The write_tables function was void. It is a bit more
useful for loading payloads from the romstage
if it returns a pointer to the table it creates.
Change-Id: I6eeaf3e16bcbaf1e7ec3eada8026c466d2fb6f5a
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Reviewed-on: https://review.coreboot.org/27537
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Fix a linking problem with VBOOT and USE_OPTION_TABLE enabled.
Make use of cbfs_locate_file_in_region() and always search the
cmos_layout.bin in the 'COREBOOT' region.
With this change applied there's no need to include the vboot_locator
in SMM any more, we can't break NVRAM with different CMOS layouts,
and we keep VBOOT and non VBOOT behaviour the same.
Only include cmos_layout.bin and cmos.default in RO region.
Add notes explaining the decisions.
Tested on Intel Sandybridge, builds and boots with vboot enabled.
Change-Id: I10ae94d7936581bbb5ea49384122062bd4934ea5
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/26863
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
* Add support for parsing and booting FIT payloads.
* Build fit loader code from depthcharge.
* Fix coding style.
* Add Kconfig option to add compiletime support for FIT.
* Add support for initrd.
* Add default compat strings
* Apply optional devicetree fixups using dt_apply_fixups
Starting at this point the CBFS payload/ can be either SELF or FIT.
Tested on Cavium SoC: Parses and loads a Linux kernel 4.16.3.
Tested on Cavium SoC: Parses and loads a Linux kernel 4.15.0.
Tested on Cavium SoC: Parses and loads a Linux kernel 4.1.52.
Change-Id: I0f27b92a5e074966f893399eb401eb97d784850d
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/25019
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
Fix regression (supposedly) after commit:
23d62dd lib/bootmem: Add more bootmem tags
Without RELOCATABLE_RAMSTAGE, payload is allowed to overwrite
memory regions of the running ramstage. This case is handled
gracefully via a bounce-buffer implementation in arch/x86/boot.c.
Change-Id: I1c9bbdb963a7210d0817a7a990a70a1e4fc03624
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/26935
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Making exceptions for some payload to be loaded near
and under 1 MiB boundary sounds like a legacy 16-bit
x86 BIOS thing we generally do not want under lib/.
Change-Id: I8e8336a03d6f06d8f022c880a8334fe19a777f0a
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/26934
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
If we have POSTCAR_STAGE there is no need for
romstage to include LZMA decompression code.
Reduces romstage by about 4 kiB on x86.
Change-Id: I4c475986b2a94e5cd540c3eead433ed6c0a815ed
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/26759
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Masked ROMs are the silent killers of boot speed on devices without
memory-mapped SPI flash. They often contain awfully slow SPI drivers
(presumably bit-banged) that take hundreds of milliseconds to load our
bootblock, and every extra kilobyte of bootblock size has a hugely
disproportionate impact on boot speed. The coreboot timestamps can never
show that component, but it impacts our users all the same.
This patch tries to alleviate that issue a bit by allowing us to
compress the bootblock with LZ4, which can cut its size down to nearly
half. Of course, masked ROMs usually don't come with decompression
algorithms built in, so we need to introduce a little decompression stub
that can decompress the rest of the bootblock. This is done by creating
a new "decompressor" stage which runs before the bootblock, but includes
the compressed bootblock code in its data section. It needs to be as
small as possible to get a real benefit from this approach, which means
no device drivers, no console output, no exception handling, etc.
Besides the decompression algorithm itself we only include the timer
driver so that we can measure the boot speed impact of decompression. On
ARM and ARM64 systems, we also need to give SoC code a chance to
initialize the MMU, since running decompression without MMU is
prohibitively slow on these architectures.
This feature is implemented for ARM and ARM64 architectures for now,
although most of it is architecture-independent and it should be
relatively simple to port to other platforms where a masked ROM loads
the bootblock into SRAM. It is also supposed to be a clean starting
point from which later optimizations can hopefully cut down the
decompression stub size (currently ~4K on RK3399) a bit more.
NOTE: Bootblock compression is not for everyone. Possible side effects
include trying to run LZ4 on CPUs that come out of reset extremely
underclocked or enabling this too early in SoC bring-up and getting
frustrated trying to find issues in an undebuggable environment. Ask
your SoC vendor if bootblock compression is right for you.
Change-Id: I0dc1cad9ae7508892e477739e743cd1afb5945e8
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/26340
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
This patch adds more parameters to bootblock_main_with_timestamp() to
give callers the opportunity to add additional timestamps that were
recorded in the platform-specific initialization phase.
Change-Id: Idf3a0fcf5aee88a33747afc69e055b95bd38750c
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/26339
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
It turns out that even with the `-gnatp` switch to suppress runtime
checks, the compiler is still allowed to generate them (it only doesn't
have to). If we can't control generation of checks, we also can't
make assumptions about propagation of their exceptions.
The compiler warning that led to this change seems spurious, though
(the check might be generated, but is dropped later). So we might
revert this decision if the compiler can be fixed.
Change-Id: I7470d74b1f96f90d0d15b24dfd636d5f1c778d46
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/26350
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
I'm increasing the max because when AGESA tracing is enabled it will use
over 120 entries. I added some padding to the number incase more probes
are added. This only affects ramstage so the extra ram shouldn't matter.
BUG=b:64549506
TEST=boot on grunt and ran cbmem -t
Change-Id: I7a3d2d09c91c9e302d139e7f65fa9c85c4594de4
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Reviewed-on: https://review.coreboot.org/26234
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
The patch series ending in 64049be (lib/bootmem: Add method to walk OS
POV memory tables) expanded the bootmem framework to also keep track of
memory regions that are only relevant while coreboot is still executing,
such as the ramstage code and data. Mixing this into the exsting bootmem
ranges has already caused an issue on CONFIG_RELOCATEABLE_RAMSTAGE
boards, because the ramstage code in CBMEM is marked as BM_RAMSTAGE
which ends up getting translated back to LB_RAM in the OS tables. This
was fixed in 1ecec5f (lib/bootmem: ensure ramstage memory isn't given to
OS) for this specific case, but unfortunately Arm boards can have a
similar problem where their stack space is sometimes located in an SRAM
region that should not be made available as RAM to the OS.
Since both the resources made available to the OS and the regions
reserved for coreboot can be different for each platform, we should find
a generic solution to this rather than trying to deal with each issue
individually. This patch solves the problem by keeping the OS point of
view and the coreboot-specific ranges separate from the start, rather
than cloning it out later. Ranges only relevant to the coreboot view
will never touch the OS-specific layout, to avoid the problem of losing
information about the original memory type of the underlying region that
needs to be restored for the OS view. This both supersedes the
RELOCATABLE_RAMSTAGE fix and resolves the problems on Arm boards.
Change-Id: I7bb018456b58ad9b0cfb0b8da8c26b791b487fbb
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/26182
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Add a method to walk bootmem memory tables and call a function
for each memory range. The tables might not match with OS sight
of view.
Return true if the callback function returned false.
Required for FIT support in coreboot to find a usable RAM region.
Tested on Cavium SoC.
Change-Id: I0004e5ad5fe2289827f370f0d0f9979d3cbd3926
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/25583
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Introduce new bootmem tags to allow more fine grained control over buffer
allocation on various platforms. The new tags are:
BM_MEM_RAMSTAGE : Memory where any kind of boot firmware resides and that
should not be touched by bootmem (by example: stack,
TTB, program, ...).
BM_MEM_PAYLOAD : Memory where any kind of payload resides and that should
not be touched by bootmem.
Starting with this commit all bootmem methods will no longer see memory
that is used by coreboot as usable RAM.
Bootmem changes:
* Introduce a weak function to add platform specific memranges.
* Mark memory allocated by bootmem as BM_TAG_PAYLOAD.
* Assert on failures.
* Add _stack and _program as BM_MEM_RAMSTAGE.
ARMv7 and ARMv8 specific changes:
* Add _ttb and _postram_cbfs_cache as BM_MEM_RAMSTAGE.
ARMv7 specific changes:
* Add _ttb_subtables as BM_MEM_RAMSTAGE.
Change-Id: I0c983ce43616147c519a43edee3b61d54eadbb9a
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/25383
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Introduce bootmem custom memory tags and use them instead of reusing
LB_MEM tags.
Use asserts in bootmem_add_range to verify parameters.
Tested with uImage payload on Cavium SoC.
Change-Id: I7be8fa792fc7933ca218ecd43d250d3a9c55caa6
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/25633
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
* Adapt to coreboot coding style.
* Use coreboot's endian conversion functions.
* Fix header and header guards.
* Get rid of unused functions.
* Add Kconfig to build it on ramstage.
* Replace size32 with ALIGN_UP and DIV_ROUND_UP.
* Add NULL pointer checks
* Convert constants to defines
Required for Cavium's BDK and uImage FIT support.
Change-Id: I6e6cd9f78fb402bd54d684097326d26eb78d552a
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/25523
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
In commit d87e4b34 (stage_cache: Add rmodule params in metadata)
the cbmem stage cache was updatd to keep track of the arg from
struct prog in the metadata. However, external stage cache did
not get the same change. Fix that.
BUG=b:72728953
Change-Id: Ifdaaa255cac0f413856410ff61bfb411a9554a31
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/25794
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Justin TerAvest <teravest@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
In https://review.coreboot.org/25383 people were confused about the
ordering of bootmem calls w.r.t. when entries are exposed to the OS. To
alleviate this add a notion of bootmem being initialized. In addition to
that, only mark bootmem initialized when bootmem_write_memory_table() is
called. Any other calls to bootmem before that will report an error on
the console.
Change-Id: I5bc31f555038ccabb82d902c54f95858679b1695
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/25503
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
AMD AGESA returns DIMM info in SMBIOS format. dimm_info expects the data
in SPD format. These methods will be used to update amd_late_init.c so
it sets the correct values.
BUG=b:65403853
TEST=Built and booted grunt. Methods are not called in this commit. So
they were tested with the later commit by verifying the output of
dmidecode.
Change-Id: Id9fa98e9aad83dfd0a86f45e18b3c312665dce9b
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Reviewed-on: https://review.coreboot.org/25412
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martinroth@google.com>
Fix the error below.
```
src/lib/lzmadecode.c: In function 'LzmaDecode':
src/lib/lzmadecode.c:77:2: error: macro expands to multiple statements \
[-Werror=multistatement-macros]
Range -= bound; \
^~~~~
src/lib/lzmadecode.c:300:7: note: in expansion of macro 'UpdateBit1'
UpdateBit1(prob);
^~~~~~~~~~
src/lib/lzmadecode.c:299:8: note: some parts of macro expansion are not \
guarded by this 'else' clause
} else
^~~~
cc1: all warnings being treated as errors
```
The macro is defined as below.
#define UpdateBit1(p) \
Range -= bound; \
Code -= bound; \
*(p) -= (*(p)) >> kNumMoveBits
Found-by: gcc-8 (Debian 8-20180402-1) 8.0.1 20180402 (experimental) [trunk revision 259004]
Fixes: 35af5c47 (src/lib: Fix spacing)
Change-Id: Ife0688541e23c05e26e429a6d8caee7e2d425b1b
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Signed-off-by: Paul Menzel <pmenzel@molgen.mpg.de>
Reviewed-on: https://review.coreboot.org/25549
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Leverage the stage_cache mechanism to store a non-specific type
of data. This is not interesting when the location for the cache
is in cbmem. However it will be more useful when an external
location is used, e.g. when the cache is in TSEG, locked from user
modification.
Change-Id: Iaf0b25ebe14c176bbd24fc8942f902f627ca8e6f
Signed-off-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Reviewed-on: https://review.coreboot.org/23518
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
