Re: [PATCH bpf-next 0/5] bpf, x86: enable EXECMEM_ROX_CACHE for BPF allocations

[Daniel Borkmann]

Hi Mike,

On 7/9/26 2:55 PM, Mike Rapoport wrote:
> On Thu, Jul 09, 2026 at 02:12:04PM +0200, Daniel Borkmann wrote:
> > On 7/9/26 10:03 AM, Mike Rapoport wrote:
> > > Gentle ping
> > >
> > > On Fri, Jun 26, 2026 at 09:42:27PM +0300, Mike Rapoport (Microsoft) wrote:
> > > > Hi,
> > > >
> > > > BPF allocations of executable memory on x86 are essentially read-only. Most
> > > > paths that call bpf_jit_alloc_exec() immediately make it ROX with
> > > > set_memory_rox().
> > > >
> > > > The code generation, at least on x86, uses separately allocated writable
> > > > buffers and then updates the actual text memory with text_poke().
> > > >
> > > > These patches do several small adjustments to how BPF allocates executable
> > > > memory and enable EXECMEM_ROX_CACHE for BPF allocations on x86.
> >
> > I think there was an open Q from Alexei wrt the sashiko reports [0] and
> > whether you could address them. Once done, please resend to bpf list.
>
> Thanks for the link!
> I'm not subscribed to bpf@ and it took me a few minutes to realize why I
> don't see this in my inbox.

Ok, np. Despite the sashiko false positives, could you pls resend against
latest bpf-next, so your set can run through the upstream BPF CI?

Thanks,
Daniel