Re: [PATCH 0/2] execve scalability issues, part 1

[Matthew Wilcox]

On Mon, Aug 21, 2023 at 10:28:27PM +0200, Mateusz Guzik wrote:
> > While with the patch these allocations remain a significant problem,
> > the primary bottleneck shifts to:
> > 
> >     __pv_queued_spin_lock_slowpath+1
> >     _raw_spin_lock_irqsave+57
> >     folio_lruvec_lock_irqsave+91
> >     release_pages+590
> >     tlb_batch_pages_flush+61
> >     tlb_finish_mmu+101
> >     exit_mmap+327
> >     __mmput+61
> >     begin_new_exec+1245
> >     load_elf_binary+712
> >     bprm_execve+644
> >     do_execveat_common.isra.0+429
> >     __x64_sys_execve+50
> >     do_syscall_64+46
> >     entry_SYSCALL_64_after_hwframe+110
> 
> I intend to do more work on the area to mostly sort it out, but I would
> not mind if someone else took the hammer to folio. :)

Funny you should ask ... these patches are from ~3 weeks ago.  They may
or may not apply to a current tree, but I'm working on this area.