From: Andy Lutomirski <luto@amacapital.net> To: Dave Hansen <dave.hansen@intel.com> Cc: "Christopherson, Sean J" <sean.j.christopherson@intel.com>, Andrew Lutomirski <luto@kernel.org>, Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>, X86 ML <x86@kernel.org>, Platform Driver <platform-driver-x86@vger.kernel.org>, nhorman@redhat.com, npmccallum@redhat.com, "Ayoun, Serge" <serge.ayoun@intel.com>, shay.katz-zamir@intel.com, linux-sgx@vger.kernel.org, Andy Shevchenko <andriy.shevchenko@linux.intel.com>, Dave Hansen <dave.hansen@linux.intel.com>, Peter Zijlstra <peterz@infradead.org>, Thomas Gleixner <tglx@linutronix.de>, Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>, "H. Peter Anvin" <hpa@zytor.com>, LKML <linux-kernel@vger.kernel.org> Subject: Re: [PATCH v14 09/19] x86/mm: x86/sgx: Signal SEGV_SGXERR for #PFs w/ PF_SGX Date: Wed, 26 Sep 2018 15:37:45 -0700 [thread overview] Message-ID: <CALCETrXyjRh1sBk1V21iTtjdhiWGCYsHnUNktkJC5xFiZy6c9A@mail.gmail.com> (raw) In-Reply-To: <992b1d6d-cc0f-776f-d938-2a1f7cad52c8@intel.com> On Wed, Sep 26, 2018 at 2:45 PM Dave Hansen <dave.hansen@intel.com> wrote: > > On 09/26/2018 02:15 PM, Andy Lutomirski wrote: > > Could we perhaps have a little vDSO entry (or syscall, I suppose) that > > runs an enclave an returns an error code, and rig up the #PF handler > > to check if the error happened in the vDSO entry and fix it up rather > > than sending a signal? > > Yeah, signals suck. > > So, instead of doing the enclave entry instruction (EENTER is it?), the > app would do the vDSO call. It would have some calling convention, like > "set %rax to 0 before entering". Then, we just teach the page fault > handler about the %RIP in the vDSO that can fault and how to move one > instruction later, munge %RIP to a value that tells about the error, > then return from the fault. It would basically be like the kernel > exception tables, but for userspace. Right? Yeah. Maybe like this: xorl %eax,%eax eenter_insn: ENCLU[whatever] eenter_landing_pad: ret And the kernel would use the existing vdso2c vdso-symbol-finding mechanism to do the fixup. > > How would a syscall work, though? I assume we can't just enter the > enclave from ring0. My understanding of how AEX works is a bit vague, but maybe a syscall could reuse the mechanism? The vDSO approach seems considerably simpler. We do need to make sure that a fault that happens on or after return from an AEX event does the right thing. But I'm still vague on how that works, sigh. --Andy
WARNING: multiple messages have this Message-ID (diff)
From: Andy Lutomirski <luto@amacapital.net> To: Dave Hansen <dave.hansen@intel.com> Cc: "Christopherson, Sean J" <sean.j.christopherson@intel.com>, "Andrew Lutomirski" <luto@kernel.org>, Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>, X86 ML <x86@kernel.org>, Platform Driver <platform-driver-x86@vger.kernel.org>, <nhorman@redhat.com>, <npmccallum@redhat.com>, "Ayoun, Serge" <serge.ayoun@intel.com>, <shay.katz-zamir@intel.com>, <linux-sgx@vger.kernel.org>, Andy Shevchenko <andriy.shevchenko@linux.intel.com>, Dave Hansen <dave.hansen@linux.intel.com>, Peter Zijlstra <peterz@infradead.org>, "Thomas Gleixner" <tglx@linutronix.de>, Ingo Molnar <mingo@redhat.com>, "Borislav Petkov" <bp@alien8.de>, "H. Peter Anvin" <hpa@zytor.com>, LKML <linux-kernel@vger.kernel.org> Subject: Re: [PATCH v14 09/19] x86/mm: x86/sgx: Signal SEGV_SGXERR for #PFs w/ PF_SGX Date: Wed, 26 Sep 2018 15:37:45 -0700 [thread overview] Message-ID: <CALCETrXyjRh1sBk1V21iTtjdhiWGCYsHnUNktkJC5xFiZy6c9A@mail.gmail.com> (raw) In-Reply-To: <992b1d6d-cc0f-776f-d938-2a1f7cad52c8@intel.com> On Wed, Sep 26, 2018 at 2:45 PM Dave Hansen <dave.hansen@intel.com> wrote: > > On 09/26/2018 02:15 PM, Andy Lutomirski wrote: > > Could we perhaps have a little vDSO entry (or syscall, I suppose) that > > runs an enclave an returns an error code, and rig up the #PF handler > > to check if the error happened in the vDSO entry and fix it up rather > > than sending a signal? > > Yeah, signals suck. > > So, instead of doing the enclave entry instruction (EENTER is it?), the > app would do the vDSO call. It would have some calling convention, like > "set %rax to 0 before entering". Then, we just teach the page fault > handler about the %RIP in the vDSO that can fault and how to move one > instruction later, munge %RIP to a value that tells about the error, > then return from the fault. It would basically be like the kernel > exception tables, but for userspace. Right? Yeah. Maybe like this: xorl %eax,%eax eenter_insn: ENCLU[whatever] eenter_landing_pad: ret And the kernel would use the existing vdso2c vdso-symbol-finding mechanism to do the fixup. > > How would a syscall work, though? I assume we can't just enter the > enclave from ring0. My understanding of how AEX works is a bit vague, but maybe a syscall could reuse the mechanism? The vDSO approach seems considerably simpler. We do need to make sure that a fault that happens on or after return from an AEX event does the right thing. But I'm still vague on how that works, sigh. --Andy
next prev parent reply other threads:[~2018-09-26 22:38 UTC|newest] Thread overview: 184+ messages / expand[flat|nested] mbox.gz Atom feed top 2018-09-25 13:06 [PATCH v14 00/19] Intel SGX1 support Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 01/19] x86/sgx: Update MAINTAINERS Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 02/19] x86/sgx: Architectural structures Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 03/19] x86/cpufeature: Add SGX and SGX_LC CPU features Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 04/19] x86/cpufeatures: Add SGX feature bits Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 16:48 ` Borislav Petkov 2018-09-25 16:48 ` Borislav Petkov 2018-09-25 16:48 ` Borislav Petkov 2018-09-26 11:11 ` Jarkko Sakkinen 2018-09-26 11:11 ` Jarkko Sakkinen 2018-09-26 11:11 ` Jarkko Sakkinen 2018-09-26 11:36 ` Borislav Petkov 2018-09-26 11:36 ` Borislav Petkov 2018-09-27 13:16 ` Jarkko Sakkinen 2018-09-27 13:16 ` Jarkko Sakkinen 2018-09-27 13:51 ` Borislav Petkov 2018-09-27 13:51 ` Borislav Petkov 2018-09-27 14:52 ` Jarkko Sakkinen 2018-09-27 14:52 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 05/19] x86/msr: Add SGX definitions to msr-index.h Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 06/19] x86/mm: x86/sgx: Add new 'PF_SGX' page fault error code bit Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 07/19] x86/fault: x86/mm/pkeys: relocate stale comment regarding OSPKE Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 08/19] signal: x86/sgx: Add SIGSEGV siginfo code for SGX EPCM fault Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-26 19:14 ` Sean Christopherson 2018-09-26 19:14 ` Sean Christopherson 2018-09-26 19:14 ` Sean Christopherson 2018-09-27 18:41 ` Eric W. Biederman 2018-09-27 18:41 ` Eric W. Biederman 2018-09-27 18:41 ` Eric W. Biederman 2018-09-25 13:06 ` [PATCH v14 09/19] x86/mm: x86/sgx: Signal SEGV_SGXERR for #PFs w/ PF_SGX Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 22:53 ` Andy Lutomirski 2018-09-25 22:53 ` Andy Lutomirski 2018-09-26 17:35 ` Sean Christopherson 2018-09-26 17:35 ` Sean Christopherson 2018-09-26 18:12 ` Andy Lutomirski 2018-09-26 18:12 ` Andy Lutomirski 2018-09-26 20:16 ` Dave Hansen 2018-09-26 20:16 ` Dave Hansen 2018-09-26 20:44 ` Sean Christopherson 2018-09-26 20:44 ` Sean Christopherson 2018-09-26 20:49 ` Dave Hansen 2018-09-26 20:49 ` Dave Hansen 2018-09-26 21:15 ` Andy Lutomirski 2018-09-26 21:15 ` Andy Lutomirski 2018-09-26 21:45 ` Dave Hansen 2018-09-26 21:45 ` Dave Hansen 2018-09-26 22:37 ` Andy Lutomirski [this message] 2018-09-26 22:37 ` Andy Lutomirski 2018-09-27 14:21 ` Jarkko Sakkinen 2018-09-27 14:21 ` Jarkko Sakkinen 2018-09-27 14:41 ` Andy Lutomirski 2018-09-27 14:41 ` Andy Lutomirski 2018-09-27 13:56 ` Jarkko Sakkinen 2018-09-27 13:56 ` Jarkko Sakkinen 2018-10-01 21:42 ` Jethro Beekman 2018-10-01 22:03 ` Dave Hansen 2018-10-31 21:30 ` Sean Christopherson 2018-10-31 21:30 ` Sean Christopherson 2018-10-31 21:35 ` Dave Hansen 2018-10-31 21:35 ` Dave Hansen 2018-10-31 21:53 ` Jethro Beekman 2018-10-31 21:58 ` Dave Hansen 2018-10-31 22:52 ` Andy Lutomirski 2018-10-31 22:52 ` Andy Lutomirski 2018-11-01 17:51 ` Dave Hansen 2018-11-01 17:52 ` Andy Lutomirski 2018-11-01 17:52 ` Andy Lutomirski 2018-11-01 17:42 ` Jarkko Sakkinen 2018-11-01 17:42 ` Jarkko Sakkinen 2018-11-01 17:42 ` Jarkko Sakkinen 2018-11-01 17:44 ` Jarkko Sakkinen 2018-11-01 17:44 ` Jarkko Sakkinen 2018-11-01 17:44 ` Jarkko Sakkinen 2018-10-02 0:31 ` Jarkko Sakkinen 2018-10-02 0:31 ` Jarkko Sakkinen 2018-10-01 14:29 ` Sean Christopherson 2018-10-01 14:29 ` Sean Christopherson 2018-10-01 14:41 ` Dave Hansen 2018-10-01 14:41 ` Dave Hansen 2018-10-02 0:07 ` Jarkko Sakkinen 2018-10-02 0:07 ` Jarkko Sakkinen 2018-09-27 13:42 ` Jarkko Sakkinen 2018-09-27 13:42 ` Jarkko Sakkinen 2018-09-27 14:58 ` Dave Hansen 2018-09-27 14:58 ` Dave Hansen 2018-09-27 15:39 ` Jarkko Sakkinen 2018-09-27 15:39 ` Jarkko Sakkinen 2018-09-27 15:53 ` Dave Hansen 2018-09-27 15:53 ` Dave Hansen 2018-09-27 13:14 ` Jarkko Sakkinen 2018-09-27 13:14 ` Jarkko Sakkinen 2018-09-27 19:43 ` Eric W. Biederman 2018-09-27 19:43 ` Eric W. Biederman 2018-09-27 19:43 ` Eric W. Biederman 2018-09-28 12:17 ` Jarkko Sakkinen 2018-09-28 12:17 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 10/19] x86/sgx: Detect Intel SGX Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 20:02 ` Randy Dunlap 2018-09-25 20:02 ` Randy Dunlap 2018-09-25 20:02 ` Randy Dunlap 2018-09-27 13:13 ` Jarkko Sakkinen 2018-09-27 13:13 ` Jarkko Sakkinen 2018-09-27 13:13 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 11/19] x86/sgx: Add wrappers for ENCLS leaf functions Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 20:01 ` Randy Dunlap 2018-09-25 20:01 ` Randy Dunlap 2018-09-25 20:01 ` Randy Dunlap 2018-09-27 13:12 ` Jarkko Sakkinen 2018-09-27 13:12 ` Jarkko Sakkinen 2018-09-27 13:12 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 12/19] x86/sgx: Add data structures for tracking the EPC pages Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 20:00 ` Randy Dunlap 2018-09-25 20:00 ` Randy Dunlap 2018-09-25 20:00 ` Randy Dunlap 2018-09-27 13:11 ` Jarkko Sakkinen 2018-09-27 13:11 ` Jarkko Sakkinen 2018-09-27 13:11 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 13/19] x86/sgx: Enclave Page Cache (EPC) memory manager Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 14/19] x86/sgx: Add sgx_einit() for initializing enclaves Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 15/19] platform/x86: Intel SGX driver Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-10-04 18:01 ` Sean Christopherson 2018-10-04 18:01 ` Sean Christopherson 2018-10-04 18:01 ` Sean Christopherson 2018-10-05 11:32 ` Jarkko Sakkinen 2018-10-05 11:32 ` Jarkko Sakkinen 2018-10-05 11:32 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 16/19] platform/x86: Add swapping functionality to the " Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 17/19] x86/sgx: Add a simple swapper for the EPC memory manager Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 18/19] platform/x86: ptrace() support for the SGX driver Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` [PATCH v14 19/19] x86/sgx: Driver documentation Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:06 ` Jarkko Sakkinen 2018-09-25 13:27 ` Jonathan Corbet 2018-09-25 13:27 ` Jonathan Corbet 2018-09-25 13:27 ` Jonathan Corbet 2018-10-15 20:54 ` Pavel Machek 2018-10-15 20:54 ` Pavel Machek 2018-10-17 23:45 ` Jarkko Sakkinen 2018-10-17 23:45 ` Jarkko Sakkinen 2018-10-18 9:57 ` Pavel Machek 2018-10-18 9:57 ` Pavel Machek 2018-10-19 23:59 ` Jarkko Sakkinen 2018-10-19 23:59 ` Jarkko Sakkinen 2018-10-17 23:56 ` Dave Hansen 2018-10-17 23:56 ` Dave Hansen
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=CALCETrXyjRh1sBk1V21iTtjdhiWGCYsHnUNktkJC5xFiZy6c9A@mail.gmail.com \ --to=luto@amacapital.net \ --cc=andriy.shevchenko@linux.intel.com \ --cc=bp@alien8.de \ --cc=dave.hansen@intel.com \ --cc=dave.hansen@linux.intel.com \ --cc=hpa@zytor.com \ --cc=jarkko.sakkinen@linux.intel.com \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-sgx@vger.kernel.org \ --cc=luto@kernel.org \ --cc=mingo@redhat.com \ --cc=nhorman@redhat.com \ --cc=npmccallum@redhat.com \ --cc=peterz@infradead.org \ --cc=platform-driver-x86@vger.kernel.org \ --cc=sean.j.christopherson@intel.com \ --cc=serge.ayoun@intel.com \ --cc=shay.katz-zamir@intel.com \ --cc=tglx@linutronix.de \ --cc=x86@kernel.org \ /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: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.