summaryrefslogtreecommitdiff
path: root/src/cpu
AgeCommit message (Collapse)Author
2014-02-06mtrr: retry fitting w/o WRCOMB if usage exceeds BIOS allocationAaron Durbin
If the MTRR usage exceeds the BIOS allocation for MTRR usage re-try without the WRCOMB type. Change-Id: Ie70ce84994428ff6700c36310264c3c44d9ed128 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5151 Reviewed-by: Vladimir Serbinenko <phcoder@gmail.com> Tested-by: build bot (Jenkins)
2014-02-03cpu/amd/model_fxx: Add coolnquiet for two new (old) AMD K8 modelsOskar Enoksson
The added CPU's are OSA248CEP5AU and a OSP280 processors. The OSP280 VID/FID numbers have been found by experimentation and extrapolation/guesses from similar models. It has been verified to work fine under Linux (OpenSuse 12.2, kernel 3.4.63-2.44) with four different test-processors. Windows is untested. Change-Id: I3afa1cba5f55c8a78917b3636382af7706a80fee Signed-off-by: Oskar Enoksson <enok@lysator.liu.se> Reviewed-on: http://review.coreboot.org/5095 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Rudolf Marek <r.marek@assembler.cz>
2014-02-01cpu/intel/model_2065x: Add model 20652Vladimir Serbinenko
Found in some X201t. Tested on X201t. Change-Id: I3fc4c3f5b1abf9fe61746ab8f401d1b6ee67f3ea Signed-off-by: Vladimir Serbinenko <phcoder@gmail.com> Reviewed-on: http://review.coreboot.org/5090 Tested-by: build bot (Jenkins) Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
2014-01-30cpu/intel: allow non-packaged scoped turbo settingAaron Durbin
In the past the turbo disable setting (bit 38) of the IA32_MISC_ENABLES msr has been package scoped. That means knocking the turbo disable bit down enabled turbo for the entire package. Sadly, that's no longer true on all Intel processors. Therefore, allow non-packaged scoped turbo setting by introducing the CPU_INTEL_TURBO_NOT_PACKAGE_SCOPED Kconfig option. It defaults to false which was the original assumption. BUG=chrome-os-partner:25014 BRANCH=baytrail TEST=Built and ran both ways successfully. Change-Id: I71a31e76ff47878023081fc47da643187517b597 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/182405 Reviewed-by: Duncan Laurie <dlaurie@chromium.org> Reviewed-on: http://review.coreboot.org/5047 Tested-by: build bot (Jenkins) Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
2014-01-30x86: Add SMM helper functions to MP infrastructureAaron Durbin
In order for the cpu code to start SMM relocation 2 new functions are added to be shared: - void smm_initiate_relocation_parallel() - void smm_initiate_relocation() The both initiate an SMI on the currently running cpu. The 2 variants allow for parallel relocation or serialized relocation. BUG=chrome-os-partner:22862 BRANCH=None TEST=Built and booted rambi using these functions. Change-Id: I325777bac27e9a0efc3f54f7223c38310604c5a2 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/173982 Reviewed-on: http://review.coreboot.org/4891 Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Tested-by: build bot (Jenkins)
2014-01-30x86: parallel MP initializationAaron Durbin
Provide a common entry point for bringing up the APs in parallel. This work is based off of the Haswell one which can be moved over to this in the future. The APs are brought up and have the BSP's MTRRs duplicated in their own MTRRs. Additionally, Microcode is loaded before enabling caching. However, the current microcode loading support assumes Intel's mechanism. The infrastructure provides a notion of a flight plan for the BSP and APs. This allows for flexibility in the order of operations for a given architecture/chip without providing any specific policy. Therefore, the chipset caller can provide the order that is required. BUG=chrome-os-partner:22862 BRANCH=None TEST=Built and booted on rambi with baytrail specific patches. Change-Id: I0539047a1b24c13ef278695737cdba3b9344c820 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/173703 Reviewed-on: http://review.coreboot.org/4888 Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Tested-by: build bot (Jenkins)
2014-01-30coreboot: config to cache ramstage outside CBMEMAaron Durbin
Haswell was the original chipset to store the cache in another area besides CBMEM. However, it was specific to the implementation. Instead, provide a generic way to obtain the location of the ramstage cache. This option is selected using the CACHE_RELOCATED_RAMSTAGE_OUTSIDE_CBMEM Kconfig option. BUG=chrome-os-partner:23249 BRANCH=None TEST=Built and booted with baytrail support. Also built for falco successfully. Change-Id: I70d0940f7a8f73640c92a75fd22588c2c234241b Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/172602 Reviewed-by: Stefan Reinauer <reinauer@google.com> Reviewed-on: http://review.coreboot.org/4876 Tested-by: build bot (Jenkins) Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
2014-01-30vboot: provide empty vboot_verify_firmware()Aaron Durbin
In the case of CONFIG_VBOOT_VERIFY_FIRMWARE not being selected allow for calling vboot_verify_firmware() with an empty implementation. This allows for one not to clutter the source with ifdefs. BUG=chrome-os-partner:23249 BRANCH=None TEST=Built with a !CONFIG_VBOOT_VERIFY_FIRMWARE and non-guarded call to vboot_verify_firmware(). Change-Id: I72af717ede3c5d1db2a1f8e586fefcca82b191d5 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/172711 Reviewed-by: Shawn Nematbakhsh <shawnn@chromium.org> Reviewed-on: http://review.coreboot.org/4879 Tested-by: build bot (Jenkins) Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
2014-01-28x86: add common definitions for control registersAaron Durbin
The access to control registers were scattered about. Provide a single header file to provide the correct access function and definitions. BUG=chrome-os-partner:22991 BRANCH=None TEST=Built and booted using this infrastructure. Also objdump'd the assembly to ensure consistency (objdump -d -r -S | grep xmm). Change-Id: Iff7a043e4e5ba930a6a77f968f1fcc14784214e9 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/172641 Reviewed-by: Stefan Reinauer <reinauer@google.com> Reviewed-on: http://review.coreboot.org/4873 Tested-by: build bot (Jenkins) Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
2014-01-28intel: fix microcode compilation failure in bootblockAaron Durbin
When not building with CONFIG_SSE there are not enough registers for ROMCC to use for spilling. The previous changes to this file had too many local variables that needed to be tracked -- thus causing romcc compilation issues. Change-Id: I3dd4b48be707f41ce273285e98ebd397c32a6a25 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/4845 Tested-by: build bot (Jenkins) Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
2014-01-26src/cpu: Fix spelling of MTTR to MTRRPaul Menzel
Change-Id: Ia4718ac31a5b2bd12f8cda5e107aa878d74d2a03 Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-on: http://review.coreboot.org/4805 Tested-by: build bot (Jenkins) Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
2014-01-23intel/microcode: Remove leftover MICROCODE_INCLUDE_PATH.Vladimir Serbinenko
Not used anymore since microcode was moved. Change-Id: Id666c80cb20e90e3664c4dcfcc0c41a4aeb4864c Signed-off-by: Vladimir Serbinenko <phcoder@gmail.com> Reviewed-on: http://review.coreboot.org/4788 Tested-by: build bot (Jenkins) Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
2014-01-22AMD K8/fam10: Fix CBMEM on S3 resumeKyösti Mälkki
Change to use cbmem_recovery() to wipe CBMEM region and reset ACPI wakeup if CBMEM TOC was not found. Change-Id: Ic362253eaa00bd442d4cc0514632f9096e20bfa6 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/4673 Tested-by: build bot (Jenkins) Reviewed-by: Aaron Durbin <adurbin@google.com>
2014-01-22AMD AGESA: Fix CBMEM on S3 resumeKyösti Mälkki
Change to use cbmem_recovery() to wipe CBMEM region and reset ACPI wakeup if CBMEM TOC was not found. Change-Id: I6648570d76b5c137f50addcc5bce9c126d179c65 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/4672 Tested-by: build bot (Jenkins) Reviewed-by: Aaron Durbin <adurbin@google.com>
2014-01-16cpu/intel: Remove dummy terminators from microcode blobsAlexandru Gagniuc
Now that CBFS microcode no longer requires a NULL termination, remove the dummy terminators from all microcode blobs. This also enables microcode blobs from different CPU models to be linked in the same cpu_microcode_blob.bin without the terminators getting in the way. Change-Id: I25a6454780fd5d56ae7660b0733ac4f8c4d90096 Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4506 Tested-by: build bot (Jenkins)
2014-01-16cpu/intel: Make all Intel CPUs load microcode from CBFSAlexandru Gagniuc
The sequence to inject microcode updates is virtually the same for all Intel CPUs. The same function is used to inject the update in both CBFS and hardcoded cases, and in both of these cases, the microcode resides in the ROM. This should be a safe change across the board. The function which loaded compiled-in microcode is also removed here in order to prevent it from being used in the future. The dummy terminators from microcode need to be removed if this change is to work when generating microcode from several microcode_blob.c files, as is the case for older socketed CPUs. Removal of dummy terminators is done in a subsequent patch. Change-Id: I2cc8220cc4cd4a87aa7fc750e6c60ccdfa9986e9 Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4495 Tested-by: build bot (Jenkins) Reviewed-by: Aaron Durbin <adurbin@gmail.com>
2014-01-15cpu/amd/model_fxx/powernow_acpi.c: Comment out set but unused variable ↵Paul Menzel
`Start_vid` When adding support for PSS object generation for AMD pre Family Fh CPUs (199c694f) the function `pstates_algorithm` was copied and adapted, but `Start_vid` is not needed anymore as a static table is used. I’d remove the variable, but Ron Minnich requested to leave it there for documentation purposes. So just comment it out. Change-Id: I3002951d168cade6461941c16d78373c47792e13 Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-on: http://review.coreboot.org/4036 Tested-by: build bot (Jenkins) Reviewed-by: Rudolf Marek <r.marek@assembler.cz>
2014-01-15nehalem/sandy/ivy/haswell: Enable WRPROT cache for all of flashKyösti Mälkki
CBFS could start from below 4MB, and should be cacheable for the purpose of early microcode update and CBFS search for romstage file. Change-Id: Ia2a1c6e5fdcc3201fafc8cf5c841cebbbf0b30c9 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/4626 Tested-by: build bot (Jenkins) Reviewed-by: Vladimir Serbinenko <phcoder@gmail.com> Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Aaron Durbin <adurbin@google.com>
2014-01-15Re-declare CACHE_ROM_SIZE as aligned ROM_SIZE for MTRRKyösti Mälkki
This change allows Kconfig options ROM_SIZE and CBFS_SIZE to be set with values that are not power of 2. The region programmed as WB cacheable will include all of ROM_SIZE. Side-effects to consider: Memory region below flash may be tagged WRPROT cacheable. As an example, with ROM_SIZE of 12 MB, CACHE_ROM_SIZE would be 16 MB. Since this can overlap CAR, we add an explicit test and fail on compile should this happen. To work around this problem, one needs to use CACHE_ROM_SIZE_OVERRIDE in the mainboard Kconfig and define a smaller region for WB cache. With this change flash regions outside CBFS are also tagged WRPROT cacheable. This covers IFD and ME and sections ChromeOS may use. Change-Id: I5e577900ff7e91606bef6d80033caaed721ce4bf Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/4625 Tested-by: build bot (Jenkins) Reviewed-by: Vladimir Serbinenko <phcoder@gmail.com>
2014-01-14cpu/allwinner/a10: Clarify positioning of boot stagesAlexandru Gagniuc
This fixes a number of potential issues, such as generating a build failure if the bootblock is too large, and making sure romstage and ramstage cannot overlap in memory. Change-Id: I4ca9ad097b145445316bcd962e007731b08a7fda Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4687 Tested-by: build bot (Jenkins)
2014-01-14cpu/allwinner/a10: Add helper to configure CPU clockAlexandru Gagniuc
Change-Id: I5a3bb3220aeefdd6822a7dbecf210ff77095dad6 Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4685 Tested-by: build bot (Jenkins) Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-01-13cpu/allwinner/a10: Provide utility to make a bootable imageAlexandru Gagniuc
Up until now, we relied on mksunxiboot to prepend the header which makes coreboot.rom bootable on Allwinner SoCs. If that tool was not present, the build silently failed. Integrate this tool into our util/ package, so that we do not have to rely on mksunxiboot being in PATH. Our version of mksunxiboot also eliminates some limitations of the original tool, so we no longer have to use 'dd' to limit the file size. Change-Id: Id5a4b1e2a3cb00cd1d6c70e6cbc3cfd8587e8a24 Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4656 Tested-by: build bot (Jenkins) Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-01-13cpu/allwinner/a10: Clean up include order in Makefile.incAlexandru Gagniuc
Alphabetize the sources for each stage (bootblock, rom, ram), and include twi.c in both romstage and ramstage. Change-Id: I5526f5a66f6600560005731a3ee536eb858f4ff0 Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4639 Tested-by: build bot (Jenkins) Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-01-12CBFS: use cbfs_get_file_content whenever possible rather than cbfs_get_fileVladimir Serbinenko
Number one reason to use cbfs_get_file was to get file length. With previous patch no more need for this. Change-Id: I330dda914d800c991757c5967b11963276ba9e00 Signed-off-by: Vladimir Serbinenko <phcoder@gmail.com> Reviewed-on: http://review.coreboot.org/4674 Reviewed-by: Patrick Georgi <patrick@georgi-clan.de> Tested-by: build bot (Jenkins)
2014-01-11intel/fsp: Fix microcode includingPatrick Georgi
IS_ENABLED() requires the full define (incl. CONFIG_ prefix) but isn't needed here. Change-Id: I91d504367c75ce3fcecc6fa2499afaa0896595d3 Signed-off-by: Patrick Georgi <patrick@georgi-clan.de> Reviewed-on: http://review.coreboot.org/4646 Tested-by: build bot (Jenkins) Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
2014-01-09cpu/allwinner/a10: Import raminit code from ubootAlexandru Gagniuc
The memory initialization code is a work in progress for uboot, so we only import the bits needed to get RAM up and running. Any refactoring is cosmetic, and any functional refactoring should be done in separate patches, and preferably, in coordination with the sunxi team. Since it's not yet determined if we should initialize memory during the bootblock or romstage, we don't add raminit to the build just yet. Change-Id: I2ec1821942c6970150a02fa3806a257da649e1c9 Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4597 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: David Hendricks <dhendrix@chromium.org>
2014-01-09cpu/allwinner/a10: Add low-level helpers for DRAM clock controlAlexandru Gagniuc
PLL5 is special in that it controls the DRAM clock, and requires a fine-grained low-level control which will be needed by raminit code. This change also brings functionality which will be needed by raminit. Change-Id: I25ecc91aa2154e504ceebb9003a5e5728d47f4a3 Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4593 Tested-by: build bot (Jenkins)
2014-01-09cubieboard: Initialize memory in bootblockAlexandru Gagniuc
Even though the Allwinner A10 is limited to a 24KiB bootblock, the memory initialization takes only about 3KiB and leaves enough room for an MMC or NAND driver, so init the memory early on. The advantage is that we can eliminate complicated logistics of where to cache CBFS and where to load the ramstage in SRAM. Change-Id: Id549552ed509434e831db60deaef28e04d62417f Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4630 Tested-by: build bot (Jenkins) Reviewed-by: David Hendricks <dhendrix@chromium.org>
2014-01-09cpu/allwinner/a10: Add functions for driving GPIO pinsAlexandru Gagniuc
Change-Id: I9473a6e574c3af02d154a7e30245f0dc0b238300 Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4599 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: David Hendricks <dhendrix@chromium.org>
2014-01-09cpu/allwinner/a10: Add definitions for in/output GPIO functionsAlexandru Gagniuc
Change-Id: I2b857d3b4c01e39c62e54f753e400e6049f1dbc9 Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4598 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: David Hendricks <dhendrix@chromium.org>
2014-01-09cpu/allwinner/a10: Add function for reading chip revisionAlexandru Gagniuc
Change-Id: Iafbd253235db3914b9382fdb41de2622ef83c6d8 Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4596 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: David Hendricks <dhendrix@chromium.org>
2014-01-09cpu/allwinner/a10: Implement udelay using timer 0Alexandru Gagniuc
Change-Id: I4825f0d57696cd28751c59ae133b7e3315fb78e5 Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4595 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: David Hendricks <dhendrix@chromium.org>
2014-01-09cpu/allwinner/a10: Add definition for gating GPIO S clockAlexandru Gagniuc
This bit is not documented in the datasheet, but is used in the upcoming RAM init code. Change-Id: I697ec222496236ac7690460ee62313ab8b1a2f0b Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4592 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: David Hendricks <dhendrix@chromium.org>
2014-01-08cpu/allwinner/a10: Add basic TWI (I²C) driverAlexandru Gagniuc
Change-Id: I11b10301199e5ff1a45d9b7d2958cc7b6667a29c Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4588 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2014-01-08cpu/allwinner/a10: Refactor API for gating clocks to peripheralsAlexandru Gagniuc
Rather than having to track which bit in which register should be cleared or set to gate or ungate the clock to a certain peripheral, provide a simplified enum which encodes the register and bit. This change comes with a function which decodes the enum and gates/ungates the clock. This also removes the register-dependent bitmasks for APB0 and APB1 gating registers. Change-Id: Ib3ca16e54eb37eadc3ceb88f4ccc497829ac34bc Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4571 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2014-01-08cpu/allwinner/a10: Refactor and document pinmux APIAlexandru Gagniuc
Include a function to multiplex more than one pin at a time. This is useful for peripherals that have the same function number for all their pins. Since we now have two functions for muxing pins, also document them. Change-Id: I53997cc3a2586e3cf749cd672f69fb427659c67f Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4565 Tested-by: build bot (Jenkins) Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2014-01-08cpu/allwinner/a10: Clarify the usage of SRAM during bootblockAlexandru Gagniuc
We have 32KiB of usable SRAM right when we boot. The first 24KiB can be loaded with our bootblock, while the other 8KiB can be used as stack during the bootblock stage. Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Change-Id: I48d3a37869031c3c1dbc1fab71204d473d64deeb Reviewed-on: http://review.coreboot.org/4563 Tested-by: build bot (Jenkins) Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2014-01-08cpu: Add initial support for Allwinner A10 SoCAlexandru Gagniuc
Add minimal support needed to get a bootblock capable of initialising a serial console. Change-Id: I50dd85544549baf9c5ea0aa3b4296972136c02a4 Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4549 Tested-by: build bot (Jenkins) Reviewed-by: David Hendricks <dhendrix@chromium.org> Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2013-12-30AMD K8: Define MEM_TRAIN_SEQ only with K8_REV_F_SUPPORTKyösti Mälkki
Change-Id: I601efbff03d0f0f59557b33be8d6928ede310b62 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/4558 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
2013-12-29AMD fam10: Drop RAMINIT_SYSINFOKyösti Mälkki
AMD fam10 raminit cannot be built without RAMINIT_SYSINFO, this is not a true option but copy-paste remainder from AMD K8. Change-Id: Id8edc112f3bacebd1732304ac9ee6e77cc6263b7 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/4581 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <patrick@georgi-clan.de> Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
2013-12-29AMD K8: Socket implies K8_REV_F_SUPPORTKyösti Mälkki
K8_REV_F_SUPPORT is already set by all affected sockets, (AM2, F, S1G1). Change-Id: If42a4178263d90a4e195fae0c78943ac9eda1ad6 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/4557 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
2013-12-26AMD boards (non-AGESA): Cleanup post_cache_as_ram.c includesKyösti Mälkki
Change-Id: Ib3a69e3364418426438f88ba14e5cf744e2414fa Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/4524 Tested-by: build bot (Jenkins) Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-by: Bruce Griffith <Bruce.Griffith@se-eng.com> Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
2013-12-26AMD boards (non-AGESA): Cleanup earlymtrr.c includesKyösti Mälkki
Change-Id: I5f4bf9dbaf3470dc83d3e980bb6cab10801e15c1 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/4523 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Bruce Griffith <Bruce.Griffith@se-eng.com>
2013-12-22Fix linking microcode with more than one microcode fileAlexandru Gagniuc
When assembling microcode , the rule to link individual object files into one larger file only passed the first dependency to the linker. As a results only microcode from one object file would actually get linked. This is fixed by replacing $^ with $+ inside the make rule. Change-Id: I65c0565f2e03777af23e530c08d1241804ca19b1 Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com> Reviewed-on: http://review.coreboot.org/4500 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
2013-12-21pit: disable LCD FETs before doing any graphics initDavid Hendricks
This ensures that the LCD FETs are off before we do graphics init. FIXME: The location of the code is sub-optimal and should probably be done in romstage, but there are __PRE_RAM__ considerations to take into account. Signed-off-by: David Hendricks <dhendrix@chromium.org> Change-Id: I0844030d0a0e51eee1d29f1762f0b495777268df Reviewed-on: https://gerrit.chromium.org/gerrit/64305 Reviewed-by: Ronald G. Minnich <rminnich@chromium.org> Commit-Queue: Ronald G. Minnich <rminnich@chromium.org> Tested-by: Ronald G. Minnich <rminnich@chromium.org> Reviewed-on: http://review.coreboot.org/4470 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
2013-12-21exynos5420: Assign corect parent PLLsDavid Hendricks
Assign correct parent PLL's for the following clocks: ACLK_400_WCORE (MPLL->CPLL) (400 -> 333MHz) PCLK_200_FSYS (MPLL->DPLL) (200 -> 200MHz) MUX_ACLK_100_NOC_SEL (MPLL -> DPLL) (100 -> 100MHz) ACLK_266 (DPLL->MPLL) (300 -> 266MHz) ACLK_200_DISP1(MPLL->DPLL) (200 -> 200MHz) ACLK_400_MSCL(MPLL->CPLL) (400 -> 333MHz) ACLK_66 (MPLL->CPLL) (66.666 -> 66.6MHz) MUX_ACLK_400_DISP1_SEL (CPLL->DPLL) (666 -> 300MHz) MUX_MPHY_REFCLK (MPLL->OSC) MUX_UNIPRO (MPLL->OSC) MUX_MIPI1 (EPLL->OSC) MUX_DP1_EXT_VID (EPLL->OSC) MUX_FIMD1_OPT (EPLL->OSC) MUX_IPLL(IPLL->OSC) This also corrects the clock dividers for few of the clocks, as the clock parent changes affect the final frequency of the clocks. This is ported from: https://gerrit.chromium.org/gerrit/#/c/62437/ Signed-off-by: David Hendricks <dhendrix@chromium.org> Change-Id: Ie833c01913d0961a6190446bd573511de8dee5f8 Reviewed-on: https://gerrit.chromium.org/gerrit/65620 Commit-Queue: Ronald G. Minnich <rminnich@chromium.org> Reviewed-by: Ronald G. Minnich <rminnich@chromium.org> Tested-by: Ronald G. Minnich <rminnich@chromium.org> Reviewed-on: http://review.coreboot.org/4469 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
2013-12-21exynos5420: don't assume MPLL for i2c parent clockDavid Hendricks
This reads the clock select field for MUX_ACLK_66_SEL in the CLK_SRC_TOP1 register in order to obtain the source clock rate for I2C peripherals. Before we were always assuming that the source was the MPLL. Unfortunately not all fields in the CLK_SRC_TOPn registers are enumerated the same with regard to clock select. So this is just a one-off for now. This is basically ported from https://gerrit.chromium.org/gerrit/#/c/62443. Signed-off-by: David Hendricks <dhendrix@chromium.org> Change-Id: I9fa85194ae1a1fadab79695f059efdc2e2f1f75f Reviewed-on: https://gerrit.chromium.org/gerrit/65611 Reviewed-by: Ronald G. Minnich <rminnich@chromium.org> Tested-by: David Hendricks <dhendrix@chromium.org> Commit-Queue: David Hendricks <dhendrix@chromium.org> Reviewed-on: http://review.coreboot.org/4468 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
2013-12-21exynos5420: Set SPLL to 400MHzDavid Hendricks
Increase SPLL to 400MHz from 300MHz as we set SPLL as the switching parent for ARM and KFC. This value is as per recommendation of the hardware team. This is ported from https://gerrit.chromium.org/gerrit/62618 Signed-off-by: David Hendricks <dhendrix@chromium.org> Change-Id: I8a5a5b957083b0b1f3e3e318fe5753cf7ae19223 Reviewed-on: https://gerrit.chromium.org/gerrit/65432 Reviewed-by: Gabe Black <gabeblack@chromium.org> Tested-by: David Hendricks <dhendrix@chromium.org> Commit-Queue: Gabe Black <gabeblack@chromium.org> Reviewed-on: http://review.coreboot.org/4464 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
2013-12-21exynos5420: re-factor clock_get_periph_rate()David Hendricks
This re-factors clock_get_periph_rate() to be a simpler and also make a few corrections along the way. To summarize: - clk_bit_info is no longer used. It had numerous errors and was really painful anyway since it was just a bunch of opaque magic numbers that made bugs non-obvious. - Clock source bitfields for peripherals handled in the switch statement are 3 bits, not 4. Some divider values are 3 bits, some are 4. The earlier code always assumed 4 bits for both which included reserved bits in many cases. - UART source clock and divider shift values were wrong. - PWM clock divider was being read from the wrong register. - SPI3 divider value was being read from the wrong register. - There was a really confusing calculation for SDMMC0 and SDMMC2 clock rates, but it was never actually used since the switch statement never handled PERIPH_ID_SDMMC{0,2} and would thus return if they were ever passed into this function. Signed-off-by: David Hendricks <dhendrix@chromium.org> Change-Id: I0a03a64d8b42fbe83dbf377292597ce681b22f4b Reviewed-on: https://gerrit.chromium.org/gerrit/65284 Commit-Queue: David Hendricks <dhendrix@chromium.org> Tested-by: David Hendricks <dhendrix@chromium.org> Reviewed-by: Gabe Black <gabeblack@chromium.org> Reviewed-on: http://review.coreboot.org/4463 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
2013-12-21exynos5420: add a peripheral clock select --> PLL decoderDavid Hendricks
This adds a helper function to translate between peripheral clock select fields in clock source registers and PLLs. Some of this was already done to handle a few special cases, this generalizes the earlier work so that follow-up patches can do further clean-up. Unfortunately, the PLLs represented by clock select fields in various modules are not uniformly ordered. So for now we focus on peripheral clock sources only. Signed-off-by: David Hendricks <dhendrix@chromium.org> Change-Id: Id58a3e488650d09e6a35c22d5394fcbf0ee9ddff Reviewed-on: https://gerrit.chromium.org/gerrit/65283 Commit-Queue: David Hendricks <dhendrix@chromium.org> Tested-by: David Hendricks <dhendrix@chromium.org> Reviewed-by: Gabe Black <gabeblack@chromium.org> Reviewed-on: http://review.coreboot.org/4462 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
generated by cgit v1.2.3 (git 2.25.1) at 2025-03-10 13:24:36 +0000