Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Content-Type: text/plain;
        charset=utf-8
Mime-Version: 1.0 (1.0)
Subject: Re: [PATCH 06/13] arc: define syscall_get_arch()
From:   Andy Lutomirski <luto@amacapital.net>
In-Reply-To: <C2D7FE5348E1B147BCA15975FBA23075013A59239B@us01wembx1.internal.synopsys.com>
Date:   Fri, 9 Nov 2018 11:03:32 -0800
Cc:     "Dmitry V. Levin" <ldv@altlinux.org>,
        Andy Lutomirski <luto@kernel.org>,
        Paul Moore <paul@paul-moore.com>,
        Eric Paris <eparis@redhat.com>,
        Elvira Khabirova <lineprinter@altlinux.org>,
        "linux-snps-arc@lists.infradead.org" 
        <linux-snps-arc@lists.infradead.org>,
        "linux-audit@redhat.com" <linux-audit@redhat.com>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Content-Transfer-Encoding: quoted-printable
Message-Id: <6C398ED8-30BC-4483-B5CF-1EB2F2699B0D@amacapital.net>
References: <20181109031605.GF21511@altlinux.org> <C2D7FE5348E1B147BCA15975FBA23075013A59239B@us01wembx1.internal.synopsys.com>
To:     Vineet Gupta <vineet.gupta1@synopsys.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> On Nov 9, 2018, at 8:50 AM, Vineet Gupta <vineet.gupta1@synopsys.com> wrot=
e:
>=20
>> On 11/8/18 7:16 PM, Dmitry V. Levin wrote:
>> syscall_get_arch() is required to be implemented on all architectures
>> that use tracehook_report_syscall_entry() in order to extend
>> the generic ptrace API with PTRACE_GET_SYSCALL_INFO request.
>>=20
>> Signed-off-by: Dmitry V. Levin <ldv@altlinux.org>
>> ---
>> arch/arc/include/asm/syscall.h | 6 ++++++
>> include/uapi/linux/audit.h     | 1 +
>> 2 files changed, 7 insertions(+)
>>=20
>> diff --git a/arch/arc/include/asm/syscall.h b/arch/arc/include/asm/syscal=
l.h
>> index 29de09804306..5662778a7411 100644
>> --- a/arch/arc/include/asm/syscall.h
>> +++ b/arch/arc/include/asm/syscall.h
>> @@ -9,6 +9,7 @@
>> #ifndef _ASM_ARC_SYSCALL_H
>> #define _ASM_ARC_SYSCALL_H  1
>>=20
>> +#include <uapi/linux/audit.h>
>> #include <linux/err.h>
>> #include <linux/sched.h>
>> #include <asm/unistd.h>
>> @@ -68,4 +69,9 @@ syscall_get_arguments(struct task_struct *task, struct p=
t_regs *regs,
>>    }
>> }
>>=20
>> +static inline int syscall_get_arch(void)
>> +{
>> +    return AUDIT_ARCH_ARC;
>> +}
>> +
>=20
> Does ptrace (or user of this API) need a unique value per arch. Otherwise i=
nstead
> of adding the boilerplate code to all arches, they could simply define AUD=
IT_ARCH
> and common code could return it. Also the EM_xxx are not there in
> include/uapi/linux/elf.h to begin with since libc elf.h already defines th=
em.

A lot of architectures allow multiple audit_arches at runtime due to compat s=
upport and similar features, so it really does want to be a function.  The g=
oal of this patch set is to get it supported everywhere.

>> #endif
>> diff --git a/include/uapi/linux/audit.h b/include/uapi/linux/audit.h
>> index 818ae690ab79..a7149ceb5b98 100644
>> --- a/include/uapi/linux/audit.h
>> +++ b/include/uapi/linux/audit.h
>> @@ -375,6 +375,7 @@ enum {
>>=20
>> #define AUDIT_ARCH_AARCH64    (EM_AARCH64|__AUDIT_ARCH_64BIT|__AUDIT_ARCH=
_LE)
>> #define AUDIT_ARCH_ALPHA    (EM_ALPHA|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE)=

>> +#define AUDIT_ARCH_ARC        (EM_ARC)
>> #define AUDIT_ARCH_ARM        (EM_ARM|__AUDIT_ARCH_LE)
>> #define AUDIT_ARCH_ARMEB    (EM_ARM)
>> #define AUDIT_ARCH_CRIS        (EM_CRIS|__AUDIT_ARCH_LE)
>=20
> So I don't have the context of this patch (or coverletter) but what exactl=
y are we
> trying to do with this (adding LE to audit)  - what happens when an arch i=
s
> capable of either and is say built for BE ?

The primary intent is that the triple (audit_arch, syscall_nr, arg1, ..., ar=
g6) should describe what system call is being called and what its arguments a=
re.  I=E2=80=99m personally not sure what, if any, technical value there is i=
n the LE bit.

I do think it makes sense for BE and LE to have different values.=