GitLab

This is mostly achieved by fixing or refining the inclusion of header
files and replacing glibc-specific ifdefs with more generic ones.

 - <sys/io.h>: Contains iopl(2) and x86 I/O port access functions (inb, outb etc).
               Generally Linux-specific but also availble on debian/kFreeBSD.
               Provided by glibc as well as musl and uclibc.
               Include it if we are running Linux or if glibc is detected.
 - <sys/fcntl.h>: should be (and is) replaced by <fcntl.h> (without the
    			  "sys" prefix).
 - <linux/spi/spidev.h>: Does not include all necessary headers, namely
                         _IOC_SIZEBITS that is used in the definition of
                         SPI_MSGSIZE is not brought in via <linux/ioctl.h>
                         but instead we relied so far on glibc's including
                         it via <sys/ioctl.h>. Change that to explicitly
                         including <linux/ioctl.h>.
 - <endian.h>: Would also be available in musl but there is no easy way
               to detect it so we do not try yet.

Corresponding to flashrom svn r1898.

The bug report and initial patches were
Signed-off-by: Gwenhael Goavec-Merou <gwenhael.goavec-merou@trabucayre.com>
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Some chips such as the ENE KB9012 internal flash require a write
granularity of 128 bytes.

Corresponding to flashrom svn r1897.
Signed-off-by: Paul Kocialkowski <contact@paulk.fr>
Acked-by: Nico Huber <nico.h@gmx.de>

Use names for the commands and request types instead of magic
numbers and remove some of the unnecessary unexplained arguments.
Also, cleanup the nonsense code left over from RE. Most of it can
not be explained by official documentation and was recorded with
ancient firmware/software.

Based on the following chromiumos changes:
Change-Id: I80a0dcdf40eedc89da48fb2c54cd9d9fd13e6fa1
Change-Id: If61bac2c8194b3ec30a80422d871842c66f0cd74

Corresponding to flashrom svn r1896.
Signed-off-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: David Hendricks <dhendrix@chromium.org>

The serprog protocol does only transmit 24 bit-wide address and ignores the
top 8 bit. This is fine as long as the underlying hardware ignores the latter
anyway (which is the case for parallel chips that even lack the respective pins).
FWH/LPC chips, however, operate on a full 32-bit (LPC) or 28-bit (FWH) address
space and would fail with the fallback mapping to NULL.

Corresponding to flashrom svn r1895.
Signed-off-by: Urja Rannikko <urjaman@gmail.com>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Previously we have used low-active macros (because the hardware and
old protocol were so too) and set every single LED explicitly although we
only used a limited number of combinations. Using an enumeration for
commonly used values instead makes things easier.

Based on the following chromiumos change:
Change-Id: Ie481a583e623cdc45e3649a4db69b15570f65a7b

Corresponding to flashrom svn r1894.
Signed-off-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: David Hendricks <dhendrix@chromium.org>

Saying that manufacturer id 0x00 is an "unknown SPI chip"
just confuses people with external programmers without a
proper connection to a chip and makes them think flashrom
doesn't support the chip they're trying to use.
Also causes unnecessary -c requirement with a multiple-slot
(FWH/LPC and SPI) serprog device i was testing.

Corresponding to flashrom svn r1893.
Signed-off-by: Urja Rannikko <urjaman@gmail.com>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Tested with a W29EE512P-70 (PLCC32, 5V)  found on an RTL8169 network card.
The ID for the chip was already in flashchips.h with the name W29C512A
and a note for "also W29EE512". The datasheets are almost identical.

Corresponding to flashrom svn r1892.
Signed-off-by: Urja Rannikko <urjaman@gmail.com>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Corresponding to flashrom svn r1891.
Signed-off-by: Steven Honeyman <stevenhoneyman@gmail.com>
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

DirectHW provides prototypes for out[bwl] and in[bwl], but we still
need to define the respective OUT[BWL] and IN[BWL] binding macros.

Apparently this has been broken since r1864 (or January 2015).

Corresponding to flashrom svn r1890.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Namely GD25LQ40, GD25LQ80, GD25LQ16, GD25LQ64(B), GD25LQ128.

Corresponding to flashrom svn r1889.
Signed-off-by: Roman Titov <titovroman@gmail.com>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Corresponding to flashrom svn r1887.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Tested mainboards:
OK:
 - Elitegroup GF7050VT-M
   Reported by Alex
 - Fujitsu D2724-A1x (used in ESPRIMO E5625)
   Reported by Rainer Spillmann
 - Teclast X98 Air 3G
   Reported by Antonio Ospite

Flash chips:
 - Fix MX25L6405(D) definition by splitting it.
   Reported by Reggie McMurtrey
 - Add Macronix MX25L..08E family and rearrange MX25L6436E.
 - Pm49FL004 to PREW (+EW)
   Reported by Georg Sauthoff

Miscellaneous:
 - Add board enable for abit KN9 Ultra.
 - Mark ARM-USB-OCD as working OK.
 - Use "mobile devices" instead of "laptops" in output.
 - Tiny other stuff.

Corresponding to flashrom svn r1886.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Coverity has brought up the following problems:

mstarddc_spi_send_command():
 - CID 1270702: bad comparison of malloced pointer 'cmd'.
 - CID 1270701: a NULL pointer dereference possible because of above.

Simply checking the return value of malloc in a valid way fixes both problems.

mstarddc_spi_init():
 - CID 1270699 and 1270700: Memory leak of i2c_device.

This patch revamps the function in various ways to fix these issues and some
other irritating bits.
It reduces scopes of variables where possible, pushes the code towards our
coding standards and introduces a label-based resource cleanup at the end.

Corresponding to flashrom svn r1885.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Alexandre Boeglin <alex@boeglin.org>

Corresponding to flashrom svn r1883.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>

Was implemented by SPARC newbies, does (cross-)compile but is not run-tested.

Corresponding to flashrom svn r1882.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>

This patch was inspired by the code in AVRDude (open source Atmel AVR
programmer) to support the PICkit2 written by Doug Brown [1]. The
Dediprog code in flashrom was used as the template for this code with
some reference to the ft2232 code as well.

[1] - https://github.com/steve-m/avrdude/blob/master/pickit2.c



Corresponding to flashrom svn r1881.
Signed-off-by: Justin Chevrier <jchevrier@gmail.com>
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

The way more elegant check for the header fails unfortunately on CentOS 4.9
because PCI_LIB_VERSION is not defined at all although the domain parameter
is present. This patch jumps through the hoops via an additional check in the
Makefile to determine if the function accepts 5 parameters (new version) or
not (old version).

Corresponding to flashrom svn r1880.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Tested mainboards:
OK:
 - AOpen UK79G-1394 (used in EZ18 barebones)
   Reported by Lawrence Gough
 - ASUS M4N78 SE
   Reported by Dima Veselov
 - ASUS P5LD2-VM
   Mark board enable as tested (reported by Dima Veselov)
 - GIGABYTE GA-970A-UD3P (rev. 2.0)
   Reported by trucmar on IRC
 - GIGABYTE GA-990FXA-UD3 (rev. 4.0)
   Reported by ROKO__ on IRC
 - GIGABYTE GA-H77-DS3H (rev. 1.1)
   Reported by Evgeniy Edigarev
 - GIGABYTE GA-P55-USB3 (rev. 2.0)
   Reported by Måns Thörnqvist
 - MSI MS-7817 (H81M-E33)
   Reported by Igor Kolker

Chipsets:
 - Marked Intel Bay Trail (0x0f1c) as tested OK
   Reported by Antonio Ospite
 - Refine Intel IDs
    * Add IDs for Braswell
    * Add IDs for 9 Series PCHs (e.g. H97, Z97)
    * Rename Wellsburg devices slightly

Flash chips:
 - Atmel AT25DF041A to PREW (+PREW)
   Reported by Tai-hwa Liang
 - Atmel AT26DF161 to PREW (+EW)
   Reported by Steve Shenton
 - Atmel AT45DB011D to PREW (+PREW)
   Reported by The R...

Strnlen() is in POSIX 2008 but was a GNU extension up to glibc 2.10
requiring to define _GNU_SOURCE. This fixes compilation on CentOS 4.9.
Also, move our implementation of strnlen() that was added to support
DJGPP to helpers.c.

Corresponding to flashrom svn r1878.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>

libusb-win32 is using a different header file name (lusb0_usb.h) for
a while. Use that on Windows builds to make clear that this is
currently the correct header to include.

Hopefully this will change soonish by migrating away from libusb-0.

Corresponding to flashrom svn r1877.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>

In cross-compiler setups pkg-config often needs some help to
determine the correct libraries. This can be done for example by
setting PKG_CONFIG_LIBDIR to point to the directory where the
cross-compile .pc files are located. If PKG_CONFIG_LIBDIR was not
set already outside of the Makefile we set it according to LIBS_BASE
and add the respective path to the linker's rpath. This makes it
possible to easily cross-compile by only setting CC and LIBS_BASE on
the command line.

This patch also removes the explicit default for LIBS_BASE again
because it does not play well with this change and was a bad idea to
begin with.

Corresponding to flashrom svn r1876.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>

Add a workaround for the GNU make that shipped with CentOS 4.9, which
apparently does not like semicolons in shell code (and which also
ignores info functions altogether by the way).

Corresponding to flashrom svn r1875.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Very old versions (<2.2) of pciutils had a typedef named "word" in
types.h. That does not play well with previous local variable names
of nicintel_eeprom.c.

Corresponding to flashrom svn r1874.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

In r1872 we replaced some magic values with constant macros from
ftdi.h. Among them was DIS_DIV_5 that represents the opcode that
disables the use of the 5x prescaler on newer devices. Unfortunately
this macro was only introduced with support for FT232H and hence is
not available in older versions of the library. Revert back to using
the magic constant.

Corresponding to flashrom svn r1873.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Also, improve documentation of static variables cs_bits and pindir.

Corresponding to flashrom svn r1872.
Signed-off-by: Antony Pavlov <antonynpavlov@gmail.com>
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

We refactored board_match_name()/--mainboard handling in r1577 and
apparently broke it because since then we were always comparing to
the respective coreboot values of board and vendor names instead of
the user-supplied strings.

Corresponding to flashrom svn r1871.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

These are very similar to Eon's EN25B series but unlike those the
EN25P has uniform sectors (of 32 and 64 kB). They can not be
distinguished by RDID alone but the RES and REMS IDs are different
and hence could be detected eventually in the future. This patch
also refines tiny bits of the EN25B series.

The series includes:
 - EN25P05
 - EN25P10
 - EN25P20
 - EN25P40
 - EN25P80
 - EN25P16
 - EN25P32
 - EN25P64

Corresponding to flashrom svn r1870.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

This patch adds a CONFIG_EVERYTHING flag to the Makefile
that sets all CONFIG_* flags to "yes" if they were "no".

This provides a comfortable way to ensure maximum code exposure
to tools like coverity[0] or clang's scan-build.

[0] https://scan.coverity.com/projects/1020



Corresponding to flashrom svn r1869.
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>

Corresponding to flashrom svn r1868.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

This is based on the code sent to the flashrom mailing list
by Alex Lu (alexlu6@mxic.com.tw) without sign-off.

Corresponding to flashrom svn r1867.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

 - /usr/include/cpu/param.h defines PAGE_MASK already, hence use another
   name for the respective macro in nicintel_eeprom.c.
 - Since DragonflyBSD 3.6 DPorts is used as the default package manager.
   Therefore we should use /usr/local/ instead of /usr/pkg/ on default
   to fetch libraries.

Corresponding to flashrom svn r1866.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

AR defaults to "ar" anyway in GNU make and instead of gcc it is probably a
better idea to use just cc. The latter allows to build on freebsd 10 without
overriding CC or doing symlink tricks (because it uses clang as cc, provides
multiple gcc packages but only versioned binaries without any actual "gcc"
command name in the path).

Corresponding to flashrom svn r1865.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

We do CPU architecture checks once for the makefile in arch.h and
once for HW access abstraction in hwaccess.c. This patch unifies
related files so that they can share the checks to improve
maintainability and reduce the chance of inconsistencies.
Furthermore, it refines some of the definitions, which
 - adds "support" for AARCH64 and PPC64,
 - adds big-endian handling on arm as well as LE handling on PPC64,
 - fixes compilation of internal.c on AARCH64 and PPC64.

Additionally, this patch continues to unify all OS checks in
flashrom by adding a new helper macro IS_WINDOWS.

The old header file for architecture checking is renamed to platform.h
to reflect its broader scope and all new macros are add in there.

Corresponding to flashrom svn r1864.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Gets rid of a few DOS-specific Makefile hacks.

Corresponding to flashrom svn r1863.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Up to now, when compiling flashrom outside a VCS it would print two
warnings that are not very clear to the user. This patch adds a new
auxilary function to getrevision.sh and uses it in the makefile to print a
single and more meaningful message to the user while hiding the
warnings from getrevision.sh.

Corresponding to flashrom svn r1862.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Silvermont (Bay Trail, Rangeley, Avoton) seems to still use the old
density encoding with 3 bits per chip. Documentation is unavailable
(held concealed by Intel) but thanks to the efforts of Tai-Hong
(Type) Wu the layout is clear now. This patch is based on his one
but solves the issue differently thus reducing the code complexity.

Corresponding to flashrom svn r1861.
Signed-off-by: Tai-Hong Wu <thwu@lunartoday.com>
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Basically, among other chips, MSTAR manufactures SoCs that equip TV sets
and computer screens, and it seems that all of their products use the
same in-system programming protocol. Basically, they use the DDC channel
of VGA or DVI connectors, which is actually an I2C bus, to encapsulate
SPI frames (the flash chip is connected to the SoC through an SPI bus).

I wrote this patch since the screen I bought had a software bug, and the
manufacturer only released a new firmware binary, but no tool or
instructions on flashing it.

More details can be found here:
http://boeglin.org/blog/index.php?entry=Flashing-a-BenQ-Z-series-for-free(dom)



I only read code from Linux kernel archives published by Acer to figure
out the protocol (for a touchscreen controller and an NFC chip, both by
MSTAR, that share the same ISP protocol), so I don't think there are
any legal problems with it.

Compilation is currently disabled by default in the Makefile.
If in doubt, additional Makefile bugs were added by Stefan.

Corresponding to flashrom svn r1860.
Signed-off-by: Alexandre Boeglin <alex@boeglin.org>
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Since r1833 we added the offset of the virtual register in several
functions, which produced segfaults. This patch renames a few
parameters and reorganizes/fixes various parts of the
changelock_regspace2_block() function - hence the rather big diff.

Thanks to Roman Lebedev for reporting this issue and testing numerous
revisions of this patch.

Corresponding to flashrom svn r1859.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

This flash chip can be configured (one time) to use 64 KiB or 256 KiB sectors.
Additionally, in the 64 KiB mode it supports 16 4 KiB sub-sectors that can be
(one time) programmed to be on the top or bottom of the device. The sub-sectors
can be erased with the 0x20 opcode but because this opcode does not work
with the remaining sectors and flashrom can not cope with that the 0x20
opcode is not supported yet.

This patch adds two definitions, one for the 64 KiB and 256 KiB configuration
respectively. The device also shares the RDID with the various S25FL128
devices so we have to increase the maximum number of successfully probed
chips to 8.

The 64 KiB mode was tested on real hardware.

Corresponding to flashrom svn r1858.
Signed-off-by: Jernej Škrabec <jernej.skrabec@planet.si>
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>

Both use the same ID and are mostly identical.

Corresponding to flashrom svn r1857.
Tested-by: Denis 'GNUtoo' Carikli <GNUtoo@no-log.org>
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@no-log.org>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>