Re: [PATCH v1] seccomp: simplify secure_computing()

2019-10-11 Thread Christian Brauner
On Thu, Oct 10, 2019 at 02:53:24PM -0700, Kees Cook wrote: > On Tue, Sep 24, 2019 at 08:44:20AM +0200, Christian Brauner wrote: > > Afaict, the struct seccomp_data argument to secure_computing() is unused > > by all current callers. So let's remove it. > > The argument was added in [1]. It was adde

Re: [PATCH v1] seccomp: simplify secure_computing()

2019-10-10 Thread Kees Cook
On Tue, Sep 24, 2019 at 08:44:20AM +0200, Christian Brauner wrote: > Afaict, the struct seccomp_data argument to secure_computing() is unused > by all current callers. So let's remove it. > The argument was added in [1]. It was added because having the arch > supply the syscall arguments used to be

Re: [PATCH v1] seccomp: simplify secure_computing()

2019-09-24 Thread Andy Lutomirski
On Mon, Sep 23, 2019 at 11:44 PM Christian Brauner wrote: > > Afaict, the struct seccomp_data argument to secure_computing() is unused > by all current callers. So let's remove it. > The argument was added in [1]. It was added because having the arch > supply the syscall arguments used to be faste

Re: [PATCH v1] seccomp: simplify secure_computing()

2019-09-24 Thread Borislav Petkov
On Tue, Sep 24, 2019 at 08:44:20AM +0200, Christian Brauner wrote: > Afaict, the struct seccomp_data argument to secure_computing() is unused > by all current callers. So let's remove it. > The argument was added in [1]. It was added because having the arch > supply the syscall arguments used to be

[PATCH v1] seccomp: simplify secure_computing()

2019-09-23 Thread Christian Brauner
Afaict, the struct seccomp_data argument to secure_computing() is unused by all current callers. So let's remove it. The argument was added in [1]. It was added because having the arch supply the syscall arguments used to be faster than having it done by secure_computing() (cf. Andy's comment in [2