From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Thu, 27 Sep 2018 17:21:58 +0300 From: Jarkko Sakkinen To: Andy Lutomirski CC: Dave Hansen , "Christopherson, Sean J" , Andrew Lutomirski , "X86 ML" , Platform Driver , , , "Ayoun, Serge" , , , Andy Shevchenko , Dave Hansen , Peter Zijlstra , "Thomas Gleixner" , Ingo Molnar , "Borislav Petkov" , "H. Peter Anvin" , LKML Subject: Re: [PATCH v14 09/19] x86/mm: x86/sgx: Signal SEGV_SGXERR for #PFs w/ PF_SGX Message-ID: <20180927142158.GG8242@linux.intel.com> References: <20180925130845.9962-10-jarkko.sakkinen@linux.intel.com> <20180926173516.GA10920@linux.intel.com> <2D60780F-ADB4-48A4-AB74-15683493D369@amacapital.net> <9835e288-ba98-2f9e-ac73-504db9512bb9@intel.com> <20180926204400.GA11446@linux.intel.com> <992b1d6d-cc0f-776f-d938-2a1f7cad52c8@intel.com> Content-Type: text/plain; charset="us-ascii" In-Reply-To: Return-Path: jarkko.sakkinen@linux.intel.com MIME-Version: 1.0 List-ID: On Wed, Sep 26, 2018 at 03:37:45PM -0700, Andy Lutomirski wrote: > 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 Returning from AEX does not differ from any other memory access event so AFAIK it should be handled right with the proposed solution already. For convenience I think we could have a fixed trampoline for AEX e.g. this how it is implemented in the open source LE that I did: sgx_get_token: push %rbx mov $0x02, %rax mov %rsi, %rbx mov %rdx, %rsi mov $sgx_async_exit, %rcx sgx_async_exit: ENCLU pop %rbx ret BTW, if I converted the in-kernel LE as a standalone test program, would that be useful for basic testing of the series? /Jarkko From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jarkko Sakkinen Subject: Re: [PATCH v14 09/19] x86/mm: x86/sgx: Signal SEGV_SGXERR for #PFs w/ PF_SGX Date: Thu, 27 Sep 2018 17:21:58 +0300 Message-ID: <20180927142158.GG8242@linux.intel.com> References: <20180925130845.9962-10-jarkko.sakkinen@linux.intel.com> <20180926173516.GA10920@linux.intel.com> <2D60780F-ADB4-48A4-AB74-15683493D369@amacapital.net> <9835e288-ba98-2f9e-ac73-504db9512bb9@intel.com> <20180926204400.GA11446@linux.intel.com> <992b1d6d-cc0f-776f-d938-2a1f7cad52c8@intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org To: Andy Lutomirski Cc: Dave Hansen , "Christopherson, Sean J" , Andrew Lutomirski , X86 ML , Platform Driver , nhorman@redhat.com, npmccallum@redhat.com, "Ayoun, Serge" , shay.katz-zamir@intel.com, linux-sgx@vger.kernel.org, Andy Shevchenko , Dave Hansen , Peter Zijlstra , Thomas Gleixner , Ingo Molnar , Borislav Petkov , "H. Peter Anvin" , LKML List-Id: platform-driver-x86.vger.kernel.org On Wed, Sep 26, 2018 at 03:37:45PM -0700, Andy Lutomirski wrote: > 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 Returning from AEX does not differ from any other memory access event so AFAIK it should be handled right with the proposed solution already. For convenience I think we could have a fixed trampoline for AEX e.g. this how it is implemented in the open source LE that I did: sgx_get_token: push %rbx mov $0x02, %rax mov %rsi, %rbx mov %rdx, %rsi mov $sgx_async_exit, %rcx sgx_async_exit: ENCLU pop %rbx ret BTW, if I converted the in-kernel LE as a standalone test program, would that be useful for basic testing of the series? /Jarkko