Skip to content

Commit 860fc2f

Browse files
author
Ingo Molnar
committed
Merge branch 'perf/urgent' into perf/core
Pick up the latest fixes, refresh the development tree. Signed-off-by: Ingo Molnar <mingo@kernel.org>
2 parents 1977499 + bee09ed commit 860fc2f

File tree

565 files changed

+5338
-2771
lines changed

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

565 files changed

+5338
-2771
lines changed

‎Documentation/block/null_blk.txt

Lines changed: 72 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,72 @@
1+
Null block device driver
2+
================================================================================
3+
4+
I. Overview
5+
6+
The null block device (/dev/nullb*) is used for benchmarking the various
7+
block-layer implementations. It emulates a block device of X gigabytes in size.
8+
The following instances are possible:
9+
10+
Single-queue block-layer
11+
- Request-based.
12+
- Single submission queue per device.
13+
- Implements IO scheduling algorithms (CFQ, Deadline, noop).
14+
Multi-queue block-layer
15+
- Request-based.
16+
- Configurable submission queues per device.
17+
No block-layer (Known as bio-based)
18+
- Bio-based. IO requests are submitted directly to the device driver.
19+
- Directly accepts bio data structure and returns them.
20+
21+
All of them have a completion queue for each core in the system.
22+
23+
II. Module parameters applicable for all instances:
24+
25+
queue_mode=[0-2]: Default: 2-Multi-queue
26+
Selects which block-layer the module should instantiate with.
27+
28+
0: Bio-based.
29+
1: Single-queue.
30+
2: Multi-queue.
31+
32+
home_node=[0--nr_nodes]: Default: NUMA_NO_NODE
33+
Selects what CPU node the data structures are allocated from.
34+
35+
gb=[Size in GB]: Default: 250GB
36+
The size of the device reported to the system.
37+
38+
bs=[Block size (in bytes)]: Default: 512 bytes
39+
The block size reported to the system.
40+
41+
nr_devices=[Number of devices]: Default: 2
42+
Number of block devices instantiated. They are instantiated as /dev/nullb0,
43+
etc.
44+
45+
irq_mode=[0-2]: Default: 1-Soft-irq
46+
The completion mode used for completing IOs to the block-layer.
47+
48+
0: None.
49+
1: Soft-irq. Uses IPI to complete IOs across CPU nodes. Simulates the overhead
50+
when IOs are issued from another CPU node than the home the device is
51+
connected to.
52+
2: Timer: Waits a specific period (completion_nsec) for each IO before
53+
completion.
54+
55+
completion_nsec=[ns]: Default: 10.000ns
56+
Combined with irq_mode=2 (timer). The time each completion event must wait.
57+
58+
submit_queues=[0..nr_cpus]:
59+
The number of submission queues attached to the device driver. If unset, it
60+
defaults to 1 on single-queue and bio-based instances. For multi-queue,
61+
it is ignored when use_per_node_hctx module parameter is 1.
62+
63+
hw_queue_depth=[0..qdepth]: Default: 64
64+
The hardware queue depth of the device.
65+
66+
III: Multi-queue specific parameters
67+
68+
use_per_node_hctx=[0/1]: Default: 0
69+
0: The number of submit queues are set to the value of the submit_queues
70+
parameter.
71+
1: The multi-queue block layer is instantiated with a hardware dispatch
72+
queue for each CPU node in the system.

‎Documentation/devicetree/bindings/clock/exynos5250-clock.txt

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -159,6 +159,8 @@ clock which they consume.
159159
mixer 343
160160
hdmi 344
161161
g2d 345
162+
mdma0 346
163+
smmu_mdma0 347
162164

163165

164166
[Clock Muxes]

‎Documentation/kernel-parameters.txt

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1529,6 +1529,8 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
15291529

15301530
* atapi_dmadir: Enable ATAPI DMADIR bridge support
15311531

1532+
* disable: Disable this device.
1533+
15321534
If there are multiple matching configurations changing
15331535
the same attribute, the last one is used.
15341536

‎Documentation/module-signing.txt

Lines changed: 240 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,240 @@
1+
==============================
2+
KERNEL MODULE SIGNING FACILITY
3+
==============================
4+
5+
CONTENTS
6+
7+
- Overview.
8+
- Configuring module signing.
9+
- Generating signing keys.
10+
- Public keys in the kernel.
11+
- Manually signing modules.
12+
- Signed modules and stripping.
13+
- Loading signed modules.
14+
- Non-valid signatures and unsigned modules.
15+
- Administering/protecting the private key.
16+
17+
18+
========
19+
OVERVIEW
20+
========
21+
22+
The kernel module signing facility cryptographically signs modules during
23+
installation and then checks the signature upon loading the module. This
24+
allows increased kernel security by disallowing the loading of unsigned modules
25+
or modules signed with an invalid key. Module signing increases security by
26+
making it harder to load a malicious module into the kernel. The module
27+
signature checking is done by the kernel so that it is not necessary to have
28+
trusted userspace bits.
29+
30+
This facility uses X.509 ITU-T standard certificates to encode the public keys
31+
involved. The signatures are not themselves encoded in any industrial standard
32+
type. The facility currently only supports the RSA public key encryption
33+
standard (though it is pluggable and permits others to be used). The possible
34+
hash algorithms that can be used are SHA-1, SHA-224, SHA-256, SHA-384, and
35+
SHA-512 (the algorithm is selected by data in the signature).
36+
37+
38+
==========================
39+
CONFIGURING MODULE SIGNING
40+
==========================
41+
42+
The module signing facility is enabled by going to the "Enable Loadable Module
43+
Support" section of the kernel configuration and turning on
44+
45+
CONFIG_MODULE_SIG "Module signature verification"
46+
47+
This has a number of options available:
48+
49+
(1) "Require modules to be validly signed" (CONFIG_MODULE_SIG_FORCE)
50+
51+
This specifies how the kernel should deal with a module that has a
52+
signature for which the key is not known or a module that is unsigned.
53+
54+
If this is off (ie. "permissive"), then modules for which the key is not
55+
available and modules that are unsigned are permitted, but the kernel will
56+
be marked as being tainted.
57+
58+
If this is on (ie. "restrictive"), only modules that have a valid
59+
signature that can be verified by a public key in the kernel's possession
60+
will be loaded. All other modules will generate an error.
61+
62+
Irrespective of the setting here, if the module has a signature block that
63+
cannot be parsed, it will be rejected out of hand.
64+
65+
66+
(2) "Automatically sign all modules" (CONFIG_MODULE_SIG_ALL)
67+
68+
If this is on then modules will be automatically signed during the
69+
modules_install phase of a build. If this is off, then the modules must
70+
be signed manually using:
71+
72+
scripts/sign-file
73+
74+
75+
(3) "Which hash algorithm should modules be signed with?"
76+
77+
This presents a choice of which hash algorithm the installation phase will
78+
sign the modules with:
79+
80+
CONFIG_SIG_SHA1 "Sign modules with SHA-1"
81+
CONFIG_SIG_SHA224 "Sign modules with SHA-224"
82+
CONFIG_SIG_SHA256 "Sign modules with SHA-256"
83+
CONFIG_SIG_SHA384 "Sign modules with SHA-384"
84+
CONFIG_SIG_SHA512 "Sign modules with SHA-512"
85+
86+
The algorithm selected here will also be built into the kernel (rather
87+
than being a module) so that modules signed with that algorithm can have
88+
their signatures checked without causing a dependency loop.
89+
90+
91+
=======================
92+
GENERATING SIGNING KEYS
93+
=======================
94+
95+
Cryptographic keypairs are required to generate and check signatures. A
96+
private key is used to generate a signature and the corresponding public key is
97+
used to check it. The private key is only needed during the build, after which
98+
it can be deleted or stored securely. The public key gets built into the
99+
kernel so that it can be used to check the signatures as the modules are
100+
loaded.
101+
102+
Under normal conditions, the kernel build will automatically generate a new
103+
keypair using openssl if one does not exist in the files:
104+
105+
signing_key.priv
106+
signing_key.x509
107+
108+
during the building of vmlinux (the public part of the key needs to be built
109+
into vmlinux) using parameters in the:
110+
111+
x509.genkey
112+
113+
file (which is also generated if it does not already exist).
114+
115+
It is strongly recommended that you provide your own x509.genkey file.
116+
117+
Most notably, in the x509.genkey file, the req_distinguished_name section
118+
should be altered from the default:
119+
120+
[ req_distinguished_name ]
121+
O = Magrathea
122+
CN = Glacier signing key
123+
emailAddress = slartibartfast@magrathea.h2g2
124+
125+
The generated RSA key size can also be set with:
126+
127+
[ req ]
128+
default_bits = 4096
129+
130+
131+
It is also possible to manually generate the key private/public files using the
132+
x509.genkey key generation configuration file in the root node of the Linux
133+
kernel sources tree and the openssl command. The following is an example to
134+
generate the public/private key files:
135+
136+
openssl req -new -nodes -utf8 -sha256 -days 36500 -batch -x509 \
137+
-config x509.genkey -outform DER -out signing_key.x509 \
138+
-keyout signing_key.priv
139+
140+
141+
=========================
142+
PUBLIC KEYS IN THE KERNEL
143+
=========================
144+
145+
The kernel contains a ring of public keys that can be viewed by root. They're
146+
in a keyring called ".system_keyring" that can be seen by:
147+
148+
[root@deneb ~]# cat /proc/keys
149+
...
150+
223c7853 I------ 1 perm 1f030000 0 0 keyring .system_keyring: 1
151+
302d2d52 I------ 1 perm 1f010000 0 0 asymmetri Fedora kernel signing key: d69a84e6bce3d216b979e9505b3e3ef9a7118079: X509.RSA a7118079 []
152+
...
153+
154+
Beyond the public key generated specifically for module signing, any file
155+
placed in the kernel source root directory or the kernel build root directory
156+
whose name is suffixed with ".x509" will be assumed to be an X.509 public key
157+
and will be added to the keyring.
158+
159+
Further, the architecture code may take public keys from a hardware store and
160+
add those in also (e.g. from the UEFI key database).
161+
162+
Finally, it is possible to add additional public keys by doing:
163+
164+
keyctl padd asymmetric "" [.system_keyring-ID] <[key-file]
165+
166+
e.g.:
167+
168+
keyctl padd asymmetric "" 0x223c7853 <my_public_key.x509
169+
170+
Note, however, that the kernel will only permit keys to be added to
171+
.system_keyring _if_ the new key's X.509 wrapper is validly signed by a key
172+
that is already resident in the .system_keyring at the time the key was added.
173+
174+
175+
=========================
176+
MANUALLY SIGNING MODULES
177+
=========================
178+
179+
To manually sign a module, use the scripts/sign-file tool available in
180+
the Linux kernel source tree. The script requires 4 arguments:
181+
182+
1. The hash algorithm (e.g., sha256)
183+
2. The private key filename
184+
3. The public key filename
185+
4. The kernel module to be signed
186+
187+
The following is an example to sign a kernel module:
188+
189+
scripts/sign-file sha512 kernel-signkey.priv \
190+
kernel-signkey.x509 module.ko
191+
192+
The hash algorithm used does not have to match the one configured, but if it
193+
doesn't, you should make sure that hash algorithm is either built into the
194+
kernel or can be loaded without requiring itself.
195+
196+
197+
============================
198+
SIGNED MODULES AND STRIPPING
199+
============================
200+
201+
A signed module has a digital signature simply appended at the end. The string
202+
"~Module signature appended~." at the end of the module's file confirms that a
203+
signature is present but it does not confirm that the signature is valid!
204+
205+
Signed modules are BRITTLE as the signature is outside of the defined ELF
206+
container. Thus they MAY NOT be stripped once the signature is computed and
207+
attached. Note the entire module is the signed payload, including any and all
208+
debug information present at the time of signing.
209+
210+
211+
======================
212+
LOADING SIGNED MODULES
213+
======================
214+
215+
Modules are loaded with insmod, modprobe, init_module() or finit_module(),
216+
exactly as for unsigned modules as no processing is done in userspace. The
217+
signature checking is all done within the kernel.
218+
219+
220+
=========================================
221+
NON-VALID SIGNATURES AND UNSIGNED MODULES
222+
=========================================
223+
224+
If CONFIG_MODULE_SIG_FORCE is enabled or enforcemodulesig=1 is supplied on
225+
the kernel command line, the kernel will only load validly signed modules
226+
for which it has a public key. Otherwise, it will also load modules that are
227+
unsigned. Any module for which the kernel has a key, but which proves to have
228+
a signature mismatch will not be permitted to load.
229+
230+
Any module that has an unparseable signature will be rejected.
231+
232+
233+
=========================================
234+
ADMINISTERING/PROTECTING THE PRIVATE KEY
235+
=========================================
236+
237+
Since the private key is used to sign modules, viruses and malware could use
238+
the private key to sign modules and compromise the operating system. The
239+
private key must be either destroyed or moved to a secure location and not kept
240+
in the root node of the kernel source tree.

‎Documentation/networking/ip-sysctl.txt

Lines changed: 6 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -16,8 +16,12 @@ ip_default_ttl - INTEGER
1616
Default: 64 (as recommended by RFC1700)
1717

1818
ip_no_pmtu_disc - BOOLEAN
19-
Disable Path MTU Discovery.
20-
default FALSE
19+
Disable Path MTU Discovery. If enabled and a
20+
fragmentation-required ICMP is received, the PMTU to this
21+
destination will be set to min_pmtu (see below). You will need
22+
to raise min_pmtu to the smallest interface MTU on your system
23+
manually if you want to avoid locally generated fragments.
24+
Default: FALSE
2125

2226
min_pmtu - INTEGER
2327
default 552 - minimum discovered Path MTU

0 commit comments

Comments
 (0)