RE: [PATCH net-next] tls: Add opt-in zerocopy mode of sendfile()

[David Laight <David.Laight@ACULAB.COM>]

From: Maxim Mikityanskiy
> Sent: 05 May 2022 13:40
> 
> On 2022-05-04 12:49, David Laight wrote:
> >>> If you declare the union on the stack in the callers, and pass by value
> >>> - is the compiler not going to be clever enough to still DDRT?
> >>
> >> Ah, OK, it should do the thing. I thought you wanted me to ditch the
> >> union altogether.
> >
> > Some architectures always pass struct/union by address.
> > Which is probably not what you had in mind.
> 
> Do you have any specific architecture in mind? I couldn't find any
> information that it happens anywhere, x86_64 ABI [1] (pages 20-21)
> aligns with my expectations, and my common sense can't explain why would
> some architectures do what you say.
> 
> In C, when the caller passes a struct as a parameter, the callee can
> freely modify it. If the compiler silently replaced it with a pointer,
> the callee would corrupt the caller's local variable, so such approach
> requires the caller to make an extra copy.

Yes, that is what happens.

> Making an extra copy on the
> stack and passing a pointer doesn't make any sense to me if you can just
> make a copy on the stack (or to a register) and call it a parameter.
> 
> If you know any specific architecture supported by Linux that passes all
> unions by a pointer, could you please point me to it? Maybe I'm missing
> something in my logic, and a real-world example will explain things, but
> at the moment it sounds unrealistic to me.

Look at any old architecture, m68k almost certainly passes all structures
by address.
i386 would - but I think the 'regparm' option includes passing small
structures by value.
I think sparc32 used to, but that might have changed in the last 30 years.

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)