linux-kernel - Re: [patch 51/60] x86/mm: Allow flushing for future ASID switches

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Mon, 4 Dec 2017 14:34:53 -0800
From:   Dave Hansen <dave.hansen@...el.com>
To:     Andy Lutomirski <luto@...nel.org>,
        Thomas Gleixner <tglx@...utronix.de>
Cc:     LKML <linux-kernel@...r.kernel.org>, X86 ML <x86@...nel.org>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Peter Zijlstra <peterz@...radead.org>,
        Borislav Petkov <bpetkov@...e.de>,
        Greg KH <gregkh@...uxfoundation.org>,
        Kees Cook <keescook@...gle.com>,
        Hugh Dickins <hughd@...gle.com>,
        Brian Gerst <brgerst@...il.com>,
        Josh Poimboeuf <jpoimboe@...hat.com>,
        Denys Vlasenko <dvlasenk@...hat.com>,
        Rik van Riel <riel@...hat.com>,
        Boris Ostrovsky <boris.ostrovsky@...cle.com>,
        Juergen Gross <jgross@...e.com>,
        David Laight <David.Laight@...lab.com>,
        Eduardo Valentin <eduval@...zon.com>, aliguori@...zon.com,
        Will Deacon <will.deacon@....com>,
        Daniel Gruss <daniel.gruss@...k.tugraz.at>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Ingo Molnar <mingo@...nel.org>, michael.schwarz@...k.tugraz.at,
        Borislav Petkov <bp@...en8.de>, moritz.lipp@...k.tugraz.at,
        richard.fellner@...dent.tugraz.at
Subject: Re: [patch 51/60] x86/mm: Allow flushing for future ASID switches

On 12/04/2017 02:22 PM, Andy Lutomirski wrote:
>> +
>> +       this_cpu_write(cpu_tlbstate.invalidate_other, true);
> 
> Why do we need this extra variable instead of just looping over all
> other ASIDs and invalidating them?  It would be something like:
> 
>         for (i = 1; i < TLB_NR_DYN_ASIDS; i++) {
>                 if (i != this_cpu_read(cpu_tlbstate.loaded_mm_asid))
>                        this_cpu_write(cpu_tlbstate.ctxs[i].ctx_id, 0);
>         }

We have loops like this:

	for (addr = start; addr < end; addr += PAGE_SIZE)
		flush_tlb_single();

Where flush_tlb_single() does a invalidate_pcid_other().  So, inlining
flush_tlb_single() rougly looks like:

	for (addr = start; addr < end; addr += PAGE_SIZE) {
		invlpg;
		for (i = 1; i < TLB_NR_DYN_ASIDS; i++) {
			this_cpu_write(cpu_tlbstate.ctxs[i].ctx_id, 0);
	}

or, with a "invalidate_other" variable:

	for (addr = start; addr < end; addr += PAGE_SIZE) {
		invlpg;
		this_cpu_write(cpu_tlbstate.ctxs.invalidate_other, 1);
	}

The double-for-loop looks a bit wasteful to me.


>>  static inline void __flush_tlb_one(unsigned long addr)
>>  {
>>         count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ONE);
>>         __flush_tlb_single(addr);
>> +       /*
>> +        * Invalidate other address spaces inaccessible to single-page
>> +        * invalidation:
>> +        */
> 
> Ugh.  If I'm reading this right, __flush_tlb_single() means "flush one
> user address" and __flush_tlb_one() means "flush one kernel address".
> That's, um, not exactly obvious.  Could this be at least commented
> better?

That sounds sane, but let me take a look at it.

Didn't Peter have some patches to do some of that rename?