[cfe-dev] [llvm-dev] RFC: Implementing -fno-delete-null-pointer-checks in clang

Manoj Gupta via cfe-dev cfe-dev at lists.llvm.org
Thu Apr 26 19:29:19 PDT 2018


Hi Tim,

> You've certainly argued against them. You haven't provided adequate
> semantics for IR without them though.

I am thinking to use the approach suggested by Eli previously.
i.e. to use a function attribute, lets say "null-pointer-is-valid".

Thanks,
Manoj

On Thu, Apr 26, 2018 at 1:58 PM Tim Northover <t.p.northover at gmail.com>
wrote:

> > In addition, It is already not easy to convince Linux Kernel
maintainers to
> > accept clang specific patches.
> > Worse performance when compared to GCC may make it even harder to push
more
> > patches.
> > (There are already many complains about clang not supporting
optimizations
> > that Linux kernel is used to.
> > As a side note: x86 maintainers deliberately broke clang support in
upstream
> > (https://lkml.org/lkml/2018/4/2/115)

> Yeah, elements of the Linux community seem actively hostile to Clang.
> We shouldn't let their hostility dictate our technical policy.

> > I hope that I have made the case for not using address spaces.

> You've certainly argued against them. You haven't provided adequate
> semantics for IR without them though.

> Cheers.

> Tim.


More information about the cfe-dev mailing list