Re: [PATCH] x86/asm/entry: (Re-)rename __NR_entry_INT80_compat_max to __NR_syscall_compat_max

[Ingo Molnar <mingo@kernel.org>]

* Brian Gerst <brgerst@gmail.com> wrote:

> On Mon, Jun 8, 2015 at 3:25 PM, Ingo Molnar <mingo@kernel.org> wrote:
> >
> > * Brian Gerst <brgerst@gmail.com> wrote:
> >
> >> > diff --git a/arch/x86/entry/syscall_32.c b/arch/x86/entry/syscall_32.c
> >> > index 3777189..e398d03 100644
> >> > --- a/arch/x86/entry/syscall_32.c
> >> > +++ b/arch/x86/entry/syscall_32.c
> >> > @@ -10,7 +10,7 @@
> >> >  #else
> >> >  #define SYM(sym, compat) sym
> >> >  #define ia32_sys_call_table sys_call_table
> >> > -#define __NR_ia32_syscall_max __NR_syscall_max
> >> > +#define __NR_entry_INT80_compat_max __NR_syscall_max
> >> >  #endif
> >>
> >> This change doesn't look intended.  __NR_ia32_syscall_max is used by
> >> all compat entry points, not just INT80,
> >
> > Indeed!
> >
> > How about the patch below?
> >
> > Thanks,
> >
> >         Ingo
> >
> > ================>
> > From 42b9d9a463bfb7252ede4eac24e50a4f013c6717 Mon Sep 17 00:00:00 2001
> > From: Ingo Molnar <mingo@kernel.org>
> > Date: Mon, 8 Jun 2015 21:20:26 +0200
> > Subject: [PATCH] x86/asm/entry: (Re-)rename __NR_entry_INT80_compat_max to __NR_syscall_compat_max
> >
> > Brian Gerst noticed that I did a weird rename in the following commit:
> >
> >    b2502b418e63 ("x86/asm/entry: Untangle 'system_call' into two entry points: entry_SYSCALL_64 and entry_INT80_32")
> >
> > which renamed __NR_ia32_syscall_max to __NR_entry_INT80_compat_max.
> >
> > Now the original name was a misnomer, but the new one is a misnomer as well,
> > as all compat syscall ABIs (such as x32) share the system call table, not just
> > INT80 based ones.
> 
> Not quite true.  X32 is really an extension of the 64-bit ABI to support the 
> ILP32 data model.  It uses the 64-bit SYSCALL entry point and table, but with 
> different syscall numbers for the syscalls that need 32-bit pointer handling in 
> data structures.
> 
> So, __NR_syscall_compat_max is only for the 32-bit ABI compat syscall entries 
> (int80, sysenter, and syscall from 32-bit mode), and not X32.

Indeed, and I knew this...

Fixed the brainfart in the changelog.

Thanks,

	Ingo