linux-kernel - [RFC PATCH v5 0/5] x86: Add vDSO exception fixup for SGX

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date:   Fri, 14 Dec 2018 13:57:24 -0800
From:   Sean Christopherson <sean.j.christopherson@...el.com>
To:     Andy Lutomirski <luto@...nel.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        x86@...nel.org, Dave Hansen <dave.hansen@...ux.intel.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Jarkko Sakkinen <jarkko.sakkinen@...ux.intel.com>
Cc:     "H. Peter Anvin" <hpa@...or.com>, linux-kernel@...r.kernel.org,
        linux-sgx@...r.kernel.org, Andy Lutomirski <luto@...capital.net>,
        Josh Triplett <josh@...htriplett.org>,
        Haitao Huang <haitao.huang@...ux.intel.com>,
        Jethro Beekman <jethro@...tanix.com>,
        "Dr . Greg Wettstein" <greg@...ellic.com>
Subject: [RFC PATCH v5 0/5] x86: Add vDSO exception fixup for SGX

__vdso_sgx_enter_enclave() gets another rewrite, this time to strip
it down to the bare minimum and explicitly break compliance with the
x86-64 ABI.  Feedback from v4 revealed that __vdso_sgx_enter_enclave()
would need to save (a lot) more than just the non-volatile GPRs to be
compliant with the x86-64 ABI, at which point breaking from the ABI
completely became much more palatable.

The non-standard ABI also solves the question of "which GPRs should be
marshalled to/from the enclave" by getting out of the way entirely and
letting userspace have free reign (except for RSP, which has a big ol'
DO NOT TOUCH sign on it).

[1] https://lkml.kernel.org/r/cda13cff-1a56-a40f-7d69-f0f1ab752f8e@fortanix.com


v1: https://lkml.kernel.org/r/20181205232012.28920-1-sean.j.christopherson@intel.com
v2: https://lkml.kernel.org/r/20181206221922.31012-1-sean.j.christopherson@intel.com
v3: https://lkml.kernel.org/r/20181210232141.5425-1-sean.j.christopherson@intel.com
v4: https://lkml.kernel.org/r/20181213213135.12913-1-sean.j.christopherson@intel.com
v5:
  - Strip down  __vdso_sgx_enter_enclave() so it only touches RSP and
    GPRs that are already collateral damage of ENCLU[EENTER/ERESUME].
  - Add a 16-byte reserved field to 'struct sgx_enclave_exception' so
    the struct can grow in a backwards compatible fashion.
  - Add a blurb at the end of the changelog for patch 1/5 explaining
    why the vDSO fixup macros look different than the equivalent kernel
    macros.

Sean Christopherson (5):
  x86/vdso: Add support for exception fixup in vDSO functions
  x86/fault: Add helper function to sanitize error code
  x86/fault: Attempt to fixup unhandled #PF on ENCLU before signaling
  x86/traps: Attempt to fixup exceptions in vDSO before signaling
  x86/vdso: Add __vdso_sgx_enter_enclave() to wrap SGX enclave
    transitions

 arch/x86/entry/vdso/Makefile             |  6 +-
 arch/x86/entry/vdso/extable.c            | 37 +++++++++
 arch/x86/entry/vdso/extable.h            | 29 +++++++
 arch/x86/entry/vdso/vdso-layout.lds.S    |  9 ++-
 arch/x86/entry/vdso/vdso.lds.S           |  1 +
 arch/x86/entry/vdso/vdso2c.h             | 58 ++++++++++++--
 arch/x86/entry/vdso/vsgx_enter_enclave.S | 97 ++++++++++++++++++++++++
 arch/x86/include/asm/vdso.h              |  5 ++
 arch/x86/include/uapi/asm/sgx.h          | 18 +++++
 arch/x86/kernel/traps.c                  | 14 ++++
 arch/x86/mm/fault.c                      | 33 +++++---
 11 files changed, 284 insertions(+), 23 deletions(-)
 create mode 100644 arch/x86/entry/vdso/extable.c
 create mode 100644 arch/x86/entry/vdso/extable.h
 create mode 100644 arch/x86/entry/vdso/vsgx_enter_enclave.S

-- 
2.19.2