From: Jarkko Sakkinen <jarkko@kernel.org>
To: Kai Huang <kai.huang@intel.com>
Cc: Reinette Chatre <reinette.chatre@intel.com>,
Dave Hansen <dave.hansen@intel.com>,
dave.hansen@linux.intel.com, linux-sgx@vger.kernel.org,
haitao.huang@intel.com
Subject: Re: [RFC PATCH 1/4] x86/sgx: Do not free backing memory on ENCLS[ELDU] failure
Date: Wed, 11 May 2022 13:26:15 +0300 [thread overview]
Message-ID: <YnuPR3OW7xdEXnVF@kernel.org> (raw)
In-Reply-To: <bb918c317ebb96768365b38f00239013b67ad607.camel@intel.com>
On Tue, May 10, 2022 at 12:36:19PM +1200, Kai Huang wrote:
> On Mon, 2022-05-09 at 10:17 -0700, Reinette Chatre wrote:
> > Hi Jarkko,
> >
> > On 5/7/2022 10:25 AM, Jarkko Sakkinen wrote:
> > > On Thu, Apr 28, 2022 at 04:49:00PM -0700, Reinette Chatre wrote:
> > > > > I also looked a little deeper at this transient failure problem. The
> > > > > ELDU documentation also mentions a possible error code of:
> > > > >
> > > > > SGX_EPC_PAGE_CONFLICT
> > > > >
> > > > > It *looks* like there can be conflicts on the SECS page as well as the
> > > > > EPC page being explicitly accessed. Is that a possible problem here?
> > > >
> > > > I went down this path myself. SGX_EPC_PAGE_CONFLICT is an error code
> > > > supported by newer ELDUC - the ELDU used in current code would indeed
> > > > #GP in this case. The SDM text describing ELDUC as "This leaf function
> > > > behaves like ELDU but with improved conflict handling for oversubscription"
> > > > really does seem relevant to the test that triggers this issue.
> > > >
> > > > I stopped pursuing this because from what I understand if
> > > > SGX_EPC_PAGE_CONFLICT is encountered with commit 08999b2489b4 ("x86/sgx:
> > > > Free backing memory after faulting the enclave page") then it should
> > > > also be encountered without it. The issue is not present with
> > > > 08999b2489b4 ("x86/sgx: Free backing memory after faulting the
> > > > enclave page") removed. I am thus currently investigating based on
> > > > the assumption that the #GP is encountered because of MAC
> > > > verification problem. I may be wrong here also and need more information
> > > > since the SDM documents two seemingly related errors:
> > > > #GP(0) -> If the instruction fails to verify MAC.
> > > > SGX_MAC_COMPARE_FAIL -> If the MAC check fails.
> > >
> > > This part puzzles me in the pseudo-code.
> > >
> > > The version is read first:
> > >
> > > TMP_VER := DS:RDX[63:0];
> > >
> > > Then there's MAC calculation, comparison, and finally this check:
> > >
> > > (* Check version before committing *)
> > > IF (DS:RDX ≠ 0)
> > > THEN #GP(0);
> > > ELSE
> > > DS:RDX := TMP_VER;
> > > FI;
> > >
> > > For me it is a mystery what does zero the slot and in what condition
> > > it would be non-zero. Perhaps the #GP refers anyway to this check?
> >
> > RDX contains the VA slot information and that appears to be correct
> > in these scenarios. The issue is the PCMD data pointed to by the
> > PAGEINFO.PCMD (link to PAGEINFO found in RBX) is all zeroes.
> >
> > There are two scenarios under which the PCMD data could be zeroes. They
> > are documented in:
> > https://lore.kernel.org/linux-sgx/8157fa40-8d02-8819-e1b6-fd2d8863fb56@intel.com/
> > https://lore.kernel.org/linux-sgx/da387afc-e666-45d0-1e99-066d8c4aab03@intel.com/
> >
> > I understand that context may be lost by pointing you to various emails
> > in this thread - I'll wrap up all learnings when I submit the new version
> > of this series today.
> >
>
> Hi Reinette,
>
> Regardless the root cause of this problem, I agree with Jarkko above pseudo-code
> in the spec is quite confusing. I can try to get it clarified from Intel
> internally if you want.
It is :-) Yeah, it would be great if it could be made a bit more punctual!
BR, Jarkko
next prev parent reply other threads:[~2022-05-11 10:27 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-04-28 20:11 [RFC PATCH 0/4] SGX shmem backing store issue Reinette Chatre
2022-04-28 20:11 ` [RFC PATCH 1/4] x86/sgx: Do not free backing memory on ENCLS[ELDU] failure Reinette Chatre
2022-04-28 21:30 ` Dave Hansen
2022-04-28 22:20 ` Reinette Chatre
2022-04-28 22:53 ` Dave Hansen
2022-04-28 23:49 ` Reinette Chatre
2022-05-03 2:01 ` Kai Huang
2022-05-07 17:25 ` Jarkko Sakkinen
2022-05-09 17:17 ` Reinette Chatre
2022-05-10 0:36 ` Kai Huang
2022-05-11 10:26 ` Jarkko Sakkinen [this message]
2022-05-11 18:29 ` Haitao Huang
2022-05-11 22:00 ` Kai Huang
2022-05-12 21:14 ` Jarkko Sakkinen
2022-05-06 22:09 ` Jarkko Sakkinen
2022-04-28 20:11 ` [RFC PATCH 2/4] x86/sgx: Set dirty bit after modifying page contents Reinette Chatre
2022-04-28 21:40 ` Dave Hansen
2022-04-28 22:41 ` Reinette Chatre
2022-05-06 22:27 ` Jarkko Sakkinen
2022-05-06 22:40 ` Reinette Chatre
2022-05-07 18:01 ` Jarkko Sakkinen
2022-04-28 20:11 ` [RFC PATCH 3/4] x86/sgx: Obtain backing storage page with enclave mutex held Reinette Chatre
2022-04-28 21:58 ` Dave Hansen
2022-04-28 22:44 ` Reinette Chatre
2022-05-06 22:43 ` Jarkko Sakkinen
2022-04-28 20:11 ` [RFC PATCH 4/4] x86/sgx: Do not allocate backing pages when loading from backing store Reinette Chatre
2022-04-28 21:12 ` [RFC PATCH 0/4] SGX shmem backing store issue Dave Hansen
2022-04-29 18:50 ` Reinette Chatre
2022-04-29 19:45 ` Dave Hansen
2022-04-30 3:22 ` Reinette Chatre
2022-04-30 15:52 ` Reinette Chatre
2022-05-02 14:36 ` Dave Hansen
2022-05-02 17:11 ` Reinette Chatre
2022-05-02 21:33 ` Dave Hansen
2022-05-04 22:13 ` Reinette Chatre
2022-05-04 22:58 ` Dave Hansen
2022-05-04 23:36 ` Reinette Chatre
2022-05-04 23:50 ` Dave Hansen
2022-05-05 0:08 ` Reinette Chatre
2022-05-04 23:05 ` Dave Hansen
2022-05-07 17:46 ` Jarkko Sakkinen
2022-05-07 17:48 ` Jarkko Sakkinen
2022-05-09 17:09 ` Reinette Chatre
2022-05-10 22:28 ` Jarkko Sakkinen
2022-05-11 17:23 ` Reinette Chatre
2022-05-12 14:10 ` Jarkko Sakkinen
2022-04-28 21:29 ` Dave Hansen
2022-04-28 22:20 ` Reinette Chatre
2022-05-04 6:40 ` Jarkko Sakkinen
2022-05-05 6:09 ` Jarkko Sakkinen
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=YnuPR3OW7xdEXnVF@kernel.org \
--to=jarkko@kernel.org \
--cc=dave.hansen@intel.com \
--cc=dave.hansen@linux.intel.com \
--cc=haitao.huang@intel.com \
--cc=kai.huang@intel.com \
--cc=linux-sgx@vger.kernel.org \
--cc=reinette.chatre@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).