Messages in this thread | | | Subject | Re: [PATCH v2] x86: Alias memset to __builtin_memset. | From | Joe Perches <> | Date | Tue, 24 Mar 2020 08:08:04 -0700 |
| |
On Tue, 2020-03-24 at 15:07 +0100, Clement Courbet wrote: > Recent compilers know the meaning of builtins (`memset`, > `memcpy`, ...) and can replace calls by inline code when > deemed better. For example, `memset(p, 0, 4)` will be lowered > to a four-byte zero store. > > When using -ffreestanding (this is the case e.g. building on > clang), these optimizations are disabled. This means that **all** > memsets, including those with small, constant sizes, will result > in an actual call to memset. [] > In terms of code size, this grows the clang-built kernel a > bit (+0.022%): > 440285512 vmlinux.clang.after > 440383608 vmlinux.clang.before
This shows the kernel getting smaller not larger. Is this still reversed or is this correct?
| |