On Sat, 2021-10-02 at 18:53 +0800, Huacai Chen wrote:
> Hi, Ruoyao,
>
> On Thu, Sep 30, 2021 at 11:43 PM Xi Ruoyao <xry111@xxxxxxxxxxxxxxxx>
> wrote:
> >
> > On Mon, 2021-09-27 at 14:42 +0800, Huacai Chen wrote:
> > > diff --git a/arch/loongarch/vdso/gen_vdso_offsets.sh
> > > b/arch/loongarch/vdso/gen_vdso_offsets.sh
> > > new file mode 100755
> > > index 000000000000..7da255fea213
> > > --- /dev/null
> > > +++ b/arch/loongarch/vdso/gen_vdso_offsets.sh
> > > @@ -0,0 +1,14 @@
> > > +#!/bin/sh
> > > +# SPDX-License-Identifier: GPL-2.0
> > > +
> > > +#
> > > +# Derived from RISC-V and ARM64:
> > > +# Author: Will Deacon <will.deacon@xxxxxxx>
> > > +#
> > > +# Match symbols in the DSO that look like VDSO_*; produce a
> > > header
> > > file
> > > +# of constant offsets into the shared object.
> > > +#
> > > +
> > > +LC_ALL=C
> >
> > I'm wondering whether this line is really useful... There is no
> > "export"
> > here so the variable won't be passed to the environment of the sed
> > command below.
> Have you encountered some problems with this? It just works for me,
> and both ARM64 and RISCV are the same.
No problems, and I also seen those in ARM64 & RISCV. But AFAIK this
line really does nothing and can be removed.
If LC_ALL=C is really necessary for the sed to operate correctly, it
should be "exported" as
LC_ALL=C
export LC_ALL
("export LC_ALL=C" will work under bash, but it's not POSIX.)
Or, explicitly pass LC_ALL to sed with:
LC_ALL=C sed ...
--
Xi Ruoyao <xry111@xxxxxxxxxxxxxxxx>
School of Aerospace Science and Technology, Xidian University
