GNUBoot
/
coreboot-kgpe-d16
2
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
coreboot-kgpe-d16/src/arch/x86/cbmem.c

80 lines
1.9 KiB
C
Raw Normal View History

CBMEM AMD: Remove references to global high_tables_base Prepare for removal of globals high_tables_base and _size by replacing the references with a helper function. Added set_top_of_ram_once() may be called several times, but only the first call (with non-zero argument) takes effect. Change-Id: I5b5f71630f03b6a01e9c8ff96cb78e9da03e5cc3 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/3894 Tested-by: build bot (Jenkins) Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@google.com>
2013-06-27 07:20:09 +02:00
/*
* This file is part of the coreboot project.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; version 2 of the License.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*/
CBMEM: Fix allocation for static CBMEM CBMEM console buffer size is adjustable in menuconfig, but this would not correctly adjust the overall allocation made for CBMEM. HIGH_MEMORY_SIZE is aligned to 64kB and definitions are moved down in the header file as HIGH_MEMORY_SIZE is not used with DYNAMIC_CBMEM. Try to continue boot even if CBMEM cannot be created. This error would only occur during development of new ports anyways and more log output is better. Change-Id: I4ee2df601b12ab6532ffcae8897775ecaa2fc05f Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/4621 Reviewed-by: Aaron Durbin <adurbin@google.com> Tested-by: build bot (Jenkins)
2014-01-04 08:42:02 +01:00
#include <stdlib.h>
CBMEM AMD: Remove references to global high_tables_base Prepare for removal of globals high_tables_base and _size by replacing the references with a helper function. Added set_top_of_ram_once() may be called several times, but only the first call (with non-zero argument) takes effect. Change-Id: I5b5f71630f03b6a01e9c8ff96cb78e9da03e5cc3 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/3894 Tested-by: build bot (Jenkins) Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@google.com>
2013-06-27 07:20:09 +02:00
#include <cbmem.h>
CBMEM ACPI: Move resume handler Handler is ACPI/x86 specific so move details out of cbmem code. With static CBMEM initialisation, ramstage will need to test for S3 wakeup condition so publish also acpi_is_wakeup(). Change-Id: If591535448cdd24a54262b534c1a828fc13da759 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/4619 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Aaron Durbin <adurbin@google.com>
2014-01-03 14:15:22 +01:00
#include <arch/acpi.h>
CBMEM AMD: Remove references to global high_tables_base Prepare for removal of globals high_tables_base and _size by replacing the references with a helper function. Added set_top_of_ram_once() may be called several times, but only the first call (with non-zero argument) takes effect. Change-Id: I5b5f71630f03b6a01e9c8ff96cb78e9da03e5cc3 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/3894 Tested-by: build bot (Jenkins) Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@google.com>
2013-06-27 07:20:09 +02:00
CBMEM: Support DYNAMIC_CBMEM with LATE_CBMEM_INIT We can now create CBMEM with dynamic allocation even if CBMEM location is resolved late in ramstage. Change-Id: I8529ccbcd4a0e567ebe0a46232ac5d16476e81a8 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/7861 Reviewed-by: Aaron Durbin <adurbin@google.com> Tested-by: build bot (Jenkins)
2014-12-19 07:20:45 +01:00
#if IS_ENABLED(CONFIG_LATE_CBMEM_INIT)
CBMEM: Add cbmem_locate_table() For both romstage and ramstage, this calls an arch-specific function get_cbmem_table() to resolve the base and size of CBMEM region. In ramstage, the result is cached as the query may be relatively slow involving multiple PCI configuration reads. For x86 CBMEM tables are located right below top of low ram and have fixed size of HIGH_MEMORY_SIZE in EARLY_CBMEM_INIT implementation. Change-Id: Ie8d16eb30cd5c3860fff243f36bd4e7d8827a782 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/3558 Tested-by: build bot (Jenkins) Reviewed-by: Aaron Durbin <adurbin@google.com>
2013-09-04 12:31:39 +02:00
compiler.h: add __weak macro Instead of writing out '__attribute__((weak))' use a shorter form. Change-Id: If418a1d55052780077febd2d8f2089021f414b91 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: https://review.coreboot.org/25767 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-by: Justin TerAvest <teravest@chromium.org>
2018-04-21 22:45:32 +02:00
void __weak backup_top_of_low_cacheable(uintptr_t ramtop)
CBMEM: Backup top_of_ram instead of cbmem_toc AMD northbridges have a complex way to resolve top_of_ram. Once it is resolved, it is stored in NVRAM to be used on resume. TODO: Redesign these get_top_of_ram() functions from scratch. Change-Id: I3cceb7e9b8b07620dacf138e99f98dc818c65341 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/3557 Tested-by: build bot (Jenkins) Reviewed-by: Aaron Durbin <adurbin@google.com>
2013-09-04 12:26:11 +02:00
{
arch/x86: Wrap lines at 80 columns Fix the following warning detected by checkpatch.pl: WARNING: line over 80 characters TEST=Build and run on Galileo Gen2 Change-Id: I3495cd30d1737d9ee728c8a9e72bd426d7a69c37 Signed-off-by: Lee Leahy <Leroy.P.Leahy@intel.com> Reviewed-on: https://review.coreboot.org/18864 Tested-by: build bot (Jenkins) Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2017-03-16 23:18:22 +01:00
/* Do nothing. Chipset may have implementation to save ramtop in NVRAM.
*/
CBMEM: Backup top_of_ram instead of cbmem_toc AMD northbridges have a complex way to resolve top_of_ram. Once it is resolved, it is stored in NVRAM to be used on resume. TODO: Redesign these get_top_of_ram() functions from scratch. Change-Id: I3cceb7e9b8b07620dacf138e99f98dc818c65341 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/3557 Tested-by: build bot (Jenkins) Reviewed-by: Aaron Durbin <adurbin@google.com>
2013-09-04 12:26:11 +02:00
}
compiler.h: add __weak macro Instead of writing out '__attribute__((weak))' use a shorter form. Change-Id: If418a1d55052780077febd2d8f2089021f414b91 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: https://review.coreboot.org/25767 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-by: Justin TerAvest <teravest@chromium.org>
2018-04-21 22:45:32 +02:00
uintptr_t __weak restore_top_of_low_cacheable(void)
CBMEM AMD: Remove references to global high_tables_base Prepare for removal of globals high_tables_base and _size by replacing the references with a helper function. Added set_top_of_ram_once() may be called several times, but only the first call (with non-zero argument) takes effect. Change-Id: I5b5f71630f03b6a01e9c8ff96cb78e9da03e5cc3 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/3894 Tested-by: build bot (Jenkins) Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@google.com>
2013-06-27 07:20:09 +02:00
{
CBMEM: Add config CBMEM_TOP_BACKUP AGESA and binaryPI boards have no easy way to determine correct cbmem_top() location early enough when GFXUMA is enabled, so they will use these functions with EARLY_CBMEM_INIT as well. At the end of AmdInitPost() the decisions of UMA base and size have not been written to hardware yet. The decisions are stored inside AGESA heap object we cannot locate from coreboot proper until after AmdInitEnv(). Modify code such that weak backup functions are only defined for LATE_CBMEM_INIT; they are somewhat troublesome to handle. Change-Id: Ifef4f75b36bc6dee6cd56d1d9164281d9b2a4f2a Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19306 Reviewed-by: Aaron Durbin <adurbin@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2017-04-19 06:37:38 +02:00
return 0;
x86: provide consistent cbmem_top() for CONFIG_LATE_CBMEM_INIT For x86 systems employing CONFIG_LATE_CBMEM_INIT, set_top_of_ram() is called in ramstage to note the upper address of the 32-bit address space. This in turn is consumed by cbmem. However, in this scenario cbmem_top() cannot always be relied upon because get_top_of_ram() doesn't return the same value provided to set_top_of_ram(). To fix the inconsistency in ramstage save the value passed in to set_top_of_ram() and defer to it as the return value for cbmem_top(). Change-Id: Ida796fb836c59b9776019e7f8b3f2cd71156f0e5 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/10313 Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Tested-by: build bot (Jenkins)
2015-05-26 18:15:45 +02:00
}
CBMEM: Add config CBMEM_TOP_BACKUP AGESA and binaryPI boards have no easy way to determine correct cbmem_top() location early enough when GFXUMA is enabled, so they will use these functions with EARLY_CBMEM_INIT as well. At the end of AmdInitPost() the decisions of UMA base and size have not been written to hardware yet. The decisions are stored inside AGESA heap object we cannot locate from coreboot proper until after AmdInitEnv(). Modify code such that weak backup functions are only defined for LATE_CBMEM_INIT; they are somewhat troublesome to handle. Change-Id: Ifef4f75b36bc6dee6cd56d1d9164281d9b2a4f2a Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19306 Reviewed-by: Aaron Durbin <adurbin@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2017-04-19 06:37:38 +02:00
#endif /* LATE_CBMEM_INIT */
CBMEM: Fail builds on missing get_top_of_ram() Dummy get_top_of_ram() is removed from romstage to fail already at build-time for cases where cbmem_initialize() would not complete. The mechanisms behind CAR_GLOBAL migration only work correctly when romstage can succesfully make the cbmem_initialize() call. Change-Id: I359820fb196ef187b9aa2e8a3e8f658a0550f237 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/3969 Tested-by: build bot (Jenkins) Reviewed-by: Aaron Durbin <adurbin@google.com>
2013-10-11 23:22:57 +02:00
CBMEM: Add config CBMEM_TOP_BACKUP AGESA and binaryPI boards have no easy way to determine correct cbmem_top() location early enough when GFXUMA is enabled, so they will use these functions with EARLY_CBMEM_INIT as well. At the end of AmdInitPost() the decisions of UMA base and size have not been written to hardware yet. The decisions are stored inside AGESA heap object we cannot locate from coreboot proper until after AmdInitEnv(). Modify code such that weak backup functions are only defined for LATE_CBMEM_INIT; they are somewhat troublesome to handle. Change-Id: Ifef4f75b36bc6dee6cd56d1d9164281d9b2a4f2a Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19306 Reviewed-by: Aaron Durbin <adurbin@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2017-04-19 06:37:38 +02:00
#if IS_ENABLED(CONFIG_CBMEM_TOP_BACKUP)
CBMEM: Clarify CBMEM_TOP_BACKUP function usage The deprecated LATE_CBMEM_INIT function is renamed: set_top_of_ram -> set_late_cbmem_top Obscure term top_of_ram is replaced: backup_top_of_ram -> backup_top_of_low_cacheable get_top_of_ram -> restore_top_of_low_cacheable New function that always resolves to CBMEM top boundary, with or without SMM, is named restore_cbmem_top(). Change-Id: I61d20f94840ad61e9fd55976e5aa8c27040b8fb7 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19377 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org> Reviewed-by: Philippe Mathieu-Daudé <philippe.mathieu.daude@gmail.com>
2017-04-19 18:57:01 +02:00
static void *cbmem_top_backup;
CBMEM: Add config CBMEM_TOP_BACKUP AGESA and binaryPI boards have no easy way to determine correct cbmem_top() location early enough when GFXUMA is enabled, so they will use these functions with EARLY_CBMEM_INIT as well. At the end of AmdInitPost() the decisions of UMA base and size have not been written to hardware yet. The decisions are stored inside AGESA heap object we cannot locate from coreboot proper until after AmdInitEnv(). Modify code such that weak backup functions are only defined for LATE_CBMEM_INIT; they are somewhat troublesome to handle. Change-Id: Ifef4f75b36bc6dee6cd56d1d9164281d9b2a4f2a Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19306 Reviewed-by: Aaron Durbin <adurbin@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2017-04-19 06:37:38 +02:00
CBMEM: Clarify CBMEM_TOP_BACKUP function usage The deprecated LATE_CBMEM_INIT function is renamed: set_top_of_ram -> set_late_cbmem_top Obscure term top_of_ram is replaced: backup_top_of_ram -> backup_top_of_low_cacheable get_top_of_ram -> restore_top_of_low_cacheable New function that always resolves to CBMEM top boundary, with or without SMM, is named restore_cbmem_top(). Change-Id: I61d20f94840ad61e9fd55976e5aa8c27040b8fb7 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19377 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org> Reviewed-by: Philippe Mathieu-Daudé <philippe.mathieu.daude@gmail.com>
2017-04-19 18:57:01 +02:00
void set_late_cbmem_top(uintptr_t ramtop)
CBMEM: Fail builds on missing get_top_of_ram() Dummy get_top_of_ram() is removed from romstage to fail already at build-time for cases where cbmem_initialize() would not complete. The mechanisms behind CAR_GLOBAL migration only work correctly when romstage can succesfully make the cbmem_initialize() call. Change-Id: I359820fb196ef187b9aa2e8a3e8f658a0550f237 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/3969 Tested-by: build bot (Jenkins) Reviewed-by: Aaron Durbin <adurbin@google.com>
2013-10-11 23:22:57 +02:00
{
CBMEM: Clarify CBMEM_TOP_BACKUP function usage The deprecated LATE_CBMEM_INIT function is renamed: set_top_of_ram -> set_late_cbmem_top Obscure term top_of_ram is replaced: backup_top_of_ram -> backup_top_of_low_cacheable get_top_of_ram -> restore_top_of_low_cacheable New function that always resolves to CBMEM top boundary, with or without SMM, is named restore_cbmem_top(). Change-Id: I61d20f94840ad61e9fd55976e5aa8c27040b8fb7 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19377 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org> Reviewed-by: Philippe Mathieu-Daudé <philippe.mathieu.daude@gmail.com>
2017-04-19 18:57:01 +02:00
backup_top_of_low_cacheable(ramtop);
CBMEM: Add config CBMEM_TOP_BACKUP AGESA and binaryPI boards have no easy way to determine correct cbmem_top() location early enough when GFXUMA is enabled, so they will use these functions with EARLY_CBMEM_INIT as well. At the end of AmdInitPost() the decisions of UMA base and size have not been written to hardware yet. The decisions are stored inside AGESA heap object we cannot locate from coreboot proper until after AmdInitEnv(). Modify code such that weak backup functions are only defined for LATE_CBMEM_INIT; they are somewhat troublesome to handle. Change-Id: Ifef4f75b36bc6dee6cd56d1d9164281d9b2a4f2a Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19306 Reviewed-by: Aaron Durbin <adurbin@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2017-04-19 06:37:38 +02:00
if (ENV_RAMSTAGE)
CBMEM: Clarify CBMEM_TOP_BACKUP function usage The deprecated LATE_CBMEM_INIT function is renamed: set_top_of_ram -> set_late_cbmem_top Obscure term top_of_ram is replaced: backup_top_of_ram -> backup_top_of_low_cacheable get_top_of_ram -> restore_top_of_low_cacheable New function that always resolves to CBMEM top boundary, with or without SMM, is named restore_cbmem_top(). Change-Id: I61d20f94840ad61e9fd55976e5aa8c27040b8fb7 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19377 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org> Reviewed-by: Philippe Mathieu-Daudé <philippe.mathieu.daude@gmail.com>
2017-04-19 18:57:01 +02:00
cbmem_top_backup = (void *)ramtop;
}
/* Top of CBMEM is at highest usable DRAM address below 4GiB. */
compiler.h: add __weak macro Instead of writing out '__attribute__((weak))' use a shorter form. Change-Id: If418a1d55052780077febd2d8f2089021f414b91 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: https://review.coreboot.org/25767 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-by: Justin TerAvest <teravest@chromium.org>
2018-04-21 22:45:32 +02:00
uintptr_t __weak restore_cbmem_top(void)
CBMEM: Clarify CBMEM_TOP_BACKUP function usage The deprecated LATE_CBMEM_INIT function is renamed: set_top_of_ram -> set_late_cbmem_top Obscure term top_of_ram is replaced: backup_top_of_ram -> backup_top_of_low_cacheable get_top_of_ram -> restore_top_of_low_cacheable New function that always resolves to CBMEM top boundary, with or without SMM, is named restore_cbmem_top(). Change-Id: I61d20f94840ad61e9fd55976e5aa8c27040b8fb7 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19377 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org> Reviewed-by: Philippe Mathieu-Daudé <philippe.mathieu.daude@gmail.com>
2017-04-19 18:57:01 +02:00
{
AGESA boards: Fix regressions with LATE_CBMEM_INIT Some configurations of AGESA boards fail to boot after commit 61be360 AGESA: Fix UMA calculations Implementation of cbmem_find() for ENV_ROMSTAGE expects that CBMEM has already been initialized. In the case of LATE_CBMEM_INIT boards, this is not the case and cbmem_top() returned NULL prior to the offending commmit. By definition LATE_CBMEM_INIT does not have known cbmem_top() in ENV_ROMSTAGE except for possible ACPI S3 resume path. Change-Id: Icb8f44661d479e5ad43b123600305dcbc3ce11e1 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/20590 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Martin Roth <martinroth@google.com> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2017-07-15 12:03:11 +02:00
if (IS_ENABLED(CONFIG_LATE_CBMEM_INIT) && ENV_ROMSTAGE)
if (!acpi_is_wakeup_s3())
return 0;
CBMEM: Clarify CBMEM_TOP_BACKUP function usage The deprecated LATE_CBMEM_INIT function is renamed: set_top_of_ram -> set_late_cbmem_top Obscure term top_of_ram is replaced: backup_top_of_ram -> backup_top_of_low_cacheable get_top_of_ram -> restore_top_of_low_cacheable New function that always resolves to CBMEM top boundary, with or without SMM, is named restore_cbmem_top(). Change-Id: I61d20f94840ad61e9fd55976e5aa8c27040b8fb7 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19377 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org> Reviewed-by: Philippe Mathieu-Daudé <philippe.mathieu.daude@gmail.com>
2017-04-19 18:57:01 +02:00
return restore_top_of_low_cacheable();
CBMEM: Fail builds on missing get_top_of_ram() Dummy get_top_of_ram() is removed from romstage to fail already at build-time for cases where cbmem_initialize() would not complete. The mechanisms behind CAR_GLOBAL migration only work correctly when romstage can succesfully make the cbmem_initialize() call. Change-Id: I359820fb196ef187b9aa2e8a3e8f658a0550f237 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/3969 Tested-by: build bot (Jenkins) Reviewed-by: Aaron Durbin <adurbin@google.com>
2013-10-11 23:22:57 +02:00
}
CBMEM: Define cbmem_top() just once for x86 It is expected this will always be a casted get_top_of_ram() call on x86, no reason to do that under chipset. Change-Id: I3a49abe13ca44bf4ca1e26d1b3baf954bc5a29b7 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/3972 Tested-by: build bot (Jenkins) Reviewed-by: Aaron Durbin <adurbin@google.com>
2013-10-13 03:15:40 +02:00
void *cbmem_top(void)
{
CBMEM: Clarify CBMEM_TOP_BACKUP function usage The deprecated LATE_CBMEM_INIT function is renamed: set_top_of_ram -> set_late_cbmem_top Obscure term top_of_ram is replaced: backup_top_of_ram -> backup_top_of_low_cacheable get_top_of_ram -> restore_top_of_low_cacheable New function that always resolves to CBMEM top boundary, with or without SMM, is named restore_cbmem_top(). Change-Id: I61d20f94840ad61e9fd55976e5aa8c27040b8fb7 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19377 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org> Reviewed-by: Philippe Mathieu-Daudé <philippe.mathieu.daude@gmail.com>
2017-04-19 18:57:01 +02:00
uintptr_t top_backup;
CBMEM: Add config CBMEM_TOP_BACKUP AGESA and binaryPI boards have no easy way to determine correct cbmem_top() location early enough when GFXUMA is enabled, so they will use these functions with EARLY_CBMEM_INIT as well. At the end of AmdInitPost() the decisions of UMA base and size have not been written to hardware yet. The decisions are stored inside AGESA heap object we cannot locate from coreboot proper until after AmdInitEnv(). Modify code such that weak backup functions are only defined for LATE_CBMEM_INIT; they are somewhat troublesome to handle. Change-Id: Ifef4f75b36bc6dee6cd56d1d9164281d9b2a4f2a Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19306 Reviewed-by: Aaron Durbin <adurbin@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2017-04-19 06:37:38 +02:00
CBMEM: Clarify CBMEM_TOP_BACKUP function usage The deprecated LATE_CBMEM_INIT function is renamed: set_top_of_ram -> set_late_cbmem_top Obscure term top_of_ram is replaced: backup_top_of_ram -> backup_top_of_low_cacheable get_top_of_ram -> restore_top_of_low_cacheable New function that always resolves to CBMEM top boundary, with or without SMM, is named restore_cbmem_top(). Change-Id: I61d20f94840ad61e9fd55976e5aa8c27040b8fb7 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19377 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org> Reviewed-by: Philippe Mathieu-Daudé <philippe.mathieu.daude@gmail.com>
2017-04-19 18:57:01 +02:00
if (ENV_RAMSTAGE && cbmem_top_backup != NULL)
return cbmem_top_backup;
x86: provide consistent cbmem_top() for CONFIG_LATE_CBMEM_INIT For x86 systems employing CONFIG_LATE_CBMEM_INIT, set_top_of_ram() is called in ramstage to note the upper address of the 32-bit address space. This in turn is consumed by cbmem. However, in this scenario cbmem_top() cannot always be relied upon because get_top_of_ram() doesn't return the same value provided to set_top_of_ram(). To fix the inconsistency in ramstage save the value passed in to set_top_of_ram() and defer to it as the return value for cbmem_top(). Change-Id: Ida796fb836c59b9776019e7f8b3f2cd71156f0e5 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/10313 Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Tested-by: build bot (Jenkins)
2015-05-26 18:15:45 +02:00
CBMEM: Clarify CBMEM_TOP_BACKUP function usage The deprecated LATE_CBMEM_INIT function is renamed: set_top_of_ram -> set_late_cbmem_top Obscure term top_of_ram is replaced: backup_top_of_ram -> backup_top_of_low_cacheable get_top_of_ram -> restore_top_of_low_cacheable New function that always resolves to CBMEM top boundary, with or without SMM, is named restore_cbmem_top(). Change-Id: I61d20f94840ad61e9fd55976e5aa8c27040b8fb7 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19377 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org> Reviewed-by: Philippe Mathieu-Daudé <philippe.mathieu.daude@gmail.com>
2017-04-19 18:57:01 +02:00
top_backup = restore_cbmem_top();
x86: provide consistent cbmem_top() for CONFIG_LATE_CBMEM_INIT For x86 systems employing CONFIG_LATE_CBMEM_INIT, set_top_of_ram() is called in ramstage to note the upper address of the 32-bit address space. This in turn is consumed by cbmem. However, in this scenario cbmem_top() cannot always be relied upon because get_top_of_ram() doesn't return the same value provided to set_top_of_ram(). To fix the inconsistency in ramstage save the value passed in to set_top_of_ram() and defer to it as the return value for cbmem_top(). Change-Id: Ida796fb836c59b9776019e7f8b3f2cd71156f0e5 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/10313 Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Tested-by: build bot (Jenkins)
2015-05-26 18:15:45 +02:00
CBMEM: Add config CBMEM_TOP_BACKUP AGESA and binaryPI boards have no easy way to determine correct cbmem_top() location early enough when GFXUMA is enabled, so they will use these functions with EARLY_CBMEM_INIT as well. At the end of AmdInitPost() the decisions of UMA base and size have not been written to hardware yet. The decisions are stored inside AGESA heap object we cannot locate from coreboot proper until after AmdInitEnv(). Modify code such that weak backup functions are only defined for LATE_CBMEM_INIT; they are somewhat troublesome to handle. Change-Id: Ifef4f75b36bc6dee6cd56d1d9164281d9b2a4f2a Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19306 Reviewed-by: Aaron Durbin <adurbin@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2017-04-19 06:37:38 +02:00
if (ENV_RAMSTAGE)
CBMEM: Clarify CBMEM_TOP_BACKUP function usage The deprecated LATE_CBMEM_INIT function is renamed: set_top_of_ram -> set_late_cbmem_top Obscure term top_of_ram is replaced: backup_top_of_ram -> backup_top_of_low_cacheable get_top_of_ram -> restore_top_of_low_cacheable New function that always resolves to CBMEM top boundary, with or without SMM, is named restore_cbmem_top(). Change-Id: I61d20f94840ad61e9fd55976e5aa8c27040b8fb7 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19377 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org> Reviewed-by: Philippe Mathieu-Daudé <philippe.mathieu.daude@gmail.com>
2017-04-19 18:57:01 +02:00
cbmem_top_backup = (void *)top_backup;
CBMEM: Add config CBMEM_TOP_BACKUP AGESA and binaryPI boards have no easy way to determine correct cbmem_top() location early enough when GFXUMA is enabled, so they will use these functions with EARLY_CBMEM_INIT as well. At the end of AmdInitPost() the decisions of UMA base and size have not been written to hardware yet. The decisions are stored inside AGESA heap object we cannot locate from coreboot proper until after AmdInitEnv(). Modify code such that weak backup functions are only defined for LATE_CBMEM_INIT; they are somewhat troublesome to handle. Change-Id: Ifef4f75b36bc6dee6cd56d1d9164281d9b2a4f2a Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19306 Reviewed-by: Aaron Durbin <adurbin@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2017-04-19 06:37:38 +02:00
CBMEM: Clarify CBMEM_TOP_BACKUP function usage The deprecated LATE_CBMEM_INIT function is renamed: set_top_of_ram -> set_late_cbmem_top Obscure term top_of_ram is replaced: backup_top_of_ram -> backup_top_of_low_cacheable get_top_of_ram -> restore_top_of_low_cacheable New function that always resolves to CBMEM top boundary, with or without SMM, is named restore_cbmem_top(). Change-Id: I61d20f94840ad61e9fd55976e5aa8c27040b8fb7 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19377 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org> Reviewed-by: Philippe Mathieu-Daudé <philippe.mathieu.daude@gmail.com>
2017-04-19 18:57:01 +02:00
return (void *)top_backup;
CBMEM: Define cbmem_top() just once for x86 It is expected this will always be a casted get_top_of_ram() call on x86, no reason to do that under chipset. Change-Id: I3a49abe13ca44bf4ca1e26d1b3baf954bc5a29b7 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/3972 Tested-by: build bot (Jenkins) Reviewed-by: Aaron Durbin <adurbin@google.com>
2013-10-13 03:15:40 +02:00
}
CBMEM: Support DYNAMIC_CBMEM with LATE_CBMEM_INIT We can now create CBMEM with dynamic allocation even if CBMEM location is resolved late in ramstage. Change-Id: I8529ccbcd4a0e567ebe0a46232ac5d16476e81a8 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/7861 Reviewed-by: Aaron Durbin <adurbin@google.com> Tested-by: build bot (Jenkins)
2014-12-19 07:20:45 +01:00
CBMEM: Add config CBMEM_TOP_BACKUP AGESA and binaryPI boards have no easy way to determine correct cbmem_top() location early enough when GFXUMA is enabled, so they will use these functions with EARLY_CBMEM_INIT as well. At the end of AmdInitPost() the decisions of UMA base and size have not been written to hardware yet. The decisions are stored inside AGESA heap object we cannot locate from coreboot proper until after AmdInitEnv(). Modify code such that weak backup functions are only defined for LATE_CBMEM_INIT; they are somewhat troublesome to handle. Change-Id: Ifef4f75b36bc6dee6cd56d1d9164281d9b2a4f2a Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: https://review.coreboot.org/19306 Reviewed-by: Aaron Durbin <adurbin@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2017-04-19 06:37:38 +02:00
#endif /* CBMEM_TOP_BACKUP */
CBMEM ACPI: Move resume handler Handler is ACPI/x86 specific so move details out of cbmem code. With static CBMEM initialisation, ramstage will need to test for S3 wakeup condition so publish also acpi_is_wakeup(). Change-Id: If591535448cdd24a54262b534c1a828fc13da759 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/4619 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Aaron Durbin <adurbin@google.com>
2014-01-03 14:15:22 +01:00
CBMEM: Implement cbmem_run_init_hooks() stub Until we completely can unify early_variables, use these to handle CBMEM update hooks for both romstage and ramstage. For x86, CAR_MIGRATE serves the purpose of romstage hooks. Change-Id: I100ebc0e35e1b7091b4f287ca37f539fd7c9fa7a Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/7876 Reviewed-by: Aaron Durbin <adurbin@google.com> Tested-by: build bot (Jenkins)
2014-12-18 17:30:29 +01:00
/* Something went wrong, our high memory area got wiped */
CBMEM ACPI: Move resume handler Handler is ACPI/x86 specific so move details out of cbmem code. With static CBMEM initialisation, ramstage will need to test for S3 wakeup condition so publish also acpi_is_wakeup(). Change-Id: If591535448cdd24a54262b534c1a828fc13da759 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/4619 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Aaron Durbin <adurbin@google.com>
2014-01-03 14:15:22 +01:00
void cbmem_fail_resume(void)
{
CBMEM: Implement cbmem_run_init_hooks() stub Until we completely can unify early_variables, use these to handle CBMEM update hooks for both romstage and ramstage. For x86, CAR_MIGRATE serves the purpose of romstage hooks. Change-Id: I100ebc0e35e1b7091b4f287ca37f539fd7c9fa7a Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/7876 Reviewed-by: Aaron Durbin <adurbin@google.com> Tested-by: build bot (Jenkins)
2014-12-18 17:30:29 +01:00
#if !defined(__PRE_RAM__) && IS_ENABLED(CONFIG_HAVE_ACPI_RESUME)
/* ACPI resume needs to be cleared in the fail-to-recover case, but that
* condition is only handled during ramstage. */
CBMEM ACPI: Move resume handler Handler is ACPI/x86 specific so move details out of cbmem code. With static CBMEM initialisation, ramstage will need to test for S3 wakeup condition so publish also acpi_is_wakeup(). Change-Id: If591535448cdd24a54262b534c1a828fc13da759 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/4619 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Aaron Durbin <adurbin@google.com>
2014-01-03 14:15:22 +01:00
acpi_fail_wakeup();
#endif
}