rsa: Allow to directly compile in rsa public keys ...

So far we relied on the U-Boot mkimage tool to generate us device tree
snippets containing rsa public keys which we then compiled into barebox.
Make this easier and allow to directly specify a filename or PKCS#11 URI
in Kconfig. With this we no longer need the U-Boot mkimage tool here and
no more external steps to prepare device tree snippets.

With this rsa public keys can be directly compiled as C structs into
barebox which is much more direct than putting it into the device tree.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

rsa: let rsa_of_read_key() return a fully allocated key ...

Until now rsa_of_read_key() took a pointer to a key and filled the
struct rsa_public_key members with allocated values. So far we have
never freed these values. Change rsa_of_read_key() to always return
a fully allocated key and provide rsa_key_free() to free it. Let the
FIT image code free the key after usage.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

scripts: Add rsatoc tool ...

The rsatoc tool converts rsa public keys into C structs suitable to
compile with barebox. Most of the openssl rsa related stuff has been
taken from the U-Boot mkimage tool.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Kbuild: Add config_filename macro from kernel ...

The config_filename allows to extract a filename from a Kconfig string
option.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

lds: Add and use RO_DATA_SECTION macro ...

We have many different pointer arrays which we put into linker sections
and each time there's one added we have to adjust all linker scripts.
This adds a common RO_DATA_SECTION define and uses it for all
architectures. This makes it easier to add a new linker array.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

lds: remove more unnecessary alignments ...

The alignment in the DTB and clk_of_table shouldn't be needed. Remove
them.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Sam Ravnborg <sam@ravnborg.org>

X86: lds: remove unnecessary alignments ...

We do not seem to need all these alignments in the rodata sections, so
remove them.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

lds: create and use BAREBOX_PCI_FIXUP macro ...

No need to repeat the pci fixup sections in each linker script. Add a
define for it.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

lds: Move start/end address variables into defines ...

For the older section defines we specify the start and end addresses
outside the macro which means we have to repeat them in each linker
script. Make the start/end addresses part of the define to simplify
things. While at it, add a BAREBOX_ prefix to the INITCALLS and EXITCALLS
macros for consistency to the other defines.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Sam Ravnborg <sam@ravnborg.org>

ppc: remove unused variables from linker scripts ...

__initcall_entries and __exitcall_entries are unused. Remove them.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

barebox.lds: Remove unnecessary braces ...

The BAREBOX_CLK_TABLE and BAREBOX_DTB macros are defines that do not
take a parameter, so we can remove the braces.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

command: Use array of pointers to commands ...

We used to store the commands as a linker array. One problem with this
is that on X86_64 for unknown reasons the linker uses a different struct
alignment than the compiler, so when we use the linker to compose the
array and the compiler to iterate over it we have to play tricks with
manually adjusting the alignment. The other problem is that we declare
the commands as const (and also put it in .rodata), but in fact we do
not treat it as const: we put the commands onto a list which modifies
the struct list_head list member of struct command.

With this patch we no longer put the command themselves into an array,
but instead create an array of pointers to the commands. This inherently
solves the second issue as well.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ARM: aarch64: add image markers to linker script ...

The __image_start and __image_end marker are used by barebox to
calculate the size of the barebox binary inside the RAM. Without the
definitions barebox calculates its own size to zero, which works until a
certain size is exceeded, than the PBL will place the binary lower into
memory than the calculation inside barebox.
Fix this by adding the appropriate image markers to the linker script.

Signed-off-by: Rouven Czerwinski <r.czerwinski@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

input: specialkeys: remove unused printk argument ...

Two pr_info calls have no format specifiers, but are still passed an
argument. Remove the unused arguments.

Signed-off-by: Ahmad Fatoum <ahmad@a3f.at>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mci: imx-esdhc-pbl: Fix watermark level value for i.MX ...

Layerscape and i.MX have different semantics of Watermark Level
Register. Whereas the former uses "0" to signify maximum allowed
value, the latter does not.

According to the RM (i.MX8MQ, i.MX6):

"...The read burst length must be less than or equal to the read
watermark level.."

Setting Watermark Level Register to zero violates that limitation. It
appears that, on i.MX8MQ, not following that rule causes certain
configs + toolchains to result in non-bootable image. Specifically,
polling for CICHB, CIDHB and DLA to clear in esdhc_send_cmd() times
out. There doesn't appear to be any clear relationship as to what kind
of image will have the problem, but the following combinations failed
to boot on ZII i.MX8MQ Zest board:

  - gcc version 9.2.1 20190827 (Red Hat Cross 9.2.1-1) (GCC) +
    imx_v8_defconfig + CONFIG_DEBUG_LL and CONFIG_PBL_CONSOLE

  - gcc version 5.5.0 (Timesys 20190405) (custom toolchain) +
    imx_v8_defconfig

Setting WML's *_BRST_LE to 16 and *_WML to 128 on i.MX resolves the
issue (same setting that's selected by writing 0 on Layerscape).

Fixes: 48562aeaa8 ("esdhc-xload: check for PRSSTAT_BREN only after each block")
Cc: Chris Healy <cphealy@gmail.com>
Cc: Ruslan Sushko <ruslan.sushko@zii.aero>
Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Tested-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: nand: nand_omap_gpmc: Fix mtd_info usage ...

The mtd_info has not been set correctly in the omap gpmc probe.
Remove the mtd_info in the private struct like it has been done in
other nand drivers. And update the driver accordingly.

Fixes: 76b6f74c5397 ("mtd: nand: Embed struct mtd_info into struct nand_chip")

Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: nand: Fix retrieving nand_chip from mtd_info ...

With 76b6f74c53 we begin embedding a struct mtd_info into struct
nand_chip. With this mtd_to_nand() must be changed accordingly.

Fixes: 76b6f74c53 ("mtd: nand: Embed struct mtd_info into struct nand_chip")
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

clk: i.MX6qp: Fix location of the enfc_sel mux ...

On the i.MX6qp the enfc_sel mux is at bits 15-17, not on 16-17. Fix
this. During clock tree initialization we reparented the enfc_sel to:

	clk_set_parent(clks[IMX6QDL_CLK_ENFC_SEL], clks[IMX6QDL_CLK_PLL2_PFD2_396M]);

This resulted in a register setting 0b110 for the enfc_sel mux which is
reserved. Apparently this reserved setting resulted in the enfc clock
being driven from pll3_pfd3_454m. This means our enfc clock was the
factor 454/396 too high.

With b534f79112f0 ("clk: imx6: Fix procedure to switch the parent of
LDB_DI_CLK") we happened to disable pll3_pfd3_454m during init, so with
this commit NAND stopped working entirely on i.MX6qp.

Both issues are fixed with this patch

Fixes: b534f79112f0 ("clk: imx6: Fix procedure to switch the parent of LDB_DI_CLK")
Fixes: 92fd61d12723 ("clk: i.MX6: Fix enfc_sel for i.MX6dqp")

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

bootm: release tee_res ...

When a tee_res has been acquired it must be released when bootm returns
(for dryrun)

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Reviewed-by: Rouven Czerwinski <r.czerwinski@pengutronix.de>

gcc-9: silence 'address-of-packed-member' warning ...

As done in the Linux Kernel. Most of the packed structures we have
are optimized to not generate any unaligned accesses, so this warning
produces a lot of false positives which are hard to fix in the code.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Merge branch 'for-next/tftp'

Merge branch 'for-next/stm32'

Merge branch 'for-next/socfpga'

Merge branch 'for-next/sandbox'

Merge branch 'for-next/reset-source'

Merge branch 'for-next/qemu'

Merge branch 'for-next/omap'

Merge branch 'for-next/net'

Merge branch 'for-next/mtd'

Merge branch 'for-next/misc'