It's x86 specific.
This is inspired by the commit listed below, but rewritten to match
upstream, and split in smaller pieces to keep intent clear.
Change-Id: Iacb91b47c89041435dd27c2c9ad34a231adf21d2
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Based-On-Change-Id: I50af7dacf616e0f8ff4c43f4acc679089ad7022b
Based-On-Signed-off-by: Julius Werner <jwerner@chromium.org>
Based-On-Reviewed-on: https://chromium-review.googlesource.com/219170
Reviewed-on: http://review.coreboot.org/9115
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
We have .lb, .lds, and .ld in the tree. Go for .ld everywhere.
This is inspired by the commit listed below, but rewritten to match
upstream, and split in smaller pieces to keep intent clear.
Change-Id: I3126af608afe4937ec4551a78df5a7824e09b04b
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Based-On-Change-Id: I50af7dacf616e0f8ff4c43f4acc679089ad7022b
Based-On-Signed-off-by: Julius Werner <jwerner@chromium.org>
Based-On-Reviewed-on: https://chromium-review.googlesource.com/219170
Reviewed-on: http://review.coreboot.org/9107
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Instead of two headers for payload and ramstage loading
combine the 2 files into one. This also allows for easier
refactoring by keeping header files consistent.
Change-Id: I4a6dffb78ad84c78e6e96c886d361413f9b4a17d
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/8708
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
From GCC's documentation:
Optimize debugging experience. -Og enables optimizations that do not interfere
with debugging. It should be the optimization level of choice for the standard
edit-compile-debug cycle, offering a reasonable level of optimization while
maintaining fast compilation and a good debugging experience.
Change-Id: I9a3dadbf8e894cb28e29d7b2f4e9add252e7bbb3
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Signed-off-by: Scott Duplichan <scott@notabs.org>
Reviewed-on: http://review.coreboot.org/8689
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
This was added to handle cases of Intel FSP platforms that had
EARLY_CBMEM_INIT but could not migrate CAR variables to CBMEM.
These boards were recently fixed.
To support combination of EARLY_CBMEM_INIT without CAR migration was
added maintenance effort with little benefits. You had no CBMEM
console for romstage and the few timestamps you could store were
circulated via PCI scratchpads or CMOS nvram.
Change-Id: I5cffb7f2b14c45b67ee70cf48be4d7a4c9e5f761
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/8636
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
1) Save the pointer to the FSP HOB list to low memory at address 0x614.
This is the same location as CBMEM_RESUME_BACKUP - the two aren't used
in the same platform, so overlapping should be OK. I didn't see any
documentation that actually said that this location was free to use, and
didn't need to be restored after use in S3 resume, but it looks like
the DOS boot vector gets loaded juat above this location, so it SHOULD
be ok. The alternative is to copy the memory out and store it in cbmem
until we're ready to restore it.
2) When a request for the pointer to a CAR variable comes in, pass back
the location inside the FSP hob structure.
3) Skip the memcopy of the CAR Data. The CAR variables do not
get transitioned back into cbmem, but used out of the HOB structure.
4) Remove the BROKEN_CAR_MIGRATE Kconfig option from the FSP platform.
Change-Id: Iaf566dce1b41a3bcb17e4134877f68262b5e113f
Signed-off-by: Martin Roth <gaumless@gmail.com>
Reviewed-on: http://review.coreboot.org/8196
Reviewed-by: Aaron Durbin <adurbin@google.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Move the CAR migration call to arch -specific part of CBMEM init,
it is truly a x86 specific thing.
Change-Id: I715417e54f197b8745e0670d6b900a5660178141
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/7860
Reviewed-by: Aaron Durbin <adurbin@google.com>
Tested-by: build bot (Jenkins)
With the change it becomes irrelevant if memcpy() car.global_data or
cbmemc_reinit() is done first.
Change-Id: Ie479eef346c959e97dcc55861ccb0db1321fb7b2
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/8032
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@google.com>
Tested-by: build bot (Jenkins)
This adds a block in the SMI handler to call init functions for
drivers which may be used in SMM. A static variable is used to
ensure the init functions are only called once.
BUG=chrome-os-partner:29580
BRANCH=mccloud
TEST=Built and booted on mccloud, system no longer hangs when
pressing power button at the dev mode screen. Also tested on parrot.
Original-Signed-off-by: David Hendricks <dhendrix@chromium.org>
Original-Change-Id: I225f572f7b3072bec2bc06aac3fb50d90a2e30ee
Original-Reviewed-on: https://chromium-review.googlesource.com/204764
Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org>
(cherry picked from commit 9315c485deb5f24df753e2d69f4819b2cb6accc2)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I8d2b21765c35c7ac7746986d5334dca17dcd6861
Reviewed-on: http://review.coreboot.org/8134
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
We relocate GDT to CBMEM, this can be done late in ramstage.
Note: We currently do this for BSP CPU only.
Change-Id: I626faaf22f846433f25ca2253d6a2a5230f50b6b
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/7858
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Always build CBMEM for romstage, even for boards that will not use it.
We further restrict car_migrate_variables() runs to non-ROMCC boards without
BROKEN_CAR_MIGRATE.
This fixes regression of commit 71b21455 that broke CBMEM console support
for boards with a combination of !EARLY_CBMEM_INIT && !HAVE_ACPI_RESUME.
Change-Id: Ife91d7baebdc9bd1e086896400059a165d3aa90f
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/7877
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
BOOT_MEDIA_SPI_BUS is a Kconfig variable used on some ARM-based
platforms to set up CBFS media. It turns out it can also be helpful
for setting up the eventlog which is intended to reside on the same
SPI flash as CBFS. Setting it for x86 will allow us to remove an
assumption about which SPI bus is used for this flash device.
Long term this can go away as we come up with a better abstraction
for the eventlog's backing store. This is only intended to help us
get from here to there.
BUG=none
BRANCH=none
TEST=built and booted on Link
Signed-off-by: David Hendricks <dhendrix@chromium.org>
Original-Change-Id: I1d84dc28592fbece33a70167be59e83bca9cd7bc
Original-Reviewed-on: https://chromium-review.googlesource.com/191202
Original-Tested-by: David Hendricks <dhendrix@chromium.org>
Original-Reviewed-by: Gabe Black <gabeblack@chromium.org>
Original-Commit-Queue: David Hendricks <dhendrix@chromium.org>
(cherry picked from commit 200aa7c5b1b1f4c74412893cf7231a12e2702463)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: If988bcff5244ec6a82580203471b25fac49c45ef
Reviewed-on: http://review.coreboot.org/7752
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Tested-by: build bot (Jenkins)
No need to pass calls through gcc in one case and
directly to binutils in another. Just always call
binutils.
Change-Id: Icf9660ce40d3c23f96dfab6a73c169ff07d3e42b
Signed-off-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-on: http://review.coreboot.org/7610
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Let's just call ld directly for gcc, too.
Change-Id: I305eb92ed0d21b098134a7eb5a9f9fe3b126aeea
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/7553
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
We build with either gcc or clang, no need to keep both around
Change-Id: I9af2cc7636bdc791a68ba8ed6e7c5a81973c5dfd
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/7552
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Broken linkage rule for Clang builds on one side of a
branch. Hence refactor out common rules from branch.
Change-Id: I00e5a2f5f9af1b7882a453caebb378ef74d2d51e
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: http://review.coreboot.org/7425
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins)
Ensure that the SMM build stage links properly using the compiler-rt
runtime under a Clang build.
Change-Id: Iead28c46d63f5bbb27757b7dc66fe06b4813d03c
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: http://review.coreboot.org/6462
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Fix up the following commit by enabling the MTRR's before enabling caching.
7756fe7 x86: Minimize work done with the caches disabled in mtrr functions.
Also fix two typos in comments.
Change-Id: If751b815f9dab781fc38c898cf692f0940c57695
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6969
Reviewed-by: Aaron Durbin <adurbin@google.com>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Tested-by: build bot (Jenkins)
The code in src/cpu/x86/mtrr/mtrr.c disables caching in a few places when
changing mtrr settings. While I can't find anything that says that's actually
required, I can believe it's necessary. With that said, other code around the
wrmsr instructions which actually modify the settings should be able to run
with caching enabled with no ill effects.
This is particularly true for two calls to printk, one in the fixed mtrr code
and one in the variable, which could result in an arbitrary amount of work
being done without caching. When changing the implementation of the cbmem
console, these two printks caused a significant regression in boot performance
on link of about 70ms which is about 10% of total firmware boot time. When the
window where the cache is disabled is minimized, both this and the new
implementation were about 30ms faster than the original boot time.
For the variable MTRRs, we now store what we want to set the MSRs to and then
write them all at once at the end of commit_var_mtrrs(). This way we don't
have some set and some not, but we still minimize the time we spend with the
caches disabled.
Change-Id: I5139b262bd2d13f79afd88e2e2c0f514fb3e27c9
Signed-off-by: Gabe Black <gabeblack@google.com>
Reviewed-on: https://chromium-review.googlesource.com/187811
Reviewed-by: Ronald Minnich <rminnich@chromium.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Commit-Queue: Gabe Black <gabeblack@chromium.org>
Tested-by: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit 31529d6d965676c6cedeb62137eabc26819956fc)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6952
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
This is essentially a revert of commit 10bd772d. The CAR_MIGRATE
mechanism is only useful to migrate variables from a special region
(e.g. cache as RAM) into DRAM-backed CBMEM between different parts of
the romstage (it does not persist into ramstage). Since ARM devices use
SRAM for which there is no reason to become inaccessible in later parts
of the romstage, this mechanism isn't useful for them. Removing it makes
the romstage.ld script much simpler, which has the nice side-effect of
putting the BSS at the end of the memory image (so that cbfstool can
actually figure out that it doesn't need to be part of the ROM image).
Old-Change-Id: I50e91d8bd51b5deb19446d9da48699edecbef6ea
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/176761
Reviewed-by: David Hendricks <dhendrix@chromium.org>
(cherry picked from commit ebfd698e57c902e2f39a0cfc1bc2b02665e47ec6)
console: Make cbmem depend on x86.
The cbmem implementation isn't supported on anything other than x86 right now
and actually causes memory corruption on ARM machines. Until that's fixed, this
will prevent people from turning it on and causing hard to track down errors.
Old-Change-Id: I00e8aacf008acfe2f76d4eab82570f7c1cc89cab
Signed-off-by: Gabe Black <gabeblack@google.com>
Reviewed-on: https://chromium-review.googlesource.com/191107
Reviewed-by: Julius Werner <jwerner@chromium.org>
Commit-Queue: Gabe Black <gabeblack@chromium.org>
Tested-by: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit e54f16e346a7f2c66d802fb78a6b24e53b732b83)
Squashed two related commits for cbmem support on arm.
Change-Id: I2be48cea348ee5dc8ca3632d743500aa111bab08
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6888
Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Tested-by: build bot (Jenkins)
The case doesn't look like a deliberate fall-through,
since the next case (SNB/IVB/HSW) is more specific
than the one before it, so break out.
Change-Id: I55497aefe9e835842a82121270f2b2a9952f560d
Found-by: Coverity Scan
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/6571
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@gmail.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Provide functionality to create dynamic classes based on program name and
architecture for which the program needs to be compiled/linked. define_class
takes program_name and arch as its arguments and adds the program_name to
classes-y to create dynamic class. Also, compiler toolset is created for the
specified arch. All the files for this program can then be added to
program_name-y += .. Ensure that define_class is called before any files are
added to the class. Check subdirs-y for order of directory inclusion.
One such example of dynamic class is rmodules. Multiple rmodules can be used
which need to be compiled for different architectures. With dynamic classes,
this is possible.
Change-Id: Ie143ed6f79ced5f58c200394cff89b006bc9b342
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: http://review.coreboot.org/6426
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
The preprocessor symbol has only one "L".
Change-Id: I3ec302f18d3bcc81bb45a9d53140f8aedd019317
Signed-off-by: Daniele Forsi <dforsi@gmail.com>
Reviewed-on: http://review.coreboot.org/6469
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Based on damo22's work and my X230 tracing.
Works for my X230 in a variety of RAM configs.
Also-By: Damien Zammit <damien@zamaudio.com>
Change-Id: I1aa024c55a8416fc53b25e7123037df0e55a2769
Signed-off-by: Vladimir Serbinenko <phcoder@gmail.com>
Signed-off-by: Damien Zammit <damien@zamaudio.com>
Reviewed-on: http://review.coreboot.org/5786
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Fixed spelling and added empty lines to separate the help
from the text automatically added during make menuconfig.
Change-Id: I6eee2c86e30573deb8cf0d42fda8b8329e1156c7
Signed-off-by: Daniele Forsi <dforsi@gmail.com>
Reviewed-on: http://review.coreboot.org/6313
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
The objective here is to tighten coreboot up a bit by not repeating
common helpers. This makes the code base more consistent and
unified/tight.
Change-Id: Ia163eae68b4a84a00ed118125e70308fab1cea0c
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: http://review.coreboot.org/6215
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
It was never well-defined what value this function should return.
Change-Id: If84aff86e0b556591d7ad557842910a2dfcd3b46
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/6166
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Comparison of unsigned expression < 0 is always false.
Change-Id: Idf4e7846b50f4376a5d33515681efbd773d1caca
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: http://review.coreboot.org/6146
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
We don't need exception handlers and they waste space.
Change-Id: I98a34d1c9638e8c4168edbfb4b1cddde8a64623f
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/6105
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
After moving out -m32 from CC_*, 64bit compilers need
CFLAGS_* in more places to handle everything in 32bit
as appropriate.
Change-Id: I692a46836fc0ba29a3a9eb47b123e3712691b45d
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/5789
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Lines with 'select SERIAL_CPU_INIT' where redundant with the
default being yes. Since there is no 'unselect SERIAL_CPU_INIT'
possibility, invert the default and rename option.
This squelches Kconfig warnings about unmet dependencies.
Change-Id: Iae546c56006278489ebae10f2daa627af48abe94
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/5700
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
- Move the non chipset-specific fsp pieces out of the chipset into a
shared area. This is used by northbridge / southbrige / SOC code. It
pulls in pieces from Kconfig, Makefile and FSP specific code.
- Enabled in the CPU code with a Kconfig "select PLATFORM_USES_FSP"
Change-Id: I7ffa934c1df09b71d48a876a56e3b888685870b8
Signed-off-by: Martin Roth <gaumless@gmail.com>
Signed-off-by: Martin Roth <martin.roth@se-eng.com>
Reviewed-on: http://review.coreboot.org/5635
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
Make all three coreboot stages (bootblock, romstage and ramstage) aware of the
architecture specific to that stage i.e. we will have CONFIG_ARCH variables for
each of the three stages. This allows us to have an SOC with any combination of
architectures and thus every stage can be made to run on a completely different
architecture independent of others. Thus, bootblock can have an x86 arch whereas
romstage and ramstage can have arm32 and arm64 arch respectively. These stage
specific CONFIG_ARCH_ variables enable us to select the proper set of toolchain
and compiler flags for every stage.
These options can be considered as either arch or modes eg: x86 running in
different modes or ARM having different arch types (v4, v7, v8). We have got rid
of the original CONFIG_ARCH option completely as every stage can have any
architecture of its own. Thus, almost all the components of coreboot are
identified as being part of one of the three stages (bootblock, romstage or
ramstage). The components which cannot be classified as such e.g. smm, rmodules
can have their own compiler toolset which is for now set to *_i386. Hence, all
special classes are treated in a similar way and the compiler toolset is defined
using create_class_compiler defined in Makefile.
In order to meet these requirements, changes have been made to CC, LD, OBJCOPY
and family to add CC_bootblock, CC_romstage, CC_ramstage and similarly others.
Additionally, CC_x86_32 and CC_armv7 handle all the special classes. All the
toolsets are defined using create_class_compiler.
Few additional macros have been introduced to identify the class to be used at
various points, e.g.: CC_$(class) derives the $(class) part from the name of
the stage being compiled.
We have also got rid of COREBOOT_COMPILER, COREBOOT_ASSEMBLER and COREBOOT_LINKER
as they do not make any sense for coreboot as a whole. All these attributes are
associated with each of the stages.
Change-Id: I923f3d4fb097d21071030b104c372cc138c68c7b
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: http://review.coreboot.org/5577
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@gmail.com>
Rename coreboot_ram stage to ramstage. This is done in order to provide
consistency with other stage names (bootblock, romstage) and to allow any
Makefile rule generalization, required for patches to be submitted later.
Change-Id: Ib66e43b7e17b9c48b2d099670ba7e7d857673386
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: http://review.coreboot.org/5567
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Console is arch-agnostic and there is no need for separate
implementations for romstage and ramstage.
For SMM there is console only if DEBUG_SMI is selected.
Change-Id: I7028eeeff8bfbb9c8552972436b29a7508834d87
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/5338
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
This is the minimal setup needed to be able to execute SMI handlers.
Only support for ASEG handlers is added, which should be sufficient
for Trinity (up to 4 cores).
There are a few hacks which need to be introduced in generic code in
order to make this work properly, but these hacks are self-contained.
They are a not a result of any special needs of this CPU, but rather
from a poorly designed infrastructure. Comments are added to explain
how such code could be refactored in the future.
Change-Id: Iefd4ae17cf0206cae8848cadba3a12cbe3b2f8b6
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/5493
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@gmail.com>
This gives us completely transparent low-level function to transmit
data.
Change-Id: I706791ff43d80a36a7252a4da0e6f3af92520db7
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/5336
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Option DRIVERS_UART builds with support for UART hardware.
Option CONSOLE_SERIAL enables the console output for UART.
Those x86 boards that do not have serial port on SuperIO should select
NO_UART_ON_SUPERIO to disable 8250 UART for the default configuration.
Removes:
CONSOLE_SERIAL_UART
HAVE_UART_IO_MAPPED
HAVE_UART_MEMORY_MAPPED
Renames:
CONSOLE_SERIAL8250 -> DRIVERS_UART_8250IO
CONSOLE_SERIAL8250MEM -> DRIVERS_UART_8250MEM
Change-Id: Id3afa05f85c0d6849746886db8b6c2ed6c846b61
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/5311
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
The timer_fsb variable was not correctly being accessed in the
presence of cache-as-ram. The cache-as-ram backing store could
be torn down but then udelay() could be called causing hangs from
accessing variables that have unknown values.
Instead change the timer_fsb variable to g_timer_fsb and obtain
the value through a local access method that does the correct things
to obtain the correct value.
Change-Id: Ia3e30808498cbe4a7f6f116c17a8cf1240a807a3
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5411
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Tested-by: build bot (Jenkins)
Use of CAR_GLOBAL is not safe after CAR is torn down, unless the
board properly implements EARLY_CBMEM_INIT.
Flag vulnerable boards that only do cbmem_recovery() in romstage on S3
resume and implementation with Intel FSP that invalidates cache before
we have a chance to copy the contents.
Change-Id: Iecd10dee9b73ab3f1f66826950fa0945675ff39f
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/5419
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
Start using the rmodtool for generating rmodules.
rmodule_link() has been changed to create 2 rules:
one for the passed in <name>, the other for creating
<name>.rmod which is an ELF file in the format of
an rmodule.
Since the header is not compiled and linked together
with an rmodule there needs to be a way of marking
which symbol is the entry point. __rmodule_entry is
the symbol used for knowing the entry point. There
was a little churn in SMM modules to ensure an
rmodule entry point symbol takes a single argument.
Change-Id: Ie452ed866f6596bf13f137f5b832faa39f48d26e
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5379
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Tested-by: build bot (Jenkins)
Boot speeds can be sped up by mirroring the payload into
main memory before doing the actual loading. Systems that
would benefit from this are typically Intel ones whose SPI
are memory mapped. Without the SPI being cached all accesses
to the payload in SPI while being loaded result in uncacheable
accesses. Instead take advantage of the on-board SPI controller
which has an internal cache and prefetcher by copying 64-byte
cachelines using 32-bit word copies.
Change-Id: I4aac856b1b5130fa2d68a6c45a96cfeead472a52
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5305
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Vladimir Serbinenko <phcoder@gmail.com>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Existing code compiled serial communication and printk() for SMM
even when DEBUG_SMI was not selected.
Change-Id: Ic5e25cd7453cb2243f7ac592b093fba752a299f7
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/5142
Tested-by: build bot (Jenkins)
Reviewed-by: Vladimir Serbinenko <phcoder@gmail.com>
NOTE: UART base for SMM continues to be broken, as it does not use
the address resource allocator has assigned.
Change-Id: I79f2ca8427a33a3c719adfe277c24dab79a33ef3
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/5235
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Do not pull in console hw-specific prototypes everywhere
with console.h as those are not needed for higher levels.
Move prototypes for UARTs next to other consoles.
Change-Id: Icbc9cd3e5bdfdab85d7dccd7c3827bba35248fb8
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/5232
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
With the recent improvement 3d6ffe76f8,
speedup by CACHE_ROM is reduced a lot.
On the other hand this makes coreboot run out of MTRRs depending on
system configuration, hence screwing up I/O access and cache
coherency in worst cases.
CACHE_ROM requires the user to sanity check their boot output because
the feature is brittle. The working configuration is dependent on I/O
hole size, ram size, and chipset. Because of this the current
implementation can leave a system configured in an inconsistent state
leading to unexpected results such as poor performance and/or
inconsistent cache-coherency
Remove this as a buggy feature until we figure out how to do it properly
if necessary.
Change-Id: I858d78a907bf042fcc21fdf7a2bf899e9f6b591d
Signed-off-by: Vladimir Serbinenko <phcoder@gmail.com>
Reviewed-on: http://review.coreboot.org/5146
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
There are some fun rules C compilers can use to optimize their code.
One of them is the assumption that two symbols point to two different
addresses.
In this case this wasn't true, resulting in unintended code execution
(and later, a crash) with a clang build.
Change-Id: I1496b22e1d1869ed0610e321b6ec6a83252e9d8b
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: http://review.coreboot.org/4719
Tested-by: build bot (Jenkins)
Get the required UART includes directly.
The ne2k part is old copy-paste leftover.
Change-Id: Ifd9253abb5a50b515887459faf06b63f907eeda9
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/5258
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Aaron Durbin <adurbin@google.com>
Its linker doesn't like "." arithmetics, so use .org,
while its assembler doesn't like data32 prefixes.
Change-Id: I3f5bbb350493d6510b8013df15d44c44c5db63c7
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/4714
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Certain CPUs require the default SMM region to be backed up
on resume after a suspend. The reason is that in order to
relocate the SMM region the default SMM region has to be used.
As coreboot is unaware of how that memory is used it needs to
be backed up. Therefore provide a common method for doing this.
Change-Id: I65fe1317dc0b2203cb29118564fdba995770ffea
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5216
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
We should not have x86 specific includes in lib/.
Change-Id: I18fa9c8017d65c166ffd465038d71f35b30d6f3d
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/5156
Tested-by: build bot (Jenkins)
Reviewed-by: David Hendricks <dhendrix@chromium.org>
Reviewed-by: Aaron Durbin <adurbin@google.com>
Be more conservative and only add VGA devices' prefetchable
resources as write-combining in the address space. Previously
all prefetchable memory was added as a write-combining memory
type. Some hardware incorrectly advertises its BAR as
prefetchable when it shouldn't be.
A new memranges_add_resources_filter() function is added
to provide additional filtering on device and resource.
Change-Id: I3fc55b90d8c5b694c5aa9e2f34db1b4ef845ce10
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5169
Reviewed-by: Vladimir Serbinenko <phcoder@gmail.com>
Tested-by: build bot (Jenkins)
If the MTRR usage exceeds the BIOS allocation for MTRR usage
re-try without the WRCOMB type.
Change-Id: Ie70ce84994428ff6700c36310264c3c44d9ed128
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5151
Reviewed-by: Vladimir Serbinenko <phcoder@gmail.com>
Tested-by: build bot (Jenkins)
In order for the cpu code to start SMM relocation 2 new
functions are added to be shared:
- void smm_initiate_relocation_parallel()
- void smm_initiate_relocation()
The both initiate an SMI on the currently running cpu.
The 2 variants allow for parallel relocation or serialized
relocation.
BUG=chrome-os-partner:22862
BRANCH=None
TEST=Built and booted rambi using these functions.
Change-Id: I325777bac27e9a0efc3f54f7223c38310604c5a2
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/173982
Reviewed-on: http://review.coreboot.org/4891
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Tested-by: build bot (Jenkins)
Provide a common entry point for bringing up the APs
in parallel. This work is based off of the Haswell one
which can be moved over to this in the future. The APs
are brought up and have the BSP's MTRRs duplicated in
their own MTRRs. Additionally, Microcode is loaded before
enabling caching. However, the current microcode loading
support assumes Intel's mechanism.
The infrastructure provides a notion of a flight plan
for the BSP and APs. This allows for flexibility in the
order of operations for a given architecture/chip without
providing any specific policy. Therefore, the chipset
caller can provide the order that is required.
BUG=chrome-os-partner:22862
BRANCH=None
TEST=Built and booted on rambi with baytrail specific patches.
Change-Id: I0539047a1b24c13ef278695737cdba3b9344c820
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/173703
Reviewed-on: http://review.coreboot.org/4888
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Tested-by: build bot (Jenkins)
The access to control registers were scattered about.
Provide a single header file to provide the correct
access function and definitions.
BUG=chrome-os-partner:22991
BRANCH=None
TEST=Built and booted using this infrastructure. Also objdump'd the
assembly to ensure consistency (objdump -d -r -S | grep xmm).
Change-Id: Iff7a043e4e5ba930a6a77f968f1fcc14784214e9
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/172641
Reviewed-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/4873
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
This change allows Kconfig options ROM_SIZE and CBFS_SIZE to be
set with values that are not power of 2. The region programmed
as WB cacheable will include all of ROM_SIZE.
Side-effects to consider:
Memory region below flash may be tagged WRPROT cacheable. As an
example, with ROM_SIZE of 12 MB, CACHE_ROM_SIZE would be 16 MB.
Since this can overlap CAR, we add an explicit test and fail
on compile should this happen. To work around this problem, one
needs to use CACHE_ROM_SIZE_OVERRIDE in the mainboard Kconfig and
define a smaller region for WB cache.
With this change flash regions outside CBFS are also tagged WRPROT
cacheable. This covers IFD and ME and sections ChromeOS may use.
Change-Id: I5e577900ff7e91606bef6d80033caaed721ce4bf
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/4625
Tested-by: build bot (Jenkins)
Reviewed-by: Vladimir Serbinenko <phcoder@gmail.com>
Add the FSP northbridge and southbridge includes.
Change-Id: I5c7f395dc033caa8d0bf0313382769595d77f2a5
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Reviewed-on: http://review.coreboot.org/4019
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Part of X201 port.
Change-Id: If17d707004aba9f08459dbd8f3a146fa3c076aa9
Signed-off-by: Vladimir Serbinenko <phcoder@gmail.com>
Reviewed-on: http://review.coreboot.org/4052
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Tested-by: build bot (Jenkins)
With the LynxPoint chipset there are more than 16
possible GPIOs that can trigger an SMI so we need
a mainboard handler that can support this.
There are only a handful of users of this function
so just change them all to use the new prototype.
Change-Id: I3d96da0397d6584f713fcf6003054b25c1c92939
Signed-off-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49530
Reviewed-by: Stefan Reinauer <reinauer@google.com>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/4145
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
The mdelay is not necessarry on 2065x.
Tested on X201 that it works without delay.
Change-Id: Ida9e85be7c214f3ba4c9476b5d8a0351e7980e5e
Signed-off-by: Vladimir Serbinenko <phcoder@gmail.com>
Reviewed-on: http://review.coreboot.org/4083
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Tested-by: build bot (Jenkins)
This reverts commit de1fe7f655.
While things appeared to work, there were actually invalid references
to CAR storage after CAR was torn down on boards without
EARLY_CBMEM_INIT. It was discussed use of CAR_GLOBAL should be
restricted to boards that handle CAR migration properly.
Change-Id: I9969d2ea79c334a7f95a0dbb7c78065720e6ccae
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3968
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
When building coreboot with the Clang static analyzer scan-build,
it reports »Value stored to 'type_index' is never read«. Indeed,
in `memranges_each_entry()` `type_index` is assigned a value
before being read. So remove that line.
Change-Id: I6da2fb8be7157bb98c57281babd4a08ca0d9f7a7
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-on: http://review.coreboot.org/3953
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
If romstage does not make cbmem_initialize() call, linker should
optimize the code for CAR migration away.
This simplifies design of CBMEM console by a considerable amount.
As console buffer is now migrated within cbmem_initialize() call there
is no longer need for cbmemc_reinit() call made at end of romstage.
Change-Id: I8675ecaafb641fa02675e9ba3f374caa8e240f1d
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3916
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
Letting SMI handler touch EHCI controller is an excellent source
of USB problems. Remove usbdebug entirely from SMM.
It may be possible to make usbdebug console work from SMM
after hard work and coordination with payloads and even
OS drivers. But we are not there.
Change-Id: Id50586758ee06e8d76e682dc6f64f756ab5b79f5
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3858
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Tested-by: build bot (Jenkins)
The code to figure out how to set num_starts was
starting to get kludgy. It's a constant for a given
CPU; constants should be constant; make it a config variable.
This change includes an example of how to override it.
Build but not boot tested; drivers welcome.
Change-Id: Iddd906a707bb16251615c7b42f2bfb5a044379b4
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Reviewed-on: http://review.coreboot.org/3796
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
Reviewed-by: Bruce Griffith <Bruce.Griffith@se-eng.com>
Non-S3 resume paths of sandy/ivybridge call cbmem_initialize()
more than once. Doing car_migrate_variables() more than twice caused
at least loss of some lines in CBMEM console.
Change-Id: Idd14aba9384984aa3a7d38937a4b3572aa5dc088
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3512
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
There are some boards that do a significant amount of
work after cache-as-ram is torn down but before ramstage
is loaded. For example, using vboot to verify the ramstage
is one such operation. However, there are pieces of code
that are executed that reference global variables that
are linked in the cache-as-ram region. If those variables
are referenced after cache-as-ram is torn down then the
values observed will most likely be incorrect.
Therefore provide a Kconfig option to select cache-as-ram
migration to memory using cbmem. This option is named
CAR_MIGRATION. When enabled, the address of cache-as-ram
variables may be obtained dynamically. Additionally,
when cache-as-ram migration occurs the cache-as-ram
data region for global variables is copied into cbmem.
There are also automatic callbacks for other modules
to perform their own migration, if necessary.
Change-Id: I2e77219647c2bd2b1aa845b262be3b2543f1fcb7
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/3232
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Thread support is added for the x86 architecture. Both
the local apic and the tsc udelay() functions have a
call to thread_yield_microseconds() so as to provide an
opportunity to run pending threads.
Change-Id: Ie39b9eb565eb189676c06645bdf2a8720fe0636a
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/3207
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Since the TSC udelay() function can be used in SMM that means the
TSC can count up to whatever value. The current loop was not handling
TSC rollover properly. In most cases this should not matter as the TSC
typically starts ticking at value 0, and it would take a very long time
to roll it over. However, it is my understanding that this behavior is
not guaranteed. Theoretically the TSC could start or be be written to
with a large value that would cause the rollover.
Change-Id: I2f11a5bc4f27d5543e74f8224811fa91e4a55484
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/3171
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Some boards use the local apic for udelay(), but they also provide
their own implementation of udelay() for SMM. The reason for using
the local apic for udelay() in ramstage is to not have to pay the
penalty of calibrating the TSC frequency. Therefore provide a
TSC_CONSTANT_RATE option to indicate that TSC calibration is not
needed. Instead rely on the presence of a tsc_freq_mhz() function
provided by the cpu/board. Additionally, assume that if
TSC_CONSTANT_RATE is selected the udelay() function in SMM will
be the tsc.
Change-Id: I1629c2fbe3431772b4e80495160584fb6f599e9e
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/3168
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Implement the timer_monotonic_get() using the TSC.
Change-Id: I5118da6fb9bccc75d2ce012317612e0ab20a2cac
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/3155
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Implement the timer_monotonic_get() functionality based off of
the local apic timer.
Change-Id: I1aa1ff64d15a3056d6abd1372be13da682c5ee2e
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/3154
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
On x86 systems there is a concept of cachings the ROM. However,
the typical policy is that the boot cpu is the only one with
it enabled. In order to ensure the MTRRs are the same across cores
the rom cache needs to be disabled prior to OS resume or boot handoff.
Therefore, utilize the boot state callbacks to schedule the disabling
of the ROM cache at the ramstage exit points.
Change-Id: I4da5886d9f1cf4c6af2f09bb909f0d0f0faa4e62
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/3138
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Based on comments in cpu/x86/msr.h for wrmsr/rdmsr, and for symmetry,
I have added __attribute__((always_inline)) for these.
Change-Id: Ia0a34c15241f9fbc8c78763386028ddcbe6690b1
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/2898
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
On certain architectures such as x86 the bootstrap processor
does most of the work. When CACHE_ROM is employed it's appropriate
to ensure that the caching enablement of the ROM is disabled so that
the caching settings are symmetric before booting the payload or OS.
Tested this on an x86 machine that turned on ROM caching. Linux did not
complain about asymmetric MTRR settings nor did the ROM show up as
cached in the MTRR settings.
Change-Id: Ia32ff9fdb1608667a0e9a5f23b9c8af27d589047
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/2980
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
There is an optimization that can take place when hole
carving in ranges above 4GiB. If the range is the last
range then there is no need to carve UC holes out from
the larger WB range.
This optimization also has the same assumption of choosing
WB as the default MTRR type: the OS needs to properly
handle accessing realloacted MMIO resources with PAT so
that the MTRR type can be overidden.
Below are results using a combination of options. The
board this was tested on has 10 variable MTRRs at its
disposal. It has 4GiB of RAM.
IO hole config #1: hole starts at 0xad800000
No CACHE_ROM and no WRCOMB resources (takes 4 MTRRs):
MTRR: Physical address space:
0x0000000000000000 - 0x00000000000a0000 size 0x000a0000 type 6
0x00000000000a0000 - 0x00000000000c0000 size 0x00020000 type 0
0x00000000000c0000 - 0x00000000ad800000 size 0xad740000 type 6
0x00000000ad800000 - 0x0000000100000000 size 0x52800000 type 0
0x0000000100000000 - 0x000000014f600000 size 0x4f600000 type 6
MTRR: default type WB/UC MTRR counts: 4/6.
MTRR: WB selected as default type.
MTRR: 0 base 0x00000000ad800000 mask 0x0000007fff800000 type 0
MTRR: 1 base 0x00000000ae000000 mask 0x0000007ffe000000 type 0
MTRR: 2 base 0x00000000b0000000 mask 0x0000007ff0000000 type 0
MTRR: 3 base 0x00000000c0000000 mask 0x0000007fc0000000 type 0
No CACHE_ROM and 1 WRCOMB resource (takes 6 MTRRs):
MTRR: Physical address space:
0x0000000000000000 - 0x00000000000a0000 size 0x000a0000 type 6
0x00000000000a0000 - 0x00000000000c0000 size 0x00020000 type 0
0x00000000000c0000 - 0x00000000ad800000 size 0xad740000 type 6
0x00000000ad800000 - 0x00000000d0000000 size 0x22800000 type 0
0x00000000d0000000 - 0x00000000e0000000 size 0x10000000 type 1
0x00000000e0000000 - 0x0000000100000000 size 0x20000000 type 0
0x0000000100000000 - 0x000000014f600000 size 0x4f600000 type 6
MTRR: default type WB/UC MTRR counts: 6/7.
MTRR: WB selected as default type.
MTRR: 0 base 0x00000000ad800000 mask 0x0000007fff800000 type 0
MTRR: 1 base 0x00000000ae000000 mask 0x0000007ffe000000 type 0
MTRR: 2 base 0x00000000b0000000 mask 0x0000007ff0000000 type 0
MTRR: 3 base 0x00000000c0000000 mask 0x0000007ff0000000 type 0
MTRR: 4 base 0x00000000d0000000 mask 0x0000007ff0000000 type 1
MTRR: 5 base 0x00000000e0000000 mask 0x0000007fe0000000 type 0
CACHE_ROM and no WRCOMB resources (takes 7 MTRRs):
MTRR: Physical address space:
0x0000000000000000 - 0x00000000000a0000 size 0x000a0000 type 6
0x00000000000a0000 - 0x00000000000c0000 size 0x00020000 type 0
0x00000000000c0000 - 0x00000000ad800000 size 0xad740000 type 6
0x00000000ad800000 - 0x00000000ff800000 size 0x52000000 type 0
0x00000000ff800000 - 0x0000000100000000 size 0x00800000 type 5
0x0000000100000000 - 0x000000014f600000 size 0x4f600000 type 6
MTRR: default type WB/UC MTRR counts: 11/7.
MTRR: UC selected as default type.
MTRR: 0 base 0x0000000000000000 mask 0x0000007f80000000 type 6
MTRR: 1 base 0x0000000080000000 mask 0x0000007fe0000000 type 6
MTRR: 2 base 0x00000000a0000000 mask 0x0000007ff0000000 type 6
MTRR: 3 base 0x00000000ad800000 mask 0x0000007fff800000 type 0
MTRR: 4 base 0x00000000ae000000 mask 0x0000007ffe000000 type 0
MTRR: 5 base 0x00000000ff800000 mask 0x0000007fff800000 type 0
MTRR: 6 base 0x0000000100000000 mask 0x0000007f00000000 type 6
CACHE_ROM and 1 WRCOMB resource (takes 8 MTRRs):
Previously this combination was impossible without the optimization.
MTRR: Physical address space:
0x0000000000000000 - 0x00000000000a0000 size 0x000a0000 type 6
0x00000000000a0000 - 0x00000000000c0000 size 0x00020000 type 0
0x00000000000c0000 - 0x00000000ad800000 size 0xad740000 type 6
0x00000000ad800000 - 0x00000000d0000000 size 0x22800000 type 0
0x00000000d0000000 - 0x00000000e0000000 size 0x10000000 type 1
0x00000000e0000000 - 0x00000000ff800000 size 0x1f800000 type 0
0x00000000ff800000 - 0x0000000100000000 size 0x00800000 type 5
0x0000000100000000 - 0x000000014f600000 size 0x4f600000 type 6
MTRR: default type WB/UC MTRR counts: 12/8.
MTRR: UC selected as default type.
MTRR: 0 base 0x0000000000000000 mask 0x0000007f80000000 type 6
MTRR: 1 base 0x0000000080000000 mask 0x0000007fe0000000 type 6
MTRR: 2 base 0x00000000a0000000 mask 0x0000007ff0000000 type 6
MTRR: 3 base 0x00000000ad800000 mask 0x0000007fff800000 type 0
MTRR: 4 base 0x00000000ae000000 mask 0x0000007ffe000000 type 0
MTRR: 5 base 0x00000000d0000000 mask 0x0000007ff0000000 type 1
MTRR: 6 base 0x00000000ff800000 mask 0x0000007fff800000 type 0
MTRR: 7 base 0x0000000100000000 mask 0x0000007f00000000 type 6
IO hole config #1: hole starts at 0x80000000
No CACHE_ROM and no WRCOMB resources (takes 1 MTRRs):
MTRR: Physical address space:
0x0000000000000000 - 0x00000000000a0000 size 0x000a0000 type 6
0x00000000000a0000 - 0x00000000000c0000 size 0x00020000 type 0
0x00000000000c0000 - 0x0000000080000000 size 0x7ff40000 type 6
0x0000000080000000 - 0x0000000100000000 size 0x80000000 type 0
0x0000000100000000 - 0x000000017ce00000 size 0x7ce00000 type 6
MTRR: default type WB/UC MTRR counts: 1/2.
MTRR: WB selected as default type.
MTRR: 0 base 0x0000000080000000 mask 0x0000007f80000000 type 0
No CACHE_ROM and 1 WRCOMB resource (takes 3 MTRRs):
MTRR: Physical address space:
0x0000000000000000 - 0x00000000000a0000 size 0x000a0000 type 6
0x00000000000a0000 - 0x00000000000c0000 size 0x00020000 type 0
0x00000000000c0000 - 0x0000000080000000 size 0x7ff40000 type 6
0x0000000080000000 - 0x00000000d0000000 size 0x50000000 type 0
0x00000000d0000000 - 0x00000000e0000000 size 0x10000000 type 1
0x00000000e0000000 - 0x0000000100000000 size 0x20000000 type 0
0x0000000100000000 - 0x000000017ce00000 size 0x7ce00000 type 6
MTRR: default type WB/UC MTRR counts: 4/3.
MTRR: UC selected as default type.
MTRR: 0 base 0x0000000000000000 mask 0x0000007f80000000 type 6
MTRR: 1 base 0x00000000d0000000 mask 0x0000007ff0000000 type 1
MTRR: 2 base 0x0000000100000000 mask 0x0000007f00000000 type 6
CACHE_ROM and no WRCOMB resources (takes 3 MTRRs):
MTRR: Physical address space:
0x0000000000000000 - 0x00000000000a0000 size 0x000a0000 type 6
0x00000000000a0000 - 0x00000000000c0000 size 0x00020000 type 0
0x00000000000c0000 - 0x0000000080000000 size 0x7ff40000 type 6
0x0000000080000000 - 0x00000000ff800000 size 0x7f800000 type 0
0x00000000ff800000 - 0x0000000100000000 size 0x00800000 type 5
0x0000000100000000 - 0x000000017ce00000 size 0x7ce00000 type 6
MTRR: default type WB/UC MTRR counts: 9/3.
MTRR: UC selected as default type.
MTRR: 0 base 0x0000000000000000 mask 0x0000007f80000000 type 6
MTRR: 1 base 0x00000000ff800000 mask 0x0000007fff800000 type 0
MTRR: 2 base 0x0000000100000000 mask 0x0000007f00000000 type 6
CACHE_ROM and 1 WRCOMB resource (takes 4 MTRRs):
MTRR: Physical address space:
0x0000000000000000 - 0x00000000000a0000 size 0x000a0000 type 6
0x00000000000a0000 - 0x00000000000c0000 size 0x00020000 type 0
0x00000000000c0000 - 0x0000000080000000 size 0x7ff40000 type 6
0x0000000080000000 - 0x00000000d0000000 size 0x50000000 type 0
0x00000000d0000000 - 0x00000000e0000000 size 0x10000000 type 1
0x00000000e0000000 - 0x00000000ff800000 size 0x1f800000 type 0
0x00000000ff800000 - 0x0000000100000000 size 0x00800000 type 5
0x0000000100000000 - 0x000000017ce00000 size 0x7ce00000 type 6
MTRR: default type WB/UC MTRR counts: 10/4.
MTRR: UC selected as default type.
MTRR: 0 base 0x0000000000000000 mask 0x0000007f80000000 type 6
MTRR: 1 base 0x00000000d0000000 mask 0x0000007ff0000000 type 1
MTRR: 2 base 0x00000000ff800000 mask 0x0000007fff800000 type 0
MTRR: 3 base 0x0000000100000000 mask 0x0000007f00000000 type 6
Change-Id: Ia3195af686c3f0603b21f713cfb2d9075eb02806
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/2959
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Some ranges would use less variable MTRRs if an UC area
can be carved off the top of larger WB range. Implement this
approach by doing 3 passes over each region in the addres space:
1. UC default type. Cover non-UC and non-WB regions with respectie type.
Punch UC hole at upper end of larger WB regions with WB type.
2. UC default type. Cover non-UC regions with respective type.
3. WB default type. Cover non-WB regions with respective type.
The hole at upper end of a region uses the same min alignment of 64MiB.
Below are results using a combination of options. The board this was
tested on has 10 variable MTRRs at its disposal. It has 4GiB of RAM.
IO hole config #1: hole starts at 0xad800000
No CACHE_ROM or WRCOMB resources (takes 4 MTRRs):
MTRR: Physical address space:
0x0000000000000000 - 0x00000000000a0000 size 0x000a0000 type 6
0x00000000000a0000 - 0x00000000000c0000 size 0x00020000 type 0
0x00000000000c0000 - 0x00000000ad800000 size 0xad740000 type 6
0x00000000ad800000 - 0x0000000100000000 size 0x52800000 type 0
0x0000000100000000 - 0x000000014f600000 size 0x4f600000 type 6
MTRR: default type WB/UC MTRR counts: 4/9.
MTRR: WB selected as default type.
MTRR: 0 base 0x00000000ad800000 mask 0x0000007fff800000 type 0
MTRR: 1 base 0x00000000ae000000 mask 0x0000007ffe000000 type 0
MTRR: 2 base 0x00000000b0000000 mask 0x0000007ff0000000 type 0
MTRR: 3 base 0x00000000c0000000 mask 0x0000007fc0000000 type 0
No CACHE_ROM. 1 WRCOMB resource (takes 6 MTRRs):
MTRR: Physical address space:
0x0000000000000000 - 0x00000000000a0000 size 0x000a0000 type 6
0x00000000000a0000 - 0x00000000000c0000 size 0x00020000 type 0
0x00000000000c0000 - 0x00000000ad800000 size 0xad740000 type 6
0x00000000ad800000 - 0x00000000d0000000 size 0x22800000 type 0
0x00000000d0000000 - 0x00000000e0000000 size 0x10000000 type 1
0x00000000e0000000 - 0x0000000100000000 size 0x20000000 type 0
0x0000000100000000 - 0x000000014f600000 size 0x4f600000 type 6
MTRR: default type WB/UC MTRR counts: 6/10.
MTRR: WB selected as default type.
MTRR: 0 base 0x00000000ad800000 mask 0x0000007fff800000 type 0
MTRR: 1 base 0x00000000ae000000 mask 0x0000007ffe000000 type 0
MTRR: 2 base 0x00000000b0000000 mask 0x0000007ff0000000 type 0
MTRR: 3 base 0x00000000c0000000 mask 0x0000007ff0000000 type 0
MTRR: 4 base 0x00000000d0000000 mask 0x0000007ff0000000 type 1
MTRR: 5 base 0x00000000e0000000 mask 0x0000007fe0000000 type 0
CACHE_ROM and no WRCOMB resources (taks 10 MTRRs):
MTRR: Physical address space:
0x0000000000000000 - 0x00000000000a0000 size 0x000a0000 type 6
0x00000000000a0000 - 0x00000000000c0000 size 0x00020000 type 0
0x00000000000c0000 - 0x00000000ad800000 size 0xad740000 type 6
0x00000000ad800000 - 0x00000000ff800000 size 0x52000000 type 0
0x00000000ff800000 - 0x0000000100000000 size 0x00800000 type 5
0x0000000100000000 - 0x000000014f600000 size 0x4f600000 type 6
MTRR: default type WB/UC MTRR counts: 11/10.
MTRR: UC selected as default type.
MTRR: 0 base 0x0000000000000000 mask 0x0000007f80000000 type 6
MTRR: 1 base 0x0000000080000000 mask 0x0000007fe0000000 type 6
MTRR: 2 base 0x00000000a0000000 mask 0x0000007ff0000000 type 6
MTRR: 3 base 0x00000000ad800000 mask 0x0000007fff800000 type 0
MTRR: 4 base 0x00000000ae000000 mask 0x0000007ffe000000 type 0
MTRR: 5 base 0x00000000ff800000 mask 0x0000007fff800000 type 0
MTRR: 6 base 0x0000000100000000 mask 0x0000007fc0000000 type 6
MTRR: 7 base 0x0000000140000000 mask 0x0000007ff0000000 type 6
Taking a reserved OS MTRR.
MTRR: 8 base 0x000000014f600000 mask 0x0000007fffe00000 type 0
Taking a reserved OS MTRR.
MTRR: 9 base 0x000000014f800000 mask 0x0000007fff800000 type 0
A combination of CACHE_ROM and WRCOMB just won't work.
IO hole config #2: hole starts at 0x80000000:
No CACHE_ROM or WRCOMB resources (takes 1 MTRR):
MTRR: Physical address space:
0x0000000000000000 - 0x00000000000a0000 size 0x000a0000 type 6
0x00000000000a0000 - 0x00000000000c0000 size 0x00020000 type 0
0x00000000000c0000 - 0x0000000080000000 size 0x7ff40000 type 6
0x0000000080000000 - 0x0000000100000000 size 0x80000000 type 0
0x0000000100000000 - 0x000000017ce00000 size 0x7ce00000 type 6
MTRR: default type WB/UC MTRR counts: 1/5.
MTRR: WB selected as default type.
MTRR: 0 base 0x0000000080000000 mask 0x0000007f80000000 type 0
No CACHE_ROM. 1 WRCOMB resource (takes 4 MTRRs):
MTRR: Physical address space:
0x0000000000000000 - 0x00000000000a0000 size 0x000a0000 type 6
0x00000000000a0000 - 0x00000000000c0000 size 0x00020000 type 0
0x00000000000c0000 - 0x0000000080000000 size 0x7ff40000 type 6
0x0000000080000000 - 0x00000000d0000000 size 0x50000000 type 0
0x00000000d0000000 - 0x00000000e0000000 size 0x10000000 type 1
0x00000000e0000000 - 0x0000000100000000 size 0x20000000 type 0
0x0000000100000000 - 0x000000017ce00000 size 0x7ce00000 type 6
MTRR: default type WB/UC MTRR counts: 4/6.
MTRR: WB selected as default type.
MTRR: 0 base 0x0000000080000000 mask 0x0000007fc0000000 type 0
MTRR: 1 base 0x00000000c0000000 mask 0x0000007ff0000000 type 0
MTRR: 2 base 0x00000000d0000000 mask 0x0000007ff0000000 type 1
MTRR: 3 base 0x00000000e0000000 mask 0x0000007fe0000000 type 0
CACHE_ROM and no WRCOMB resources (takes 6 MTRRs):
MTRR: Physical address space:
0x0000000000000000 - 0x00000000000a0000 size 0x000a0000 type 6
0x00000000000a0000 - 0x00000000000c0000 size 0x00020000 type 0
0x00000000000c0000 - 0x0000000080000000 size 0x7ff40000 type 6
0x0000000080000000 - 0x00000000ff800000 size 0x7f800000 type 0
0x00000000ff800000 - 0x0000000100000000 size 0x00800000 type 5
0x0000000100000000 - 0x000000017ce00000 size 0x7ce00000 type 6
MTRR: default type WB/UC MTRR counts: 9/6.
MTRR: UC selected as default type.
MTRR: 0 base 0x0000000000000000 mask 0x0000007f80000000 type 6
MTRR: 1 base 0x00000000ff800000 mask 0x0000007fff800000 type 0
MTRR: 2 base 0x0000000100000000 mask 0x0000007f80000000 type 6
MTRR: 3 base 0x000000017ce00000 mask 0x0000007fffe00000 type 0
MTRR: 4 base 0x000000017d000000 mask 0x0000007fff000000 type 0
MTRR: 5 base 0x000000017e000000 mask 0x0000007ffe000000 type 0
CACHE_ROM and 1 WRCOMB resource (takes 7 MTRRs):
MTRR: Physical address space:
0x0000000000000000 - 0x00000000000a0000 size 0x000a0000 type 6
0x00000000000a0000 - 0x00000000000c0000 size 0x00020000 type 0
0x00000000000c0000 - 0x0000000080000000 size 0x7ff40000 type 6
0x0000000080000000 - 0x00000000d0000000 size 0x50000000 type 0
0x00000000d0000000 - 0x00000000e0000000 size 0x10000000 type 1
0x00000000e0000000 - 0x00000000ff800000 size 0x1f800000 type 0
0x00000000ff800000 - 0x0000000100000000 size 0x00800000 type 5
0x0000000100000000 - 0x000000017ce00000 size 0x7ce00000 type 6
MTRR: default type WB/UC MTRR counts: 10/7.
MTRR: UC selected as default type.
MTRR: 0 base 0x0000000000000000 mask 0x0000007f80000000 type 6
MTRR: 1 base 0x00000000d0000000 mask 0x0000007ff0000000 type 1
MTRR: 2 base 0x00000000ff800000 mask 0x0000007fff800000 type 0
MTRR: 3 base 0x0000000100000000 mask 0x0000007f80000000 type 6
MTRR: 4 base 0x000000017ce00000 mask 0x0000007fffe00000 type 0
MTRR: 5 base 0x000000017d000000 mask 0x0000007fff000000 type 0
MTRR: 6 base 0x000000017e000000 mask 0x0000007ffe000000 type 0
Change-Id: Iceb9b64991accf558caae2e7b0205951e9bcde44
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/2925
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Downstream payloads may need to take advantage of caching the
ROM for performance reasons. Add the ability to communicate the
variable range MTRR index to use to perform the caching enablement.
An example usage implementation would be to obtain the variable MTRR
index that covers the ROM from the coreboot tables. Then one would
disable caching and change the MTRR type from uncacheable to
write-protect and enable caching. The opposite sequence is required
to tearn down the caching.
Change-Id: I4d486cfb986629247ab2da7818486973c6720ef5
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/2919
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
The CONFIG_CACHE_ROM support in the MTRR code allocates an MTRR
specifically for setting up write-protect cachine of the ROM. It is
assumed that CONFIG_ROM_SIZE is the size of the ROM and the whole
area should be cached just under 4GiB. If enabled, the MTRR code
will allocate but not enable rom caching. It is up to the callers
of the MTRR code to explicitly enable (and disable afterwards) through
the use of 2 new functions:
- x86_mtrr_enable_rom_caching()
- x86_mtrr_disable_rom_caching()
Additionally, the CACHE_ROM option is exposed to the config menu so
that it is not just selected by the chipset or board. The reasoning
is that through a multitude of options CACHE_ROM may not be appropriate
for enabling.
Change-Id: I4483df850f442bdcef969ffeaf7608ed70b88085
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/2918
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
All resources that set the IORESOURCE_WRCOMB attribute which are
also marked as IORESOURCE_PREFETCH will have a MTRR set up that
is of the write-combining cacheable type. The only resources on
x86 that can be set to write-combining are prefetchable ones.
Change-Id: Iba7452cff3677e07d7e263b79982a49c93be9c54
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/2892
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Patrick Georgi