Signed-off-by: Nico Huber <nico.h@gmx.de>

The 32-bit library is called libi386, not libx86. Fixes building with
Meson on i386 NetBSD.

Change-Id: Ib9e1aa16558dc24f0bf14dd573934cf5164d67e8
Signed-off-by: Nico Huber <nico.h@gmx.de>
Original-Reviewed-on: https://review.sourcearcade.org/c/flashprog/+/129
Original-Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.sourcearcade.org/c/flashprog/+/146

Change-Id: Idc6f442361ea3860d7ad3932ad1b72c6ff56ba45
Signed-off-by: Nico Huber <nico.h@gmx.de>
Original-Reviewed-on: https://review.sourcearcade.org/c/flashprog/+/130
Original-Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.sourcearcade.org/c/flashprog/+/147

Same as the Makefile does with $(DEPENDS_ON_RAW_MEM_ACCESS), we should
tag all the programmers that require raw memory access. Otherwise, we
might compile these drivers for architectures that require explicit
memory barriers that we haven't implemented.

As there seems to be no notion of intersection in Meson, we have to make
`cpu_families' a list of lists and check them all.

Change-Id: I3bd6bd7b5cb871a506178496241bacdeb40febcf
Signed-off-by: Nico Huber <nico.h@gmx.de>
Original-Reviewed-on: https://review.sourcearcade.org/c/flashprog/+/131
Original-Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.sourcearcade.org/c/flashprog/+/148

Signed-off-by: Nico Huber <nico.h@gmx.de>

Signed-off-by: Nico Huber <nico.h@gmx.de>

The opmenu is not authoritative when we can re-program it on the fly.
Hence always return true in ich_spi_probe_opcode() when it isn't locked.

Tested by `cobra` on IRC.

Change-Id: I636f00acbdfa1661c13c2d82d7d7e61fbe6b543b
Signed-off-by: Nico Huber <nico.h@gmx.de>
Original-Reviewed-on: https://review.sourcearcade.org/c/flashprog/+/151
Original-Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.sourcearcade.org/c/flashprog/+/166

…ces"

This reverts commit a32d47f.  The idea
was right: In theory, these Dediprogs can support 4BA modes. Due to the
lack of test hardware, however,  nobody actually implemented it for the
newer protocol versions.

So make sure once again, that we bail out before trying to write to the
flash with wrong command sequences.

Change-Id: Ic5cb426f9b8ed8e1df803387a0d1a22909707874
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.sourcearcade.org/c/flashprog/+/167