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

65 lines
1.5 KiB
C
Raw Normal View History

src/arch/x86: Convert to SPDX license header This also drops individual copyright notices, all mentioned authors in that part of the tree are listed in AUTHORS. Change-Id: Ib5a92bb46ff2b9d2928aae3763daec71747044c2 Signed-off-by: Patrick Georgi <pgeorgi@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/39284 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
2020-03-04 15:10:45 +01:00
/* SPDX-License-Identifier: GPL-2.0-only */
Prepare the BIOS data areas before device init. Since we do not run option roms in normal mode nothing was initializing the BDA/EBDA and yet Linux depends very much on it having sane values here. For the most part the kernel tries to work around this not being initialized, but every once in awhile (1/300 boots or so) it would end up reading something that looked sane from BDA but was not and then it would panic. In this change the EBDA is unconditionally setup before devices are initialized. I'm not set on the location in dev_initialize() but there does not seem to be another place to hook it in so that it runs just once for ALL platforms regardless of whether they use option roms or not. (possibly hardwaremain?) The EBDA setup code has been moved into its own location in arch/x86/lib/ebda.c so it can be compiled in even if the option rom code is not. The low memory size is still set to 1MB which is enough to make linux happy without having to hook into each mainboard to get a more appropriate value. The setup_ebda() function takes inputs so it could be changed for a mainboard if needed. OLD/BROKEN would read garbage. Examples from different boots: ebda_addr=0x75e80 lowmem=0x1553400 ebda_addr=0x5e080 lowmem=0x3e51400 ebda_addr=0x7aa80 lowmem=0x2f8a800 NEW/FIXED now reads consistent values: ebda_addr=0xf6000 lowmem=0x100000 Change-Id: I6cb79f0e3e43cc65f7e5fe98b6cad1a557ccd949 Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: http://review.coreboot.org/769 Tested-by: build bot (Jenkins) Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2012-01-17 18:03:11 +01:00
acpi: Move ACPI table support out of arch/x86 (3/5) This change moves all ACPI table support in coreboot currently living under arch/x86 into common code to make it architecture independent. ACPI table generation is not really tied to any architecture and hence it makes sense to move this to its own directory. In order to make it easier to review, this change is being split into multiple CLs. This is change 3/5 which basically is generated by running the following command: $ git grep -iIl "arch/acpi" | xargs sed -i 's/arch\/acpi/acpi\/acpi/g' BUG=b:155428745 Change-Id: I16b1c45d954d6440fb9db1d3710063a47b582eae Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/40938 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
2020-05-02 19:24:23 +02:00
#include <acpi/acpi.h>
arch/x86/ebda.c: Move setting up ebda to a BS hook device.c should not hold arch specific code. Change-Id: I9dfdb905a83916c0e9d298e1c38da89f6bc5e038 Signed-off-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-on: https://review.coreboot.org/c/coreboot/+/64297 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Elyes Haouas <ehaouas@noos.fr> Reviewed-by: Werner Zeh <werner.zeh@siemens.com>
2022-05-12 20:39:16 +02:00
#include <bootstate.h>
arch/x86/ebda: Change memcpy() to endian wrappers Change memcpy()s and memset()s to endian.h wrappers for consistency and safety. Add zero_n() wrapper to safely clear memory. Change-Id: If155d82608c81992f05eae4d2223de6f506e98c5 Signed-off-by: Ryan Salsamendi <rsalsamendi@hotmail.com> Reviewed-on: https://review.coreboot.org/20240 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2017-06-16 22:15:31 +02:00
#include <commonlib/endian.h>
arch/x86: Clean up includes Signed-off-by: Elyes Haouas <ehaouas@noos.fr> Change-Id: Id2db229dec2ed44333faaa8c53f3a2f9d66d52e0 Reviewed-on: https://review.coreboot.org/c/coreboot/+/68200 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Martin Roth <martin.roth@amd.corp-partner.google.com>
2022-10-07 10:02:38 +02:00
#include <types.h>
arch/x86/ebda.c: Move setting up ebda to a BS hook device.c should not hold arch specific code. Change-Id: I9dfdb905a83916c0e9d298e1c38da89f6bc5e038 Signed-off-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-on: https://review.coreboot.org/c/coreboot/+/64297 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Elyes Haouas <ehaouas@noos.fr> Reviewed-by: Werner Zeh <werner.zeh@siemens.com>
2022-05-12 20:39:16 +02:00
#define X86_BDA_SIZE 0x200
#define X86_BDA_BASE ((void *)0x400)
#define X86_EBDA_SEGMENT ((void *)0x40e)
#define X86_EBDA_LOWMEM ((void *)0x413)
#define DEFAULT_EBDA_LOWMEM (1024 << 10)
#define DEFAULT_EBDA_SEGMENT 0xF600
#define DEFAULT_EBDA_SIZE 0x400
arch/x86: Add ebda read/write functions into EBDA library This patch provides new APIs to write into EBDA area and read from EBDA area based on user input structure. Change-Id: I26d5c0ba82c842f0b734a8e0f03abf148737c5c4 Signed-off-by: Subrata Banik <subrata.banik@intel.com> Reviewed-on: https://review.coreboot.org/21536 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org> Reviewed-by: Sumeet R Pawnikar <sumeet.r.pawnikar@intel.com>
2017-09-10 16:36:22 +02:00
static void *get_ebda_start(void)
{
return (void *)((uintptr_t)DEFAULT_EBDA_SEGMENT << 4);
}
/*
* EBDA area is representing a 1KB memory area just below
* the top of conventional memory (below 1MB)
*/
arch/x86: Remove EARLY_EBDA_INIT support This is unused now. Change-Id: Ie8bc1d6761d66c5e1dda40c34c940cdba90646d2 Signed-off-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-on: https://review.coreboot.org/c/coreboot/+/36363 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Michael Niewöhner Reviewed-by: Nico Huber <nico.h@gmx.de> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2019-10-26 20:31:41 +02:00
static void setup_ebda(u32 low_memory_size, u16 ebda_segment, u16 ebda_size)
Prepare the BIOS data areas before device init. Since we do not run option roms in normal mode nothing was initializing the BDA/EBDA and yet Linux depends very much on it having sane values here. For the most part the kernel tries to work around this not being initialized, but every once in awhile (1/300 boots or so) it would end up reading something that looked sane from BDA but was not and then it would panic. In this change the EBDA is unconditionally setup before devices are initialized. I'm not set on the location in dev_initialize() but there does not seem to be another place to hook it in so that it runs just once for ALL platforms regardless of whether they use option roms or not. (possibly hardwaremain?) The EBDA setup code has been moved into its own location in arch/x86/lib/ebda.c so it can be compiled in even if the option rom code is not. The low memory size is still set to 1MB which is enough to make linux happy without having to hook into each mainboard to get a more appropriate value. The setup_ebda() function takes inputs so it could be changed for a mainboard if needed. OLD/BROKEN would read garbage. Examples from different boots: ebda_addr=0x75e80 lowmem=0x1553400 ebda_addr=0x5e080 lowmem=0x3e51400 ebda_addr=0x7aa80 lowmem=0x2f8a800 NEW/FIXED now reads consistent values: ebda_addr=0xf6000 lowmem=0x100000 Change-Id: I6cb79f0e3e43cc65f7e5fe98b6cad1a557ccd949 Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: http://review.coreboot.org/769 Tested-by: build bot (Jenkins) Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2012-01-17 18:03:11 +01:00
{
arch/x86: Fix undefined behavior Fixes report found by undefined behavior sanitizer. Dereferencing a pointer that is not aligned to the size of access is undefined behavior. Switch to memcpy() for unaligned write to EBDA_LOWMEM. Change other write16()s in setup_ebda() to memcpy() for consistency. Change-Id: I79814bd47a14ec59d84068b11d094dc2531995d9 Signed-off-by: Ryan Salsamendi <rsalsamendi@hotmail.com> Reviewed-on: https://review.coreboot.org/20132 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>
2017-06-09 21:01:39 +02:00
u16 low_memory_kb;
u16 ebda_kb;
void *ebda;
Prepare the BIOS data areas before device init. Since we do not run option roms in normal mode nothing was initializing the BDA/EBDA and yet Linux depends very much on it having sane values here. For the most part the kernel tries to work around this not being initialized, but every once in awhile (1/300 boots or so) it would end up reading something that looked sane from BDA but was not and then it would panic. In this change the EBDA is unconditionally setup before devices are initialized. I'm not set on the location in dev_initialize() but there does not seem to be another place to hook it in so that it runs just once for ALL platforms regardless of whether they use option roms or not. (possibly hardwaremain?) The EBDA setup code has been moved into its own location in arch/x86/lib/ebda.c so it can be compiled in even if the option rom code is not. The low memory size is still set to 1MB which is enough to make linux happy without having to hook into each mainboard to get a more appropriate value. The setup_ebda() function takes inputs so it could be changed for a mainboard if needed. OLD/BROKEN would read garbage. Examples from different boots: ebda_addr=0x75e80 lowmem=0x1553400 ebda_addr=0x5e080 lowmem=0x3e51400 ebda_addr=0x7aa80 lowmem=0x2f8a800 NEW/FIXED now reads consistent values: ebda_addr=0xf6000 lowmem=0x100000 Change-Id: I6cb79f0e3e43cc65f7e5fe98b6cad1a557ccd949 Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: http://review.coreboot.org/769 Tested-by: build bot (Jenkins) Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2012-01-17 18:03:11 +01:00
if (!low_memory_size || !ebda_segment || !ebda_size)
return;
arch/x86: Fix undefined behavior Fixes report found by undefined behavior sanitizer. Dereferencing a pointer that is not aligned to the size of access is undefined behavior. Switch to memcpy() for unaligned write to EBDA_LOWMEM. Change other write16()s in setup_ebda() to memcpy() for consistency. Change-Id: I79814bd47a14ec59d84068b11d094dc2531995d9 Signed-off-by: Ryan Salsamendi <rsalsamendi@hotmail.com> Reviewed-on: https://review.coreboot.org/20132 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>
2017-06-09 21:01:39 +02:00
low_memory_kb = low_memory_size >> 10;
ebda_kb = ebda_size >> 10;
arch/x86: Add ebda read/write functions into EBDA library This patch provides new APIs to write into EBDA area and read from EBDA area based on user input structure. Change-Id: I26d5c0ba82c842f0b734a8e0f03abf148737c5c4 Signed-off-by: Subrata Banik <subrata.banik@intel.com> Reviewed-on: https://review.coreboot.org/21536 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org> Reviewed-by: Sumeet R Pawnikar <sumeet.r.pawnikar@intel.com>
2017-09-10 16:36:22 +02:00
ebda = get_ebda_start();
arch/x86: Fix undefined behavior Fixes report found by undefined behavior sanitizer. Dereferencing a pointer that is not aligned to the size of access is undefined behavior. Switch to memcpy() for unaligned write to EBDA_LOWMEM. Change other write16()s in setup_ebda() to memcpy() for consistency. Change-Id: I79814bd47a14ec59d84068b11d094dc2531995d9 Signed-off-by: Ryan Salsamendi <rsalsamendi@hotmail.com> Reviewed-on: https://review.coreboot.org/20132 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>
2017-06-09 21:01:39 +02:00
Prepare the BIOS data areas before device init. Since we do not run option roms in normal mode nothing was initializing the BDA/EBDA and yet Linux depends very much on it having sane values here. For the most part the kernel tries to work around this not being initialized, but every once in awhile (1/300 boots or so) it would end up reading something that looked sane from BDA but was not and then it would panic. In this change the EBDA is unconditionally setup before devices are initialized. I'm not set on the location in dev_initialize() but there does not seem to be another place to hook it in so that it runs just once for ALL platforms regardless of whether they use option roms or not. (possibly hardwaremain?) The EBDA setup code has been moved into its own location in arch/x86/lib/ebda.c so it can be compiled in even if the option rom code is not. The low memory size is still set to 1MB which is enough to make linux happy without having to hook into each mainboard to get a more appropriate value. The setup_ebda() function takes inputs so it could be changed for a mainboard if needed. OLD/BROKEN would read garbage. Examples from different boots: ebda_addr=0x75e80 lowmem=0x1553400 ebda_addr=0x5e080 lowmem=0x3e51400 ebda_addr=0x7aa80 lowmem=0x2f8a800 NEW/FIXED now reads consistent values: ebda_addr=0xf6000 lowmem=0x100000 Change-Id: I6cb79f0e3e43cc65f7e5fe98b6cad1a557ccd949 Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: http://review.coreboot.org/769 Tested-by: build bot (Jenkins) Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2012-01-17 18:03:11 +01:00
/* clear BIOS DATA AREA */
arch/x86/ebda: Change memcpy() to endian wrappers Change memcpy()s and memset()s to endian.h wrappers for consistency and safety. Add zero_n() wrapper to safely clear memory. Change-Id: If155d82608c81992f05eae4d2223de6f506e98c5 Signed-off-by: Ryan Salsamendi <rsalsamendi@hotmail.com> Reviewed-on: https://review.coreboot.org/20240 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2017-06-16 22:15:31 +02:00
zero_n(X86_BDA_BASE, X86_BDA_SIZE);
Prepare the BIOS data areas before device init. Since we do not run option roms in normal mode nothing was initializing the BDA/EBDA and yet Linux depends very much on it having sane values here. For the most part the kernel tries to work around this not being initialized, but every once in awhile (1/300 boots or so) it would end up reading something that looked sane from BDA but was not and then it would panic. In this change the EBDA is unconditionally setup before devices are initialized. I'm not set on the location in dev_initialize() but there does not seem to be another place to hook it in so that it runs just once for ALL platforms regardless of whether they use option roms or not. (possibly hardwaremain?) The EBDA setup code has been moved into its own location in arch/x86/lib/ebda.c so it can be compiled in even if the option rom code is not. The low memory size is still set to 1MB which is enough to make linux happy without having to hook into each mainboard to get a more appropriate value. The setup_ebda() function takes inputs so it could be changed for a mainboard if needed. OLD/BROKEN would read garbage. Examples from different boots: ebda_addr=0x75e80 lowmem=0x1553400 ebda_addr=0x5e080 lowmem=0x3e51400 ebda_addr=0x7aa80 lowmem=0x2f8a800 NEW/FIXED now reads consistent values: ebda_addr=0xf6000 lowmem=0x100000 Change-Id: I6cb79f0e3e43cc65f7e5fe98b6cad1a557ccd949 Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: http://review.coreboot.org/769 Tested-by: build bot (Jenkins) Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2012-01-17 18:03:11 +01:00
arch/x86: Fix undefined behavior Fixes report found by undefined behavior sanitizer. Dereferencing a pointer that is not aligned to the size of access is undefined behavior. Switch to memcpy() for unaligned write to EBDA_LOWMEM. Change other write16()s in setup_ebda() to memcpy() for consistency. Change-Id: I79814bd47a14ec59d84068b11d094dc2531995d9 Signed-off-by: Ryan Salsamendi <rsalsamendi@hotmail.com> Reviewed-on: https://review.coreboot.org/20132 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>
2017-06-09 21:01:39 +02:00
/* Avoid unaligned write16() since it's undefined behavior */
arch/x86/ebda: Change memcpy() to endian wrappers Change memcpy()s and memset()s to endian.h wrappers for consistency and safety. Add zero_n() wrapper to safely clear memory. Change-Id: If155d82608c81992f05eae4d2223de6f506e98c5 Signed-off-by: Ryan Salsamendi <rsalsamendi@hotmail.com> Reviewed-on: https://review.coreboot.org/20240 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2017-06-16 22:15:31 +02:00
write_le16(X86_EBDA_LOWMEM, low_memory_kb);
write_le16(X86_EBDA_SEGMENT, ebda_segment);
Prepare the BIOS data areas before device init. Since we do not run option roms in normal mode nothing was initializing the BDA/EBDA and yet Linux depends very much on it having sane values here. For the most part the kernel tries to work around this not being initialized, but every once in awhile (1/300 boots or so) it would end up reading something that looked sane from BDA but was not and then it would panic. In this change the EBDA is unconditionally setup before devices are initialized. I'm not set on the location in dev_initialize() but there does not seem to be another place to hook it in so that it runs just once for ALL platforms regardless of whether they use option roms or not. (possibly hardwaremain?) The EBDA setup code has been moved into its own location in arch/x86/lib/ebda.c so it can be compiled in even if the option rom code is not. The low memory size is still set to 1MB which is enough to make linux happy without having to hook into each mainboard to get a more appropriate value. The setup_ebda() function takes inputs so it could be changed for a mainboard if needed. OLD/BROKEN would read garbage. Examples from different boots: ebda_addr=0x75e80 lowmem=0x1553400 ebda_addr=0x5e080 lowmem=0x3e51400 ebda_addr=0x7aa80 lowmem=0x2f8a800 NEW/FIXED now reads consistent values: ebda_addr=0xf6000 lowmem=0x100000 Change-Id: I6cb79f0e3e43cc65f7e5fe98b6cad1a557ccd949 Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: http://review.coreboot.org/769 Tested-by: build bot (Jenkins) Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2012-01-17 18:03:11 +01:00
/* Set up EBDA */
arch/x86/ebda: Change memcpy() to endian wrappers Change memcpy()s and memset()s to endian.h wrappers for consistency and safety. Add zero_n() wrapper to safely clear memory. Change-Id: If155d82608c81992f05eae4d2223de6f506e98c5 Signed-off-by: Ryan Salsamendi <rsalsamendi@hotmail.com> Reviewed-on: https://review.coreboot.org/20240 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2017-06-16 22:15:31 +02:00
zero_n(ebda, ebda_size);
write_le16(ebda, ebda_kb);
Prepare the BIOS data areas before device init. Since we do not run option roms in normal mode nothing was initializing the BDA/EBDA and yet Linux depends very much on it having sane values here. For the most part the kernel tries to work around this not being initialized, but every once in awhile (1/300 boots or so) it would end up reading something that looked sane from BDA but was not and then it would panic. In this change the EBDA is unconditionally setup before devices are initialized. I'm not set on the location in dev_initialize() but there does not seem to be another place to hook it in so that it runs just once for ALL platforms regardless of whether they use option roms or not. (possibly hardwaremain?) The EBDA setup code has been moved into its own location in arch/x86/lib/ebda.c so it can be compiled in even if the option rom code is not. The low memory size is still set to 1MB which is enough to make linux happy without having to hook into each mainboard to get a more appropriate value. The setup_ebda() function takes inputs so it could be changed for a mainboard if needed. OLD/BROKEN would read garbage. Examples from different boots: ebda_addr=0x75e80 lowmem=0x1553400 ebda_addr=0x5e080 lowmem=0x3e51400 ebda_addr=0x7aa80 lowmem=0x2f8a800 NEW/FIXED now reads consistent values: ebda_addr=0xf6000 lowmem=0x100000 Change-Id: I6cb79f0e3e43cc65f7e5fe98b6cad1a557ccd949 Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: http://review.coreboot.org/769 Tested-by: build bot (Jenkins) Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2012-01-17 18:03:11 +01:00
}
arch/x86/ebda.c: Move setting up ebda to a BS hook device.c should not hold arch specific code. Change-Id: I9dfdb905a83916c0e9d298e1c38da89f6bc5e038 Signed-off-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-on: https://review.coreboot.org/c/coreboot/+/64297 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Elyes Haouas <ehaouas@noos.fr> Reviewed-by: Werner Zeh <werner.zeh@siemens.com>
2022-05-12 20:39:16 +02:00
static void setup_default_ebda(void *unused)
Prepare the BIOS data areas before device init. Since we do not run option roms in normal mode nothing was initializing the BDA/EBDA and yet Linux depends very much on it having sane values here. For the most part the kernel tries to work around this not being initialized, but every once in awhile (1/300 boots or so) it would end up reading something that looked sane from BDA but was not and then it would panic. In this change the EBDA is unconditionally setup before devices are initialized. I'm not set on the location in dev_initialize() but there does not seem to be another place to hook it in so that it runs just once for ALL platforms regardless of whether they use option roms or not. (possibly hardwaremain?) The EBDA setup code has been moved into its own location in arch/x86/lib/ebda.c so it can be compiled in even if the option rom code is not. The low memory size is still set to 1MB which is enough to make linux happy without having to hook into each mainboard to get a more appropriate value. The setup_ebda() function takes inputs so it could be changed for a mainboard if needed. OLD/BROKEN would read garbage. Examples from different boots: ebda_addr=0x75e80 lowmem=0x1553400 ebda_addr=0x5e080 lowmem=0x3e51400 ebda_addr=0x7aa80 lowmem=0x2f8a800 NEW/FIXED now reads consistent values: ebda_addr=0xf6000 lowmem=0x100000 Change-Id: I6cb79f0e3e43cc65f7e5fe98b6cad1a557ccd949 Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: http://review.coreboot.org/769 Tested-by: build bot (Jenkins) Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2012-01-17 18:03:11 +01:00
{
arch/x86/ebda: Don't trash the EBDA on the resume path Clearing the EBDA was introduced with b4aaaa "Prepare the BIOS data areas before device init." which states that the purpose of setting up these area's is just to make sure they are sane. On the S3 path doing this is not needed and can even thrash data set up by payloads (mostly SeaBIOS) that used that memory. Change-Id: I9c54156bd8247e8a34dec6edc27cfc2d33cde595 Signed-off-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-on: https://review.coreboot.org/c/30789 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Patrick Rudolph <siro@das-labor.org> Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Nico Huber <nico.h@gmx.de>
2019-01-09 18:30:33 +01:00
if (acpi_is_wakeup_s3())
return;
Prepare the BIOS data areas before device init. Since we do not run option roms in normal mode nothing was initializing the BDA/EBDA and yet Linux depends very much on it having sane values here. For the most part the kernel tries to work around this not being initialized, but every once in awhile (1/300 boots or so) it would end up reading something that looked sane from BDA but was not and then it would panic. In this change the EBDA is unconditionally setup before devices are initialized. I'm not set on the location in dev_initialize() but there does not seem to be another place to hook it in so that it runs just once for ALL platforms regardless of whether they use option roms or not. (possibly hardwaremain?) The EBDA setup code has been moved into its own location in arch/x86/lib/ebda.c so it can be compiled in even if the option rom code is not. The low memory size is still set to 1MB which is enough to make linux happy without having to hook into each mainboard to get a more appropriate value. The setup_ebda() function takes inputs so it could be changed for a mainboard if needed. OLD/BROKEN would read garbage. Examples from different boots: ebda_addr=0x75e80 lowmem=0x1553400 ebda_addr=0x5e080 lowmem=0x3e51400 ebda_addr=0x7aa80 lowmem=0x2f8a800 NEW/FIXED now reads consistent values: ebda_addr=0xf6000 lowmem=0x100000 Change-Id: I6cb79f0e3e43cc65f7e5fe98b6cad1a557ccd949 Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: http://review.coreboot.org/769 Tested-by: build bot (Jenkins) Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2012-01-17 18:03:11 +01:00
setup_ebda(DEFAULT_EBDA_LOWMEM,
DEFAULT_EBDA_SEGMENT,
DEFAULT_EBDA_SIZE);
}
arch/x86/ebda.c: Move setting up ebda to a BS hook device.c should not hold arch specific code. Change-Id: I9dfdb905a83916c0e9d298e1c38da89f6bc5e038 Signed-off-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-on: https://review.coreboot.org/c/coreboot/+/64297 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Elyes Haouas <ehaouas@noos.fr> Reviewed-by: Werner Zeh <werner.zeh@siemens.com>
2022-05-12 20:39:16 +02:00
/* Ensure EBDA is prepared before Option ROMs. */
BOOT_STATE_INIT_ENTRY(BS_DEV_INIT, BS_ON_ENTRY, setup_default_ebda, NULL);
Reference in a new issue Copy permalink