There are many good reasons why we may want to run some sort of generic
callback before we're executing a reset. Unfortunateley, that is really
hard right now: code that wants to reset simply calls the hard_reset()
function (or one of its ill-differentiated cousins) which is directly
implemented by a myriad of different mainboards, northbridges, SoCs,
etc. More recent x86 SoCs have tried to solve the problem in their own
little corner of soc/intel/common, but it's really something that would
benefit all of coreboot.
This patch expands the concept onto all boards: hard_reset() and friends
get implemented in a generic location where they can run hooks before
calling the platform-specific implementation that is now called
do_hard_reset(). The existing Intel reset_prepare() gets generalized as
soc_reset_prepare() (and other hooks for arch, mainboard, etc. can now
easily be added later if necessary). We will also use this central point
to ensure all platforms flush their cache before reset, which is
generally useful for all cases where we're trying to persist information
in RAM across reboots (like the new persistent CBMEM console does).
Also remove cpu_reset() completely since it's not used anywhere and
doesn't seem very useful compared to the others.
Change-Id: I41b89ce4a923102f0748922496e1dd9bce8a610f
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/19789
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
The comparison value was obviously wrong here. One too many 'f'
characters.
Found-by: Coverity Scan #1229588 & 1229604
Change-Id: Iedd4f956d846f1c8661390b346c7397346def86b
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/18100
Tested-by: build bot (Jenkins)
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Previously, all romstages for this northbridge family
would compile via 1 single C file with everything
included into the romstage.c file (!)
This patch separates the build into separate .o modules
and links them accordingly.
Currently compiles and links all fam10 roms without
breaking other roms.
Both DDR2 and DDR3 have been completed
TESTED on REACTS: passes all boot tests for 2 boards
ASUS KGPE-D16
ASUS KFSN4-DRE
Some extra changes were required to make it compile
otherwise there were unused functions in included "c" files.
This is because I needed to exchange CIMX
for the native southbridge routines. See in particular:
advansus/a785e-i
asus/m5a88-v
avalue/eax-785e
A followup patch may be required to fix the above boards.
See FIXME, XXX tags
Change-Id: Id0f9849578fd0f8b1eab83aed910902c27354426
Signed-off-by: Damien Zammit <damien@zamaudio.com>
Reviewed-on: https://review.coreboot.org/17625
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Timothy Pearson <tpearson@raptorengineering.com>
amd_generate_powernow is never called by in lpc_slave_ops.
Move it to lpc_ops like on all other AMD southbridges.
TESTED on Gigabyte ga-m57sli-s4
Change-Id: I7db036e681d591a19e15dd3eaafb88b72a41bea1
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/17977
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
The code originates from times before __SIMPLE_DEVICE__ was
introduced. To keep behaviour unchanged, use explicit PCI
IO operations here.
Change-Id: I44851633115f9aee4c308fd3711571a4b14c5f2f
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/17720
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Replace the use of the old device_t definition inside
southbridge/nvidia/mcp55.
Change-Id: I98ac468940eaf6c456fa95540ec3e718edfe26a7
Signed-off-by: Antonello Dettori <dev@dettori.io>
Reviewed-on: https://review.coreboot.org/16487
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Non-code flow assembly stubs do not have to be included in
bootblock.S, now that we have more freedom in bootblock linking.
Rather than bringing these stubs to the config system, just link them
in the bootblock.
Note that we cannot fully remove CHIPSET_BOOTBLOCK_INCLUDE at this
point, as some intel SOCs use this stub for code flow.
objdump -h build/cbfs/fallback/bootblock.debug on a few random boards
confirms that the appropriate sections are still included in the
final binary.
Change-Id: Id3f9ece14e399c1cc83090f407780c4a05a076f0
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: https://review.coreboot.org/11856
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
It encourages users from writing to the FSF without giving an address.
Linux also prefers to drop that and their checkpatch.pl (that we
imported) looks out for that.
This is the result of util/scripts/no-fsf-addresses.sh with no further
editing.
Change-Id: Ie96faea295fe001911d77dbc51e9a6789558fbd6
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/11888
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
The x86 bootblock linking is a mess. The bootblock is treated in
a very special manner, and never received the update to link-time
garbage collection.
On newer x86 platforms, the boot media is no longer memory-mapped.
That means we need to do a lot more setup in the bootblock. ROMCC is
unsuitable for this task, and walkcbfs only works on memory-mapped
CBFS. We need to revise the x86 bootflow for this new case.
The approach this patch series takes is to perform CAR setup in the
bootblock, and load the following stage (either romstage or verstage)
from the boot media. This approach is not new, but has been done on
our ARM ports for years.
Since we will be adding .c files to the bootblock, it is prudent to
use link-time garbage collection. This is also consistent to how we
do things on other architectures. Unification FTW!
Change-Id: I16b78456df56e0053984a9aca9367e2542adfdc9
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/11781
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
The build system was previously determining the flow
and linking scripts bootblock code by the order of files
added to the bootblock_inc bootblock-y variables.Those
files were then concatenated together and built by a myriad of
make rules.
Now bootblock.S and bootblock.ld is added so that bootblock
can be built and linked using the default build rules.
CHIPSET_BOOTBLOCK_INCLUDE is introduced in order to allow the
chipset code to place include files in the path of the bootblock
program -- a replacement for the chipset_bootblock_inc
make variable.
BUG=chrome-os-partner:44827
BRANCH=None
TEST=Built vortex, rambi, and some asus boards.
Change-Id: Ida4571cbe6eed65e77ade98b8d9ad056353c53f9
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/11495
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
As acpi_write_hpet() uses CONFIG_HPET_ADDRESS in the HPET table we
need to use CONFIG_HPET_ADDRESS when assigning it to the device.
Change-Id: I656f917658f1c1717bb3653fa048a6d36fca2454
Signed-off-by: Jonathan A. Kollasch <jakllsch@kollasch.net>
Reviewed-on: http://review.coreboot.org/10925
Tested-by: build bot (Jenkins)
Reviewed-by: Idwer Vollering <vidwer@gmail.com>
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Based on the observations that AMD Fam10h with both Nvidia CK804 (Asus
KFSN4-DRE) and MCP55 (Sun Ultra 40 M2) need to avoid adjusting the LPC
bridge register 0x78 (particularly the 0x7b byte) to get to ramstage:
Assume that there's something about this register that adjusting it the
way we do for K8 is something that can/should be universally avoided on
all Fam10h systems with these chipsets.
Change-Id: I1eceeb20ecaefef4c61c11e19d1f5a59f91a0a2f
Signed-off-by: Jonathan A. Kollasch <jakllsch@kollasch.net>
Reviewed-on: http://review.coreboot.org/10984
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Should fix regression in HDA verb setup on nvidia mcp55 and intel sch
southbridges. The mcp55 code could not find the mainboard's verb table
because the table was not even being compiled in. The sch boards appeared
to have the same issue.
Intel broadwell and fsp_bd82x6x seemed to have not gotten the boilerplate
shrink, so apply it to those too.
Followup-to: Ib3e09644c0ee71aacb067adaa85653d151b52078
(azalia: Shrink boilerplate)
Change-Id: If7aae69f5171db67055ffe220bdff392caaa5d9f
Signed-off-by: Jonathan A. Kollasch <jakllsch@kollasch.net>
Reviewed-on: http://review.coreboot.org/10826
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
`device_t device` is missing as argument. Every device_op function
should have a `device_t device` argument.
Change-Id: I7fca8c3fa15c1be672e50e4422d7ac8e4aaa1e36
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
Reviewed-on: http://review.coreboot.org/9598
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Use of scan_static_bus() and tree traversals is somewhat convoluted.
Start cleaning this up by assigning each path type with separate
static scan_bus() function.
For ME, SMBus and LPC paths a bus cannot expose bridges, as those would
add to the number of encountered PCI buses.
Change-Id: I8bb11450516faad4fa33b8f69bce5b9978ec75e5
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/8534
Tested-by: build bot (Jenkins)
Reviewed-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
All boards now use per-device ACPI. This patch finishes migration
by removing transitional kludges.
Change-Id: Ie4577f89bf3bb17b310b7b0a84b2c54e404b1606
Signed-off-by: Vladimir Serbinenko <phcoder@gmail.com>
Reviewed-on: http://review.coreboot.org/7372
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
As per discussion with lawyers[tm], it's not a good idea to
shorten the license header too much - not for legal reasons
but because there are tools that look for them, and giving
them a standard pattern simplifies things.
However, we got confirmation that we don't have to update
every file ever added to coreboot whenever the FSF gets a
new lease, but can drop the address instead.
util/kconfig is excluded because that's imported code that
we may want to synchronize every now and then.
$ find * -type f -exec sed -i "s:Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, *MA[, ]*02110-1301[, ]*USA:Foundation, Inc.:" {} +
$ find * -type f -exec sed -i "s:Foundation, Inc., 51 Franklin Street, Suite 500, Boston, MA 02110-1335, USA:Foundation, Inc.:" {} +
$ find * -type f -exec sed -i "s:Foundation, Inc., 59 Temple Place[-, ]*Suite 330, Boston, MA *02111-1307[, ]*USA:Foundation, Inc.:" {} +
$ find * -type f -exec sed -i "s:Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.:Foundation, Inc.:" {} +
$ find * -type f
-a \! -name \*.patch \
-a \! -name \*_shipped \
-a \! -name LICENSE_GPL \
-a \! -name LGPL.txt \
-a \! -name COPYING \
-a \! -name DISCLAIMER \
-exec sed -i "/Foundation, Inc./ N;s:Foundation, Inc.* USA\.* *:Foundation, Inc. :;s:Foundation, Inc. $:Foundation, Inc.:" {} +
Change-Id: Icc968a5a5f3a5df8d32b940f9cdb35350654bef9
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/9233
Tested-by: build bot (Jenkins)
Reviewed-by: Vladimir Serbinenko <phcoder@gmail.com>
This change switches all southbridge vendors and southbridges
to be autoincluded by Makefile.inc, rather than having to be
mentioned explicitly in southbridge/Makefile.inc or in
southbridge/<vendor>/Makefile.inc.
In order to be able to drop southbridge/amd/Makefile.inc, some
scattered source files had to be moved to a southbridge/amd/common
directory, in accordance to what we are doing on other architectures
already.
This means, vendor and southbridge directories are now "drop
in", e.g. be placed in the coreboot directory hierarchy
without having to modify any higher level coreboot files.
The long term plan is to enable out of tree components to be
built with a given coreboot version (given that the API did not
change).
Change-Id: I79bd644a0a3c4e8320c80f8cc7a7f8ffd65d32f2
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: http://review.coreboot.org/9796
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Instead of keeping this separate variable around, add linker scripts
to the $(class)-y source lists and let the build system sort things out.
This is inspired by the commit listed below, but rewritten to match
upstream, and split in smaller pieces to keep intent clear.
Change-Id: I4af687becf2971e009cb077debc902d2f0722cfb
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Based-On-Change-Id: I50af7dacf616e0f8ff4c43f4acc679089ad7022b
Based-On-Signed-off-by: Julius Werner <jwerner@chromium.org>
Based-On-Reviewed-on: https://chromium-review.googlesource.com/219170
Reviewed-on: http://review.coreboot.org/9289
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
We have .lb, .lds, and .ld in the tree. Go for .ld everywhere.
This is inspired by the commit listed below, but rewritten to match
upstream, and split in smaller pieces to keep intent clear.
Change-Id: I3126af608afe4937ec4551a78df5a7824e09b04b
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Based-On-Change-Id: I50af7dacf616e0f8ff4c43f4acc679089ad7022b
Based-On-Signed-off-by: Julius Werner <jwerner@chromium.org>
Based-On-Reviewed-on: https://chromium-review.googlesource.com/219170
Reviewed-on: http://review.coreboot.org/9107
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
On x86, change the type of the address parameter in
read8()/read16/read32()/write8()/write16()/write32() to be a
pointer, instead of unsigned long.
Change-Id: Ic26dd8a72d82828b69be3c04944710681b7bd330
Signed-off-by: Kevin Paul Herbert <kph@meraki.net>
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/7784
Tested-by: build bot (Jenkins)
Because we had no stack on romcc boards, we had a separate, not as
powerful clone of printk: print_*. Back in the day, like more than
half a decade ago, we migrated a lot of boards to printk, but we never
cleaned up the existing code to be consistent. instead, we worked around
the problem with a very messy console.h (nowadays the mess is hidden in
romstage_console.c and early_print.h)
This patch cleans up the southbridge code to use printk() on all non-ROMCC
boards.
Change-Id: I312406257e66bbdc3940e206b5256460559a2c98
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: http://review.coreboot.org/8110
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Tested-by: build bot (Jenkins)
Most of the code related to the mc146818 is not related to the RTC and is
really for managing the CMOS storage. Since we intend to add a generic API
for RTC drivers it's inconvenient for those functions to have an rtc_ prefix.
This CL renames those functions so they start with cmos_ instead. There are
some places where rtc_init was called with a comment that says something about
starting the RTC. That wasn't correct before (the RTC is always running), but
it looks a little odd now that the function is called cmos_init.
This CL also opportunistically cleans up some style problems in this file.
Signed-off-by: Gabe Black <gabeblack@google.com>
Reviewed-on: https://chromium-review.googlesource.com/197794
Reviewed-by: Gabe Black <gabeblack@chromium.org>
Tested-by: Gabe Black <gabeblack@chromium.org>
Commit-Queue: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit 9a9ad24888b185fb58965457704e326bb508d788)
Removed the addition of stdint.h to mc146818rtc.h since
types.h is now included. Changed rtc_init to cmos_init for
fsp_bd82x6x, fsp_rangeley, fsp_baytrail, ibexpeak, vortex86ex.
Change-Id: Id4b9f6bea93e8bd5eaef2cb17f296adb9697114c
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6977
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
As currently many systems would be barely functional without ACPI,
always generate ACPI tables if supported.
Change-Id: I372dbd03101030c904dab153552a1291f3b63518
Signed-off-by: Vladimir Serbinenko <phcoder@gmail.com>
Reviewed-on: http://review.coreboot.org/4609
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
Also relocate and split header files, there is some interest
for EHCI debug support without PCI.
Change-Id: Ibe91730eb72dfe0634fb38bdd184043495e2fb08
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/5129
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
The main purpose of option rom is to supply int* handlers.
But supplying those is outside of coreboot scope and if someone needs those
they should run SeaBIOS anyway which runs the option roms wonderfully.
Running VGA oprom is kept because they're needed to init graphics.
This patch still keeps the options to include the option roms to make them
available to SeaBIOS.
Change-Id: I646334cf88094d3bf8f527779a68a07e0b4b93ec
Signed-off-by: Vladimir Serbinenko <phcoder@gmail.com>
Reviewed-on: http://review.coreboot.org/4545
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-by: Kevin O'Connor <kevin@koconnor.net>
The main usbdebug file lib/usbdebug.c was removed from romstage
build with commit f8bf5a10 but the chipset-specific parts were not,
leading to unresolved symbol errors for AMD platforms.
Add a silent Kconfig variable USBDEBUG_IN_ROMSTAGE for convenient
use of this feature.
Change-Id: I0cd3fccf2612cf08497aa5c3750c89bf43ff69be
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3983
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
Nowadays, chipsets or boards do not only have one USB port with the
capabilities of a debug port but several ones. Some of these ports are
easier accessible than others, so making them configurable is also necessary.
This change adds infrastructure to switch between EHCI controllers,
but does not implement it for any chipset.
Change-Id: I079643870104fbc64091a54e1bfd56ad24422c9f
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3438
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
On AMD platforms, setting of USBDEBUG_DEFAULT_PORT=0 tries to scan
all physical ports one after other in incrementing order. To avoid
possible problems with other USB devices, one can select the port
number here and bypass the scan.
Intel platforms can communicate with usbdebug dongle on one
physical port only, and this option makes no difference there.
Change-Id: I45be6cc3aa91b74650eda2d444c9fcad39d58897
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3872
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
Declare the functions that may be used in both romstage and ramstage
with simple device model. This will later allow to define PCI access
functions for ramstage using the inlined functions from romstage.
Change-Id: I32ff622883ceee4628e6b1b01023b970e379113f
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3508
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
Letting SMI handler touch EHCI controller is an excellent source
of USB problems. Remove usbdebug entirely from SMM.
It may be possible to make usbdebug console work from SMM
after hard work and coordination with payloads and even
OS drivers. But we are not there.
Change-Id: Id50586758ee06e8d76e682dc6f64f756ab5b79f5
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3858
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Tested-by: build bot (Jenkins)
With USBDEBUG selected, the file is built for both romstage and
ramstage. For the ramstage build, we need to explicitly use the
simple PCI config operations without devicetree.
Change-Id: I2de8d9c77bb458ba797c3aac9e2cd0d653e06684
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3437
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
All the additional work that needs to be done in EHCI BAR relocation
is independent of the hardware platform and was functionally identical
in all the copies removed.
When USBDEBUG is not selected, PCI EHCI controllers use standard
pci_dev_read_resources() call.
With USBDEBUG selected, PCI EHCI controller's device_operations
.read_resources is replaced with pci_ehci_read_resources() call,
which in turn will replace the device_operations .set_resources call.
The replacement for .set_resources reconfigures usbdebug driver side,
and calls the original .set_resources to configure hardware side.
Change-Id: I8e136a5da4efedf60b6dd7068c0488153efaaf8e
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3412
Reviewed-by: Aaron Durbin <adurbin@google.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Commit "romcc: Don't fail on function prototypes" (11a7db3b) [1]
made romcc not choke on function prototypes anymore. This
allows us to get rid of a lot of ifdefs guarding __ROMCC__ .
[1] http://review.coreboot.org/2424
Change-Id: Ib1be3b294e5b49f5101f2e02ee1473809109c8ac
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/3216
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Julius Werner