"Christopher M. Riedl" <c...@codefail.de> writes:
> On Mon Feb 1, 2021 at 10:54 AM CST, Gabriel Paubert wrote:
>> On Mon, Feb 01, 2021 at 09:55:44AM -0600, Christopher M. Riedl wrote:
>> > On Thu Jan 28, 2021 at 4:38 AM CST, David Laight wrote:
>> > > From: Christopher M. Riedl
>> > > > Sent: 28 January 2021 04:04
>> > > >
>> > > > Reuse the "safe" implementation from signal.c except for calling
>> > > > unsafe_copy_from_user() to copy into a local buffer.
>> > > >
>> > > > Signed-off-by: Christopher M. Riedl <c...@codefail.de>
>> > > > ---
>> > > > arch/powerpc/kernel/signal.h | 33 +++++++++++++++++++++++++++++++++
>> > > > 1 file changed, 33 insertions(+)
>> > > >
>> > > > diff --git a/arch/powerpc/kernel/signal.h
>> > > > b/arch/powerpc/kernel/signal.h
>> > > > index 2559a681536e..c18402d625f1 100644
>> > > > --- a/arch/powerpc/kernel/signal.h
>> > > > +++ b/arch/powerpc/kernel/signal.h
>> > > > @@ -53,6 +53,33 @@ unsigned long copy_ckfpr_from_user(struct
>> > > > task_struct *task, void __user *from);
>> > > > &buf[i], label);\
>> > > > } while (0)
>> > > >
>> > > > +#define unsafe_copy_fpr_from_user(task, from, label) do {
>> > > > \
>> > > > + struct task_struct *__t = task;
>> > > > \
>> > > > + u64 __user *__f = (u64 __user *)from;
>> > > > \
>> > > > + u64 buf[ELF_NFPREG];
>> > > > \
>> > >
>> > > How big is that buffer?
>> > > Isn't is likely to be reasonably large compared to a reasonable
>> > > kernel stack frame.
>> > > Especially since this isn't even a leaf function.
>> > >
>> >
>> > I think Christophe answered this - I don't really have an opinion either
>> > way. What would be a 'reasonable' kernel stack frame for reference?
>>
>> See include/linux/poll.h, where the limit is of the order of 800 bytes
>> and the number of entries in an on stack array is chosen at compile time
>> (different between 32 and 64 bit for example).
>>
>> The values are used in do_sys_poll, which, with almost 1000 bytes of
>> stack footprint, appears close to the top of "make checkstack". In
>> addition do_sys_poll has to call the ->poll function of every file
>> descriptor in its table, so it is not a tail function.
>>
>> This 264 bytes array looks reasonable, but please use 'make checkstack'
>> to verify that the function's total stack usage stays within reason.
>
> Neat, looks like total usage is a bit larger but still reasonable and
> less than half of 800B:
>
> 0xc000000000017e900 __unsafe_restore_sigcontext.constprop.0 [vmlinux]:352
We warn for frames larger than 2KB on 64-bit, see FRAME_WARN in
lib/Kconfig.debug.
So 264 bytes is entirely reasonable IMHO.
cheers
