summaryrefslogtreecommitdiff
path: root/util/cbfstool/flashmap/valstr.c
diff options
context:
space:
mode:
authorFurquan Shaikh <furquan@google.com>2021-01-12 16:09:43 -0800
committerPatrick Georgi <pgeorgi@google.com>2021-01-18 07:27:43 +0000
commitea3e6b06cc82a64d16616399362d373bfcef8a60 (patch)
treef9c15dd02ba417fe448ac4ba9b6535936176816c /util/cbfstool/flashmap/valstr.c
parent7cc513557d41ea86e4c5ce4a9dba14b879c8b7b2 (diff)
soc/intel/common/pcie: Allow pcie_rp_group table to be non-contiguous
In case of CPU PCIe RPs, the RP numbers might not be contiguous for all the functions in a slot. Example: In ADL, RP1 is 00:06.0, RP2 is 00:01.0 and RP3 is 00:06.2 as per the FSP expectations. Hence, this change updates the defintion of `struct pcie_rp_group` to include a `start` member which indicates the starting PCI function number within the group. All common functions for PCIe RP are accordingly updated to take the `start` member into account. Thus, in the above example, ADL can provide a cpu_rp_table as follows: { { .slot = PCIE_SLOT_6, .start = 0, .count = 1 }, { .slot = PCIE_SLOT_1, .start = 0, .count = 1 }, { .slot = PCIE_SLOT_6, .start = 2, .count = 1 }, } Since start defaults to 0 when uninitialized, current PCH RP group tables don't need to be updated. Change-Id: Idf80a0f29e7c315105f76a7460c8e1e8f9a10d25 Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/49370 Reviewed-by: Nico Huber <nico.h@gmx.de> Reviewed-by: EricR Lai <ericr_lai@compal.corp-partner.google.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Diffstat (limited to 'util/cbfstool/flashmap/valstr.c')
0 files changed, 0 insertions, 0 deletions