Re: [PATCHv2 bpf-next 2/8] bpf: Add user memory kfuncs

[Jiri Olsa <olsajiri@gmail.com>]

On Tue, Nov 03, 2026 at 06:05:46PM +0000, Alexei Starovoitov wrote:
> On Mon, Nov 02, 2026 at 11:59 PM Jiri Olsa <jolsa@kernel.org> wrote:
> > +#ifdef CONFIG_IA32_EMULATION
> > +static unsigned long bpf_user_vm_mmap(unsigned long len, unsigned long prot)
> > +{
> > +	struct thread_info *ti = current_thread_info();
> > +	bool compat = test_thread_flag(TIF_ADDR32);
> > +	unsigned long addr;
> > +
> > +	/*
> > +	 * We are not in a syscall, but x86 get_unmapped_area() paths
> > +	 * ignore TIF_ADDR32 and rely on in_32bit_syscall().
> > +	 */
> > +	if (compat)
> > +		ti->status |= TS_COMPAT;
> > +	addr = vm_mmap(NULL, 0, len, prot, MAP_PRIVATE | MAP_ANONYMOUS, 0);
> > +	if (compat)
> > +		ti->status &= ~TS_COMPAT;
> 
> Do you have users with ia32 tasks ?

nope

> Without this code bpf_user_anon_mmap() returns -ENOMEM in an ia32 task.
> Nothing breaks.
> 
> arch_uprobe_get_xol_area() calls it a HACK, and that one is in arch/x86.
> kernel/bpf/helpers.c has no arch #ifdef today. Let's keep it this way.
> 
> Drop it and say in the comment that ia32 tasks are not supported.
> Pls don't add code just because a bot said so.

ok, will drop it

jirka