L!TF Bulletin #8: The final bits

L!TF Bulletin #8: The final bits

[Thomas Gleixner]

[-- Attachment #1: Type: text/plain, Size: 655 bytes --]

Hi!

I've merged the SMT fix and parts of Andis PTE inversion series:

958f79b9ee55 ("x86/mm/pat: Make set_memory_np() L1TF safe")
0768f91530ff ("x86/speculation/l1tf: Make pmd/pud_mknotpresent() invert")
f22cc87f6c1f ("x86/speculation/l1tf: Invert all not present mappings")
bc2d8d262cba ("cpu/hotplug: Fix SMT supported evaluation")

The branches are updated and pushed out. The resulting bundles are
attached.

I'm leaving for the summer camp now, but I'll have a look once per day into
the list and if no reports come in until the weekend I'll prepare the pull
request for Linus from what we have now.

Please give it a proper testing.

Thanks,

	tglx

[-- Attachment #2: Type: application/octet-stream, Size: 110637 bytes --]

[-- Attachment #3: Type: application/octet-stream, Size: 106239 bytes --]

[-- Attachment #4: Type: application/octet-stream, Size: 112897 bytes --]

[MODERATED] Re: L!TF Bulletin #8: The final bits

[David Woodhouse]

On Wed, 2018-08-08 at 09:37 +0200, speck for Thomas Gleixner wrote:
> Hi!
> 
> I've merged the SMT fix and parts of Andis PTE inversion series:
> 
> 958f79b9ee55 ("x86/mm/pat: Make set_memory_np() L1TF safe")
> 0768f91530ff ("x86/speculation/l1tf: Make pmd/pud_mknotpresent() invert")
> f22cc87f6c1f ("x86/speculation/l1tf: Invert all not present mappings")
> bc2d8d262cba ("cpu/hotplug: Fix SMT supported evaluation")
> 
> The branches are updated and pushed out. The resulting bundles are
> attached.
> 
> I'm leaving for the summer camp now, but I'll have a look once per day into
> the list and if no reports come in until the weekend I'll prepare the pull
> request for Linus from what we have now.
> 
> Please give it a proper testing.

I've updated the 4.9 branch. I also have this to add on top, with
slight tweaks for each of the stable branches. Want me to just push
those?

From 9d9051c006c988e06c64c808e8133ba674cc2ec1 Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw@amazon.co.uk>
Date: Wed, 8 Aug 2018 11:00:16 +0100
Subject: [PATCH] tools headers: Synchronise x86 cpufeatures.h for L1TF additions

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
---
 tools/arch/x86/include/asm/cpufeatures.h | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/tools/arch/x86/include/asm/cpufeatures.h b/tools/arch/x86/include/asm/cpufeatures.h
index 5701f5c..64aaa3f 100644
--- a/tools/arch/x86/include/asm/cpufeatures.h
+++ b/tools/arch/x86/include/asm/cpufeatures.h
@@ -219,6 +219,7 @@
 #define X86_FEATURE_IBPB		( 7*32+26) /* Indirect Branch Prediction Barrier */
 #define X86_FEATURE_STIBP		( 7*32+27) /* Single Thread Indirect Branch Predictors */
 #define X86_FEATURE_ZEN			( 7*32+28) /* "" CPU is AMD family 0x17 (Zen) */
+#define X86_FEATURE_L1TF_PTEINV		( 7*32+29) /* "" L1TF workaround PTE inversion */
 
 /* Virtualization flags: Linux defined, word 8 */
 #define X86_FEATURE_TPR_SHADOW		( 8*32+ 0) /* Intel TPR Shadow */
@@ -341,6 +342,7 @@
 #define X86_FEATURE_PCONFIG		(18*32+18) /* Intel PCONFIG */
 #define X86_FEATURE_SPEC_CTRL		(18*32+26) /* "" Speculation Control (IBRS + IBPB) */
 #define X86_FEATURE_INTEL_STIBP		(18*32+27) /* "" Single Thread Indirect Branch Predictors */
+#define X86_FEATURE_FLUSH_L1D		(18*32+28) /* Flush L1D cache */
 #define X86_FEATURE_ARCH_CAPABILITIES	(18*32+29) /* IA32_ARCH_CAPABILITIES MSR (Intel) */
 #define X86_FEATURE_SPEC_CTRL_SSBD	(18*32+31) /* "" Speculative Store Bypass Disable */
 
@@ -373,5 +375,6 @@
 #define X86_BUG_SPECTRE_V1		X86_BUG(15) /* CPU is affected by Spectre variant 1 attack with conditional branches */
 #define X86_BUG_SPECTRE_V2		X86_BUG(16) /* CPU is affected by Spectre variant 2 attack with indirect branches */
 #define X86_BUG_SPEC_STORE_BYPASS	X86_BUG(17) /* CPU is affected by speculative store bypass attack */
+#define X86_BUG_L1TF			X86_BUG(18) /* CPU is affected by L1 Terminal Fault */
 
 #endif /* _ASM_X86_CPUFEATURES_H */
-- 
2.7.4

[MODERATED] Re: L!TF Bulletin #8: The final bits

[Greg KH]

On Wed, Aug 08, 2018 at 11:34:40AM +0100, speck for David Woodhouse wrote:
> 
> 
> On Wed, 2018-08-08 at 09:37 +0200, speck for Thomas Gleixner wrote:
> > Hi!
> > 
> > I've merged the SMT fix and parts of Andis PTE inversion series:
> > 
> > 958f79b9ee55 ("x86/mm/pat: Make set_memory_np() L1TF safe")
> > 0768f91530ff ("x86/speculation/l1tf: Make pmd/pud_mknotpresent() invert")
> > f22cc87f6c1f ("x86/speculation/l1tf: Invert all not present mappings")
> > bc2d8d262cba ("cpu/hotplug: Fix SMT supported evaluation")
> > 
> > The branches are updated and pushed out. The resulting bundles are
> > attached.
> > 
> > I'm leaving for the summer camp now, but I'll have a look once per day into
> > the list and if no reports come in until the weekend I'll prepare the pull
> > request for Linus from what we have now.
> > 
> > Please give it a proper testing.
> 
> I've updated the 4.9 branch. I also have this to add on top, with
> slight tweaks for each of the stable branches. Want me to just push
> those?

I have some initial patches that Guenter made up for 4.4 that ignores
the KVM stuff, as odds are those will be good to get merged.  Should I
post them here or push them to the git tree?  I haven't checked to see
if I have write permissions...

Or should I just not worry about it and queue them up in my stable trees
when the embargo lifts?


thanks,

greg k-h

Re: L!TF Bulletin #8: The final bits

[Thomas Gleixner]

On Wed, 8 Aug 2018, speck for Greg KH wrote:
> On Wed, Aug 08, 2018 at 11:34:40AM +0100, speck for David Woodhouse wrote:
> > 
> > 
> > On Wed, 2018-08-08 at 09:37 +0200, speck for Thomas Gleixner wrote:
> > > Hi!
> > > 
> > > I've merged the SMT fix and parts of Andis PTE inversion series:
> > > 
> > > 958f79b9ee55 ("x86/mm/pat: Make set_memory_np() L1TF safe")
> > > 0768f91530ff ("x86/speculation/l1tf: Make pmd/pud_mknotpresent() invert")
> > > f22cc87f6c1f ("x86/speculation/l1tf: Invert all not present mappings")
> > > bc2d8d262cba ("cpu/hotplug: Fix SMT supported evaluation")
> > > 
> > > The branches are updated and pushed out. The resulting bundles are
> > > attached.
> > > 
> > > I'm leaving for the summer camp now, but I'll have a look once per day into
> > > the list and if no reports come in until the weekend I'll prepare the pull
> > > request for Linus from what we have now.
> > > 
> > > Please give it a proper testing.
> > 
> > I've updated the 4.9 branch. I also have this to add on top, with
> > slight tweaks for each of the stable branches. Want me to just push
> > those?
> 
> I have some initial patches that Guenter made up for 4.4 that ignores
> the KVM stuff, as odds are those will be good to get merged.  Should I
> post them here or push them to the git tree?  I haven't checked to see
> if I have write permissions...

You have now.

Thanks,

	tglx

[MODERATED] Re: L!TF Bulletin #8: The final bits

[Greg KH]

On Wed, Aug 08, 2018 at 09:33:18PM +0200, speck for Thomas Gleixner wrote:
> On Wed, 8 Aug 2018, speck for Greg KH wrote:
> > On Wed, Aug 08, 2018 at 11:34:40AM +0100, speck for David Woodhouse wrote:
> > > 
> > > 
> > > On Wed, 2018-08-08 at 09:37 +0200, speck for Thomas Gleixner wrote:
> > > > Hi!
> > > > 
> > > > I've merged the SMT fix and parts of Andis PTE inversion series:
> > > > 
> > > > 958f79b9ee55 ("x86/mm/pat: Make set_memory_np() L1TF safe")
> > > > 0768f91530ff ("x86/speculation/l1tf: Make pmd/pud_mknotpresent() invert")
> > > > f22cc87f6c1f ("x86/speculation/l1tf: Invert all not present mappings")
> > > > bc2d8d262cba ("cpu/hotplug: Fix SMT supported evaluation")
> > > > 
> > > > The branches are updated and pushed out. The resulting bundles are
> > > > attached.
> > > > 
> > > > I'm leaving for the summer camp now, but I'll have a look once per day into
> > > > the list and if no reports come in until the weekend I'll prepare the pull
> > > > request for Linus from what we have now.
> > > > 
> > > > Please give it a proper testing.
> > > 
> > > I've updated the 4.9 branch. I also have this to add on top, with
> > > slight tweaks for each of the stable branches. Want me to just push
> > > those?
> > 
> > I have some initial patches that Guenter made up for 4.4 that ignores
> > the KVM stuff, as odds are those will be good to get merged.  Should I
> > post them here or push them to the git tree?  I haven't checked to see
> > if I have write permissions...
> 
> You have now.

Wonderful, I've pushed out the patches that Guenter sent me and tested
to the linux-4.4.y branch there.

thanks,

greg k-h

Re: L!TF Bulletin #8: The final bits

[Thomas Gleixner]

On Wed, 8 Aug 2018, speck for David Woodhouse wrote:
> On Wed, 2018-08-08 at 09:37 +0200, speck for Thomas Gleixner wrote:
> I've updated the 4.9 branch. I also have this to add on top, with
> slight tweaks for each of the stable branches. Want me to just push
> those?

Yes. Please add:

Acked-by: Thomas Gleixner <tglx@linutronix.de>