Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1632561

Re: [PATCH v2] x86/KASLR: Use old ident map page table if physical randomization failed

From Baoquan He <bhe@redhat.com>
Newsgroups linux.kernel
Subject Re: [PATCH v2] x86/KASLR: Use old ident map page table if physical randomization failed
Date 2017-04-28 09:00 +0200
Message-ID <tB82K-3Nr-3@gated-at.bofh.it> (permalink)
References <tAMlA-5Ov-11@gated-at.bofh.it> <tB7zH-3BN-1@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


On 04/28/17 at 08:25am, Ingo Molnar wrote:
> 
> * Baoquan He <bhe@redhat.com> wrote:
> 
> > Dave found kdump kernel with kaslr enabled will reset to bios immediately
> > if physical randomization failed to find a new position for kernel. But
> > kernel with 'nokaslr' option works in this case.
> > 
> > The reason is kaslr will install a new page table for ident mapping,
> > while it missed to consider building ident mapping for original area
> > of kernel if kaslr failed on physical randomization.
> > 
> > This only happens in kexec/kdump kernel. Since ident mapping has been
> > built for kexec/kdump in 1st kernel for the whole memory by calling
> > init_pgtable(). Here if physical randomizaiton failed, it won't build
> > ident mapping for the original area of kernel but change to new page
> > table '_pgtable'. Then kernel will reset to bios immediately caused by
> > no ident mapping.
> > 
> > While normal kernel won't be impacted because it comes here via
> > startup_32() and cr3 will be _pgtable already. In startup_32() ident
> > mapping is built for 0~4G area. In kaslr We just append to the existing
> > area instead of entirely overwriting it for on-demand ident mapping
> > building. So ident mapping for the original area of kernel is still
> > there.
> > 
> > So for fixing it,  We just switch to the new ident mapping page table
> > when physical kaslr succeeds. Otherwise we keep the old page table
> > unchanged just like nokaslr does.
> > 
> > Signed-off-by: Baoquan He <bhe@redhat.com>
> > Signed-off-by: Dave Young <dyoung@redhat.com>
> > Acked-by: Kees Cook <keescook@chromium.org>
> 
> That's not a valid signoff chain. I made it:
> 
>  Reported-by: Dave Young <dyoung@redhat.com>
>  Signed-off-by: Baoquan He <bhe@redhat.com>
>  Acked-by: Dave Young <dyoung@redhat.com>
>  Acked-by: Kees Cook <keescook@chromium.org>
> 
> (Let me know if that's not the proper authorship chain.)

Yeah, this is fine, thanks. Next time I will notice this.

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

[PATCH v2] x86/KASLR: Use old ident map page table if physical randomization failed Baoquan He <bhe@redhat.com> - 2017-04-27 09:50 +0200
  Re: [PATCH v2] x86/KASLR: Use old ident map page table if physical  randomization failed Ingo Molnar <mingo@kernel.org> - 2017-04-28 08:30 +0200
    Re: [PATCH v2] x86/KASLR: Use old ident map page table if physical  randomization failed Baoquan He <bhe@redhat.com> - 2017-04-28 09:00 +0200
  [tip:x86/boot] x86/KASLR: Fix kexec kernel boot crash when KASLR  randomization fails tip-bot for Baoquan He <tipbot@zytor.com> - 2017-04-28 10:50 +0200

csiph-web