Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

ignore errors, the application will detect whatever is missing
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

The spec carrier can easy take the dma_device without knowing the DMA
details implemented in its driver.
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

The driver was using the wrong buffer
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

CONFIG_FPGA_MGR_BACKPORT_PATH was set but not use, the actual variable
to be used is FPGA_MGR
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

This reverts commit ac524017.

A fix is necessary but this is the wrong place, next patch will fix
it properly.

CONFIG_FPGA_MGR_BACKPORT_PATH is used only in combination with DKMS

Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

Signed-off-by: Dimitris Lampridis <dimitris.lampridis@cern.ch>

The issue comes from the fact that boolean'image() returns lowercase "true"/"false" strings, which
are then not properly matched by the DDR controller core. To fix this, we pull the to_upper()
function from general-cores and use it to convert the output of bolean'image() to uppercase before
passing it to the DDR controller.
Signed-off-by: Dimitris Lampridis <dimitris.lampridis@cern.ch>

spec_fpga_app_init() declares too many local buffers that are allocated on the stack.

On some versions of GCC this is detected and reported, causing a warning that is promoted to an
error. Example:

spec-core-fpga.c:792:1: error: the frame size of 1888 bytes is larger than 1024 bytes
[-Werror=frame-larger-than=]

Since apparently it is not a very good idea to increase the stack frame size limit [1], the solution
implemented here is to switch to dynimically allocating the biggest buffer (struct resource res[]).

[1]: https://gcc.gnu.org/onlinedocs/gccint/Stack-Checking.html#Stack-Checking
"The maximum size of a stack frame, in bytes. GCC will generate probe instructions in non-leaf
functions to ensure at least this many bytes of stack are available. If a stack frame is larger than
this size, stack checking will not be reliable and GCC will issue a warning. The default is chosen
so that GCC only generates one instruction on most systems. You should normally not change the
default value of this macro."
Signed-off-by: Dimitris Lampridis <dimitris.lampridis@cern.ch>

Signed-off-by: Dimitris Lampridis <dimitris.lampridis@cern.ch>

Signed-off-by: Dimitris Lampridis <dimitris.lampridis@cern.ch>

Signed-off-by: Dimitris Lampridis <dimitris.lampridis@cern.ch>

Signed-off-by: Dimitris Lampridis <dimitris.lampridis@cern.ch>

Kernel v4.0 rehauled the DMA engine. See also:
https://www.kernel.org/doc/html/latest/driver-api/dmaengine/provider.htmlSigned-off-by: Dimitris Lampridis <dimitris.lampridis@cern.ch>

Signed-off-by: Dimitris Lampridis <dimitris.lampridis@cern.ch>

Signed-off-by: Dimitris Lampridis <dimitris.lampridis@cern.ch>

Signed-off-by: Dimitris Lampridis <dimitris.lampridis@cern.ch>

Signed-off-by: Dimitris Lampridis <dimitris.lampridis@cern.ch>