coreboot-kgpe-d16/util/sconfig/main.c

2131 lines
58 KiB
C
Raw Normal View History

/* sconfig, coreboot device tree compiler */
util/: Replace GPLv2 boiler plate with SPDX header Used commands: perl -i -p0e 's|\/\*[\s*]*.*is free software[:;][\s*]*you[\s*]*can[\s*]*redistribute[\s*]*it[\s*]*and\/or[\s*]*modify[\s*]*it[\s*]*under[\s*]*the[\s*]*terms[\s*]*of[\s*]*the[\s*]*GNU[\s*]*General[\s*]*Public[\s*]*License[\s*]*as[\s*]*published[\s*]*by[\s*]*the[\s*]*Free[\s*]*Software[\s*]*Foundation[;,][\s*]*version[\s*]*2[\s*]*of[\s*]*the[\s*]*License.[\s*]*This[\s*]*program[\s*]*is[\s*]*distributed[\s*]*in[\s*]*the[\s*]*hope[\s*]*that[\s*]*it[\s*]*will[\s*]*be[\s*]*useful,[\s*]*but[\s*]*WITHOUT[\s*]*ANY[\s*]*WARRANTY;[\s*]*without[\s*]*even[\s*]*the[\s*]*implied[\s*]*warranty[\s*]*of[\s*]*MERCHANTABILITY[\s*]*or[\s*]*FITNESS[\s*]*FOR[\s*]*A[\s*]*PARTICULAR[\s*]*PURPOSE.[\s*]*See[\s*]*the[\s*]*GNU[\s*]*General[\s*]*Public[\s*]*License[\s*]*for[\s*]*more[\s*]*details.[\s*]*\*\/|/* SPDX-License-Identifier: GPL-2.0-only */|' $(cat filelist) perl -i -p0e 's|This[\s*]*program[\s*]*is[\s*]*free[\s*]*software[:;][\s*]*you[\s*]*can[\s*]*redistribute[\s*]*it[\s*]*and/or[\s*]*modify[\s*]*it[\s*]*under[\s*]*the[\s*]*terms[\s*]*of[\s*]*the[\s*]*GNU[\s*]*General[\s*]*Public[\s*]*License[\s*]*as[\s*]*published[\s*]*by[\s*]*the[\s*]*Free[\s*]*Software[\s*]*Foundation[;,][\s*]*either[\s*]*version[\s*]*2[\s*]*of[\s*]*the[\s*]*License,[\s*]*or[\s*]*.at[\s*]*your[\s*]*option.*[\s*]*any[\s*]*later[\s*]*version.[\s*]*This[\s*]*program[\s*]*is[\s*]*distributed[\s*]*in[\s*]*the[\s*]*hope[\s*]*that[\s*]*it[\s*]*will[\s*]*be[\s*]*useful,[\s*]*but[\s*]*WITHOUT[\s*]*ANY[\s*]*WARRANTY;[\s*]*without[\s*]*even[\s*]*the[\s*]*implied[\s*]*warranty[\s*]*of[\s*]*MERCHANTABILITY[\s*]*or[\s*]*FITNESS[\s*]*FOR[\s*]*A[\s*]*PARTICULAR[\s*]*PURPOSE.[\s*]*See[\s*]*the[\s*]*GNU[\s*]*General[\s*]*Public[\s*]*License[\s*]*for[\s*]*more[\s*]*details.[\s*]*\*\/|/* SPDX-License-Identifier: GPL-2.0-or-later */|' $(cat filelist) perl -i -p0e 's|\/\*[\s*]*.*This[\s*#]*program[\s*#]*is[\s*#]*free[\s*#]*software[;:,][\s*#]*you[\s*#]*can[\s*#]*redistribute[\s*#]*it[\s*#]*and/or[\s*#]*modify[\s*#]*it[\s*#]*under[\s*#]*the[\s*#]*terms[\s*#]*of[\s*#]*the[\s*#]*GNU[\s*#]*General[\s*#]*Public[\s*#]*License[\s*#]*as[\s*#]*published[\s*#]*by[\s*#]*the[\s*#]*Free[\s*#]*Software[\s*#]*Foundation[;:,][\s*#]*either[\s*#]*version[\s*#]*3[\s*#]*of[\s*#]*the[\s*#]*License[;:,][\s*#]*or[\s*#]*.at[\s*#]*your[\s*#]*option.*[\s*#]*any[\s*#]*later[\s*#]*version.[\s*#]*This[\s*#]*program[\s*#]*is[\s*#]*distributed[\s*#]*in[\s*#]*the[\s*#]*hope[\s*#]*that[\s*#]*it[\s*#]*will[\s*#]*be[\s*#]*useful[;:,][\s*#]*but[\s*#]*WITHOUT[\s*#]*ANY[\s*#]*WARRANTY[;:,][\s*#]*without[\s*#]*even[\s*#]*the[\s*#]*implied[\s*#]*warranty[\s*#]*of[\s*#]*MERCHANTABILITY[\s*#]*or[\s*#]*FITNESS[\s*#]*FOR[\s*#]*A[\s*#]*PARTICULAR[\s*#]*PURPOSE.[\s*#]*See[\s*#]*the[\s*#]*GNU[\s*#]*General[\s*#]*Public[\s*#]*License[\s*#]*for[\s*#]*more[\s*#]*details.[\s*]*\*\/|/* SPDX-License-Identifier: GPL-3.0-or-later */|' $(cat filelist) perl -i -p0e 's|(\#\#*)[\w]*.*is free software[:;][\#\s]*you[\#\s]*can[\#\s]*redistribute[\#\s]*it[\#\s]*and\/or[\#\s]*modify[\#\s]*it[\s\#]*under[\s \#]*the[\s\#]*terms[\s\#]*of[\s\#]*the[\s\#]*GNU[\s\#]*General[\s\#]*Public[\s\#]*License[\s\#]*as[\s\#]*published[\s\#]*by[\s\#]*the[\s\#]*Free[\s\#]*Software[\s\#]*Foundation[;,][\s\#]*version[\s\#]*2[\s\#]*of[\s\#]*the[\s\#]*License.*[\s\#]*This[\s\#]*program[\s\#]*is[\s\#]*distributed[\s\#]*in[\s\#]*the[\s\#]*hope[\s\#]*that[\s\#]*it[\s\#]*will[\#\s]*be[\#\s]*useful,[\#\s]*but[\#\s]*WITHOUT[\#\s]*ANY[\#\s]*WARRANTY;[\#\s]*without[\#\s]*even[\#\s]*the[\#\s]*implied[\#\s]*warranty[\#\s]*of[\#\s]*MERCHANTABILITY[\#\s]*or[\#\s]*FITNESS[\#\s]*FOR[\#\s]*A[\#\s]*PARTICULAR[\#\s]*PURPOSE.[\#\s]*See[\#\s]*the[\#\s]*GNU[\#\s]*General[\#\s]*Public[\#\s]*License[\#\s]*for[\#\s]*more[\#\s]*details.\s(#* *\n)*|\1 SPDX-License-Identifier: GPL-2.0-only\n\n|' $(cat filelist) perl -i -p0e 's|(\#\#*)[\w*]*.*is free software[:;][\s*]*you[\s*]*can[\s*]*redistribute[\s*]*it[\s*]*and\/or[\s*]*modify[\s*]*it[\s*]*under[\s*]*the[\s*]*terms[\s*]*of[\s*]*the[\s*]*GNU[\s*]*General[\s*]*Public[\s*]*License[\s*]*as[\s*]*published[\s*]*by[\s*]*the[\s*]*Free[\s*]*Software[\s*]*Foundation[;,][\s*]*version[\s*]*2[\s*]*of[\s*]*the[\s*]*License.[\s*]*This[\s*]*program[\s*]*is[\s*]*distributed[\s*]*in[\s*]*the[\s*]*hope[\s*]*that[\s*]*it[\s*]*will[\s*]*be[\s*]*useful,[\s*]*but[\s*]*WITHOUT[\s*]*ANY[\s*]*WARRANTY;[\s*]*without[\s*]*even[\s*]*the[\s*]*implied[\s*]*warranty[\s*]*of[\s*]*MERCHANTABILITY[\s*]*or[\s*]*FITNESS[\s*]*FOR[\s*]*A[\s*]*PARTICULAR[\s*]*PURPOSE.[\s*]*See[\s*]*the[\s*]*GNU[\s*]*General[\s*]*Public[\s*]*License[\s*]*for[\s*]*more[\s*]*details.\s(#* *\n)*|\1 SPDX-License-Identifier: GPL-2.0-only\n\n|' $(cat filelist) Change-Id: I1008a63b804f355a916221ac994701d7584f60ff Signed-off-by: Patrick Georgi <pgeorgi@google.com> Signed-off-by: Elyes HAOUAS <ehaouas@noos.fr> Reviewed-on: https://review.coreboot.org/c/coreboot/+/41177 Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2020-05-08 20:48:04 +02:00
/* SPDX-License-Identifier: GPL-2.0-only */
#include <assert.h>
#include <ctype.h>
#include <getopt.h>
#include <inttypes.h>
#include <libgen.h>
/* stat.h needs to be included before commonlib/helpers.h to avoid errors.*/
#include <sys/stat.h>
#include <commonlib/helpers.h>
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
#include <stdint.h>
#include "sconfig.h"
#include "sconfig.tab.h"
extern int linenum;
/*
* Maintains list of all the unique chip structures for the board.
* This is shared across base and override device trees since we need to
* generate headers for all chips added by both the trees.
*/
static struct chip chip_header;
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
typedef enum {
UNSLASH,
SPLIT_1ST,
TO_LOWER,
TO_UPPER,
} translate_t;
/*
* Mainboard is assumed to have a root device whose bus is the parent of all the
* devices that are added by parsing the devicetree file. This device has a
* mainboard chip instance associated with it.
*
*
*
* +------------------------+ +----------------------+
* | Root device | | Mainboard |
* +---------+ (base_root_dev) +--------------->+ instance +
* | | | chip_instance | (mainboard_instance)|
* | +------------------------+ | |
* | | +----------------------+
* | | bus |
* | parent v |
* | +-------------------+ |
* | | Root bus | |
* +----------->+ (base_root_bus) | |
* | | |
* +-------------------+ |
* | |
* | children | chip
* v |
* X |
* (new devices will |
* be added here as |
* children) |
* |
* |
* |
* +-------+----------+
* | |
* | Mainboard chip +----------->X (new chips will be
* | (mainboard_chip) | added here)
* | |
* +------------------+
*
*
*/
/* Root device of primary tree. */
static struct device base_root_dev;
/* Root device of chipset tree. */
static struct device chipset_root_dev;
/* Root device of override tree (if applicable). */
static struct device override_root_dev;
static struct chip_instance mainboard_instance;
static struct bus base_root_bus = {
.id = 0,
.dev = &base_root_dev,
};
static struct device base_root_dev = {
.name = "dev_root",
.chip_instance = &mainboard_instance,
.path = " .type = DEVICE_PATH_ROOT ",
.parent = &base_root_bus,
.enabled = 1,
.bus = &base_root_bus,
};
static struct bus chipset_root_bus = {
.id = 0,
.dev = &chipset_root_dev,
};
static struct device chipset_root_dev = {
.name = "chipset_root",
.chip_instance = &mainboard_instance,
.path = " .type = DEVICE_PATH_ROOT ",
.parent = &chipset_root_bus,
.enabled = 1,
.bus = &chipset_root_bus,
};
static struct bus override_root_bus = {
.id = 0,
.dev = &override_root_dev,
};
static struct device override_root_dev = {
.name = "override_root",
/*
* Override tree root device points to the same mainboard chip instance
* as the base tree root device. It should not cause any side-effects
* since the mainboard chip instance pointer in override tree will just
* be ignored.
*/
.chip_instance = &mainboard_instance,
.path = " .type = DEVICE_PATH_ROOT ",
.parent = &override_root_bus,
.enabled = 1,
.bus = &override_root_bus,
};
static struct chip mainboard_chip = {
.name = "mainboard",
.name_underscore = "mainboard",
.instance = &mainboard_instance,
};
static struct chip_instance mainboard_instance = {
.id = 0,
.chip = &mainboard_chip,
};
/* This is the parent of all devices added by parsing the devicetree file. */
struct bus *root_parent;
struct queue_entry {
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
void *data;
struct queue_entry *next;
struct queue_entry *prev;
};
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
sconfig: Allow to specify device operations Currently we only have runtime mechanisms to assign device operations to a node in our devicetree (with one exception: the root device). The most common method is to map PCI IDs to the device operations with a `struct pci_driver`. Another accustomed way is to let a chip driver assign them. For very common drivers, e.g. those in soc/intel/common/blocks/, the PCI ID lists grew very large and are incredibly error-prone. Often, IDs are missing and sometimes IDs are added almost mechanically without checking the code for compatibility. Maintaining these lists in a central place also reduces flexibility. Now, for onboard devices it is actually unnecessary to assign the device operations at runtime. We already know exactly what operations should be assigned. And since we are using chipset devicetrees, we have a perfect place to put that information. This patch adds a simple mechanism to `sconfig`. It allows us to speci- fy operations per device, e.g. device pci 00.0 alias system_agent on ops system_agent_ops end The operations are given as a C identifier. In this example, we simply assume that a global `struct device_operations system_agent_ops` exists. Change-Id: I2833d2f2450fde3206c33393f58b86fd4280b566 Signed-off-by: Nico Huber <nico.h@gmx.de> Reviewed-on: https://review.coreboot.org/c/coreboot/+/66483 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
2022-08-06 19:02:59 +02:00
/* Global list of all `struct device_operations` identifiers to declare. */
static struct identifier *device_operations;
#define S_ALLOC(_s) s_alloc(__func__, _s)
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
static void *s_alloc(const char *f, size_t s)
{
void *data = calloc(1, s);
if (!data) {
fprintf(stderr, "%s: Failed to alloc mem!\n", f);
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
exit(1);
}
return data;
}
static struct queue_entry *new_queue_entry(void *data)
{
struct queue_entry *e = S_ALLOC(sizeof(*e));
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
e->data = data;
e->next = e->prev = e;
return e;
}
static void enqueue_tail(struct queue_entry **q_head, void *data)
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
{
struct queue_entry *tmp = new_queue_entry(data);
struct queue_entry *q = *q_head;
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
if (!q) {
*q_head = tmp;
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
return;
}
q->prev->next = tmp;
tmp->prev = q->prev;
q->prev = tmp;
tmp->next = q;
}
static void *dequeue_tail(struct queue_entry **q_head)
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
{
struct queue_entry *q = *q_head;
struct queue_entry *tmp;
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
void *data;
if (!q)
return NULL;
tmp = q->prev;
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
if (tmp == q)
*q_head = NULL;
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
else {
tmp->prev->next = q;
q->prev = tmp->prev;
}
data = tmp->data;
free(tmp);
return data;
}
static void *dequeue_head(struct queue_entry **q_head)
{
struct queue_entry *q = *q_head;
struct queue_entry *tmp = q;
void *data;
if (!q)
return NULL;
if (q->next == q)
*q_head = NULL;
else {
q->next->prev = q->prev;
q->prev->next = q->next;
*q_head = q->next;
}
data = tmp->data;
free(tmp);
return data;
}
static void *peek_queue_head(struct queue_entry *q_head)
{
if (!q_head)
return NULL;
return q_head->data;
}
static struct queue_entry *chip_q_head;
void chip_enqueue_tail(void *data)
{
enqueue_tail(&chip_q_head, data);
}
void *chip_dequeue_tail(void)
{
return dequeue_tail(&chip_q_head);
}
int yywrap(void)
{
return 1;
}
void yyerror(char const *str)
{
extern char *yytext;
fprintf(stderr, "line %d: %s: %s\n", linenum + 1, yytext, str);
exit(1);
}
char *translate_name(const char *str, translate_t mode)
{
char *b, *c;
b = c = strdup(str);
while (c && *c) {
if ((mode == SPLIT_1ST) && (*c == '/')) {
*c = 0;
break;
}
if (*c == '/')
*c = '_';
if (*c == '-')
*c = '_';
if (mode == TO_UPPER)
*c = toupper(*c);
if (mode == TO_LOWER)
*c = tolower(*c);
c++;
}
return b;
}
static struct chip *get_chip(char *path)
{
struct chip *h = &chip_header;
while (h->next) {
int result = strcmp(path, h->next->name);
if (result == 0)
return h->next;
if (result < 0)
break;
h = h->next;
}
struct chip *new_chip = S_ALLOC(sizeof(struct chip));
new_chip->next = h->next;
h->next = new_chip;
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
new_chip->chiph_exists = 1;
new_chip->name = path;
new_chip->name_underscore = translate_name(path, UNSLASH);
struct stat st;
char *chip_h = S_ALLOC(strlen(path) + 18);
sprintf(chip_h, "src/%s", path);
if ((stat(chip_h, &st) == -1) && (errno == ENOENT)) {
/* root_complex gets away without a separate directory, but
* exists on pretty much all AMD chipsets.
*/
if (!strstr(path, "/root_complex")) {
fprintf(stderr, "ERROR: Chip component %s does not exist.\n",
path);
exit(1);
}
}
sprintf(chip_h, "src/%s/chip.h", path);
if ((stat(chip_h, &st) == -1) && (errno == ENOENT))
new_chip->chiph_exists = 0;
free(chip_h);
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
return new_chip;
}
struct chip_instance *new_chip_instance(char *path)
{
struct chip *chip = get_chip(path);
struct chip_instance *instance = S_ALLOC(sizeof(*instance));
instance->chip = chip;
instance->next = chip->instance;
chip->instance = instance;
return instance;
}
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
/* List of fw_config fields added during parsing. */
static struct fw_config_field *fw_config_fields;
static struct fw_config_option *find_fw_config_option(struct fw_config_field *field,
const char *name)
{
struct fw_config_option *option = field->options;
while (option && option->name) {
if (!strcmp(option->name, name))
return option;
option = option->next;
}
return NULL;
}
static struct fw_config_field *find_fw_config_field(const char *name)
{
struct fw_config_field *field = fw_config_fields;
while (field && field->name) {
if (!strcmp(field->name, name))
return field;
field = field->next;
}
return NULL;
}
struct fw_config_field *get_fw_config_field(const char *name)
{
struct fw_config_field *field = find_fw_config_field(name);
/* Fail if the field does not exist, new fields must be added with a mask. */
if (!field) {
printf("ERROR: fw_config field not found: %s\n", name);
exit(1);
}
return field;
}
static void append_fw_config_field(struct fw_config_field *add)
{
struct fw_config_field *field = fw_config_fields;
if (!fw_config_fields) {
fw_config_fields = add;
} else {
while (field && field->next)
field = field->next;
field->next = add;
}
}
util/sconfig: Add support for discontiguous FW_CONFIG fields Sooner or later, some board was going to need extra FW_CONFIG bits for a field that was already in production, so this patch adds support for adding extra (unused) bits to a field. The extra are appended via a syntax like: `field FIELD_NAME START0 END0 | START1 END1 | START2 END2 ...` and the suffixed bits are all treated as if they are contiguous when defining option values. BUG=b:185190978 TEST=Modified volteer fw_config to the following: field AUDIO 8 10 | 29 29 | 31 31 option NONE 0 option MAX98357_ALC5682I_I2S 1 option MAX98373_ALC5682I_I2S 2 option MAX98373_ALC5682_SNDW 3 option MAX98373_ALC5682I_I2S_UP4 4 option MAX98360_ALC5682I_I2S 5 option RT1011_ALC5682I_I2S 6 option AUDIO_FOO 7 option AUDIO_BAR 8 option AUDIO_QUUX 9 option AUDIO_BLAH1 10 option AUDIO_BLAH2 15 option AUDIO_BLAH3 16 option AUDIO_BLAH4 31 end which yielded (in static_fw_config.h): FW_CONFIG_FIELD_AUDIO_MASK 0xa0000700 FW_CONFIG_FIELD_AUDIO_OPTION_NONE_VALUE 0x0 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98357_ALC5682I_I2S_VALUE 0x100 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_VALUE 0x200 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682_SNDW_VALUE 0x300 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_UP4_VALUE 0x400 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98360_ALC5682I_I2S_VALUE 0x500 FW_CONFIG_FIELD_AUDIO_OPTION_RT1011_ALC5682I_I2S_VALUE 0x600 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_FOO_VALUE 0x700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BAR_VALUE 0x20000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_QUUX_VALUE 0x20000100 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH1_VALUE 0x20000200 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH2_VALUE 0x20000700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH3_VALUE 0x80000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH4_VALUE 0xa0000700 Change-Id: I5ed76706347ee9642198efc77139abdc3af1b8a6 Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52747 Reviewed-by: Duncan Laurie <duncan@iceblink.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28 22:03:07 +02:00
void append_fw_config_bits(struct fw_config_field_bits **bits,
unsigned int start_bit, unsigned int end_bit)
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
{
util/sconfig: Add support for discontiguous FW_CONFIG fields Sooner or later, some board was going to need extra FW_CONFIG bits for a field that was already in production, so this patch adds support for adding extra (unused) bits to a field. The extra are appended via a syntax like: `field FIELD_NAME START0 END0 | START1 END1 | START2 END2 ...` and the suffixed bits are all treated as if they are contiguous when defining option values. BUG=b:185190978 TEST=Modified volteer fw_config to the following: field AUDIO 8 10 | 29 29 | 31 31 option NONE 0 option MAX98357_ALC5682I_I2S 1 option MAX98373_ALC5682I_I2S 2 option MAX98373_ALC5682_SNDW 3 option MAX98373_ALC5682I_I2S_UP4 4 option MAX98360_ALC5682I_I2S 5 option RT1011_ALC5682I_I2S 6 option AUDIO_FOO 7 option AUDIO_BAR 8 option AUDIO_QUUX 9 option AUDIO_BLAH1 10 option AUDIO_BLAH2 15 option AUDIO_BLAH3 16 option AUDIO_BLAH4 31 end which yielded (in static_fw_config.h): FW_CONFIG_FIELD_AUDIO_MASK 0xa0000700 FW_CONFIG_FIELD_AUDIO_OPTION_NONE_VALUE 0x0 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98357_ALC5682I_I2S_VALUE 0x100 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_VALUE 0x200 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682_SNDW_VALUE 0x300 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_UP4_VALUE 0x400 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98360_ALC5682I_I2S_VALUE 0x500 FW_CONFIG_FIELD_AUDIO_OPTION_RT1011_ALC5682I_I2S_VALUE 0x600 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_FOO_VALUE 0x700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BAR_VALUE 0x20000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_QUUX_VALUE 0x20000100 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH1_VALUE 0x20000200 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH2_VALUE 0x20000700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH3_VALUE 0x80000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH4_VALUE 0xa0000700 Change-Id: I5ed76706347ee9642198efc77139abdc3af1b8a6 Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52747 Reviewed-by: Duncan Laurie <duncan@iceblink.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28 22:03:07 +02:00
struct fw_config_field_bits *new_bits = S_ALLOC(sizeof(*new_bits));
new_bits->start_bit = start_bit;
new_bits->end_bit = end_bit;
new_bits->next = NULL;
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
util/sconfig: Add support for discontiguous FW_CONFIG fields Sooner or later, some board was going to need extra FW_CONFIG bits for a field that was already in production, so this patch adds support for adding extra (unused) bits to a field. The extra are appended via a syntax like: `field FIELD_NAME START0 END0 | START1 END1 | START2 END2 ...` and the suffixed bits are all treated as if they are contiguous when defining option values. BUG=b:185190978 TEST=Modified volteer fw_config to the following: field AUDIO 8 10 | 29 29 | 31 31 option NONE 0 option MAX98357_ALC5682I_I2S 1 option MAX98373_ALC5682I_I2S 2 option MAX98373_ALC5682_SNDW 3 option MAX98373_ALC5682I_I2S_UP4 4 option MAX98360_ALC5682I_I2S 5 option RT1011_ALC5682I_I2S 6 option AUDIO_FOO 7 option AUDIO_BAR 8 option AUDIO_QUUX 9 option AUDIO_BLAH1 10 option AUDIO_BLAH2 15 option AUDIO_BLAH3 16 option AUDIO_BLAH4 31 end which yielded (in static_fw_config.h): FW_CONFIG_FIELD_AUDIO_MASK 0xa0000700 FW_CONFIG_FIELD_AUDIO_OPTION_NONE_VALUE 0x0 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98357_ALC5682I_I2S_VALUE 0x100 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_VALUE 0x200 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682_SNDW_VALUE 0x300 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_UP4_VALUE 0x400 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98360_ALC5682I_I2S_VALUE 0x500 FW_CONFIG_FIELD_AUDIO_OPTION_RT1011_ALC5682I_I2S_VALUE 0x600 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_FOO_VALUE 0x700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BAR_VALUE 0x20000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_QUUX_VALUE 0x20000100 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH1_VALUE 0x20000200 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH2_VALUE 0x20000700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH3_VALUE 0x80000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH4_VALUE 0xa0000700 Change-Id: I5ed76706347ee9642198efc77139abdc3af1b8a6 Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52747 Reviewed-by: Duncan Laurie <duncan@iceblink.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28 22:03:07 +02:00
if (*bits == NULL) {
*bits = new_bits;
return;
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
}
util/sconfig: Add support for discontiguous FW_CONFIG fields Sooner or later, some board was going to need extra FW_CONFIG bits for a field that was already in production, so this patch adds support for adding extra (unused) bits to a field. The extra are appended via a syntax like: `field FIELD_NAME START0 END0 | START1 END1 | START2 END2 ...` and the suffixed bits are all treated as if they are contiguous when defining option values. BUG=b:185190978 TEST=Modified volteer fw_config to the following: field AUDIO 8 10 | 29 29 | 31 31 option NONE 0 option MAX98357_ALC5682I_I2S 1 option MAX98373_ALC5682I_I2S 2 option MAX98373_ALC5682_SNDW 3 option MAX98373_ALC5682I_I2S_UP4 4 option MAX98360_ALC5682I_I2S 5 option RT1011_ALC5682I_I2S 6 option AUDIO_FOO 7 option AUDIO_BAR 8 option AUDIO_QUUX 9 option AUDIO_BLAH1 10 option AUDIO_BLAH2 15 option AUDIO_BLAH3 16 option AUDIO_BLAH4 31 end which yielded (in static_fw_config.h): FW_CONFIG_FIELD_AUDIO_MASK 0xa0000700 FW_CONFIG_FIELD_AUDIO_OPTION_NONE_VALUE 0x0 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98357_ALC5682I_I2S_VALUE 0x100 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_VALUE 0x200 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682_SNDW_VALUE 0x300 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_UP4_VALUE 0x400 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98360_ALC5682I_I2S_VALUE 0x500 FW_CONFIG_FIELD_AUDIO_OPTION_RT1011_ALC5682I_I2S_VALUE 0x600 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_FOO_VALUE 0x700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BAR_VALUE 0x20000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_QUUX_VALUE 0x20000100 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH1_VALUE 0x20000200 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH2_VALUE 0x20000700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH3_VALUE 0x80000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH4_VALUE 0xa0000700 Change-Id: I5ed76706347ee9642198efc77139abdc3af1b8a6 Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52747 Reviewed-by: Duncan Laurie <duncan@iceblink.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28 22:03:07 +02:00
struct fw_config_field_bits *tmp = *bits;
while (tmp->next)
tmp = tmp->next;
tmp->next = new_bits;
}
int fw_config_masks_overlap(struct fw_config_field *existing,
unsigned int start_bit, unsigned int end_bit)
{
struct fw_config_field_bits *bits = existing->bits;
while (bits) {
if (start_bit <= bits->end_bit && end_bit >= bits->start_bit) {
printf("ERROR: fw_config field [%u-%u] overlaps %s[%u-%u]\n",
start_bit, end_bit,
existing->name, bits->start_bit, bits->end_bit);
return 1;
}
bits = bits->next;
}
return 0;
}
struct fw_config_field *new_fw_config_field(const char *name, struct fw_config_field_bits *bits)
{
struct fw_config_field *field = find_fw_config_field(name);
struct fw_config_field_bits *tmp;
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
/* Don't allow re-defining a field, only adding new fields. */
if (field) {
util/sconfig: Add support for discontiguous FW_CONFIG fields Sooner or later, some board was going to need extra FW_CONFIG bits for a field that was already in production, so this patch adds support for adding extra (unused) bits to a field. The extra are appended via a syntax like: `field FIELD_NAME START0 END0 | START1 END1 | START2 END2 ...` and the suffixed bits are all treated as if they are contiguous when defining option values. BUG=b:185190978 TEST=Modified volteer fw_config to the following: field AUDIO 8 10 | 29 29 | 31 31 option NONE 0 option MAX98357_ALC5682I_I2S 1 option MAX98373_ALC5682I_I2S 2 option MAX98373_ALC5682_SNDW 3 option MAX98373_ALC5682I_I2S_UP4 4 option MAX98360_ALC5682I_I2S 5 option RT1011_ALC5682I_I2S 6 option AUDIO_FOO 7 option AUDIO_BAR 8 option AUDIO_QUUX 9 option AUDIO_BLAH1 10 option AUDIO_BLAH2 15 option AUDIO_BLAH3 16 option AUDIO_BLAH4 31 end which yielded (in static_fw_config.h): FW_CONFIG_FIELD_AUDIO_MASK 0xa0000700 FW_CONFIG_FIELD_AUDIO_OPTION_NONE_VALUE 0x0 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98357_ALC5682I_I2S_VALUE 0x100 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_VALUE 0x200 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682_SNDW_VALUE 0x300 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_UP4_VALUE 0x400 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98360_ALC5682I_I2S_VALUE 0x500 FW_CONFIG_FIELD_AUDIO_OPTION_RT1011_ALC5682I_I2S_VALUE 0x600 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_FOO_VALUE 0x700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BAR_VALUE 0x20000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_QUUX_VALUE 0x20000100 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH1_VALUE 0x20000200 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH2_VALUE 0x20000700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH3_VALUE 0x80000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH4_VALUE 0xa0000700 Change-Id: I5ed76706347ee9642198efc77139abdc3af1b8a6 Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52747 Reviewed-by: Duncan Laurie <duncan@iceblink.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28 22:03:07 +02:00
printf("ERROR: fw_config field %s already exists\n", name);
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
exit(1);
}
util/sconfig: Add support for discontiguous FW_CONFIG fields Sooner or later, some board was going to need extra FW_CONFIG bits for a field that was already in production, so this patch adds support for adding extra (unused) bits to a field. The extra are appended via a syntax like: `field FIELD_NAME START0 END0 | START1 END1 | START2 END2 ...` and the suffixed bits are all treated as if they are contiguous when defining option values. BUG=b:185190978 TEST=Modified volteer fw_config to the following: field AUDIO 8 10 | 29 29 | 31 31 option NONE 0 option MAX98357_ALC5682I_I2S 1 option MAX98373_ALC5682I_I2S 2 option MAX98373_ALC5682_SNDW 3 option MAX98373_ALC5682I_I2S_UP4 4 option MAX98360_ALC5682I_I2S 5 option RT1011_ALC5682I_I2S 6 option AUDIO_FOO 7 option AUDIO_BAR 8 option AUDIO_QUUX 9 option AUDIO_BLAH1 10 option AUDIO_BLAH2 15 option AUDIO_BLAH3 16 option AUDIO_BLAH4 31 end which yielded (in static_fw_config.h): FW_CONFIG_FIELD_AUDIO_MASK 0xa0000700 FW_CONFIG_FIELD_AUDIO_OPTION_NONE_VALUE 0x0 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98357_ALC5682I_I2S_VALUE 0x100 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_VALUE 0x200 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682_SNDW_VALUE 0x300 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_UP4_VALUE 0x400 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98360_ALC5682I_I2S_VALUE 0x500 FW_CONFIG_FIELD_AUDIO_OPTION_RT1011_ALC5682I_I2S_VALUE 0x600 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_FOO_VALUE 0x700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BAR_VALUE 0x20000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_QUUX_VALUE 0x20000100 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH1_VALUE 0x20000200 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH2_VALUE 0x20000700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH3_VALUE 0x80000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH4_VALUE 0xa0000700 Change-Id: I5ed76706347ee9642198efc77139abdc3af1b8a6 Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52747 Reviewed-by: Duncan Laurie <duncan@iceblink.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28 22:03:07 +02:00
/* Check that each field is within 64 bits. */
tmp = bits;
while (tmp) {
if (tmp->start_bit > tmp->end_bit || tmp->end_bit > 63) {
printf("ERROR: fw_config field %s has invalid range %u-%u\n", name,
util/sconfig: Add support for discontiguous FW_CONFIG fields Sooner or later, some board was going to need extra FW_CONFIG bits for a field that was already in production, so this patch adds support for adding extra (unused) bits to a field. The extra are appended via a syntax like: `field FIELD_NAME START0 END0 | START1 END1 | START2 END2 ...` and the suffixed bits are all treated as if they are contiguous when defining option values. BUG=b:185190978 TEST=Modified volteer fw_config to the following: field AUDIO 8 10 | 29 29 | 31 31 option NONE 0 option MAX98357_ALC5682I_I2S 1 option MAX98373_ALC5682I_I2S 2 option MAX98373_ALC5682_SNDW 3 option MAX98373_ALC5682I_I2S_UP4 4 option MAX98360_ALC5682I_I2S 5 option RT1011_ALC5682I_I2S 6 option AUDIO_FOO 7 option AUDIO_BAR 8 option AUDIO_QUUX 9 option AUDIO_BLAH1 10 option AUDIO_BLAH2 15 option AUDIO_BLAH3 16 option AUDIO_BLAH4 31 end which yielded (in static_fw_config.h): FW_CONFIG_FIELD_AUDIO_MASK 0xa0000700 FW_CONFIG_FIELD_AUDIO_OPTION_NONE_VALUE 0x0 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98357_ALC5682I_I2S_VALUE 0x100 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_VALUE 0x200 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682_SNDW_VALUE 0x300 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_UP4_VALUE 0x400 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98360_ALC5682I_I2S_VALUE 0x500 FW_CONFIG_FIELD_AUDIO_OPTION_RT1011_ALC5682I_I2S_VALUE 0x600 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_FOO_VALUE 0x700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BAR_VALUE 0x20000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_QUUX_VALUE 0x20000100 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH1_VALUE 0x20000200 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH2_VALUE 0x20000700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH3_VALUE 0x80000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH4_VALUE 0xa0000700 Change-Id: I5ed76706347ee9642198efc77139abdc3af1b8a6 Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52747 Reviewed-by: Duncan Laurie <duncan@iceblink.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28 22:03:07 +02:00
tmp->start_bit, tmp->end_bit);
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
exit(1);
}
util/sconfig: Add support for discontiguous FW_CONFIG fields Sooner or later, some board was going to need extra FW_CONFIG bits for a field that was already in production, so this patch adds support for adding extra (unused) bits to a field. The extra are appended via a syntax like: `field FIELD_NAME START0 END0 | START1 END1 | START2 END2 ...` and the suffixed bits are all treated as if they are contiguous when defining option values. BUG=b:185190978 TEST=Modified volteer fw_config to the following: field AUDIO 8 10 | 29 29 | 31 31 option NONE 0 option MAX98357_ALC5682I_I2S 1 option MAX98373_ALC5682I_I2S 2 option MAX98373_ALC5682_SNDW 3 option MAX98373_ALC5682I_I2S_UP4 4 option MAX98360_ALC5682I_I2S 5 option RT1011_ALC5682I_I2S 6 option AUDIO_FOO 7 option AUDIO_BAR 8 option AUDIO_QUUX 9 option AUDIO_BLAH1 10 option AUDIO_BLAH2 15 option AUDIO_BLAH3 16 option AUDIO_BLAH4 31 end which yielded (in static_fw_config.h): FW_CONFIG_FIELD_AUDIO_MASK 0xa0000700 FW_CONFIG_FIELD_AUDIO_OPTION_NONE_VALUE 0x0 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98357_ALC5682I_I2S_VALUE 0x100 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_VALUE 0x200 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682_SNDW_VALUE 0x300 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_UP4_VALUE 0x400 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98360_ALC5682I_I2S_VALUE 0x500 FW_CONFIG_FIELD_AUDIO_OPTION_RT1011_ALC5682I_I2S_VALUE 0x600 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_FOO_VALUE 0x700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BAR_VALUE 0x20000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_QUUX_VALUE 0x20000100 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH1_VALUE 0x20000200 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH2_VALUE 0x20000700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH3_VALUE 0x80000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH4_VALUE 0xa0000700 Change-Id: I5ed76706347ee9642198efc77139abdc3af1b8a6 Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52747 Reviewed-by: Duncan Laurie <duncan@iceblink.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28 22:03:07 +02:00
/* Check for overlap with an existing field. */
struct fw_config_field *existing = fw_config_fields;
while (existing) {
if (fw_config_masks_overlap(existing, tmp->start_bit, tmp->end_bit))
exit(1);
existing = existing->next;
}
tmp = tmp->next;
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
}
field = S_ALLOC(sizeof(*field));
field->name = name;
util/sconfig: Add support for discontiguous FW_CONFIG fields Sooner or later, some board was going to need extra FW_CONFIG bits for a field that was already in production, so this patch adds support for adding extra (unused) bits to a field. The extra are appended via a syntax like: `field FIELD_NAME START0 END0 | START1 END1 | START2 END2 ...` and the suffixed bits are all treated as if they are contiguous when defining option values. BUG=b:185190978 TEST=Modified volteer fw_config to the following: field AUDIO 8 10 | 29 29 | 31 31 option NONE 0 option MAX98357_ALC5682I_I2S 1 option MAX98373_ALC5682I_I2S 2 option MAX98373_ALC5682_SNDW 3 option MAX98373_ALC5682I_I2S_UP4 4 option MAX98360_ALC5682I_I2S 5 option RT1011_ALC5682I_I2S 6 option AUDIO_FOO 7 option AUDIO_BAR 8 option AUDIO_QUUX 9 option AUDIO_BLAH1 10 option AUDIO_BLAH2 15 option AUDIO_BLAH3 16 option AUDIO_BLAH4 31 end which yielded (in static_fw_config.h): FW_CONFIG_FIELD_AUDIO_MASK 0xa0000700 FW_CONFIG_FIELD_AUDIO_OPTION_NONE_VALUE 0x0 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98357_ALC5682I_I2S_VALUE 0x100 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_VALUE 0x200 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682_SNDW_VALUE 0x300 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_UP4_VALUE 0x400 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98360_ALC5682I_I2S_VALUE 0x500 FW_CONFIG_FIELD_AUDIO_OPTION_RT1011_ALC5682I_I2S_VALUE 0x600 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_FOO_VALUE 0x700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BAR_VALUE 0x20000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_QUUX_VALUE 0x20000100 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH1_VALUE 0x20000200 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH2_VALUE 0x20000700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH3_VALUE 0x80000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH4_VALUE 0xa0000700 Change-Id: I5ed76706347ee9642198efc77139abdc3af1b8a6 Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52747 Reviewed-by: Duncan Laurie <duncan@iceblink.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28 22:03:07 +02:00
field->bits = bits;
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
append_fw_config_field(field);
return field;
}
static void append_fw_config_option_to_field(struct fw_config_field *field,
struct fw_config_option *add)
{
struct fw_config_option *option = field->options;
if (!option) {
field->options = add;
} else {
while (option && option->next)
option = option->next;
option->next = add;
}
}
util/sconfig: Add support for discontiguous FW_CONFIG fields Sooner or later, some board was going to need extra FW_CONFIG bits for a field that was already in production, so this patch adds support for adding extra (unused) bits to a field. The extra are appended via a syntax like: `field FIELD_NAME START0 END0 | START1 END1 | START2 END2 ...` and the suffixed bits are all treated as if they are contiguous when defining option values. BUG=b:185190978 TEST=Modified volteer fw_config to the following: field AUDIO 8 10 | 29 29 | 31 31 option NONE 0 option MAX98357_ALC5682I_I2S 1 option MAX98373_ALC5682I_I2S 2 option MAX98373_ALC5682_SNDW 3 option MAX98373_ALC5682I_I2S_UP4 4 option MAX98360_ALC5682I_I2S 5 option RT1011_ALC5682I_I2S 6 option AUDIO_FOO 7 option AUDIO_BAR 8 option AUDIO_QUUX 9 option AUDIO_BLAH1 10 option AUDIO_BLAH2 15 option AUDIO_BLAH3 16 option AUDIO_BLAH4 31 end which yielded (in static_fw_config.h): FW_CONFIG_FIELD_AUDIO_MASK 0xa0000700 FW_CONFIG_FIELD_AUDIO_OPTION_NONE_VALUE 0x0 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98357_ALC5682I_I2S_VALUE 0x100 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_VALUE 0x200 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682_SNDW_VALUE 0x300 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_UP4_VALUE 0x400 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98360_ALC5682I_I2S_VALUE 0x500 FW_CONFIG_FIELD_AUDIO_OPTION_RT1011_ALC5682I_I2S_VALUE 0x600 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_FOO_VALUE 0x700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BAR_VALUE 0x20000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_QUUX_VALUE 0x20000100 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH1_VALUE 0x20000200 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH2_VALUE 0x20000700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH3_VALUE 0x80000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH4_VALUE 0xa0000700 Change-Id: I5ed76706347ee9642198efc77139abdc3af1b8a6 Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52747 Reviewed-by: Duncan Laurie <duncan@iceblink.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28 22:03:07 +02:00
static uint64_t calc_max_field_value(const struct fw_config_field *field)
{
unsigned int bit_count = 0;
const struct fw_config_field_bits *bits = field->bits;
while (bits) {
bit_count += 1 + bits->end_bit - bits->start_bit;
bits = bits->next;
};
return (1ull << bit_count) - 1ull;
}
void add_fw_config_option(struct fw_config_field *field, const char *name, uint64_t value)
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
{
struct fw_config_option *option;
/* Check that option value fits within field mask. */
util/sconfig: Add support for discontiguous FW_CONFIG fields Sooner or later, some board was going to need extra FW_CONFIG bits for a field that was already in production, so this patch adds support for adding extra (unused) bits to a field. The extra are appended via a syntax like: `field FIELD_NAME START0 END0 | START1 END1 | START2 END2 ...` and the suffixed bits are all treated as if they are contiguous when defining option values. BUG=b:185190978 TEST=Modified volteer fw_config to the following: field AUDIO 8 10 | 29 29 | 31 31 option NONE 0 option MAX98357_ALC5682I_I2S 1 option MAX98373_ALC5682I_I2S 2 option MAX98373_ALC5682_SNDW 3 option MAX98373_ALC5682I_I2S_UP4 4 option MAX98360_ALC5682I_I2S 5 option RT1011_ALC5682I_I2S 6 option AUDIO_FOO 7 option AUDIO_BAR 8 option AUDIO_QUUX 9 option AUDIO_BLAH1 10 option AUDIO_BLAH2 15 option AUDIO_BLAH3 16 option AUDIO_BLAH4 31 end which yielded (in static_fw_config.h): FW_CONFIG_FIELD_AUDIO_MASK 0xa0000700 FW_CONFIG_FIELD_AUDIO_OPTION_NONE_VALUE 0x0 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98357_ALC5682I_I2S_VALUE 0x100 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_VALUE 0x200 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682_SNDW_VALUE 0x300 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_UP4_VALUE 0x400 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98360_ALC5682I_I2S_VALUE 0x500 FW_CONFIG_FIELD_AUDIO_OPTION_RT1011_ALC5682I_I2S_VALUE 0x600 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_FOO_VALUE 0x700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BAR_VALUE 0x20000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_QUUX_VALUE 0x20000100 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH1_VALUE 0x20000200 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH2_VALUE 0x20000700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH3_VALUE 0x80000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH4_VALUE 0xa0000700 Change-Id: I5ed76706347ee9642198efc77139abdc3af1b8a6 Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52747 Reviewed-by: Duncan Laurie <duncan@iceblink.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28 22:03:07 +02:00
uint64_t field_max_value = calc_max_field_value(field);
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
if (value > field_max_value) {
printf("ERROR: fw_config option %s:%s value %" PRIx64 " larger than field max %"
PRIx64 "\n",
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
field->name, name, value, field_max_value);
exit(1);
}
/* Check for existing option with this name or value. */
option = field->options;
while (option) {
if (!strcmp(option->name, name)) {
printf("ERROR: fw_config option name %s:%s already exists\n",
field->name, name);
exit(1);
}
/* Compare values. */
if (value == option->value) {
printf("ERROR: fw_config option %s:%s[%" PRIx64 "] redefined as %s\n",
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
field->name, option->name, value, name);
exit(1);
}
option = option->next;
}
option = S_ALLOC(sizeof(*option));
option->name = name;
option->value = value;
/* Add option to the current field. */
append_fw_config_option_to_field(field, option);
}
static void append_fw_config_probe_to_dev(struct device *dev, struct fw_config_probe *add)
{
struct fw_config_probe *probe = dev->probe;
if (!probe) {
dev->probe = add;
} else {
while (probe && probe->next)
probe = probe->next;
probe->next = add;
}
}
static int check_probe_exists(struct fw_config_probe *probe, const char *field,
const char *option)
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
{
while (probe) {
if (!strcmp(probe->field, field) && !strcmp(probe->option, option)) {
return 1;
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
}
probe = probe->next;
}
return 0;
}
void add_fw_config_probe(struct bus *bus, const char *field, const char *option)
{
struct fw_config_probe *probe;
if (check_probe_exists(bus->dev->probe, field, option)) {
printf("ERROR: fw_config probe %s:%s already exists\n", field, option);
exit(1);
}
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
probe = S_ALLOC(sizeof(*probe));
probe->field = field;
probe->option = option;
append_fw_config_probe_to_dev(bus->dev, probe);
}
util/sconfig: Add support for discontiguous FW_CONFIG fields Sooner or later, some board was going to need extra FW_CONFIG bits for a field that was already in production, so this patch adds support for adding extra (unused) bits to a field. The extra are appended via a syntax like: `field FIELD_NAME START0 END0 | START1 END1 | START2 END2 ...` and the suffixed bits are all treated as if they are contiguous when defining option values. BUG=b:185190978 TEST=Modified volteer fw_config to the following: field AUDIO 8 10 | 29 29 | 31 31 option NONE 0 option MAX98357_ALC5682I_I2S 1 option MAX98373_ALC5682I_I2S 2 option MAX98373_ALC5682_SNDW 3 option MAX98373_ALC5682I_I2S_UP4 4 option MAX98360_ALC5682I_I2S 5 option RT1011_ALC5682I_I2S 6 option AUDIO_FOO 7 option AUDIO_BAR 8 option AUDIO_QUUX 9 option AUDIO_BLAH1 10 option AUDIO_BLAH2 15 option AUDIO_BLAH3 16 option AUDIO_BLAH4 31 end which yielded (in static_fw_config.h): FW_CONFIG_FIELD_AUDIO_MASK 0xa0000700 FW_CONFIG_FIELD_AUDIO_OPTION_NONE_VALUE 0x0 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98357_ALC5682I_I2S_VALUE 0x100 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_VALUE 0x200 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682_SNDW_VALUE 0x300 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_UP4_VALUE 0x400 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98360_ALC5682I_I2S_VALUE 0x500 FW_CONFIG_FIELD_AUDIO_OPTION_RT1011_ALC5682I_I2S_VALUE 0x600 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_FOO_VALUE 0x700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BAR_VALUE 0x20000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_QUUX_VALUE 0x20000100 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH1_VALUE 0x20000200 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH2_VALUE 0x20000700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH3_VALUE 0x80000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH4_VALUE 0xa0000700 Change-Id: I5ed76706347ee9642198efc77139abdc3af1b8a6 Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52747 Reviewed-by: Duncan Laurie <duncan@iceblink.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28 22:03:07 +02:00
static uint64_t compute_fw_config_mask(const struct fw_config_field_bits *bits)
{
uint64_t mask = 0;
while (bits) {
/* Compute mask from start and end bit. */
uint64_t tmp = ((1ull << (1ull + bits->end_bit - bits->start_bit)) - 1ull);
tmp <<= bits->start_bit;
mask |= tmp;
bits = bits->next;
}
return mask;
}
static unsigned int bits_width(const struct fw_config_field_bits *bits)
{
return 1 + bits->end_bit - bits->start_bit;
}
static uint64_t calc_option_value(const struct fw_config_field *field,
const struct fw_config_option *option)
{
uint64_t value = 0;
uint64_t original = option->value;
struct fw_config_field_bits *bits = field->bits;
while (bits) {
const unsigned int width = bits_width(bits);
const uint64_t orig_mask = (1ull << width) - 1ull;
const uint64_t orig = (original & orig_mask);
value |= (orig << bits->start_bit);
original >>= width;
bits = bits->next;
}
return value;
}
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
static void emit_fw_config(FILE *fil)
{
struct fw_config_field *field = fw_config_fields;
if (!field)
return;
while (field) {
struct fw_config_option *option = field->options;
uint64_t mask;
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
fprintf(fil, "#define FW_CONFIG_FIELD_%s_NAME \"%s\"\n",
field->name, field->name);
util/sconfig: Add support for discontiguous FW_CONFIG fields Sooner or later, some board was going to need extra FW_CONFIG bits for a field that was already in production, so this patch adds support for adding extra (unused) bits to a field. The extra are appended via a syntax like: `field FIELD_NAME START0 END0 | START1 END1 | START2 END2 ...` and the suffixed bits are all treated as if they are contiguous when defining option values. BUG=b:185190978 TEST=Modified volteer fw_config to the following: field AUDIO 8 10 | 29 29 | 31 31 option NONE 0 option MAX98357_ALC5682I_I2S 1 option MAX98373_ALC5682I_I2S 2 option MAX98373_ALC5682_SNDW 3 option MAX98373_ALC5682I_I2S_UP4 4 option MAX98360_ALC5682I_I2S 5 option RT1011_ALC5682I_I2S 6 option AUDIO_FOO 7 option AUDIO_BAR 8 option AUDIO_QUUX 9 option AUDIO_BLAH1 10 option AUDIO_BLAH2 15 option AUDIO_BLAH3 16 option AUDIO_BLAH4 31 end which yielded (in static_fw_config.h): FW_CONFIG_FIELD_AUDIO_MASK 0xa0000700 FW_CONFIG_FIELD_AUDIO_OPTION_NONE_VALUE 0x0 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98357_ALC5682I_I2S_VALUE 0x100 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_VALUE 0x200 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682_SNDW_VALUE 0x300 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_UP4_VALUE 0x400 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98360_ALC5682I_I2S_VALUE 0x500 FW_CONFIG_FIELD_AUDIO_OPTION_RT1011_ALC5682I_I2S_VALUE 0x600 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_FOO_VALUE 0x700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BAR_VALUE 0x20000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_QUUX_VALUE 0x20000100 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH1_VALUE 0x20000200 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH2_VALUE 0x20000700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH3_VALUE 0x80000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH4_VALUE 0xa0000700 Change-Id: I5ed76706347ee9642198efc77139abdc3af1b8a6 Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52747 Reviewed-by: Duncan Laurie <duncan@iceblink.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28 22:03:07 +02:00
mask = compute_fw_config_mask(field->bits);
fprintf(fil, "#define FW_CONFIG_FIELD_%s_MASK 0x%" PRIx64 "\n",
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
field->name, mask);
while (option) {
util/sconfig: Add support for discontiguous FW_CONFIG fields Sooner or later, some board was going to need extra FW_CONFIG bits for a field that was already in production, so this patch adds support for adding extra (unused) bits to a field. The extra are appended via a syntax like: `field FIELD_NAME START0 END0 | START1 END1 | START2 END2 ...` and the suffixed bits are all treated as if they are contiguous when defining option values. BUG=b:185190978 TEST=Modified volteer fw_config to the following: field AUDIO 8 10 | 29 29 | 31 31 option NONE 0 option MAX98357_ALC5682I_I2S 1 option MAX98373_ALC5682I_I2S 2 option MAX98373_ALC5682_SNDW 3 option MAX98373_ALC5682I_I2S_UP4 4 option MAX98360_ALC5682I_I2S 5 option RT1011_ALC5682I_I2S 6 option AUDIO_FOO 7 option AUDIO_BAR 8 option AUDIO_QUUX 9 option AUDIO_BLAH1 10 option AUDIO_BLAH2 15 option AUDIO_BLAH3 16 option AUDIO_BLAH4 31 end which yielded (in static_fw_config.h): FW_CONFIG_FIELD_AUDIO_MASK 0xa0000700 FW_CONFIG_FIELD_AUDIO_OPTION_NONE_VALUE 0x0 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98357_ALC5682I_I2S_VALUE 0x100 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_VALUE 0x200 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682_SNDW_VALUE 0x300 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_UP4_VALUE 0x400 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98360_ALC5682I_I2S_VALUE 0x500 FW_CONFIG_FIELD_AUDIO_OPTION_RT1011_ALC5682I_I2S_VALUE 0x600 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_FOO_VALUE 0x700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BAR_VALUE 0x20000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_QUUX_VALUE 0x20000100 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH1_VALUE 0x20000200 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH2_VALUE 0x20000700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH3_VALUE 0x80000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH4_VALUE 0xa0000700 Change-Id: I5ed76706347ee9642198efc77139abdc3af1b8a6 Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52747 Reviewed-by: Duncan Laurie <duncan@iceblink.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28 22:03:07 +02:00
const uint64_t value = calc_option_value(field, option);
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
fprintf(fil, "#define FW_CONFIG_FIELD_%s_OPTION_%s_NAME \"%s\"\n",
field->name, option->name, option->name);
fprintf(fil, "#define FW_CONFIG_FIELD_%s_OPTION_%s_VALUE 0x%"
util/sconfig: Add support for discontiguous FW_CONFIG fields Sooner or later, some board was going to need extra FW_CONFIG bits for a field that was already in production, so this patch adds support for adding extra (unused) bits to a field. The extra are appended via a syntax like: `field FIELD_NAME START0 END0 | START1 END1 | START2 END2 ...` and the suffixed bits are all treated as if they are contiguous when defining option values. BUG=b:185190978 TEST=Modified volteer fw_config to the following: field AUDIO 8 10 | 29 29 | 31 31 option NONE 0 option MAX98357_ALC5682I_I2S 1 option MAX98373_ALC5682I_I2S 2 option MAX98373_ALC5682_SNDW 3 option MAX98373_ALC5682I_I2S_UP4 4 option MAX98360_ALC5682I_I2S 5 option RT1011_ALC5682I_I2S 6 option AUDIO_FOO 7 option AUDIO_BAR 8 option AUDIO_QUUX 9 option AUDIO_BLAH1 10 option AUDIO_BLAH2 15 option AUDIO_BLAH3 16 option AUDIO_BLAH4 31 end which yielded (in static_fw_config.h): FW_CONFIG_FIELD_AUDIO_MASK 0xa0000700 FW_CONFIG_FIELD_AUDIO_OPTION_NONE_VALUE 0x0 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98357_ALC5682I_I2S_VALUE 0x100 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_VALUE 0x200 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682_SNDW_VALUE 0x300 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_UP4_VALUE 0x400 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98360_ALC5682I_I2S_VALUE 0x500 FW_CONFIG_FIELD_AUDIO_OPTION_RT1011_ALC5682I_I2S_VALUE 0x600 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_FOO_VALUE 0x700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BAR_VALUE 0x20000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_QUUX_VALUE 0x20000100 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH1_VALUE 0x20000200 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH2_VALUE 0x20000700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH3_VALUE 0x80000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH4_VALUE 0xa0000700 Change-Id: I5ed76706347ee9642198efc77139abdc3af1b8a6 Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52747 Reviewed-by: Duncan Laurie <duncan@iceblink.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28 22:03:07 +02:00
PRIx64 "\n", field->name, option->name, value);
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
option = option->next;
}
field = field->next;
}
fprintf(fil, "\n");
}
static int emit_fw_config_probe(FILE *fil, struct device *dev)
{
struct fw_config_probe *probe = dev->probe;
fprintf(fil, "STORAGE struct fw_config %s_probe_list[] = {\n", dev->name);
while (probe) {
/* Find matching field. */
struct fw_config_field *field;
struct fw_config_option *option;
uint64_t mask, value;
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
field = find_fw_config_field(probe->field);
if (!field) {
printf("ERROR: fw_config_probe field %s not found\n", probe->field);
return -1;
}
option = find_fw_config_option(field, probe->option);
if (!option) {
printf("ERROR: fw_config_probe field %s option %s not found\n",
probe->field, probe->option);
return -1;
}
/* Fill out the probe structure with values from emit_fw_config(). */
fprintf(fil, "\t{\n");
fprintf(fil, "\t\t.field_name = FW_CONFIG_FIELD_%s_NAME,\n", probe->field);
fprintf(fil, "\t\t.option_name = FW_CONFIG_FIELD_%s_OPTION_%s_NAME,\n",
probe->field, probe->option);
fprintf(fil, "\t\t.mask = FW_CONFIG_FIELD_%s_MASK,\n", probe->field);
fprintf(fil, "\t\t.value = FW_CONFIG_FIELD_%s_OPTION_%s_VALUE,\n",
probe->field, probe->option);
fprintf(fil, "\t},\n");
probe = probe->next;
}
/* Add empty entry to mark end of list. */
fprintf(fil, "\t{ }\n};\n");
return 0;
}
sconfig: Allow to specify device operations Currently we only have runtime mechanisms to assign device operations to a node in our devicetree (with one exception: the root device). The most common method is to map PCI IDs to the device operations with a `struct pci_driver`. Another accustomed way is to let a chip driver assign them. For very common drivers, e.g. those in soc/intel/common/blocks/, the PCI ID lists grew very large and are incredibly error-prone. Often, IDs are missing and sometimes IDs are added almost mechanically without checking the code for compatibility. Maintaining these lists in a central place also reduces flexibility. Now, for onboard devices it is actually unnecessary to assign the device operations at runtime. We already know exactly what operations should be assigned. And since we are using chipset devicetrees, we have a perfect place to put that information. This patch adds a simple mechanism to `sconfig`. It allows us to speci- fy operations per device, e.g. device pci 00.0 alias system_agent on ops system_agent_ops end The operations are given as a C identifier. In this example, we simply assume that a global `struct device_operations system_agent_ops` exists. Change-Id: I2833d2f2450fde3206c33393f58b86fd4280b566 Signed-off-by: Nico Huber <nico.h@gmx.de> Reviewed-on: https://review.coreboot.org/c/coreboot/+/66483 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
2022-08-06 19:02:59 +02:00
/* Enqueue identifier to list with head `*it`, if not already present. */
void add_identifier(struct identifier **it, const char *id)
{
for (; *it != NULL; it = &(*it)->next) {
if (!strcmp((*it)->id, id))
return;
}
*it = S_ALLOC(sizeof(**it));
(*it)->id = id;
}
void add_device_ops(struct bus *bus, char *ops_id)
{
if (bus->dev->ops_id) {
printf("ERROR: Device operations may only be specified once,\n"
" found '%s', '%s'.\n", bus->dev->ops_id, ops_id);
exit(1);
}
add_identifier(&device_operations, ops_id);
bus->dev->ops_id = ops_id;
}
/*
* Allocate a new bus for the provided device.
* - If this is the first bus being allocated under this device, then its id
* is set to 0 and bus and last_bus are pointed to the newly allocated bus.
* - If this is not the first bus under this device, then its id is set to 1
* plus the id of last bus and newly allocated bus is added to the list of
* buses under the device. last_bus is updated to point to the newly
* allocated bus.
*/
static void alloc_bus(struct device *dev)
{
struct bus *bus = S_ALLOC(sizeof(*bus));
bus->dev = dev;
if (dev->last_bus == NULL) {
bus->id = 0;
dev->bus = bus;
} else {
bus->id = dev->last_bus->id + 1;
dev->last_bus->next_bus = bus;
}
dev->last_bus = bus;
}
/*
* Allocate a new device under the given parent. This function allocates a new
* device structure under the provided parent bus and allocates a bus structure
* under the newly allocated device.
*/
static struct device *alloc_dev(struct bus *parent)
{
struct device *dev = S_ALLOC(sizeof(*dev));
dev->parent = parent;
dev->subsystem_vendor = -1;
dev->subsystem_device = -1;
alloc_bus(dev);
return dev;
}
/*
* This function scans the children of given bus to see if any device matches
* the new device that is requested.
*
* Returns pointer to the node if found, else NULL.
*/
static struct device *get_dev(struct bus *parent, int path_a, int path_b,
int bustype, struct chip_instance *chip_instance)
{
struct device *child = parent->children;
while (child) {
if ((child->path_a == path_a) && (child->path_b == path_b) &&
(child->bustype == bustype) &&
(child->chip_instance == chip_instance))
return child;
child = child->sibling;
}
return NULL;
}
/*
* Add given node as child of the provided parent. If this is the first child of
* the parent, update parent->children pointer as well.
*/
static void set_new_child(struct bus *parent, struct device *child)
{
struct device *c = parent->children;
if (c) {
while (c->sibling)
c = c->sibling;
c->sibling = child;
} else
parent->children = child;
child->sibling = NULL;
child->parent = parent;
}
sconfig: Allow to link devices to other device's drivers Rarely, the driver of one device needs to know about another device that can be anywhere in the device hierarchy. Current applications boil down to EEPROMs that store information that is consumed by some code (e.g. MAC address). The idea is to give device nodes in the `devicetree.cb` an alias that can later be used to link it to a device driver's `config` structure. The driver has to declare a field of type `struct device *`, e.g. struct some_chip_driver_config { DEVTREE_CONST struct device *needed_eeprom; }; In the devicetree, the referenced device gets an alias, e.g. device i2c 0x50 alias my_eeprom on end The author of the devicetree is free to choose any alias name that is unique in the devicetree. Later, when configuring the driver the alias can be used to link the device with the field of a driver's config: chip some/chip/driver use my_eeprom as needed_eeprom end Override devices can add an alias if it does not exist, but cannot change the alias for a device that already exists. Alias names are checked for conflicts both in the base tree and in the override tree. References are resolved after the tree is parsed so aliases and references do not need to be in a specific order in the tree. Change-Id: I058a319f9b968924fbef9485a96c9e3f900a3ee8 Signed-off-by: Nico Huber <nico.huber@secunet.com> Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/35456 Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2020-06-03 19:20:07 +02:00
static const struct device *find_alias(const struct device *const parent,
const char *const alias)
{
if (parent->alias && !strcmp(parent->alias, alias))
return parent;
const struct bus *bus;
for (bus = parent->bus; bus; bus = bus->next_bus) {
const struct device *child;
for (child = bus->children; child; child = child->sibling) {
const struct device *const ret = find_alias(child, alias);
if (ret)
return ret;
}
}
return NULL;
}
static struct device *new_device_with_path(struct bus *parent,
struct chip_instance *chip_instance,
const int bustype, int path_a, int path_b,
char *alias, int status)
{
struct device *new_d;
/* If device is found under parent, no need to allocate new device. */
new_d = get_dev(parent, path_a, path_b, bustype, chip_instance);
if (new_d) {
alloc_bus(new_d);
return new_d;
}
new_d = alloc_dev(parent);
new_d->bustype = bustype;
new_d->path_a = path_a;
new_d->path_b = path_b;
sconfig: Allow to link devices to other device's drivers Rarely, the driver of one device needs to know about another device that can be anywhere in the device hierarchy. Current applications boil down to EEPROMs that store information that is consumed by some code (e.g. MAC address). The idea is to give device nodes in the `devicetree.cb` an alias that can later be used to link it to a device driver's `config` structure. The driver has to declare a field of type `struct device *`, e.g. struct some_chip_driver_config { DEVTREE_CONST struct device *needed_eeprom; }; In the devicetree, the referenced device gets an alias, e.g. device i2c 0x50 alias my_eeprom on end The author of the devicetree is free to choose any alias name that is unique in the devicetree. Later, when configuring the driver the alias can be used to link the device with the field of a driver's config: chip some/chip/driver use my_eeprom as needed_eeprom end Override devices can add an alias if it does not exist, but cannot change the alias for a device that already exists. Alias names are checked for conflicts both in the base tree and in the override tree. References are resolved after the tree is parsed so aliases and references do not need to be in a specific order in the tree. Change-Id: I058a319f9b968924fbef9485a96c9e3f900a3ee8 Signed-off-by: Nico Huber <nico.huber@secunet.com> Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/35456 Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2020-06-03 19:20:07 +02:00
new_d->alias = alias;
new_d->enabled = status & 0x01;
new_d->hidden = (status >> 1) & 0x01;
new_d->mandatory = (status >> 2) & 0x01;
new_d->chip_instance = chip_instance;
set_new_child(parent, new_d);
switch (bustype) {
case PCI:
new_d->path = ".type=DEVICE_PATH_PCI,{.pci={ .devfn = PCI_DEVFN(0x%x,%d)}}";
break;
case PNP:
new_d->path = ".type=DEVICE_PATH_PNP,{.pnp={ .port = 0x%x, .device = 0x%x }}";
break;
case I2C:
new_d->path = ".type=DEVICE_PATH_I2C,{.i2c={ .device = 0x%x, .mode_10bit = %d }}";
break;
case CPU_CLUSTER:
new_d->path = ".type=DEVICE_PATH_CPU_CLUSTER,{.cpu_cluster={ .cluster = 0x%x }}";
break;
case CPU:
new_d->path = ".type=DEVICE_PATH_CPU,{.cpu={ .id = 0x%x }}";
break;
case DOMAIN:
new_d->path = ".type=DEVICE_PATH_DOMAIN,{.domain={ .domain = 0x%x }}";
break;
case GENERIC:
new_d->path = ".type=DEVICE_PATH_GENERIC,{.generic={ .id = 0x%x, .subid = 0x%x }}";
break;
case SPI:
new_d->path = ".type=DEVICE_PATH_SPI,{.spi={ .cs = 0x%x }}";
break;
case USB:
new_d->path = ".type=DEVICE_PATH_USB,{.usb={ .port_type = %d, .port_id = %d }}";
break;
case MMIO:
new_d->path = ".type=DEVICE_PATH_MMIO,{.mmio={ .addr = 0x%x }}";
break;
device + util/sconfig: introduce new device `gpio` Introduce a new device `gpio` that is going to be used for generic abstraction of gpio operations in the devicetree. The general idea behind this is that every chip can have gpios that shall be accessible in a very generic way by any driver through the devicetree. The chip that implements the chip-specific gpio operations has to assign them to the generic device operations struct, which then gets assigned to the gpio device during device probing. See CB:48583 for how this gets done for the SoCs using intelblocks/gpio. The gpio device then can be added to the devicetree with an alias name like in the following example: chip soc/whateverlake device gpio 0 alias soc_gpio on end ... end Any driver that requires access to this gpio device needs to have a device pointer (or multiple) and an option for specifying the gpio to be used in its chip config like this: struct drivers_ipmi_config { ... DEVTREE_CONST struct device *gpio_dev; u16 post_complete_gpio; ... }; The device `soc_gpio` can then be linked to the chip driver's `gpio_dev` above by using the syntax `use ... as ...`, which was introduced in commit 8e1ea52: chip drivers/ipmi use soc_gpio as gpio_dev register "bmc_jumper_gpio" = "GPP_D22" ... end The IPMI driver can then use the generic gpio operations without any knowlege of the chip's specifics: unsigned int gpio_val; const struct gpio_operations *gpio_ops; gpio_ops = dev_get_gpio_ops(conf->gpio_dev); gpio_val = gpio_ops->get(conf->bmc_jumper_gpio); For a full example have a look at CB:48096 and CB:48095. This change adds the new device type to sconfig and adds generic gpio operations to the `device_operations` struct. Also, a helper for getting the gpio operations from a device after checking them for NULL pointers gets added. Successfully tested on Supermicro X11SSM-F with CB:48097, X11SSH-TF with CB:48711 and OCP DeltaLake with CB:48672. Change-Id: Ic4572ad8b37bd1afd2fb213b2c67fb8aec536786 Tested-by: Johnny Lin <Johnny_Lin@wiwynn.com> Tested-by: Michael Niewöhner <foss@mniewoehner.de> Tested-by: Patrick Rudolph <siro@das-labor.org> Signed-off-by: Michael Niewöhner <foss@mniewoehner.de> Reviewed-on: https://review.coreboot.org/c/coreboot/+/48582 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nico Huber <nico.h@gmx.de>
2020-12-11 21:26:02 +01:00
case GPIO:
new_d->path = ".type=DEVICE_PATH_GPIO,{.gpio={ .id = 0x%x }}";
break;
case MDIO:
new_d->path = ".type=DEVICE_PATH_MDIO,{.mdio={ .addr = 0x%x }}";
break;
}
return new_d;
}
struct device *new_device_reference(struct bus *parent,
struct chip_instance *chip_instance,
const char *reference, int status)
{
const struct device *dev = find_alias(&base_root_dev, reference);
if (!dev) {
printf("ERROR: Unable to find device reference %s\n", reference);
exit(1);
}
return new_device_with_path(parent, chip_instance, dev->bustype, dev->path_a,
dev->path_b, NULL, status);
}
struct device *new_device_raw(struct bus *parent,
struct chip_instance *chip_instance,
const int bustype, const char *devnum,
char *alias, int status)
{
char *tmp;
int path_a;
int path_b = 0;
/* Check for alias name conflicts. */
if (alias && find_alias(root_parent->dev, alias)) {
printf("ERROR: Alias already exists: %s\n", alias);
exit(1);
}
path_a = strtol(devnum, &tmp, 16);
if (*tmp == '.') {
tmp++;
path_b = strtol(tmp, NULL, 16);
}
return new_device_with_path(parent, chip_instance, bustype, path_a, path_b, alias,
status);
}
static void new_resource(struct device *dev, int type, int index, int base)
{
struct resource *r = S_ALLOC(sizeof(struct resource));
r->type = type;
r->index = index;
r->base = base;
if (dev->res) {
struct resource *head = dev->res;
while (head->next)
head = head->next;
head->next = r;
} else {
dev->res = r;
}
}
void add_resource(struct bus *bus, int type, int index, int base)
{
new_resource(bus->dev, type, index, base);
}
static void add_reg(struct reg **const head, char *const name, char *const val)
{
struct reg *const r = S_ALLOC(sizeof(struct reg));
struct reg *prev = NULL;
struct reg *cur;
r->key = name;
r->value = val;
for (cur = *head; cur != NULL; prev = cur, cur = cur->next) {
const int sort = strcmp(r->key, cur->key);
if (sort == 0) {
printf("ERROR: duplicate 'register' key '%s'.\n", r->key);
exit(1);
}
if (sort < 0)
break;
}
r->next = cur;
if (prev)
prev->next = r;
else
*head = r;
}
void add_register(struct chip_instance *chip_instance, char *name, char *val)
{
add_reg(&chip_instance->reg, name, val);
}
sconfig: Allow to link devices to other device's drivers Rarely, the driver of one device needs to know about another device that can be anywhere in the device hierarchy. Current applications boil down to EEPROMs that store information that is consumed by some code (e.g. MAC address). The idea is to give device nodes in the `devicetree.cb` an alias that can later be used to link it to a device driver's `config` structure. The driver has to declare a field of type `struct device *`, e.g. struct some_chip_driver_config { DEVTREE_CONST struct device *needed_eeprom; }; In the devicetree, the referenced device gets an alias, e.g. device i2c 0x50 alias my_eeprom on end The author of the devicetree is free to choose any alias name that is unique in the devicetree. Later, when configuring the driver the alias can be used to link the device with the field of a driver's config: chip some/chip/driver use my_eeprom as needed_eeprom end Override devices can add an alias if it does not exist, but cannot change the alias for a device that already exists. Alias names are checked for conflicts both in the base tree and in the override tree. References are resolved after the tree is parsed so aliases and references do not need to be in a specific order in the tree. Change-Id: I058a319f9b968924fbef9485a96c9e3f900a3ee8 Signed-off-by: Nico Huber <nico.huber@secunet.com> Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/35456 Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2020-06-03 19:20:07 +02:00
void add_reference(struct chip_instance *const chip_instance,
char *const name, char *const alias)
{
add_reg(&chip_instance->ref, name, alias);
}
static void set_reference(struct chip_instance *const chip_instance,
char *const name, char *const alias)
{
const struct device *const dev = find_alias(&base_root_dev, alias);
if (!dev) {
printf("ERROR: Cannot find device alias '%s'.\n", alias);
exit(1);
}
char *const ref_name = S_ALLOC(strlen(dev->name) + 2);
sprintf(ref_name, "&%s", dev->name);
add_register(chip_instance, name, ref_name);
}
static void update_references(FILE *file, FILE *head, struct device *dev,
struct device *next)
{
struct reg *ref;
for (ref = dev->chip_instance->ref; ref; ref = ref->next)
set_reference(dev->chip_instance, ref->key, ref->value);
}
void add_slot_desc(struct bus *bus, char *type, char *length, char *designation,
char *data_width)
{
struct device *dev = bus->dev;
if (dev->bustype != PCI && dev->bustype != DOMAIN) {
printf("ERROR: 'slot_type' only allowed for PCI devices\n");
exit(1);
}
dev->smbios_slot_type = type;
dev->smbios_slot_length = length;
dev->smbios_slot_data_width = data_width;
dev->smbios_slot_designation = designation;
}
void add_smbios_dev_info(struct bus *bus, long instance_id, const char *refdes)
{
struct device *dev = bus->dev;
if (dev->bustype != PCI && dev->bustype != DOMAIN) {
printf("ERROR: 'dev_info' only allowed for PCI devices\n");
exit(1);
}
if (instance_id < 0 || instance_id > UINT8_MAX) {
printf("ERROR: SMBIOS dev info instance ID '%ld' out of range\n", instance_id);
exit(1);
}
dev->smbios_instance_id_valid = 1;
dev->smbios_instance_id = (unsigned int)instance_id;
dev->smbios_refdes = refdes;
}
void add_pci_subsystem_ids(struct bus *bus, int vendor, int device,
int inherit)
{
struct device *dev = bus->dev;
if (dev->bustype != PCI && dev->bustype != DOMAIN) {
printf("ERROR: 'subsystem' only allowed for PCI devices\n");
exit(1);
}
dev->subsystem_vendor = vendor;
dev->subsystem_device = device;
dev->inherit_subsystem = inherit;
}
static int dev_has_children(struct device *dev)
{
struct bus *bus = dev->bus;
while (bus) {
if (bus->children)
return 1;
bus = bus->next_bus;
}
return 0;
}
static void pass0(FILE *fil, FILE *head, struct device *ptr, struct device *next)
{
static int dev_id;
if (ptr == &base_root_dev) {
fprintf(fil, "STORAGE struct bus %s_links[];\n",
ptr->name);
return;
}
Make the device tree available in the rom stage We thought about two ways to do this change. The way we decided to try was to 1. drop all ops from devices in romstage 2. constify all devices in romstage (make them read-only) so we can compile static.c into romstage 3. the device tree "devices" can be used to read configuration from the device tree (and nothing else, really) 4. the device tree devices are accessed through struct device * in romstage only. device_t stays the typedef to int in romstage 5. Use the same static.c file in ramstage and romstage We declare structs as follows: ROMSTAGE_CONST struct bus dev_root_links[]; ROMSTAGE_CONST is const in romstage and empty in ramstage; This forces all of the device tree into the text area. So a struct looks like this: static ROMSTAGE_CONST struct device _dev21 = { #ifndef __PRE_RAM__ .ops = 0, #endif .bus = &_dev7_links[0], .path = {.type=DEVICE_PATH_PCI,{.pci={ .devfn = PCI_DEVFN(0x1c,3)}}}, .enabled = 0, .on_mainboard = 1, .subsystem_vendor = 0x1ae0, .subsystem_device = 0xc000, .link_list = NULL, .sibling = &_dev22, #ifndef __PRE_RAM__ .chip_ops = &southbridge_intel_bd82x6x_ops, #endif .chip_info = &southbridge_intel_bd82x6x_info_10, .next=&_dev22 }; Change-Id: I722454d8d3c40baf7df989f5a6891f6ba7db5727 Signed-off-by: Ronald G. Minnich <rminnich@chromium.org> Signed-off-by: Stefan Reinauer <reinauer@google.com> Reviewed-on: http://review.coreboot.org/1398 Tested-by: build bot (Jenkins) Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2012-08-01 01:47:25 +02:00
char *name;
if (ptr->alias) {
name = S_ALLOC(6 + strlen(ptr->alias));
sprintf(name, "_dev_%s", ptr->alias);
} else {
name = S_ALLOC(11);
sprintf(name, "_dev_%d", dev_id++);
}
ptr->name = name;
fprintf(fil, "STORAGE struct device %s;\n", ptr->name);
if (ptr->res)
fprintf(fil, "STORAGE struct resource %s_res[];\n",
ptr->name);
if (dev_has_children(ptr))
fprintf(fil, "STORAGE struct bus %s_links[];\n",
ptr->name);
if (next)
return;
fprintf(fil,
"DEVTREE_CONST struct device * DEVTREE_CONST last_dev = &%s;\n",
ptr->name);
}
static void emit_smbios_data(FILE *fil, struct device *ptr)
{
fprintf(fil, "#if !DEVTREE_EARLY\n");
fprintf(fil, "#if CONFIG(GENERATE_SMBIOS_TABLES)\n");
/* SMBIOS types start at 1, if zero it hasn't been set */
if (ptr->smbios_slot_type)
fprintf(fil, "\t.smbios_slot_type = %s,\n",
ptr->smbios_slot_type);
if (ptr->smbios_slot_data_width)
fprintf(fil, "\t.smbios_slot_data_width = %s,\n",
ptr->smbios_slot_data_width);
if (ptr->smbios_slot_designation)
fprintf(fil, "\t.smbios_slot_designation = \"%s\",\n",
ptr->smbios_slot_designation);
if (ptr->smbios_slot_length)
fprintf(fil, "\t.smbios_slot_length = %s,\n",
ptr->smbios_slot_length);
/* Fill in SMBIOS type41 fields */
if (ptr->smbios_instance_id_valid) {
fprintf(fil, "\t.smbios_instance_id_valid = true,\n");
fprintf(fil, "\t.smbios_instance_id = %u,\n", ptr->smbios_instance_id);
if (ptr->smbios_refdes)
fprintf(fil, "\t.smbios_refdes = \"%s\",\n", ptr->smbios_refdes);
}
fprintf(fil, "#endif\n");
fprintf(fil, "#endif\n");
}
static void emit_resources(FILE *fil, struct device *ptr)
{
if (ptr->res == NULL)
return;
int i = 1;
fprintf(fil, "STORAGE struct resource %s_res[] = {\n", ptr->name);
struct resource *r = ptr->res;
while (r) {
fprintf(fil,
"\t\t{ .flags=IORESOURCE_FIXED | IORESOURCE_ASSIGNED | IORESOURCE_");
if (r->type == IRQ)
fprintf(fil, "IRQ");
if (r->type == DRQ)
fprintf(fil, "DRQ");
if (r->type == IO)
fprintf(fil, "IO");
fprintf(fil, ", .index=0x%x, .base=0x%x,", r->index,
r->base);
if (r->next)
fprintf(fil, ".next=&%s_res[%d]},\n", ptr->name,
i++);
else
fprintf(fil, ".next=NULL },\n");
r = r->next;
}
fprintf(fil, "\t };\n");
}
static void emit_bus(FILE *fil, struct bus *bus)
{
fprintf(fil, "\t\t[%d] = {\n", bus->id);
fprintf(fil, "\t\t\t.link_num = %d,\n", bus->id);
fprintf(fil, "\t\t\t.dev = &%s,\n", bus->dev->name);
if (bus->children)
fprintf(fil, "\t\t\t.children = &%s,\n", bus->children->name);
if (bus->next_bus)
fprintf(fil, "\t\t\t.next=&%s_links[%d],\n", bus->dev->name,
bus->id + 1);
else
fprintf(fil, "\t\t\t.next = NULL,\n");
fprintf(fil, "\t\t},\n");
}
static void emit_dev_links(FILE *fil, struct device *ptr)
{
fprintf(fil, "STORAGE struct bus %s_links[] = {\n",
ptr->name);
struct bus *bus = ptr->bus;
while (bus) {
emit_bus(fil, bus);
bus = bus->next_bus;
}
fprintf(fil, "\t};\n");
}
static struct chip_instance *get_chip_instance(const struct device *dev)
{
struct chip_instance *chip_ins = dev->chip_instance;
util/sconfig: Drop use of ref_count for chip_instance chip_instance structure currently uses a ref_count to determine how many devices hold reference to that instance. If the count drops to zero, then it is assumed that the chip instance is a duplicate in override tree and has a similar instance that is already overriden in base device tree. ref_count is currently decremented whenever a device in override tree matches the one in base device tree and the registers from the override tree instance are copied over to the base tree instance. On the other hand, if a device in override tree does not match any device in base tree under a given parent, then the device is added to base tree and all the devices in its subtree that hold pointers to its parent chip instance are updated to point to the parent's chip instance in base tree. This is done as part of update_chip_pointers. However, there are a couple of issues that this suffers from: a) If a device is present only in override tree and it does not have its own chip (i.e. pointing to parent's chip instance), then it results in sconfig emiiting parent's chip instance (which can be the SoC chip instance) in static.c even though it is unused. This is because update_chip_pointers() does not call delete_chip_instance() before reassigning the chip instance pointer. b) If a device is added under root device only in the override tree and it does not have its own chip instance (i.e. uses SoC chip instance), then it results in sconfig emitting a copy of the SoC chip instance and setting that as chip_ops for this new device in the override tree. In order to fix the above issues, this change drops the ref_count field from chip_instance structure and instead adds a forwarding pointer `base_chip_instance`. This is setup as per the following rules: 1. If the instance belongs to base devicetree, base_chip_instance is set to NULL. 2. If the instance belongs to override tree, then it is set to its corresponding chip instance in base tree (if present), else set to NULL. State of base_chip_instance is then used when emitting chips and devices using the following rules: 1. If a chip_instance has non-NULL base_chip_instance, then that chip instance is not emitted to static.c 2. When emitting chip_ops for a device, base_chip_instance is used to determine the correct chip instance name to emit. BUG=b:155549176 TEST=Verified that the static.c file generated for base/override tree combination is correct when new devices without chips are added only to override tree. Change-Id: Idbb5b34f49bf874da3f30ebb6a6a0e2d8d091fe5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/41007 Reviewed-by: Aaron Durbin <adurbin@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2020-05-03 01:05:29 +02:00
/*
* If the chip instance of device has base_chip_instance pointer set, then follow that
* to update the chip instance for current device.
*/
if (chip_ins->base_chip_instance)
chip_ins = chip_ins->base_chip_instance;
return chip_ins;
}
static void pass1(FILE *fil, FILE *head, struct device *ptr, struct device *next)
{
struct chip_instance *chip_ins = get_chip_instance(ptr);
int has_children = dev_has_children(ptr);
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
/* Emit probe structures. */
if (ptr->probe && (emit_fw_config_probe(fil, ptr) < 0)) {
if (head)
fclose(head);
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
fclose(fil);
exit(1);
}
if (ptr == &base_root_dev)
fprintf(fil, "DEVTREE_CONST struct device %s = {\n", ptr->name);
else
fprintf(fil, "STORAGE struct device %s = {\n", ptr->name);
fprintf(fil, "#if !DEVTREE_EARLY\n");
/*
sconfig: Allow to specify device operations Currently we only have runtime mechanisms to assign device operations to a node in our devicetree (with one exception: the root device). The most common method is to map PCI IDs to the device operations with a `struct pci_driver`. Another accustomed way is to let a chip driver assign them. For very common drivers, e.g. those in soc/intel/common/blocks/, the PCI ID lists grew very large and are incredibly error-prone. Often, IDs are missing and sometimes IDs are added almost mechanically without checking the code for compatibility. Maintaining these lists in a central place also reduces flexibility. Now, for onboard devices it is actually unnecessary to assign the device operations at runtime. We already know exactly what operations should be assigned. And since we are using chipset devicetrees, we have a perfect place to put that information. This patch adds a simple mechanism to `sconfig`. It allows us to speci- fy operations per device, e.g. device pci 00.0 alias system_agent on ops system_agent_ops end The operations are given as a C identifier. In this example, we simply assume that a global `struct device_operations system_agent_ops` exists. Change-Id: I2833d2f2450fde3206c33393f58b86fd4280b566 Signed-off-by: Nico Huber <nico.h@gmx.de> Reviewed-on: https://review.coreboot.org/c/coreboot/+/66483 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
2022-08-06 19:02:59 +02:00
* ops field can be set in the devicetree. If unspecified, it is set
* to default_dev_ops_root only for the root device, other devices
* get it set by the driver at runtime.
*/
sconfig: Allow to specify device operations Currently we only have runtime mechanisms to assign device operations to a node in our devicetree (with one exception: the root device). The most common method is to map PCI IDs to the device operations with a `struct pci_driver`. Another accustomed way is to let a chip driver assign them. For very common drivers, e.g. those in soc/intel/common/blocks/, the PCI ID lists grew very large and are incredibly error-prone. Often, IDs are missing and sometimes IDs are added almost mechanically without checking the code for compatibility. Maintaining these lists in a central place also reduces flexibility. Now, for onboard devices it is actually unnecessary to assign the device operations at runtime. We already know exactly what operations should be assigned. And since we are using chipset devicetrees, we have a perfect place to put that information. This patch adds a simple mechanism to `sconfig`. It allows us to speci- fy operations per device, e.g. device pci 00.0 alias system_agent on ops system_agent_ops end The operations are given as a C identifier. In this example, we simply assume that a global `struct device_operations system_agent_ops` exists. Change-Id: I2833d2f2450fde3206c33393f58b86fd4280b566 Signed-off-by: Nico Huber <nico.h@gmx.de> Reviewed-on: https://review.coreboot.org/c/coreboot/+/66483 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
2022-08-06 19:02:59 +02:00
if (ptr->ops_id)
fprintf(fil, "\t.ops = &%s,\n", ptr->ops_id);
else if (ptr == &base_root_dev)
fprintf(fil, "\t.ops = &default_dev_ops_root,\n");
else
fprintf(fil, "\t.ops = NULL,\n");
fprintf(fil, "#endif\n");
fprintf(fil, "\t.bus = &%s_links[%d],\n", ptr->parent->dev->name,
ptr->parent->id);
fprintf(fil, "\t.path = {");
fprintf(fil, ptr->path, ptr->path_a, ptr->path_b);
fprintf(fil, "},\n");
fprintf(fil, "\t.enabled = %d,\n", ptr->enabled);
fprintf(fil, "\t.hidden = %d,\n", ptr->hidden);
fprintf(fil, "\t.mandatory = %d,\n", ptr->mandatory);
fprintf(fil, "\t.on_mainboard = 1,\n");
if (ptr->subsystem_vendor > 0)
fprintf(fil, "\t.subsystem_vendor = 0x%04x,\n",
ptr->subsystem_vendor);
if (ptr->subsystem_device > 0)
fprintf(fil, "\t.subsystem_device = 0x%04x,\n",
ptr->subsystem_device);
if (ptr->res) {
fprintf(fil, "\t.resource_list = &%s_res[0],\n",
ptr->name);
}
if (has_children)
fprintf(fil, "\t.link_list = &%s_links[0],\n",
ptr->name);
else
fprintf(fil, "\t.link_list = NULL,\n");
if (ptr->sibling)
fprintf(fil, "\t.sibling = &%s,\n", ptr->sibling->name);
else
fprintf(fil, "\t.sibling = NULL,\n");
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
if (ptr->probe)
fprintf(fil, "\t.probe_list = %s_probe_list,\n", ptr->name);
fprintf(fil, "#if !DEVTREE_EARLY\n");
fprintf(fil, "\t.chip_ops = &%s_ops,\n",
chip_ins->chip->name_underscore);
if (chip_ins == &mainboard_instance)
fprintf(fil, "\t.name = mainboard_name,\n");
fprintf(fil, "#endif\n");
if (chip_ins->chip->chiph_exists)
fprintf(fil, "\t.chip_info = &%s_info_%d,\n",
chip_ins->chip->name_underscore, chip_ins->id);
if (next)
fprintf(fil, "\t.next=&%s,\n", next->name);
emit_smbios_data(fil, ptr);
fprintf(fil, "};\n");
emit_resources(fil, ptr);
if (has_children)
emit_dev_links(fil, ptr);
}
static void expose_device_names(FILE *fil, FILE *head, struct device *ptr, struct device *next)
{
struct chip_instance *chip_ins = get_chip_instance(ptr);
/* Only devices on root bus here. */
if (ptr->bustype == PCI && ptr->parent->dev->bustype == DOMAIN) {
fprintf(head, "extern DEVTREE_CONST struct device *const __pci_0_%02x_%d;\n",
ptr->path_a, ptr->path_b);
fprintf(fil, "DEVTREE_CONST struct device *const __pci_0_%02x_%d = &%s;\n",
ptr->path_a, ptr->path_b, ptr->name);
if (chip_ins->chip->chiph_exists) {
fprintf(head, "extern DEVTREE_CONST void *const __pci_0_%02x_%d_config;\n",
ptr->path_a, ptr->path_b);
fprintf(fil, "DEVTREE_CONST void *const __pci_0_%02x_%d_config = &%s_info_%d;\n",
ptr->path_a, ptr->path_b, chip_ins->chip->name_underscore, chip_ins->id);
}
}
if (ptr->bustype == PNP) {
fprintf(head, "extern DEVTREE_CONST struct device *const __pnp_%04x_%02x;\n",
ptr->path_a, ptr->path_b);
fprintf(fil, "DEVTREE_CONST struct device *const __pnp_%04x_%02x = &%s;\n",
ptr->path_a, ptr->path_b, ptr->name);
}
if (ptr->alias) {
fprintf(head, "extern DEVTREE_CONST struct device *const %s_ptr;\n", ptr->name);
fprintf(fil, "DEVTREE_CONST struct device *const %s_ptr = &%s;\n",
ptr->name, ptr->name);
}
}
static void add_siblings_to_queue(struct queue_entry **bfs_q_head,
struct device *d)
{
while (d) {
enqueue_tail(bfs_q_head, d);
d = d->sibling;
}
}
static void add_children_to_queue(struct queue_entry **bfs_q_head,
struct device *d)
{
struct bus *bus = d->bus;
while (bus) {
if (bus->children)
add_siblings_to_queue(bfs_q_head, bus->children);
bus = bus->next_bus;
}
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
}
static void walk_device_tree(FILE *fil, FILE *head, struct device *ptr,
void (*func)(FILE *, FILE *, struct device *,
struct device *))
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
{
struct queue_entry *bfs_q_head = NULL;
enqueue_tail(&bfs_q_head, ptr);
while ((ptr = dequeue_head(&bfs_q_head))) {
add_children_to_queue(&bfs_q_head, ptr);
func(fil, head, ptr, peek_queue_head(bfs_q_head));
}
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
}
static void emit_chip_headers(FILE *fil, struct chip *chip)
{
struct chip *tmp = chip;
while (chip) {
if (chip->chiph_exists)
fprintf(fil, "#include \"%s/chip.h\"\n", chip->name);
chip = chip->next;
}
fprintf(fil, "\n#if !DEVTREE_EARLY\n");
fprintf(fil,
"__attribute__((weak)) struct chip_operations mainboard_ops = {};\n");
chip = tmp;
while (chip) {
/* A lot of cpus do not define chip_operations at all, and the ones
that do only initialise .name. */
if (strstr(chip->name_underscore, "cpu_") == chip->name_underscore) {
fprintf(fil,
"__attribute__((weak)) struct chip_operations %s_ops = {};\n",
chip->name_underscore);
} else {
fprintf(fil, "extern struct chip_operations %s_ops;\n",
chip->name_underscore);
}
chip = chip->next;
}
fprintf(fil, "#endif\n");
}
static void emit_chip_instance(FILE *fil, struct chip_instance *instance)
{
fprintf(fil, "STORAGE struct %s_config %s_info_%d = {",
instance->chip->name_underscore,
instance->chip->name_underscore,
instance->id);
if (instance->reg) {
fprintf(fil, "\n");
struct reg *r = instance->reg;
while (r) {
fprintf(fil, "\t.%s = %s,\n", r->key, r->value);
r = r->next;
}
}
fprintf(fil, "};\n\n");
}
sconfig: Allow to link devices to other device's drivers Rarely, the driver of one device needs to know about another device that can be anywhere in the device hierarchy. Current applications boil down to EEPROMs that store information that is consumed by some code (e.g. MAC address). The idea is to give device nodes in the `devicetree.cb` an alias that can later be used to link it to a device driver's `config` structure. The driver has to declare a field of type `struct device *`, e.g. struct some_chip_driver_config { DEVTREE_CONST struct device *needed_eeprom; }; In the devicetree, the referenced device gets an alias, e.g. device i2c 0x50 alias my_eeprom on end The author of the devicetree is free to choose any alias name that is unique in the devicetree. Later, when configuring the driver the alias can be used to link the device with the field of a driver's config: chip some/chip/driver use my_eeprom as needed_eeprom end Override devices can add an alias if it does not exist, but cannot change the alias for a device that already exists. Alias names are checked for conflicts both in the base tree and in the override tree. References are resolved after the tree is parsed so aliases and references do not need to be in a specific order in the tree. Change-Id: I058a319f9b968924fbef9485a96c9e3f900a3ee8 Signed-off-by: Nico Huber <nico.huber@secunet.com> Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/35456 Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2020-06-03 19:20:07 +02:00
static void emit_chip_configs(FILE *fil)
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
{
struct chip *chip = chip_header.next;
struct chip_instance *instance;
int chip_id;
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
for (; chip; chip = chip->next) {
util/sconfig: Re-factor sconfig to not assume chip as device This change adds a new structure "struct chip" to identify elements of type chip rather than re-using the structure for device. Until now chip was treated as a device while generating the parse tree and then device tree postprocessing skipped over all the chip entries in children and sibling pointers of device nodes. With this change, the device tree will only contain struct device in the parsed tree. It helps by avoiding unnecessary pointers to chip structure as children or next_sibling and then skipping those elements in post processing. Every device can then hold a pointer to its chip. When generating static.c, chip structure is emitted before device structure to ensure that the device structure has chip within its scope. Externally, the only visible change in static.c should be the order in which chip/device elements are emitted i.e. previously all chips under a particular device were emitted to static.c and then the devices using those chips. Now, all chips are emitted before all the devices in static.c BUG=b:80081934 TEST=Verified that abuild is successful for all boards. Also, verified that static.c generated for eve, kahlee, scarlet, asrock imb_a180 is unchanged from before in node definitions. Change-Id: I255092f527c8eecb144385eb681df20e54caf8f5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/26720 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2018-05-31 00:09:09 +02:00
if (!chip->chiph_exists)
continue;
chip_id = 1;
instance = chip->instance;
while (instance) {
util/sconfig: Drop use of ref_count for chip_instance chip_instance structure currently uses a ref_count to determine how many devices hold reference to that instance. If the count drops to zero, then it is assumed that the chip instance is a duplicate in override tree and has a similar instance that is already overriden in base device tree. ref_count is currently decremented whenever a device in override tree matches the one in base device tree and the registers from the override tree instance are copied over to the base tree instance. On the other hand, if a device in override tree does not match any device in base tree under a given parent, then the device is added to base tree and all the devices in its subtree that hold pointers to its parent chip instance are updated to point to the parent's chip instance in base tree. This is done as part of update_chip_pointers. However, there are a couple of issues that this suffers from: a) If a device is present only in override tree and it does not have its own chip (i.e. pointing to parent's chip instance), then it results in sconfig emiiting parent's chip instance (which can be the SoC chip instance) in static.c even though it is unused. This is because update_chip_pointers() does not call delete_chip_instance() before reassigning the chip instance pointer. b) If a device is added under root device only in the override tree and it does not have its own chip instance (i.e. uses SoC chip instance), then it results in sconfig emitting a copy of the SoC chip instance and setting that as chip_ops for this new device in the override tree. In order to fix the above issues, this change drops the ref_count field from chip_instance structure and instead adds a forwarding pointer `base_chip_instance`. This is setup as per the following rules: 1. If the instance belongs to base devicetree, base_chip_instance is set to NULL. 2. If the instance belongs to override tree, then it is set to its corresponding chip instance in base tree (if present), else set to NULL. State of base_chip_instance is then used when emitting chips and devices using the following rules: 1. If a chip_instance has non-NULL base_chip_instance, then that chip instance is not emitted to static.c 2. When emitting chip_ops for a device, base_chip_instance is used to determine the correct chip instance name to emit. BUG=b:155549176 TEST=Verified that the static.c file generated for base/override tree combination is correct when new devices without chips are added only to override tree. Change-Id: Idbb5b34f49bf874da3f30ebb6a6a0e2d8d091fe5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/41007 Reviewed-by: Aaron Durbin <adurbin@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2020-05-03 01:05:29 +02:00
/*
* Emit this chip instance only if there is no forwarding pointer to the
* base tree chip instance.
*/
if (instance->base_chip_instance == NULL) {
instance->id = chip_id++;
emit_chip_instance(fil, instance);
}
instance = instance->next;
}
}
}
sconfig: Allow to specify device operations Currently we only have runtime mechanisms to assign device operations to a node in our devicetree (with one exception: the root device). The most common method is to map PCI IDs to the device operations with a `struct pci_driver`. Another accustomed way is to let a chip driver assign them. For very common drivers, e.g. those in soc/intel/common/blocks/, the PCI ID lists grew very large and are incredibly error-prone. Often, IDs are missing and sometimes IDs are added almost mechanically without checking the code for compatibility. Maintaining these lists in a central place also reduces flexibility. Now, for onboard devices it is actually unnecessary to assign the device operations at runtime. We already know exactly what operations should be assigned. And since we are using chipset devicetrees, we have a perfect place to put that information. This patch adds a simple mechanism to `sconfig`. It allows us to speci- fy operations per device, e.g. device pci 00.0 alias system_agent on ops system_agent_ops end The operations are given as a C identifier. In this example, we simply assume that a global `struct device_operations system_agent_ops` exists. Change-Id: I2833d2f2450fde3206c33393f58b86fd4280b566 Signed-off-by: Nico Huber <nico.h@gmx.de> Reviewed-on: https://review.coreboot.org/c/coreboot/+/66483 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
2022-08-06 19:02:59 +02:00
static void emit_identifiers(FILE *fil, const char *decl, const struct identifier *it)
{
for (; it != NULL; it = it->next)
fprintf(fil, "extern %s %s;\n", decl, it->id);
}
static void inherit_subsystem_ids(FILE *file, FILE *head, struct device *dev,
struct device *next)
{
struct device *p;
if (dev->subsystem_vendor != -1 && dev->subsystem_device != -1) {
/* user already gave us a subsystem vendor/device */
return;
}
for (p = dev; p && p->parent->dev != p; p = p->parent->dev) {
if (p->bustype != PCI && p->bustype != DOMAIN)
continue;
if (p->inherit_subsystem) {
dev->subsystem_vendor = p->subsystem_vendor;
dev->subsystem_device = p->subsystem_device;
break;
}
}
}
static void usage(void)
{
printf("usage: sconfig <options>\n");
printf(" -c | --output_c : Path to output static.c file (required)\n");
printf(" -r | --output_h : Path to header static.h file (required)\n");
printf(" -d | --output_d : Path to header static_devices.h file (required)\n");
printf(" -f | --output_f : Path to header static_fw_config.h file (required)\n");
printf(" -m | --mainboard_devtree : Path to mainboard devicetree file (required)\n");
printf(" -o | --override_devtree : Path to override devicetree file (optional)\n");
printf(" -p | --chipset_devtree : Path to chipset/SOC devicetree file (optional)\n");
exit(1);
}
static void parse_devicetree(const char *file, struct bus *parent)
{
FILE *filec = fopen(file, "r");
if (!filec) {
perror(NULL);
exit(1);
}
yyrestart(filec);
root_parent = parent;
linenum = 0;
yyparse();
fclose(filec);
}
static int device_probe_count(struct fw_config_probe *probe)
{
int count = 0;
while (probe) {
probe = probe->next;
count++;
}
return count;
}
/*
* When overriding devices, use the following rules:
* 1. If probe count matches and:
* a. Entire probe list matches for both devices -> Same device, override.
* b. No probe entries match -> Different devices, do not override.
* c. Partial list matches -> Bad device tree entries, fail build.
*
* 2. If probe counts do not match and:
* a. No probe entries match -> Different devices, do not override.
* b. Partial list matches -> Bad device tree entries, fail build.
*/
static int device_probes_match(struct device *a, struct device *b)
{
struct fw_config_probe *a_probe = a->probe;
struct fw_config_probe *b_probe = b->probe;
int a_probe_count = device_probe_count(a_probe);
int b_probe_count = device_probe_count(b_probe);
int match_count = 0;
while (a_probe) {
if (check_probe_exists(b_probe, a_probe->field, a_probe->option))
match_count++;
a_probe = a_probe->next;
}
if ((a_probe_count == b_probe_count) && (a_probe_count == match_count))
return 1;
if (match_count) {
printf("ERROR: devices with overlapping probes: ");
printf(a->path, a->path_a, a->path_b);
printf(b->path, b->path_a, b->path_b);
printf("\n");
exit(1);
}
return 0;
}
/*
* Match device nodes from base and override tree to see if they are the same
* node.
*/
static int device_match(struct device *a, struct device *b)
{
return ((a->path_a == b->path_a) &&
(a->path_b == b->path_b) &&
(a->bustype == b->bustype) &&
(a->chip_instance->chip ==
b->chip_instance->chip));
}
/*
* Match resource nodes from base and override tree to see if they are the same
* node.
*/
static int res_match(struct resource *a, struct resource *b)
{
return ((a->type == b->type) &&
(a->index == b->index));
}
/*
* Add resource to device. If resource is already present, then update its base
* and index. If not, then add a new resource to the device.
*/
static void update_resource(struct device *dev, struct resource *res)
{
struct resource *base_res = dev->res;
while (base_res) {
if (res_match(base_res, res)) {
base_res->base = res->base;
return;
}
base_res = base_res->next;
}
new_resource(dev, res->type, res->index, res->base);
}
/*
* Add register to chip instance. If register is already present, then update
* its value. If not, then add a new register to the chip instance.
*/
sconfig: Allow to link devices to other device's drivers Rarely, the driver of one device needs to know about another device that can be anywhere in the device hierarchy. Current applications boil down to EEPROMs that store information that is consumed by some code (e.g. MAC address). The idea is to give device nodes in the `devicetree.cb` an alias that can later be used to link it to a device driver's `config` structure. The driver has to declare a field of type `struct device *`, e.g. struct some_chip_driver_config { DEVTREE_CONST struct device *needed_eeprom; }; In the devicetree, the referenced device gets an alias, e.g. device i2c 0x50 alias my_eeprom on end The author of the devicetree is free to choose any alias name that is unique in the devicetree. Later, when configuring the driver the alias can be used to link the device with the field of a driver's config: chip some/chip/driver use my_eeprom as needed_eeprom end Override devices can add an alias if it does not exist, but cannot change the alias for a device that already exists. Alias names are checked for conflicts both in the base tree and in the override tree. References are resolved after the tree is parsed so aliases and references do not need to be in a specific order in the tree. Change-Id: I058a319f9b968924fbef9485a96c9e3f900a3ee8 Signed-off-by: Nico Huber <nico.huber@secunet.com> Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/35456 Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2020-06-03 19:20:07 +02:00
static void update_register(struct reg **const head, struct reg *reg)
{
sconfig: Allow to link devices to other device's drivers Rarely, the driver of one device needs to know about another device that can be anywhere in the device hierarchy. Current applications boil down to EEPROMs that store information that is consumed by some code (e.g. MAC address). The idea is to give device nodes in the `devicetree.cb` an alias that can later be used to link it to a device driver's `config` structure. The driver has to declare a field of type `struct device *`, e.g. struct some_chip_driver_config { DEVTREE_CONST struct device *needed_eeprom; }; In the devicetree, the referenced device gets an alias, e.g. device i2c 0x50 alias my_eeprom on end The author of the devicetree is free to choose any alias name that is unique in the devicetree. Later, when configuring the driver the alias can be used to link the device with the field of a driver's config: chip some/chip/driver use my_eeprom as needed_eeprom end Override devices can add an alias if it does not exist, but cannot change the alias for a device that already exists. Alias names are checked for conflicts both in the base tree and in the override tree. References are resolved after the tree is parsed so aliases and references do not need to be in a specific order in the tree. Change-Id: I058a319f9b968924fbef9485a96c9e3f900a3ee8 Signed-off-by: Nico Huber <nico.huber@secunet.com> Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/35456 Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2020-06-03 19:20:07 +02:00
struct reg *base_reg = *head;
while (base_reg) {
if (!strcmp(base_reg->key, reg->key)) {
base_reg->value = reg->value;
return;
}
base_reg = base_reg->next;
}
sconfig: Allow to link devices to other device's drivers Rarely, the driver of one device needs to know about another device that can be anywhere in the device hierarchy. Current applications boil down to EEPROMs that store information that is consumed by some code (e.g. MAC address). The idea is to give device nodes in the `devicetree.cb` an alias that can later be used to link it to a device driver's `config` structure. The driver has to declare a field of type `struct device *`, e.g. struct some_chip_driver_config { DEVTREE_CONST struct device *needed_eeprom; }; In the devicetree, the referenced device gets an alias, e.g. device i2c 0x50 alias my_eeprom on end The author of the devicetree is free to choose any alias name that is unique in the devicetree. Later, when configuring the driver the alias can be used to link the device with the field of a driver's config: chip some/chip/driver use my_eeprom as needed_eeprom end Override devices can add an alias if it does not exist, but cannot change the alias for a device that already exists. Alias names are checked for conflicts both in the base tree and in the override tree. References are resolved after the tree is parsed so aliases and references do not need to be in a specific order in the tree. Change-Id: I058a319f9b968924fbef9485a96c9e3f900a3ee8 Signed-off-by: Nico Huber <nico.huber@secunet.com> Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/35456 Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2020-06-03 19:20:07 +02:00
add_reg(head, reg->key, reg->value);
}
static void override_devicetree(struct bus *base_parent,
struct bus *override_parent);
/*
* Update the base device properties using the properties of override device. In
* addition to that, call override_devicetree for all the buses under the
* override device.
*
* Override Rules:
* +--------------------+--------------------------------------------+
* | | |
* |struct device member| Rule |
* | | |
* +-----------------------------------------------------------------+
* | | |
* | id | Unchanged. This is used to generate device |
* | | structure name in static.c. So, no need to |
* | | override. |
* | | |
* +-----------------------------------------------------------------+
* | | |
* | enabled | Copy enabled state from override device. |
* | | This allows variants to override device |
* | | state. |
* | | |
* +-----------------------------------------------------------------+
* | | |
* | subsystem_vendor | Copy from override device only if any one |
* | subsystem_device | of the ids is non-zero. |
* | | |
* +-----------------------------------------------------------------+
* | | |
* | inherit_subsystem | Copy from override device only if it is |
* | | non-zero. This allows variant to only |
* | | enable inherit flag for a device. |
* | | |
* +-----------------------------------------------------------------+
* | | |
* | path | Unchanged since these are same for both |
* | path_a | base and override device (Used for |
* | path_b | matching devices). |
* | | |
* +-----------------------------------------------------------------+
* | | |
* | bustype | Unchanged since this is same for both base |
* | | and override device (User for matching |
* | | devices). |
* | | |
* +-----------------------------------------------------------------+
* | | |
* | pci_irq_info | Unchanged. |
* | | |
* +-----------------------------------------------------------------+
* | | |
* | parent | Unchanged. This is meaningful only within |
* | sibling | the parse tree, hence not being copied. |
* | | |
* +-----------------------------------------------------------------+
* | | |
* | res | Each resource that is present in override |
* | | device is copied over to base device: |
* | | 1. If resource of same type and index is |
* | | present in base device, then base of |
* | | the resource is copied. |
* | | 2. If not, then a new resource is allocated|
* | | under the base device using type, index |
* | | and base from override res. |
* | | |
* +-----------------------------------------------------------------+
* | | |
sconfig: Allow to link devices to other device's drivers Rarely, the driver of one device needs to know about another device that can be anywhere in the device hierarchy. Current applications boil down to EEPROMs that store information that is consumed by some code (e.g. MAC address). The idea is to give device nodes in the `devicetree.cb` an alias that can later be used to link it to a device driver's `config` structure. The driver has to declare a field of type `struct device *`, e.g. struct some_chip_driver_config { DEVTREE_CONST struct device *needed_eeprom; }; In the devicetree, the referenced device gets an alias, e.g. device i2c 0x50 alias my_eeprom on end The author of the devicetree is free to choose any alias name that is unique in the devicetree. Later, when configuring the driver the alias can be used to link the device with the field of a driver's config: chip some/chip/driver use my_eeprom as needed_eeprom end Override devices can add an alias if it does not exist, but cannot change the alias for a device that already exists. Alias names are checked for conflicts both in the base tree and in the override tree. References are resolved after the tree is parsed so aliases and references do not need to be in a specific order in the tree. Change-Id: I058a319f9b968924fbef9485a96c9e3f900a3ee8 Signed-off-by: Nico Huber <nico.huber@secunet.com> Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/35456 Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2020-06-03 19:20:07 +02:00
* | ref | Each reference that is present in override |
* | | device is copied over to base device with |
* | | the same rules as registers. |
* | | |
* +-----------------------------------------------------------------+
* | | |
* | alias | Base device alias is copied to override. |
* | | Override devices cannot change/remove an |
* | | existing alias, but they can add an alias |
* | | if one does not exist. |
* | | |
* +-----------------------------------------------------------------+
* | | |
* | smbios_slot info | Copy SMBIOS slot information from override.|
* | | This allows variants to override PCI(e) |
* | | slot information in SMBIOS tables. |
* | | |
* +-----------------------------------------------------------------+
* | | |
* | chip_instance | Each register of chip_instance is copied |
* | | over from override device to base device: |
* | | 1. If register with same key is present in |
* | | base device, then value of the register |
* | | is copied. |
* | | 2. If not, then a new register is allocated|
* | | under the base chip_instance using key |
* | | and value from override register. |
* | | |
* +-----------------------------------------------------------------+
* | | |
* | bus | Recursively call override_devicetree on |
* | last_bus | each bus of override device. It is assumed |
* | | that bus with id X under base device |
* | | to bus with id X under override device. If |
* | | override device has more buses than base |
* | | device, then new buses are allocated under |
* | | base device. |
* | | |
* +-----------------------------------------------------------------+
*/
static void update_device(struct device *base_dev, struct device *override_dev)
{
/*
* Copy the enabled state of override device to base device. This allows
* override tree to enable or disable a particular device.
*/
base_dev->enabled = override_dev->enabled;
/*
* Copy the hidden state of override device to base device. This allows
* override tree to hide or unhide a particular device.
*/
base_dev->hidden = override_dev->hidden;
/*
* Copy subsystem vendor and device ids from override device to base
* device only if the ids are non-zero in override device. Else, honor
* the values in base device.
*/
if (override_dev->subsystem_vendor ||
override_dev->subsystem_device) {
base_dev->subsystem_vendor = override_dev->subsystem_vendor;
base_dev->subsystem_device = override_dev->subsystem_device;
}
/*
* Copy value of inherity_subsystem from override device to base device
* only if it is non-zero in override device. This allows override
* tree to only enable inhert flag for a device.
*/
if (override_dev->inherit_subsystem)
base_dev->inherit_subsystem = override_dev->inherit_subsystem;
/*
* Copy resources of override device to base device.
* 1. If resource is already present in base device, then index and base
* of the resource will be copied over.
* 2. If resource is not already present in base device, a new resource
* will be allocated.
*/
struct resource *res = override_dev->res;
while (res) {
update_resource(base_dev, res);
res = res->next;
}
/*
* Copy registers of override chip instance to base chip instance.
* 1. If register key is already present in base chip instance, then
* value for the register is copied over.
* 2. If register key is not already present in base chip instance, then
* a new register will be allocated.
*/
struct reg *reg = override_dev->chip_instance->reg;
while (reg) {
sconfig: Allow to link devices to other device's drivers Rarely, the driver of one device needs to know about another device that can be anywhere in the device hierarchy. Current applications boil down to EEPROMs that store information that is consumed by some code (e.g. MAC address). The idea is to give device nodes in the `devicetree.cb` an alias that can later be used to link it to a device driver's `config` structure. The driver has to declare a field of type `struct device *`, e.g. struct some_chip_driver_config { DEVTREE_CONST struct device *needed_eeprom; }; In the devicetree, the referenced device gets an alias, e.g. device i2c 0x50 alias my_eeprom on end The author of the devicetree is free to choose any alias name that is unique in the devicetree. Later, when configuring the driver the alias can be used to link the device with the field of a driver's config: chip some/chip/driver use my_eeprom as needed_eeprom end Override devices can add an alias if it does not exist, but cannot change the alias for a device that already exists. Alias names are checked for conflicts both in the base tree and in the override tree. References are resolved after the tree is parsed so aliases and references do not need to be in a specific order in the tree. Change-Id: I058a319f9b968924fbef9485a96c9e3f900a3ee8 Signed-off-by: Nico Huber <nico.huber@secunet.com> Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/35456 Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2020-06-03 19:20:07 +02:00
update_register(&base_dev->chip_instance->reg, reg);
reg = reg->next;
}
sconfig: Allow to link devices to other device's drivers Rarely, the driver of one device needs to know about another device that can be anywhere in the device hierarchy. Current applications boil down to EEPROMs that store information that is consumed by some code (e.g. MAC address). The idea is to give device nodes in the `devicetree.cb` an alias that can later be used to link it to a device driver's `config` structure. The driver has to declare a field of type `struct device *`, e.g. struct some_chip_driver_config { DEVTREE_CONST struct device *needed_eeprom; }; In the devicetree, the referenced device gets an alias, e.g. device i2c 0x50 alias my_eeprom on end The author of the devicetree is free to choose any alias name that is unique in the devicetree. Later, when configuring the driver the alias can be used to link the device with the field of a driver's config: chip some/chip/driver use my_eeprom as needed_eeprom end Override devices can add an alias if it does not exist, but cannot change the alias for a device that already exists. Alias names are checked for conflicts both in the base tree and in the override tree. References are resolved after the tree is parsed so aliases and references do not need to be in a specific order in the tree. Change-Id: I058a319f9b968924fbef9485a96c9e3f900a3ee8 Signed-off-by: Nico Huber <nico.huber@secunet.com> Signed-off-by: Duncan Laurie <dlaurie@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/35456 Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2020-06-03 19:20:07 +02:00
/* Copy references just as with registers. */
reg = override_dev->chip_instance->ref;
while (reg) {
update_register(&base_dev->chip_instance->ref, reg);
reg = reg->next;
}
/* Check for alias name conflicts. */
if (override_dev->alias && find_alias(&base_root_dev, override_dev->alias)) {
printf("ERROR: alias already exists: %s\n", override_dev->alias);
exit(1);
}
/*
* Copy alias from base device.
*
* Override devices cannot change/remove an existing alias,
* but they can add an alias to a device if one does not exist yet.
*/
if (base_dev->alias)
override_dev->alias = base_dev->alias;
else
base_dev->alias = override_dev->alias;
sconfig: Add support for firmware configuration This change adds support to sconfig for generating the firmware configuration field and option definitions in devicetree.cb. In addition these fields and options can be used to probe for a device and have that device be disabled if it is not found at boot time. New tokens: fw_config: top level token, table can be defined before chips field: define field in the mask with the start and end bits option: define option in a field with the value of the field probe: indicate that a device should probe by field and option Example: fw_config field FEATURE 0 0 option DISABLE 0 option ENABLE 1 end end chip drivers/generic/feature device generic 0 on probe FEATURE ENABLE end end Variants can add new fields and add new options to existing fields in overridetree.cb but cannot redefine an existing option. Devices can have multiple probe tokens, and the device will be considered to be found if any of them return true. The output from defining this field are: 1) the various fields and options will be added as macro constants to static.h and can be used by fw_config for probing. 2) the probe entries will result in a list of fields/options to probe that is added to the resulting struct device and handled by coreboot. BUG=b:147462631 Signed-off-by: Duncan Laurie <dlaurie@google.com> Change-Id: I8aea63e577d933aea09e0d0b09470929cc96e0de Reviewed-on: https://review.coreboot.org/c/coreboot/+/41440 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2020-05-16 00:39:08 +02:00
/*
* Use probe list from override device in place of base device, in order
* to allow an override to remove a probe from the base device.
*/
base_dev->probe = override_dev->probe;
/* Copy SMBIOS slot information from base device */
base_dev->smbios_slot_type = override_dev->smbios_slot_type;
base_dev->smbios_slot_length = override_dev->smbios_slot_length;
base_dev->smbios_slot_data_width = override_dev->smbios_slot_data_width;
base_dev->smbios_slot_designation = override_dev->smbios_slot_designation;
util/sconfig: Drop use of ref_count for chip_instance chip_instance structure currently uses a ref_count to determine how many devices hold reference to that instance. If the count drops to zero, then it is assumed that the chip instance is a duplicate in override tree and has a similar instance that is already overriden in base device tree. ref_count is currently decremented whenever a device in override tree matches the one in base device tree and the registers from the override tree instance are copied over to the base tree instance. On the other hand, if a device in override tree does not match any device in base tree under a given parent, then the device is added to base tree and all the devices in its subtree that hold pointers to its parent chip instance are updated to point to the parent's chip instance in base tree. This is done as part of update_chip_pointers. However, there are a couple of issues that this suffers from: a) If a device is present only in override tree and it does not have its own chip (i.e. pointing to parent's chip instance), then it results in sconfig emiiting parent's chip instance (which can be the SoC chip instance) in static.c even though it is unused. This is because update_chip_pointers() does not call delete_chip_instance() before reassigning the chip instance pointer. b) If a device is added under root device only in the override tree and it does not have its own chip instance (i.e. uses SoC chip instance), then it results in sconfig emitting a copy of the SoC chip instance and setting that as chip_ops for this new device in the override tree. In order to fix the above issues, this change drops the ref_count field from chip_instance structure and instead adds a forwarding pointer `base_chip_instance`. This is setup as per the following rules: 1. If the instance belongs to base devicetree, base_chip_instance is set to NULL. 2. If the instance belongs to override tree, then it is set to its corresponding chip instance in base tree (if present), else set to NULL. State of base_chip_instance is then used when emitting chips and devices using the following rules: 1. If a chip_instance has non-NULL base_chip_instance, then that chip instance is not emitted to static.c 2. When emitting chip_ops for a device, base_chip_instance is used to determine the correct chip instance name to emit. BUG=b:155549176 TEST=Verified that the static.c file generated for base/override tree combination is correct when new devices without chips are added only to override tree. Change-Id: Idbb5b34f49bf874da3f30ebb6a6a0e2d8d091fe5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/41007 Reviewed-by: Aaron Durbin <adurbin@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2020-05-03 01:05:29 +02:00
/*
* Update base_chip_instance member in chip instance of override tree to forward it to
* the chip instance in base tree.
*/
override_dev->chip_instance->base_chip_instance = get_chip_instance(base_dev);
util/sconfig: Drop use of ref_count for chip_instance chip_instance structure currently uses a ref_count to determine how many devices hold reference to that instance. If the count drops to zero, then it is assumed that the chip instance is a duplicate in override tree and has a similar instance that is already overriden in base device tree. ref_count is currently decremented whenever a device in override tree matches the one in base device tree and the registers from the override tree instance are copied over to the base tree instance. On the other hand, if a device in override tree does not match any device in base tree under a given parent, then the device is added to base tree and all the devices in its subtree that hold pointers to its parent chip instance are updated to point to the parent's chip instance in base tree. This is done as part of update_chip_pointers. However, there are a couple of issues that this suffers from: a) If a device is present only in override tree and it does not have its own chip (i.e. pointing to parent's chip instance), then it results in sconfig emiiting parent's chip instance (which can be the SoC chip instance) in static.c even though it is unused. This is because update_chip_pointers() does not call delete_chip_instance() before reassigning the chip instance pointer. b) If a device is added under root device only in the override tree and it does not have its own chip instance (i.e. uses SoC chip instance), then it results in sconfig emitting a copy of the SoC chip instance and setting that as chip_ops for this new device in the override tree. In order to fix the above issues, this change drops the ref_count field from chip_instance structure and instead adds a forwarding pointer `base_chip_instance`. This is setup as per the following rules: 1. If the instance belongs to base devicetree, base_chip_instance is set to NULL. 2. If the instance belongs to override tree, then it is set to its corresponding chip instance in base tree (if present), else set to NULL. State of base_chip_instance is then used when emitting chips and devices using the following rules: 1. If a chip_instance has non-NULL base_chip_instance, then that chip instance is not emitted to static.c 2. When emitting chip_ops for a device, base_chip_instance is used to determine the correct chip instance name to emit. BUG=b:155549176 TEST=Verified that the static.c file generated for base/override tree combination is correct when new devices without chips are added only to override tree. Change-Id: Idbb5b34f49bf874da3f30ebb6a6a0e2d8d091fe5 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/41007 Reviewed-by: Aaron Durbin <adurbin@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2020-05-03 01:05:29 +02:00
sconfig: Allow to specify device operations Currently we only have runtime mechanisms to assign device operations to a node in our devicetree (with one exception: the root device). The most common method is to map PCI IDs to the device operations with a `struct pci_driver`. Another accustomed way is to let a chip driver assign them. For very common drivers, e.g. those in soc/intel/common/blocks/, the PCI ID lists grew very large and are incredibly error-prone. Often, IDs are missing and sometimes IDs are added almost mechanically without checking the code for compatibility. Maintaining these lists in a central place also reduces flexibility. Now, for onboard devices it is actually unnecessary to assign the device operations at runtime. We already know exactly what operations should be assigned. And since we are using chipset devicetrees, we have a perfect place to put that information. This patch adds a simple mechanism to `sconfig`. It allows us to speci- fy operations per device, e.g. device pci 00.0 alias system_agent on ops system_agent_ops end The operations are given as a C identifier. In this example, we simply assume that a global `struct device_operations system_agent_ops` exists. Change-Id: I2833d2f2450fde3206c33393f58b86fd4280b566 Signed-off-by: Nico Huber <nico.h@gmx.de> Reviewed-on: https://review.coreboot.org/c/coreboot/+/66483 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
2022-08-06 19:02:59 +02:00
/* Allow to override the ops of a device */
if (override_dev->ops_id)
base_dev->ops_id = override_dev->ops_id;
/*
* Now that the device properties are all copied over, look at each bus
* of the override device and run override_devicetree in a recursive
* manner. The assumption here is that first bus of override device
* corresponds to first bus of base device and so on. If base device has
* lesser buses than override tree, then new buses are allocated for it.
*/
struct bus *override_bus = override_dev->bus;
struct bus *base_bus = base_dev->bus;
while (override_bus) {
/*
* If we have more buses in override tree device, then allocate
* a new bus for the base tree device as well.
*/
if (!base_bus) {
alloc_bus(base_dev);
base_bus = base_dev->last_bus;
}
override_devicetree(base_dev->bus, override_dev->bus);
override_bus = override_bus->next_bus;
base_bus = base_bus->next_bus;
}
}
/*
* Perform copy of device and properties from override parent to base parent.
* This function walks through the override tree in a depth-first manner
* performing following actions:
* 1. If matching device is found in base tree, then copy the properties of
* override device to base tree device. Call override_devicetree recursively on
* the bus of override device.
* 2. If matching device is not found in base tree, then set override tree
* device as new child of base_parent and update the chip pointers in override
* device subtree to ensure the nodes do not point to override tree chip
* instance.
*/
static void override_devicetree(struct bus *base_parent,
struct bus *override_parent)
{
struct device *base_child;
struct device *override_child = override_parent->children;
struct device *next_child;
while (override_child) {
/* Look for a matching device in base tree. */
for (base_child = base_parent->children;
base_child; base_child = base_child->sibling) {
if (!device_match(base_child, override_child))
continue;
/* If base device has no probe statement, nothing else to compare. */
if (base_child->probe == NULL)
break;
/*
* If base device has probe statements, ensure that all probe conditions
* match for base and override device.
*/
if (device_probes_match(base_child, override_child))
break;
}
next_child = override_child->sibling;
/*
* If matching device is found, copy properties of
* override_child to base_child.
*/
if (base_child)
update_device(base_child, override_child);
else {
/*
* If matching device is not found, set override_child
* as a new child of base_parent.
*/
set_new_child(base_parent, override_child);
}
override_child = next_child;
}
}
static void parse_override_devicetree(const char *file, struct device *dev)
{
parse_devicetree(file, dev->bus);
if (!dev_has_children(dev)) {
fprintf(stderr, "ERROR: Override tree needs at least one device!\n");
exit(1);
}
override_devicetree(&base_root_bus, dev->bus);
}
static void generate_outputh(FILE *f, const char *fw_conf_header, const char *device_header)
{
fprintf(f, "#ifndef __STATIC_DEVICE_TREE_H\n");
fprintf(f, "#define __STATIC_DEVICE_TREE_H\n\n");
fprintf(f, "#include <%s>\n", fw_conf_header);
fprintf(f, "#include <%s>\n\n", device_header);
fprintf(f, "\n#endif /* __STATIC_DEVICE_TREE_H */\n");
}
static void generate_outputc(FILE *f, const char *static_header)
{
fprintf(f, "#include <boot/coreboot_tables.h>\n");
fprintf(f, "#include <device/device.h>\n");
fprintf(f, "#include <device/pci.h>\n");
fprintf(f, "#include <fw_config.h>\n");
fprintf(f, "#include <identity.h>\n");
fprintf(f, "#include <%s>\n", static_header);
emit_chip_headers(f, chip_header.next);
sconfig: Allow to specify device operations Currently we only have runtime mechanisms to assign device operations to a node in our devicetree (with one exception: the root device). The most common method is to map PCI IDs to the device operations with a `struct pci_driver`. Another accustomed way is to let a chip driver assign them. For very common drivers, e.g. those in soc/intel/common/blocks/, the PCI ID lists grew very large and are incredibly error-prone. Often, IDs are missing and sometimes IDs are added almost mechanically without checking the code for compatibility. Maintaining these lists in a central place also reduces flexibility. Now, for onboard devices it is actually unnecessary to assign the device operations at runtime. We already know exactly what operations should be assigned. And since we are using chipset devicetrees, we have a perfect place to put that information. This patch adds a simple mechanism to `sconfig`. It allows us to speci- fy operations per device, e.g. device pci 00.0 alias system_agent on ops system_agent_ops end The operations are given as a C identifier. In this example, we simply assume that a global `struct device_operations system_agent_ops` exists. Change-Id: I2833d2f2450fde3206c33393f58b86fd4280b566 Signed-off-by: Nico Huber <nico.h@gmx.de> Reviewed-on: https://review.coreboot.org/c/coreboot/+/66483 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
2022-08-06 19:02:59 +02:00
emit_identifiers(f, "struct device_operations", device_operations);
fprintf(f, "\n#define STORAGE static __maybe_unused DEVTREE_CONST\n\n");
walk_device_tree(NULL, NULL, &base_root_dev, inherit_subsystem_ids);
fprintf(f, "\n/* pass 0 */\n");
walk_device_tree(f, NULL, &base_root_dev, pass0);
walk_device_tree(NULL, NULL, &base_root_dev, update_references);
fprintf(f, "\n/* chip configs */\n");
emit_chip_configs(f);
fprintf(f, "\n/* pass 1 */\n");
walk_device_tree(f, NULL, &base_root_dev, pass1);
}
static void generate_outputd(FILE *gen, FILE *dev)
{
fprintf(dev, "#ifndef __STATIC_DEVICES_H\n");
fprintf(dev, "#define __STATIC_DEVICES_H\n\n");
fprintf(dev, "#include <device/device.h>\n\n");
fprintf(dev, "/* expose_device_names */\n");
walk_device_tree(gen, dev, &base_root_dev, expose_device_names);
fprintf(dev, "\n#endif /* __STATIC_DEVICE_NAMES_H */\n");
}
static void generate_outputf(FILE *f)
{
fprintf(f, "#ifndef __STATIC_FW_CONFIG_H\n");
fprintf(f, "#define __STATIC_FW_CONFIG_H\n\n");
emit_fw_config(f);
fprintf(f, "\n#endif /* __STATIC_FW_CONFIG_H */\n");
}
int main(int argc, char **argv)
{
static const struct option long_options[] = {
{ "mainboard_devtree", 1, NULL, 'm' },
{ "override_devtree", 1, NULL, 'o' },
{ "chipset_devtree", 1, NULL, 'p' },
{ "output_c", 1, NULL, 'c' },
{ "output_h", 1, NULL, 'r' },
{ "output_d", 1, NULL, 'd' },
{ "output_f", 1, NULL, 'f' },
{ "help", 1, NULL, 'h' },
{ }
};
const char *override_devtree = NULL;
const char *base_devtree = NULL;
const char *chipset_devtree = NULL;
const char *outputc = NULL;
const char *outputh = NULL;
const char *outputd = NULL;
const char *outputf = NULL;
int opt, option_index;
while ((opt = getopt_long(argc, argv, "m:o:p:c:r:d:f:h", long_options,
&option_index)) != EOF) {
switch (opt) {
case 'm':
base_devtree = strdup(optarg);
break;
case 'o':
override_devtree = strdup(optarg);
break;
case 'p':
chipset_devtree = strdup(optarg);
break;
case 'c':
outputc = strdup(optarg);
break;
case 'r':
outputh = strdup(optarg);
break;
case 'd':
outputd = strdup(optarg);
break;
case 'f':
outputf = strdup(optarg);
break;
case 'h':
default:
usage();
}
}
if (!base_devtree || !outputc || !outputh || !outputd || !outputf)
usage();
if (chipset_devtree) {
/* Use the chipset devicetree as the base, then override
with the mainboard "base" devicetree. */
parse_devicetree(chipset_devtree, &base_root_bus);
parse_override_devicetree(base_devtree, &chipset_root_dev);
} else {
parse_devicetree(base_devtree, &base_root_bus);
}
if (override_devtree)
parse_override_devicetree(override_devtree, &override_root_dev);
util/sconfig: Add support for discontiguous FW_CONFIG fields Sooner or later, some board was going to need extra FW_CONFIG bits for a field that was already in production, so this patch adds support for adding extra (unused) bits to a field. The extra are appended via a syntax like: `field FIELD_NAME START0 END0 | START1 END1 | START2 END2 ...` and the suffixed bits are all treated as if they are contiguous when defining option values. BUG=b:185190978 TEST=Modified volteer fw_config to the following: field AUDIO 8 10 | 29 29 | 31 31 option NONE 0 option MAX98357_ALC5682I_I2S 1 option MAX98373_ALC5682I_I2S 2 option MAX98373_ALC5682_SNDW 3 option MAX98373_ALC5682I_I2S_UP4 4 option MAX98360_ALC5682I_I2S 5 option RT1011_ALC5682I_I2S 6 option AUDIO_FOO 7 option AUDIO_BAR 8 option AUDIO_QUUX 9 option AUDIO_BLAH1 10 option AUDIO_BLAH2 15 option AUDIO_BLAH3 16 option AUDIO_BLAH4 31 end which yielded (in static_fw_config.h): FW_CONFIG_FIELD_AUDIO_MASK 0xa0000700 FW_CONFIG_FIELD_AUDIO_OPTION_NONE_VALUE 0x0 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98357_ALC5682I_I2S_VALUE 0x100 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_VALUE 0x200 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682_SNDW_VALUE 0x300 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98373_ALC5682I_I2S_UP4_VALUE 0x400 FW_CONFIG_FIELD_AUDIO_OPTION_MAX98360_ALC5682I_I2S_VALUE 0x500 FW_CONFIG_FIELD_AUDIO_OPTION_RT1011_ALC5682I_I2S_VALUE 0x600 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_FOO_VALUE 0x700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BAR_VALUE 0x20000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_QUUX_VALUE 0x20000100 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH1_VALUE 0x20000200 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH2_VALUE 0x20000700 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH3_VALUE 0x80000000 FW_CONFIG_FIELD_AUDIO_OPTION_AUDIO_BLAH4_VALUE 0xa0000700 Change-Id: I5ed76706347ee9642198efc77139abdc3af1b8a6 Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52747 Reviewed-by: Duncan Laurie <duncan@iceblink.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28 22:03:07 +02:00
FILE *autogen = fopen(outputc, "w");
if (!autogen) {
fprintf(stderr, "Could not open file '%s' for writing: ",
outputc);
perror(NULL);
exit(1);
}
FILE *autohead = fopen(outputh, "w");
if (!autohead) {
fprintf(stderr, "Could not open file '%s' for writing: ", outputh);
perror(NULL);
fclose(autogen);
exit(1);
}
FILE *autodev = fopen(outputd, "w");
if (!autodev) {
fprintf(stderr, "Could not open file '%s' for writing: ", outputd);
perror(NULL);
fclose(autogen);
fclose(autohead);
exit(1);
}
FILE *autofwconf = fopen(outputf, "w");
if (!autofwconf) {
fprintf(stderr, "Could not open file '%s' for writing: ", outputf);
perror(NULL);
fclose(autogen);
fclose(autohead);
fclose(autodev);
exit(1);
}
char *f = strdup(outputf);
assert(f);
char *d = strdup(outputd);
assert(d);
char *h = strdup(outputh);
assert(h);
const char *fw_conf_header = basename(f);
const char *device_header = basename(d);
const char *static_header = basename(h);
generate_outputh(autohead, fw_conf_header, device_header);
generate_outputc(autogen, static_header);
generate_outputd(autogen, autodev);
generate_outputf(autofwconf);
fclose(autohead);
fclose(autogen);
fclose(autodev);
fclose(autofwconf);
free(f);
free(d);
free(h);
return 0;
}