linux-kernel - Re: [PATCH v35 21/24] x86/vdso: Implement a vDSO for Intel SGX enclave call

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-prev] [thread-next>] [day] [month] [year] [list]
Date:   Tue, 14 Jul 2020 12:07:54 +0200
From:   Jethro Beekman <jethro@...tanix.com>
To:     Jarkko Sakkinen <jarkko.sakkinen@...ux.intel.com>
Cc:     x86@...nel.org, linux-sgx@...r.kernel.org,
        linux-kernel@...r.kernel.org,
        Sean Christopherson <sean.j.christopherson@...el.com>,
        Andy Lutomirski <luto@...capital.net>,
        Cedric Xing <cedric.xing@...el.com>, akpm@...ux-foundation.org,
        andriy.shevchenko@...ux.intel.com, asapek@...gle.com, bp@...en8.de,
        chenalexchen@...gle.com, conradparker@...gle.com,
        cyhanish@...gle.com, dave.hansen@...el.com, haitao.huang@...el.com,
        josh@...htriplett.org, kai.huang@...el.com, kai.svahn@...el.com,
        kmoy@...gle.com, ludloff@...gle.com, luto@...nel.org,
        nhorman@...hat.com, npmccallum@...hat.com, puiterwijk@...hat.com,
        rientjes@...gle.com, tglx@...utronix.de, yaozhangx@...gle.com
Subject: Re: [PATCH v35 21/24] x86/vdso: Implement a vDSO for Intel SGX
 enclave call

On 2020-07-14 11:56, Jarkko Sakkinen wrote:
> On Tue, Jul 14, 2020 at 09:30:03AM +0200, Jethro Beekman wrote:
>> On 2020-07-07 05:37, Jarkko Sakkinen wrote:
>>> From: Sean Christopherson <sean.j.christopherson@...el.com>
>>>
>>> An SGX runtime must be aware of the exceptions, which happen inside an
>>> enclave. Introduce a vDSO call that wraps EENTER/ERESUME cycle and returns
>>> the CPU exception back to the caller exactly when it happens.
>>>
>>> Kernel fixups the exception information to RDI, RSI and RDX. The SGX call
>>> vDSO handler fills this information to the user provided buffer or
>>> alternatively trigger user provided callback at the time of the exception.
>>>
>>> The calling convention is custom and does not follow System V x86-64 ABI.
>>>
>>> Suggested-by: Andy Lutomirski <luto@...capital.net>
>>> Acked-by: Jethro Beekman <jethro@...tanix.com>
>>> Tested-by: Jethro Beekman <jethro@...tanix.com>
>>> Signed-off-by: Sean Christopherson <sean.j.christopherson@...el.com>
>>> Co-developed-by: Cedric Xing <cedric.xing@...el.com>
>>> Signed-off-by: Cedric Xing <cedric.xing@...el.com>
>>> Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen@...ux.intel.com>
>>> ---
>>>  arch/x86/entry/vdso/Makefile             |   2 +
>>>  arch/x86/entry/vdso/vdso.lds.S           |   1 +
>>>  arch/x86/entry/vdso/vsgx_enter_enclave.S | 131 +++++++++++++++++++++++
>>>  arch/x86/include/asm/enclu.h             |   8 ++
>>>  arch/x86/include/uapi/asm/sgx.h          |  98 +++++++++++++++++
>>>  5 files changed, 240 insertions(+)
>>>  create mode 100644 arch/x86/entry/vdso/vsgx_enter_enclave.S
>>>  create mode 100644 arch/x86/include/asm/enclu.h
>>>
>>> diff --git a/arch/x86/entry/vdso/Makefile b/arch/x86/entry/vdso/Makefile
>>> index ebe82b7aecda..f71ad5ebd0c4 100644
>>> --- a/arch/x86/entry/vdso/Makefile
>>> +++ b/arch/x86/entry/vdso/Makefile
>>> @@ -29,6 +29,7 @@ VDSO32-$(CONFIG_IA32_EMULATION)	:= y
>>>  vobjs-y := vdso-note.o vclock_gettime.o vgetcpu.o
>>>  vobjs32-y := vdso32/note.o vdso32/system_call.o vdso32/sigreturn.o
>>>  vobjs32-y += vdso32/vclock_gettime.o
>>> +vobjs-$(VDSO64-y)		+= vsgx_enter_enclave.o
>>>  
>>>  # files to link into kernel
>>>  obj-y				+= vma.o extable.o
>>> @@ -100,6 +101,7 @@ $(vobjs): KBUILD_CFLAGS := $(filter-out $(GCC_PLUGINS_CFLAGS) $(RETPOLINE_CFLAGS
>>>  CFLAGS_REMOVE_vclock_gettime.o = -pg
>>>  CFLAGS_REMOVE_vdso32/vclock_gettime.o = -pg
>>>  CFLAGS_REMOVE_vgetcpu.o = -pg
>>> +CFLAGS_REMOVE_vsgx_enter_enclave.o = -pg
>>>  
>>>  #
>>>  # X32 processes use x32 vDSO to access 64bit kernel data.
>>> diff --git a/arch/x86/entry/vdso/vdso.lds.S b/arch/x86/entry/vdso/vdso.lds.S
>>> index 36b644e16272..4bf48462fca7 100644
>>> --- a/arch/x86/entry/vdso/vdso.lds.S
>>> +++ b/arch/x86/entry/vdso/vdso.lds.S
>>> @@ -27,6 +27,7 @@ VERSION {
>>>  		__vdso_time;
>>>  		clock_getres;
>>>  		__vdso_clock_getres;
>>> +		__vdso_sgx_enter_enclave;
>>>  	local: *;
>>>  	};
>>>  }
>>> diff --git a/arch/x86/entry/vdso/vsgx_enter_enclave.S b/arch/x86/entry/vdso/vsgx_enter_enclave.S
>>> new file mode 100644
>>> index 000000000000..be7e467e1efb
>>> --- /dev/null
>>> +++ b/arch/x86/entry/vdso/vsgx_enter_enclave.S
>>> @@ -0,0 +1,131 @@
>>> +/* SPDX-License-Identifier: GPL-2.0 */
>>> +
>>> +#include <linux/linkage.h>
>>> +#include <asm/export.h>
>>> +#include <asm/errno.h>
>>> +#include <asm/enclu.h>
>>> +
>>> +#include "extable.h"
>>> +
>>> +#define EX_LEAF		0*8
>>> +#define EX_TRAPNR	0*8+4
>>> +#define EX_ERROR_CODE	0*8+6
>>> +#define EX_ADDRESS	1*8
>>> +
>>> +.code64
>>> +.section .text, "ax"
>>> +
>>> +SYM_FUNC_START(__vdso_sgx_enter_enclave)
>>> +	/* Prolog */
>>> +	.cfi_startproc
>>> +	push	%rbp
>>> +	.cfi_adjust_cfa_offset	8
>>> +	.cfi_rel_offset		%rbp, 0
>>> +	mov	%rsp, %rbp
>>> +	.cfi_def_cfa_register	%rbp
>>> +	push	%rbx
>>> +	.cfi_rel_offset		%rbx, -8
>>> +
>>> +	mov	%ecx, %eax
>>> +.Lenter_enclave:
>>> +	/* EENTER <= leaf <= ERESUME */
>>> +	cmp	$EENTER, %eax
>>> +	jb	.Linvalid_leaf
>>> +	cmp	$ERESUME, %eax
>>> +	ja	.Linvalid_leaf
>>> +
>>> +	/* Load TCS and AEP */
>>> +	mov	0x10(%rbp), %rbx
>>> +	lea	.Lasync_exit_pointer(%rip), %rcx
>>> +
>>> +	/* Single ENCLU serving as both EENTER and AEP (ERESUME) */
>>> +.Lasync_exit_pointer:
>>> +.Lenclu_eenter_eresume:
>>> +	enclu
>>
>> After thinking about this some more, I'd like to come back to this
>> setup. Prior discussion at https://lkml.org/lkml/2018/11/2/597 . I
>> hope I'm not derailing the discussion so much as to delay the patch
>> set :(
>>
>> I previously mentioned “Userspace may want fine-grained control over
>> enclave scheduling” as a reason userspace may want to specify a
>> different AEP, but gave a bad example. Here's a better example: If I'm
>> running my enclave in an M:N threading model (where M user threads run
>> N TCSs, with N > M), an AEX is a good oppurtunity to switch contexts.
>> Yes, I could implement this with alarm() or so, but that adds overhead
>> while missing out on a lot of opportunities for context switching.
> 
> The vDSO interface also provides optional callback. Wonder if that
> works for this or can it be refined to work for this?

Yeah I think if the callback was called instead of ENCLU, the callback has the opportunity to return non-positive which will trigger a return from __vdso_sgx_enter_enclave. Moving .Lasync_exit_pointer to .Lhandle_exit might be sufficient. But I imagine not all users would want this behavior (although calling the few userspace instructions is likely negligible compared to the actual ERESUME).

--
Jethro Beekman | Fortanix


Download attachment "smime.p7s" of type "application/pkcs7-signature" (4054 bytes)