huang
2023-10-16 10:57 ` Huang, Kai
2023-10-16 19:52 ` Haitao Huang
2023-10-16 21:09 ` Huang, Kai
2023-10-17 0:10 ` Haitao Huang
2023-10-17 1:34 ` Huang, Kai
2023-10-17 12:58 ` Haitao Huang
2023-10-17 18:54 ` Michal Koutný
2023-10-17 19:13 ` Michal Koutný
2023-10-18 4:39 ` Haitao Huang
2023-10-18 4:37 ` Haitao Huang
2023-10-18 13:55 ` Dave Hansen
2023-10-18 15:26 ` Haitao Huang
2023-10-18 15:37 ` Dave Hansen
2023-10-18 15:52 ` Michal Koutný
2023-10-18 16:25 ` Haitao Huang
2023-10-16 21:32 ` Sean Christopherson
2023-10-17 0:09 ` Haitao Huang
2023-10-17 15:43 ` Sean Christopherson
2023-10-17 11:49 ` Mikko Ylinen
2023-10-11 1:14 ` Huang, Kai
2023-10-16 11:02 ` Huang, Kai
2023-10-10 1:42 ` Haitao Huang
2023-10-10 2:23 ` Huang, Kai
2023-10-10 13:26 ` Haitao Huang
2023-10-11 0:01 ` Sean Christopherson
2023-10-11 15:02 ` Haitao Huang
2023-10-10 1:04 ` Haitao Huang [this message]
2023-10-10 1:18 ` Huang, Kai
2023-10-10 1:38 ` Haitao Huang
2023-10-10 2:12 ` Huang, Kai
2023-10-10 17:05 ` Haitao Huang
2023-10-11 0:31 ` Huang, Kai
2023-10-11 16:04 ` Haitao Huang
2023-09-23 3:06 ` [PATCH v5 13/18] x86/sgx: Expose sgx_reclaim_pages() for use by EPC cgroup Haitao Huang
2023-10-05 12:24 ` Huang, Kai
2023-10-05 19:23 ` Haitao Huang
2023-10-05 20:25 ` Huang, Kai
2023-09-23 3:06 ` [PATCH v5 14/18] x86/sgx: Add helper to grab pages from an arbitrary EPC LRU Haitao Huang
2023-09-23 3:06 ` [PATCH v5 15/18] x86/sgx: Prepare for multiple LRUs Haitao Huang
2023-10-05 12:30 ` Huang, Kai
2023-10-05 19:33 ` Haitao Huang
2023-10-05 20:38 ` Huang, Kai
2023-09-23 3:06 ` [PATCH v5 16/18] x86/sgx: Limit process EPC usage with misc cgroup controller Haitao Huang
2023-09-25 17:15 ` Jarkko Sakkinen
2023-10-05 21:01 ` Huang, Kai
2023-10-10 0:12 ` Huang, Kai
2023-10-10 0:16 ` Huang, Kai
2023-10-10 0:26 ` Huang, Kai
2023-10-22 18:26 ` Haitao Huang
2023-10-10 9:19 ` Huang, Kai
2023-10-10 9:32 ` Huang, Kai
2023-10-17 18:54 ` Michal Koutný
2023-10-19 16:05 ` Haitao Huang
2023-09-23 3:06 ` [PATCH v5 17/18] Docs/x86/sgx: Add description for cgroup support Haitao Huang
2023-09-23 3:06 ` [PATCH v5 18/18] selftests/sgx: Add scripts for EPC cgroup testing 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=op.2ckqmgs9wjvjmi@hhuan26-mobl.amr.corp.intel.com \
--to=haitao.huang@linux.intel.com \
--cc=anakrish@microsoft.com \
--cc=bp@alien8.de \
--cc=cgroups@vger.kernel.org \
--cc=dave.hansen@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=mikko.ylinen@linux.intel.com \
--cc=mingo@redhat.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).