- flashrom - mail.coreboot.org

Re: [flashrom] Flashrom fails on write. P4 motherboard
by Stefan Tauner 25 Jul '13

25 Jul '13

On Fri, 19 Jul 2013 10:55:31 +0200 san <san(a)plusnet.pl> wrote: > Yes, i did try it, but can not find any useful informations. Still can not > write. Can not find any gpio ports and anything to use with pciset. Found > lpt port and files according to this page are uploaded on server. > > Please, help. According to what our community found out so far you need to lower GPIO 19. Please send/upload the output of lspci -nn too so that I can prepare a proper patch for you to test. -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

2 1

[PATCH] Add support for Loongson-3A.
by Stefan Tauner 25 Jul '13

25 Jul '13

Also be less restrictive regarding Loongson 2, because these strings seem to be rather volatile: http://www.linux-mips.org/archives/linux-mips/2013-04/msg00179.html http://www.loongson.cn/dev/gitweb/?p=linux-loongson-all;a=history;f=arch/mi… Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> --- Let's pretend I know what I am doing here... if I read the chinese data sheets correctly then Loongson-3B has moved to use SPI primarily for its firmware but there is always a chipset attached too... In any case I would rather not touch it yet. 3A OTOH seems to be more similar to Loongson-2 so this might actually work. This patch applies w/o conflicts to r1565 and later. processor_enable.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/processor_enable.c b/processor_enable.c index 1361dd5..92c6b18 100644 --- a/processor_enable.c +++ b/processor_enable.c @@ -58,8 +58,9 @@ static int is_loongson(void) while (*ptr && isspace((unsigned char)*ptr)) ptr++; fclose(cpuinfo); - return (strncmp(ptr, "ICT Loongson-2 V0.3", strlen("ICT Loongson-2 V0.3")) == 0) || - (strncmp(ptr, "Godson2 V0.3 FPU V0.1", strlen("Godson2 V0.3 FPU V0.1")) == 0); + return (strncmp(ptr, "ICT Loongson-2", strlen("ICT Loongson-2")) == 0) || + (strncmp(ptr, "Godson2 V0.3 FPU V0.1", strlen("Godson2 V0.3 FPU V0.1")) == 0) || + (strncmp(ptr, "ICT Loongson-3A", strlen("ICT Loongson-3A")) == 0); } fclose(cpuinfo); return 0; -- Kind regards, Stefan Tauner

1 0

Re: [flashrom] W836xx: Supermicro X8DTG-D: flashrom -V
by Stefan Tauner 22 Jul '13

22 Jul '13

On Mon, 22 Jul 2013 16:01:55 -0400 "Mueller, Daniel" <DANIEL.MUELLER(a)saic.com> wrote: > Today I tested flashrom on a Supermicro X8DTG-D motherboard. Below are > the commands I ran, a link to their output, and any of my notes on that > step. > > > Test environment: 64 bit CentOS 6 image booted via PXE, linux kernel > 2.6.32-358.6.1, Supermicro X8DTG-D with BIOS Rev. 2.1a > > `flashrom -p internal -V` http://paste.flashrom.org/view.php?id=1702 > > `lspci -nnvvxxx` http://paste.flashrom.org/view.php?id=1703 > > `superiotool -deV` http://paste.flashrom.org/view.php?id=1704 > > `flashrom -V -r x8dtg-bak.bin -p internal` > http://paste.flashrom.org/view.php?id=1707 > > `flashrom -V -w X8DTG.21A` http://paste.flashrom.org/view.php?id=1708 > > Note: The BIOS binary released by Supermicro, Rev. 2.1a. > > At this point, I rebooted the system to verify the write was successful > and see if any settings were changed. > > `flashrom -V -v X8DTG.21A -p internal` > http://pase.flashrom.org/view.php?id=1705 > > Note: The return code was 255 > > > > Is it normal behavior to have the verify operation succeed after a > write, but fail after a reboot? Hello Daniel, thanks for your report! I have added your board to our list of supported hardware and will commit that later together with other small changes. The behavior you have been seeing is normal. The firmware often writes some kind of log data to the flash chip on boot, hence the verification error. If you compare a file read after writing and after rebooting in a hex editor then they should be very similar. -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

1 0

W836xx: Supermicro X8DTG-D: flashrom -V
by Mueller, Daniel 22 Jul '13

22 Jul '13

Good afternoon, Today I tested flashrom on a Supermicro X8DTG-D motherboard. Below are the commands I ran, a link to their output, and any of my notes on that step. Test environment: 64 bit CentOS 6 image booted via PXE, linux kernel 2.6.32-358.6.1, Supermicro X8DTG-D with BIOS Rev. 2.1a `flashrom -p internal -V` http://paste.flashrom.org/view.php?id=1702 `lspci -nnvvxxx` http://paste.flashrom.org/view.php?id=1703 `superiotool -deV` http://paste.flashrom.org/view.php?id=1704 `flashrom -V -r x8dtg-bak.bin -p internal` http://paste.flashrom.org/view.php?id=1707 `flashrom -V -w X8DTG.21A` http://paste.flashrom.org/view.php?id=1708 Note: The BIOS binary released by Supermicro, Rev. 2.1a. At this point, I rebooted the system to verify the write was successful and see if any settings were changed. `flashrom -V -v X8DTG.21A -p internal` http://pase.flashrom.org/view.php?id=1705 Note: The return code was 255 Is it normal behavior to have the verify operation succeed after a write, but fail after a reboot? Daniel Mueller Systems Engineering co-op

1 0

Flashrom works with GA-P55A-UD7
by goodgod261 22 Jul '13

22 Jul '13

Hi, I've successfully flashed my Gigabyte GA-P55A-UD7 with flashrom (svn 1701)! Flashrom reported an error when erasing chip, but everything worked fine. Three flash chips have been detected with one physical address and the same contents. Flashing the first one changed all of them, so it's probably a bug. It might related to Gigabyte's DualBIOS. By the way, wouldn't it be relatively easy to port Coreboot to motherboards with DualBIOS? I attach output of: * flashrom -p internal -V * lspci -nnvvxxx * superiotool -deV * flashrom -p internal -c MX25L1605 -w P55AUD7.F8B -V Best regards, goodgod261

2 2

Re: [flashrom] Sapphire IPC-E350M1 mainboard read success
by Stefan Tauner 21 Jul '13

21 Jul '13

On Fri, 19 Jul 2013 19:24:36 +0400 Антон Кочков <anton.kochkov(a)gmail.com> wrote: > Just downloaded flashrom trunk and tried to read > http://paste.flashrom.org/view.php?id=1695 - as you can see it fails very interesting result (with r1701). Thanks for testing that. Can you please test r1700, and if that does not help r1697? -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

1 1

serprog with arduino UNO
by Hugo Petrucci 20 Jul '13

20 Jul '13

Hi, I followed the steps in this link http://flashrom.org/Serprog/Arduino_flasher for write the SST25LF040A flash chip run first time and asked to specify which chip was using run specifying the chip, the error not found the chip follows the commands executed hugo@hugo-linux ~/flashrom $ sudo ./flashrom -p serprog:dev=/dev/ttyACM0:2000000 flashrom v0.9.6.1-r1701 on Linux 3.5.0-17-generic (i686) flashrom is free software, get the source code at http://www.flashrom.org Calibrating delay loop... OK. serprog: Programmer name is "serprog-duino" Found SST flash chip "SST25LF040A" (512 kB, SPI) on serprog. Found SST flash chip "SST25VF040" (512 kB, SPI) on serprog. Multiple flash chips were detected: "SST25LF040A", "SST25VF040" Please specify which chip to use with the -c <chipname> option. specified the chip and verbose hugo@hugo-linux ~/flashrom $ sudo ./flashrom -c "SST25LF040A" -V -p serprog:dev=/dev/ttyACM0:2000000 flashrom v0.9.6.1-r1701 on Linux 3.5.0-17-generic (i686) flashrom is free software, get the source code at http://www.flashrom.org flashrom was built with libpci 3.1.9, GCC 4.7.2, little endian Command line (5 args): ./flashrom -c SST25LF040A -V -p serprog:dev=/dev/ttyACM0:2000000 Calibrating delay loop... OS timer resolution is 1 usecs, 1004M loops per second, 10 myus = 10 us, 100 myus = 100 us, 1000 myus = 999 us, 10000 myus = 10002 us, 4 myus = 4 us, OK. Initializing serprog programmer Baud rate is 2000000. serprog: connected - attempting to synchronize ..... serprog: Synchronized serprog: Interface version ok. serprog: Bus support: parallel=off, LPC=off, FWH=off, SPI=on Warning: Automatic command availability check failed for cmd 0x08 - won't execute cmd Warning: Automatic command availability check failed for cmd 0x11 - won't execute cmd serprog: Programmer name is "serprog-duino" serprog: Serial buffer size is 65535 serprog: Warning: Programmer does not support toggling its output drivers The following protocols are supported: SPI. Probing for SST SST25LF040A, 512 kB: probe_spi_res2: id1 0x0, id2 0x0 No EEPROM/flash device found. Note: flashrom can never write if the flash chip isn't found automatically.

3 4

[PATCH] Makefile: Print data model of C types.
by Stefan Tauner 20 Jul '13

20 Jul '13

Since we are starting to support more data models and are reworking some core data types in flashrom, it might become useful to know the size of some types, hence print them at compile and run time. Because compilers do not define suitable macros for this purpose, we implement our own in arch.h and the Makefile. Because sizeof is not evaluated by the CPP, the implementation uses macros from limits.h, stdint.h and hardcoded constants. See also http://sourceforge.net/p/predef/wiki/DataModels/ Also, simplify the __FLASHROM_ARCH__ generation and parsing, and refine sysinfo printing if uname is unavailable. New output w/o uname: flashrom v0.9.6.1-r1701 on Windows 5.1 (x86 ILP32) flashrom v0.9.6.1-r1701 on Windows 5.1 (x86_64 LLP64) flashrom v0.9.6.1-r1701 on unknown OS (arm ILP32) flashrom v0.9.6.1-r1701 on unknown OS (ppc ILP32) flashrom v0.9.6.1-r1701 on unknown OS (x86 LP64) flashrom v0.9.6.1-r1701 on unknown OS (x86 ILP32) w/ uname: flashrom v0.9.6.1-r1701 on Windows 5.1 (x86 ILP32) flashrom v0.9.6.1-r1701 on Windows 5.1 (x86_64 LLP64) flashrom v0.9.6.1-r1701 on Linux 2.6.32 (armv7l ILP32) flashrom v0.9.6.1-r1701 on Linux 2.6.32 (ppc ILP32) flashrom v0.9.6.1-r1701 on Linux 3.8.0-6-generic (x86_64 LP64) flashrom v0.9.6.1-r1701 on Linux 3.8.0-6-generic (x86_64 ILP32) Handcrafted "unknown" example: flashrom v0.9.6.1-r1701 on Linux 3.8.0-6-generic (x86_64 unknown data model) Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> --- Makefile | 9 +++++++-- arch.h | 22 +++++++++++++++++++++- flashrom.c | 19 ++++++++++--------- 3 files changed, 38 insertions(+), 12 deletions(-) diff --git a/Makefile b/Makefile index 0cd4e75..5e54c9f 100644 --- a/Makefile +++ b/Makefile @@ -267,7 +267,12 @@ endif # IMPORTANT: The following line must be placed before ARCH is ever used # (of course), but should come after any lines setting CC because the line # below uses CC itself. -override ARCH := $(strip $(shell LC_ALL=C $(CC) $(CPPFLAGS) -E arch.h 2>/dev/null | grep -v '^\#' | grep '"' | cut -f 2 -d'"')) +override ARCH := $(shell LC_ALL=C $(CC) $(CPPFLAGS) -E arch.h 2>/dev/null | grep '"__FLASHROM_ARCH__"' | cut -f 4 -d'"') +FLASHROM_CFLAGS += -D'FLASHROM_ARCH="$(ARCH)"' + +# As above but for the C data model +override DATAMODEL := $(shell LC_ALL=C $(CC) $(CPPFLAGS) -E arch.h 2>/dev/null | grep '"__FLASHROM_DATAMODEL__"' | cut -f 4 -d'"') +FLASHROM_CFLAGS += -D'FLASHROM_DATAMODEL="$(DATAMODEL)"' # PCI port I/O support is unimplemented on PPC/MIPS and unavailable on ARM. # Right now this means the drivers below only work on x86. @@ -692,7 +697,7 @@ compiler: featuresavailable @# FreeBSD wc will output extraneous whitespace. @echo $(ARCH)|wc -w|grep -q '^[[:blank:]]*1[[:blank:]]*$$' || \ ( echo "unknown. Aborting."; exit 1) - @printf "%s\n" '$(ARCH)' + @printf "%s %s\n" '$(ARCH)' '$(DATAMODEL)' @printf "Target OS is " @# FreeBSD wc will output extraneous whitespace. @echo $(TARGET_OS)|wc -w|grep -q '^[[:blank:]]*1[[:blank:]]*$$' || \ diff --git a/arch.h b/arch.h index f3f3b0d..1ed13fb 100644 --- a/arch.h +++ b/arch.h @@ -21,6 +21,25 @@ * Header file for CPU architecture checking. */ +#include <limits.h> +#include <stdint.h> +#define __16b 0xFFFFULL +#define __32b 0xFFFFFFFFULL +#define __64b 0xFFFFFFFFFFFFFFFFULL +#if (UINT_MAX == __16b && ULONG_MAX == __32b && UINTPTR_MAX == __32b) +#define __FLASHROM_DATAMODEL__ "LP32" +#elif (UINT_MAX == __32b && ULONG_MAX == __32b && UINTPTR_MAX == __32b) +#define __FLASHROM_DATAMODEL__ "ILP32" +#elif (UINT_MAX == __32b && ULONG_MAX == __32b && UINTPTR_MAX == __64b) +#define __FLASHROM_DATAMODEL__ "LLP64" +#elif (UINT_MAX == __32b && ULONG_MAX == __64b && UINTPTR_MAX == __64b) +#define __FLASHROM_DATAMODEL__ "LP64" +#elif (UINT_MAX == 8 && ULONG_MAX == __64b && UINTPTR_MAX == __64b) +#define __FLASHROM_DATAMODEL__ "ILP64" +#else +#define __FLASHROM_DATAMODEL__ "unknown data model" +#endif + #if defined (__i386__) || defined (__x86_64__) #define __FLASHROM_ARCH__ "x86" #elif defined (__mips) || defined (__mips__) || defined (_mips) || defined (mips) @@ -30,4 +49,5 @@ #elif defined(__arm__) #define __FLASHROM_ARCH__ "arm" #endif -__FLASHROM_ARCH__ +"__FLASHROM_ARCH__"=__FLASHROM_ARCH__ +"__FLASHROM_DATAMODEL__"=__FLASHROM_DATAMODEL__ diff --git a/flashrom.c b/flashrom.c index b05faf3..8a9b185 100644 --- a/flashrom.c +++ b/flashrom.c @@ -1578,40 +1578,41 @@ void list_programmers_linebreak(int startcol, int cols, int paren) void print_sysinfo(void) { +msg_ginfo(" on "); #ifdef _WIN32 SYSTEM_INFO si; OSVERSIONINFOEX osvi; memset(&si, 0, sizeof(SYSTEM_INFO)); memset(&osvi, 0, sizeof(OSVERSIONINFOEX)); - msg_ginfo(" on Windows"); + msg_ginfo("Windows "); /* Tell Windows which version of the structure we want. */ osvi.dwOSVersionInfoSize = sizeof(OSVERSIONINFOEX); if (GetVersionEx((OSVERSIONINFO*) &osvi)) - msg_ginfo(" %lu.%lu", osvi.dwMajorVersion, osvi.dwMinorVersion); + msg_ginfo("%lu.%lu", osvi.dwMajorVersion, osvi.dwMinorVersion); else - msg_ginfo(" unknown version"); + msg_ginfo("unknown version"); GetSystemInfo(&si); switch (si.wProcessorArchitecture) { case PROCESSOR_ARCHITECTURE_AMD64: - msg_ginfo(" (x86_64)"); + msg_ginfo(" (x86_64"); break; case PROCESSOR_ARCHITECTURE_INTEL: - msg_ginfo(" (x86)"); + msg_ginfo(" (x86"); break; default: - msg_ginfo(" (unknown arch)"); + msg_ginfo(" (unknown arch"); break; } #elif HAVE_UTSNAME == 1 struct utsname osinfo; uname(&osinfo); - msg_ginfo(" on %s %s (%s)", osinfo.sysname, osinfo.release, - osinfo.machine); + msg_ginfo("%s %s (%s", osinfo.sysname, osinfo.release, osinfo.machine); #else - msg_ginfo(" on unknown machine"); + msg_ginfo("unknown OS (" FLASHROM_ARCH); #endif + msg_ginfo(" " FLASHROM_DATAMODEL ")"); } void print_buildinfo(void) -- Kind regards, Stefan Tauner

1 0

Re: [flashrom] New development guidelines after 0.9.7
by Stefan Tauner 20 Jul '13

20 Jul '13

On Fri, 19 Jul 2013 16:35:14 -0700 David Hendricks <dhendrix(a)google.com> wrote: > On Fri, Jul 19, 2013 at 4:31 AM, Stefan Tauner < > stefan.tauner(a)student.tuwien.ac.at> wrote: > > > I think it is a bit sad that almost all discussion in this thread is > > focused on the VCS to use and not about the guidelines. Apparently > > emotions regarding the tools are much more intense than about the > > created work. > > > > Maybe it's a sign? > > (Hint: re-read *all* the stuff you wrote, and then reconsider what Marc > said about automation) It certainly is a sign of frustration, but if I look at coreboot reviews I would even feel vindicated that the tools are not so important and one needs to set some rules, because frankly speaking... the coreboot review process has become pretty crappy lately AFAICS (but I am not that involved, maybe it was always like that). Things like Peter's patch that got merged although he did not want that and that he reverted afterwards etc. are a clear sign that there are some missing rules IMHO :) And, even if we would switch to gerrit and git, I don't see how that would influence the guidelines I proposed very much. Yes, maybe there would be more reviewing and that would ease the problem of endless patch queues, but that would just be the same uncontrolled process - just faster. So while I recognize the wish for changing tools, the pain would not go away just by switching them and we should discuss the process nevertheless :) -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

2 1

Re: [flashrom] [PATCH 5/7] Combine serialport_write and serialport_read into serialport_rw.
by Carl-Daniel Hailfinger 20 Jul '13

20 Jul '13

Am 10.07.2013 21:21 schrieb Stefan Tauner: > Because they share almost all code, combine them into serialport_rw > and export the old functions as wrappers. This makes reads "empty reads"- > aware and programmers able to detect such problems even on "blocking" > reads. Also, this fixes error detection on Windows. > > Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> > --- > serial.c | 83 ++++++++++++++++++++++++++++++---------------------------------- > 1 file changed, 39 insertions(+), 44 deletions(-) > > diff --git a/serial.c b/serial.c > index 1b394cd..5f74fd1 100644 > --- a/serial.c > +++ b/serial.c > @@ -328,66 +330,59 @@ int serialport_shutdown(void *data) > return 0; > } > > -int serialport_write(unsigned char *buf, unsigned int writecnt) > +#define EMPTY_TRIES 10 A total of 10 ms timeout is too fast for the Bus Pirate and some slower serprog implementations AFAICS. I'd prefer a 1 s timeout before signaling an error. That should be sufficient, and callers are supposed to check for errors, so the 1 s timeout won't happen multiple times. > +/* Writes (if \c outdir is true) or reads \c todo bytes from/into \c buf. > + * Tries up to \c EMPTY_TRIES times with a timeout of \c ms ms between each try. */ > +static int serialport_rw(unsigned char *buf, unsigned int todo, bool outdir, unsigned int ms) > { > -#ifdef _WIN32 > - DWORD tmp = 0; > -#else > - ssize_t tmp = 0; > -#endif > - unsigned int empty_writes = 250; /* results in a ca. 125ms timeout */ > + const char * const op = outdir ? "write" : "read"; > + bool err = false; > + unsigned int empty_tries = EMPTY_TRIES; > > - while (writecnt > 0) { > + while (todo > 0) { > #ifdef _WIN32 > - WriteFile(sp_fd, buf, writecnt, &tmp, NULL); > + DWORD cur = 0; > + if (outdir) > + err = !WriteFile(sp_fd, buf, todo, &cur, NULL); > + else > + err = !ReadFile(sp_fd, buf, todo, &cur, NULL); Should we check for cur == 0 here? Can that happen? > #else > - tmp = write(sp_fd, buf, writecnt); > -#endif > - if (tmp == -1) { > - msg_perr("Serial port write error!\n"); > + ssize_t cur = 0; > + if (outdir) > + err = (cur = write(sp_fd, buf, todo)) < 0; OK. > + else > + err = (cur = read(sp_fd, buf, todo)) < 0; Should be <= 0 instead of < 0 AFAICS. See the POSIX man page of read(): "When attempting to read from an empty pipe or FIFO: If no process has the pipe open for writing, read() shall return 0 to indicate end-of-file." > +#endif > + if (err) { > + msg_perr("Serial port %s error!\n", op); > return 1; > } > - if (!tmp) { > - msg_pdbg2("Empty write\n"); > - empty_writes--; > - programmer_delay(500); > - if (empty_writes == 0) { > - msg_perr("Serial port is unresponsive!\n"); > + if (cur == 0) { > + msg_pdbg2("Empty %s.\n", op); > + empty_tries--; > + programmer_delay(ms * 1000); > + if (empty_tries == 0) { > + msg_perr("Serial port is unresponsive or disappeared!\n"); > return 1; > } > + continue; > } > - writecnt -= tmp; > - buf += tmp; > + todo -= cur; > + buf += cur; > + empty_tries = EMPTY_TRIES; > } > > return 0; > } > > -int serialport_read(unsigned char *buf, unsigned int readcnt) > +int serialport_write(unsigned char *buf, unsigned int writecnt) > { > -#ifdef _WIN32 > - DWORD tmp = 0; > -#else > - ssize_t tmp = 0; > -#endif > - > - while (readcnt > 0) { > -#ifdef _WIN32 > - ReadFile(sp_fd, buf, readcnt, &tmp, NULL); > -#else > - tmp = read(sp_fd, buf, readcnt); > -#endif > - if (tmp == -1) { > - msg_perr("Serial port read error!\n"); > - return 1; > - } > - if (!tmp) > - msg_pdbg2("Empty read\n"); > - readcnt -= tmp; > - buf += tmp; > - } > + return serialport_rw(buf, writecnt, true, 1); > +} > > - return 0; > +int serialport_read(unsigned char *buf, unsigned int readcnt) > +{ > + return serialport_rw(buf, readcnt, false, 10); > } > > /* Tries up to timeout ms to read readcnt characters and places them into the array starting at c. Returns Regards, Carl-Daniel -- http://www.hailfinger.org/

2 1

New development guidelines after 0.9.7
by Stefan Tauner 20 Jul '13

20 Jul '13

Our current development model with elaborate reviews has not been working for a while now. Therefore, beginning with the release of flashrom 0.9.7 our strategy on merging patches will change radically. The current rules regarding committing to subversion can be found here: http://flashrom.org/Development_Guidelines#Committing Essentially they prohibit merging non-trivial patches without explicit consent from another community member and is not too specific on the requirements for an ack. Below is my proposal for the new rules which works with implicit consent after timeouts. The exact time values are completely open for discussion of course. I would also like to introduce some kind of stable branch where we can commit important backported patches, if need be, without releasing all non-critical fixes committed in the meantime (as was the case with 0.9.6.1 for example). I would not have a problem doing this with git, but I am not so sure if I want to try that with svn... let's pretend there is a community consensus to continue with svn for now. Is that sensible to do? AFAIK this would produce odd revision numbers (which is one of the main arguments pro svn). I would also like to clean up the commit permissions and name the ones with commit rights in the wiki. IMHO the following ppl should have commit rights: carldani, stefanct, dhendrix, kmalkki, idwer I am not sure about uwe and twice11, because they are (sadly) not even active on IRC or the ml. Everyone else's permission should be revoked. = Reviews = Contributors without commit rights should receive at least a preliminary review within one week of submission by some committer. At minimum this should include a broad indication of acceptance or rejection of... * the idea/rationale/motivation, * the implementation respectively. In general, reviews should focus on the architectural changes and things that affect flashrom as a whole. This includes (but is by no means limited to) changes in APIs and types, safety, portability, extensibility, and maintainability. The purpose of reviews is not to create perfect patches, but to steer development in the right direction and produce consensus within the community. The goal of each patch should be to improve the state of the project - it does not need to fix all problems of the respective field perfectly. NB: New contributors may need more detailed advices and should be told about minor issues like formatting problems more precisely. The result of a review should either be a == Acked-by == * Trivial stuff like... ** compile fixes (that are tested on all or at least some affected and some ''unaffected'' platforms), ** marking boards/chips/programmers as tested, and ** typo and whitespace fixes etc. :do not need an Acked-by line from someone else than the author. * If the patch is something like a board enable, new chip(set) support etc. (i.e. no refactoring, no architectural impact), you can get an ack from someone who is not an experienced developer after he/she has tested the patch, or ack yourself if proof exists that it worked. * However, for more complicated patches you should get a review from someone who knows the code well if possible. ** If no one replies to the patch submission on the mailing list within 2 weeks, you may assume consent by the community and ack yourself. ** If, however, one of the known community members comments on your submission and requests changes or further information/a rationale etc., you need to come to some consensus with that member about the points complained about (only). *** If your answer (e.g. in form of a new iteration of the patch including the requested changes or a mail containing the requested information) does not get any replies for three weeks, you may ack yourself. *** When consensus in all points have been reached, but no explicit ack has been sent, then the submission falls under the same timeout rule again; i.e. if no one replies to your last e-mail within 2 weeks, you may ack yourself. * Contributions from people without commit rights should not be treated much differently. : Maintainers should merge them if there is no objection from the community within the timeframes and according to the rule set described above. = Committing = Those with commit rights to the (currently subversion-based) source repository should follow the following rules. * Reviews ** Changes should generally be discussed and reviewed before committing, see [[Development_Guidelines#Reviews|above]] * The commit log ** Should start with a short sentence summing up the changes of the patch. ** Optionally this should be prefixed with a topic or file name if the changes are targeting one area only. ** End the sentence with a full stop followed by an empty line. ** The body of the commit log should be short, but descriptive: If anyone involved in flashrom reads your comment in a year, she/he shall still be able to understand what your commit is about, without analyzing the code (what was the motivation for the change? what are the differences to the previous behavior?). ** At the end it has to list Signed-off-by lines of all authors and at least one Acked-by line. ** Optionally one can add Tested-by lines to give credit to testers. * Committing ** If you ack something and the sender has commit rights, let him commit unless he asks you to commit. *: If the sender does not have commit rights, wait at least 24 hours after sending the ack to give him/her an opportunity to add final refinements or comments. ** After committing please reply to the mailing thread that contains the patch with at least the revision in the body to notify anyone involved of the commit and document it for references in the future and in patchwork or similar facilities. * Don't share your login(s)/password(s) (should be obvious, but sadly it is not for everyone :) -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

4 5

Re: [flashrom] Sapphire IPC-E350M1 mainboard read success
by Антон Кочков 19 Jul '13

19 Jul '13

Just downloaded flashrom trunk and tried to read http://paste.flashrom.org/view.php?id=1695 - as you can see it fails Best regards, Anton Kochkov. On Tue, Apr 16, 2013 at 4:20 PM, Антон Кочков <anton.kochkov(a)gmail.com> wrote: > Reading flash http://paste.flashrom.org/view.php?id=1607 > Writing flash http://paste.flashrom.org/view.php?id=1608 > Best regards, > Anton Kochkov. > > > On Tue, Apr 16, 2013 at 4:06 PM, Антон Кочков <anton.kochkov(a)gmail.com> wrote: >> Here is full read log on trunk flashrom >> https://mega.co.nz/#!ndQhQLxL!NgMxOTwHi_oCGUD0663wjXKIcQoGWO1VU_XnXM_sY50 >> >> Sorry for using fileshare service, but full log is about 75Mb. >> >> Best regards, >> Anton Kochkov.

1 0

Re: [flashrom] New development guidelines after 0.9.7
by Carl-Daniel Hailfinger 19 Jul '13

19 Jul '13

Am 18.07.2013 19:15 schrieb Marc Jones: > On Wed, Jul 17, 2013 at 7:21 PM, Stefan Tauner < > stefan.tauner(a)student.tuwien.ac.at> wrote: > >> Our current development model with elaborate reviews has not been >> working for a while now. Therefore, beginning with the release of >> flashrom 0.9.7 our strategy on merging patches will change radically. >> >> The current rules regarding committing to subversion can be found here: >> http://flashrom.org/Development_Guidelines#Committing >> Essentially they prohibit merging non-trivial patches without explicit >> consent from another community member and is not too specific on the >> requirements for an ack. >> > Not to add undue overhead to the project, but why not go to git and gerrit? > Automates a lot of this stuff and is already setup on coreboot.org. <rant> Because I think git sucks? I might be open for a switch to mercurial because it has usable revision numbers and a user-friendly interface. But git... without a command line and the right tree, I'll never know if commit hash deadbeef is before or after badc0ffe. Besides that, I managed to learn mercurial in a few hours and gave up on git after a few days. And gerrit needs known insecure OpenID for authentication. I'm not going to get any OpenID capable account just to satisfy a defective login mechanism. Apart from that, it seems gerrit pretty much killed all patch discussion on the coreboot mailing list. It also was pretty much unsuitable for offline use (no idea if that changed). While those properties may work for coreboot development (with very infrequent contributions/reviews by outsiders), I fear it will introduce an additional barrier for flashrom. </rant> Regards, Carl-Daniel

3 2

Re: [flashrom] Flashrom fails on write. P4 motherboard
by Stefan Tauner 19 Jul '13

19 Jul '13

On Thu, 18 Jul 2013 11:31:12 +0200 san <san(a)plusnet.pl> wrote: > Ok, i found my compiled version does not work. Will check this out later. > > Flashrom.log and other possibly useful informations can be found here > http://e-san.info/Flashrom-P4/ I am quite sure that you will need a board enable that changes the state of the WP and TBL pins. You can verify that by measuring the voltage of said pins against ground. You can read more about board enables here: http://flashrom.org/Board_Enable -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

2 1

Re: [flashrom] [PATCH 04/10] sbxxx: Handle active IMCs in AMD chipsets.
by Carl-Daniel Hailfinger 18 Jul '13

18 Jul '13

Am 10.07.2013 21:17 schrieb Stefan Tauner: > Detect and temporarily disable the IMC while accessing the flash. > Disable writes on default, but allow the user to enforce it. > > Signed-off-by: Rudolf Marek <r.marek(a)assembler.cz> > Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> > --- > imc.c | 171 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ I already mentioned the suboptimal file name and public function names on IRC. You said: "[...] best would be to name it amd_imc.c and name the one external visible function amd_imc_shutdown [...]" That change would be appreciated. > diff --git a/chipset_enable.c b/chipset_enable.c > index 3979347..c884c80 100644 > --- a/chipset_enable.c > +++ b/chipset_enable.c > @@ -922,6 +922,10 @@ static int enable_flash_sb600(struct pci_dev *dev, const char *name) > uint8_t reg; > int ret; > > + if (imc_shutdown(dev) != 0) { Hm. Can we avoid calling this on SB600 (if SB600 doesn't support IMC at all)? > + return ERROR_FATAL; > + } > + > /* Clear ROM protect 0-3. */ > for (reg = 0x50; reg < 0x60; reg += 4) { > prot = pci_read_long(dev, reg); > diff --git a/sb600spi.c b/sb600spi.c > index a5c00d8..7043aec 100644 > --- a/sb600spi.c > +++ b/sb600spi.c > @@ -210,10 +211,26 @@ int sb600_probe_spi(struct pci_dev *dev) > struct pci_dev *smbus_dev; > uint32_t tmp; > uint8_t reg; > + bool amd_imc_force = false; Ah yes, the joy of having a bool type in C99. It makes sense here. >From a quick glance the code looks sane otherwise. I'll try to evaluate it in light of the recent discussions about AMD IMC and associated problems. Regards, Carl-Daniel -- http://www.hailfinger.org/

2 2

Re: [flashrom] Flashrom fails on write. P4 motherboard
by san 18 Jul '13

18 Jul '13

Ok, i found my compiled version does not work. Will check this out later. Flashrom.log and other possibly useful informations can be found here http://e-san.info/Flashrom-P4/ Thanks for quick response. > Unicorn wishes! ;) Cherry lollipops! 2013/7/18 san <san(a)plusnet.pl> > Ok, i found my compiled version does not work. Will check this out later. > > Flashrom.log and other possibly useful informations can be found here > http://e-san.info/Flashrom-P4/ > > Thanks for quick response. > > > Unicorn wishes! ;) > Cherry lollipops! > > > 2013/7/18 Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> > >> On Thu, 18 Jul 2013 02:14:22 +0200 >> san <san(a)plusnet.pl> wrote: >> >> > Hi! >> > >> > First of all, thank You for providing such a great software. >> > >> > I have problems with flashing bios on my no-name motherboard. >> > >> > It is propably Saphire or PCPartner (a bit different: >> > http://www.compass.ua/products/view1.asp?code=28915 ) >> > But almost sure it is PP: >> > http://support.pcpartner.com/products/old-intel/DM-262E.htm >> > >> > Flashrom does recognise chipset, flash chip and read rom but does not >> write. >> > >> > I use TinyCore Linux LiveCD (http://distro.ibiblio.org/tinycorelinux/), >> > Flashrom from ubuntu (binary) and current version (compiled). >> > >> > I hot swap chips but they are same model. >> > >> > Manual for motherboard found here: >> > http://support.pcpartner.com/support/man/intel/p262.pdf ( >> > http://support.pcpartner.com/support/bios/i845.htm#845e => 845EAS4-262E >> > [DDR]) doeas not provide any information about bios flash protection. >> > >> > Please help! >> >> Please provide flashrom's verbose output when it is trying to write. >> With newer versions you can create a log file easily by using the -o >> parameter. Without the detailed output we can only guess what happens. >> >> > Beast regards! >> >> Unicorn wishes! ;) >> >> -- >> Kind regards/Mit freundlichen Grüßen, Stefan Tauner >> > > > > -- > [ e-San.info | San(a)plusnet.pl ] > -- [ e-San.info | San(a)plusnet.pl ]

1 0

Re: [flashrom] New development guidelines after 0.9.7
by David Hendricks 18 Jul '13

18 Jul '13

On Thu, Jul 18, 2013 at 10:15 AM, Marc Jones <marcj303(a)gmail.com> wrote: > On Wed, Jul 17, 2013 at 7:21 PM, Stefan Tauner < > stefan.tauner(a)student.tuwien.ac.at> wrote: > >> Our current development model with elaborate reviews has not been >> working for a while now. Therefore, beginning with the release of >> flashrom 0.9.7 our strategy on merging patches will change radically. >> >> The current rules regarding committing to subversion can be found here: >> http://flashrom.org/Development_Guidelines#Committing >> Essentially they prohibit merging non-trivial patches without explicit >> consent from another community member and is not too specific on the >> requirements for an ack. >> > > Not to add undue overhead to the project, but why not go to git and > gerrit? Automates a lot of this stuff and is already setup on coreboot.org > . > +1. -- David Hendricks (dhendrix) Systems Software Engineer, Google Inc.

1 0

Re: [flashrom] Flashrom fails on write. P4 motherboard
by Stefan Tauner 18 Jul '13

18 Jul '13

On Thu, 18 Jul 2013 02:14:22 +0200 san <san(a)plusnet.pl> wrote: > Hi! > > First of all, thank You for providing such a great software. > > I have problems with flashing bios on my no-name motherboard. > > It is propably Saphire or PCPartner (a bit different: > http://www.compass.ua/products/view1.asp?code=28915 ) > But almost sure it is PP: > http://support.pcpartner.com/products/old-intel/DM-262E.htm > > Flashrom does recognise chipset, flash chip and read rom but does not write. > > I use TinyCore Linux LiveCD (http://distro.ibiblio.org/tinycorelinux/), > Flashrom from ubuntu (binary) and current version (compiled). > > I hot swap chips but they are same model. > > Manual for motherboard found here: > http://support.pcpartner.com/support/man/intel/p262.pdf ( > http://support.pcpartner.com/support/bios/i845.htm#845e => 845EAS4-262E > [DDR]) doeas not provide any information about bios flash protection. > > Please help! Please provide flashrom's verbose output when it is trying to write. With newer versions you can create a log file easily by using the -o parameter. Without the detailed output we can only guess what happens. > Beast regards! Unicorn wishes! ;) -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

2 1

Flashrom fails on write. P4 motherboard
by san 18 Jul '13

18 Jul '13

Hi! First of all, thank You for providing such a great software. I have problems with flashing bios on my no-name motherboard. It is propably Saphire or PCPartner (a bit different: http://www.compass.ua/products/view1.asp?code=28915 ) But almost sure it is PP: http://support.pcpartner.com/products/old-intel/DM-262E.htm Flashrom does recognise chipset, flash chip and read rom but does not write. I use TinyCore Linux LiveCD (http://distro.ibiblio.org/tinycorelinux/), Flashrom from ubuntu (binary) and current version (compiled). I hot swap chips but they are same model. Manual for motherboard found here: http://support.pcpartner.com/support/man/intel/p262.pdf ( http://support.pcpartner.com/support/bios/i845.htm#845e => 845EAS4-262E [DDR]) doeas not provide any information about bios flash protection. Please help! Beast regards! -- [ e-San.info | San(a)plusnet.pl ]

1 0

Re: [flashrom] [PATCH] uintptr_t-ify map_flash_region functions.
by Carl-Daniel Hailfinger 18 Jul '13

18 Jul '13

Am 17.07.2013 15:58 schrieb Stefan Tauner: > unsigned long is not the right type for manipulating pointer values. > Since C99 there are suitable unsigned and signed types available, namely > uintptr_t and intptr_t respectively. > > Use them in functions assigned to programmers' map_flash_region fields and > their callers where applicable. > > This patch also changes the display width of all associated address values in > physmap.c to 16/8 hex characters depending on the actual size by > introducing a macro PRIxPTR_WIDTH and exploiting printf's * field width > specifier. > > Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> > Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> > --- > > Can you please verify that this is what you want? Yes. Exactly. > I put the _WIDTH macro in flash.h now for obvious reasons. Thanks! Regards, Carl-Daniel -- http://www.hailfinger.org/

2 1

[commit] r1701 - trunk
by repository service 18 Jul '13

18 Jul '13

Author: stefanct Date: Thu Jul 18 01:46:44 2013 New Revision: 1701 URL: http://flashrom.org/trac/flashrom/changeset/1701 Log: uintptr_t-ify map_flash_region functions. unsigned long is not the right type for manipulating pointer values. Since C99 there are suitable unsigned and signed types available, namely uintptr_t and intptr_t respectively. Use them in functions assigned to programmers' map_flash_region fields and their callers where applicable. This patch also changes the display width of all associated address values in physmap.c to 16/8 hex characters depending on the actual size by introducing a macro PRIxPTR_WIDTH and exploiting printf's * field width specifier. Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Modified: trunk/dummyflasher.c trunk/flash.h trunk/flashrom.c trunk/physmap.c trunk/programmer.c trunk/programmer.h Modified: trunk/dummyflasher.c ============================================================================== --- trunk/dummyflasher.c Wed Jul 17 11:28:00 2013 (r1700) +++ trunk/dummyflasher.c Thu Jul 18 01:46:44 2013 (r1701) @@ -412,10 +412,10 @@ return 0; } -void *dummy_map(const char *descr, unsigned long phys_addr, size_t len) +void *dummy_map(const char *descr, uintptr_t phys_addr, size_t len) { - msg_pspew("%s: Mapping %s, 0x%lx bytes at 0x%08lx\n", - __func__, descr, (unsigned long)len, phys_addr); + msg_pspew("%s: Mapping %s, 0x%lx bytes at 0x%*" PRIxPTR "\n", + __func__, descr, (unsigned long)len, PRIxPTR_WIDTH, phys_addr); return (void *)phys_addr; } Modified: trunk/flash.h ============================================================================== --- trunk/flash.h Wed Jul 17 11:28:00 2013 (r1700) +++ trunk/flash.h Thu Jul 18 01:46:44 2013 (r1701) @@ -43,10 +43,10 @@ /* TODO: check using code for correct usage of types */ typedef uintptr_t chipaddr; +#define PRIxPTR_WIDTH ((int)(sizeof(uintptr_t)*2)) int register_shutdown(int (*function) (void *data), void *data); -void *programmer_map_flash_region(const char *descr, unsigned long phys_addr, - size_t len); +void *programmer_map_flash_region(const char *descr, uintptr_t phys_addr, size_t len); void programmer_unmap_flash_region(void *virt_addr, size_t len); void programmer_delay(int usecs); Modified: trunk/flashrom.c ============================================================================== --- trunk/flashrom.c Wed Jul 17 11:28:00 2013 (r1700) +++ trunk/flashrom.c Thu Jul 18 01:46:44 2013 (r1701) @@ -417,8 +417,7 @@ return ret; } -void *programmer_map_flash_region(const char *descr, unsigned long phys_addr, - size_t len) +void *programmer_map_flash_region(const char *descr, uintptr_t phys_addr, size_t len) { return programmer_table[programmer].map_flash_region(descr, phys_addr, len); Modified: trunk/physmap.c ============================================================================== --- trunk/physmap.c Wed Jul 17 11:28:00 2013 (r1700) +++ trunk/physmap.c Thu Jul 18 01:46:44 2013 (r1701) @@ -42,7 +42,7 @@ static void *realmem_map; -static void *map_first_meg(unsigned long phys_addr, size_t len) +static void *map_first_meg(uintptr_t phys_addr, size_t len) { if (realmem_map) return realmem_map + phys_addr; @@ -61,7 +61,7 @@ return realmem_map + phys_addr; } -static void *sys_physmap(unsigned long phys_addr, size_t len) +static void *sys_physmap(uintptr_t phys_addr, size_t len) { int ret; __dpmi_meminfo mi; @@ -109,7 +109,7 @@ #define MEM_DEV "" -void *sys_physmap(unsigned long phys_addr, size_t len) +void *sys_physmap(uintptr_t phys_addr, size_t len) { return (void *)phys_to_virt(phys_addr); } @@ -124,7 +124,7 @@ #define MEM_DEV "DirectHW" -static void *sys_physmap(unsigned long phys_addr, size_t len) +static void *sys_physmap(uintptr_t phys_addr, size_t len) { /* The short form of ?: is a GNU extension. * FIXME: map_physical returns NULL both for errors and for success @@ -156,7 +156,7 @@ static int fd_mem_cached = -1; /* For MMIO access. Must be uncached, doesn't make sense to restrict to ro. */ -static void *sys_physmap_rw_uncached(unsigned long phys_addr, size_t len) +static void *sys_physmap_rw_uncached(uintptr_t phys_addr, size_t len) { void *virt_addr; @@ -176,7 +176,7 @@ /* For reading DMI/coreboot/whatever tables. We should never write, and we * do not care about caching. */ -static void *sys_physmap_ro_cached(unsigned long phys_addr, size_t len) +static void *sys_physmap_ro_cached(uintptr_t phys_addr, size_t len) { void *virt_addr; @@ -209,25 +209,24 @@ #define PHYSMAP_RW 0 #define PHYSMAP_RO 1 -static void *physmap_common(const char *descr, unsigned long phys_addr, +static void *physmap_common(const char *descr, uintptr_t phys_addr, size_t len, int mayfail, int readonly) { void *virt_addr; if (len == 0) { - msg_pspew("Not mapping %s, zero size at 0x%08lx.\n", - descr, phys_addr); + msg_pspew("Not mapping %s, zero size at 0x%0*" PRIxPTR ".\n", descr, PRIxPTR_WIDTH, phys_addr); return ERROR_PTR; } if ((getpagesize() - 1) & len) { - msg_perr("Mapping %s at 0x%08lx, unaligned size 0x%lx.\n", - descr, phys_addr, (unsigned long)len); + msg_perr("Mapping %s at 0x%0*" PRIxPTR ", unaligned size 0x%lx.\n", + descr, PRIxPTR_WIDTH, phys_addr, (unsigned long)len); } if ((getpagesize() - 1) & phys_addr) { - msg_perr("Mapping %s, 0x%lx bytes at unaligned 0x%08lx.\n", - descr, (unsigned long)len, phys_addr); + msg_perr("Mapping %s, 0x%lx bytes at unaligned 0x%0*" PRIxPTR ".\n", + descr, (unsigned long)len, PRIxPTR_WIDTH, phys_addr); } if (readonly) @@ -238,8 +237,8 @@ if (ERROR_PTR == virt_addr) { if (NULL == descr) descr = "memory"; - msg_perr("Error accessing %s, 0x%lx bytes at 0x%08lx\n", descr, - (unsigned long)len, phys_addr); + msg_perr("Error accessing %s, 0x%lx bytes at 0x%0*" PRIxPTR "\n", + descr, (unsigned long)len, PRIxPTR_WIDTH, phys_addr); msg_perr(MEM_DEV " mmap failed: %s\n", strerror(errno)); #ifdef __linux__ if (EINVAL == errno) { @@ -261,13 +260,13 @@ return virt_addr; } -void *physmap(const char *descr, unsigned long phys_addr, size_t len) +void *physmap(const char *descr, uintptr_t phys_addr, size_t len) { return physmap_common(descr, phys_addr, len, PHYSMAP_NOFAIL, PHYSMAP_RW); } -void *physmap_try_ro(const char *descr, unsigned long phys_addr, size_t len) +void *physmap_try_ro(const char *descr, uintptr_t phys_addr, size_t len) { return physmap_common(descr, phys_addr, len, PHYSMAP_MAYFAIL, PHYSMAP_RO); Modified: trunk/programmer.c ============================================================================== --- trunk/programmer.c Wed Jul 17 11:28:00 2013 (r1700) +++ trunk/programmer.c Thu Jul 18 01:46:44 2013 (r1701) @@ -28,7 +28,7 @@ } /* Fallback map() for programmers which don't need special handling */ -void *fallback_map(const char *descr, unsigned long phys_addr, size_t len) +void *fallback_map(const char *descr, uintptr_t phys_addr, size_t len) { /* FIXME: Should return phys_addr. */ return NULL; Modified: trunk/programmer.h ============================================================================== --- trunk/programmer.h Wed Jul 17 11:28:00 2013 (r1700) +++ trunk/programmer.h Thu Jul 18 01:46:44 2013 (r1701) @@ -117,7 +117,7 @@ int (*init) (void); - void *(*map_flash_region) (const char *descr, unsigned long phys_addr, size_t len); + void *(*map_flash_region) (const char *descr, uintptr_t phys_addr, size_t len); void (*unmap_flash_region) (void *virt_addr, size_t len); void (*delay) (int usecs); @@ -275,8 +275,8 @@ #endif /* physmap.c */ -void *physmap(const char *descr, unsigned long phys_addr, size_t len); -void *physmap_try_ro(const char *descr, unsigned long phys_addr, size_t len); +void *physmap(const char *descr, uintptr_t phys_addr, size_t len); +void *physmap_try_ro(const char *descr, uintptr_t phys_addr, size_t len); void physunmap(void *virt_addr, size_t len); #if CONFIG_INTERNAL == 1 int setup_cpu_msr(int cpu); @@ -359,7 +359,7 @@ /* dummyflasher.c */ #if CONFIG_DUMMY == 1 int dummy_init(void); -void *dummy_map(const char *descr, unsigned long phys_addr, size_t len); +void *dummy_map(const char *descr, uintptr_t phys_addr, size_t len); void dummy_unmap(void *virt_addr, size_t len); #endif @@ -606,7 +606,7 @@ /* programmer.c */ int noop_shutdown(void); -void *fallback_map(const char *descr, unsigned long phys_addr, size_t len); +void *fallback_map(const char *descr, uintptr_t phys_addr, size_t len); void fallback_unmap(void *virt_addr, size_t len); void noop_chip_writeb(const struct flashctx *flash, uint8_t val, chipaddr addr); void fallback_chip_writew(const struct flashctx *flash, uint16_t val, chipaddr addr);

1 0

Re: [flashrom] [PATCH 06/10] uintptr_t-ify physmap functions.
by Carl-Daniel Hailfinger 17 Jul '13

17 Jul '13

Am 10.07.2013 21:17 schrieb Stefan Tauner: > unsigned long is not the right type for manipulating pointer values. > Since C99 there are suitable unsigned and signed types available, namely > uintptr_t and intptr_t respectively. > > Use them in physmap.c and its callers where applicable. > > This patch also changes the display width of all address values in > physmap.c to 16/8 hex characters depending on the actual size. > > Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> Why don't you use PRIxPTR_WIDTH everywhere including dummyflasher.c? Put another way, is there a reason to use PRIxPTR_WIDTH at all? Do we care that C99 doesn't guarantee a 0x prefix for printed addresses with PRIxPTR? Will that cause confusion? I think PRIxPTR is a good idea, but I'm not so sure about the side effects. There's also the problem of programmer_map_flash_region. It should not take an uintptr_t. The address parameter is a horrible mess: A physical address from the CPU point of view for the internal programmer, a top-of-4-GB aligned fake address for some other programmers, and a bottom-aligned address for the rest of our programmers. If we want to do this the right way, we need to have a wrapper for physmap in the "internal" programmer case instead of setting map_flash_region = physmap. And then I need to update my patch which abstracts the flash bus address vs programmer address conflict into something that won't cause madness. > --- > dummyflasher.c | 4 ++-- > flash.h | 3 +-- > flashrom.c | 3 +-- > physmap.c | 36 +++++++++++++++++++----------------- > programmer.c | 2 +- > programmer.h | 10 +++++----- > 6 files changed, 29 insertions(+), 29 deletions(-) > > diff --git a/dummyflasher.c b/dummyflasher.c > index e2ec036..8e861ee 100644 > --- a/dummyflasher.c > +++ b/dummyflasher.c > @@ -412,9 +412,9 @@ dummy_init_out: > return 0; > } > > -void *dummy_map(const char *descr, unsigned long phys_addr, size_t len) > +void *dummy_map(const char *descr, uintptr_t phys_addr, size_t len) > { > - msg_pspew("%s: Mapping %s, 0x%lx bytes at 0x%08lx\n", > + msg_pspew("%s: Mapping %s, 0x%lx bytes at 0x%08" PRIxPTR "\n", > __func__, descr, (unsigned long)len, phys_addr); > return (void *)phys_addr; > } > diff --git a/flash.h b/flash.h > index 9e787c0..a752d6b 100644 > --- a/flash.h > +++ b/flash.h > @@ -45,8 +45,7 @@ > typedef uintptr_t chipaddr; > > int register_shutdown(int (*function) (void *data), void *data); > -void *programmer_map_flash_region(const char *descr, unsigned long phys_addr, > - size_t len); > +void *programmer_map_flash_region(const char *descr, uintptr_t phys_addr, size_t len); > void programmer_unmap_flash_region(void *virt_addr, size_t len); > void programmer_delay(int usecs); > > diff --git a/flashrom.c b/flashrom.c > index c298748..43b7b11 100644 > --- a/flashrom.c > +++ b/flashrom.c > @@ -417,8 +417,7 @@ int programmer_shutdown(void) > return ret; > } > > -void *programmer_map_flash_region(const char *descr, unsigned long phys_addr, > - size_t len) > +void *programmer_map_flash_region(const char *descr, uintptr_t phys_addr, size_t len) > { > return programmer_table[programmer].map_flash_region(descr, > phys_addr, len); > diff --git a/physmap.c b/physmap.c > index 644ee35..8be4f06 100644 > --- a/physmap.c > +++ b/physmap.c > @@ -20,6 +20,7 @@ > * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA > */ > > +#include <inttypes.h> > #include <unistd.h> > #include <stdio.h> > #include <stdlib.h> > @@ -34,6 +35,8 @@ > #include <fcntl.h> > #endif > > +#define PRIxPTR_WIDTH ((int)(sizeof(uintptr_t)*2)) > + > #ifdef __DJGPP__ > #include <dpmi.h> > #include <sys/nearptr.h> > @@ -42,7 +45,7 @@ > > static void *realmem_map; > > -static void *map_first_meg(unsigned long phys_addr, size_t len) > +static void *map_first_meg(uintptr_t phys_addr, size_t len) > { > if (realmem_map) > return realmem_map + phys_addr; > @@ -61,7 +64,7 @@ static void *map_first_meg(unsigned long phys_addr, size_t len) > return realmem_map + phys_addr; > } > > -static void *sys_physmap(unsigned long phys_addr, size_t len) > +static void *sys_physmap(uintptr_t phys_addr, size_t len) > { > int ret; > __dpmi_meminfo mi; > @@ -109,7 +112,7 @@ void physunmap(void *virt_addr, size_t len) > > #define MEM_DEV "" > > -void *sys_physmap(unsigned long phys_addr, size_t len) > +void *sys_physmap(uintptr_t phys_addr, size_t len) > { > return (void *)phys_to_virt(phys_addr); > } > @@ -124,7 +127,7 @@ void physunmap(void *virt_addr, size_t len) > > #define MEM_DEV "DirectHW" > > -static void *sys_physmap(unsigned long phys_addr, size_t len) > +static void *sys_physmap(uintptr_t phys_addr, size_t len) > { > /* The short form of ?: is a GNU extension. > * FIXME: map_physical returns NULL both for errors and for success > @@ -156,7 +159,7 @@ static int fd_mem = -1; > static int fd_mem_cached = -1; > > /* For MMIO access. Must be uncached, doesn't make sense to restrict to ro. */ > -static void *sys_physmap_rw_uncached(unsigned long phys_addr, size_t len) > +static void *sys_physmap_rw_uncached(uintptr_t phys_addr, size_t len) > { > void *virt_addr; > > @@ -176,7 +179,7 @@ static void *sys_physmap_rw_uncached(unsigned long phys_addr, size_t len) > /* For reading DMI/coreboot/whatever tables. We should never write, and we > * do not care about caching. > */ > -static void *sys_physmap_ro_cached(unsigned long phys_addr, size_t len) > +static void *sys_physmap_ro_cached(uintptr_t phys_addr, size_t len) > { > void *virt_addr; > > @@ -209,25 +212,24 @@ void physunmap(void *virt_addr, size_t len) > #define PHYSMAP_RW 0 > #define PHYSMAP_RO 1 > > -static void *physmap_common(const char *descr, unsigned long phys_addr, > +static void *physmap_common(const char *descr, uintptr_t phys_addr, > size_t len, int mayfail, int readonly) > { > void *virt_addr; > > if (len == 0) { > - msg_pspew("Not mapping %s, zero size at 0x%08lx.\n", > - descr, phys_addr); > + msg_pspew("Not mapping %s, zero size at 0x%0*" PRIxPTR ".\n", descr, PRIxPTR_WIDTH, phys_addr); > return ERROR_PTR; > } > > if ((getpagesize() - 1) & len) { > - msg_perr("Mapping %s at 0x%08lx, unaligned size 0x%lx.\n", > - descr, phys_addr, (unsigned long)len); > + msg_perr("Mapping %s at 0x%0*" PRIxPTR ", unaligned size 0x%lx.\n", > + descr, PRIxPTR_WIDTH, phys_addr, (unsigned long)len); > } > > if ((getpagesize() - 1) & phys_addr) { > - msg_perr("Mapping %s, 0x%lx bytes at unaligned 0x%08lx.\n", > - descr, (unsigned long)len, phys_addr); > + msg_perr("Mapping %s, 0x%lx bytes at unaligned 0x%0*" PRIxPTR ".\n", > + descr, (unsigned long)len, PRIxPTR_WIDTH, phys_addr); > } > > if (readonly) > @@ -238,8 +240,8 @@ static void *physmap_common(const char *descr, unsigned long phys_addr, > if (ERROR_PTR == virt_addr) { > if (NULL == descr) > descr = "memory"; > - msg_perr("Error accessing %s, 0x%lx bytes at 0x%08lx\n", descr, > - (unsigned long)len, phys_addr); > + msg_perr("Error accessing %s, 0x%0zx bytes at 0x%0*" PRIxPTR "\n", > + descr, len, PRIxPTR_WIDTH, phys_addr); > msg_perr(MEM_DEV " mmap failed: %s\n", strerror(errno)); > #ifdef __linux__ > if (EINVAL == errno) { > @@ -261,13 +263,13 @@ static void *physmap_common(const char *descr, unsigned long phys_addr, > return virt_addr; > } > > -void *physmap(const char *descr, unsigned long phys_addr, size_t len) > +void *physmap(const char *descr, uintptr_t phys_addr, size_t len) > { > return physmap_common(descr, phys_addr, len, PHYSMAP_NOFAIL, > PHYSMAP_RW); > } > > -void *physmap_try_ro(const char *descr, unsigned long phys_addr, size_t len) > +void *physmap_try_ro(const char *descr, uintptr_t phys_addr, size_t len) > { > return physmap_common(descr, phys_addr, len, PHYSMAP_MAYFAIL, > PHYSMAP_RO); > diff --git a/programmer.c b/programmer.c > index 3b4def0..bf7dca1 100644 > --- a/programmer.c > +++ b/programmer.c > @@ -28,7 +28,7 @@ int noop_shutdown(void) > } > > /* Fallback map() for programmers which don't need special handling */ > -void *fallback_map(const char *descr, unsigned long phys_addr, size_t len) > +void *fallback_map(const char *descr, uintptr_t phys_addr, size_t len) > { > /* FIXME: Should return phys_addr. */ > return NULL; > diff --git a/programmer.h b/programmer.h > index 7de8f2c..4ed7c90 100644 > --- a/programmer.h > +++ b/programmer.h > @@ -117,7 +117,7 @@ struct programmer_entry { > > int (*init) (void); > > - void *(*map_flash_region) (const char *descr, unsigned long phys_addr, size_t len); > + void *(*map_flash_region) (const char *descr, uintptr_t phys_addr, size_t len); > void (*unmap_flash_region) (void *virt_addr, size_t len); > > void (*delay) (int usecs); > @@ -275,8 +275,8 @@ int processor_flash_enable(void); > #endif > > /* physmap.c */ > -void *physmap(const char *descr, unsigned long phys_addr, size_t len); > -void *physmap_try_ro(const char *descr, unsigned long phys_addr, size_t len); > +void *physmap(const char *descr, uintptr_t phys_addr, size_t len); > +void *physmap_try_ro(const char *descr, uintptr_t phys_addr, size_t len); > void physunmap(void *virt_addr, size_t len); > #if CONFIG_INTERNAL == 1 > int setup_cpu_msr(int cpu); > @@ -359,7 +359,7 @@ void rmmio_vall(void *addr); > /* dummyflasher.c */ > #if CONFIG_DUMMY == 1 > int dummy_init(void); > -void *dummy_map(const char *descr, unsigned long phys_addr, size_t len); > +void *dummy_map(const char *descr, uintptr_t phys_addr, size_t len); > void dummy_unmap(void *virt_addr, size_t len); > #endif > > @@ -609,7 +609,7 @@ int register_opaque_programmer(const struct opaque_programmer *pgm); > > /* programmer.c */ > int noop_shutdown(void); > -void *fallback_map(const char *descr, unsigned long phys_addr, size_t len); > +void *fallback_map(const char *descr, uintptr_t phys_addr, size_t len); > void fallback_unmap(void *virt_addr, size_t len); > void noop_chip_writeb(const struct flashctx *flash, uint8_t val, chipaddr addr); > void fallback_chip_writew(const struct flashctx *flash, uint16_t val, chipaddr addr); Overall, while I'd like to have this patch in, it conceptually breaks as much stuff as it fixes. I don't see this getting fixed within the release timeframe, so I'd like to postpone. Regards, Carl-Daniel -- http://www.hailfinger.org/

2 4

Re: [flashrom] [PATCH 08/10] print.c: Fix multiline support.
by Carl-Daniel Hailfinger 17 Jul '13

17 Jul '13

Am 10.07.2013 21:17 schrieb Stefan Tauner: > - Use the reentrant tokenizer version strtok_r to break up vendor and model > names in print.c > - Add implementation of strtok_r for mingw (posix only) > - Free allocated temporary memory again. > > Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> > --- > print.c | 49 ++++++++++++++++++++++++++++++++++++------------- > 1 file changed, 36 insertions(+), 13 deletions(-) > > diff --git a/print.c b/print.c > index 6ca2a57..b163cbf 100644 > --- a/print.c > +++ b/print.c > @@ -25,6 +25,26 @@ > #include "flash.h" > #include "programmer.h" > > +#ifdef __MINGW32__ > +static char* strtok_r(char *str, const char *delim, char **nextp) > +{ > + if (str == NULL) > + str = *nextp; > + > + str += strspn(str, delim); /* Skip leading delimiters */ > + if (*str == '\0') > + return NULL; > + > + char *ret = str; > + str += strcspn(str, delim); /* Find end of token */ > + if (*str != '\0') > + *str++ = '\0'; > + > + *nextp = str; > + return ret; > +} > +#endif OK. In the future, we may want to make strtok_r a wrapper around strtok_s (which is in C11). It might take some time for the C library on various platforms to catch up on this, though. > + > /* > * Return a string corresponding to the bustype parameter. > * Memory is obtained with malloc() and must be freed with free() by the caller. > @@ -69,7 +89,8 @@ static int print_supported_chips(void) > int maxtypelen = strlen("Type") + 1; > const struct flashchip *chip; > char *s; > - char *tmpven, *tmpdev; > + char *ven, *dev; > + char *tmpven, *tmpdev, *tmpven_save, *tmpdev_save; > int tmpvenlen, tmpdevlen, curvenlen, curdevlen; > > /* calculate maximum column widths and by iterating over all chips */ > @@ -179,17 +200,17 @@ static int print_supported_chips(void) > * - after all other values are printed print the surplus tokens > * on fresh lines > */ > - tmpven = malloc(strlen(chip->vendor) + 1); > - if (tmpven == NULL) { > + ven = malloc(strlen(chip->vendor) + 1); > + if (ven == NULL) { > msg_gerr("Out of memory!\n"); > return 1; > } > - strcpy(tmpven, chip->vendor); > + strcpy(ven, chip->vendor); > > - tmpven = strtok(tmpven, delim); > + tmpven = strtok_r(ven, delim, &tmpven_save); > msg_ginfo("%s", tmpven); > curvenlen = strlen(tmpven); > - while ((tmpven = strtok(NULL, delim)) != NULL) { > + while ((tmpven = strtok_r(NULL, delim, &tmpven_save)) != NULL) { > msg_ginfo("%s", delim); > curvenlen++; > tmpvenlen = strlen(tmpven); > @@ -203,17 +224,17 @@ static int print_supported_chips(void) > msg_ginfo(" "); > > /* support for multiline device names as above */ > - tmpdev = malloc(strlen(chip->name) + 1); > - if (tmpdev == NULL) { > + dev = malloc(strlen(chip->name) + 1); > + if (dev == NULL) { > msg_gerr("Out of memory!\n"); > return 1; > } > - strcpy(tmpdev, chip->name); > + strcpy(dev, chip->name); > > - tmpdev = strtok(tmpdev, delim); > + tmpdev = strtok_r(dev, delim, &tmpdev_save); > msg_ginfo("%s", tmpdev); > curdevlen = strlen(tmpdev); > - while ((tmpdev = strtok(NULL, delim)) != NULL) { > + while ((tmpdev = strtok_r(NULL, delim, &tmpdev_save)) != NULL) { > msg_ginfo("%s", delim); > curdevlen++; > tmpdevlen = strlen(tmpdev); > @@ -287,7 +308,7 @@ static int print_supported_chips(void) > if (tmpven != NULL){ > msg_ginfo("%s", tmpven); > curvenlen = strlen(tmpven); > - while ((tmpven = strtok(NULL, delim)) != NULL) { > + while ((tmpven = strtok_r(NULL, delim, &tmpven_save)) != NULL) { > msg_ginfo("%s", delim); > curvenlen++; > tmpvenlen = strlen(tmpven); > @@ -306,7 +327,7 @@ static int print_supported_chips(void) > if (tmpdev != NULL){ > msg_ginfo("%s", tmpdev); > curdevlen = strlen(tmpdev); > - while ((tmpdev = strtok(NULL, delim)) != NULL) { > + while ((tmpdev = strtok_r(NULL, delim, &tmpdev_save)) != NULL) { > msg_ginfo("%s", delim); > curdevlen++; > tmpdevlen = strlen(tmpdev); > @@ -319,6 +340,8 @@ static int print_supported_chips(void) > } > } > msg_ginfo("\n"); > + free(ven); > + free(dev); > } > > return 0; Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Thanks for fixing this! Regards, Carl-Daniel -- http://www.hailfinger.org/

2 1

[commit] r1700 - trunk
by repository service 17 Jul '13

17 Jul '13

Author: stefanct Date: Wed Jul 17 11:28:00 2013 New Revision: 1700 URL: http://flashrom.org/trac/flashrom/changeset/1700 Log: print.c: Fix multiline support. - Use the reentrant tokenizer version strtok_r to break up vendor and model names in print.c. - Add implementation of strtok_r for mingw (strtok_r is POSIX only). - Free allocated temporary memory again. Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Modified: trunk/print.c Modified: trunk/print.c ============================================================================== --- trunk/print.c Mon Jul 15 12:47:53 2013 (r1699) +++ trunk/print.c Wed Jul 17 11:28:00 2013 (r1700) @@ -3,6 +3,7 @@ * * Copyright (C) 2009 Uwe Hermann <uwe(a)hermann-uwe.de> * Copyright (C) 2009 Carl-Daniel Hailfinger + * Copyright (C) 2011-2013 Stefan Tauner * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -25,6 +26,27 @@ #include "flash.h" #include "programmer.h" +/* FIXME: Find a better solution for MinGW. Maybe wrap strtok_s (C11) below if it becomes available */ +#ifdef __MINGW32__ +static char* strtok_r(char *str, const char *delim, char **nextp) +{ + if (str == NULL) + str = *nextp; + + str += strspn(str, delim); /* Skip leading delimiters */ + if (*str == '\0') + return NULL; + + char *ret = str; + str += strcspn(str, delim); /* Find end of token */ + if (*str != '\0') + *str++ = '\0'; + + *nextp = str; + return ret; +} +#endif + /* * Return a string corresponding to the bustype parameter. * Memory is obtained with malloc() and must be freed with free() by the caller. @@ -69,7 +91,8 @@ int maxtypelen = strlen("Type") + 1; const struct flashchip *chip; char *s; - char *tmpven, *tmpdev; + char *ven, *dev; + char *tmpven, *tmpdev, *tmpven_save, *tmpdev_save; int tmpvenlen, tmpdevlen, curvenlen, curdevlen; /* calculate maximum column widths and by iterating over all chips */ @@ -179,17 +202,17 @@ * - after all other values are printed print the surplus tokens * on fresh lines */ - tmpven = malloc(strlen(chip->vendor) + 1); - if (tmpven == NULL) { + ven = malloc(strlen(chip->vendor) + 1); + if (ven == NULL) { msg_gerr("Out of memory!\n"); return 1; } - strcpy(tmpven, chip->vendor); + strcpy(ven, chip->vendor); - tmpven = strtok(tmpven, delim); + tmpven = strtok_r(ven, delim, &tmpven_save); msg_ginfo("%s", tmpven); curvenlen = strlen(tmpven); - while ((tmpven = strtok(NULL, delim)) != NULL) { + while ((tmpven = strtok_r(NULL, delim, &tmpven_save)) != NULL) { msg_ginfo("%s", delim); curvenlen++; tmpvenlen = strlen(tmpven); @@ -203,17 +226,17 @@ msg_ginfo(" "); /* support for multiline device names as above */ - tmpdev = malloc(strlen(chip->name) + 1); - if (tmpdev == NULL) { + dev = malloc(strlen(chip->name) + 1); + if (dev == NULL) { msg_gerr("Out of memory!\n"); return 1; } - strcpy(tmpdev, chip->name); + strcpy(dev, chip->name); - tmpdev = strtok(tmpdev, delim); + tmpdev = strtok_r(dev, delim, &tmpdev_save); msg_ginfo("%s", tmpdev); curdevlen = strlen(tmpdev); - while ((tmpdev = strtok(NULL, delim)) != NULL) { + while ((tmpdev = strtok_r(NULL, delim, &tmpdev_save)) != NULL) { msg_ginfo("%s", delim); curdevlen++; tmpdevlen = strlen(tmpdev); @@ -287,7 +310,7 @@ if (tmpven != NULL){ msg_ginfo("%s", tmpven); curvenlen = strlen(tmpven); - while ((tmpven = strtok(NULL, delim)) != NULL) { + while ((tmpven = strtok_r(NULL, delim, &tmpven_save)) != NULL) { msg_ginfo("%s", delim); curvenlen++; tmpvenlen = strlen(tmpven); @@ -306,7 +329,7 @@ if (tmpdev != NULL){ msg_ginfo("%s", tmpdev); curdevlen = strlen(tmpdev); - while ((tmpdev = strtok(NULL, delim)) != NULL) { + while ((tmpdev = strtok_r(NULL, delim, &tmpdev_save)) != NULL) { msg_ginfo("%s", delim); curdevlen++; tmpdevlen = strlen(tmpdev); @@ -319,6 +342,8 @@ } } msg_ginfo("\n"); + free(ven); + free(dev); } return 0;

1 0

Windows and flashrom
by CJ HUDSON 15 Jul '13

15 Jul '13

Hi, I have successfully used flashrom before and donated to the project. However, I have stopped using Ubuntu because I was experimenting with overclocking and am now building a new PC and have to do a lot of restarts which was a pain with dual boot Windows/ Ubuntu. Hence I am now trying to get flashrom working with windows 7. Please can you help? I get the error message described in the attached .jpg file and cannot seem to install flashrom, despite downloading WinZip and Open Freely (http://www.winzip.com/win/en/index.htm and http://www.openfreely.com/ respectively.) Chris Hudson Esq. A non-text attachment has been stripped. Its content was: flashrom.exe - System Error The program can't start because libusb0.dll is missing from your computer. Try reinstalling the program to fix this problem.

2 2

Re: [flashrom] [PATCH 10/10] Differentiate ultimate error messages depending on programmer used.
by Carl-Daniel Hailfinger 15 Jul '13

15 Jul '13

Am 10.07.2013 21:17 schrieb Stefan Tauner: > With every newly supported programmer the information regarding reboots > on failures becomes more ridiculous. With this patch it is only shown when > the internal programmer module was selected. > > Example outputs for external programmers: > 1) non-fatal: > […] > Reading current flash chip contents... done. FAILED at 0x00000000! Expected=0xff, Found=0x28, failed byte count from 0x00000000-0x0001ffff: 0x1fde7 > ERASE FAILED! > FAILED! > Uh oh. Erase/write failed. Checking if anything changed. > Good. It seems nothing was changed. > Writing to the flash chip apparently didn't do anything. > Please check the connections (especially those to write protection pins) between > the programmer and the flash chip. If you think the error is caused by flashrom > please report this on IRC at chat.freenode.net (channel #flashrom) or > mail flashrom(a)flashrom.org, thanks! > > 2) fatal: > […] > Verifying flash... FAILED at 0x00000000! Expected=0x0f, Found=0xff, failed byte count from 0x00000000-0x0001ffff: 0x1fde6 > Your flash chip is in an unknown state. > Please report this on IRC at chat.freenode.net (channel #flashrom) or > mail flashrom(a)flashrom.org, thanks! > > Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> > Acked-by: David Hendricks <dhendrix(a)google.com> Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Regards, Carl-Daniel -- http://www.hailfinger.org/

2 1

[commit] r1699 - trunk
by repository service 15 Jul '13

15 Jul '13

Author: stefanct Date: Mon Jul 15 12:47:53 2013 New Revision: 1699 URL: http://flashrom.org/trac/flashrom/changeset/1699 Log: Differentiate ultimate error messages depending on programmer used. With every newly supported programmer the information regarding reboots on failures becomes more ridiculous. With this patch it is only shown when the internal programmer module was selected. Example outputs for external programmers: 1) non-fatal: […] Reading current flash chip contents... done. FAILED at 0x00000000! Expected=0xff, Found=0x28, failed byte count from 0x00000000-0x0001ffff: 0x1fde7 ERASE FAILED! FAILED! Uh oh. Erase/write failed. Checking if anything changed. Good. It seems nothing was changed. Writing to the flash chip apparently didn't do anything. Please check the connections (especially those to write protection pins) between the programmer and the flash chip. If you think the error is caused by flashrom please report this on IRC at chat.freenode.net (channel #flashrom) or mail flashrom(a)flashrom.org, thanks! 2) fatal: […] Verifying flash... FAILED at 0x00000000! Expected=0x0f, Found=0xff, failed byte count from 0x00000000-0x0001ffff: 0x1fde6 Your flash chip is in an unknown state. Please report this on IRC at chat.freenode.net (channel #flashrom) or mail flashrom(a)flashrom.org, thanks! Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> Acked-by: David Hendricks <dhendrix(a)google.com> Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Modified: trunk/flashrom.c Modified: trunk/flashrom.c ============================================================================== --- trunk/flashrom.c Sun Jul 14 01:31:37 2013 (r1698) +++ trunk/flashrom.c Mon Jul 15 12:47:53 2013 (r1699) @@ -1493,26 +1493,37 @@ return ret; } -void nonfatal_help_message(void) +static void nonfatal_help_message(void) { - msg_gerr("Writing to the flash chip apparently didn't do anything.\n" - "This means we have to add special support for your board, " - "programmer or flash chip.\n" - "Please report this on IRC at irc.freenode.net (channel " - "#flashrom) or\n" - "mail flashrom(a)flashrom.org!\n" - "-------------------------------------------------------------" - "------------------\n" - "You may now reboot or simply leave the machine running.\n"); -} - -void emergency_help_message(void) -{ - msg_gerr("Your flash chip is in an unknown state.\n" - "Get help on IRC at chat.freenode.net (channel #flashrom) or\n" - "mail flashrom(a)flashrom.org with the subject \"FAILED: <your board name>\"!\n" - "-------------------------------------------------------------------------------\n" - "DO NOT REBOOT OR POWEROFF!\n"); + msg_gerr("Writing to the flash chip apparently didn't do anything.\n"); +#if CONFIG_INTERNAL == 1 + if (programmer == PROGRAMMER_INTERNAL) + msg_gerr("This means we have to add special support for your board, programmer or flash\n" + "chip. Please report this on IRC at chat.freenode.net (channel #flashrom) or\n" + "mail flashrom(a)flashrom.org, thanks!\n" + "-------------------------------------------------------------------------------\n" + "You may now reboot or simply leave the machine running.\n"); + else +#endif + msg_gerr("Please check the connections (especially those to write protection pins) between\n" + "the programmer and the flash chip. If you think the error is caused by flashrom\n" + "please report this on IRC at chat.freenode.net (channel #flashrom) or\n" + "mail flashrom(a)flashrom.org, thanks!\n"); +} + +static void emergency_help_message(void) +{ + msg_gerr("Your flash chip is in an unknown state.\n"); +#if CONFIG_INTERNAL == 1 + if (programmer == PROGRAMMER_INTERNAL) + msg_gerr("Get help on IRC at chat.freenode.net (channel #flashrom) or\n" + "mail flashrom(a)flashrom.org with the subject \"FAILED: <your board name>\"!\n" + "-------------------------------------------------------------------------------\n" + "DO NOT REBOOT OR POWEROFF!\n"); + else +#endif + msg_gerr("Please report this on IRC at chat.freenode.net (channel #flashrom) or\n" + "mail flashrom(a)flashrom.org, thanks!\n"); } /* The way to go if you want a delimited list of programmers */ @@ -1973,8 +1984,7 @@ "anything changed.\n"); if (!flash->chip->read(flash, newcontents, 0, size)) { if (!memcmp(oldcontents, newcontents, size)) { - msg_cinfo("Good. It seems nothing was " - "changed.\n"); + msg_cinfo("Good. It seems nothing was changed.\n"); nonfatal_help_message(); ret = 1; goto out;

1 0

Re: [flashrom] [PATCH 08/10] print.c: Fix multiline support.
by Carl-Daniel Hailfinger 15 Jul '13

15 Jul '13

Am 10.07.2013 21:17 schrieb Stefan Tauner: > - Use the reentrant tokenizer version strtok_r to break up vendor and model > names in print.c > - Add implementation of strtok_r for mingw (posix only) Self-written or copied from a place with comptaible license? > - Free allocated temporary memory again. > > Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> Regards, Carl-Daniel -- http://www.hailfinger.org/

2 1

Re: [flashrom] [PATCH 05/10] Use uintptr_t for chipaddr instead of unsigned long.
by Carl-Daniel Hailfinger 14 Jul '13

14 Jul '13

Am 10.07.2013 21:17 schrieb Stefan Tauner: > Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> > > diff --git a/flash.h b/flash.h > index ba93245..9e787c0 100644 > --- a/flash.h > +++ b/flash.h > @@ -41,7 +41,8 @@ > #define ERROR_OOM -100 > #define TIMEOUT_ERROR -101 > > -typedef unsigned long chipaddr; > +/* TODO: check using code for correct usage of types */ > +typedef uintptr_t chipaddr; This looks too good to be viable. I have a bad feeling about this in the context of 64-bit resources on x86 (32bit), but drivers accessing resources shouldn't be using chipaddr anyway.. That said, it should not result in any breakage on 32bit platforms. We may see pretty explosions once we start supporting the x32 pseudo-arch. But that will need an audit of the codebase anyway. Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Regards, Carl-Daniel -- http://www.hailfinger.org/

2 2

[commit] r1698 - trunk
by repository service 14 Jul '13

14 Jul '13

Author: stefanct Date: Sun Jul 14 01:31:37 2013 New Revision: 1698 URL: http://flashrom.org/trac/flashrom/changeset/1698 Log: Use uintptr_t for chipaddr instead of unsigned long. Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Modified: trunk/dummyflasher.c trunk/flash.h trunk/jedec.c trunk/serprog.c trunk/sst49lfxxxc.c Modified: trunk/dummyflasher.c ============================================================================== --- trunk/dummyflasher.c Sun Jul 14 01:21:05 2013 (r1697) +++ trunk/dummyflasher.c Sun Jul 14 01:31:37 2013 (r1698) @@ -428,26 +428,26 @@ static void dummy_chip_writeb(const struct flashctx *flash, uint8_t val, chipaddr addr) { - msg_pspew("%s: addr=0x%lx, val=0x%02x\n", __func__, addr, val); + msg_pspew("%s: addr=0x%" PRIxPTR ", val=0x%02x\n", __func__, addr, val); } static void dummy_chip_writew(const struct flashctx *flash, uint16_t val, chipaddr addr) { - msg_pspew("%s: addr=0x%lx, val=0x%04x\n", __func__, addr, val); + msg_pspew("%s: addr=0x%" PRIxPTR ", val=0x%04x\n", __func__, addr, val); } static void dummy_chip_writel(const struct flashctx *flash, uint32_t val, chipaddr addr) { - msg_pspew("%s: addr=0x%lx, val=0x%08x\n", __func__, addr, val); + msg_pspew("%s: addr=0x%" PRIxPTR ", val=0x%08x\n", __func__, addr, val); } static void dummy_chip_writen(const struct flashctx *flash, uint8_t *buf, chipaddr addr, size_t len) { size_t i; - msg_pspew("%s: addr=0x%lx, len=0x%08lx, writing data (hex):", + msg_pspew("%s: addr=0x%" PRIxPTR ", len=0x%08lx, writing data (hex):", __func__, addr, (unsigned long)len); for (i = 0; i < len; i++) { if ((i % 16) == 0) @@ -459,28 +459,28 @@ static uint8_t dummy_chip_readb(const struct flashctx *flash, const chipaddr addr) { - msg_pspew("%s: addr=0x%lx, returning 0xff\n", __func__, addr); + msg_pspew("%s: addr=0x%" PRIxPTR ", returning 0xff\n", __func__, addr); return 0xff; } static uint16_t dummy_chip_readw(const struct flashctx *flash, const chipaddr addr) { - msg_pspew("%s: addr=0x%lx, returning 0xffff\n", __func__, addr); + msg_pspew("%s: addr=0x%" PRIxPTR ", returning 0xffff\n", __func__, addr); return 0xffff; } static uint32_t dummy_chip_readl(const struct flashctx *flash, const chipaddr addr) { - msg_pspew("%s: addr=0x%lx, returning 0xffffffff\n", __func__, addr); + msg_pspew("%s: addr=0x%" PRIxPTR ", returning 0xffffffff\n", __func__, addr); return 0xffffffff; } static void dummy_chip_readn(const struct flashctx *flash, uint8_t *buf, const chipaddr addr, size_t len) { - msg_pspew("%s: addr=0x%lx, len=0x%lx, returning array of 0xff\n", + msg_pspew("%s: addr=0x%" PRIxPTR ", len=0x%lx, returning array of 0xff\n", __func__, addr, (unsigned long)len); memset(buf, 0xff, len); return; Modified: trunk/flash.h ============================================================================== --- trunk/flash.h Sun Jul 14 01:21:05 2013 (r1697) +++ trunk/flash.h Sun Jul 14 01:31:37 2013 (r1698) @@ -41,7 +41,8 @@ #define ERROR_OOM -100 #define TIMEOUT_ERROR -101 -typedef unsigned long chipaddr; +/* TODO: check using code for correct usage of types */ +typedef uintptr_t chipaddr; int register_shutdown(int (*function) (void *data), void *data); void *programmer_map_flash_region(const char *descr, unsigned long phys_addr, Modified: trunk/jedec.c ============================================================================== --- trunk/jedec.c Sun Jul 14 01:21:05 2013 (r1697) +++ trunk/jedec.c Sun Jul 14 01:31:37 2013 (r1698) @@ -376,7 +376,7 @@ dst++, src++; } if (failed) - msg_cerr(" writing sector at 0x%lx failed!\n", olddst); + msg_cerr(" writing sector at 0x%" PRIxPTR " failed!\n", olddst); return failed; } @@ -417,8 +417,7 @@ goto retry; } if (failed) { - msg_cerr(" page 0x%lx failed!\n", - (d - bios) / page_size); + msg_cerr(" page 0x%" PRIxPTR " failed!\n", (d - bios) / page_size); } return failed; } Modified: trunk/serprog.c ============================================================================== --- trunk/serprog.c Sun Jul 14 01:21:05 2013 (r1697) +++ trunk/serprog.c Sun Jul 14 01:31:37 2013 (r1698) @@ -789,7 +789,7 @@ sp_flush_stream(); if (serialport_read(&c, 1) != 0) sp_die("readb byteread"); - msg_pspew("%s addr=0x%lx returning 0x%02X\n", __func__, addr, c); + msg_pspew("%s addr=0x%" PRIxPTR " returning 0x%02X\n", __func__, addr, c); return c; } @@ -797,7 +797,7 @@ static void sp_do_read_n(uint8_t * buf, const chipaddr addr, size_t len) { unsigned char sbuf[6]; - msg_pspew("%s: addr=0x%lx len=%lu\n", __func__, addr, (unsigned long)len); + msg_pspew("%s: addr=0x%" PRIxPTR " len=%lu\n", __func__, addr, (unsigned long)len); /* Stream the read-n -- as above. */ if ((sp_opbuf_usage) || (sp_max_write_n && sp_write_n_bytes)) sp_execute_opbuf_noflush(); Modified: trunk/sst49lfxxxc.c ============================================================================== --- trunk/sst49lfxxxc.c Sun Jul 14 01:21:05 2013 (r1697) +++ trunk/sst49lfxxxc.c Sun Jul 14 01:31:37 2013 (r1698) @@ -41,7 +41,7 @@ unsigned int i, left = flash->chip->total_size * 1024; unsigned long address; - msg_cdbg("\nbios=0x%08lx\n", registers); + msg_cdbg("\nbios=0x%08" PRIxPTR "\n", registers); for (i = 0; left > 65536; i++, left -= 65536) { write_lockbits_block_49lfxxxc(flash, i * 65536, bits); }

1 0

Re: [flashrom] [PATCH 03/10] Various cross-platform fixes.
by Carl-Daniel Hailfinger 14 Jul '13

14 Jul '13

Am 10.07.2013 21:17 schrieb Stefan Tauner: > Improve compilation with libpayload (compiling flashrom.c and > linking is still broken): > - disable Ponyprog (which enforced serial.c compilation) > - make errno available where it is needed > > Fix internal.c for non-x86 and enable cb parsing on ARM. > > Fix mingw builds by using its __USE_MINGW_ANSI_STDIO macro > and gnu_printf definition for printf format style checking. > See http://sourceforge.net/apps/trac/mingw-w64/wiki/gnu%20printf > This requires inclusion of stdio.h in flash.h. > > Fix order of libraries in the Makefile: > FEATURE_LIBS needs to come *after* PCILIBS in case ZLIB is needed by it. > > Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> > Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> Apart from the library order, I'm happy with it. > > diff --git a/Makefile b/Makefile > index 805290c..34bbab8 100644 > --- a/Makefile > +++ b/Makefile > @@ -632,7 +641,7 @@ ifeq ($(ARCH), x86) > endif > > $(PROGRAM)$(EXEC_SUFFIX): $(OBJS) > - $(CC) $(LDFLAGS) -o $(PROGRAM)$(EXEC_SUFFIX) $(OBJS) $(FEATURE_LIBS) $(LIBS) $(PCILIBS) $(USBLIBS) > + $(CC) $(LDFLAGS) -o $(PROGRAM)$(EXEC_SUFFIX) $(OBJS) $(LIBS) $(PCILIBS) $(USBLIBS) $(FEATURE_LIBS) I believe this is partially wrong. AFAICS the order should be: LIBS, PCILIBS, FEATURE_LIBS, USBLIBS. Explanation: The linker doesn't remember defined symbols if they were not on its list of needed symbols. This means you have to first specify a library and then its dependencies. LIBS is a misnomer (not generic libraries needed for everything, but rather stuff like getopt and socket for some OS). This confused me initially. The place of LIBS can be pretty much anywhere because it is not needed by any other library. FEATURE_LIBS can contain libftdi (so it needs to come before USBLIBS which satisifies libftdi dependencies), but it also can contain libz (so it needs to come after PCILIBS which may have libz as dependency). To really get this in good shape, we should move libz into LIBS instead of FEATURE_LIBS, but that breaks other parts of the Makefile. I really really want a separate configure script to solve this. > > libflashrom.a: $(LIBFLASHROM_OBJS) > $(AR) rcs $@ $^ > diff --git a/physmap.c b/physmap.c > index 524f558..644ee35 100644 > --- a/physmap.c > +++ b/physmap.c > @@ -120,15 +120,6 @@ void *sys_physmap(unsigned long phys_addr, size_t len) > void physunmap(void *virt_addr, size_t len) > { > } > - > -int setup_cpu_msr(int cpu) > -{ > - return 0; > -} > - > -void cleanup_cpu_msr(void) > -{ > -} > #elif defined(__MACH__) && defined(__APPLE__) > > #define MEM_DEV "DirectHW" > @@ -568,6 +559,15 @@ int libpayload_wrmsr(int addr, msr_t msr) > _wrmsr(addr, msr.lo | ((unsigned long long)msr.hi << 32)); > return 0; > } > + > +int setup_cpu_msr(int cpu) > +{ > + return 0; > +} > + > +void cleanup_cpu_msr(void) > +{ > +} > #else > /* default MSR implementation */ > msr_t rdmsr(int addr) The two hunks above are what I wanted in patch 2 of this series. They fit a lot better here, though, so don't change it. Sorry. With the library order fixed, this is Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Regards, Carl-Daniel -- http://www.hailfinger.org/

2 1

[commit] r1697 - trunk
by repository service 14 Jul '13

14 Jul '13

Author: stefanct Date: Sun Jul 14 01:21:05 2013 New Revision: 1697 URL: http://flashrom.org/trac/flashrom/changeset/1697 Log: Various cross-platform fixes. Improve compilation with libpayload (compiling flashrom.c and linking is still broken): - disable Ponyprog (which enforced serial.c compilation) - make errno available where it is needed Fix internal.c for non-x86 and enable cb parsing on ARM. Fix mingw builds by using its __USE_MINGW_ANSI_STDIO macro and gnu_printf definition for printf format style checking. See http://sourceforge.net/apps/trac/mingw-w64/wiki/gnu%20printf This requires inclusion of stdio.h in flash.h. Fix order of libraries in the Makefile: FEATURE_LIBS needs to come *after* PCILIBS in case ZLIB is needed by it. Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Modified: trunk/Makefile trunk/flash.h trunk/flashrom.c trunk/hwaccess.c trunk/internal.c trunk/physmap.c Modified: trunk/Makefile ============================================================================== --- trunk/Makefile Sat Jul 13 23:18:32 2013 (r1696) +++ trunk/Makefile Sun Jul 14 01:21:05 2013 (r1697) @@ -140,6 +140,9 @@ EXEC_SUFFIX := .exe # MinGW doesn't have the ffs() function, but we can use gcc's __builtin_ffs(). FLASHROM_CFLAGS += -Dffs=__builtin_ffs +# Some functions provided by Microsoft do not work as described in C99 specifications. This macro fixes that +# for MinGW. See http://sourceforge.net/apps/trac/mingw-w64/wiki/printf%20and%20scanf%20fami… */ +FLASHROM_CFLAGS += -D__USE_MINGW_ANSI_STDIO=1 # libusb-win32/libftdi stuff is usually installed in /usr/local. CPPFLAGS += -I/usr/local/include LDFLAGS += -L/usr/local/lib @@ -218,6 +221,7 @@ else override CONFIG_DUMMY = no endif +# Bus Pirate, Serprog and PonyProg are not supported with libpayload (missing serial support). ifeq ($(CONFIG_BUSPIRATE_SPI), yes) UNSUPPORTED_FEATURES += CONFIG_BUSPIRATE_SPI=yes else @@ -228,6 +232,11 @@ else override CONFIG_SERPROG = no endif +ifeq ($(CONFIG_PONY_SPI), yes) +UNSUPPORTED_FEATURES += CONFIG_PONY_SPI=yes +else +override CONFIG_PONY_SPI = no +endif # Dediprog and FT2232 are not supported with libpayload (missing libusb support) ifeq ($(CONFIG_DEDIPROG), yes) UNSUPPORTED_FEATURES += CONFIG_DEDIPROG=yes @@ -632,7 +641,7 @@ endif $(PROGRAM)$(EXEC_SUFFIX): $(OBJS) - $(CC) $(LDFLAGS) -o $(PROGRAM)$(EXEC_SUFFIX) $(OBJS) $(FEATURE_LIBS) $(LIBS) $(PCILIBS) $(USBLIBS) + $(CC) $(LDFLAGS) -o $(PROGRAM)$(EXEC_SUFFIX) $(OBJS) $(LIBS) $(PCILIBS) $(FEATURE_LIBS) $(USBLIBS) libflashrom.a: $(LIBFLASHROM_OBJS) $(AR) rcs $@ $^ Modified: trunk/flash.h ============================================================================== --- trunk/flash.h Sat Jul 13 23:18:32 2013 (r1696) +++ trunk/flash.h Sun Jul 14 01:21:05 2013 (r1697) @@ -25,6 +25,7 @@ #define __FLASH_H__ 1 #include <inttypes.h> +#include <stdio.h> #include <stdint.h> #include <stddef.h> #include <stdbool.h> @@ -286,7 +287,12 @@ MSG_SPEW = 5, }; /* Let gcc and clang check for correct printf-style format strings. */ -int print(enum msglevel level, const char *fmt, ...) __attribute__((format(printf, 2, 3))); +int print(enum msglevel level, const char *fmt, ...) +#ifdef __MINGW32__ +__attribute__((format(gnu_printf, 2, 3))); +#else +__attribute__((format(printf, 2, 3))); +#endif #define msg_gerr(...) print(MSG_ERROR, __VA_ARGS__) /* general errors */ #define msg_perr(...) print(MSG_ERROR, __VA_ARGS__) /* programmer errors */ #define msg_cerr(...) print(MSG_ERROR, __VA_ARGS__) /* chip errors */ Modified: trunk/flashrom.c ============================================================================== --- trunk/flashrom.c Sat Jul 13 23:18:32 2013 (r1696) +++ trunk/flashrom.c Sun Jul 14 01:21:05 2013 (r1697) @@ -1168,7 +1168,7 @@ return 1; } if (image_stat.st_size != size) { - msg_gerr("Error: Image size (%jd B) doesn't match the flash chip's size (%ld B)!\n", + msg_gerr("Error: Image size (%jd B) doesn't match the flash chip's size (%lu B)!\n", (intmax_t)image_stat.st_size, size); fclose(image); return 1; Modified: trunk/hwaccess.c ============================================================================== --- trunk/hwaccess.c Sat Jul 13 23:18:32 2013 (r1696) +++ trunk/hwaccess.c Sun Jul 14 01:21:05 2013 (r1697) @@ -35,14 +35,13 @@ #include <stdint.h> #include <string.h> #include <stdlib.h> +#include <errno.h> #include <sys/types.h> #if !defined (__DJGPP__) && !defined(__LIBPAYLOAD__) +/* No file access needed/possible to get hardware access permissions. */ #include <unistd.h> #include <fcntl.h> #endif -#if !defined (__DJGPP__) -#include <errno.h> -#endif #include "flash.h" #include "hwaccess.h" Modified: trunk/internal.c ============================================================================== --- trunk/internal.c Sat Jul 13 23:18:32 2013 (r1696) +++ trunk/internal.c Sun Jul 14 01:21:05 2013 (r1697) @@ -172,8 +172,10 @@ int not_a_laptop = 0; const char *board_vendor = NULL; const char *board_model = NULL; +#if defined (__i386__) || defined (__x86_64__) || defined (__arm__) const char *cb_vendor = NULL; const char *cb_model = NULL; +#endif char *arg; arg = extract_programmer_param("boardenable"); @@ -254,7 +256,7 @@ return 1; } -#if defined(__i386__) || defined(__x86_64__) +#if defined(__i386__) || defined(__x86_64__) || defined (__arm__) if ((cb_parse_table(&cb_vendor, &cb_model) == 0) && (board_vendor != NULL) && (board_model != NULL)) { if (strcasecmp(board_vendor, cb_vendor) || strcasecmp(board_model, cb_model)) { msg_pwarn("Warning: The mainboard IDs set by -p internal:mainboard (%s:%s) do not\n" @@ -265,7 +267,9 @@ msg_pinfo("Continuing anyway.\n"); } } +#endif +#if defined(__i386__) || defined(__x86_64__) dmi_init(); /* In case Super I/O probing would cause pretty explosions. */ Modified: trunk/physmap.c ============================================================================== --- trunk/physmap.c Sat Jul 13 23:18:32 2013 (r1696) +++ trunk/physmap.c Sun Jul 14 01:21:05 2013 (r1697) @@ -24,14 +24,14 @@ #include <stdio.h> #include <stdlib.h> #include <string.h> +#include <errno.h> #include "flash.h" #include "hwaccess.h" -/* Do we need any file access or ioctl for physmap or MSR? */ #if !defined(__DJGPP__) && !defined(__LIBPAYLOAD__) +/* No file access needed/possible to get mmap access permissions or access MSR. */ #include <sys/stat.h> #include <fcntl.h> -#include <errno.h> #endif #ifdef __DJGPP__ @@ -120,15 +120,6 @@ void physunmap(void *virt_addr, size_t len) { } - -int setup_cpu_msr(int cpu) -{ - return 0; -} - -void cleanup_cpu_msr(void) -{ -} #elif defined(__MACH__) && defined(__APPLE__) #define MEM_DEV "DirectHW" @@ -569,6 +560,15 @@ _wrmsr(addr, msr.lo | ((unsigned long long)msr.hi << 32)); return 0; } + +int setup_cpu_msr(int cpu) +{ + return 0; +} + +void cleanup_cpu_msr(void) +{ +} #else /* default MSR implementation */ msr_t rdmsr(int addr)

1 0

Re: [flashrom] [PATCH 02/10] Add support for AMD Geode's MSR on OpenBSD.
by Carl-Daniel Hailfinger 13 Jul '13

13 Jul '13

Am 10.07.2013 21:17 schrieb Stefan Tauner: > Previously the default implementation was used, which always failed. > On other systems than the Geode the code is not equal to that anymore, > but should not explode because setup_cpu_msr() returns an error and > therefore no other MSR functions should be called. > > Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> > Tested-by Leonardo Guardati <leonardo(a)guardati.it> > Acked-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> > > Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> Please kill one of your signoffs. I like the simplification of the previous #ifdef mess. Cosmetic suggestions (commit is OK without them as well): Move the libpayload setup_cpu_msr/cleanup_cpu_msr from the top of the file down to the block right after libpayload_wrmsr. This has the advantage of making the msr setup code for libpayload dependent on x86 instead of having it enabled for all architectures. Add a comment to the OSX MSR functions which mentions that rdmsr/wrmsr are provided by DirectHW. Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Regards, Carl-Daniel -- http://www.hailfinger.org/

2 1

[commit] r1696 - trunk
by repository service 13 Jul '13

13 Jul '13

Author: stefanct Date: Sat Jul 13 23:18:32 2013 New Revision: 1696 URL: http://flashrom.org/trac/flashrom/changeset/1696 Log: Add support for AMD Geode's MSR on OpenBSD. Previously the default implementation was used, which always failed. On other systems than the Geode the code is not equal to that anymore, but should not explode because setup_cpu_msr() returns an error and therefore no other MSR functions should be called. Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> Tested-by Leonardo Guardati <leonardo(a)guardati.it> Acked-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Modified: trunk/physmap.c Modified: trunk/physmap.c ============================================================================== --- trunk/physmap.c Sat Jul 13 22:55:33 2013 (r1695) +++ trunk/physmap.c Sat Jul 13 23:18:32 2013 (r1696) @@ -282,6 +282,8 @@ PHYSMAP_RO); } +/* MSR abstraction implementations for Linux, OpenBSD, FreeBSD/Dragonfly, OSX, libpayload + * and a non-working default implemenation on the bottom. See also hwaccess.h for some (re)declarations. */ #if defined(__i386__) || defined(__x86_64__) #ifdef __linux__ @@ -381,8 +383,83 @@ /* Clear MSR file descriptor. */ fd_msr = -1; } -#else -#if defined(__FreeBSD__) || defined(__FreeBSD_kernel__) || defined(__DragonFly__) +#elif defined(__OpenBSD__) /* This does only work for certain AMD Geode LX systems see amdmsr(4). */ +#include <sys/ioctl.h> +#include <machine/amdmsr.h> + +static int fd_msr = -1; + +msr_t rdmsr(int addr) +{ + struct amdmsr_req args; + + msr_t msr = { 0xffffffff, 0xffffffff }; + + args.addr = (uint32_t)addr; + + if (ioctl(fd_msr, RDMSR, &args) < 0) { + msg_perr("Error while executing RDMSR ioctl: %s\n", strerror(errno)); + close(fd_msr); + exit(1); + } + + msr.lo = args.val & 0xffffffff; + msr.hi = args.val >> 32; + + return msr; +} + +int wrmsr(int addr, msr_t msr) +{ + struct amdmsr_req args; + + args.addr = addr; + args.val = (((uint64_t)msr.hi) << 32) | msr.lo; + + if (ioctl(fd_msr, WRMSR, &args) < 0) { + msg_perr("Error while executing WRMSR ioctl: %s\n", strerror(errno)); + close(fd_msr); + exit(1); + } + + return 0; +} + +int setup_cpu_msr(int cpu) +{ + char msrfilename[64]; + memset(msrfilename, 0, sizeof(msrfilename)); + snprintf(msrfilename, sizeof(msrfilename), "/dev/amdmsr"); + + if (fd_msr != -1) { + msg_pinfo("MSR was already initialized\n"); + return -1; + } + + fd_msr = open(msrfilename, O_RDWR); + + if (fd_msr < 0) { + msg_perr("Error while opening %s: %s\n", msrfilename, strerror(errno)); + return -1; + } + + return 0; +} + +void cleanup_cpu_msr(void) +{ + if (fd_msr == -1) { + msg_pinfo("No MSR initialized.\n"); + return; + } + + close(fd_msr); + + /* Clear MSR file descriptor. */ + fd_msr = -1; +} + +#elif defined(__FreeBSD__) || defined(__FreeBSD_kernel__) || defined(__DragonFly__) #include <sys/ioctl.h> typedef struct { @@ -465,9 +542,8 @@ fd_msr = -1; } -#else - -#if defined(__MACH__) && defined(__APPLE__) +#elif defined(__MACH__) && defined(__APPLE__) +/* rdmsr() and wrmsr() are provided by DirectHW which needs neither setup nor cleanup. */ int setup_cpu_msr(int cpu) { // Always succeed for now @@ -494,6 +570,7 @@ return 0; } #else +/* default MSR implementation */ msr_t rdmsr(int addr) { msr_t ret = { 0xffffffff, 0xffffffff }; @@ -516,9 +593,7 @@ { // Nothing, yet. } -#endif -#endif -#endif -#else +#endif // OS switches for MSR code +#else // x86 /* Does MSR exist on non-x86 architectures? */ -#endif +#endif // arch switches for MSR code

1 0

Re: [flashrom] [PATCH 01/10] Fix hwaccess (again).
by Carl-Daniel Hailfinger 13 Jul '13

13 Jul '13

Am 10.07.2013 21:17 schrieb Stefan Tauner: > Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> This changes the behaviour for FreeBSD (and derivatives). Is this what you wanted? If yes, please change the commit message to Fix hwaccess for FreeBSD Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Regards, Carl-Daniel -- http://www.hailfinger.org/

2 1

[commit] r1695 - trunk
by repository service 13 Jul '13

13 Jul '13

Author: stefanct Date: Sat Jul 13 22:55:33 2013 New Revision: 1695 URL: http://flashrom.org/trac/flashrom/changeset/1695 Log: Fix hwaccess (again). I completely forgot about NetBSD in r1638 which was broken since then. Also, on OpenBSD use iopl again and refine the warning in case we have insufficient privileges. Thanks to Jonathan A. Kollasch for his help in getting the details right. Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Modified: trunk/hwaccess.c Modified: trunk/hwaccess.c ============================================================================== --- trunk/hwaccess.c Thu Jul 11 15:48:52 2013 (r1694) +++ trunk/hwaccess.c Sat Jul 13 22:55:33 2013 (r1695) @@ -26,10 +26,9 @@ #error Unknown architecture #endif -#define IS_BSD (defined(__FreeBSD__) || defined(__FreeBSD_kernel__) || defined(__DragonFly__) || defined(__OpenBSD__)) #define IS_LINUX (defined(__gnu_linux__) || defined(__linux__)) #define IS_MACOSX (defined(__APPLE__) && defined(__MACH__)) -#if !(IS_BSD || IS_LINUX || IS_MACOSX || defined(__DJGPP__) || defined(__LIBPAYLOAD__) || defined(__sun)) +#if !(IS_LINUX || IS_MACOSX || defined(__NetBSD__) || defined(__OpenBSD__) || defined(__FreeBSD__) || defined(__FreeBSD_kernel__) || defined(__DragonFly__) || defined(__DJGPP__) || defined(__LIBPAYLOAD__) || defined(__sun)) #error "Unknown operating system" #endif @@ -47,7 +46,10 @@ #include "flash.h" #include "hwaccess.h" -#if IS_X86 && IS_BSD +#define USE_IOPL (IS_LINUX || IS_MACOSX || defined(__NetBSD__) || defined(__OpenBSD__)) +#define USE_DEV_IO (defined(__FreeBSD__) || defined(__FreeBSD_kernel__) || defined(__DragonFly__)) + +#if IS_X86 && USE_DEV_IO int io_fd; #endif @@ -71,9 +73,9 @@ { #if defined (__sun) sysi86(SI86V86, V86SC_IOPL, 0); -#elif IS_BSD +#elif USE_DEV_IO close(io_fd); -#elif IS_LINUX || IS_MACOSX +#elif USE_IOPL iopl(0); #endif return 0; @@ -86,16 +88,19 @@ #if IS_X86 && !(defined(__DJGPP__) || defined(__LIBPAYLOAD__)) #if defined (__sun) if (sysi86(SI86V86, V86SC_IOPL, PS_IOPL) != 0) { -#elif IS_BSD +#elif USE_DEV_IO if ((io_fd = open("/dev/io", O_RDWR)) < 0) { -#elif IS_LINUX || IS_MACOSX +#elif USE_IOPL if (iopl(3) != 0) { #endif - msg_perr("ERROR: Could not get I/O privileges (%s).\n" - "You need to be root.\n", strerror(errno)); + msg_perr("ERROR: Could not get I/O privileges (%s).\n", strerror(errno)); + msg_perr("You need to be root.\n"); #if defined (__OpenBSD__) - msg_perr("Please set securelevel=-1 in /etc/rc.securelevel and reboot, or reboot into \n"); - msg_perr("single user mode.\n"); + msg_perr("If you are root already please set securelevel=-1 in /etc/rc.securelevel and\n" + "reboot, or reboot into single user mode.\n"); +#elif defined(__NetBSD__) + msg_perr("If you are root already please reboot into single user mode or make sure\n" + "that your kernel configuration has the option INSECURE enabled.\n"); #endif return 1; } else {

1 0

Re: [flashrom] Want to donate 110 Euro to flashrom.org thru paypal if can get 45 minute help with flashing BIOS
by Stefan Tauner 12 Jul '13

12 Jul '13

On Fri, 12 Jul 2013 10:15:36 +0700 jakob neubert <jakobneubert(a)fastmail.fm> wrote: > Well, am a non-programmer so just trying to figure out what to write > correctly in the terminal window is very difficult. Spent 3 hours on > that yesterday. > > Can't make it save the BIOS - and - can't make it erase BIOS either. > When doing that then it says that write protected. > > When trying to flash BIOS then it says that the motehrbnoard is not > supported - but, its a common mother board with common parts on it from > Realtek, Broadcom and so on. You are describing the things you tried and their outcome very sketchy. You have not even mentioned if you have used flashrom or the vendor tools(!). We can not diagnose anything that way. We need to know exactly what commands you used and what the response was. If you have flashrom installed already send us the log file generated by "flashrom -p internal -o flashrom.log" please. That would be a good start. -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

1 0

Re: [flashrom] [RFC] Layout *file* handling...
by Stefan Tauner 12 Jul '13

12 Jul '13

On Thu, 11 Jul 2013 12:31:55 -0700 David Hendricks <dhendrix(a)google.com> wrote: > On Thu, Jul 11, 2013 at 9:42 AM, Stefan Tauner < > stefan.tauner(a)student.tuwien.ac.at> wrote: > > > > - Default inclusion status. It would be nice to be able to mark some > > regions as to be included by default so that one does not even need > > to use -i. (no idea about syntax yet) > > > > This one could get complicated... Maybe for the layout 2.0 you can add a > column to specify include options like "always" or "never". That could be > made into a comma-separated list with other region attributes (TBD). Maybe we should think more broadly about what we want to achieve. The layout file as it is and the features I enumerated are very much focused on describing attributes of address ranges. That is fine if it is our main concern, but I am not sure it is. Eventually we want to make it easier to describe access patterns to a memory device, so maybe a more imperative approach would be better that describes what flashrom should do. We would end up with a configuration file that would be able to replace everything we can declare on the command line now + current layout files + your write protection commands + a schedule of what of the above should be used/done when. The more I think about it the more I think this is wrong and way more easily handled by real scipting languages together with a libflashrom binding. :) > > - Includes of other layout files (include .*) > > > > Interesting! I like that idea. Do you have use cases for that? We dont want to implement features just because they sound nice to have :) > - Some kind of priority if one selects overlapping regions. > > Interesting... Not sure I like it though. It seems too easy to botch this > IMO. Yes indeed. :) These ideas are a food for thought and nothing else. I would like to get more feedback from the potential users regarding (these) features and not invent stuff no one actually needs. As Stefan wrote already the normal work flow is to get a complete image out of the build system... -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

1 0

[RFC] Layout handling...
by Carl-Daniel Hailfinger 12 Jul '13

12 Jul '13

Hi, before we change flashrom to work with current layout requirements, we should summarize the features we need/want, and then decide how to handle the individual layout regions internally. Terms: "write strategy" describes a combination of erase+write commands with a given block size, touching some blocks. "read protected" describes a region which can not be read. "write protected" describes a region which can not be written. "write once" describes a region which can be written exactly once in the chip lifetime. If you add a feature to the list, please give it a nickname so we know which feature people are referring to. Some of the features below might not be desirable, but I want to list them anyway so we can explicitly declare them as unsupported if needed. fullwrite-unrestricted: Write a chip-sized image, no special read/write restrictions of the chip, no layout file needs to be specified. Default write case right now. fullwrite-noread: Write a chip-sized image, reading anything from the chip is not possible. Many DVD writers fall in that category. No verification, violates our reliability guarantees. fullwrite-postread: Write a chip-sized image, reading anything from the chip is only possible after write. Chips which are read-locked until a full erase/write fall in that category. Do those exist as standalone flash chips or only integrated into processors? fullwrite-partialread: Write a chip-sized image, reading is only possible in some chip regions. Only partial verification, violates our reliability guarantees. fullwrite-partialwrite: Write a chip-sized image, but writing is only possible in some regions. This is obviously a conflict unless the image has the same contents as the chip in the write-protected regions and there is a possible write strategy for the whole image which does not touch the write-protected regions. Should flashrom always refuse this scenario, or only refuse it in case of conflicts? partialwrite-unrestricted: Write only parts of an image, the rest of the chip contents is kept, no special read/write restrictions. partialwrite-partialread: Write only parts of an image, the rest of the chip contents is kept, reading is only possible in some chip regions. If no read-protected regions are written and a suitable write strategy exists, should flashrom warn? If a read-protected region is written, should flashrom warn/refuse due to reliability requirements? partialwrite-partialwrite: Write only parts of an image, the rest of the chip contents is kept, writing is only possible in some chip regions. If no write-protected regions are written and a suitable write strategy exists, should flashrom warn? flashrom will refuse to write a write-protected region. fullread-unrestricted: Read the full chip, no special read restrictions of the chip. partialread-unrestricted: Read only parts of a chip, no special read restrictions of the chip. partialread-partialread: Read only parts of a chip, some regions are read-protected. flashrom should refuse to read any read-protected regions. partialread-imagefiller: If only parts of a chip are read and the read image has full chip size, what should be used as filler for unread regions? 0x00 or 0xff? partialread-layout-imagesize: If only parts of a chip are read, should the read image still have full chip size with all read regions filled in at the respective locations? partialread-layout-split: If only parts of a chip are read, should it be possible to write each region (or a combination thereof) to a separate image file, and would that mapping be specified in the layout file? partialwrite-layout-split: If only parts of a chip are written, should it be possible to collect each part of the new image from a separate image file, and would that mapping be specified in the layout file? readwrite-protection-time: Which read protection and write protection times exist? Temporary lock until unlock, temporary lock until chip soft reset, temporary lock until chip/programmer hard reset (powerdown or reset line), permanent eternal lock. readwrite-protection-type: Which read protection and write protection types exist? Programmer lock (e.g. Intel SPI), hardware chip lock (WP pin), software chip lock (chip command). readwrite-protection-interaction: How should we express this situation: A region is write-locked with a software chip lock, but to remove that software chip lock, a hardware chip lock has to be disabled first, then the software chip lock can be disabled. partialaccess-crash: Some regions in the chip are really off-limits and will cause an unrecoverable error (system crash) when accessed (read or write). That seems to be the case for some EC/flash interactions. Comments? Any forks of flashrom (e.g. chromium) which need infrastructure features not mentioned above? Regards, Carl-Daniel -- http://www.hailfinger.org/

5 12

Re: [flashrom] [RFC] Layout *file* handling...
by Stefan Reinauer 12 Jul '13

12 Jul '13

On 7/11/13 12:31 PM, David Hendricks wrote: > > in the cli) > > > +1. It's pretty silly to need to specify a ROM-sized file for "-w" > just to update a single region. > While this is true, ROM-sized files are typically what falls out of pretty much everyone's build system, so we should retain this option. Stefan

2 1

Re: [flashrom] Want to donate 110 Euro to flashrom.org thru paypal if can get 45 minute help with flashing BIOS
by Stefan Tauner 12 Jul '13

12 Jul '13

On Wed, 10 Jul 2013 17:22:24 +0700 jakob neubert <jakobneubert(a)fastmail.fm> wrote: > I am un-able to flash my computers with flashrom and have already been > using gparted, rescuecd, linuxmint, Ubuntu and so on. > > > > They are Acer aspire 4755G and Acer aspire v3-471g laptops (I7 CPU, 8 > GB ram). Please provide some verbose output and/or error messages, for example the complete output of flashrom -p internal -VV -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

2 1

Re: [flashrom] [RFC] Layout *file* handling...
by Stefan Tauner 11 Jul '13

11 Jul '13

On Thu, 11 Jul 2013 13:55:38 -0700 David Hendricks <dhendrix(a)google.com> wrote: > On Thu, Jul 11, 2013 at 1:12 PM, Stefan Tauner < > stefan.tauner(a)student.tuwien.ac.at> wrote: > > > On Thu, 11 Jul 2013 12:31:55 -0700 > > David Hendricks <dhendrix(a)google.com> wrote: > > > > > > - Default image names. For example one might want to always use me.bin > > > > for the ME region without the need to specify it all the time on the > > > > cli. (add the name after the region name, separated by : just like > > > > in the cli) > > > > > > > > > > +1. It's pretty silly to need to specify a ROM-sized file for "-w" just > > to > > > update a single region. > > > > That needs to be dealt otherwise. > > > Darn, I was hoping to rope you into doing work for me ;-) > > What i meant is that you specify in > > the layout file the image you would normally specify by -i > > regionname:filename. > > > > if there is no file name in the cli, look in the layout file. > > > So far so good. > > > > if there > > is none there either, use the one of the complete image (for the -w > > case, -r would be slightly different). > > > > So the behavior is the same as it is now? That is, the -w argument is the > complete ROM-sized file and flashrom will use the layout info to seek to > the correct offset in the file? It is the same regarding the rom-sized file... because that has absolutely nothing to do with the layout file per se. That solely depends on the CLI which is free to prepare the image otherwise (if that's not possible then it should be eventually). You talked about -w (etc.) no longer requiring an argument... that's where I would put the focus to solve the problem of requiring rom-sized images. If I'll beat you to implement that, you will notice it ;) -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

1 0

Re: [flashrom] [RFC] Layout *file* handling...
by Stefan Tauner 11 Jul '13

11 Jul '13

On Thu, 11 Jul 2013 12:31:55 -0700 David Hendricks <dhendrix(a)google.com> wrote: > > - Default image names. For example one might want to always use me.bin > > for the ME region without the need to specify it all the time on the > > cli. (add the name after the region name, separated by : just like > > in the cli) > > > > +1. It's pretty silly to need to specify a ROM-sized file for "-w" just to > update a single region. That needs to be dealt otherwise. What i meant is that you specify in the layout file the image you would normally specify by -i regionname:filename. if there is no file name in the cli, look in the layout file. if there is none there either, use the one of the complete image (for the -w case, -r would be slightly different). -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

2 1

[commit] r1694 - trunk
by repository service 11 Jul '13

11 Jul '13

Author: stefanct Date: Thu Jul 11 15:48:52 2013 New Revision: 1694 URL: http://flashrom.org/trac/flashrom/changeset/1694 Log: print_wiki: Fixes, explanation re. flash chips, links to footnotes. - Escape mainboard notes in the wiki output with <nowiki> (this became necessary because MediaWiki started to convert <mark(a)zl2tod.net> to ). - Add explanations at the top of the flash chip table in the wiki. - Add links to the footnotes of mainboards and laptops similarly to what MediaWiki does: link from the reference to the footnote and vice versa too. Also, check for truncations of the footnotes due to the limited buffer. - Fix a typo introduced in r1555. Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> Acked-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> Modified: trunk/print_wiki.c Modified: trunk/print_wiki.c ============================================================================== --- trunk/print_wiki.c Fri Jun 28 23:29:51 2013 (r1693) +++ trunk/print_wiki.c Thu Jul 11 15:48:52 2013 (r1694) @@ -76,6 +76,18 @@ | Probe\n| Read\n| Erase\n| Write\n\ | align=\"center\" | Min \n| align=\"center\" | Max\n\n"; +static const char chip_intro[] = "\ +\n== Supported flash chips ==\n\n\ +The list below contains all chips that have some kind of explicit support added to flashrom and their last \ +known test status. Newer SPI flash chips might work even without explicit support if they implement SFDP ([\ +http://www.jedec.org/standards-documents/docs/jesd216 Serial Flash Discoverable Parameters - JESD216]). \ +Flashrom will detect this automatically and inform you about it.\n\n\ +The names used below are designed to be as concise as possible and hence contain only the characters \ +describing properties that are relevant to flashrom. Irrelevant characters specify attributes flashrom can not \ +use or even detect by itself (e.g. the physical package) and have no effect on flashrom's operation. They are \ +replaced by dots ('.') functioning as wildcards (like in Regular Expressions) or are completely omitted at the \ +end of a name.\n"; + static const char programmer_th[] = "\ ! align=\"left\" | Programmer\n\ ! align=\"left\" | Vendor\n\ @@ -159,7 +171,7 @@ /* +1 to force the resulting number of columns to be < cols */ lines_per_col = boardcount / cols + ((boardcount%cols) > 0 ? 1 : 0); - printf("\n\nTotal amount of known good boards %s: '''%d'''; " + printf("\n\nTotal amount of known good %s: '''%d'''; " "Untested (e.g. user vanished before testing new code): '''%d'''; " "Not yet supported (i.e. known-bad): '''%d'''.\n\n" "{| border=\"0\" valign=\"top\"\n", devicetype, boardcount_good, boardcount_nt, boardcount_bad); @@ -194,9 +206,17 @@ (boards[i].working == NT) ? "?3" : "No"); if (boards[i].note) { - printf("%d\n", num_notes + 1); - snprintf(tmp, sizeof(tmp), "%d %s \n", - 1 + num_notes++, boards[i].note); + num_notes++; + printf(" [[#%s_note%d|%d]]\n", + devicetype, num_notes, devicetype, num_notes, num_notes); + int ret = snprintf(tmp, sizeof(tmp), + "%d. [[#%s_ref%d|↑]]" + " <nowiki>%s</nowiki> \n", devicetype, num_notes, num_notes, + devicetype, num_notes, boards[i].note); + if (ret < 0 || ret >= sizeof(tmp)) { + fprintf(stderr, "Footnote text #%d of %s truncated (ret=%d, sizeof(tmp)=%zu)\n", + num_notes, devicetype, ret, sizeof(tmp)); + } notes = strcat_realloc(notes, tmp); } else { printf("\n"); @@ -248,7 +268,8 @@ /* +1 to force the resulting number of columns to be < cols */ lines_per_col = chipcount / cols + ((chipcount%cols) > 0 ? 1 : 0); - printf("\n== Supported chips ==\n\nTotal amount of supported chips: '''%d'''\n\n" + printf("%s", chip_intro); + printf("\nTotal amount of supported chips: '''%d'''\n\n" "{| border=\"0\" valign=\"top\"\n", chipcount); for (f = flashchips; f->name != NULL; f++) {

1 0

Re: [flashrom] WG: SB600, shutting down IMC failed
by Stefan Tauner 11 Jul '13

11 Jul '13

On Wed, 10 Jul 2013 13:00:43 +0200 Rudolf Marek <r.marek(a)assembler.cz> wrote: > Please can you provide the debug output too (i think it is -VVV)? I'm > CCing Stefan he did this patch. > > > I guess it might be that IMC runs but firmware does not ack the command. > This may happen if IMC strap is enabled but no firmware is in BIOS, or > if firmware crashed. > > Lets see what is in the debug output. So... you got mail from Markus too... :) Markus: that's exactly why I said I would prefer the mailing list. Let's continue there. Markus: you should be able to just cc to the mailing list even if you are not subscribed (someone has to moderate this once) and I have put it into CC in this mail. Markus sent me this log of what seems to be vanilla r1563 + patch 3967: > flashrom v0.9.6.1-r1563 on Linux 3.0.1pae (i686) > flashrom was built with libpci 3.1.7, GCC 4.4.5, little endian > Command line (4 args): flashrom -p internal:amc_imc_force=yes -o log.txt > Calibrating delay loop... OS timer resolution is 1 usecs, 824M loops per second, 10 myus = 10 us, 100 myus = 101 us, 1000 myus = 999 us, 10000 myus = 10003 us, 4 myus = 5 us, OK. > Initializing internal programmer > No coreboot table found. > DMI string system-manufacturer: "To be filled by O.E.M." > DMI string system-product-name: "To be filled by O.E.M." > DMI string system-version: "To be filled by O.E.M." > DMI string baseboard-manufacturer: "abc" > DMI string baseboard-product-name: "test" > DMI string baseboard-version: "1.1.0" > DMI string chassis-type: "Desktop" > Found chipset "AMD SB7x0/SB8x0/SB9x0" with PCI ID 1002:439d. Enabling flash write... IMC SIO is at 0x6e. > IMC MBOX is at 0x3e. > Shutting down IMC failed. > FAILED! > FATAL ERROR! > Unhandled programmer parameters: amc_imc_force=yes > Error: Programmer initialization failed. Apart from the obvious typo this means that mbox_wait_ack returned a positive value which happens only in the case of a timeout... and in that case we should see "IMC MBOX: Timeout!" i the log but we don't! Then we talked about applying Rudolf's patch to see if my refinements broke something. > ich möchte nur kurz sicherstellen, dass ich auch die richtigen Ausgangs-Sourcen verwende: > flashrom_0.9.6.1+r1563.orig.tar.gz. Ist dieses Paket korrekt? (in der Beschreibung von Patch 3894 war einmal die Rede von r1564. > > Diese Sourcen (ohne eine andere Änderung) habe ich mit patch 3894 "patch < AMD---SP5100---take-SPI-ownership-1-2.patch modifiziert. > > Folgende Fehlermeldung erscheint beim Kompilieren mit make: > chipset_enable.o: In function "enable_flash_sb600": > chipset_enable.c:(.text+0x13e8) : undefined reference to "imc_try_shutdown" > collect2: ld returned 1 exit status Either patching or compiling did not work as intended. I have tried to reproduce the problem with my git tree reset to "Increase flashrom release number to 0.9.6.1" aka r1563 and apply patch 3894. This worked fine (more or less): patch < AMD---SP5100---take-SPI-ownership-1-2.patch patching file Makefile Hunk #1 succeeded at 379 (offset -38 lines). patching file imc.c patching file chipset_enable.c Hunk #1 succeeded at 839 (offset -83 lines). patching file sb600spi.c patching file programmer.h Hunk #1 succeeded at 563 with fuzz 2 (offset 4 lines). But r1563 does not contain msg_pwarn which was introduced in r1643, so I really don't think it makes much sense to continue there anyway. But in any case your result should be different to what you got... > Habe nun die flashrom_0,9.6.1+r1564-Sourcen gefunden und diese mit patch 3894 modifiziert: => keine Fehlermeldung beim Patchen. Kompilieren klappte ebenfalls. I don't think that to be honest, see above. Applying patch 3894 to r1643 and compilation worked fine, maybe you wanna try that. But I think Rudolf might be able to tell us something from the log at the top. In any case please do *always* use at least -VV in all flashrom invocations so that we get useful output! -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

3 6

[PATCH 7/7] Get rid of sp_die().
by Stefan Tauner 10 Jul '13

10 Jul '13

- Export msg_perr_strerror() - Add return values to sp_flush_stream(), sp_pass_writen(), sp_execute_opbuf(), sp_execute_opbuf_noflush(), sp_check_opbuf_usage(), sp_do_read_n(). - Use those return values to propagate errors instead of exiting. In some places this has to wait for core API changes (error handling for chip_readb, chip_readn, chip_write) hence comments are added instead. Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> --- programmer.h | 3 +- serial.c | 8 +-- serprog.c | 155 ++++++++++++++++++++++++++++++++++++++--------------------- 3 files changed, 102 insertions(+), 64 deletions(-) diff --git a/programmer.h b/programmer.h index 4f590ab..b50afd5 100644 --- a/programmer.h +++ b/programmer.h @@ -659,10 +659,11 @@ typedef int fdtype; #define SER_INV_FD -1 #endif +void msg_perr_strerror(const char *msg); + void sp_flush_incoming(void); fdtype sp_openserport(char *dev, unsigned int baud); int serialport_config(fdtype fd, unsigned int baud); -void __attribute__((noreturn)) sp_die(char *msg); extern fdtype sp_fd; /* expose serialport_shutdown as it's currently used by buspirate */ int serialport_shutdown(void *data); diff --git a/serial.c b/serial.c index 0504aaa..ba886ad 100644 --- a/serial.c +++ b/serial.c @@ -43,12 +43,6 @@ fdtype sp_fd = SER_INV_FD; -void __attribute__((noreturn)) sp_die(char *msg) -{ - perror(msg); - exit(1); -} - #ifdef _WIN32 struct baudentry { DWORD flag; @@ -142,7 +136,7 @@ const struct baudentry *round_baud(unsigned int baud) /* Uses msg_perr to print the last system error. * Prints "Error: " followed first by \c msg and then by the description of the last error retrieved via * strerror() or FormatMessage() and ending with a linebreak. */ -static void msg_perr_strerror(const char *msg) +void msg_perr_strerror(const char *msg) { msg_perr("Error: %s", msg); #ifdef _WIN32 diff --git a/serprog.c b/serprog.c index 26b580c..439d821 100644 --- a/serprog.c +++ b/serprog.c @@ -237,43 +237,58 @@ static int sp_docommand(uint8_t command, uint32_t parmlen, return 0; } -static void sp_flush_stream(void) +static int sp_flush_stream(void) { if (sp_streamed_transmit_ops) do { unsigned char c; if (serialport_read(&c, 1) != 0) { - sp_die("Error: cannot read from device (flushing stream)"); + msg_perr("Error: cannot read from device (flushing stream)"); + return 1; } if (c == S_NAK) { msg_perr("Error: NAK to a stream buffer operation\n"); - exit(1); + return 1; } if (c != S_ACK) { msg_perr("Error: Invalid reply 0x%02X from device\n", c); - exit(1); + return 1; } } while (--sp_streamed_transmit_ops); sp_streamed_transmit_ops = 0; sp_streamed_transmit_bytes = 0; + return 0; } -static int sp_stream_buffer_op(uint8_t cmd, uint32_t parmlen, uint8_t * parms) +static int sp_stream_buffer_op(uint8_t cmd, uint32_t parmlen, uint8_t *parms) { uint8_t *sp; if (sp_automatic_cmdcheck(cmd)) return 1; + sp = malloc(1 + parmlen); - if (!sp) sp_die("Error: cannot malloc command buffer"); + if (!sp) { + msg_perr("Error: cannot malloc command buffer\n"); + return 1; + } sp[0] = cmd; memcpy(&(sp[1]), parms, parmlen); - if (sp_streamed_transmit_bytes >= (1 + parmlen + sp_device_serbuf_size)) - sp_flush_stream(); - if (serialport_write(sp, 1 + parmlen) != 0) - sp_die("Error: cannot write command"); - free(sp); + + if (sp_streamed_transmit_bytes >= (1 + parmlen + sp_device_serbuf_size)) { + if (sp_flush_stream() != 0) { + free(sp); + return 1; + } + } + if (serialport_write(sp, 1 + parmlen) != 0) { + msg_perr("Error: cannot write command\n"); + free(sp); + return 1; + } sp_streamed_transmit_ops += 1; sp_streamed_transmit_bytes += 1 + parmlen; + + free(sp); return 0; } @@ -656,16 +671,16 @@ int serprog_init(void) return 0; } -/* Move an in flashrom buffer existing write-n operation to * - * the on-device operation buffer. */ -static void sp_pass_writen(void) +/* Move an in flashrom buffer existing write-n operation to the on-device operation buffer. */ +static int sp_pass_writen(void) { unsigned char header[7]; - msg_pspew(MSGHEADER "Passing write-n bytes=%d addr=0x%x\n", - sp_write_n_bytes, sp_write_n_addr); - if (sp_streamed_transmit_bytes >= - (7 + sp_write_n_bytes + sp_device_serbuf_size)) - sp_flush_stream(); + msg_pspew(MSGHEADER "Passing write-n bytes=%d addr=0x%x\n", sp_write_n_bytes, sp_write_n_addr); + if (sp_streamed_transmit_bytes >= (7 + sp_write_n_bytes + sp_device_serbuf_size)) { + if (sp_flush_stream() != 0) { + return 1; + } + } /* In case it's just a single byte send it as a single write. */ if (sp_write_n_bytes == 1) { sp_write_n_bytes = 0; @@ -673,9 +688,10 @@ static void sp_pass_writen(void) header[1] = (sp_write_n_addr >> 8) & 0xFF; header[2] = (sp_write_n_addr >> 16) & 0xFF; header[3] = sp_write_n_buf[0]; - sp_stream_buffer_op(S_CMD_O_WRITEB, 4, header); + if (sp_stream_buffer_op(S_CMD_O_WRITEB, 4, header) != 0) + return 1; sp_opbuf_usage += 5; - return; + return 0; } header[0] = S_CMD_O_WRITEN; header[1] = (sp_write_n_bytes >> 0) & 0xFF; @@ -684,39 +700,55 @@ static void sp_pass_writen(void) header[4] = (sp_write_n_addr >> 0) & 0xFF; header[5] = (sp_write_n_addr >> 8) & 0xFF; header[6] = (sp_write_n_addr >> 16) & 0xFF; - if (serialport_write(header, 7) != 0) - sp_die("Error: cannot write write-n command\n"); - if (serialport_write(sp_write_n_buf, sp_write_n_bytes) != 0) - sp_die("Error: cannot write write-n data"); + if (serialport_write(header, 7) != 0) { + msg_perr(MSGHEADER "Error: cannot write write-n command\n"); + return 1; + } + if (serialport_write(sp_write_n_buf, sp_write_n_bytes) != 0) { + msg_perr(MSGHEADER "Error: cannot write write-n data"); + return 1; + } sp_streamed_transmit_bytes += 7 + sp_write_n_bytes; sp_streamed_transmit_ops += 1; sp_opbuf_usage += 7 + sp_write_n_bytes; sp_write_n_bytes = 0; sp_prev_was_write = 0; + return 0; } -static void sp_execute_opbuf_noflush(void) +static int sp_execute_opbuf_noflush(void) { - if ((sp_max_write_n) && (sp_write_n_bytes)) - sp_pass_writen(); - sp_stream_buffer_op(S_CMD_O_EXEC, 0, NULL); - msg_pspew(MSGHEADER "Executed operation buffer of %d bytes\n", - sp_opbuf_usage); + if ((sp_max_write_n) && (sp_write_n_bytes)) { + if (sp_pass_writen() != 0) { + msg_perr("Error: could not transfer write buffer\n"); + return 1; + } + } + if (sp_stream_buffer_op(S_CMD_O_EXEC, 0, NULL) != 0) { + msg_perr("Error: could not execute command buffer\n"); + return 1; + } + msg_pspew(MSGHEADER "Executed operation buffer of %d bytes\n", sp_opbuf_usage); sp_opbuf_usage = 0; sp_prev_was_write = 0; - return; + return 0; } -static void sp_execute_opbuf(void) +static int sp_execute_opbuf(void) { - sp_execute_opbuf_noflush(); - sp_flush_stream(); + if (sp_execute_opbuf_noflush() != 0) + return 1; + if (sp_flush_stream() != 0) + return 1; + + return 0; } static int serprog_shutdown(void *data) { if ((sp_opbuf_usage) || (sp_max_write_n && sp_write_n_bytes)) - sp_execute_opbuf(); + if (sp_execute_opbuf() != 0) + msg_pwarn("Could not flush command buffer.\n"); if (sp_check_commandavail(S_CMD_S_PIN_STATE)) { uint8_t dis = 0; if (sp_docommand(S_CMD_S_PIN_STATE, 1, &dis, 0, NULL) == 0) @@ -731,14 +763,15 @@ static int serprog_shutdown(void *data) return 0; } -static void sp_check_opbuf_usage(int bytes_to_be_added) +static int sp_check_opbuf_usage(int bytes_to_be_added) { if (sp_device_opbuf_size <= (sp_opbuf_usage + bytes_to_be_added)) { - sp_execute_opbuf(); - /* If this happens in the mid of an page load the page load * - * will probably fail. */ - msg_pdbg(MSGHEADER "Warning: executed operation buffer due to size reasons\n"); + /* If this happens in the middle of a page load the page load will probably fail. */ + msg_pwarn(MSGHEADER "Warning: executed operation buffer due to size reasons\n"); + if (sp_execute_opbuf() != 0) + return 1; } + return 0; } static void serprog_chip_writeb(const struct flashctx *flash, uint8_t val, @@ -768,7 +801,7 @@ static void serprog_chip_writeb(const struct flashctx *flash, uint8_t val, writeb_parm[1] = (addr >> 8) & 0xFF; writeb_parm[2] = (addr >> 16) & 0xFF; writeb_parm[3] = val; - sp_stream_buffer_op(S_CMD_O_WRITEB, 4, writeb_parm); + sp_stream_buffer_op(S_CMD_O_WRITEB, 4, writeb_parm); // FIXME: return error sp_opbuf_usage += 5; } } @@ -785,16 +818,16 @@ static uint8_t serprog_chip_readb(const struct flashctx *flash, buf[0] = ((addr >> 0) & 0xFF); buf[1] = ((addr >> 8) & 0xFF); buf[2] = ((addr >> 16) & 0xFF); - sp_stream_buffer_op(S_CMD_R_BYTE, 3, buf); - sp_flush_stream(); + sp_stream_buffer_op(S_CMD_R_BYTE, 3, buf); // FIXME: return error + sp_flush_stream(); // FIXME: return error if (serialport_read(&c, 1) != 0) - sp_die("readb byteread"); + msg_perr(MSGHEADER "readb byteread"); // FIXME: return error msg_pspew("%s addr=0x%" PRIxPTR " returning 0x%02X\n", __func__, addr, c); return c; } /* Local version that really does the job, doesn't care of max_read_n. */ -static void sp_do_read_n(uint8_t * buf, const chipaddr addr, size_t len) +static int sp_do_read_n(uint8_t * buf, const chipaddr addr, size_t len) { unsigned char sbuf[6]; msg_pspew("%s: addr=0x%" PRIxPTR " len=%lu\n", __func__, addr, (unsigned long)len); @@ -808,10 +841,13 @@ static void sp_do_read_n(uint8_t * buf, const chipaddr addr, size_t len) sbuf[4] = ((len >> 8) & 0xFF); sbuf[5] = ((len >> 16) & 0xFF); sp_stream_buffer_op(S_CMD_R_NBYTES, 6, sbuf); - sp_flush_stream(); - if (serialport_read(buf, len) != 0) - sp_die("Error: cannot read read-n data"); - return; + if (sp_flush_stream() != 0) + return 1; + if (serialport_read(buf, len) != 0) { + msg_perr(MSGHEADER "Error: cannot read read-n data"); + return 1; + } + return 0; } /* The externally called version that makes sure that max_read_n is obeyed. */ @@ -821,12 +857,12 @@ static void serprog_chip_readn(const struct flashctx *flash, uint8_t * buf, size_t lenm = len; chipaddr addrm = addr; while ((sp_max_read_n != 0) && (lenm > sp_max_read_n)) { - sp_do_read_n(&(buf[addrm-addr]), addrm, sp_max_read_n); + sp_do_read_n(&(buf[addrm-addr]), addrm, sp_max_read_n); // FIXME: return error addrm += sp_max_read_n; lenm -= sp_max_read_n; } if (lenm) - sp_do_read_n(&(buf[addrm-addr]), addrm, lenm); + sp_do_read_n(&(buf[addrm-addr]), addrm, lenm); // FIXME: return error } void serprog_delay(int usecs) @@ -858,11 +894,18 @@ static int serprog_spi_send_command(struct flashctx *flash, unsigned char *parmbuf; int ret; msg_pspew("%s, writecnt=%i, readcnt=%i\n", __func__, writecnt, readcnt); - if ((sp_opbuf_usage) || (sp_max_write_n && sp_write_n_bytes)) - sp_execute_opbuf(); + if ((sp_opbuf_usage) || (sp_max_write_n && sp_write_n_bytes)) { + if (sp_execute_opbuf() != 0) { + msg_perr("Error: could not execute command buffer before sending SPI commands.\n"); + return 1; + } + } + parmbuf = malloc(writecnt + 6); - if (!parmbuf) - sp_die("Error: cannot malloc SPI send param buffer"); + if (!parmbuf) { + msg_perr_strerror("cannot malloc SPI send param buffer"); + return 1; + } parmbuf[0] = (writecnt >> 0) & 0xFF; parmbuf[1] = (writecnt >> 8) & 0xFF; parmbuf[2] = (writecnt >> 16) & 0xFF; -- Kind regards, Stefan Tauner

1 0

[PATCH 6/7] Introduce serialport_config().
by Stefan Tauner 10 Jul '13

10 Jul '13

This allows to easily reconfigure a serial port as needed in the Bus Pirate speedup. Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> --- programmer.h | 1 + serial.c | 105 +++++++++++++++++++++++++++++++++-------------------------- 2 files changed, 60 insertions(+), 46 deletions(-) diff --git a/programmer.h b/programmer.h index b32b85f..4f590ab 100644 --- a/programmer.h +++ b/programmer.h @@ -661,6 +661,7 @@ typedef int fdtype; void sp_flush_incoming(void); fdtype sp_openserport(char *dev, unsigned int baud); +int serialport_config(fdtype fd, unsigned int baud); void __attribute__((noreturn)) sp_die(char *msg); extern fdtype sp_fd; /* expose serialport_shutdown as it's currently used by buspirate */ diff --git a/serial.c b/serial.c index 5f74fd1..0504aaa 100644 --- a/serial.c +++ b/serial.c @@ -160,35 +160,18 @@ static void msg_perr_strerror(const char *msg) #endif } -fdtype sp_openserport(char *dev, unsigned int baud) +int serialport_config(fdtype fd, unsigned int baud) { -#ifdef _WIN32 - HANDLE fd; - char *dev2 = dev; - if ((strlen(dev) > 3) && - (tolower((unsigned char)dev[0]) == 'c') && - (tolower((unsigned char)dev[1]) == 'o') && - (tolower((unsigned char)dev[2]) == 'm')) { - dev2 = malloc(strlen(dev) + 5); - if (!dev2) { - msg_perr_strerror("Out of memory: "); - return SER_INV_FD; - } - strcpy(dev2, "\\\\.\\"); - strcpy(dev2 + 4, dev); - } - fd = CreateFile(dev2, GENERIC_READ | GENERIC_WRITE, 0, NULL, - OPEN_EXISTING, 0, NULL); - if (dev2 != dev) - free(dev2); - if (fd == INVALID_HANDLE_VALUE) { - msg_perr_strerror("Cannot open serial port: "); - return SER_INV_FD; + if (fd == SER_INV_FD) { + msg_perr("%s: File descriptor is invalid.\n", __func__); + return 1; } + +#ifdef _WIN32 DCB dcb; if (!GetCommState(fd, &dcb)) { msg_perr_strerror("Could not fetch original serial port configuration: "); - goto out_close; + return 1; } const struct baudentry *entry = round_baud(baud); dcb.BaudRate = entry->flag; @@ -197,35 +180,25 @@ fdtype sp_openserport(char *dev, unsigned int baud) dcb.StopBits = ONESTOPBIT; if (!SetCommState(fd, &dcb)) { msg_perr_strerror("Could not change serial port configuration: "); - goto out_close; + return 1; } if (!GetCommState(fd, &dcb)) { msg_perr_strerror("Could not fetch new serial port configuration: "); - goto out_close; + return 1; } msg_pdbg("Baud rate is %ld.\n", dcb.BaudRate); - return fd; -out_close: - CloseHandle(sp_fd); - return SER_INV_FD; #else struct termios wanted, observed; - int fd; - fd = open(dev, O_RDWR | O_NOCTTY | O_NDELAY); - if (fd < 0) { - msg_perr_strerror("Cannot open serial port: "); - return SER_INV_FD; - } fcntl(fd, F_SETFL, 0); if (tcgetattr(fd, &observed) != 0) { msg_perr_strerror("Could not fetch original serial port configuration: "); - goto out_close; + return 1; } wanted = observed; const struct baudentry *entry = round_baud(baud); if (cfsetispeed(&wanted, entry->flag) != 0 || cfsetospeed(&wanted, entry->flag) != 0) { msg_perr_strerror("Could not set serial baud rate: "); - goto out_close; + return 1; } wanted.c_cflag &= ~(PARENB | CSTOPB | CSIZE | CRTSCTS); wanted.c_cflag |= (CS8 | CLOCAL | CREAD); @@ -234,11 +207,11 @@ out_close: wanted.c_oflag &= ~OPOST; if (tcsetattr(fd, TCSANOW, &wanted) != 0) { msg_perr_strerror("Could not change serial port configuration: "); - goto out_close; + return 1; } if (tcgetattr(fd, &observed) != 0) { msg_perr_strerror("Could not fetch new serial port configuration: "); - goto out_close; + return 1; } if (observed.c_cflag != wanted.c_cflag || observed.c_lflag != wanted.c_lflag || @@ -246,14 +219,54 @@ out_close: observed.c_oflag != wanted.c_oflag || cfgetispeed(&observed) != cfgetispeed(&wanted)) { msg_perr("%s: Some requested options did not stick.\n", __func__); - goto out_close; + return 1; } - msg_pdbg("Baud rate is %d.\n", entry->baud); - return fd; + msg_pdbg("Baud rate is %d now.\n", entry->baud); +#endif + return 0; +} -out_close: - close(sp_fd); - return SER_INV_FD; +fdtype sp_openserport(char *dev, unsigned int baud) +{ + fdtype fd; +#ifdef _WIN32 + char *dev2 = dev; + if ((strlen(dev) > 3) && + (tolower((unsigned char)dev[0]) == 'c') && + (tolower((unsigned char)dev[1]) == 'o') && + (tolower((unsigned char)dev[2]) == 'm')) { + dev2 = malloc(strlen(dev) + 5); + if (!dev2) { + msg_perr_strerror("Out of memory: "); + return SER_INV_FD; + } + strcpy(dev2, "\\\\.\\"); + strcpy(dev2 + 4, dev); + } + fd = CreateFile(dev2, GENERIC_READ | GENERIC_WRITE, 0, NULL, + OPEN_EXISTING, 0, NULL); + if (dev2 != dev) + free(dev2); + if (fd == INVALID_HANDLE_VALUE) { + msg_perr_strerror("Cannot open serial port: "); + return SER_INV_FD; + } + if (serialport_config(fd, baud) != 0) { + CloseHandle(fd); + return SER_INV_FD; + } + return fd; +#else + fd = open(dev, O_RDWR | O_NOCTTY | O_NDELAY); + if (fd < 0) { + msg_perr_strerror("Cannot open serial port: "); + return SER_INV_FD; + } + if (serialport_config(fd, baud) != 0) { + close(fd); + return SER_INV_FD; + } + return fd; #endif } -- Kind regards, Stefan Tauner

1 0

[PATCH 5/7] Combine serialport_write and serialport_read into serialport_rw.
by Stefan Tauner 10 Jul '13

10 Jul '13

Because they share almost all code, combine them into serialport_rw and export the old functions as wrappers. This makes reads "empty reads"- aware and programmers able to detect such problems even on "blocking" reads. Also, this fixes error detection on Windows. Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> --- serial.c | 83 ++++++++++++++++++++++++++++++---------------------------------- 1 file changed, 39 insertions(+), 44 deletions(-) diff --git a/serial.c b/serial.c index 1b394cd..5f74fd1 100644 --- a/serial.c +++ b/serial.c @@ -3,6 +3,7 @@ * * Copyright (C) 2009 Urja Rannikko <urjaman(a)gmail.com> * Copyright (C) 2009,2010 Carl-Daniel Hailfinger + * Copyright (C) 2013 Stefan Tauner * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -28,6 +29,7 @@ #include <sys/stat.h> #include <errno.h> #include <inttypes.h> +#include <stdbool.h> #ifdef _WIN32 #include <conio.h> #else @@ -328,66 +330,59 @@ int serialport_shutdown(void *data) return 0; } -int serialport_write(unsigned char *buf, unsigned int writecnt) +#define EMPTY_TRIES 10 +/* Writes (if \c outdir is true) or reads \c todo bytes from/into \c buf. + * Tries up to \c EMPTY_TRIES times with a timeout of \c ms ms between each try. */ +static int serialport_rw(unsigned char *buf, unsigned int todo, bool outdir, unsigned int ms) { -#ifdef _WIN32 - DWORD tmp = 0; -#else - ssize_t tmp = 0; -#endif - unsigned int empty_writes = 250; /* results in a ca. 125ms timeout */ + const char * const op = outdir ? "write" : "read"; + bool err = false; + unsigned int empty_tries = EMPTY_TRIES; - while (writecnt > 0) { + while (todo > 0) { #ifdef _WIN32 - WriteFile(sp_fd, buf, writecnt, &tmp, NULL); + DWORD cur = 0; + if (outdir) + err = !WriteFile(sp_fd, buf, todo, &cur, NULL); + else + err = !ReadFile(sp_fd, buf, todo, &cur, NULL); #else - tmp = write(sp_fd, buf, writecnt); -#endif - if (tmp == -1) { - msg_perr("Serial port write error!\n"); + ssize_t cur = 0; + if (outdir) + err = (cur = write(sp_fd, buf, todo)) < 0; + else + err = (cur = read(sp_fd, buf, todo)) < 0; +#endif + if (err) { + msg_perr("Serial port %s error!\n", op); return 1; } - if (!tmp) { - msg_pdbg2("Empty write\n"); - empty_writes--; - programmer_delay(500); - if (empty_writes == 0) { - msg_perr("Serial port is unresponsive!\n"); + if (cur == 0) { + msg_pdbg2("Empty %s.\n", op); + empty_tries--; + programmer_delay(ms * 1000); + if (empty_tries == 0) { + msg_perr("Serial port is unresponsive or disappeared!\n"); return 1; } + continue; } - writecnt -= tmp; - buf += tmp; + todo -= cur; + buf += cur; + empty_tries = EMPTY_TRIES; } return 0; } -int serialport_read(unsigned char *buf, unsigned int readcnt) +int serialport_write(unsigned char *buf, unsigned int writecnt) { -#ifdef _WIN32 - DWORD tmp = 0; -#else - ssize_t tmp = 0; -#endif - - while (readcnt > 0) { -#ifdef _WIN32 - ReadFile(sp_fd, buf, readcnt, &tmp, NULL); -#else - tmp = read(sp_fd, buf, readcnt); -#endif - if (tmp == -1) { - msg_perr("Serial port read error!\n"); - return 1; - } - if (!tmp) - msg_pdbg2("Empty read\n"); - readcnt -= tmp; - buf += tmp; - } + return serialport_rw(buf, writecnt, true, 1); +} - return 0; +int serialport_read(unsigned char *buf, unsigned int readcnt) +{ + return serialport_rw(buf, readcnt, false, 10); } /* Tries up to timeout ms to read readcnt characters and places them into the array starting at c. Returns -- Kind regards, Stefan Tauner

1 0

[PATCH 4/7] Add an internal DMI decoder
by Stefan Tauner 10 Jul '13

10 Jul '13

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 setting DMI_INTERNAL to 0. 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. 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). Signed-off-by: Sean Nelson<audiohacked(a)gmail.com> Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> Tested-by: Maciej Pijanka <maciej.pijanka(a)gmail.com> Tested-by: Idwer Vollering <vidwer(a)gmail.com> --- dmi.c | 299 ++++++++++++++++++++++++++++++++++++++++++++++++----------- physmap.c | 5 + programmer.h | 1 + 3 files changed, 253 insertions(+), 52 deletions(-) diff --git a/dmi.c b/dmi.c index 242889f..47b5795 100644 --- a/dmi.c +++ b/dmi.c @@ -1,7 +1,10 @@ /* * This file is part of the flashrom project. * + * Copyright (C) 2000-2002 Alan Cox <alan(a)redhat.com> + * Copyright (C) 2002-2010 Jean Delvare <khali(a)linux-fr.org> * Copyright (C) 2009,2010 Michael Karcher + * Copyright (C) 2011-2013 Stefan Tauner * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -26,33 +29,28 @@ #include "flash.h" #include "programmer.h" -int has_dmi_support = 0; - -#if STANDALONE - -/* Stub to indicate missing DMI functionality. - * has_dmi_support is 0 by default, so nothing to do here. - * Because dmidecode is not available on all systems, the goal is to implement - * the DMI subset we need directly in this file. - */ -void dmi_init(void) -{ -} +/* Select internal DMI decoder instead of relying on dmidecode */ +#define DMI_INTERNAL 1 +/* Enable SMBIOS decoding. Currently legacy DMI decoding is enough. */ +#define SM_SUPPORT 0 -int dmi_match(const char *pattern) -{ - return 0; -} +/* Strings longer than 4096 in DMI are just insane. */ +#define DMI_MAX_ANSWER_LEN 4096 -#else /* STANDALONE */ +int has_dmi_support = 0; -static const char *dmidecode_names[] = { - "system-manufacturer", - "system-product-name", - "system-version", - "baseboard-manufacturer", - "baseboard-product-name", - "baseboard-version", +static struct { + const char *const keyword; + const uint8_t type; + const uint8_t offset; + char *value; +} dmi_strings[] = { + { "system-manufacturer", 1, 0x04 }, + { "system-product-name", 1, 0x05 }, + { "system-version", 1, 0x06 }, + { "baseboard-manufacturer", 2, 0x04 }, + { "baseboard-product-name", 2, 0x05 }, + { "baseboard-version", 2, 0x06 }, }; /* This list is used to identify supposed laptops. The is_laptop field has the @@ -66,9 +64,9 @@ static const char *dmidecode_names[] = { * The types below are the most common ones. */ static const struct { - unsigned char type; - unsigned char is_laptop; - const char *name; + uint8_t type; + uint8_t is_laptop; + char *name; } dmi_chassis_types[] = { {0x01, 2, "Other"}, {0x02, 2, "Unknown"}, @@ -86,20 +84,199 @@ static const struct { {0x18, 0, "Sealed-case PC"}, /* used by Supermicro (X8SIE) */ }; -#define DMI_COMMAND_LEN_MAX 260 -static const char *dmidecode_command = "dmidecode"; +#if DMI_INTERNAL +static int dmi_checksum(const uint8_t const *buf, size_t len) +{ + uint8_t sum = 0; + size_t a; -static char *dmistrings[ARRAY_SIZE(dmidecode_names)]; + for (a = 0; a < len; a++) + sum += buf[a]; + return (sum == 0); +} -/* Strings longer than 4096 in DMI are just insane. */ -#define DMI_MAX_ANSWER_LEN 4096 +static char *dmi_string(uint8_t *buf, uint8_t string_id, uint8_t *limit) +{ + size_t i, len; + + if (string_id == 0) + return "Not Specified"; + + while (string_id > 1 && string_id--) { + if (buf > limit) { + msg_perr("DMI table is broken (string portion out of bounds)!\n"); + return "<OUT OF BOUNDS>"; + } + buf += strnlen((char *)buf, limit - buf) + 1; + } + + if (!*buf) /* as long as the current byte we're on isn't null */ + return "<BAD INDEX>"; + + len = strnlen((char *)buf, limit - buf); + if (len > DMI_MAX_ANSWER_LEN) + len = DMI_MAX_ANSWER_LEN; + + /* fix junk bytes in the string */ + for (i = 0; i < len; i++) + if (buf[i] < 32 || buf[i] >= 127) + buf[i] = ' '; + + return (char *)buf; +} + +static void dmi_chassis_type(uint8_t code) +{ + int i; + code &= 0x7f; /* bits 6:0 are chassis type, 7th bit is the lock bit */ + is_laptop = 2; + for (i = 0; i < ARRAY_SIZE(dmi_chassis_types); i++) { + if (code == dmi_chassis_types[i].type) { + msg_pdbg("DMI string chassis-type: \"%s\"\n", dmi_chassis_types[i].name); + is_laptop = dmi_chassis_types[i].is_laptop; + break; + } + } +} + +static void dmi_table(uint32_t base, uint16_t len, uint16_t num) +{ + int i = 0, j = 0; + + uint8_t *dmi_table_mem = physmap_round("DMI Table", base, len); + //uint8_t *dmi_table_mem = rphysmap("DMI Table", base, len); + if (dmi_table_mem == NULL) { + msg_perr("Unable to access DMI Table\n"); + return; + } + + uint8_t *data = dmi_table_mem; + uint8_t *limit = dmi_table_mem + len; + + /* SMBIOS structure header is always 4 B long and contains: + * - uint8_t type; // see dmi_chassis_types's type + * - uint8_t length; // data section w/ header w/o strings + * - uint16_t handle; + */ + while (i < num && data + 4 <= limit) { + /* - If a short entry is found (less than 4 bytes), not only it + * is invalid, but we cannot reliably locate the next entry. + * - If the length value indicates that this structure spreads + * accross the table border, something is fishy too. + * Better stop at this point, and let the user know his/her + * table is broken. + */ + if (data[1] < 4 || data + data[1] > limit) { + msg_perr("DMI table is broken (bogus header)!\n"); + break; + } + + if(data[0] == 3) { + if (data + 5 <= limit) + dmi_chassis_type(data[5]); + /* else the table is broken, but laptop detection is optional, hence continue. */ + } else + for (j = 0; j < ARRAY_SIZE(dmi_strings); j++) { + uint8_t offset = dmi_strings[j].offset; + uint8_t type = dmi_strings[j].type; + + if (data[0] != type) + continue; + + if (data[1] <= offset || data+offset > limit) { + msg_perr("DMI table is broken (offset out of bounds)!\n"); + goto out; + } + + /* Table will be unmapped, hence fill the struct with duplicated strings. */ + dmi_strings[j].value = strdup(dmi_string(data + data[1], data[offset], limit)); + } + /* Find next structure by skipping data and string sections */ + data += data[1]; + while (data + 1 <= limit) { + if (data[0] == 0 && data[1] == 0) + break; + data++; + } + data += 2; + i++; + } +out: + physunmap(dmi_table_mem, len); + ; +} + +#if SM_SUPPORT +static int smbios_decode(uint8_t *buf) +{ + /* TODO: other checks mentioned in the conformance guidelines? */ + if (!dmi_checksum(buf, buf[0x05]) || + (memcmp(buf + 0x10, "_DMI_", 5) != 0) || + !dmi_checksum(buf + 0x10, 0x0F)) + return 0; + + dmi_table(mmio_readl(buf + 0x18), + mmio_readw(buf + 0x16), + mmio_readw(buf + 0x1C)); + + return 1; +} +#endif + +static int legacy_decode(uint8_t *buf) +{ + if (!dmi_checksum(buf, 0x0F)) + return 0; + + dmi_table(mmio_readl(buf + 0x08), mmio_readw(buf + 0x06), mmio_readw(buf + 0x0C)); + + return 1; +} + +int dmi_fill(void) +{ + size_t fp; + uint8_t *dmi_mem; + + msg_pdbg("Using Internal DMI decoder.\n"); + /* There are two ways specified to gain access to the SMBIOS table: + * - EFI's configuration table contains a pointer to the SMBIOS table. + * On linux it can be obtained from sysfs. EFI's SMBIOS GUID is: + * {0xeb9d2d31,0x2d88,0x11d3,0x9a,0x16,0x0,0x90,0x27,0x3f,0xc1,0x4d} + * - Scanning physical memory address range 0x000F0000h to 0x000FFFFF + * for the anchor-string(s). */ + dmi_mem = physmap_try_ro("DMI", 0xF0000, 0x10000); + if (!dmi_mem) + return -1; + + for (fp = 0; fp <= 0xFFF0; fp += 16) { +#if SM_SUPPORT + if (memcmp(dmi_mem + fp, "_SM_", 4) == 0 && fp <= 0xFFE0) { + if (smbios_decode(dmi_mem + fp)) // FIXME: length check + goto out; + } else +#endif + if (memcmp(dmi_mem + fp, "_DMI_", 5) == 0) + if (legacy_decode(dmi_mem + fp)) + goto out; + } + msg_pinfo("No DMI table found.\n"); +out: + physunmap(dmi_mem, 0x10000); + return 0; +} + +#else /* DMI_INTERNAL */ + +#define DMI_COMMAND_LEN_MAX 300 +static const char *dmidecode_command = "dmidecode"; static char *get_dmi_string(const char *string_name) { FILE *dmidecode_pipe; char *result; char answerbuf[DMI_MAX_ANSWER_LEN]; - char commandline[DMI_COMMAND_LEN_MAX + 40]; + char commandline[DMI_COMMAND_LEN_MAX]; snprintf(commandline, sizeof(commandline), "%s -s %s", dmidecode_command, string_name); @@ -138,7 +315,6 @@ static char *get_dmi_string(const char *string_name) /* Chomp trailing newline. */ if (answerbuf[0] != 0 && answerbuf[strlen(answerbuf) - 1] == '\n') answerbuf[strlen(answerbuf) - 1] = 0; - msg_pdbg("DMI string %s: \"%s\"\n", string_name, answerbuf); result = strdup(answerbuf); if (!result) @@ -150,34 +326,36 @@ static char *get_dmi_string(const char *string_name) static int dmi_shutdown(void *data) { int i; - for (i = 0; i < ARRAY_SIZE(dmistrings); i++) { - free(dmistrings[i]); - dmistrings[i] = NULL; + for (i = 0; i < ARRAY_SIZE(dmi_strings); i++) { + free(dmi_strings[i].value); + dmi_strings[i].value = NULL; } return 0; } -void dmi_init(void) +int dmi_fill(void) { int i; char *chassis_type; - if (register_shutdown(dmi_shutdown, NULL)) - return; - - has_dmi_support = 1; - for (i = 0; i < ARRAY_SIZE(dmidecode_names); i++) { - dmistrings[i] = get_dmi_string(dmidecode_names[i]); - if (!dmistrings[i]) { - has_dmi_support = 0; - return; + msg_pdbg("Using External DMI decoder.\n"); + for (i = 0; i < ARRAY_SIZE(dmi_strings); i++) { + dmi_strings[i].value = get_dmi_string(dmi_strings[i].keyword); + if (!dmi_strings[i].value) { + dmi_shutdown(NULL); + return 1; } } + if (register_shutdown(dmi_shutdown, NULL)) + return 1; + + chassis_type = get_dmi_string("chassis-type"); if (chassis_type == NULL) - return; + return 0; /* chassis-type handling is optional anyway */ + msg_pdbg("DMI string chassis-type: \"%s\"\n", chassis_type); is_laptop = 2; for (i = 0; i < ARRAY_SIZE(dmi_chassis_types); i++) { if (strcasecmp(chassis_type, dmi_chassis_types[i].name) == 0) { @@ -185,6 +363,17 @@ void dmi_init(void) break; } } + free(chassis_type); + return 0; +} + +#endif /* DMI_INTERNAL */ + +void dmi_init(void) +{ + /* dmi_fill fills the dmi_strings array, and if possible sets the global is_laptop variable. */ + if (dmi_fill()) + return; switch (is_laptop) { case 1: @@ -194,7 +383,14 @@ void dmi_init(void) msg_pdbg("DMI chassis-type is not specific enough.\n"); break; } - free(chassis_type); + + has_dmi_support = 1; + int i; + for (i = 0; i < ARRAY_SIZE(dmi_strings); i++) { + msg_pdbg("DMI string %s: \"%s\"\n", + dmi_strings[i].keyword, + dmi_strings[i].value); + } } /** @@ -252,11 +448,10 @@ int dmi_match(const char *pattern) if (!has_dmi_support) return 0; - for (i = 0; i < ARRAY_SIZE(dmidecode_names); i++) - if (dmi_compare(dmistrings[i], pattern)) + for (i = 0; i < ARRAY_SIZE(dmi_strings); i++) + if (dmi_compare(dmi_strings[i].value, pattern)) return 1; return 0; } -#endif /* STANDALONE */ diff --git a/physmap.c b/physmap.c index b248f92..93ce30b 100644 --- a/physmap.c +++ b/physmap.c @@ -332,6 +332,11 @@ void *rphysmap(const char *descr, uintptr_t phys_addr, size_t len) return physmap_common(descr, phys_addr, len, PHYSM_NOFAIL, PHYSM_RW, PHYSM_CLEANUP, PHYSM_ROUND); } +void *physmap_round(const char *descr, uintptr_t phys_addr, size_t len) +{ + return physmap_common(descr, phys_addr, len, PHYSM_NOFAIL, PHYSM_RW, PHYSM_NOCLEANUP, PHYSM_ROUND); +} + void *physmap_try_ro(const char *descr, uintptr_t phys_addr, size_t len) { return physmap_common(descr, phys_addr, len, PHYSM_MAYFAIL, PHYSM_RO, PHYSM_NOCLEANUP, PHYSM_EXACT); diff --git a/programmer.h b/programmer.h index a434c25..b32b85f 100644 --- a/programmer.h +++ b/programmer.h @@ -277,6 +277,7 @@ int processor_flash_enable(void); /* physmap.c */ void *physmap(const char *descr, uintptr_t phys_addr, size_t len); void *rphysmap(const char *descr, uintptr_t phys_addr, size_t len); +void *physmap_round(const char *descr, uintptr_t phys_addr, size_t len); void *physmap_try_ro(const char *descr, uintptr_t phys_addr, size_t len); void physunmap(void *virt_addr, size_t len); #if CONFIG_INTERNAL == 1 -- Kind regards, Stefan Tauner

1 0

[PATCH 3/7] Allow physmap_common() to round address ranges to page boundaries.
by Stefan Tauner 10 Jul '13

10 Jul '13

Automatically round address ranges requested from rphysmap() to page boundaries. Other physmap() variants were not changed so this is safe regarding unmapping. I had also to shorten the readbility defines for the physmap_common parameters to make the calls fit a single line. Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> --- physmap.c | 61 ++++++++++++++++++++++++++++++++++++++++--------------------- 1 file changed, 40 insertions(+), 21 deletions(-) diff --git a/physmap.c b/physmap.c index 2c90091..b248f92 100644 --- a/physmap.c +++ b/physmap.c @@ -208,12 +208,35 @@ void physunmap(void *virt_addr, size_t len) } #endif -#define PHYSMAP_NOFAIL 0 -#define PHYSMAP_MAYFAIL 1 -#define PHYSMAP_RW 0 -#define PHYSMAP_RO 1 -#define PHYSMAP_NOCLEANUP 0 -#define PHYSMAP_CLEANUP 1 +#define PHYSM_NOFAIL 0 +#define PHYSM_MAYFAIL 1 +#define PHYSM_RW 0 +#define PHYSM_RO 1 +#define PHYSM_NOCLEANUP 0 +#define PHYSM_CLEANUP 1 +#define PHYSM_EXACT 0 +#define PHYSM_ROUND 1 + +/* Round start to nearest page boundary below and set len so that the resulting + * address range ends at the lowest possible page boundary where the original + * address range is still entirely contained. It returns the difference between + * the rounded start address and the original start address. */ +static uintptr_t round_to_page_boundaries(uintptr_t *start, size_t *len) +{ + uintptr_t page_size = getpagesize(); + uintptr_t page_mask = ~(page_size-1); + uintptr_t end = *start + *len; + uintptr_t old_start = *start; + msg_gspew("page_size=%" PRIxPTR "\n", page_size); + msg_gspew("pre-rounding: start=0x%0*" PRIxPTR ", len=0x%0zx, end=0x%0*" PRIxPTR "\n", + PRIxPTR_WIDTH, *start, *len, PRIxPTR_WIDTH, end); + *start = *start & page_mask; + end = (end + page_size - 1) & page_mask; + *len = end - *start; + msg_gspew("post-rounding: start=0x%0*" PRIxPTR ", len=0x%0zx, end=0x%0*" PRIxPTR "\n", + PRIxPTR_WIDTH, *start, *len, PRIxPTR_WIDTH, *start + *len); + return old_start - *start; +} struct undo_physmap_data { void *virt_addr; @@ -233,24 +256,18 @@ static int undo_physmap(void *data) } static void *physmap_common(const char *descr, uintptr_t phys_addr, size_t len, bool mayfail, - bool readonly, bool autocleanup) + bool readonly, bool autocleanup, bool round) { void *virt_addr; + uintptr_t offset = 0; if (len == 0) { msg_pspew("Not mapping %s, zero size at 0x%0*" PRIxPTR ".\n", descr, PRIxPTR_WIDTH, phys_addr); return ERROR_PTR; } - if ((getpagesize() - 1) & len) { - msg_perr("Mapping %s at 0x%0*" PRIxPTR ", unaligned size 0x%lx.\n", - descr, PRIxPTR_WIDTH, phys_addr, (unsigned long)len); - } - - if ((getpagesize() - 1) & phys_addr) { - msg_perr("Mapping %s, 0x%lx bytes at unaligned 0x%0*" PRIxPTR ".\n", - descr, (unsigned long)len, PRIxPTR_WIDTH, phys_addr); - } + if (round) + offset = round_to_page_boundaries(&phys_addr, &len); if (readonly) virt_addr = sys_physmap_ro_cached(phys_addr, len); @@ -276,7 +293,9 @@ static void *physmap_common(const char *descr, uintptr_t phys_addr, size_t len, "and reboot, or reboot into\n" "single user mode.\n"); #endif - if (!mayfail) + if (mayfail) + return ERROR_PTR; + else exit(3); } @@ -295,7 +314,7 @@ static void *physmap_common(const char *descr, uintptr_t phys_addr, size_t len, } } - return virt_addr; + return virt_addr + offset; unmap_out: physunmap(virt_addr, len); if (!mayfail) @@ -305,17 +324,17 @@ unmap_out: void *physmap(const char *descr, uintptr_t phys_addr, size_t len) { - return physmap_common(descr, phys_addr, len, PHYSMAP_NOFAIL, PHYSMAP_RW, PHYSMAP_NOCLEANUP); + return physmap_common(descr, phys_addr, len, PHYSM_NOFAIL, PHYSM_RW, PHYSM_NOCLEANUP, PHYSM_EXACT); } void *rphysmap(const char *descr, uintptr_t phys_addr, size_t len) { - return physmap_common(descr, phys_addr, len, PHYSMAP_NOFAIL, PHYSMAP_RW, PHYSMAP_CLEANUP); + return physmap_common(descr, phys_addr, len, PHYSM_NOFAIL, PHYSM_RW, PHYSM_CLEANUP, PHYSM_ROUND); } void *physmap_try_ro(const char *descr, uintptr_t phys_addr, size_t len) { - return physmap_common(descr, phys_addr, len, PHYSMAP_MAYFAIL, PHYSMAP_RO, PHYSMAP_NOCLEANUP); + return physmap_common(descr, phys_addr, len, PHYSM_MAYFAIL, PHYSM_RO, PHYSM_NOCLEANUP, PHYSM_EXACT); } /* MSR abstraction implementations for Linux, OpenBSD, FreeBSD/Dragonfly, OSX, libpayload -- Kind regards, Stefan Tauner

1 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

flashrom