Applied, thank you
On Thu, Dec 22, 2022 at 3:30 PM Sören Tempel wrote:
>
> PING.
>
> Any love for good old ed(1)?
>
> Sören Tempel wrote:
> > Pinging again as this is still unfixed and the proposed fix is rather
> > trivial.
> >
> > Sören Tempel wrote:
> > > Ping.
> > >
> > >
From: Sören Tempel
> Sent: 22 December 2022 14:23
>
> PING.
>
> Any love for good old ed(1)?
Some versions of glibc will also do 'not strictly increasing'
memcpy() on some x86 cpu.
Just copying the last 4/8 bytes first and then doing a forwards
copy is enough to break things.
David
>
PING.
Any love for good old ed(1)?
Sören Tempel wrote:
> Pinging again as this is still unfixed and the proposed fix is rather trivial.
>
> Sören Tempel wrote:
> > Ping.
> >
> > soe...@soeren-tempel.net wrote:
> > > From: Sören Tempel
> > >
> > > The memcpy invocations in the subCommand
Pinging again as this is still unfixed and the proposed fix is rather trivial.
Sören Tempel wrote:
> Ping.
>
> soe...@soeren-tempel.net wrote:
> > From: Sören Tempel
> >
> > The memcpy invocations in the subCommand function, modified by this
> > commit, previously used memcpy with overlapping
Ping.
soe...@soeren-tempel.net wrote:
> From: Sören Tempel
>
> The memcpy invocations in the subCommand function, modified by this
> commit, previously used memcpy with overlapping memory regions. This is
> undefined behavior. On Alpine Linux, it causes BusyBox ed to crash since
> we compile