diff options
author | Kyösti Mälkki <kyosti.malkki@gmail.com> | 2014-06-19 23:12:15 +0300 |
---|---|---|
committer | Kyösti Mälkki <kyosti.malkki@gmail.com> | 2014-07-03 09:47:48 +0200 |
commit | 005028e0a952b00b6184cdddf5905a1637029585 (patch) | |
tree | 75bf8496298c783731502d9444a6a064d3e9203e /src/mainboard/jetway/nf81-t56n-lf/get_bus_conf.c | |
parent | e1b468e1a7cbea55108fb106105612e1f50c9487 (diff) |
AGESA: Add agesawrapper_post_device()
NOTE: The procedure is moved across a collected timestamp
TS_WRITE_TABLES, so the delay of SPI erase/write will be accounted
for in an earlier entry in cbmem -t output.
Change-Id: I0f082e7af1769c8d7d03cdd51fdb5dacbf3402b4
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/6132
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Diffstat (limited to 'src/mainboard/jetway/nf81-t56n-lf/get_bus_conf.c')
-rw-r--r-- | src/mainboard/jetway/nf81-t56n-lf/get_bus_conf.c | 37 |
1 files changed, 0 insertions, 37 deletions
diff --git a/src/mainboard/jetway/nf81-t56n-lf/get_bus_conf.c b/src/mainboard/jetway/nf81-t56n-lf/get_bus_conf.c index 688f9dd929..69a86ab0e8 100644 --- a/src/mainboard/jetway/nf81-t56n-lf/get_bus_conf.c +++ b/src/mainboard/jetway/nf81-t56n-lf/get_bus_conf.c @@ -54,14 +54,9 @@ u32 sbdn_sb800; static u32 get_bus_conf_done = 0; -#if CONFIG_HAVE_ACPI_RESUME -extern u8 acpi_slp_type; -#endif /* CONFIG_HAVE_ACPI_RESUME */ - void get_bus_conf(void) { u32 apicid_base; - u32 status; device_t dev; int i, j; @@ -72,39 +67,7 @@ void get_bus_conf(void) get_bus_conf_done = 1; - /* - * This is the call to AmdInitLate. It is really in the wrong place, - * conceptually, but functionally within the coreboot model, this is the - * best place to make the call. The logically correct place to call - * AmdInitLate is after PCI scan is done, after the decision about S3 - * resume is made, and before the system tables are written into RAM. The - * routine that is - * responsible for writing the tables is "write_tables", called near the - * end of "main". There is no platform specific entry point between the S3 - * resume decision point and the call to "write_tables", and the next - * platform specific entry points are the calls to the ACPI table write - * functions. The first of ose would seem to be the right place, but other - * table write - * functions, e.g. the PIRQ table write function, are called before the - * ACPI tables are written. This routine is called at the beginning of each - * of the write functions called prior to the ACPI write functions, so this - * becomes the best place for this call. - */ -#if CONFIG_HAVE_ACPI_RESUME - if (acpi_slp_type != 3) { - status = agesawrapper_amdinitlate(); - if(status) - printk(BIOS_DEBUG, "agesawrapper_amdinitlate failed: %x \n", status); - status = agesawrapper_amdS3Save(); - if(status) - printk(BIOS_DEBUG, "agesawrapper_amds3save failed: %x \n", status); - } -#else - status = agesawrapper_amdinitlate(); - if(status) - printk(BIOS_DEBUG, "agesawrapper_amdinitlate failed: %x \n", status); -#endif /* CONFIG_HAVE_ACPI_RESUME */ sbdn_sb800 = 0; |