Convert all physmaps in dmi.c to use aligned readonly maps.
Convert all physmaps in cbtable.c to use unaligned readonly maps.
Make physunmap() a generic architecture-independent wrapper.
Add physunmap_unaligned() to complement physmap*_unaligned().
Corresponding to flashrom svn r1765.
Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Previously the external DMI decoder did not allow this to happen because
all possible pointers were initialized at startup by the output of
'dmidecode -s ...' which has default values for all supported types.
The now active internal DMI decoder does work differently: it scans the
complete DMI table once and copies the available strings. Therefore, strings
that are not set by the firmware are left at their default value of NULL.
A segfault would arise if the following conditions are all true:
- the firmware sets up a DMI/SMBIOS table which has at least a correct
checksum, and
- that table does *not* define at least one of the DMI strings we use
for matching (as defined by dmi_strings[] in dmi.c), and
- there exists a board enable whose PCI IDs are matched by the board,
and which has a DMI string set that ends with a $ anchor, and
- the user calls the internal programmer of flashrom without the
optional mainboard parameter.
This was first observed by Gelip on an abit BF6 using the coreboot port
for the abit BE6-II V2.0.
The segfault was reproduced by Idwer Vollering on an ASUS F2A85-M with
the default DMI values of CONFIG_MAINBOARD_SMBIOS_MANUFACTURER etc.
overwritten and a forged board enable matching his board.
Idwer also verified that this patch fixes the problem, thanks a lot!
Corresponding to flashrom svn r1763.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
The only call path where exit was reached was from physmap functions.
Callers of physmap() et al. which were not prepared to handle
ERROR_PTR return values have been adjusted.
physmap_try_ro() has been renamed to physmap_ro() and physmap_common()
slightly refactored due to the now removed *FAIL parameters.
Corresponding to flashrom svn r1745.
Signed-off-by: Niklas Söderlund <niso@kth.se>
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Previously we had to rely on dmidecode to decode the DMI/SMBIOS table.
This patch integrates a DMI decoder into flashrom. The old behavior of calling
dmidecode can be brought back by using CONFIG_INTERNAL_DMI=no.
Significant portions of this patch were taken from dmidecode, hence add its
authors to the copyright notice (dmidecode is also GPL2+). We do a few things
differently though. First of all we do more bounds checking to prevent accessing
unmapped memory. We do not support disovery via EFI (yet), but memory
scanning only. We handle the chassis-type lock bit correctly which dmidecode
did not for a long while.
The API to the rest of flashrom remains stable, but the output changes slightly.
To share as much code as possible (which actually is not much), i have added
dmi_fill methods that get called in dmi_init. They are reponsible to fill the
dmi_strings array and setting the is_laptop variable. After it is called, dmi_init
prints all dmi_strings. Previously the strings were printed in the order they were
discovered, followed by the chassis-type, which is now output earlier (in dmi_fill).
Because DJGPP does not support strnlen a simple implementation was added
for it.
This is still only available on x86; actually it is not even compiled in for other
architectures at all anymore.
Corresponding to flashrom svn r1735.
Signed-off-by: Sean Nelson <audiohacked@gmail.com>
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Tested-by: Maciej Pijanka <maciej.pijanka@gmail.com>
Tested-by: Idwer Vollering <vidwer@gmail.com>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Also, unify all outputs of "Warning:" and "Error:" to use normal
capitalization instead of mixing it with all capitals.
Corresponding to flashrom svn r1643.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Idwer Vollering <vidwer@gmail.com>
In dmi_init() we populate static char *dmistrings[] with values that get
later compared in dmi_match(). Those strings are actually strduped in
get_dmi_string() and hence need to be freed later. This patch accomplishes
this by registering another shutdown method. Also, the tangling pointers are
nulled when the memories are freed.
This bug was found thanks to valgrind.
Corresponding to flashrom svn r1604.
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 just fixes a limited number of bits not conforming to c99 by using
- __asm__ instead of just asm
- {0} instead of {} for struct initialization
- h_addr_list[0] instead of h_addr to access the host address in
struct hostent
- #include <strings.h> where needed (for ffs and strcasecmp)
Based on a previous patch by Carl-Daniel.
Corresponding to flashrom svn r1585.
Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
- Mixing uninitialized and initialized local variables leads to
confusion.
- ft2232_spi error cases should have gotten some error handling, and
that's the reason the curly braces were there.
- Fixing typos/wording in some places would have been nice given that
those places were touched anyway.
Corresponding to flashrom svn r1413.
Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
Acked-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Also, indentation fixes, e.g. due to conversion to msg_*, use ARRAY_SIZE
where possible, wrap overly long line, etc.
Compile-tested. There should be no functional changes.
Corresponding to flashrom svn r1397.
Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
We had broken laptops in the past that were not detected as such because
their DMI chassis-type was either undefined/out-of-spec, or set to
'other' or 'unknown'.
this patch tries to mitigate this problem as follows:
- if the DMI chassis-type clearly identifies the system as
laptop/notebook/mobile platform then nothing changes: the user gets
the laptop warning without a hint to the force switch.
- if the DMI chassis-type is not specific enough, we warn the user
similarly, but tell them the switch.
to reduce the number of false positives i have added a few new
chassis types that we have encountered in the last months to the list.
Corresponding to flashrom svn r1390.
Signed-off-by: Stefan Tauner <stefan.tauner@alumni.tuwien.ac.at>
Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
Half a dozen hardcoded strcmp() don't make sense if we need a
chassis-type list anyway once we merge the internal DMI decoder. Provide
and array of the most interesting chassis types and annotate them with
laptop/non-laptop status. Match the dmidecode chassis type against the
strings in the array.
Corresponding to flashrom svn r1182.
Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
Acked-by: Sean Nelson <audiohacked@gmail.com>
Dmidecode emits a warning message about unsupported SMBIOS versions
to stdout before the information asked for when using "-s". I consider
this behaviour broken, but we still need to workaround it as e.g. Fedora
currently distributes an dmidecode with this behaviour.
Corresponding to flashrom svn r1136.
Signed-off-by: Michael Karcher <flashrom@mkarcher.dialup.fu-berlin.de>
Acked-by: Sean Nelson <audiohacked@gmail.com>
Programmer specific functions are of absolutely no interest to any file
except those dealing with programmer specific actions (special SPI
commands and the generic core).
The new header structure is as follows (and yes, improvements are
possible):
flashchips.h flash chip IDs
chipdrivers.h chip-specific read/write/... functions
flash.h common header for all stuff that doesn't fit elsewhere
hwaccess.h hardware access functions
programmer.h programmer specific functions
coreboot_tables.h header from coreboot, internal programmer only
spi.h SPI command definitions
Corresponding to flashrom svn r1112.
Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
Constify variables where possible.
Initialize programmer-related variables explicitly in programmer_init to
allow running programmer_init from a clean state after
programmer_shutdown.
Prohibit registering programmer shutdown functions before init or after
shutdown.
Kill some dead code.
Rename global variables with namespace-polluting names.
Use a previously unused locking helper function in sst49lfxxxc.c.
This is needed for libflashrom.
Effects on the binary size of flashrom are minimal (300 bytes
shrinkage), but the data section shrinks by 4384 bytes, and that's a
good thing if flashrom is operating in constrained envionments.
Corresponding to flashrom svn r1068.
Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
Acked-by: Michael Karcher <flashrom@mkarcher.dialup.fu-berlin.de>
Provide empty DMI stubs for that case until someone implements our own
dmidecode subset.
Tested by Patrick Georgi on top of libpayload.
Corresponding to flashrom svn r1058.
Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
Convert programmer print messages to msg_p* convert general print messages to msg_g* a few fixes as suggested by Carl-Daniel.
Corresponding to flashrom svn r997.
Signed-off-by: Sean Nelson <audiohacked@gmail.com>
Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
- Force chip read by faking probe success.
- Force chip access even if the chip is bigger than max decode size for
the flash bus.
- Force erase even if erase is known bad.
- Force write even if write is known bad.
- Force writing even if cbtable tells us that this is the wrong image
for this board.
This patch cleans up --force usage:
- Remove any suggestions to use --force for probe/read from flashrom
output.
- Don't talk about "success" or "Found chip" if the chip is forced.
- Add a new internal programmer parameter boardmismatch=force. This
overrides any mismatch detection from cbtable/image comparisons.
- Add a new internal programmer parameter laptop=force_I_want_a_brick.
- Adjust the documentation for --force.
- Clean up the man page a bit whereever it talks about --force or
laptops.
Additional changes in this patch:
- Add warnings about laptops to the documentation.
- Abort if a laptop is detected. Can be overridden with the programmer
parameter mentioned above.
- Add "Portable" to the list of DMI strings indicating laptops.
- Check if a chip specified with -c is known to flashrom.
- Programmer parameter reliability and consistency fixes.
- More paranoid self-checks.
- Improve documentation.
Corresponding to flashrom svn r996.
Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
Acked-by: Michael Karcher <flashrom@mkarcher.dialup.fu-berlin.de>
Also further removes dead code from dmi.c and adds a missing newline.
Corresponding to flashrom svn r936.
Signed-off-by: Michael Karcher <flashrom@mkarcher.dialup.fu-berlin.de>
Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
- Fix coding-style, whitespace, and indentation in a few places.
- Consistently use the same spelling ("Super I/O") everywhere.
Corresponding to flashrom svn r933.
- Make some flashrom stdout output look a bit nicer.
Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
This also replaces the meaningless numbers in the DMI debug printout
with the parameter names.
Corresponding to flashrom svn r912.
Signed-off-by: Michael Karcher <flashrom@mkarcher.dialup.fu-berlin.de>
Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
This bug slipped in on changing back match-specific to match-any
Corresponding to flashrom svn r896.
Signed-off-by: Michael Karcher <flashrom@mkarcher.dialup.fu-berlin.de>
Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
Recent gcc/glibc combinations warn about ignoring the fgets() result.
The problem exists on Ubuntu 9.10 with current updates. This "fix" of
the non-problem (as I check ferror() afterwards) should even be a
(negligible) performance optimization.
Corresponding to flashrom svn r884.
Signed-off-by: Michael Karcher <flashrom@mkarcher.dialup.fu-berlin.de>
Acked-by: Andrew Morgan <ziltro@ziltro.com>
If a board is not uniquely identifiable by PCI device/subsystem IDs, a
string can be specified to be looked for (case-sensitive, substring or
anchored) for now in one of the following DMI items in addition to matching
the PCI IDs:
- System Manufacturer
- System Product Name
- System Version
- Baseboard Manufacturer
- Baseboard Product Name
- Baseboard Version
Strings are anchored re-like (^ at the beginning, $ at the end), but
there are no plans to support full regular expressions and matched to any
of the mentioned fields.
The match is only made if DMI info is available and the string matches.
If no DMI info is available and the PCI IDs match, a warning is printed
as the board can not be autodetected.
It's still open to discussion whether we add an DMI override switch to
specify a string that will definitely match, and whether this switch is
only used if no DMI is available or whether it overrides or augments DMI
data.
DMI data is currently read using dmidecode. This tool is available for
all major platforms except MacOS X. I heard that there also is a MacOS X
version of dmidecode, but didn't investigate that.
Corresponding to flashrom svn r874.
Signed-off-by: Michael Karcher <flashrom@mkarcher.dialup.fu-berlin.de>
Acked-by: Luc Verhaegen <libv@skynet.be>
Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
