coreboot-kgpe-d16/util/options/build_opt_tbl.c

526 lines
15 KiB
C
Raw Normal View History

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#include <ctype.h>
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
#include <errno.h>
#include <libgen.h>
#include "../../src/include/pc80/mc146818rtc.h"
#include "../../src/include/boot/coreboot_tables.h"
#define CMOS_IMAGE_BUFFER_SIZE 256
#define INPUT_LINE_MAX 256
#define MAX_VALUE_BYTE_LENGTH 64
#define TMPFILE_LEN 256
#define TMPFILE_TEMPLATE "/build_opt_tbl_XXXXXX"
static unsigned char cmos_table[4096];
/* This routine loops through the entried and tests if any of the fields overlap
input entry_start = the memory pointer to the start of the entries.
entry_end = the byte past the entries.
output none
if there is an overlap, the routine exits, other wise it returns.
*/
CMOS: Add set_option and rework get_option. To ease some of my debugging pain on the unichrome, i decided i needed to move FB size selection into cmos, so i could test a size and then reset it to the default after loading this value so that the next reboot uses the (working) default again. This meant implementing set_option in parallel to get_option. get_option was then found to have inversed argument ordering (like outb) and passing char * and then depending on the cmos layout length, which made me feel quite uncomfortable. Since we either have reserved space (which we shouldn't do anything with in these two functions), an enum or a hexadecimal value, unsigned int seemed like the way to go. So all users of get_option now have their arguments inversed and switched from using ints to unsigned ints now. The way get_cmos_value was implemented forced us to not overlap byte and to have multibyte values be byte aligned. This logic is now adapted to do a full uint32_t read (when needed) at any offset and any length up to 32, and the shifting all happens inside an uint32_t as well. set_cmos_value was implemented similarly. Both routines have been extensively tested in a quick separate little program as it is not easy to get this stuff right. build_opt_tbl.c was altered to function correctly within these new parameters. The enum value retrieval has been changed strol(..., NULL, 10) to stroul(..., NULL, 0), so that we not only are able to use unsigned ints now but so that we also interprete hex values correctly. The 32bit limit gets imposed on all entries not marked reserved, an unused "user_data" field that appeared in a lot of cmos.layouts has been changed to reserved as well. Signed-off-by: Luc Verhaegen <libv@skynet.be> Acked-by: Peter Stuge <peter@stuge.se> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4332 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2009-06-03 12:47:19 +02:00
void test_for_entry_overlaps(void *entry_start, void *entry_end)
{
char *cptr;
CMOS: Add set_option and rework get_option. To ease some of my debugging pain on the unichrome, i decided i needed to move FB size selection into cmos, so i could test a size and then reset it to the default after loading this value so that the next reboot uses the (working) default again. This meant implementing set_option in parallel to get_option. get_option was then found to have inversed argument ordering (like outb) and passing char * and then depending on the cmos layout length, which made me feel quite uncomfortable. Since we either have reserved space (which we shouldn't do anything with in these two functions), an enum or a hexadecimal value, unsigned int seemed like the way to go. So all users of get_option now have their arguments inversed and switched from using ints to unsigned ints now. The way get_cmos_value was implemented forced us to not overlap byte and to have multibyte values be byte aligned. This logic is now adapted to do a full uint32_t read (when needed) at any offset and any length up to 32, and the shifting all happens inside an uint32_t as well. set_cmos_value was implemented similarly. Both routines have been extensively tested in a quick separate little program as it is not easy to get this stuff right. build_opt_tbl.c was altered to function correctly within these new parameters. The enum value retrieval has been changed strol(..., NULL, 10) to stroul(..., NULL, 0), so that we not only are able to use unsigned ints now but so that we also interprete hex values correctly. The 32bit limit gets imposed on all entries not marked reserved, an unused "user_data" field that appeared in a lot of cmos.layouts has been changed to reserved as well. Signed-off-by: Luc Verhaegen <libv@skynet.be> Acked-by: Peter Stuge <peter@stuge.se> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4332 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2009-06-03 12:47:19 +02:00
struct cmos_entries *ce = NULL;
unsigned char test[CMOS_IMAGE_BUFFER_SIZE];
CMOS: Add set_option and rework get_option. To ease some of my debugging pain on the unichrome, i decided i needed to move FB size selection into cmos, so i could test a size and then reset it to the default after loading this value so that the next reboot uses the (working) default again. This meant implementing set_option in parallel to get_option. get_option was then found to have inversed argument ordering (like outb) and passing char * and then depending on the cmos layout length, which made me feel quite uncomfortable. Since we either have reserved space (which we shouldn't do anything with in these two functions), an enum or a hexadecimal value, unsigned int seemed like the way to go. So all users of get_option now have their arguments inversed and switched from using ints to unsigned ints now. The way get_cmos_value was implemented forced us to not overlap byte and to have multibyte values be byte aligned. This logic is now adapted to do a full uint32_t read (when needed) at any offset and any length up to 32, and the shifting all happens inside an uint32_t as well. set_cmos_value was implemented similarly. Both routines have been extensively tested in a quick separate little program as it is not easy to get this stuff right. build_opt_tbl.c was altered to function correctly within these new parameters. The enum value retrieval has been changed strol(..., NULL, 10) to stroul(..., NULL, 0), so that we not only are able to use unsigned ints now but so that we also interprete hex values correctly. The 32bit limit gets imposed on all entries not marked reserved, an unused "user_data" field that appeared in a lot of cmos.layouts has been changed to reserved as well. Signed-off-by: Luc Verhaegen <libv@skynet.be> Acked-by: Peter Stuge <peter@stuge.se> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4332 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2009-06-03 12:47:19 +02:00
int i;
memset(test, 0, CMOS_IMAGE_BUFFER_SIZE);
for (cptr = entry_start; cptr < (char *)entry_end; cptr += ce->size) {
ce = (struct cmos_entries *)cptr;
CMOS: Add set_option and rework get_option. To ease some of my debugging pain on the unichrome, i decided i needed to move FB size selection into cmos, so i could test a size and then reset it to the default after loading this value so that the next reboot uses the (working) default again. This meant implementing set_option in parallel to get_option. get_option was then found to have inversed argument ordering (like outb) and passing char * and then depending on the cmos layout length, which made me feel quite uncomfortable. Since we either have reserved space (which we shouldn't do anything with in these two functions), an enum or a hexadecimal value, unsigned int seemed like the way to go. So all users of get_option now have their arguments inversed and switched from using ints to unsigned ints now. The way get_cmos_value was implemented forced us to not overlap byte and to have multibyte values be byte aligned. This logic is now adapted to do a full uint32_t read (when needed) at any offset and any length up to 32, and the shifting all happens inside an uint32_t as well. set_cmos_value was implemented similarly. Both routines have been extensively tested in a quick separate little program as it is not easy to get this stuff right. build_opt_tbl.c was altered to function correctly within these new parameters. The enum value retrieval has been changed strol(..., NULL, 10) to stroul(..., NULL, 0), so that we not only are able to use unsigned ints now but so that we also interprete hex values correctly. The 32bit limit gets imposed on all entries not marked reserved, an unused "user_data" field that appeared in a lot of cmos.layouts has been changed to reserved as well. Signed-off-by: Luc Verhaegen <libv@skynet.be> Acked-by: Peter Stuge <peter@stuge.se> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4332 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2009-06-03 12:47:19 +02:00
/* Only reserved space can be larger than 32bits */
if ((ce->length > 32) && (ce->config != 'r')) {
fprintf(stderr, "Error: Entry %s is longer than"
" 32bits.\n", ce->name);
exit(1);
}
/* test if entry goes past the end of the buffer */
CMOS: Add set_option and rework get_option. To ease some of my debugging pain on the unichrome, i decided i needed to move FB size selection into cmos, so i could test a size and then reset it to the default after loading this value so that the next reboot uses the (working) default again. This meant implementing set_option in parallel to get_option. get_option was then found to have inversed argument ordering (like outb) and passing char * and then depending on the cmos layout length, which made me feel quite uncomfortable. Since we either have reserved space (which we shouldn't do anything with in these two functions), an enum or a hexadecimal value, unsigned int seemed like the way to go. So all users of get_option now have their arguments inversed and switched from using ints to unsigned ints now. The way get_cmos_value was implemented forced us to not overlap byte and to have multibyte values be byte aligned. This logic is now adapted to do a full uint32_t read (when needed) at any offset and any length up to 32, and the shifting all happens inside an uint32_t as well. set_cmos_value was implemented similarly. Both routines have been extensively tested in a quick separate little program as it is not easy to get this stuff right. build_opt_tbl.c was altered to function correctly within these new parameters. The enum value retrieval has been changed strol(..., NULL, 10) to stroul(..., NULL, 0), so that we not only are able to use unsigned ints now but so that we also interprete hex values correctly. The 32bit limit gets imposed on all entries not marked reserved, an unused "user_data" field that appeared in a lot of cmos.layouts has been changed to reserved as well. Signed-off-by: Luc Verhaegen <libv@skynet.be> Acked-by: Peter Stuge <peter@stuge.se> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4332 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2009-06-03 12:47:19 +02:00
if ((ce->bit + ce->length) > (CMOS_IMAGE_BUFFER_SIZE * 8)) {
fprintf(stderr, "Error: Entry %s exceeds CMOS"
" space.\n", ce->name);
exit(1);
}
CMOS: Add set_option and rework get_option. To ease some of my debugging pain on the unichrome, i decided i needed to move FB size selection into cmos, so i could test a size and then reset it to the default after loading this value so that the next reboot uses the (working) default again. This meant implementing set_option in parallel to get_option. get_option was then found to have inversed argument ordering (like outb) and passing char * and then depending on the cmos layout length, which made me feel quite uncomfortable. Since we either have reserved space (which we shouldn't do anything with in these two functions), an enum or a hexadecimal value, unsigned int seemed like the way to go. So all users of get_option now have their arguments inversed and switched from using ints to unsigned ints now. The way get_cmos_value was implemented forced us to not overlap byte and to have multibyte values be byte aligned. This logic is now adapted to do a full uint32_t read (when needed) at any offset and any length up to 32, and the shifting all happens inside an uint32_t as well. set_cmos_value was implemented similarly. Both routines have been extensively tested in a quick separate little program as it is not easy to get this stuff right. build_opt_tbl.c was altered to function correctly within these new parameters. The enum value retrieval has been changed strol(..., NULL, 10) to stroul(..., NULL, 0), so that we not only are able to use unsigned ints now but so that we also interprete hex values correctly. The 32bit limit gets imposed on all entries not marked reserved, an unused "user_data" field that appeared in a lot of cmos.layouts has been changed to reserved as well. Signed-off-by: Luc Verhaegen <libv@skynet.be> Acked-by: Peter Stuge <peter@stuge.se> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4332 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2009-06-03 12:47:19 +02:00
/* see whether our bits were marked before */
for (i = ce->bit; i < (ce->bit + ce->length); i++)
if (test[i / 8] & (0x01 << (i % 8))) {
printf("Error: Entry %s overlaps at bit %d.\n",
ce->name, i);
exit(1);
}
CMOS: Add set_option and rework get_option. To ease some of my debugging pain on the unichrome, i decided i needed to move FB size selection into cmos, so i could test a size and then reset it to the default after loading this value so that the next reboot uses the (working) default again. This meant implementing set_option in parallel to get_option. get_option was then found to have inversed argument ordering (like outb) and passing char * and then depending on the cmos layout length, which made me feel quite uncomfortable. Since we either have reserved space (which we shouldn't do anything with in these two functions), an enum or a hexadecimal value, unsigned int seemed like the way to go. So all users of get_option now have their arguments inversed and switched from using ints to unsigned ints now. The way get_cmos_value was implemented forced us to not overlap byte and to have multibyte values be byte aligned. This logic is now adapted to do a full uint32_t read (when needed) at any offset and any length up to 32, and the shifting all happens inside an uint32_t as well. set_cmos_value was implemented similarly. Both routines have been extensively tested in a quick separate little program as it is not easy to get this stuff right. build_opt_tbl.c was altered to function correctly within these new parameters. The enum value retrieval has been changed strol(..., NULL, 10) to stroul(..., NULL, 0), so that we not only are able to use unsigned ints now but so that we also interprete hex values correctly. The 32bit limit gets imposed on all entries not marked reserved, an unused "user_data" field that appeared in a lot of cmos.layouts has been changed to reserved as well. Signed-off-by: Luc Verhaegen <libv@skynet.be> Acked-by: Peter Stuge <peter@stuge.se> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4332 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2009-06-03 12:47:19 +02:00
/* now mark our bits */
for (i = ce->bit; i < (ce->bit + ce->length); i++)
test[i / 8] |= 0x01 << (i % 8);
}
}
/* This routine displays the usage options */
void display_usage(char *name)
{
printf("Usage: %s [--config filename]\n", name);
printf(" [--option filename]\n");
printf(" [--header filename]\n\n");
printf("--config = Build the definitions table from the given file.\n");
printf("--option = Output a C source file with the definitions.\n");
printf("--header = Ouput a C header file with the definitions.\n");
exit(1);
}
static void skip_spaces(char *line, char **ptr)
{
if (!isspace(**ptr)) {
printf("Error missing whitespace in line\n%s\n", line);
exit(1);
}
while(isspace(**ptr)) {
(*ptr)++;
}
return;
}
static unsigned long get_number(char *line, char **ptr, int base)
{
unsigned long value;
char *ptr2;
value = strtoul(*ptr, &ptr2, base);
if (ptr2 == *ptr) {
printf("Error missing digits at: \n%s\n in line:\n%s\n",
*ptr, line);
exit(1);
}
*ptr = ptr2;
return value;
}
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
static int is_ident_nondigit(int c)
{
CMOS: Add set_option and rework get_option. To ease some of my debugging pain on the unichrome, i decided i needed to move FB size selection into cmos, so i could test a size and then reset it to the default after loading this value so that the next reboot uses the (working) default again. This meant implementing set_option in parallel to get_option. get_option was then found to have inversed argument ordering (like outb) and passing char * and then depending on the cmos layout length, which made me feel quite uncomfortable. Since we either have reserved space (which we shouldn't do anything with in these two functions), an enum or a hexadecimal value, unsigned int seemed like the way to go. So all users of get_option now have their arguments inversed and switched from using ints to unsigned ints now. The way get_cmos_value was implemented forced us to not overlap byte and to have multibyte values be byte aligned. This logic is now adapted to do a full uint32_t read (when needed) at any offset and any length up to 32, and the shifting all happens inside an uint32_t as well. set_cmos_value was implemented similarly. Both routines have been extensively tested in a quick separate little program as it is not easy to get this stuff right. build_opt_tbl.c was altered to function correctly within these new parameters. The enum value retrieval has been changed strol(..., NULL, 10) to stroul(..., NULL, 0), so that we not only are able to use unsigned ints now but so that we also interprete hex values correctly. The 32bit limit gets imposed on all entries not marked reserved, an unused "user_data" field that appeared in a lot of cmos.layouts has been changed to reserved as well. Signed-off-by: Luc Verhaegen <libv@skynet.be> Acked-by: Peter Stuge <peter@stuge.se> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4332 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2009-06-03 12:47:19 +02:00
if (((c >= 'A') && (c <='Z')) ||
((c >= 'a') && (c <='z')) ||
(c == '_'))
return 1;
else
return 0;
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
}
static int is_ident(char *str)
{
int result;
int ch;
ch = *str;
result = 0;
if (is_ident_nondigit(ch)) {
do {
str++;
ch = *str;
CMOS: Add set_option and rework get_option. To ease some of my debugging pain on the unichrome, i decided i needed to move FB size selection into cmos, so i could test a size and then reset it to the default after loading this value so that the next reboot uses the (working) default again. This meant implementing set_option in parallel to get_option. get_option was then found to have inversed argument ordering (like outb) and passing char * and then depending on the cmos layout length, which made me feel quite uncomfortable. Since we either have reserved space (which we shouldn't do anything with in these two functions), an enum or a hexadecimal value, unsigned int seemed like the way to go. So all users of get_option now have their arguments inversed and switched from using ints to unsigned ints now. The way get_cmos_value was implemented forced us to not overlap byte and to have multibyte values be byte aligned. This logic is now adapted to do a full uint32_t read (when needed) at any offset and any length up to 32, and the shifting all happens inside an uint32_t as well. set_cmos_value was implemented similarly. Both routines have been extensively tested in a quick separate little program as it is not easy to get this stuff right. build_opt_tbl.c was altered to function correctly within these new parameters. The enum value retrieval has been changed strol(..., NULL, 10) to stroul(..., NULL, 0), so that we not only are able to use unsigned ints now but so that we also interprete hex values correctly. The 32bit limit gets imposed on all entries not marked reserved, an unused "user_data" field that appeared in a lot of cmos.layouts has been changed to reserved as well. Signed-off-by: Luc Verhaegen <libv@skynet.be> Acked-by: Peter Stuge <peter@stuge.se> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4332 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2009-06-03 12:47:19 +02:00
} while(ch && (is_ident_nondigit(ch) || isdigit(ch)));
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
result = (ch == '\0');
}
return result;
}
/* This routine builds the cmos definition table from the cmos layout file
input The input comes from the configuration file which contains two parts
entries and enumerations. Each section is started with the key words
entries and enumerations. Records then follow in their respective
formats.
output The output of this program is the cmos definitions table. It is stored
in the cmos_table array. If this module is called, and the global
table_file has been implimented by the user, the table is also written
to the specified file.
This program exits on and error. It returns a 1 on successful
completion
*/
int main(int argc, char **argv)
{
int i;
char *config=0;
char *option=0;
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
char *header=0;
FILE *fp;
int tmpfile;
char tmpfilename[TMPFILE_LEN];
struct cmos_option_table *ct;
struct cmos_entries *ce;
struct cmos_enums *c_enums, *c_enums_start;
struct cmos_checksum *cs;
char line[INPUT_LINE_MAX];
unsigned char uc;
int entry_mode=0;
int enum_mode=0;
int checksum_mode=0;
long ptr;
int cnt;
char *cptr;
void *entry_start, *entry_end;
int entries_length;
int enum_length;
int len;
char buf[16];
for(i=1;i<argc;i++) {
if(argv[i][0]!='-') {
display_usage(argv[0]);
}
switch(argv[i][1]) {
case '-': /* data is requested from a file */
switch(argv[i][2]) {
case 'c': /* use a configuration file */
if(strcmp(&argv[i][2],"config")) {
display_usage(argv[0]);
}
config=argv[++i];
break;
case 'o': /* use a cmos definitions table file */
if(strcmp(&argv[i][2],"option")) {
display_usage(argv[0]);
}
option=argv[++i];
break;
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
case 'h': /* Output a header file */
if (strcmp(&argv[i][2], "header") != 0) {
display_usage(argv[0]);
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
}
header=argv[++i];
break;
default:
display_usage(argv[0]);
break;
}
break;
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
default:
display_usage(argv[0]);
break;
}
}
/* Has the user specified a configuration file */
if(config) { /* if yes, open it */
if((fp=fopen(config,"r"))==NULL){
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
fprintf(stderr, "Error - Can not open config file %s\n",config);
exit(1); /* exit if it can not be opened */
}
}
else { /* no configuration file specified, so try the default */
if((fp=fopen("cmos.layout","r"))==NULL){
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
fprintf(stderr, "Error - Can not open cmos.layout\n");
exit(1); /* end of no configuration file is found */
}
}
/* type cast a pointer, so we can us the structure */
ct=(struct cmos_option_table*)cmos_table;
/* start the table with the type signature */
ct->tag = LB_TAG_CMOS_OPTION_TABLE;
/* put in the header length */
ct->header_length=sizeof(*ct);
/* Get the entry records */
ce=(struct cmos_entries*)(cmos_table+(ct->header_length));
cptr = (char*)ce;
for(;;){ /* this section loops through the entry records */
if(fgets(line,INPUT_LINE_MAX,fp)==NULL)
break; /* end if no more input */
if(!entry_mode) { /* skip input until the entries key word */
if (strstr(line,"entries") != 0) {
entry_mode=1;
continue;
}
}
else{ /* Test if we are done with entries and starting enumerations */
if (strstr(line,"enumerations") != 0){
entry_mode=0;
enum_mode=1;
break;
}
if (strstr(line, "checksums") != 0) {
enum_mode=0;
checksum_mode=1;
break;
}
}
/* skip commented and blank lines */
if(line[0]=='#') continue;
if(line[strspn(line," ")]=='\n') continue;
/* scan in the input data */
sscanf(line,"%d %d %c %d %s",
&ce->bit,&ce->length,&uc,&ce->config_id,&ce->name[0]);
ce->config=(int)uc;
/* check bit and length ranges */
if(ce->bit>(CMOS_IMAGE_BUFFER_SIZE*8)) {
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
fprintf(stderr, "Error - bit is to big in line \n%s\n",line);
exit(1);
}
if((ce->length>(MAX_VALUE_BYTE_LENGTH*8))&&(uc!='r')) {
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
fprintf(stderr, "Error - Length is to long in line \n%s\n",line);
exit(1);
}
if (!is_ident((char *)ce->name)) {
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
fprintf(stderr,
"Error - Name %s is an invalid identifier in line\n %s\n",
ce->name, line);
exit(1);
}
/* put in the record type */
ce->tag=LB_TAG_OPTION;
/* calculate and save the record length */
len=strlen((char *)ce->name)+1;
/* make the record int aligned */
if(len%4)
len+=(4-(len%4));
ce->size=sizeof(struct cmos_entries)-32+len;
cptr = (char*)ce;
cptr += ce->size; /* increment to the next table position */
ce = (struct cmos_entries*) cptr;
}
/* put the length of the entries into the header section */
entries_length = (cptr - (char *)&cmos_table) - ct->header_length;
/* compute the start of the enumerations section */
entry_start = ((char*)&cmos_table) + ct->header_length;
entry_end = ((char *)entry_start) + entries_length;
c_enums_start = c_enums = (struct cmos_enums*)entry_end;
/* test for overlaps in the entry records */
test_for_entry_overlaps(entry_start, entry_end);
for(;enum_mode;){ /* loop to build the enumerations section */
if(fgets(line,INPUT_LINE_MAX,fp)==NULL)
break; /* go till end of input */
if (strstr(line, "checksums") != 0) {
enum_mode=0;
checksum_mode=1;
break;
}
/* skip commented and blank lines */
if(line[0]=='#') continue;
if(line[strspn(line," ")]=='\n') continue;
/* scan in the data */
for(ptr=0;(line[ptr]==' ')||(line[ptr]=='\t');ptr++);
c_enums->config_id=strtol(&line[ptr],(char**)NULL,10);
for(;(line[ptr]!=' ')&&(line[ptr]!='\t');ptr++);
for(;(line[ptr]==' ')||(line[ptr]=='\t');ptr++);
CMOS: Add set_option and rework get_option. To ease some of my debugging pain on the unichrome, i decided i needed to move FB size selection into cmos, so i could test a size and then reset it to the default after loading this value so that the next reboot uses the (working) default again. This meant implementing set_option in parallel to get_option. get_option was then found to have inversed argument ordering (like outb) and passing char * and then depending on the cmos layout length, which made me feel quite uncomfortable. Since we either have reserved space (which we shouldn't do anything with in these two functions), an enum or a hexadecimal value, unsigned int seemed like the way to go. So all users of get_option now have their arguments inversed and switched from using ints to unsigned ints now. The way get_cmos_value was implemented forced us to not overlap byte and to have multibyte values be byte aligned. This logic is now adapted to do a full uint32_t read (when needed) at any offset and any length up to 32, and the shifting all happens inside an uint32_t as well. set_cmos_value was implemented similarly. Both routines have been extensively tested in a quick separate little program as it is not easy to get this stuff right. build_opt_tbl.c was altered to function correctly within these new parameters. The enum value retrieval has been changed strol(..., NULL, 10) to stroul(..., NULL, 0), so that we not only are able to use unsigned ints now but so that we also interprete hex values correctly. The 32bit limit gets imposed on all entries not marked reserved, an unused "user_data" field that appeared in a lot of cmos.layouts has been changed to reserved as well. Signed-off-by: Luc Verhaegen <libv@skynet.be> Acked-by: Peter Stuge <peter@stuge.se> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4332 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2009-06-03 12:47:19 +02:00
c_enums->value = strtoul(&line[ptr],(char**)NULL,0);
for(;(line[ptr]!=' ')&&(line[ptr]!='\t');ptr++);
for(;(line[ptr]==' ')||(line[ptr]=='\t');ptr++);
for(cnt=0;(line[ptr]!='\n')&&(cnt<31);ptr++,cnt++)
c_enums->text[cnt]=line[ptr];
c_enums->text[cnt]=0;
/* make the record int aligned */
cnt++;
if(cnt%4)
cnt+=4-(cnt%4);
/* store the record length */
c_enums->size=((char *)&c_enums->text[cnt]) - (char *)c_enums;
/* store the record type */
c_enums->tag=LB_TAG_OPTION_ENUM;
/* increment to the next record */
c_enums=(struct cmos_enums*)&c_enums->text[cnt];
}
/* save the enumerations length */
enum_length= (char *)c_enums - (char *)c_enums_start;
ct->size=ct->header_length+enum_length+entries_length;
/* Get the checksum records */
cs=(struct cmos_checksum *)(cmos_table+(ct->size));
cptr = (char*)cs;
for(;checksum_mode;) { /* This section finds the checksums */
char *ptr;
if(fgets(line, INPUT_LINE_MAX,fp)==NULL)
break; /* end if no more input */
/* skip commented and blank lines */
if (line[0]=='#') continue;
if (line[strspn(line, " ")]=='\n') continue;
if (memcmp(line, "checksum", 8) != 0) continue;
/* get the information */
ptr = line + 8;
skip_spaces(line, &ptr);
cs->range_start = get_number(line, &ptr, 10);
skip_spaces(line, &ptr);
cs->range_end = get_number(line, &ptr, 10);
skip_spaces(line, &ptr);
cs->location = get_number(line, &ptr, 10);
/* Make certain there are spaces until the end of the line */
skip_spaces(line, &ptr);
if ((cs->range_start%8) != 0) {
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
fprintf(stderr, "Error - range start is not byte aligned in line\n%s\n", line);
exit(1);
}
if (cs->range_start >= (CMOS_IMAGE_BUFFER_SIZE*8)) {
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
fprintf(stderr, "Error - range start is to big in line\n%s\n", line);
exit(1);
}
if ((cs->range_end%8) != 7) {
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
fprintf(stderr, "Error - range end is not byte aligned in line\n%s\n", line);
exit(1);
}
if ((cs->range_end) >= (CMOS_IMAGE_BUFFER_SIZE*8)) {
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
fprintf(stderr, "Error - range end is to long in line\n%s\n", line);
exit(1);
}
if ((cs->location%8) != 0) {
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
fprintf(stderr, "Error - location is not byte aligned in line\n%s\n", line);
exit(1);
}
if ((cs->location >= (CMOS_IMAGE_BUFFER_SIZE*8)) ||
((cs->location + 16) > (CMOS_IMAGE_BUFFER_SIZE*8)))
{
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
fprintf(stderr, "Error - location is to big in line\n%s\n", line);
exit(1);
}
/* And since we are not ready to be fully general purpose yet.. */
if ((cs->range_start/8) != LB_CKS_RANGE_START) {
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
fprintf(stderr, "Error - Range start(%d) does not match define(%d) in line\n%s\n",
cs->range_start/8, LB_CKS_RANGE_START, line);
exit(1);
}
if ((cs->range_end/8) != LB_CKS_RANGE_END) {
fprintf(stderr, "Error - Range end (%d) does not match define (%d) in line\n%s\n",
(cs->range_end/8), LB_CKS_RANGE_END, line);
exit(1);
}
if ((cs->location/8) != LB_CKS_LOC) {
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
fprintf(stderr, "Error - Location does not match define in line\n%s\n", line);
exit(1);
}
cs->tag = LB_TAG_OPTION_CHECKSUM;
cs->size = sizeof(*cs);
cs->type = CHECKSUM_PCBIOS;
cptr = (char *)cs;
cptr += cs->size;
cs = (struct cmos_checksum *)cptr;
}
ct->size += (cptr - (char *)(cmos_table + ct->size));
fclose(fp);
/* See if we want to output a C source file */
if(option) {
int err=0;
strncpy(tmpfilename, dirname(option), TMPFILE_LEN);
strncat(tmpfilename, TMPFILE_TEMPLATE, TMPFILE_LEN);
tmpfile = mkstemp(tmpfilename);
if(tmpfile == -1) {
perror("Error - Could not create temporary file");
exit(1);
}
if((fp=fdopen(tmpfile,"w"))==NULL){
perror("Error - Could not open temporary file");
unlink(tmpfilename);
exit(1);
}
/* write the header */
if(!fwrite("unsigned char option_table[] = {",1,32,fp)) {
perror("Error - Could not write image file");
fclose(fp);
unlink(tmpfilename);
exit(1);
}
/* write the array values */
for(i=0;i<(ct->size-1);i++) {
if(!(i%10) && !err) err=!fwrite("\n\t",1,2,fp);
sprintf(buf,"0x%02x,",cmos_table[i]);
if(!err) err=!fwrite(buf,1,5,fp);
}
/* write the end */
sprintf(buf,"0x%02x\n",cmos_table[i]);
if(!err) err=!fwrite(buf,1,4,fp);
if(!fwrite("};\n",1,3,fp)) {
perror("Error - Could not write image file");
fclose(fp);
unlink(tmpfilename);
exit(1);
}
fclose(fp);
if (rename(tmpfilename, option)) {
fprintf(stderr, "Error - Could not write %s: ", option);
perror(NULL);
unlink(tmpfilename);
exit(1);
}
}
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
/* See if we also want to output a C header file */
if (header) {
struct cmos_option_table *hdr;
struct lb_record *ptr, *end;
strncpy(tmpfilename, dirname(option), TMPFILE_LEN);
strncat(tmpfilename, TMPFILE_TEMPLATE, TMPFILE_LEN);
tmpfile = mkstemp(tmpfilename);
if(tmpfile == -1) {
perror("Error - Could not create temporary file");
exit(1);
}
fp = fdopen(tmpfile, "w");
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
if (!fp) {
perror("Error - Could not open temporary file");
unlink(tmpfilename);
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
exit(1);
}
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
/* Get the cmos table header */
hdr = (struct cmos_option_table *)cmos_table;
/* Walk through the entry records */
ptr = (struct lb_record *)(cmos_table + hdr->header_length);
end = (struct lb_record *)(cmos_table + hdr->size);
for(;ptr < end; ptr = (struct lb_record *)(((char *)ptr) + ptr->size)) {
if (ptr->tag != LB_TAG_OPTION) {
continue;
}
ce = (struct cmos_entries *)ptr;
if (!is_ident((char *)ce->name)) {
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
fprintf(stderr, "Invalid identifier: %s\n",
ce->name);
fclose(fp);
unlink(tmpfilename);
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
exit(1);
}
fprintf(fp, "#define CMOS_VSTART_%s %d\n",
ce->name, ce->bit);
fprintf(fp, "#define CMOS_VLEN_%s %d\n",
ce->name, ce->length);
}
fclose(fp);
if (rename(tmpfilename, header)) {
fprintf(stderr, "Error - Could not write %s: ", header);
perror(NULL);
unlink(tmpfilename);
exit(1);
}
- Moved hlt() to it's own header. - Reworked pnp superio device support. Now complete superio support is less than 100 lines. - Added support for hard coding resource assignments in Config.lb - Minor bug fixes to romcc - Initial support for catching the x86 processor BIST error codes. I've only seen this trigger once in production during a very suspcious reset but... - added raminit_test to test the code paths in raminit.c for the Opteron - Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED so we can tell what we have really done. - Added generic AGP/IOMMU setting code to x86 - Added an implementation of memmove and removed reserved identifiers from memcpy - Added minimal support for booting on pre b3 stepping K8 cores - Moved the checksum on amd8111 boards because our default location was on top of extended RTC registers - On the Hdama added support for enabling i2c hub so we can get at the temperature sensors. Not that i2c bus was implemented well enough to make that useful. - Redid the Opteron port so we should only need one reset and most of memory initialization is done in cpu_fixup. This is much, much faster. - Attempted to make the VGA IO region assigment work. The code seems to work now... - Redid the error handling in amdk8/raminit.c to distinguish between a bad value and a smbus error, and moved memory clearing out to cpufixup. - Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to setup a legacy keyboard properly. - Reworked the register values for standard hardware, moving the defintions from chip.h into the headers of the initialization routines. This is much saner and is actually implemented. - Made the hdama port an under clockers BIOS. I debuged so many interesting problems. - On amd8111_lpc added setup of architectural/legacy hardware - Enabled PCI error reporting as much as possible. - Enhanded build_opt_tbl to generate a header of the cmos option locations so that romcc compiled code can query the cmos options. - In romcc gracefully handle function names that degenerate into function pointers - Bumped the version to 1.1.6 as we are getting closer to 2.0 TODO finish optimizing the HT links of non dual boards TODO make all Opteron board work again TODO convert all superio devices to use the new helpers TODO convert the via/epia to freebios2 conventions TODO cpu fixup/setup by cpu type git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-03-11 16:01:31 +01:00
}
return(0);
}