GitLab

Doing certain operations may need uploading and executing code
on the device. For example, such operations right now are
reading/writing ARM CP15 coprocessor registers. Uploading the
code to the device is naturally overwriting some part of SRAM
as a side effect. Right now it is not a problem, because the
CP15 coprocessor registers are only accessed as part of uploading
and executing U-Boot SPL. They are nicely timed not to cause
problems and the temporary scratch area gets overwritten by
the SPL code anyway.

But if we decide to provide access to such operations via
command line interface, then any side effects may potentially
cause problems for the users. Consider the following scenario:

  sunxi-fel clear 0x2000 0x100 \
            advanced-command-which-uploads-and-executes-code \
            hexdump 0x2000 0x100

The user may rightfully expect that clearing a buffer in SRAM
to zero and then reading it back should show all zero bytes.
But inserting advanced commands in the middle may cause data
corruption.

In order to resolve this problem, just move the scratch area
away from the 0x2000-0x5BFF addresses range. These particular
addresses are already known to the users as a safe place for
their bare metal expariments in FEL mode. The "sunxi-fel spl"
command is a special case though and it is expected to
overwrite data in this area too.

A possible alternative would be to just backup & restore data
in the scratch area. But this has some disadvantages:
  1. Extra code in the sunxi-fel tool and extra roundtrips over
     USB to do the backup/restore job.
  2. If we allow the OpenRISC core to use the 0x2000-0x5C00
     range in SRAM A1, then this becomes unsafe and racy
     (we can't really backup & restore data without causing
     a temporarily glitch for the currently running code on
     the OpenRISC core).

To sum it up. With this patch we make it so that now the
0x2000-0x5BFF range is freely available for the users of the
sunxi-fel tool. The 0x1000-0x1FFF range is off limits (the
upper part of it is used by the FEL IRQ handler, the lower
part of it is reserved for internal use by the sunxi-fel
tool). The 0x0000-0x0FFF addresses range is reserved for
passing data from the SPL to the main U-Boot binary (via
the SPL header) and is also off limits.
Signed-off-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>
Tested-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Reviewed-by: Bernhard Nortmann <bernhard.nortmann@web.de>

This is done by explicitly setting at least one member field
(preferably the one that gets tested for validity later),
instead of just using "{ 0 }".
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>

Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>

For Linux, this requires additional #defines, see
https://gist.github.com/panzi/6856583#gistcomment-1656524



Placing those #defines into portable_endian.h would mean we
depend on the specific order of #includes (whatever pulls
in endian.h first, including other system includes). Avoid
this by using symbols provided "globally" via the Makefile.

Also nuke the _NETBSD_SOURCE definition in fel.c
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>

Our previous include file only supplied a very limited set of
wrapper redefinitions for platforms that define __APPLE__ (i.e.
Mac OS). In particular, some needed big endian conversions were
missing, causing "symbol(s) not found" errors when linking.

Instead of patching the existing file, let's do away with it
completely and replace it by something more sophisticated.
The portable_endian.h used is in the public domain, and was
taken from https://gist.github.com/panzi/6856583

Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>

Commit 472ac475

 introduced
a possible regression by relying on the availability of
libusb_strerror(). There are libusb versions out there _not_
offering this function, which breaks compilation.

Introducing a separate helper function allows us to work around
this, refactor existing code and have more streamlined error
reporting - avoiding the message string retrieval where needed.
The usb_error() function can also optionally abort execution
with a given exit code.
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>

Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>

Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>

Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>

instead of memset() and strcpy() to clear / assign member fields.
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>

Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>

Gracefully shut down the FEL utility, instead of continuing
execution. This avoids a possible "ERROR: Allwinner USB FEL
device not found!" when doing a plain `./sunxi-fel` without
any arguments (and no suitable USB device connected).
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>

The A80 has the V bit in SCTLR set to 0 because of the BROM
being at 0x00000000 now, so the SCTLR check has to be relaxed.
Signed-off-by: Jens Kuske <jenskuske@gmail.com>
Reviewed-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

This workaround is necessary for A80, which sometimes
fails at reading DACR.
Signed-off-by: Jens Kuske <jenskuske@gmail.com>
Reviewed-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

See https://github.com/linux-sunxi/sunxi-tools/issues/37

The patch was originally inspired by
https://github.com/NextThingCo/sunxi-tools/commit/16386a7
and
https://github.com/NextThingCo/sunxi-tools/commit/47bafaf



It introduces a "--dev" (-d) option to specify the desired FEL
device. This is useful if multiple target devices are connected
to the same host.
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Reviewed-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

Closes #24. See https://groups.google.com/forum/#!topic/linux-sunxi/sEJ1eTBoCvw

Tested-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Reviewed-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

The aw_fel_process_spl_and_uboot() was missing a free() for the file
buffer. This patch also adds a proper libusb cleanup/shutdown.
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Reviewed-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

This patch forces relevant options to be placed at the beginning of
the command line, but also makes sure that their specific order does
not matter.
Currently this applies to "-p" (--progress) and "-v" (--verbose).
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Reviewed-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

Commit d99d59c0

 introduces a bug when the "spl" command is used with a
file that contains a U-Boot binary. "spl" will now try to execute it,
while it's not supposed to. (The command should only start the SPL.)

This patch brings back the `uboot_autostart` flag, which is meant to
remain `false` for the "spl" command, and gets set by "uboot" only.
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Reviewed-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

This patch adds an "echo-gauge" command that is intended for use with
sunxi-fel's output piped into the dialog utility. "echo-gauge" will
output its string argument in a way that updates (i.e. overwrites)
the prompt text that dialog displays, so it's possible to change that
inbetween file transfers from sunxi-fel.

Example:
	sunxi-fel uboot ${UBOOT} \
		echo-gauge "Uploading kernel (1/2)" \
		write-with-gauge 0x42000000 ${KERNEL} \
		write 0x43100000 ${SCRIPT} \
		echo-gauge "Uploading rootfs (2/2)" \
		write-with-gauge 0x44000000 ${ROOTFS} \
		| dialog --gauge "" 6 70
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Reviewed-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

This patch adds some additional commands, and implements corresponding
progress callbacks that produce output for the 'dialog' utility:
http://invisible-island.net/dialog/dialog.html



The simple "*-with-gauge" just emits percentage numbers, line by line,
while "*-with-xgauge" outputs extended information to update the
dialog prompt. Both are meant for directly piping the output:
sunxi-fel write-with-gauge <...> | dialog --gauge "FEL upload" 6 70
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Acked-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

The two new commands both imply that a (simple) progress bar
display is requested. "multi[write]" allows to upload a group
of files with a single command, resulting in their transfer
showing a common, 'overall' progress status.
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Acked-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

This patch moves some code out of the "write" handler, with the intention
to reuse it later for similar commands.

At the same time, it also discards the previous transfer rate output when
using "-v", as that has become redundant with the 'progress bar' display
available by "-p".
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Acked-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

Add "--progress" option and a progress bar display for FEL
transfers. This picks up on a suggestion from Alexander Kaplan
and the discussion at
https://groups.google.com/forum/#!topic/linux-sunxi/lz0oQBwjex0

Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Reviewed-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

This patch adds a boolean parameter "progress" to several functions.
When set (true), it will cause usb_bulk_send() to notify the progress
framework, using progress_update() to convey the current transfer
status.

The feature gets enabled for the "write" command, and is forced
inactive in most other places, e.g. aw_fel_write_uboot_image().
Also, we specifically want the internal use of aw_fel_write() to
never use it.
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Reviewed-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

This patch introduces a new function aw_write_buffer() to
allow a better distinction between FEL write operations.

aw_fel_write() is meant for "privileged" internal use, while
aw_write_buffer() now represents the preferred entry point
for user code like the "write", "fill" or "clear" commands.

There is some deliberate code duplication here that makes sense
when combined with the next patch, where the aw_usb_write()
call will differ. We want aw_fel_write() to enforce no progress
update/callback, while aw_write_buffer() will (optionally)
support it.
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Reviewed-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

This patch adds two new source files that will deal with code
related to progress callbacks and display. I have decided to
keep this separate, as there will be many smaller routines
involved, which otherwise would bloat fel.c unnecessarily.

For starters, let's also move the gettime() function there.
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Reviewed-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

When the CPU clock speed is set to 480 MHz by the U-Boot SPL,
the performance of 'sunxi-fel write' transfers to DRAM improves
from ~95 KB/s to ~510 KB/s.

When the CPU clock speed is set to 1008 MHz by the U-Boot SPL,
the performance of 'sunxi-fel write' transfers to DRAM improves
from ~180 KB/s to ~510 KB/s.

This means that the CPU is not a bottleneck for FEL transfers
anymore. Further performance improvements are possible by
increasing the AHB1 clock speed in the U-Boot SPL (up to
something like ~900 KB/s).
Signed-off-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>
Acked-by: Peter Korsgaard <peter@korsgaard.com>

The BROM in newer SoC variants doesn't enable MMU by default anymore.
So in order to benefit from e4b3da2b


("fel: Faster USB transfers via 'fel write' to DRAM"), we need to
be able to enable it from the 'sunxi-fel' tool.

This patch can be interpreted as simply reverting the changes done
by Allwinner and bringing back the MMU support in roughly the same
way as it was before. That's why the values in the hardware
registers and the translation table entries replicate the A20 setup.

Additionally, the code is now more defensive and introduces new
"canary" checks for certain known magic values in the coprocessor
registers in order to safeguard against any unpleasant surprises.

MMU tuning for A80 and A64 will probably need a more sophisticated
setup with a second level page table. Because both the SRAM and
the BROM reside in the same 1MB section there and we need finer
granularity. In other words, enabling the MMU on A80 and A64 is
not supported yet.
Signed-off-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>
Acked-by: Peter Korsgaard <peter@korsgaard.com>

This helps to reduce code duplication if we want to read and write
many different kinds of coprocessor registers.
Signed-off-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>
Acked-by: Peter Korsgaard <peter@korsgaard.com>

Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Acked-by: Peter Korsgaard <peter@korsgaard.com>

- get rid of double includes
- fix U-Boot source reference
- get rid of "uboot_autostart" variable
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Acked-by: Peter Korsgaard <peter@korsgaard.com>

This patch makes use of a new sunxi SPL header format, where
fields for passing data to U-Boot are reserved. This allows the
"fel" utility to provide specific pieces of information by
setting these fields (which will be evaluated by U-Boot then).

For now, we provide the memory location of the boot script file
(boot.scr) this way. A suitably modified U-Boot can adjust the
boot process accordingly and will auto-execute the script.
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Acked-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

Information about the SoC-specific memory layout may be required across
several places in the fel utility code. To avoid having to retrieve this
information repeatedly, this patch modifies the aw_fel_get_sram_info()
function to cache its result (via an internal static pointer).

This makes it safe to call aw_fel_get_sram_info() wherever needed, while
avoiding additional aw_fel_get_version() lookups, and potential surplus
"no 'soc_sram_info' data for your SoC" warnings.
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Acked-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

This patch moves retrieving the image header type into a function
of its own. The idea is that fel "write" can and will also make
use of this function to detect boot scripts (by IH_TYPE_SCRIPT).
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Acked-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

With glibc 2.20 (and newer), defining _BSD_SOURCE (only)
will now generate compiler warnings similar to:

/usr/include/features.h:148:3: warning: #warning "_BSD_SOURCE
and _SVID_SOURCE are deprecated, use _DEFAULT_SOURCE" [-Wcpp]

See the glibc 2.20 release notes, https://lwn.net/Articles/611162/

.

This patch adds the required _DEFAULT_SOURCE.
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Acked-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de>
Acked-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>

Add a new field 'spl_addr' to the SoC description structure and adjust
the code to honor it. This is needed for supporting Allwinner A80.

Tested on Allwinner A20 by changing the 'spl_addr' to 0x28000 in the
'fel' tool and using a custom build of U-Boot (CONFIG_SPL_TEXT_BASE
changed from 0x20 to 0x28020).
Signed-off-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>
Acked-by: Hans de Goede <hdegoede@redhat.com>

The safe area to upload and execute code via FEL may be different
for different SoC variants (0x12000 for A80, 0x2000 for everything
else), so we can't have it as a global compile time constant, but
need need to treat it as a property of the SoC.

This patch changes the code to pass the SoC info structure to a
number of functions and adds a new 'scrath_addr' struct field to it.
Signed-off-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>
Acked-by: Hans de Goede <hdegoede@redhat.com>

This is may be useful when adding support for new SoC variants.
We want to be sure that the IRQ stack and the regular stack both
reside at predictable locations in SRAM and such extra diagnostic
message can confirm this.
Signed-off-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>
Acked-by: Hans de Goede <hdegoede@redhat.com>