The only differenence betweeen the compat exec* syscalls and their
native versions is that compat_ptr sign extension, and the fact that
the pointer arithmetics for the two dimensional arrays needs to use
the compat pointer size. Instead of the compat wrappers and the
struct user_arg_ptr machinery
On Mon, Jun 15, 2020 at 2:47 PM Arnd Bergmann wrote:
>
> On Mon, Jun 15, 2020 at 4:48 PM Brian Gerst wrote:
> > On Mon, Jun 15, 2020 at 10:13 AM Christoph Hellwig wrote:
> > > On Mon, Jun 15, 2020 at 03:31:35PM +0200, Arnd Bergmann wrote:
>
> > >
> > > I'd rather keep it in common code as that a
On Mon, Jun 15, 2020 at 4:48 PM Brian Gerst wrote:
> On Mon, Jun 15, 2020 at 10:13 AM Christoph Hellwig wrote:
> > On Mon, Jun 15, 2020 at 03:31:35PM +0200, Arnd Bergmann wrote:
> >
> > I'd rather keep it in common code as that allows all the low-level
> > exec stuff to be marked static, and avo
On Mon, Jun 15, 2020 at 11:33:49AM -0400, Brian Gerst wrote:
> If you move those aliases above all the __SYSCALL_* defines it will
> work, since that will get the forward declaration too. This would be
> the simplest workaround.
That compiles and also passes my exaustive x32 tests (chroot + ls -l
On Mon, Jun 15, 2020 at 11:10 AM Christoph Hellwig wrote:
>
> On Mon, Jun 15, 2020 at 04:46:15PM +0200, Arnd Bergmann wrote:
> > How about this one:
> >
> > diff --git a/arch/x86/entry/syscall_x32.c b/arch/x86/entry/syscall_x32.c
> > index 3d8d70d3896c..0ce15807cf54 100644
> > --- a/arch/x86/entry
On Mon, Jun 15, 2020 at 04:46:15PM +0200, Arnd Bergmann wrote:
> How about this one:
>
> diff --git a/arch/x86/entry/syscall_x32.c b/arch/x86/entry/syscall_x32.c
> index 3d8d70d3896c..0ce15807cf54 100644
> --- a/arch/x86/entry/syscall_x32.c
> +++ b/arch/x86/entry/syscall_x32.c
> @@ -16,6 +16,9 @@
On Mon, Jun 15, 2020 at 10:13 AM Christoph Hellwig wrote:
>
> On Mon, Jun 15, 2020 at 03:31:35PM +0200, Arnd Bergmann wrote:
> > > #ifdef CONFIG_COMPAT
> > > - if (unlikely(argv.is_compat)) {
> > > + if (in_compat_syscall()) {
> > > + const compat_uptr_t __user *compat_a
On Mon, Jun 15, 2020 at 4:43 PM Christoph Hellwig wrote:
>
> On Mon, Jun 15, 2020 at 04:40:28PM +0200, Arnd Bergmann wrote:
> > > ld: arch/x86/entry/syscall_x32.o:(.rodata+0x1040): undefined reference to
> > > `__x32_sys_execve'
> > > ld: arch/x86/entry/syscall_x32.o:(.rodata+0x1108): undefined re
On Mon, Jun 15, 2020 at 4:12 PM Christoph Hellwig wrote:
> On Mon, Jun 15, 2020 at 03:31:35PM +0200, Arnd Bergmann wrote:
>
> > I don't really understand
> > the comment, why can't this just use this?
>
> That errors out with:
>
> ld: arch/x86/entry/syscall_x32.o:(.rodata+0x1040): undefined refer
On Mon, Jun 15, 2020 at 04:40:28PM +0200, Arnd Bergmann wrote:
> > ld: arch/x86/entry/syscall_x32.o:(.rodata+0x1040): undefined reference to
> > `__x32_sys_execve'
> > ld: arch/x86/entry/syscall_x32.o:(.rodata+0x1108): undefined reference to
> > `__x32_sys_execveat'
> > make: *** [Makefile:1139: vm
On Mon, Jun 15, 2020 at 03:31:35PM +0200, Arnd Bergmann wrote:
> > #ifdef CONFIG_COMPAT
> > - if (unlikely(argv.is_compat)) {
> > + if (in_compat_syscall()) {
> > + const compat_uptr_t __user *compat_argv =
> > + compat_ptr((unsigned long)argv);
> >
On Mon, Jun 15, 2020 at 3:00 PM Christoph Hellwig wrote:
>
> The only differenence betweeen the compat exec* syscalls and their
> native versions is that compat_ptr sign extension, and the fact that
> the pointer arithmetics for the two dimensional arrays needs to use
> the compat pointer size. I
The only differenence betweeen the compat exec* syscalls and their
native versions is that compat_ptr sign extension, and the fact that
the pointer arithmetics for the two dimensional arrays needs to use
the compat pointer size. Instead of the compat wrappers and the
struct user_arg_ptr machinery
13 matches
Mail list logo
