From: Pavel Machek <pavel@ucw.cz>
To: joe.korty@concurrent-rt.com
Cc: Andy Lutomirski <luto@amacapital.net>,
int-list-linux-mm@kvack.orglinux-mm@kvack.org
Subject: Re: [PATCH 00/31 v2] PTI support for x86_32
Date: Fri, 16 Feb 2018 15:34:01 +0100 [thread overview]
Message-ID: <20180216143401.GA3439@amd> (raw)
In-Reply-To: <20180215034444.GA18849@zipoli.concurrent-rt.com>
[-- Attachment #1: Type: text/plain, Size: 1465 bytes --]
On Wed 2018-02-14 22:44:44, joe.korty@concurrent-rt.com wrote:
> On Wed, Feb 14, 2018 at 11:43:42AM +0100, Pavel Machek wrote:
> > We have just found out that majority of 64-bit machines are broken in
> > rather fundamental ways (Spectre) and Intel does not even look
> > interested in fixing that (because it would make them look bad on
> > benchmarks).
> >
> > Even when the Spectre bug is mitigated... this looks like can of worms
> > that can not be closed.
> >
> > OTOH -- we do know that there are non-broken machines out there,
> > unfortunately they are mostly 32-bit :-). Removing support for
> > majority of working machines may not be good idea...
> >
> > [And I really hope future CPUs get at least option to treat cache miss
> > as a side-effect -- thus disalowed during speculation -- and probably
> > option to turn off speculation altogether. AFAICT, it should "only"
> > result in 50% slowdown -- or that was result in some riscv
> > presentation.]
>
> Or, future CPU designs introduce shadow caches and shadow
> TLBs which only speculation loads and sees and which
> become real only if and whend the resultant speculative
> calculations become real.
Yes, that could help.
But there's still sidechannel in the RAM: it has row buffer
or something like that.
Pavel
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]
next prev parent reply other threads:[~2018-02-16 14:34 UTC|newest]
Thread overview: 93+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-02-09 9:25 [PATCH 00/31 v2] PTI support for x86_32 Joerg Roedel
2018-02-09 9:25 ` [PATCH 01/31] x86/asm-offsets: Move TSS_sp0 and TSS_sp1 to asm-offsets.c Joerg Roedel
2018-02-09 9:25 ` [PATCH 02/31] x86/entry/32: Rename TSS_sysenter_sp0 to TSS_entry_stack Joerg Roedel
2018-02-09 9:25 ` [PATCH 03/31] x86/entry/32: Load task stack from x86_tss.sp1 in SYSENTER handler Joerg Roedel
2018-02-09 9:25 ` [PATCH 04/31] x86/entry/32: Put ESPFIX code into a macro Joerg Roedel
2018-02-09 9:25 ` [PATCH 05/31] x86/entry/32: Unshare NMI return path Joerg Roedel
2018-02-09 9:25 ` [PATCH 06/31] x86/entry/32: Split off return-to-kernel path Joerg Roedel
2018-02-09 9:25 ` [PATCH 07/31] x86/entry/32: Restore segments before int registers Joerg Roedel
2018-02-09 9:25 ` [PATCH 08/31] x86/entry/32: Enter the kernel via trampoline stack Joerg Roedel
2018-02-09 9:25 ` [PATCH 09/31] x86/entry/32: Leave " Joerg Roedel
2018-02-09 17:05 ` Linus Torvalds
2018-02-09 17:17 ` Denys Vlasenko
2018-02-10 15:26 ` David Laight
2018-02-10 20:18 ` Linus Torvalds
2018-02-09 17:43 ` Andy Lutomirski
2018-02-09 19:06 ` Joerg Roedel
2018-02-09 19:02 ` Joerg Roedel
2018-02-09 19:17 ` Linus Torvalds
2018-02-09 19:25 ` Joerg Roedel
2018-02-09 19:48 ` Linus Torvalds
2018-02-10 15:41 ` David Laight
2018-02-09 19:30 ` Denys Vlasenko
2018-02-09 9:25 ` [PATCH 10/31] x86/entry/32: Introduce SAVE_ALL_NMI and RESTORE_ALL_NMI Joerg Roedel
2018-02-09 9:25 ` [PATCH 11/31] x86/entry/32: Add PTI cr3 switches to NMI handler code Joerg Roedel
2018-02-09 9:25 ` [PATCH 12/31] x86/entry/32: Add PTI cr3 switch to non-NMI entry/exit points Joerg Roedel
2018-02-27 19:18 ` Waiman Long
2018-03-01 12:03 ` Joerg Roedel
2018-03-01 13:34 ` Joerg Roedel
2018-03-01 14:33 ` Waiman Long
2018-03-01 16:50 ` Joerg Roedel
2018-03-01 18:24 ` Brian Gerst
2018-03-01 18:36 ` Dave Hansen
2018-03-01 18:38 ` Linus Torvalds
2018-03-02 9:10 ` Joerg Roedel
2018-03-16 20:55 ` Andy Lutomirski
2018-03-02 9:07 ` Joerg Roedel
2018-02-09 9:25 ` [PATCH 13/31] x86/entry/32: Handle Entry from Kernel-Mode on Entry-Stack Joerg Roedel
2018-02-09 9:25 ` [PATCH 14/31] x86/pgtable/pae: Unshare kernel PMDs when PTI is enabled Joerg Roedel
2018-02-09 9:25 ` [PATCH 15/31] x86/pgtable/32: Allocate 8k page-tables " Joerg Roedel
2018-02-09 9:25 ` [PATCH 16/31] x86/pgtable: Move pgdp kernel/user conversion functions to pgtable.h Joerg Roedel
2018-02-09 9:25 ` [PATCH 17/31] x86/pgtable: Move pti_set_user_pgd() " Joerg Roedel
2018-02-09 9:25 ` [PATCH 18/31] x86/pgtable: Move two more functions from pgtable_64.h " Joerg Roedel
2018-02-09 9:25 ` [PATCH 19/31] x86/mm/pae: Populate valid user PGD entries Joerg Roedel
2018-02-14 9:45 ` Juergen Gross
2018-02-14 10:00 ` Joerg Roedel
2018-02-09 9:25 ` [PATCH 20/31] x86/mm/pae: Populate the user page-table with user pgd's Joerg Roedel
2018-02-09 17:48 ` Andy Lutomirski
2018-02-09 19:09 ` Joerg Roedel
2018-02-09 9:25 ` [PATCH 21/31] x86/mm/legacy: " Joerg Roedel
2018-02-09 9:25 ` [PATCH 22/31] x86/mm/pti: Add an overflow check to pti_clone_pmds() Joerg Roedel
2018-02-09 9:25 ` [PATCH 23/31] x86/mm/pti: Define X86_CR3_PTI_PCID_USER_BIT on x86_32 Joerg Roedel
2018-02-09 17:42 ` Andy Lutomirski
2018-02-09 9:25 ` [PATCH 24/31] x86/mm/pti: Clone CPU_ENTRY_AREA on PMD level " Joerg Roedel
2018-02-09 9:25 ` [PATCH 25/31] x86/mm/dump_pagetables: Define INIT_PGD Joerg Roedel
2018-02-09 9:25 ` [PATCH 26/31] x86/pgtable/pae: Use separate kernel PMDs for user page-table Joerg Roedel
2018-02-09 9:25 ` [PATCH 27/31] x86/ldt: Reserve address-space range on 32 bit for the LDT Joerg Roedel
2018-02-09 9:25 ` [PATCH 28/31] x86/ldt: Define LDT_END_ADDR Joerg Roedel
2018-02-09 9:25 ` [PATCH 29/31] x86/ldt: Split out sanity check in map_ldt_struct() Joerg Roedel
2018-02-09 9:25 ` [PATCH 30/31] x86/ldt: Enable LDT user-mapping for PAE Joerg Roedel
2018-02-09 9:25 ` [PATCH 31/31] x86/pti: Allow CONFIG_PAGE_TABLE_ISOLATION for x86_32 Joerg Roedel
2018-02-09 12:11 ` [PATCH 00/31 v2] PTI support " Juergen Gross
2018-02-09 13:35 ` Joerg Roedel
2018-02-09 13:54 ` Andrew Cooper
2018-02-09 17:47 ` Andy Lutomirski
2018-02-09 19:11 ` Joerg Roedel
2018-02-10 9:15 ` Adam Borowski
2018-02-10 20:22 ` Linus Torvalds
2018-02-11 10:59 ` Adam Borowski
2018-02-11 17:40 ` Mark D Rustad
2018-02-11 19:42 ` Andy Lutomirski
2018-02-11 20:14 ` Linus Torvalds
2018-02-11 22:12 ` James Bottomley
2018-02-11 22:30 ` Andy Lutomirski
2018-02-11 23:47 ` James Bottomley
2018-02-11 22:34 ` Pavel Machek
2018-02-11 23:25 ` Alan Cox
2018-02-12 10:16 ` Anders Larsen
2018-02-14 10:43 ` Pavel Machek
2018-02-15 3:44 ` joe.korty
2018-02-16 14:34 ` Pavel Machek [this message]
2018-02-13 8:54 ` Greg KH
2018-02-13 17:25 ` Linus Torvalds
2018-02-14 8:54 ` Greg KH
2018-02-21 10:26 ` Lorenzo Colitti
2018-02-21 16:59 ` Arnd Bergmann
2018-02-22 11:10 ` Greg KH
2018-02-22 11:18 ` Arnd Bergmann
2018-03-06 15:39 ` Jason A. Donenfeld
2018-02-11 19:13 ` Ingo Molnar
2018-02-12 14:51 ` Joerg Roedel
2018-02-09 21:09 ` Pavel Machek
2018-02-09 21:11 ` Linus Torvalds
2018-02-09 21:28 ` Andrew Cooper
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20180216143401.GA3439@amd \
--to=pavel@ucw.cz \
--cc=int-list-linux-mm@kvack.orglinux-mm \
--cc=joe.korty@concurrent-rt.com \
--cc=luto@amacapital.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).