Commit 1fb3b526 authored by Linus Torvalds's avatar Linus Torvalds

Merge tag 'docs-5.2a' of git://git.lwn.net/linux

Pull more documentation updates from Jonathan Corbet:
 "Some late arriving documentation changes. In particular, this contains
  the conversion of the x86 docs to RST, which has been in the works for
  some time but needed a couple of final tweaks"

* tag 'docs-5.2a' of git://git.lwn.net/linux: (29 commits)
  Documentation: x86: convert x86_64/machinecheck to reST
  Documentation: x86: convert x86_64/cpu-hotplug-spec to reST
  Documentation: x86: convert x86_64/fake-numa-for-cpusets to reST
  Documentation: x86: convert x86_64/5level-paging.txt to reST
  Documentation: x86: convert x86_64/mm.txt to reST
  Documentation: x86: convert x86_64/uefi.txt to reST
  Documentation: x86: convert x86_64/boot-options.txt to reST
  Documentation: x86: convert i386/IO-APIC.txt to reST
  Documentation: x86: convert usb-legacy-support.txt to reST
  Documentation: x86: convert orc-unwinder.txt to reST
  Documentation: x86: convert resctrl_ui.txt to reST
  Documentation: x86: convert microcode.txt to reST
  Documentation: x86: convert pti.txt to reST
  Documentation: x86: convert amd-memory-encryption.txt to reST
  Documentation: x86: convert intel_mpx.txt to reST
  Documentation: x86: convert protection-keys.txt to reST
  Documentation: x86: convert pat.txt to reST
  Documentation: x86: convert mtrr.txt to reST
  Documentation: x86: convert tlb.txt to reST
  Documentation: x86: convert zero-page.txt to reST
  ...
parents e290e6af afbd4d42
......@@ -112,6 +112,7 @@ implementation.
.. toctree::
:maxdepth: 2
x86/index
sh/index
Filesystem Documentation
......
......@@ -1915,7 +1915,10 @@ The following commonly-used handler.action pairs are available:
The 'matching.event' specification is simply the fully qualified
event name of the event that matches the target event for the
onmatch() functionality, in the form 'system.event_name'.
onmatch() functionality, in the form 'system.event_name'. Histogram
keys of both events are compared to find if events match. In case
multiple histogram keys are used, they all must match in the specified
order.
Finally, the number and type of variables/fields in the 'param
list' must match the number and types of the fields in the
......@@ -1978,9 +1981,9 @@ The following commonly-used handler.action pairs are available:
/sys/kernel/debug/tracing/events/sched/sched_waking/trigger
Then, when the corresponding thread is actually scheduled onto the
CPU by a sched_switch event, calculate the latency and use that
along with another variable and an event field to generate a
wakeup_latency synthetic event::
CPU by a sched_switch event (saved_pid matches next_pid), calculate
the latency and use that along with another variable and an event field
to generate a wakeup_latency synthetic event::
# echo 'hist:keys=next_pid:wakeup_lat=common_timestamp.usecs-$ts0:\
onmatch(sched.sched_waking).wakeup_latency($wakeup_lat,\
......
......@@ -249,13 +249,13 @@ essere categorizzate in:
|
2. Licenze non raccomandate:
2. Licenze deprecate:
Questo tipo di licenze dovrebbero essere usate solo per codice già esistente
o quando si prende codice da altri progetti. Le licenze sono disponibili
nei sorgenti del kernel nella cartella::
LICENSES/other/
LICENSES/deprecated/
I file in questa cartella contengono il testo completo della licenza e i
`Metatag`_. Il nome di questi file è lo stesso usato come identificatore
......@@ -263,14 +263,14 @@ essere categorizzate in:
Esempi::
LICENSES/other/ISC
LICENSES/deprecated/ISC
Contiene il testo della licenza Internet System Consortium e i suoi
metatag::
LICENSES/other/ZLib
LICENSES/deprecated/GPL-1.0
Contiene il testo della licenza ZLIB e i suoi metatag.
Contiene il testo della versione 1 della licenza GPL e i suoi metatag.
Metatag:
......@@ -294,7 +294,55 @@ essere categorizzate in:
|
3. _`Eccezioni`:
3. Solo per doppie licenze
Queste licenze dovrebbero essere usate solamente per codice licenziato in
combinazione con un'altra licenza che solitamente è quella preferita.
Queste licenze sono disponibili nei sorgenti del kernel nella cartella::
LICENSES/dual
I file in questa cartella contengono il testo completo della rispettiva
licenza e i suoi `Metatags`_. I nomi dei file sono identici agli
identificatori di licenza SPDX che dovrebbero essere usati nei file
sorgenti.
Esempi::
LICENSES/dual/MPL-1.1
Questo file contiene il testo della versione 1.1 della licenza *Mozilla
Pulic License* e i metatag necessari::
LICENSES/dual/Apache-2.0
Questo file contiene il testo della versione 2.0 della licenza Apache e i
metatag necessari.
Metatag:
I requisiti per le 'altre' ('*other*') licenze sono identici a quelli per le
`Licenze raccomandate`_.
Esempio del formato del file::
Valid-License-Identifier: MPL-1.1
SPDX-URL: https://spdx.org/licenses/MPL-1.1.html
Usage-Guide:
Do NOT use. The MPL-1.1 is not GPL2 compatible. It may only be used for
dual-licensed files where the other license is GPL2 compatible.
If you end up using this it MUST be used together with a GPL2 compatible
license using "OR".
To use the Mozilla Public License version 1.1 put the following SPDX
tag/value pair into a comment according to the placement guidelines in
the licensing rules documentation:
SPDX-License-Identifier: MPL-1.1
License-Text:
Full license text
|
4. _`Eccezioni`:
Alcune licenze possono essere corrette con delle eccezioni che forniscono
diritti aggiuntivi. Queste eccezioni sono disponibili nei sorgenti del
......
.. SPDX-License-Identifier: GPL-2.0
=====================
AMD Memory Encryption
=====================
Secure Memory Encryption (SME) and Secure Encrypted Virtualization (SEV) are
features found on AMD processors.
......@@ -34,7 +40,7 @@ is operating in 64-bit or 32-bit PAE mode, in all other modes the SEV hardware
forces the memory encryption bit to 1.
Support for SME and SEV can be determined through the CPUID instruction. The
CPUID function 0x8000001f reports information related to SME:
CPUID function 0x8000001f reports information related to SME::
0x8000001f[eax]:
Bit[0] indicates support for SME
......@@ -48,14 +54,14 @@ CPUID function 0x8000001f reports information related to SME:
addresses)
If support for SME is present, MSR 0xc00100010 (MSR_K8_SYSCFG) can be used to
determine if SME is enabled and/or to enable memory encryption:
determine if SME is enabled and/or to enable memory encryption::
0xc0010010:
Bit[23] 0 = memory encryption features are disabled
1 = memory encryption features are enabled
If SEV is supported, MSR 0xc0010131 (MSR_AMD64_SEV) can be used to determine if
SEV is active:
SEV is active::
0xc0010131:
Bit[0] 0 = memory encryption is not active
......@@ -68,6 +74,7 @@ requirements for the system. If this bit is not set upon Linux startup then
Linux itself will not set it and memory encryption will not be possible.
The state of SME in the Linux kernel can be documented as follows:
- Supported:
The CPU supports SME (determined through CPUID instruction).
......
.. SPDX-License-Identifier: GPL-2.0
============
Early Printk
============
Mini-HOWTO for using the earlyprintk=dbgp boot option with a
USB2 Debug port key and a debug cable, on x86 systems.
You need two computers, the 'USB debug key' special gadget and
and two USB cables, connected like this:
and two USB cables, connected like this::
[host/target] <-------> [USB debug key] <-------> [client/console]
1. There are a number of specific hardware requirements:
Hardware requirements
=====================
a.) Host/target system needs to have USB debug port capability.
a) Host/target system needs to have USB debug port capability.
You can check this capability by looking at a 'Debug port' bit in
the lspci -vvv output:
the lspci -vvv output::
# lspci -vvv
...
......@@ -32,20 +38,20 @@ and two USB cables, connected like this:
Kernel modules: ehci-hcd
...
( If your system does not list a debug port capability then you probably
won't be able to use the USB debug key. )
.. note::
If your system does not list a debug port capability then you probably
won't be able to use the USB debug key.
b.) You also need a NetChip USB debug cable/key:
b) You also need a NetChip USB debug cable/key:
http://www.plxtech.com/products/NET2000/NET20DC/default.asp
This is a small blue plastic connector with two USB connections;
it draws power from its USB connections.
c.) You need a second client/console system with a high speed USB 2.0
port.
c) You need a second client/console system with a high speed USB 2.0 port.
d.) The NetChip device must be plugged directly into the physical
d) The NetChip device must be plugged directly into the physical
debug port on the "host/target" system. You cannot use a USB hub in
between the physical debug port and the "host/target" system.
......@@ -65,29 +71,31 @@ and two USB cables, connected like this:
to the hardware vendor, because there is no reason not to wire
this port into one of the physically accessible ports.
e.) It is also important to note, that many versions of the NetChip
e) It is also important to note, that many versions of the NetChip
device require the "client/console" system to be plugged into the
right hand side of the device (with the product logo facing up and
readable left to right). The reason being is that the 5 volt
power supply is taken from only one side of the device and it
must be the side that does not get rebooted.
2. Software requirements:
Software requirements
=====================
a.) On the host/target system:
a) On the host/target system:
You need to enable the following kernel config option:
You need to enable the following kernel config option::
CONFIG_EARLY_PRINTK_DBGP=y
And you need to add the boot command line: "earlyprintk=dbgp".
(If you are using Grub, append it to the 'kernel' line in
.. note::
If you are using Grub, append it to the 'kernel' line in
/etc/grub.conf. If you are using Grub2 on a BIOS firmware system,
append it to the 'linux' line in /boot/grub2/grub.cfg. If you are
using Grub2 on an EFI firmware system, append it to the 'linux'
or 'linuxefi' line in /boot/grub2/grub.cfg or
/boot/efi/EFI/<distro>/grub.cfg.)
/boot/efi/EFI/<distro>/grub.cfg.
On systems with more than one EHCI debug controller you must
specify the correct EHCI debug controller number. The ordering
......@@ -96,14 +104,15 @@ and two USB cables, connected like this:
controller. To use the second EHCI debug controller, you would
use the command line: "earlyprintk=dbgp1"
NOTE: normally earlyprintk console gets turned off once the
.. note::
normally earlyprintk console gets turned off once the
regular console is alive - use "earlyprintk=dbgp,keep" to keep
this channel open beyond early bootup. This can be useful for
debugging crashes under Xorg, etc.
b.) On the client/console system:
b) On the client/console system:
You should enable the following kernel config option:
You should enable the following kernel config option::
CONFIG_USB_SERIAL_DEBUG=y
......@@ -115,27 +124,28 @@ and two USB cables, connected like this:
it up to use /dev/ttyUSB0 - or use a raw 'cat /dev/ttyUSBx' to
see the raw output.
c.) On Nvidia Southbridge based systems: the kernel will try to probe
c) On Nvidia Southbridge based systems: the kernel will try to probe
and find out which port has a debug device connected.
3. Testing that it works fine:
Testing
=======
You can test the output by using earlyprintk=dbgp,keep and provoking
kernel messages on the host/target system. You can provoke a harmless
kernel message by for example doing:
You can test the output by using earlyprintk=dbgp,keep and provoking
kernel messages on the host/target system. You can provoke a harmless
kernel message by for example doing::
echo h > /proc/sysrq-trigger
On the host/target system you should see this help line in "dmesg" output:
On the host/target system you should see this help line in "dmesg" output::
SysRq : HELP : loglevel(0-9) reBoot Crashdump terminate-all-tasks(E) memory-full-oom-kill(F) kill-all-tasks(I) saK show-backtrace-all-active-cpus(L) show-memory-usage(M) nice-all-RT-tasks(N) powerOff show-registers(P) show-all-timers(Q) unRaw Sync show-task-states(T) Unmount show-blocked-tasks(W) dump-ftrace-buffer(Z)
On the client/console system do:
On the client/console system do::
cat /dev/ttyUSB0
And you should see the help line above displayed shortly after you've
provoked it on the host system.
And you should see the help line above displayed shortly after you've
provoked it on the host system.
If it does not work then please ask about it on the linux-kernel@vger.kernel.org
mailing list or contact the x86 maintainers.
.. SPDX-License-Identifier: GPL-2.0
==============
Kernel Entries
==============
This file documents some of the kernel entries in
arch/x86/entry/entry_64.S. A lot of this explanation is adapted from
an email from Ingo Molnar:
......@@ -59,7 +65,7 @@ Now, there's a secondary complication: there's a cheap way to test
which mode the CPU is in and an expensive way.
The cheap way is to pick this info off the entry frame on the kernel
stack, from the CS of the ptregs area of the kernel stack:
stack, from the CS of the ptregs area of the kernel stack::
xorl %ebx,%ebx
testl $3,CS+8(%rsp)
......@@ -67,7 +73,7 @@ stack, from the CS of the ptregs area of the kernel stack:
SWAPGS
The expensive (paranoid) way is to read back the MSR_GS_BASE value
(which is what SWAPGS modifies):
(which is what SWAPGS modifies)::
movl $1,%ebx
movl $MSR_GS_BASE,%ecx
......@@ -76,7 +82,7 @@ The expensive (paranoid) way is to read back the MSR_GS_BASE value
js 1f /* negative -> in kernel */
SWAPGS
xorl %ebx,%ebx
1: ret
1: ret
If we are at an interrupt or user-trap/gate-alike boundary then we can
use the faster check: the stack will be a reliable indicator of
......
Kernel level exception handling in Linux
Commentary by Joerg Pommnitz <joerg@raleigh.ibm.com>
.. SPDX-License-Identifier: GPL-2.0
===============================
Kernel level exception handling
===============================
Commentary by Joerg Pommnitz <joerg@raleigh.ibm.com>
When a process runs in kernel mode, it often has to access user
mode memory whose address has been passed by an untrusted program.
......@@ -25,9 +30,9 @@ How does this work?
Whenever the kernel tries to access an address that is currently not
accessible, the CPU generates a page fault exception and calls the
page fault handler
page fault handler::
void do_page_fault(struct pt_regs *regs, unsigned long error_code)
void do_page_fault(struct pt_regs *regs, unsigned long error_code)
in arch/x86/mm/fault.c. The parameters on the stack are set up by
the low level assembly glue in arch/x86/kernel/entry_32.S. The parameter
......@@ -57,12 +62,13 @@ as an example. The definition is somewhat hard to follow, so let's peek at
the code generated by the preprocessor and the compiler. I selected
the get_user call in drivers/char/sysrq.c for a detailed examination.
The original code in sysrq.c line 587:
The original code in sysrq.c line 587::
get_user(c, buf);
The preprocessor output (edited to become somewhat readable):
The preprocessor output (edited to become somewhat readable)::
(
(
{
long __gu_err = - 14 , __gu_val = 0;
const __typeof__(*( ( buf ) )) *__gu_addr = ((buf));
......@@ -120,10 +126,10 @@ The preprocessor output (edited to become somewhat readable):
((c)) = (__typeof__(*((buf))))__gu_val;
__gu_err;
}
);
);
WOW! Black GCC/assembly magic. This is impossible to follow, so let's
see what code gcc generates:
see what code gcc generates::
> xorl %edx,%edx
> movl current_set,%eax
......@@ -154,7 +160,7 @@ understand. Can we? The actual user access is quite obvious. Thanks
to the unified address space we can just access the address in user
memory. But what does the .section stuff do?????
To understand this we have to look at the final kernel:
To understand this we have to look at the final kernel::
> objdump --section-headers vmlinux
>
......@@ -181,7 +187,7 @@ To understand this we have to look at the final kernel:
There are obviously 2 non standard ELF sections in the generated object
file. But first we want to find out what happened to our code in the
final kernel executable:
final kernel executable::
> objdump --disassemble --section=.text vmlinux
>
......@@ -199,7 +205,7 @@ final kernel executable:
The whole user memory access is reduced to 10 x86 machine instructions.
The instructions bracketed in the .section directives are no longer
in the normal execution path. They are located in a different section
of the executable file:
of the executable file::
> objdump --disassemble --section=.fixup vmlinux
>
......@@ -207,14 +213,15 @@ of the executable file:
> c0199ffa <.fixup+10ba> xorb %dl,%dl
> c0199ffc <.fixup+10bc> jmp c017e7a7 <do_con_write+e3>
And finally:
And finally::
> objdump --full-contents --section=__ex_table vmlinux
>
> c01aa7c4 93c017c0 e09f19c0 97c017c0 99c017c0 ................
> c01aa7d4 f6c217c0 e99f19c0 a5e717c0 f59f19c0 ................
> c01aa7e4 080a18c0 01a019c0 0a0a18c0 04a019c0 ................
or in human readable byte order:
or in human readable byte order::
> c01aa7c4 c017c093 c0199fe0 c017c097 c017c099 ................
> c01aa7d4 c017c2f6 c0199fe9 c017e7a5 c0199ff5 ................
......@@ -222,18 +229,22 @@ or in human readable byte order:
this is the interesting part!
> c01aa7e4 c0180a08 c019a001 c0180a0a c019a004 ................
What happened? The assembly directives
What happened? The assembly directives::
.section .fixup,"ax"
.section __ex_table,"a"
.section .fixup,"ax"
.section __ex_table,"a"
told the assembler to move the following code to the specified
sections in the ELF object file. So the instructions
3: movl $-14,%eax
sections in the ELF object file. So the instructions::
3: movl $-14,%eax
xorb %dl,%dl
jmp 2b
ended up in the .fixup section of the object file and the addresses
ended up in the .fixup section of the object file and the addresses::
.long 1b,3b
ended up in the __ex_table section of the object file. 1b and 3b
are local labels. The local label 1b (1b stands for next label 1
backward) is the address of the instruction that might fault, i.e.
......@@ -246,34 +257,38 @@ the fault, in our case the actual value is c0199ff5:
the original assembly code: > 3: movl $-14,%eax
and linked in vmlinux : > c0199ff5 <.fixup+10b5> movl $0xfffffff2,%eax
The assembly code
The assembly code::
> .section __ex_table,"a"
> .align 4
> .long 1b,3b
becomes the value pair
becomes the value pair::
> c01aa7d4 c017c2f6 c0199fe9 c017e7a5 c0199ff5 ................
^this is ^this is
1b 3b
c017e7a5,c0199ff5 in the exception table of the kernel.
So, what actually happens if a fault from kernel mode with no suitable
vma occurs?
1.) access to invalid address:
#. access to invalid address::
> c017e7a5 <do_con_write+e1> movb (%ebx),%dl
2.) MMU generates exception
3.) CPU calls do_page_fault
4.) do page fault calls search_exception_table (regs->eip == c017e7a5);
5.) search_exception_table looks up the address c017e7a5 in the
#. MMU generates exception
#. CPU calls do_page_fault
#. do page fault calls search_exception_table (regs->eip == c017e7a5);
#. search_exception_table looks up the address c017e7a5 in the
exception table (i.e. the contents of the ELF section __ex_table)
and returns the address of the associated fault handle code c0199ff5.
6.) do_page_fault modifies its own return address to point to the fault
#. do_page_fault modifies its own return address to point to the fault
handle code and returns.
7.) execution continues in the fault handling code.
8.) 8a) EAX becomes -EFAULT (== -14)
8b) DL becomes zero (the value we "read" from user space)
8c) execution continues at local label 2 (address of the
#. execution continues in the fault handling code.
#. a) EAX becomes -EFAULT (== -14)
b) DL becomes zero (the value we "read" from user space)
c) execution continues at local label 2 (address of the
instruction immediately after the faulting user access).
The steps 8a to 8c in a certain way emulate the faulting instruction.
......@@ -295,14 +310,15 @@ Things changed when 64-bit support was added to x86 Linux. Rather than
double the size of the exception table by expanding the two entries
from 32-bits to 64 bits, a clever trick was used to store addresses
as relative offsets from the table itself. The assembly code changed
from:
from::
.long 1b,3b
to:
to:
.long (from) - .
.long (to) - .
and the C-code that uses these values converts back to absolute addresses
like this:
like this::
ex_insn_addr(const struct exception_table_entry *x)
{
......@@ -313,15 +329,18 @@ In v4.6 the exception table entry was expanded with a new field "handler".
This is also 32-bits wide and contains a third relative function
pointer which points to one of:
1) int ex_handler_default(const struct exception_table_entry *fixup)
1) ``int ex_handler_default(const struct exception_table_entry *fixup)``
This is legacy case that just jumps to the fixup code
2) int ex_handler_fault(const struct exception_table_entry *fixup)
2) ``int ex_handler_fault(const struct exception_table_entry *fixup)``
This case provides the fault number of the trap that occurred at
entry->insn. It is used to distinguish page faults from machine
check.
3) int ex_handler_ext(const struct exception_table_entry *fixup)
3) ``int ex_handler_ext(const struct exception_table_entry *fixup)``
This case is used for uaccess_err ... we need to set a flag
in the task structure. Before the handler functions existed this
case was handled by adding a large offset to the fixup to tag
it as special.
More functions can easily be added.
.. SPDX-License-Identifier: GPL-2.0
=======
IO-APIC
=======
:Author: Ingo Molnar <mingo@kernel.org>
Most (all) Intel-MP compliant SMP boards have the so-called 'IO-APIC',
which is an enhanced interrupt controller. It enables us to route
hardware interrupts to multiple CPUs, or to CPU groups. Without an
......@@ -13,9 +21,8 @@ usually worked around by the kernel. If your MP-compliant SMP board does
not boot Linux, then consult the linux-smp mailing list archives first.
If your box boots fine with enabled IO-APIC IRQs, then your
/proc/interrupts will look like this one:
/proc/interrupts will look like this one::
---------------------------->
hell:~> cat /proc/interrupts
CPU0
0: 1360293 IO-APIC-edge timer
......@@ -28,7 +35,6 @@ If your box boots fine with enabled IO-APIC IRQs, then your
NMI: 0
ERR: 0
hell:~>
<----------------------------
Some interrupts are still listed as 'XT PIC', but this is not a problem;
none of those IRQ sources is performance-critical.
......@@ -37,14 +43,14 @@ none of those IRQ sources is performance-critical.
In the unlikely case that your board does not create a working mp-table,
you can use the pirq= boot parameter to 'hand-construct' IRQ entries. This
is non-trivial though and cannot be automated. One sample /etc/lilo.conf
entry:
entry::
append="pirq=15,11,10"
The actual numbers depend on your system, on your PCI cards and on their
PCI slot position. Usually PCI slots are 'daisy chained' before they are
connected to the PCI chipset IRQ routing facility (the incoming PIRQ1-4
lines):
lines)::
,-. ,-. ,-. ,-. ,-.
PIRQ4 ----| |-. ,-| |-. ,-| |-. ,-| |--------| |
......@@ -56,7 +62,7 @@ lines):
PIRQ1 ----| |- `----| |- `----| |- `----| |--------| |
`-' `-' `-' `-' `-'
Every PCI card emits a PCI IRQ, which can be INTA, INTB, INTC or INTD:
Every PCI card emits a PCI IRQ, which can be INTA, INTB, INTC or INTD::
,-.
INTD--| |
......@@ -78,19 +84,19 @@ to have non shared interrupts). Slot5 should be used for videocards, they
do not use interrupts normally, thus they are not daisy chained either.
so if you have your SCSI card (IRQ11) in Slot1, Tulip card (IRQ9) in
Slot2, then you'll have to specify this pirq= line:
Slot2, then you'll have to specify this pirq= line::
append="pirq=11,9"
the following script tries to figure out such a default pirq= line from
your PCI configuration:
your PCI configuration::
echo -n pirq=; echo `scanpci | grep T_L | cut -c56-` | sed 's/ /,/g'
note that this script won't work if you have skipped a few slots or if your
board does not do default daisy-chaining. (or the IO-APIC has the PIRQ pins
connected in some strange way). E.g. if in the above case you have your SCSI
card (IRQ11) in Slot3, and have Slot1 empty:
card (IRQ11) in Slot3, and have Slot1 empty::
append="pirq=0,9,11"
......@@ -105,7 +111,7 @@ won't function properly (e.g. if it's inserted as a module).
If you have 2 PCI buses, then you can use up to 8 pirq values, although such
boards tend to have a good configuration.
Be prepared that it might happen that you need some strange pirq line:
Be prepared that it might happen that you need some strange pirq line::
append="pirq=0,0,0,0,0,0,9,11"
......@@ -115,5 +121,3 @@ Good luck and mail to linux-smp@vger.kernel.org or
linux-kernel@vger.kernel.org if you have any problems that are not covered
by this document.
-- mingo
.. SPDX-License-Identifier: GPL-2.0
============
i386 Support
============
.. toctree::
:maxdepth: 2
IO-APIC
.. SPDX-License-Identifier: GPL-2.0
==========================
x86-specific Documentation
==========================
.. toctree::
:maxdepth: 2
:numbered:
boot
topology
exception-tables
kernel-stacks
entry_64
earlyprintk
orc-unwinder
zero-page
tlb
mtrr
pat
protection-keys
intel_mpx
amd-memory-encryption
pti
microcode
resctrl_ui
usb-legacy-support
i386/index
x86_64/index
1. Intel(R) MPX Overview
========================
.. SPDX-License-Identifier: GPL-2.0
===========================================
Intel(R) Memory Protection Extensions (MPX)
===========================================
Intel(R) MPX Overview
=====================
Intel(R) Memory Protection Extensions (Intel(R) MPX) is a new capability
introduced into Intel Architecture. Intel MPX provides hardware features
......@@ -7,7 +13,7 @@ that can be used in conjunction with compiler changes to check memory
references, for those references whose compile-time normal intentions are
usurped at runtime due to buffer overflow or underflow.
You can tell if your CPU supports MPX by looking in /proc/cpuinfo:
You can tell if your CPU supports MPX by looking in /proc/cpuinfo::
cat /proc/cpuinfo | grep ' mpx '
......@@ -21,8 +27,8 @@ can be downloaded from
http://software.intel.com/en-us/articles/intel-software-development-emulator
2. How to get the advantage of MPX
==================================
How to get the advantage of MPX
===============================
For MPX to work, changes are required in the kernel, binutils and compiler.
No source changes are required for applications, just a recompile.
......@@ -84,14 +90,15 @@ Kernel MPX Code:
is unmapped.
3. How does MPX kernel code work
================================
How does MPX kernel code work
=============================
Handling #BR faults caused by MPX
---------------------------------
When MPX is enabled, there are 2 new situations that can generate
#BR faults.
* new bounds tables (BT) need to be allocated to save bounds.
* bounds violation caused by MPX instructions.
......@@ -124,9 +131,9 @@ the kernel. It can theoretically be done completely from userspace. Here
are a few ways this could be done. We don't think any of them are practical
in the real-world, but here they are.
Q: Can virtual space simply be reserved for the bounds tables so that we
:Q: Can virtual space simply be reserved for the bounds tables so that we
never have to allocate them?
A: MPX-enabled application will possibly create a lot of bounds tables in
:A: MPX-enabled application will possibly create a lot of bounds tables in
process address space to save bounds information. These tables can take
up huge swaths of memory (as much as 80% of the memory on the system)
even if we clean them up aggressively. In the worst-case scenario, the
......@@ -140,19 +147,19 @@ A: MPX-enabled application will possibly create a lot of bounds tables in
consumes 2GB of virtual *AND* physical memory. IOW, it's completely
infeasible to prepopulate bounds directories.
Q: Can we preallocate bounds table space at the same time memory is
:Q: Can we preallocate bounds table space at the same time memory is
allocated which might contain pointers that might eventually need
bounds tables?
A: This would work if we could hook the site of each and every memory
:A: This would work if we could hook the site of each and every memory
allocation syscall. This can be done for small, constrained applications.
But, it isn't practical at a larger scale since a given app has no
way of controlling how all the parts of the app might allocate memory
(think libraries). The kernel is really the only place to intercept
these calls.
Q: Could a bounds fault be handed to userspace and the tables allocated
:Q: Could a bounds fault be handed to userspace and the tables allocated
there in a signal handler instead of in the kernel?
A: mmap() is not on the list of safe async handler functions and even
:A: mmap() is not on the list of safe async handler functions and even
if mmap() would work it still requires locking or nasty tricks to
keep track of the allocation state there.
......@@ -167,20 +174,20 @@ If a #BR is generated due to a bounds violation caused by MPX.
We need to decode MPX instructions to get violation address and
set this address into extended struct siginfo.
The _sigfault field of struct siginfo is extended as follow:
87 /* SIGILL, SIGFPE, SIGSEGV, SIGBUS */
88 struct {
89 void __user *_addr; /* faulting insn/memory ref. */
90 #ifdef __ARCH_SI_TRAPNO
91 int _trapno; /* TRAP # which caused the signal */
92 #endif
93 short _addr_lsb; /* LSB of the reported address */
94 struct {
95 void __user *_lower;
96 void __user *_upper;
97 } _addr_bnd;
98 } _sigfault;
The _sigfault field of struct siginfo is extended as follow::
87 /* SIGILL, SIGFPE, SIGSEGV, SIGBUS */
88 struct {
89 void __user *_addr; /* faulting insn/memory ref. */
90 #ifdef __ARCH_SI_TRAPNO
91 int _trapno; /* TRAP # which caused the signal */
92 #endif
93 short _addr_lsb; /* LSB of the reported address */
94 struct {
95 void __user *_lower;
96 void __user *_upper;
97 } _addr_bnd;
98 } _sigfault;
The '_addr' field refers to violation address, and new '_addr_and'
field refers to the upper/lower bounds when a #BR is caused.
......@@ -209,9 +216,10 @@ Adding new prctl commands
Two new prctl commands are added to enable and disable MPX bounds tables
management in kernel.
::
155 #define PR_MPX_ENABLE_MANAGEMENT 43
156 #define PR_MPX_DISABLE_MANAGEMENT 44
155 #define PR_MPX_ENABLE_MANAGEMENT 43
156 #define PR_MPX_DISABLE_MANAGEMENT 44
Runtime library in userspace is responsible for allocation of bounds
directory. So kernel have to use XSAVE instruction to get the base
......@@ -223,8 +231,8 @@ into struct mm_struct to be used in future during PR_MPX_ENABLE_MANAGEMENT
command execution.
4. Special rules
================
Special rules
=============
1) If userspace is requesting help from the kernel to do the management
of bounds tables, it may not create or modify entries in the bounds directory.
......
.. SPDX-License-Identifier: GPL-2.0
=============
Kernel Stacks
=============
Kernel stacks on x86-64 bit
---------------------------
===========================
Most of the text from Keith Owens, hacked by AK
......@@ -57,7 +63,7 @@ IST events with the same code to be nested. However in most cases, the
stack size allocated to an IST assumes no nesting for the same code.
If that assumption is ever broken then the stacks will become corrupt.
The currently assigned IST stacks are :-
The currently assigned IST stacks are:
* ESTACK_DF. EXCEPTION_STKSZ (PAGE_SIZE).
......@@ -103,7 +109,7 @@ For more details see the Intel IA32 or AMD AMD64 architecture manuals.
Printing backtraces on x86
--------------------------
==========================
The question about the '?' preceding function names in an x86 stacktrace
keeps popping up, here's an indepth explanation. It helps if the reader
......@@ -113,7 +119,7 @@ arch/x86/kernel/dumpstack.c.
Adapted from Ingo's mail, Message-ID: <20150521101614.GA10889@gmail.com>:
We always scan the full kernel stack for return addresses stored on
the kernel stack(s) [*], from stack top to stack bottom, and print out
the kernel stack(s) [1]_, from stack top to stack bottom, and print out
anything that 'looks like' a kernel text address.
If it fits into the frame pointer chain, we print it without a question
......@@ -141,6 +147,6 @@ that look like kernel text addresses, so if debug information is wrong,
we still print out the real call chain as well - just with more question
marks than ideal.
[*] For things like IRQ and IST stacks, we also scan those stacks, in
.. [1] For things like IRQ and IST stacks, we also scan those stacks, in
the right order, and try to cross from one stack into another
reconstructing the call chain. This works most of the time.
The Linux Microcode Loader
.. SPDX-License-Identifier: GPL-2.0
Authors: Fenghua Yu <fenghua.yu@intel.com>
Borislav Petkov <bp@suse.de>
==========================
The Linux Microcode Loader
==========================
:Authors: - Fenghua Yu <fenghua.yu@intel.com>
- Borislav Petkov <bp@suse.de>
The kernel has a x86 microcode loading facility which is supposed to
provide microcode loading methods in the OS. Potential use cases are
......@@ -10,8 +14,8 @@ and updating the microcode on long-running systems without rebooting.
The loader supports three loading methods:
1. Early load microcode
=======================
Early load microcode
====================
The kernel can update microcode very early during boot. Loading
microcode early can fix CPU issues before they are observed during
......@@ -26,8 +30,10 @@ loader parses the combined initrd image during boot.
The microcode files in cpio name space are:
on Intel: kernel/x86/microcode/GenuineIntel.bin
on AMD : kernel/x86/microcode/AuthenticAMD.bin
on Intel:
kernel/x86/microcode/GenuineIntel.bin
on AMD :
kernel/x86/microcode/AuthenticAMD.bin
During BSP (BootStrapping Processor) boot (pre-SMP), the kernel
scans the microcode file in the initrd. If microcode matching the
......@@ -42,8 +48,8 @@ Here's a crude example how to prepare an initrd with microcode (this is
normally done automatically by the distribution, when recreating the
initrd, so you don't really have to do it yourself. It is documented
here for future reference only).
::
---
#!/bin/bash
if [ -z "$1" ]; then
......@@ -76,15 +82,15 @@ here for future reference only).
cat ucode.cpio $INITRD.orig > $INITRD
rm -rf $TMPDIR
---
The system needs to have the microcode packages installed into
/lib/firmware or you need to fixup the paths above if yours are
somewhere else and/or you've downloaded them directly from the processor
vendor's site.
2. Late loading
===============
Late loading
============
There are two legacy user space interfaces to load microcode, either through
/dev/cpu/microcode or through /sys/devices/system/cpu/microcode/reload file
......@@ -94,9 +100,9 @@ The /dev/cpu/microcode method is deprecated because it needs a special
userspace tool for that.
The easier method is simply installing the microcode packages your distro
supplies and running:
supplies and running::
# echo 1 > /sys/devices/system/cpu/microcode/reload
# echo 1 > /sys/devices/system/cpu/microcode/reload
as root.
......@@ -104,29 +110,29 @@ The loading mechanism looks for microcode blobs in
/lib/firmware/{intel-ucode,amd-ucode}. The default distro installation
packages already put them there.
3. Builtin microcode
====================
Builtin microcode
=================
The loader supports also loading of a builtin microcode supplied through
the regular builtin firmware method CONFIG_EXTRA_FIRMWARE. Only 64-bit is
currently supported.
Here's an example:
Here's an example::
CONFIG_EXTRA_FIRMWARE="intel-ucode/06-3a-09 amd-ucode/microcode_amd_fam15h.bin"
CONFIG_EXTRA_FIRMWARE_DIR="/lib/firmware"
CONFIG_EXTRA_FIRMWARE="intel-ucode/06-3a-09 amd-ucode/microcode_amd_fam15h.bin"
CONFIG_EXTRA_FIRMWARE_DIR="/lib/firmware"
This basically means, you have the following tree structure locally:
This basically means, you have the following tree structure locally::
/lib/firmware/
|-- amd-ucode
...
| |-- microcode_amd_fam15h.bin
...
|-- intel-ucode
...
| |-- 06-3a-09
...
/lib/firmware/
|-- amd-ucode
...
| |-- microcode_amd_fam15h.bin
...
|-- intel-ucode
...
| |-- 06-3a-09
...
so that the build system can find those files and integrate them into
the final kernel image. The early loader finds them and applies them.
......
.. SPDX-License-Identifier: GPL-2.0
============
ORC unwinder
============
Overview
--------
========
The kernel CONFIG_UNWINDER_ORC option enables the ORC unwinder, which is
similar in concept to a DWARF unwinder. The difference is that the
......@@ -23,12 +26,12 @@ correlate instruction addresses with their stack states at run time.
ORC vs frame pointers
---------------------
=====================
With frame pointers enabled, GCC adds instrumentation code to every
function in the kernel. The kernel's .text size increases by about
3.2%, resulting in a broad kernel-wide slowdown. Measurements by Mel
Gorman [1] have shown a slowdown of 5-10% for some workloads.
Gorman [1]_ have shown a slowdown of 5-10% for some workloads.
In contrast, the ORC unwinder has no effect on text size or runtime
performance, because the debuginfo is out of band. So if you disable
......@@ -55,7 +58,7 @@ depending on the kernel config.
ORC vs DWARF
------------
============
ORC debuginfo's advantage over DWARF itself is that it's much simpler.
It gets rid of the complex DWARF CFI state machine and also gets rid of
......@@ -65,7 +68,7 @@ mission critical oops code.
The simpler debuginfo format also enables the unwinder to be much faster
than DWARF, which is important for perf and lockdep. In a basic
performance test by Jiri Slaby [2], the ORC unwinder was about 20x
performance test by Jiri Slaby [2]_, the ORC unwinder was about 20x
faster than an out-of-tree DWARF unwinder. (Note: That measurement was
taken before some performance tweaks were added, which doubled
performance, so the speedup over DWARF may be closer to 40x.)
......@@ -85,7 +88,7 @@ still be able to control the format, e.g. no complex state machines.
ORC unwind table generation
---------------------------
===========================
The ORC data is generated by objtool. With the existing compile-time
stack metadata validation feature, objtool already follows all code
......@@ -133,7 +136,7 @@ objtool follows GCC code quite well.
Unwinder implementation details
-------------------------------
===============================
Objtool generates the ORC data by integrating with the compile-time
stack metadata validation feature, which is described in detail in
......@@ -154,7 +157,7 @@ subset of the table needs to be searched.
Etymology
---------
=========
Orcs, fearsome creatures of medieval folklore, are the Dwarves' natural
enemies. Similarly, the ORC unwinder was created in opposition to the
......@@ -162,7 +165,7 @@ complexity and slowness of DWARF.
"Although Orcs rarely consider multiple solutions to a problem, they do
excel at getting things done because they are creatures of action, not
thought." [3] Similarly, unlike the esoteric DWARF unwinder, the
thought." [3]_ Similarly, unlike the esoteric DWARF unwinder, the
veracious ORC unwinder wastes no time or siloconic effort decoding
variable-length zero-extended unsigned-integer byte-coded
state-machine-based debug information entries.
......@@ -174,6 +177,6 @@ brutal, unyielding efficiency.
ORC stands for Oops Rewind Capability.
[1] https://lkml.kernel.org/r/20170602104048.jkkzssljsompjdwy@suse.de
[2] https://lkml.kernel.org/r/d2ca5435-6386-29b8-db87-7f227c2b713a@suse.cz
[3] http://dustin.wikidot.com/half-orcs-and-orcs
.. [1] https://lkml.kernel.org/r/20170602104048.jkkzssljsompjdwy@suse.de
.. [2] https://lkml.kernel.org/r/d2ca5435-6386-29b8-db87-7f227c2b713a@suse.cz
.. [3] http://dustin.wikidot.com/half-orcs-and-orcs
.. SPDX-License-Identifier: GPL-2.0
======================
Memory Protection Keys
======================
Memory Protection Keys for Userspace (PKU aka PKEYs) is a feature
which is found on Intel's Skylake "Scalable Processor" Server CPUs.
It will be avalable in future non-server parts.
......@@ -23,9 +29,10 @@ even though there is theoretically space in the PAE PTEs. These
permissions are enforced on data access only and have no effect on
instruction fetches.
=========================== Syscalls ===========================
Syscalls
========
There are 3 system calls which directly interact with pkeys:
There are 3 system calls which directly interact with pkeys::
int pkey_alloc(unsigned long flags, unsigned long init_access_rights)
int pkey_free(int pkey);
......@@ -37,6 +44,7 @@ pkey_alloc(). An application calls the WRPKRU instruction
directly in order to change access permissions to memory covered
with a key. In this example WRPKRU is wrapped by a C function
called pkey_set().
::
int real_prot = PROT_READ|PROT_WRITE;
pkey = pkey_alloc(0, PKEY_DISABLE_WRITE);
......@@ -45,43 +53,44 @@ called pkey_set().
... application runs here
Now, if the application needs to update the data at 'ptr', it can
gain access, do the update, then remove its write access:
gain access, do the update, then remove its write access::
pkey_set(pkey, 0); // clear PKEY_DISABLE_WRITE
*ptr = foo; // assign something
pkey_set(pkey, PKEY_DISABLE_WRITE); // set PKEY_DISABLE_WRITE again
Now when it frees the memory, it will also free the pkey since it
is no longer in use:
is no longer in use::
munmap(ptr, PAGE_SIZE);
pkey_free(pkey);
(Note: pkey_set() is a wrapper for the RDPKRU and WRPKRU instructions.
.. note:: pkey_set() is a wrapper for the RDPKRU and WRPKRU instructions.
An example implementation can be found in
tools/testing/selftests/x86/protection_keys.c)
tools/testing/selftests/x86/protection_keys.c.
=========================== Behavior ===========================
Behavior
========
The kernel attempts to make protection keys consistent with the
behavior of a plain mprotect(). For instance if you do this:
behavior of a plain mprotect(). For instance if you do this::
mprotect(ptr, size, PROT_NONE);
something(ptr);
you can expect the same effects with protection keys when doing this:
you can expect the same effects with protection keys when doing this::
pkey = pkey_alloc(0, PKEY_DISABLE_WRITE | PKEY_DISABLE_READ);
pkey_mprotect(ptr, size, PROT_READ|PROT_WRITE, pkey);
something(ptr);
That should be true whether something() is a direct access to 'ptr'
like:
like::
*ptr = foo;
or when the kernel does the access on the application's behalf like
with a read():
with a read()::
read(fd, ptr, 1);
......
.. SPDX-License-Identifier: GPL-2.0
==========================
Page Table Isolation (PTI)
==========================
Overview
========
Page Table Isolation (pti, previously known as KAISER[1]) is a
Page Table Isolation (pti, previously known as KAISER [1]_) is a
countermeasure against attacks on the shared user/kernel address
space such as the "Meltdown" approach[2].
space such as the "Meltdown" approach [2]_.
To mitigate this class of attacks, we create an independent set of
page tables for use only when running userspace applications. When
......@@ -60,6 +66,7 @@ Protection against side-channel attacks is important. But,
this protection comes at a cost:
1. Increased Memory Use
a. Each process now needs an order-1 PGD instead of order-0.
(Consumes an additional 4k per process).
b. The 'cpu_entry_area' structure must be 2MB in size and 2MB
......@@ -68,6 +75,7 @@ this protection comes at a cost:
is decompressed, but no space in the kernel image itself.
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,
......@@ -142,6 +150,7 @@ ideally doing all of these in parallel:
interrupted, including nested NMIs. Using "-c" boosts the rate of
NMIs, and using two -c with separate counters encourages nested NMIs
and less deterministic behavior.
::
while true; do perf record -c 10000 -e instructions,cycles -a sleep 10; done
......@@ -182,5 +191,5 @@ that are worth noting here.
tended to be TLB invalidation issues. Usually invalidating
the wrong PCID, or otherwise missing an invalidation.
1. https://gruss.cc/files/kaiser.pdf
2. https://meltdownattack.com/meltdown.pdf
.. [1] https://gruss.cc/files/kaiser.pdf
.. [2] https://meltdownattack.com/meltdown.pdf
.. SPDX-License-Identifier: GPL-2.0
=======
The TLB
=======
When the kernel unmaps or modified the attributes of a range of
memory, it has two choices:
1. Flush the entire TLB with a two-instruction sequence. This is
a quick operation, but it causes collateral damage: TLB entries
from areas other than the one we are trying to flush will be
......@@ -10,6 +17,7 @@ memory, it has two choices:
damage to other TLB entries.
Which method to do depends on a few things:
1. The size of the flush being performed. A flush of the entire
address space is obviously better performed by flushing the
entire TLB than doing 2^48/PAGE_SIZE individual flushes.
......@@ -33,7 +41,7 @@ well. There is essentially no "right" point to choose.
You may be doing too many individual invalidations if you see the
invlpg instruction (or instructions _near_ it) show up high in
profiles. If you believe that individual invalidations being
called too often, you can lower the tunable:
called too often, you can lower the tunable::
/sys/kernel/debug/x86/tlb_single_page_flush_ceiling
......@@ -43,7 +51,7 @@ Setting it to 1 is a very conservative setting and it should
never need to be 0 under normal circumstances.
Despite the fact that a single individual flush on x86 is
guaranteed to flush a full 2MB [1], hugetlbfs always uses the full
guaranteed to flush a full 2MB [1]_, hugetlbfs always uses the full
flushes. THP is treated exactly the same as normal memory.
You might see invlpg inside of flush_tlb_mm_range() show up in
......@@ -54,9 +62,9 @@ Essentially, you are balancing the cycles you spend doing invlpg
with the cycles that you spend refilling the TLB later.
You can measure how expensive TLB refills are by using
performance counters and 'perf stat', like this:
performance counters and 'perf stat', like this::
perf stat -e
perf stat -e
cpu/event=0x8,umask=0x84,name=dtlb_load_misses_walk_duration/,
cpu/event=0x8,umask=0x82,name=dtlb_load_misses_walk_completed/,
cpu/event=0x49,umask=0x4,name=dtlb_store_misses_walk_duration/,
......@@ -70,6 +78,6 @@ be there in some form. You can use pmu-tools 'ocperf list'
(https://github.com/andikleen/pmu-tools) to find the right
counters for a given CPU.
1. A footnote in Intel's SDM "4.10.4.2 Recommended Invalidation"
.. [1] A footnote in Intel's SDM "4.10.4.2 Recommended Invalidation"
says: "One execution of INVLPG is sufficient even for a page
with size greater than 4 KBytes."
.. SPDX-License-Identifier: GPL-2.0
============
x86 Topology
============
......@@ -33,14 +36,14 @@ The topology of a system is described in the units of:
- cores
- threads
* Package:
Package
=======
Packages contain a number of cores plus shared resources, e.g. DRAM
controller, shared caches etc.
Packages contain a number of cores plus shared resources, e.g. DRAM
controller, shared caches etc.
AMD nomenclature for package is 'Node'.
AMD nomenclature for package is 'Node'.
Package-related topology information in the kernel:
Package-related topology information in the kernel:
- cpuinfo_x86.x86_max_cores:
......@@ -66,6 +69,7 @@ The topology of a system is described in the units of:
- cpu_llc_id:
A per-CPU variable containing:
- On Intel, the first APIC ID of the list of CPUs sharing the Last Level
Cache
......@@ -73,33 +77,33 @@ The topology of a system is described in the units of:
Cache. In general, it is a number identifying an LLC uniquely on the
system.
* Cores:
Cores
=====
A core consists of 1 or more threads. It does not matter whether the threads
are SMT- or CMT-type threads.
A core consists of 1 or more threads. It does not matter whether the threads
are SMT- or CMT-type threads.
AMDs nomenclature for a CMT core is "Compute Unit". The kernel always uses
"core".
AMDs nomenclature for a CMT core is "Compute Unit". The kernel always uses
"core".
Core-related topology information in the kernel:
Core-related topology information in the kernel:
- smp_num_siblings:
The number of threads in a core. The number of threads in a package can be
calculated by:
calculated by::
threads_per_package = cpuinfo_x86.x86_max_cores * smp_num_siblings
* Threads:
A thread is a single scheduling unit. It's the equivalent to a logical Linux
CPU.
Threads
=======
A thread is a single scheduling unit. It's the equivalent to a logical Linux
CPU.
AMDs nomenclature for CMT threads is "Compute Unit Core". The kernel always
uses "thread".
AMDs nomenclature for CMT threads is "Compute Unit Core". The kernel always
uses "thread".
Thread-related topology information in the kernel:
Thread-related topology information in the kernel:
- topology_core_cpumask():
......@@ -129,41 +133,41 @@ The topology of a system is described in the units of:
System topology examples
========================
Note:
The alternative Linux CPU enumeration depends on how the BIOS enumerates the
threads. Many BIOSes enumerate all threads 0 first and then all threads 1.
That has the "advantage" that the logical Linux CPU numbers of threads 0 stay
the same whether threads are enabled or not. That's merely an implementation
detail and has no practical impact.
.. note::
The alternative Linux CPU enumeration depends on how the BIOS enumerates the
threads. Many BIOSes enumerate all threads 0 first and then all threads 1.
That has the "advantage" that the logical Linux CPU numbers of threads 0 stay
the same whether threads are enabled or not. That's merely an implementation
detail and has no practical impact.
1) Single Package, Single Core
1) Single Package, Single Core::
[package 0] -> [core 0] -> [thread 0] -> Linux CPU 0
2) Single Package, Dual Core
a) One thread per core
a) One thread per core::
[package 0] -> [core 0] -> [thread 0] -> Linux CPU 0
-> [core 1] -> [thread 0] -> Linux CPU 1
b) Two threads per core
b) Two threads per core::
[package 0] -> [core 0] -> [thread 0] -> Linux CPU 0
-> [thread 1] -> Linux CPU 1
-> [core 1] -> [thread 0] -> Linux CPU 2
-> [thread 1] -> Linux CPU 3
Alternative enumeration:
Alternative enumeration::
[package 0] -> [core 0] -> [thread 0] -> Linux CPU 0
-> [thread 1] -> Linux CPU 2
-> [core 1] -> [thread 0] -> Linux CPU 1
-> [thread 1] -> Linux CPU 3
AMD nomenclature for CMT systems:
AMD nomenclature for CMT systems::
[node 0] -> [Compute Unit 0] -> [Compute Unit Core 0] -> Linux CPU 0
-> [Compute Unit Core 1] -> Linux CPU 1
......@@ -172,7 +176,7 @@ detail and has no practical impact.
4) Dual Package, Dual Core
a) One thread per core
a) One thread per core::
[package 0] -> [core 0] -> [thread 0] -> Linux CPU 0
-> [core 1] -> [thread 0] -> Linux CPU 1
......@@ -180,7 +184,7 @@ detail and has no practical impact.
[package 1] -> [core 0] -> [thread 0] -> Linux CPU 2
-> [core 1] -> [thread 0] -> Linux CPU 3
b) Two threads per core
b) Two threads per core::
[package 0] -> [core 0] -> [thread 0] -> Linux CPU 0
-> [thread 1] -> Linux CPU 1
......@@ -192,7 +196,7 @@ detail and has no practical impact.
-> [core 1] -> [thread 0] -> Linux CPU 6
-> [thread 1] -> Linux CPU 7
Alternative enumeration:
Alternative enumeration::
[package 0] -> [core 0] -> [thread 0] -> Linux CPU 0
-> [thread 1] -> Linux CPU 4
......@@ -204,7 +208,7 @@ detail and has no practical impact.
-> [core 1] -> [thread 0] -> Linux CPU 3
-> [thread 1] -> Linux CPU 7
AMD nomenclature for CMT systems:
AMD nomenclature for CMT systems::
[node 0] -> [Compute Unit 0] -> [Compute Unit Core 0] -> Linux CPU 0
-> [Compute Unit Core 1] -> Linux CPU 1
......
.. SPDX-License-Identifier: GPL-2.0
==================
USB Legacy support
~~~~~~~~~~~~~~~~~~
==================
Vojtech Pavlik <vojtech@suse.cz>, January 2004
:Author: Vojtech Pavlik <vojtech@suse.cz>, January 2004
Also known as "USB Keyboard" or "USB Mouse support" in the BIOS Setup is a
......@@ -27,18 +31,20 @@ It has several drawbacks, though:
Solutions:
Problem 1) can be solved by loading the USB drivers prior to loading the
PS/2 mouse driver. Since the PS/2 mouse driver is in 2.6 compiled into
the kernel unconditionally, this means the USB drivers need to be
compiled-in, too.
Problem 2) can currently only be solved by either disabling HIGHMEM64G
in the kernel config or USB Legacy support in the BIOS. A BIOS update
could help, but so far no such update exists.
Problem 3) is usually fixed by a BIOS update. Check the board
manufacturers web site. If an update is not available, disable USB
Legacy support in the BIOS. If this alone doesn't help, try also adding
idle=poll on the kernel command line. The BIOS may be entering the SMM
on the HLT instruction as well.
Problem 1)
can be solved by loading the USB drivers prior to loading the
PS/2 mouse driver. Since the PS/2 mouse driver is in 2.6 compiled into
the kernel unconditionally, this means the USB drivers need to be
compiled-in, too.
Problem 2)
can currently only be solved by either disabling HIGHMEM64G
in the kernel config or USB Legacy support in the BIOS. A BIOS update
could help, but so far no such update exists.
Problem 3)
is usually fixed by a BIOS update. Check the board
manufacturers web site. If an update is not available, disable USB
Legacy support in the BIOS. If this alone doesn't help, try also adding
idle=poll on the kernel command line. The BIOS may be entering the SMM
on the HLT instruction as well.
== Overview ==
.. SPDX-License-Identifier: GPL-2.0
==============
5-level paging
==============
Overview
========
Original x86-64 was limited by 4-level paing to 256 TiB of virtual address
space and 64 TiB of physical address space. We are already bumping into
this limit: some vendors offers servers with 64 TiB of memory today.
......@@ -16,16 +22,17 @@ QEMU 2.9 and later support 5-level paging.
Virtual memory layout for 5-level paging is described in
Documentation/x86/x86_64/mm.txt
== Enabling 5-level paging ==
Enabling 5-level paging
=======================
CONFIG_X86_5LEVEL=y enables the feature.
Kernel with CONFIG_X86_5LEVEL=y still able to boot on 4-level hardware.
In this case additional page table level -- p4d -- will be folded at
runtime.
== User-space and large virtual address space ==
User-space and large virtual address space
==========================================
On x86, 5-level paging enables 56-bit userspace virtual address space.
Not all user space is ready to handle wide addresses. It's known that
at least some JIT compilers use higher bits in pointers to encode their
......@@ -58,4 +65,3 @@ One important case we need to handle here is interaction with MPX.
MPX (without MAWA extension) cannot handle addresses above 47-bit, so we
need to make sure that MPX cannot be enabled we already have VMA above
the boundary and forbid creating such VMAs once MPX is enabled.
.. SPDX-License-Identifier: GPL-2.0
===================================================
Firmware support for CPU hotplug under Linux/x86-64
---------------------------------------------------
===================================================
Linux/x86-64 supports CPU hotplug now. For various reasons Linux wants to
know in advance of boot time the maximum number of CPUs that could be plugged
......
.. SPDX-License-Identifier: GPL-2.0
=====================
Fake NUMA For CPUSets
=====================
:Author: David Rientjes <rientjes@cs.washington.edu>
Using numa=fake and CPUSets for Resource Management
Written by David Rientjes <rientjes@cs.washington.edu>
This document describes how the numa=fake x86_64 command-line option can be used
in conjunction with cpusets for coarse memory management. Using this feature,
......@@ -20,7 +27,7 @@ you become more familiar with using this combination for resource control,
you'll determine a better setup to minimize the number of nodes you have to deal
with.
A machine may be split as follows with "numa=fake=4*512," as reported by dmesg:
A machine may be split as follows with "numa=fake=4*512," as reported by dmesg::
Faking node 0 at 0000000000000000-0000000020000000 (512MB)
Faking node 1 at 0000000020000000-0000000040000000 (512MB)
......@@ -34,7 +41,7 @@ A machine may be split as follows with "numa=fake=4*512," as reported by dmesg:
Now following the instructions for mounting the cpusets filesystem from
Documentation/cgroup-v1/cpusets.txt, you can assign fake nodes (i.e. contiguous memory
address spaces) to individual cpusets:
address spaces) to individual cpusets::
[root@xroads /]# mkdir exampleset
[root@xroads /]# mount -t cpuset none exampleset
......@@ -47,7 +54,7 @@ Now this cpuset, 'ddset', will only allowed access to fake nodes 0 and 1 for
memory allocations (1G).
You can now assign tasks to these cpusets to limit the memory resources
available to them according to the fake nodes assigned as mems:
available to them according to the fake nodes assigned as mems::
[root@xroads /exampleset/ddset]# echo $$ > tasks
[root@xroads /exampleset/ddset]# dd if=/dev/zero of=tmp bs=1024 count=1G
......@@ -57,9 +64,13 @@ Notice the difference between the system memory usage as reported by
/proc/meminfo between the restricted cpuset case above and the unrestricted
case (i.e. running the same 'dd' command without assigning it to a fake NUMA
cpuset):
Unrestricted Restricted
MemTotal: 3091900 kB 3091900 kB
MemFree: 42113 kB 1513236 kB
======== ============ ==========
Name Unrestricted Restricted
======== ============ ==========
MemTotal 3091900 kB 3091900 kB
MemFree 42113 kB 1513236 kB
======== ============ ==========
This allows for coarse memory management for the tasks you assign to particular
cpusets. Since cpusets can form a hierarchy, you can create some pretty
......
.. SPDX-License-Identifier: GPL-2.0
==============
x86_64 Support
==============
.. toctree::
:maxdepth: 2
boot-options
uefi
mm
5level-paging
fake-numa-for-cpusets
cpu-hotplug-spec
machinecheck
.. SPDX-License-Identifier: GPL-2.0
Configurable sysfs parameters for the x86-64 machine check code.
===============================================================
Configurable sysfs parameters for the x86-64 machine check code
===============================================================
Machine checks report internal hardware error conditions detected
by the CPU. Uncorrected errors typically cause a machine check
......@@ -16,14 +19,13 @@ log then mcelog should run to collect and decode machine check entries
from /dev/mcelog. Normally mcelog should be run regularly from a cronjob.
Each CPU has a directory in /sys/devices/system/machinecheck/machinecheckN
(N = CPU number)
(N = CPU number).
The directory contains some configurable entries:
Entries:
bankNctl
(N bank number)
(N bank number)
64bit Hex bitmask enabling/disabling specific subevents for bank N
When a bit in the bitmask is zero then the respective
subevent will not be reported.
......
This diff is collapsed.
This diff is collapsed.
.. SPDX-License-Identifier: GPL-2.0
=====================================
General note on [U]EFI x86_64 support
-------------------------------------
=====================================
The nomenclature EFI and UEFI are used interchangeably in this document.
......@@ -14,29 +17,42 @@ with EFI firmware and specifications are listed below.
3. x86_64 platform with EFI/UEFI firmware.
Mechanics:
Mechanics
---------
- Build the kernel with the following configuration.
- Build the kernel with the following configuration::
CONFIG_FB_EFI=y
CONFIG_FRAMEBUFFER_CONSOLE=y
If EFI runtime services are expected, the following configuration should
be selected.
be selected::
CONFIG_EFI=y
CONFIG_EFI_VARS=y or m # optional
- Create a VFAT partition on the disk
- Copy the following to the VFAT partition:
elilo bootloader with x86_64 support, elilo configuration file,
kernel image built in first step and corresponding
initrd. Instructions on building elilo and its dependencies
can be found in the elilo sourceforge project.
- Boot to EFI shell and invoke elilo choosing the kernel image built
in first step.
- If some or all EFI runtime services don't work, you can try following
kernel command line parameters to turn off some or all EFI runtime
services.
noefi turn off all EFI runtime services
reboot_type=k turn off EFI reboot runtime service
noefi
turn off all EFI runtime services
reboot_type=k
turn off EFI reboot runtime service
- If the EFI memory map has additional entries not in the E820 map,
you can include those entries in the kernels memory map of available
physical RAM by using the following kernel command line parameter.
add_efi_memmap include EFI memory map of available physical RAM
add_efi_memmap
include EFI memory map of available physical RAM
This diff is collapsed.
This diff is collapsed.
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment