Skip to contentSkip to conversation

x86/doc: add PTI description

PATCH Superseded Dave Hansen 3 participants 4 messages Jan 5, 2018 Documentation

v2 of 4 Superseded Latest is v4 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 x86/doc: add PTI description

Changes from v1:
* update kernel-parameters.txt to clarify that the pti= option is not just for disabling. Also describe what 'pti=auto' does and why
* Add a note about the presence of NX in the user portion of the kernel page tables
* Clarify _additional_ 4k of PGD space
* Add a note about the runtime overhead of PCID without INVPCID

---

From: Dave Hansen <[email protected]>

Add some details about how PTI works, what some of the downsides are, and how to debug it when things go wrong.

Also document the kernel parameter: 'nopti'.

Cc Moritz Lipp <[email protected]>
Cc Daniel Gruss <[email protected]>
Cc Michael Schwarz <[email protected]>
Cc Richard Fellner <[email protected]>
Cc Andy Lutomirski <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Hugh Dickins <[email protected]>

---

 b/Documentation/admin-guide/kernel-parameters.txt |   22 +-
 b/Documentation/x86/pti.txt                       |  185 ++++++++++++++++++++++
 2 files changed, 200 insertions(+), 7 deletions(-)
Documentation/admin-guide/kernel-parameters.txt renamed from Documentation/admin-guide/kernel-parameters.txt~kpti-doc +15 -7
diff -puN Documentation/admin-guide/kernel-parameters.txt~kpti-doc Documentation/admin-guide/kernel-parameters.txt
@@ -2712,8 +2712,6 @@
 			steal time is computed, but won't influence scheduler
 			behaviour
 
-	nopti		[X86-64] Disable kernel page table isolation
-
 	nolapic		[X86-32,APIC] Do not enable or use the local APIC.
 
 	nolapic_timer	[X86-32,APIC] Do not use the local APIC timer.
@@ -3288,11 +3286,21 @@
 	pt.		[PARIDE]
 			See Documentation/blockdev/paride.txt.
 
-	pti=		[X86_64]
-			Control user/kernel address space isolation:
-			on - enable
-			off - disable
-			auto - default setting
+	pti=		[X86_64] Control Page Table Isolation of user and
+			kernel address spaces.  Disabling this feature
+			removes hardening, but improves performance of
+			system calls and interrupts.
+
+			on   - unconditionally enable
+			off  - unconditionally disable
+			auto - kernel detects whether your CPU model is
+			       vulnerable to issues that PTI mitigates
+
+			Not specifying this option nothing is equivalent to
+			pti=auto.
+
+	nopti		[X86_64]
+			Equivalent to pti=off
 
 	pty.legacy_count=
 			[KNL] Number of legacy pty's. Overwrites compiled-in
Documentation/x86/pti.txt new file +185 -0

185 lines not on this page · raw message

_

#1 lore raw

Documentation/admin-guide/kernel-parameters.txt line 3300

Quoted 2 lines · Dave Hansen
Documentation/admin-guide/kernel-parameters.txt renamed from Documentation/admin-guide/kernel-parameters.txt~kpti-doc +2 -0
+			Not specifying this option nothing is equivalent to
+			pti=auto.

-nothing

Documentation/x86/pti.txt line 6

Quoted 3 lines
Documentation/x86/pti.txt new file +3 -0
+Page Table Isolation (pti, previously known as KAISER[1]) is a
+countermeasure against attacks on kernel address information such
+as the "Meltdown" approach[2].

It's not really just address information, but any data. Maybe "attacks that leak kernel memory"?

Quoted 1 line

+To avoid leaking address information, we create an new, independent

Same issue here. Also an -> a.

Quoted 1 line

+copy of the page tables which are used only when running userspace

are -> is. The copy is singular.

Documentation/x86/pti.txt line 11

Quoted 2 lines
Documentation/x86/pti.txt new file +2 -0
+applications.  When the kernel is entered via syscalls, interrupts or
+exceptions, page tables are switched to the full "kernel" copy.  When

"the page tables".

Documentation/x86/pti.txt line 38

Quoted 3 lines
Documentation/x86/pti.txt new file +3 -0
+crippled by setting the NX bit in the top level.  This ensures
+that if a kernel->user CR3 switch is missed that userspace will
+crash immediately upon executing its first instruction.

"that userspace" -> "then userspace"

Signature · 2 lines
Hector Martin "marcan" ([email protected])
Public Key: https://mrcn.st/pub
#2 lore raw
Quoted 5 lines · Hector Martin 'marcan

Documentation/admin-guide/kernel-parameters.txt line 3300

Quoted 2 lines · Dave Hansen
Documentation/admin-guide/kernel-parameters.txt renamed from Documentation/admin-guide/kernel-parameters.txt~kpti-doc +2 -0
+			Not specifying this option nothing is equivalent to
+			pti=auto.

-nothing

Sure, will fix.

Quoted 6 lines

Documentation/x86/pti.txt line 6

Quoted 3 lines
Documentation/x86/pti.txt new file +3 -0
+Page Table Isolation (pti, previously known as KAISER[1]) is a
+countermeasure against attacks on kernel address information such
+as the "Meltdown" approach[2].

It's not really just address information, but any data. Maybe "attacks that leak kernel memory"?

It's not just kernel leaks either, though.

Quoted 3 lines
Quoted 1 line

+To avoid leaking address information, we create an new, independent

Same issue here. Also an -> a.

Will fix.

Quoted 3 lines
Quoted 1 line

+copy of the page tables which are used only when running userspace

are -> is. The copy is singular.

I've reworded the sentence to remove the ambiguity.

Quoted 4 lines

Documentation/x86/pti.txt line 11

Quoted 2 lines
Documentation/x86/pti.txt new file +2 -0
+applications.  When the kernel is entered via syscalls, interrupts or
+exceptions, page tables are switched to the full "kernel" copy.  When

"the page tables".

No thanks. It's fine the way it is.

Quoted 5 lines

Documentation/x86/pti.txt line 38

Quoted 3 lines
Documentation/x86/pti.txt new file +3 -0
+crippled by setting the NX bit in the top level.  This ensures
+that if a kernel->user CR3 switch is missed that userspace will
+crash immediately upon executing its first instruction.

"that userspace" -> "then userspace"

reviewer
#3 lore raw
In reply to Dave Hansen
Quoted 38 lines · Dave Hansen

Changes from v1:
* update kernel-parameters.txt to clarify that the pti= option is not just for disabling. Also describe what 'pti=auto' does and why
* Add a note about the presence of NX in the user portion of the kernel page tables
* Clarify _additional_ 4k of PGD space
* Add a note about the runtime overhead of PCID without INVPCID

---

From: Dave Hansen <[email protected]>

Add some details about how PTI works, what some of the downsides are, and how to debug it when things go wrong.

Also document the kernel parameter: 'nopti'.

Cc Moritz Lipp <[email protected]>
Cc Daniel Gruss <[email protected]>
Cc Michael Schwarz <[email protected]>
Cc Richard Fellner <[email protected]>
Cc Andy Lutomirski <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Hugh Dickins <[email protected]>

---

 b/Documentation/admin-guide/kernel-parameters.txt |   22 +-
 b/Documentation/x86/pti.txt                       |  185 ++++++++++++++++++++++
 2 files changed, 200 insertions(+), 7 deletions(-)

diff -puN /dev/null Documentation/x86/pti.txt

Documentation/x86/pti.txt line 41

Quoted 3 lines
Documentation/x86/pti.txt new file +2 -0
@@ -0,0 +1,185 @@
 
+The userspace copy is used when running userspace and mirrors the
+mapping of userspace present in the kernel copy.  It maps a only

drop: a

Documentation/x86/pti.txt line 77

Quoted 12 lines
Documentation/x86/pti.txt new file +11 -0
+the kernel data needed to enter and exit the kernel.  This data
+is entirely contained in the 'struct cpu_entry_area' structure
+which is placed in the fixmap and thus each CPU's copy of the
+area has a compile-time-fixed virtual address.
+
 
+2. Runtime Cost
+  a. CR3 manipulation to switch between the page table copies
+     must be done at interrupt, syscall, and exception entry
+     and exit (it can be skipped when the kernel is interrupted,
+     though.)  Moves to CR3 are on the order of a hundred
+     cycles, and are required every at entry and every at exit.

at every entry and at every exit.

Documentation/x86/pti.txt line 84

Quoted 2 lines
Documentation/x86/pti.txt new file +2 -0
+  d. Global pages are disabled for all kernel structures not
+     mapped in both to kernel and userspace page tables.  This

into both kernel and userspace page tables.

Quoted 4 lines
Documentation/x86/pti.txt new file +4 -0
+     feature of the MMU allows different processes to share TLB
+     entries mapping the kernel.  Losing the feature means more
+     TLB misses after a context switch.  The actual loss of
+     performance is very small, however, never exceeding 1%.

Documentation/x86/pti.txt line 112

Quoted 10 lines
Documentation/x86/pti.txt new file +9 -0
 
+  f. In addition to the fork()-time copying, there must also
+     be an update to the userspace PGD any time a set_pgd() is done
+     on a PGD used to map userspace.  This ensures that the kernel
+     and userspace copies always map the same userspace
+     memory.
+  g. On systems without PCID support, each CR3 write flushes
+     the entire TLB.  That means that each syscall, interrupt
+     or exception flushes the TLB.
+  h. On systems without INVPCID support, addresses can only be

This is the first mention of INVPCID. Probably needs more info about what it is.

Documentation/x86/pti.txt line 122

Quoted 10 lines
Documentation/x86/pti.txt new file +10 -0
+     flushed from the TLB for the current PCID.  When flushing
+     a kernel address, we need to flush all PCIDs, so a single
+     kernel address flush will require a TLB-flushing CR3 write
+     upon the next use of every PCID.
+
+Possible Future Work
+====================
+1. We can be more careful about not actually writing to CR3
+   unless its value is actually changed.
+2. Allow PTI to enabled/disabled at runtime in addition to the

to be

Quoted 4 lines
Documentation/x86/pti.txt new file +4 -0
+   boot-time switching.
+
+Testing
+========
Signature · 1 line
~Randy