MRS commands are used to tell the DRAM chip what timing and what
termination and drive strength to use, along with other parameters.
The MRS commands are defined by the DDR3 specification [1]. This
makes MRS commands hardware-independent.
MRS command creation is duplicated in various shapes and forms in any
chipset that does DDR3. This is an effort to create a generic MRS API
that can be used with any chipset.
This is used in the VX900 branch.
[1] www.jedec.org/sites/default/files/docs/JESD79-3E.pdf
Change-Id: Ia8bb593e3e28a5923a866042327243d798c3b793
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/3354
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
While we had support for updating microcode on the VIA Nano CPUs for a
while now, we never included the actual microcode. Unlike, Intel and
AMD CPUs, VIA microcode is not available for download, and was
extracted from the vendor BIOS. It was not included in coreboot since
we never had explicit permission to do so. I have just received
confirmation from VIA that we can distribute the microcode.
Change-Id: I4c15b090cd2713cfe5dc6b50db777ff89dbc0f19
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/3357
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Tested-by: build bot (Jenkins)
Move the include before static inline int spd_read_byte().
Change-Id: I4cac4b1f55368041b067422d95c09208e15d0f2d
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-on: http://review.coreboot.org/3368
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
This commit fixes problems if we build raminit.c
for romstage.
Change-Id: Ic1380f3635ac28b939fa2a8ce614814012455c44
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-on: http://review.coreboot.org/3363
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
In order to get rid of the bad #include "northbridge/amd/lx/raminit.c"
line we need to do some prepartion steps. This commit is one of them.
Change-Id: I33173660bbda8894e7672e41e1b994d254d7ae8a
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-on: http://review.coreboot.org/3362
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Take a Parmer board with 4G memory as an example.
Use 'cat /proc/meminfo' to check memory, it reads 'MemTotal 3327540kB'.
Parmer uses 512M as video memory when it has 4G.
3327540+512*1024 = 3851828(kB), so some memory is lost.
When Parmer has 4G memory, TOM2 low is 0x1F000000, TOM2 high is
0x00000001. But in e820 table or coreboot table, the last item is
6: 0000000100000000 - 0000000118000000 = 1 RAM
This is not correct, it should be
6: 0000000100000000 - 000000011f000000 = 1 RAM
This patch changes the memory layout when TOM2 is set.
Change-Id: I4e2d163ae8fe1e65ddc384b520a5112ca067b1d1
Signed-off-by: Siyuan Wang <SiYuan.Wang@amd.com>
Signed-off-by: Siyuan Wang <wangsiyuanbuaa@gmail.com>
Reviewed-on: http://review.coreboot.org/3366
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Bash case statements are terminated with ';;'.
Unlike C, bash case statements will not continue to the next case. No 'break' is needed.
Change-Id: I62e7e91f3223ac4052728a1ca12a4681af0dc036
Signed-off-by: David Hubbard <david.c.hubbard+coreboot@gmail.com>
Reviewed-on: http://review.coreboot.org/3330
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
This file was missing some definitions, so add them. Also turn the defines
into an enum. The reason for doing this is that functions can now
explicitly take an spd_memory_type as a parameter:
> int do_something_with_dram(enum spd_memory_type type, ...)
Which is a lot more explicit and readable than:
> int do_something_with_dram(u8 type, ...)
These are used in the VX900 branch.
Change-Id: Ic7871e82c2523a94eac8e07979a8e34e0b459b46
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/3266
Tested-by: build bot (Jenkins)
Use PRIx64 to print a u64 instead of "llx". Fixes the following error:
cbmem.c: In function 'parse_cbtable':
cbmem.c:135:2: error: format '%llx' expects argument of type 'long long unsigned int', but argument 2 has type 'u64' [-Werror=format=]
Change-Id: Ibc2bf8597cb86db5b2e71fba77ec837a08c5e3d4
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/3301
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Probably due to different (character) widths for a tab, sometimes only
one tab was used for aligning the define `CPU_ID_EXT_FEATURES_MSR`. For
the “correct” alignment, that means where a tab is eight characters,
two tabs are necessary. Change it accordingly.
Change-Id: I450a7796dc00b934b5a6bab8642db04a27f69f4b
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-on: http://review.coreboot.org/3263
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
On Asus F2A85-M, the Linux kernel complains that the _CRS method does
not specify the number of PCI busses.
[FIRMWARE BUG]: ACPI: no secondary bus range in _CRS
Just put there 256. This should be part of re-factoring of the whole
ACPI stuff.
The same change was already done for the AMD Brazos (SB800) boards,
based on commit »Persimmon DSDT: Add secondary bus range to PCI0«
(4733c647) [1].
[1] http://review.coreboot.org/2592
Change-Id: I06f90ec353df9198a20b2165741ea0fe94071266
Signed-off-by: Rudolf Marek <r.marek@assembler.cz>
Reviewed-on: http://review.coreboot.org/3320
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martin.roth@se-eng.com>
Reviewed-by: David Hubbard <david.c.hubbard+coreboot@gmail.com>
There is no need to use everywhere BIOS_ERR.
Change-Id: If33d72919109244a7c3bd96674a4e386c8d1a19e
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-on: http://review.coreboot.org/3307
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Denis Carikli <GNUtoo@no-log.org>
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Add support for sending debug output to an I/O port.
It can be used together with QEMU's isa-debugcon driver to log the
coreboot output to a file. The port is configurable and defaults
to 0x402 which has established as the de facto standard. For example,
SeaBIOS+OVMF [1] use that one too.
[1] http://www.linux-kvm.org/page/OVMF
Open Virtual Machine Firmware
Change-Id: I0803f7fc70030242f80003e25c9449c37d71975e
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-on: http://review.coreboot.org/3331
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
There were assumptions being made in the haswell
MP and SMM code which assumed the APIC id space
was 1:1 w.r.t. cpu number. When hyperthreading is
disabled the APIC ids of the logical processors
are all even. That means the APIC id space is sparse.
Handle this situation.
Change-Id: Ibe79ab156c0a171208a77db8a252aa5b73205d6c
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/3353
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
It's possible that the TOUUD can be set to less than
4GiB. When that is the case the size_k variable is
an extremely large value. Instead ensure TOUUD is greater
than 4GiB before adding said resources.
Change-Id: I456633d6210824e60665281538300fd15656b86d
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/3352
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Remove local copies of reading and writing I/O APIC registers by
using already available functions.
This change is similar to
commit db4f875a41
Author: Kyösti Mälkki <kyosti.malkki@gmail.com>
Date: Tue Jan 31 17:24:12 2012 +0200
IOAPIC: Divide setup_ioapic() in two parts.
Reviewed-on: http://review.coreboot.org/300
and
commit e614353194
Author: Kyösti Mälkki <kyosti.malkki@gmail.com>
Date: Tue Feb 26 17:24:41 2013 +0200
Unify setting 82801a/b/c/d IOAPIC ID
Reviewed-on: http://review.coreboot.org/2532
and uses `io_apic_read()` and `io_apic_write()` too. Define
`ACPI_EN` in the header file `pch.h`.
As commented by Aaron Durbin, a separate `pch_enable_acpi()` is
not needed: “The existing code path *in this file* is about enabling
the io apic.” [1].
[1] http://review.coreboot.org/#/c/3182/4/src/southbridge/intel/lynxpoint/lpc.c
Change-Id: I6f2559f1d134590f781bd2cb325a9560512285dc
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-on: http://review.coreboot.org/3182
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
Remove local copies of reading and writing I/O APIC registers by
using already available functions.
This change is similar to
commit db4f875a41
Author: Kyösti Mälkki <kyosti.malkki@gmail.com>
Date: Tue Jan 31 17:24:12 2012 +0200
IOAPIC: Divide setup_ioapic() in two parts.
Reviewed-on: http://review.coreboot.org/300
and
commit e614353194
Author: Kyösti Mälkki <kyosti.malkki@gmail.com>
Date: Tue Feb 26 17:24:41 2013 +0200
Unify setting 82801a/b/c/d IOAPIC ID
Reviewed-on: http://review.coreboot.org/2532
and uses `io_apic_read()` and `io_apic_write()` too. Define
`ACPI_EN` in the header file `pch.h`.
As commented by Aaron Durbin, a separate `pch_enable_acpi()` is
not needed: “The existing code path *in this file* is about enabling
the io apic.” [1].
[1] http://review.coreboot.org/#/c/3182/4/src/southbridge/intel/lynxpoint/lpc.c
Change-Id: I4478b1902d09061ca1db8eab6b71fef388c7a74c
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-on: http://review.coreboot.org/3183
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
From ISO C99 standard: »The placement of a storage-class specifier
other than at the beginning of the declaration specifiers in a
declaration is an obsolescent feature.«
Found at <http://www.approxion.com/?p=41>.
The following command was used to make the change.
$ git grep -l 'const static' src/ | xargs sed -i 's/const static/static const/'
As asked by Bruce Griffith, the changes in `src/vendorcode` were
reverted as that is what AMD prefers.
The same change was done already for AMD Persimmon in the following
commit.
commit 824e192809
Author: Jens Rottmann <JRottmann@LiPPERTembedded.de>
Date: Wed Feb 20 21:24:20 2013 +0100
Persimmon: platform_cfg.h: Declare codec arrays as `static const`
Reviewed-on: http://review.coreboot.org/2474
Change-Id: I233c83fdc95ea4f83f7296c818547beb52366a3d
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-on: http://review.coreboot.org/3197
Tested-by: build bot (Jenkins)
Reviewed-by: Bruce Griffith <Bruce.Griffith@se-eng.com>
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Some settings in the am335x Kconfig weren't actually used for anything, some
where place holders, and some where left over from another CPU. The memory
addresses are in the internal RAM in the SOC as described in the reference
manual. The stack is put where the internal ROM had its stack, and the
bootblock is put at the bottom of that region as the manual suggests. The
ROM stage offset is set to 10K which is a bit bigger than the ~7.5K the
bootblock currently takes up.
Change-Id: I1a117d789a791d7e3db1118823f8216b3361433c
Signed-off-by: Gabe Black <gabeblack@chromium.org>
Reviewed-on: http://review.coreboot.org/3327
Tested-by: build bot (Jenkins)
Reviewed-by: David Hendricks <dhendrix@chromium.org>
Without that fix we have with CONFIG_USE_OPTION_TABLE:
OPTION cmos_layout.bin
build/util/nvramtool/nvramtool -y /home/gnutoo/x86/coreboot-alix/src/mainboard/pcengines/alix1c/cmos.layout -L build/cmos_layout.bin
make: *** No rule to make target `nvramtool', needed by `build/coreboot.pre1'. Stop.
rm build/util/sconfig/sconfig.tab.c build/cbfs/fallback/bootblock.elf build/util/sconfig/lex.yy.c
That log was captured with make V=1 but the error also appear with make.
Tested on the PC Engines ALIX.1C with the following commit (Change-Id: Ia87b090) [1]:
PC Engines ALIX.1C: Add CMOS defaults.
[1] http://review.coreboot.org/#/c/3323/
Change-Id: I548005a58f430ed7b6da5249a24bbdcae440a1e9
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@no-log.org>
Reviewed-on: http://review.coreboot.org/3223
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Tested-by: build bot (Jenkins)
- SPI controller base address gets overwritten by SD controller under Linux.
- Reason for overwrite is the SPI base address isn't in a standard BAR and doesn't
get automatically reserved. Solution is to add it as a reserved memory area in
ACPI.
- This issue was found on the ASUS F2A85-M platform. Currently a workaround on this
platform was made as part of: http://review.coreboot.org/#/c/3167/3
- Once approved a follow-on patch for other southbridges using a non-standard BAR for
the spi controller.
Change-Id: I1b67da3045729a6754e245141cd83c5b3cc9009e
Signed-off-by: Steven Sherk <steven.sherk@se-eng.com>
Reviewed-on: http://review.coreboot.org/3270
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
This issue can be reproduced in Linux by the following steps:
1) use pm-suspend to suspend.
2) use USB keyboard to wake up.
3) use pm-suspend to suspend. FAIL To SUSPEND.
The cause of this issue is:
USB devices use bit 11(0x0b) of GP0_STS represents S3 wake up event,
but this bit is not clear after wake up. So OS thinks there is a
wake up signal and wake up immediately.
In this patch, I add AcpiGpe0Blk using MMIO access and write 1
on bit 11. I have tested on Parmer.
Change-Id: Iec3078bf29de99683e7cd3ef4e178fbeb4dc09c1
Signed-off-by: Siyuan Wang <SiYuan.Wang@amd.com>
Signed-off-by: Siyuan Wang <wangsiyuanbuaa@gmail.com>
Reviewed-on: http://review.coreboot.org/3347
Tested-by: build bot (Jenkins)
Reviewed-by: Dave Frodin <dave.frodin@se-eng.com>
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Change `sizeof(type) * n`, where n is the number of array
elements, to `sizeof(variable)` to directly get the size of the
variable (struct, array). Determining the size by counting array
elements is error prone and unnecessary.
Rudolf Marek’s patch »ASUS F2A85-M: Correct and clean up PCIe
config« [1] contains the same change and is ported over. In
the commit message Rudolf makes the following comment.
»Not sure why the copy is needed instead of direct reference.
Maybe it has something to do with CAR?«
Testing on the ASRock E350M1, no regressions were noticed.
[1] http://review.coreboot.org/#/c/3194/
Change-Id: I123031b3819a10c9c85577fdca96c70d9c992e87
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-on: http://review.coreboot.org/3248
Tested-by: build bot (Jenkins)
Reviewed-by: Bruce Griffith <Bruce.Griffith@se-eng.com>
Change `sizeof(type) * n`, where n is the number of array
elements, to `sizeof(variable)` to directly get the size of the
variable (struct, array). Determining the size by counting array
elements is error prone and unnecessary.
Not sure why the copy is needed instead of direct reference.
Maybe it has something to do with CAR?
These changes are based on Rudolf’s original patch »ASUS F2A85-M:
Correct and clean up PCIe config« [1], where it was just done for
the ASUS board.
[1] http://review.coreboot.org/#/c/3194/
Change-Id: I4aa4c6cde5a27b7f335a71afc21d1603f2ae814b
Signed-off-by: Rudolf Marek <r.marek@assembler.cz>
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-on: http://review.coreboot.org/3247
Tested-by: build bot (Jenkins)
Reviewed-by: David Hubbard <david.c.hubbard+coreboot@gmail.com>
Reviewed-by: Bruce Griffith <Bruce.Griffith@se-eng.com>
Extra care for the qemu vga should not be needed any more.
Since release 0.12 qemu loads the vgabios into the PCI ROM
bar, so everything works exactly like it does on real hardware.
Change-Id: I4b9bf1244cad437cbe5168600aeee52031456033
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-on: http://review.coreboot.org/3333
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Initial structure of Beaglebone port
Change-Id: Ia255ab207f424dcd525990cdc0d74953e012c087
Signed-off-by: David Hendricks <dhendrix@chromium.org>
Signed-off-by: Gabe Black <gabeblack@chromium.org>
Reviewed-on: http://review.coreboot.org/3279
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
buildgcc has many wrong choices, and two right ones,
but you would never guess that. It's even more
frustrating when it spends lots of time building a
full tool chain and you find out it's not the one you
wanted and, still worse, you've forgotten what it does want
and, even worse, it won't f-ing tell you what the two
right choices are!.
Have it tell you when you've done something wrong, and have it
make reasonable decisions when you say things like
-p arm
instead of
-p armv7a-eabi
This change lowers my blood pressure 10 points.
Change-Id: I44a59d7cb7a6260894d8bcb692a693ed25681ff8
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Reviewed-on: http://review.coreboot.org/3292
Tested-by: build bot (Jenkins)
This allows other boards to have the same choice block without confusing
kconfig.
Change-Id: Iea5a7f2d1c263aa7992f504b832ca9c862833c3f
Signed-off-by: Gabe Black <gabeblack@chromium.org>
Reviewed-on: http://review.coreboot.org/3293
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Tested-by: build bot (Jenkins)
Properly check the dependency of choices as a group.
Also fix that sym_check_deps() correctly terminates the dependency loop
error check (otherwise it would continue printing the dependency chain).
Signed-off-by: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
=======
Cherry-picked from the Linux kernel.
Change-Id: I0c98760dd0f55cf2ff70c53e0b014288b59574c8
Signed-off-by: Gabe Black <gabeblack@chromium.org>
Reviewed-on: http://review.coreboot.org/3290
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Fix reversal of dlg.border.atr and dlg.dialog.atr for draw_box()
Makes the inputbox look like expected
Signed-off-by: Roel Kluin <12o3l@tiscali.nl>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
=======
Cherry-picked from the Linux kernel.
Change-Id: I596915aab0204ef0e392fefa56fad8e25204e207
Signed-off-by: Gabe Black <gabeblack@chromium.org>
Reviewed-on: http://review.coreboot.org/3289
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
As choice dependency are now fully checked, it's quite easy to add support
for named choices. This lifts the restriction that a choice value can only
appear once, although it still has to be within the same group,
but multiple choices can be joined by giving them a name.
While at it I cleaned up a little the choice type logic to simplify it a
bit.
Signed-off-by: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
=======
Cherry-picked from the Linux kernel.
Change-Id: If0f00d1783907d606220cda5307b8960d3bfc38d
Signed-off-by: Gabe Black <gabeblack@chromium.org>
Reviewed-on: http://review.coreboot.org/3291
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
multiply_to_tsc was being copied everywhere, which is bad
practice. Put it in the tsc.h include file where it belongs.
Delete the copies of it.
Per secunet, no copyright notice is needed.
This might be a good time to get a copyright notice into tsc.h
anyway.
Change-Id: Ied0013ad4b1a9e5e2b330614bb867fd806f9a407
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Reviewed-on: http://review.coreboot.org/3242
Tested-by: build bot (Jenkins)
Reviewed-by: Nico Huber <nico.huber@secunet.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Commit »haswell: 24MHz monotonic time implementation« (c46cc6f1) [1]
added the Kconfig variable `MONOTONIC_TIMER_MSR` with a help text,
but only used one space instead of the suggested two spaces for
indentation. So add one space.
»Lines under a "config" definition are indented with one tab, while
help text is indented an additional two spaces.« [2]
[1] http://review.coreboot.org/3153
[2] https://www.kernel.org/doc/Documentation/CodingStyle
(Chapter 10: Kconfig configuration files)
Change-Id: I39cf356bfd54c66a2f1b837c6667dcc915e41f29
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-on: http://review.coreboot.org/3262
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
Currently code in `udelay.c` differs between the Intel northbridges
GM45, 945 on the one hand and Sandy Bridge on the other hand.
The reason for this is that a wrong comparison > was used.
The following commit
commit 784ffb3db6
Author: Sven Schnelle <svens@stackframe.org>
Date: Tue Jan 10 12:16:38 2012 +0100
i945: fix tsc udelay()
Reviewed-on: http://review.coreboot.org/530
fixed the sign from > to <, whereas Stefan Reinauer changed it from
> to <= before adding the Sandy Bridge port in the following commit.
commit 00636b0dae
Author: Stefan Reinauer <stefan.reinauer@coreboot.org>
Date: Wed Apr 4 00:08:51 2012 +0200
Add support for Intel Sandybridge CPU (northbridge part)
Reviewed-on: http://review.coreboot.org/854
As there are no technical reasons for this difference, unify this
between the chipsets. See the discussion of the other patch set in
Gerrit [1].
[1] http://review.coreboot.org/#/c/3220/1/src/northbridge/intel/i5000/udelay.c
Change-Id: I64f2aa1db114ad2e9f34181c5f3034f6a8414a11
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-on: http://review.coreboot.org/3259
Tested-by: build bot (Jenkins)
Reviewed-by: Nico Huber <nico.huber@secunet.com>
Add debug output for the timing values of the edges found during
read and write training.
Now, output for one DIMM of DDR3-1066 in a roda/rk9 looks like:
[...]
Lower bound for byte lane 0 on channel 0: 0.0
Upper bound for byte lane 0 on channel 0: 8.4
Final timings for byte lane 0 on channel 0: 4.2
Lower bound for byte lane 1 on channel 0: 0.0
Upper bound for byte lane 1 on channel 0: 10.2
Final timings for byte lane 1 on channel 0: 5.1
Lower bound for byte lane 2 on channel 0: 0.0
Upper bound for byte lane 2 on channel 0: 7.5
Final timings for byte lane 2 on channel 0: 3.6
Lower bound for byte lane 3 on channel 0: 0.0
Upper bound for byte lane 3 on channel 0: 11.4
Final timings for byte lane 3 on channel 0: 5.6
Lower bound for byte lane 4 on channel 0: 0.0
Upper bound for byte lane 4 on channel 0: 9.4
Final timings for byte lane 4 on channel 0: 4.6
Lower bound for byte lane 5 on channel 0: 0.0
Upper bound for byte lane 5 on channel 0: 11.2
Final timings for byte lane 5 on channel 0: 5.5
Lower bound for byte lane 6 on channel 0: 0.0
Upper bound for byte lane 6 on channel 0: 8.4
Final timings for byte lane 6 on channel 0: 4.2
Lower bound for byte lane 7 on channel 0: 0.0
Upper bound for byte lane 7 on channel 0: 10.4
Final timings for byte lane 7 on channel 0: 5.2
Lower bound for group 0 on channel 0: 1.7.5
Upper bound for group 0 on channel 0: 2.2.2
Final timings for group 0 on channel 0: 1.10.7
Lower bound for group 1 on channel 0: 1.6.1
Upper bound for group 1 on channel 0: 2.0.2
Final timings for group 1 on channel 0: 1.9.1
Lower bound for group 2 on channel 0: 2.0.7
Upper bound for group 2 on channel 0: 2.8.1
Final timings for group 2 on channel 0: 2.4.4
Lower bound for group 3 on channel 0: 2.4.7
Upper bound for group 3 on channel 0: 3.0.0
Final timings for group 3 on channel 0: 2.8.3
[...]
Final timings are always the average of the two bounds. The last dots
separate eights (not decimals) and the middles are elenvenths or twelfths
depending on the clock speed (twelfths in this case).
Change-Id: Idb7c84b514716c7265b94890c39b7225de7800dc
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/3257
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
We halted the machine on any overflow during the write training.
However, overflows during the search for a good to bad edge are
non-fatal, and should be ignored.
Change-Id: I45ccbabc214e208974039246d806b0d2ca2fdc03
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/3256
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Split some code in individual functions. It's the refactoring part of
a bigger change, following...
Change-Id: Id19be4588ad8984935040d9bcba4d7c5f2e1114f
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/3255
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Alexandru Gagniuc