flashrom August 2012

flashrom@flashrom.org

31 participants
160 discussions

Flashing M/B Bios (Asus P5LD2-VM) goes wrong
by Emil Sirbu 07 Sep '12

07 Sep '12

Hello, I have an ASUS P5LD2-VM, and I try to update bios, first with flashrom v0.9.2-r1028 ( http://paste.flashrom.org/view.php?id=489 ) then with 0.9.3 ( http://paste.flashrom.org/view.php?id=490 also, the ROM is at http://paste.flashrom.org/uploads/491.bin ) but: Writing flash chip... Erasing flash chip... Looking at blockwise erase function 0... trying... 0x000000-0x000fff, 0x001000-0x001fff, 0x002000-0x002fff, 0x003000-0xff, 0x004000-0x004fff, 0x005000-0x005fff, ERASE FAILED at 0x00005760! Expected=0xff, Read=0x50, failed byte count from 0x00005000-0x00005fff: 0x87a ERASE FAILED! Can you help whit this? I'm carco on freenode/#flashrom channel Thank you, Emil Sirbu

3 4

[PATCH] remove exit calls from sp_sync_read_timeout and sp_synchronize
by Niklas Söderlund 06 Sep '12

06 Sep '12

Add return values to sp_synchronize so we can signal a failure to the only upstream caller, which is prepared to propagate a failure. The addition of a return code to signal a failure from the read syscall in sp_sync_read_timeout is a bit of a stretch and might look strange. But the only caller is in the loop in sp_synchronize and any sequential call to read will result in the same error and end result. But please take a extra look on this to make sure we don't miss any corner case. The alternative would be to differentiate between a read failure over fast detection of a EAGAIN. The end result would in my opinion only make the code harder to read. The cost is that upon read failure the user might be spammed up to 10 times with the same error message before flashrom dies. Signed-off-by: Niklas Söderlund <niso(a)kth.se> --- serprog.c | 35 ++++++++++++++++++++++------------- 1 file changed, 22 insertions(+), 13 deletions(-) diff --git a/serprog.c b/serprog.c index 02996ea..f03d332 100644 --- a/serprog.c +++ b/serprog.c @@ -145,8 +145,10 @@ static int sp_sync_read_timeout(int loops) rv = read(sp_fd, &c, 1); if (rv == 1) return c; - if ((rv == -1) && (errno != EAGAIN)) - sp_die("read"); + if ((rv == -1) && (errno != EAGAIN)) { + msg_perr("read: %s\n", strerror(errno)); + return -1; + } usleep(10 * 1000); /* 10ms units */ } return -1; @@ -158,7 +160,7 @@ static int sp_sync_read_timeout(int loops) * blocking read - TODO: add an alarm() timer for the rest of the app on * * serial operations, though not such a big issue as the first thing to * * do is synchronize (eg. check that device is alive). */ -static void sp_synchronize(void) +static int sp_synchronize(void) { int i; int flags = fcntl(sp_fd, F_GETFL); @@ -169,8 +171,10 @@ static void sp_synchronize(void) * the device serial parser to get to a sane state, unless if it * * is waiting for a real long write-n. */ memset(buf, S_CMD_NOP, 8); - if (write(sp_fd, buf, 8) != 8) - sp_die("flush write"); + if (write(sp_fd, buf, 8) != 8) { + msg_perr("flush write: %s\n", strerror(errno)); + return 1; + } /* A second should be enough to get all the answers to the buffer */ usleep(1000 * 1000); sp_flush_incoming(); @@ -181,9 +185,11 @@ static void sp_synchronize(void) * attempt, ~1s if immediate success. */ for (i = 0; i < 8; i++) { int n; - unsigned char c = S_CMD_SYNCNOP; - if (write(sp_fd, &c, 1) != 1) - sp_die("sync write"); + int c = S_CMD_SYNCNOP; + if (write(sp_fd, &c, 1) != 1) { + msg_perr("sync write: %s\n", strerror(errno)); + return 1; + } msg_pdbg("."); fflush(stdout); for (n = 0; n < 10; n++) { @@ -194,8 +200,10 @@ static void sp_synchronize(void) if (c != S_ACK) continue; c = S_CMD_SYNCNOP; - if (write(sp_fd, &c, 1) != 1) - sp_die("sync write"); + if (write(sp_fd, &c, 1) != 1) { + msg_perr("sync write: %s\n", strerror(errno)); + return 1; + } c = sp_sync_read_timeout(50); if (c != S_NAK) break; /* fail */ @@ -206,12 +214,12 @@ static void sp_synchronize(void) flags &= ~O_NONBLOCK; fcntl(sp_fd, F_SETFL, flags); msg_pdbg("\n"); - return; + return 0; } } msg_perr("Error: cannot synchronize protocol " "- check communications and reset device?\n"); - exit(1); + return 1; } static int sp_check_commandavail(uint8_t command) @@ -432,7 +440,8 @@ int serprog_init(void) sp_check_avail_automatic = 0; - sp_synchronize(); + if (sp_synchronize()) + return 1; msg_pdbg(MSGHEADER "Synchronized\n"); -- 1.7.10.3

2 2

Gembird-PCI-VIA VT6421A-W39V040C
by Alexey 05 Sep '12

05 Sep '12

Hi! I'm going to flash Gembird PCI 3SATA+1IDE controller's BIOS, there is W39V040C. I found an excellent professionaly modified ROM for this device. I will try to do it on board of ASUS P4P800E-DELUX. Just let me know, is it possible in the moment or I have to wait for a new version of flashrom.exe? This device have unfriendly relations to Radeon X1600 AGP, on board of ASRock P4i45E MB, so I found only new modified ROM of PCI controller can help. Waiting for Your answer, Alexey.

2 4

[PATCH] Hack libpci not to exit(1) during init if PCI does not work
by Carl-Daniel Hailfinger 05 Sep '12

05 Sep '12

This patch is for testing only, and not for immediate merge. libpci unhelpfully calls exit(1) instead of reporting an error if init fails. Fortunately, that part of libpci has not changed since the year 2000, so we can use a really evil hack to avoid that. We supply an alternative error printing function (which does not exit(1)) and this function detects which error is happening and hacks internal libpci structures in a way that avoids a later segfault due to NULL pointer dereference. We MUST absolutely make sure that pci_init() in libpci will not change or this hack will explode spectacularly. Once we decide that this hack is the way to go, notifying the libpci maintainer is essential. Apply this patch, then compile flashrom with make CONFIG_DUMMY=no CONFIG_INTERNAL=o CONFIG_SERPROG=no CONFIG_RAYER_SPI=no CONFIG_NIC3COM=no CONFIG_GFXNVIDIA=no CONFIG_SATASII=no CONFIG_FT2232_SPI=no CONFIG_DRKAISER=no CONFIG_NICREALTEK=no CONFIG_NICINTEL_SPI=no CONFIG_BUSPIRATE_SPI=no CONFIG_OGP_SPI=no CONFIG_SATAMV=no CONFIG_NICINTEL=yes To check if the hack works, run "flashrom -V" and respond to this mail with the output. You should get a message near the end which says that programmer initialization failed. Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Index: flashrom-libpci_hack_avoid_exit1_during_init/pcidev.c =================================================================== --- flashrom-libpci_hack_avoid_exit1_during_init/pcidev.c (Revision 1362) +++ flashrom-libpci_hack_avoid_exit1_during_init/pcidev.c (Arbeitskopie) @@ -21,6 +21,8 @@ #include <stdlib.h> #include <string.h> +#include <stdio.h> +#include <stdarg.h> #include "flash.h" #include "programmer.h" @@ -188,6 +190,58 @@ return 0; } +/* Ugliest hack ever in flashrom! Should work for all libpci versions released + * in the year 2000 or later. Checked to make sense until at least version + * 3.1.8-test1. + */ +static void fake_libpci_method_init(struct pci_access *foo) +{ + msg_pinfo("Working around another pcilib crash\n"); + return; +} + +static struct { + char *name; /* name in all libpci versions */ + char *dummy1; /* Only present in some libpci versions */ + void *dummy2; + void (*init1)(struct pci_access *); /* init in older versions */ + void (*init2)(struct pci_access *); /* init in newer versions */ + void *dummy3; + void *dummy4; + void *dummy5; + void *dummy6; + void *dummy7; + void *dummy8; + void *dummy9; + void *dummy10; +} fake_libpci_methods = { + .name = "None", + .init1 = &fake_libpci_method_init, + .init2 = &fake_libpci_method_init, +}; + +static int pci_init_failed = 0; + +static void my_libpci_generic_error(char *msg, ...) +{ + va_list ap; + + msg_pinfo("pcilib: "); + va_start(ap, msg); + vprintf(msg, ap); + va_end(ap); + msg_pinfo("\n"); + if (!strcmp(msg, "Cannot find any working access method.")) { + /* Workaround for crash, very ugly. */ + msg_pinfo("Working around pcilib crash\n"); + pacc->methods = (void *)&fake_libpci_methods; + pci_init_failed = 1; + } else { + /* Any error in libpci would result in exit(1) by default. */ + exit(1); + } +} + uintptr_t pcidev_init(int bar, const struct pcidev_status *devs) { struct pci_dev *dev; @@ -198,7 +252,14 @@ uintptr_t addr = 0, curaddr = 0; pacc = pci_alloc(); /* Get the pci_access structure */ + pacc->error = &my_libpci_generic_error; pci_init(pacc); /* Initialize the PCI library */ + /* pci_init returns void, so we have to hack it. */ + if (pci_init_failed) { + /* Yes, technically we would have to free stuff... */ + return 1; + } + pci_scan_bus(pacc); /* We want to get the list of devices */ pci_filter_init(pacc, &filter); -- http://www.hailfinger.org/

4 7

[PATCH 1/2] libftdi complains that it is impossible to set interface on an already open device.
by Ilya A. Volynets-Evenbakh 03 Sep '12

03 Sep '12

Call to ftdi_set_interface right after ftdi_init and before call to ftdi_usb_open Signed-off-by: Ilya A. Volynets-Evenbakh <ilya(a)total-knowledge.com> --- ft2232_spi.c | 10 +++++----- 1 files changed, 5 insertions(+), 5 deletions(-) diff --git a/ft2232_spi.c b/ft2232_spi.c index aec2fd5..aaca5f3 100644 --- a/ft2232_spi.c +++ b/ft2232_spi.c @@ -276,6 +276,11 @@ int ft2232_spi_init(void) return -3; } + if (ftdi_set_interface(ftdic, ft2232_interface) < 0) { + msg_perr("Unable to select interface: %s\n", + ftdic->error_str); + } + f = ftdi_usb_open(ftdic, ft2232_vid, ft2232_type); if (f < 0 && f != -5) { @@ -290,11 +295,6 @@ int ft2232_spi_init(void) clock_5x = 0; } - if (ftdi_set_interface(ftdic, ft2232_interface) < 0) { - msg_perr("Unable to select interface: %s\n", - ftdic->error_str); - } - if (ftdi_usb_reset(ftdic) < 0) { msg_perr("Unable to reset FTDI device\n"); } -- 1.7.3.4

2 4

Re: [flashrom] About mcp73 support on flashrom
by Stefan Tauner 02 Sep '12

02 Sep '12

On Thu, 22 Mar 2012 13:53:34 +0200 Olcay Korkmaz <olcay.mz(a)gmail.com> wrote: > Hello all > I have Zotac 630i-itx board (old pcb rev model..very old i guess) > flashrom -V log details on mail attachment > is it safe to erase/write process > extra info bios dump is ok via flashrom -r command > Hello Olcay! seems to work fine: http://www.flashrom.org/pipermail/flashrom/2012-April/009073.html -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

2 1

[PATCH] Add support for AJAWe to pony_spi.
by Stefan Tauner 31 Aug '12

31 Aug '12

http://www.ajawe.pl/ajawe0208.htm http://www.ajawe.pl/files/0208_R_DOK.pdf First version was Signed-off-by: Virgil-Adrian Teaca <darkstarlinux(a)gmail.com> Refinements and general cleanup is Signed-off-by: Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at> --- Virgil: could you please take a look at this and maybe even test it? i have changed a few bits of your patch, most of it shouldnt change behavior, with one major exception: AJAWe is always assumed to be attached. this should not be a big issue the worst thing that could happen is that we toggle a few bits of an unknown device, but OTOH even if your detection would work it is not really safe, but just a good guess (this also applies to some extent to the SI-Prog detection). if i interpret the AJAWe schematic correctly, then it sets pin 6 (DSR) high if there is too much voltage on VCC and low otherwise. it does so by two times inverting the voltage level that remains after subtracting 3.3V (with the zener diode D5) from VCC. the trigger voltage of the schmitt triggers is about 2V, so this would result DSR being high only if the voltage rises above abot 5.3V. this looks more like an overvoltage signaling than a power inidicator to me. OTOH i am bad at electronics... we need to check this with someone who knows what he is doing :) pony_spi.c | 171 +++++++++++++++++++++++++++++++++++++++--------------------- print.c | 2 +- 2 files changed, 113 insertions(+), 60 deletions(-) diff --git a/pony_spi.c b/pony_spi.c index fb90a33..7875200 100644 --- a/pony_spi.c +++ b/pony_spi.c @@ -17,8 +17,27 @@ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ -/* Driver for the SI-Prog like hardware by Lancos.com. - * See http://www.lancos.com/siprogsch.html for schematics and instructions. +/* Driver for serial programmers compatible with SI-Prog or AJAWe. + * + * See http://www.lancos.com/siprogsch.html for SI-Prog schematics and instructions. + * See http://www.ajawe.pl/ajawe0208.htm for AJAWe serial programmer documentation. + * + * Pin layout for SI-Prog-like hardware: + * + * MOSI <-------< DTR + * MISO >-------> CTS + * SCK <---+---< RTS + * +---> DSR + * CS# <-------< TXD + * + * and for the AJAWe serial programmer: + * + * MOSI <-------< DTR + * MISO >-------> CTS + * SCK <-------< RTS + * CS# <-------< TXD + * + * DCE >-------> DSR */ #include <stdlib.h> @@ -28,55 +47,50 @@ #include "flash.h" #include "programmer.h" -/* We have: - - MOSI -----> DTR - MISO -----> CTS - SCK --+--> RTS - +--> DSR - CS# -----> TXD - -*/ - enum pony_type { TYPE_SI_PROG, TYPE_SERBANG, + TYPE_AJAWE }; -/* The hardware programmer used. */ -static enum pony_type pony_type = TYPE_SI_PROG; +/* Pins for master->slave direction */ +static int pony_negate_cs = 1; +static int pony_negate_sck = 0; +static int pony_negate_mosi = 0; +/* Pins for slave->master direction */ +static int pony_negate_miso = 0; static void pony_bitbang_set_cs(int val) { - if (pony_type == TYPE_SI_PROG) - { - /* CS# is negated by the SI-Prog programmer. */ + if (pony_negate_cs) val ^= 1; - } + sp_set_pin(PIN_TXD, val); } static void pony_bitbang_set_sck(int val) { + if (pony_negate_sck) + val ^= 1; + sp_set_pin(PIN_RTS, val); } static void pony_bitbang_set_mosi(int val) { + if (pony_negate_mosi) + val ^= 1; + sp_set_pin(PIN_DTR, val); } static int pony_bitbang_get_miso(void) { - int tmp; - - tmp = sp_get_pin(PIN_CTS); + int tmp = sp_get_pin(PIN_CTS); - if (pony_type == TYPE_SERBANG) - { - /* MISO is negated by the SERBANG programmer. */ + if (pony_negate_miso) tmp ^= 1; - } + return tmp; } @@ -91,15 +105,17 @@ static const struct bitbang_spi_master bitbang_spi_master_pony = { int pony_spi_init(void) { + int i, data_out; char *arg = NULL; - int i, data_in, data_out, have_device = 0; - int have_prog = 1; + enum pony_type type = TYPE_SI_PROG; + char *name; + int have_device = 0; + int have_prog = 0; - /* The parameter is on format "dev=/dev/device,type=serbang" */ + /* The parameter is in format "dev=/dev/device,type=serbang" */ arg = extract_programmer_param("dev"); - if (arg && strlen(arg)) { - sp_fd = sp_openserport( arg, 9600 ); + sp_fd = sp_openserport(arg, 9600); if (sp_fd < 0) { free(arg); return 1; @@ -110,51 +126,88 @@ int pony_spi_init(void) if (!have_device) { msg_perr("Error: No valid device specified.\n" - "Use flashrom -p pony_spi:dev=/dev/device\n"); - + "Use flashrom -p pony_spi:dev=/dev/device[,type=name]\n"); return 1; } - /* Register the shutdown callback. */ - if (register_shutdown(serialport_shutdown, NULL)) { + + arg = extract_programmer_param("type"); + if (arg && !strcasecmp(arg, "serbang")) { + type = TYPE_SERBANG; + } else if (arg && !strcasecmp(arg, "si_prog")) { + type = TYPE_SI_PROG; + } else if (arg && !strcasecmp( arg, "ajawe")) { + type = TYPE_AJAWE; + } else if (arg && !strlen(arg)) { + msg_perr("Error: Missing argument for programmer type.\n"); + free(arg); + } else if (arg){ + msg_perr("Error: Invalid programmer type specified.\n"); + free(arg); return 1; } - arg = extract_programmer_param("type"); + free(arg); - if (arg) { - if (!strcasecmp( arg, "serbang")) { - pony_type = TYPE_SERBANG; - } else if (!strcasecmp( arg, "si_prog") ) { - pony_type = TYPE_SI_PROG; - } else { - msg_perr("Error: Invalid programmer type specified.\n"); - free(arg); - return 1; - } + /* + * Configure the serial port pins, depending on the used programmer. + */ + switch (type) { + case TYPE_AJAWE: + pony_negate_cs = 1; + pony_negate_sck = 1; + pony_negate_mosi = 1; + pony_negate_miso = 1; + name = "AJAWe"; + break; + case TYPE_SERBANG: + pony_negate_cs = 0; + pony_negate_sck = 0; + pony_negate_mosi = 0; + pony_negate_miso = 1; + name = "serbang"; + break; + default: + case TYPE_SI_PROG: + pony_negate_cs = 1; + pony_negate_sck = 0; + pony_negate_mosi = 0; + pony_negate_miso = 0; + name = "SI-Prog"; + break; } - free(arg); + msg_pdbg("Using %s programmer pinout.\n", name); - msg_pdbg("Using the %s programmer.\n", ((pony_type == TYPE_SI_PROG ) ? "SI-Prog" : "SERBANG")); /* - * Detect if there is a SI-Prog compatible programmer connected. + * Detect if there is a compatible hardware programmer connected. */ pony_bitbang_set_cs(1); + pony_bitbang_set_sck(1); pony_bitbang_set_mosi(1); - /* We toggle SCK while we keep MOSI and CS# on. */ - for (i = 1; i <= 10; ++i) { - data_out = i & 1; - sp_set_pin(PIN_RTS, data_out); - programmer_delay( 1000 ); - data_in = sp_get_pin(PIN_DSR); - - if (data_out != data_in) { - have_prog = 0; - break; + switch (type) { + case TYPE_AJAWE: + have_prog = 1; + break; + case TYPE_SI_PROG: + case TYPE_SERBANG: + default: + have_prog = 1; + /* We toggle RTS/SCK a few times and see if DSR changes too. */ + for (i = 1; i <= 10; i++) { + data_out = i & 1; + sp_set_pin(PIN_RTS, data_out); + programmer_delay(1000); + + /* If DSR does not change, we are not connected to what we think */ + if (data_out != sp_get_pin(PIN_DSR)) { + have_prog = 0; + break; + } } + break; } if (!have_prog) { - msg_perr( "No SI-Prog compatible hardware detected.\n" ); + msg_perr("No programmer compatible with %s detected.\n", name); return 1; } diff --git a/print.c b/print.c index f0cd753..ae067a6 100644 --- a/print.c +++ b/print.c @@ -520,7 +520,7 @@ void print_supported(void) msg_ginfo("\nSupported devices for the %s programmer:\n", programmer_table[PROGRAMMER_PONY_SPI].name); /* FIXME */ - msg_ginfo("SI-Prog and SERBANG serial port programmer\n"); + msg_ginfo("SI-Prog, serbang and AJAWe serial port programmers\n"); #endif #if CONFIG_NICINTEL == 1 msg_ginfo("\nSupported devices for the %s programmer:\n", -- Kind regards, Stefan Tauner

1 0

Re: [flashrom] [PATCH] Add support for generic SPI RES matching.
by Carl-Daniel Hailfinger 30 Aug '12

30 Aug '12

Am 30.08.2012 23:55 schrieb Stefan Tauner: > Support generic 1-byte SPI RES matching (low match quality) and enable > read support by guessing the chip size from the ID. > Support generic 2-byte SPI RES matching (high match quality). > > 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> > --- > chipdrivers.h | 1 + > flashchips.c | 31 +++++++++++++++++++++++++++++ > flashrom.c | 4 +++- > spi25.c | 61 +++++++++++++++++++++++++++++++++++++++++++++------------ > 4 files changed, 84 insertions(+), 13 deletions(-) > > diff --git a/chipdrivers.h b/chipdrivers.h > index b43e16e..63c5d37 100644 > --- a/chipdrivers.h > +++ b/chipdrivers.h > @@ -33,6 +33,7 @@ int probe_spi_rdid4(struct flashctx *flash); > int probe_spi_rems(struct flashctx *flash); > int probe_spi_res1(struct flashctx *flash); > int probe_spi_res2(struct flashctx *flash); > +int probe_spi_automagic_res1(struct flashctx *flash); > int spi_write_enable(struct flashctx *flash); > int spi_write_disable(struct flashctx *flash); > int spi_block_erase_20(struct flashctx *flash, unsigned int addr, unsigned int blocklen); > diff --git a/flashchips.c b/flashchips.c > index 8fe50bd..c175108 100644 > --- a/flashchips.c > +++ b/flashchips.c > @@ -9662,6 +9662,24 @@ const struct flashchip flashchips[] = { > > { > .vendor = "Generic", > + .name = "unknown SPI chip (RES1)", > + .bustype = BUS_SPI, > + .manufacture_id = GENERIC_MANUF_ID, > + .model_id = GENERIC_DEVICE_ID, > + .total_size = 0, > + .page_size = 256, > + .tested = TEST_BAD_PREW, > + .probe = probe_spi_automagic_res1, > + .probe_timing = TIMING_ZERO, > + .block_erasers = {}, > + .unlock = spi_disable_blockprotect, Please kill the unlock here. AFAICS we have no way to determine how to unlock for a REMS1 chip. > + .write = NULL, > + .read = spi_chip_read, > + .voltage = {}, > + }, The RES1 chip should be below the RES2 chip to make sure the RES2 chip matches first (better match quality) and prevents RES1 matching if RES2 matches. > + > + { > + .vendor = "Generic", > .name = "unknown SPI chip (REMS)", > .bustype = BUS_SPI, > .manufacture_id = GENERIC_MANUF_ID, > @@ -9673,5 +9691,18 @@ const struct flashchip flashchips[] = { > .write = NULL, > }, > > + { > + .vendor = "Generic", > + .name = "unknown SPI chip (RES2)", > + .bustype = BUS_SPI, > + .manufacture_id = GENERIC_MANUF_ID, > + .model_id = GENERIC_DEVICE_ID, > + .total_size = 0, > + .page_size = 256, > + .tested = TEST_BAD_PREW, > + .probe = probe_spi_res2, > + .write = NULL, > + }, > + > { NULL } > }; > diff --git a/flashrom.c b/flashrom.c > index fdc5412..148f969 100644 > --- a/flashrom.c > +++ b/flashrom.c > @@ -1028,7 +1028,9 @@ int probe_flash(struct registered_programmer *pgm, int startchip, struct flashct > if (startchip == 0) > break; > /* Not the first flash chip detected on this bus, but not a generic match either. */ > - if ((flash->chip->model_id != GENERIC_DEVICE_ID) && (flash->chip->model_id != SFDP_DEVICE_ID)) > + if ((flash->chip->manufacture_id != GENERIC_MANUF_ID) && > + (flash->chip->model_id != GENERIC_DEVICE_ID) && > + (flash->chip->model_id != SFDP_DEVICE_ID)) > break; > /* Not the first flash chip detected on this bus, and it's just a generic match. Ignore it. */ > notfound: > diff --git a/spi25.c b/spi25.c > index 914b821..7d9e948 100644 > --- a/spi25.c > +++ b/spi25.c > @@ -232,6 +232,7 @@ int probe_spi_rems(struct flashctx *flash) > > int probe_spi_res1(struct flashctx *flash) > { > + const struct flashchip *chip = flash->chip; > static const unsigned char allff[] = {0xff, 0xff, 0xff}; > static const unsigned char all00[] = {0x00, 0x00, 0x00}; > unsigned char readarr[3]; > @@ -265,18 +266,23 @@ int probe_spi_res1(struct flashctx *flash) > > msg_cdbg("%s: id 0x%x\n", __func__, id2); > > - if (id2 != flash->chip->model_id) > - return 0; > + if (id2 == flash->chip->model_id) { my bad, this could have been simplified to use chip->model_id instead of flash->chip->model_id > + /* Print the status register to tell the user about possible write protection. */ > + spi_prettyprint_status_register(flash); > > - /* Print the status register to tell the > - * user about possible write protection. > - */ > - spi_prettyprint_status_register(flash); > - return 1; > + return 1; > + } > + > + /* Test if there is any device ID. */ > + if (GENERIC_MANUF_ID == chip->manufacture_id && id2 == chip->model_id) > + return 1; > + > + return 0; > } > > int probe_spi_res2(struct flashctx *flash) > { > + const struct flashchip *chip = flash->chip; > unsigned char readarr[2]; > uint32_t id1, id2; > > @@ -289,13 +295,44 @@ int probe_spi_res2(struct flashctx *flash) > > msg_cdbg("%s: id1 0x%x, id2 0x%x\n", __func__, id1, id2); > > - if (id1 != flash->chip->manufacture_id || id2 != flash->chip->model_id) > + if (id1 == chip->manufacture_id && id2 == chip->model_id) { > + /* Print the status register to tell the > + * user about possible write protection. > + */ > + spi_prettyprint_status_register(flash); > + > + return 1; > + } > + > + /* Test if this is a pure vendor match. */ > + if (id1 == chip->manufacture_id && GENERIC_DEVICE_ID == chip->model_id) > + return 1; > + > + /* Test if there is any vendor ID and if this is indeed a two-byte RES. */ > + if (GENERIC_MANUF_ID == chip->manufacture_id && id1 != 0xff && id1 != id2) > + return 1; > + > + return 0; > +} > + > +int probe_spi_automagic_res1(struct flashctx *flash) > +{ > + unsigned char readarr[1]; > + unsigned int total_size; > + uint32_t id1; IMHO we should check here that neither RDID nor REMS nor RES2 work before proceeding with RES1. > + > + if (spi_res(flash, readarr, 1)) > return 0; > > - /* Print the status register to tell the > - * user about possible write protection. > - */ > - spi_prettyprint_status_register(flash); > + id1 = readarr[0]; > + msg_cdbg("%s: id1 0x%02x\n", __func__, id1); > + if (id1 < 0x10 || id1 > 0x17) { > + msg_cdbg("RES ID out of bounds!\n"); > + return 0; > + } > + total_size = (2 << id1) / 1024; > + flash->chip->total_size = total_size; > + flash->chip->tested = TEST_OK_PR | TEST_BAD_ERASE | TEST_BAD_WRITE; > return 1; > } > I might have missed something, but right now I hope the review is complete. Regards, Carl-Daniel -- http://www.hailfinger.org/

2 1

[PATCH] Add support for generic SPI RES matching.
by Stefan Tauner 30 Aug '12

30 Aug '12

Support generic 1-byte SPI RES matching (low match quality) and enable read support by guessing the chip size from the ID. Support generic 2-byte SPI RES matching (high match quality). 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> --- chipdrivers.h | 1 + flashchips.c | 31 +++++++++++++++++++++++++++++ flashrom.c | 4 +++- spi25.c | 61 +++++++++++++++++++++++++++++++++++++++++++++------------ 4 files changed, 84 insertions(+), 13 deletions(-) diff --git a/chipdrivers.h b/chipdrivers.h index b43e16e..63c5d37 100644 --- a/chipdrivers.h +++ b/chipdrivers.h @@ -33,6 +33,7 @@ int probe_spi_rdid4(struct flashctx *flash); int probe_spi_rems(struct flashctx *flash); int probe_spi_res1(struct flashctx *flash); int probe_spi_res2(struct flashctx *flash); +int probe_spi_automagic_res1(struct flashctx *flash); int spi_write_enable(struct flashctx *flash); int spi_write_disable(struct flashctx *flash); int spi_block_erase_20(struct flashctx *flash, unsigned int addr, unsigned int blocklen); diff --git a/flashchips.c b/flashchips.c index 8fe50bd..c175108 100644 --- a/flashchips.c +++ b/flashchips.c @@ -9662,6 +9662,24 @@ const struct flashchip flashchips[] = { { .vendor = "Generic", + .name = "unknown SPI chip (RES1)", + .bustype = BUS_SPI, + .manufacture_id = GENERIC_MANUF_ID, + .model_id = GENERIC_DEVICE_ID, + .total_size = 0, + .page_size = 256, + .tested = TEST_BAD_PREW, + .probe = probe_spi_automagic_res1, + .probe_timing = TIMING_ZERO, + .block_erasers = {}, + .unlock = spi_disable_blockprotect, + .write = NULL, + .read = spi_chip_read, + .voltage = {}, + }, + + { + .vendor = "Generic", .name = "unknown SPI chip (REMS)", .bustype = BUS_SPI, .manufacture_id = GENERIC_MANUF_ID, @@ -9673,5 +9691,18 @@ const struct flashchip flashchips[] = { .write = NULL, }, + { + .vendor = "Generic", + .name = "unknown SPI chip (RES2)", + .bustype = BUS_SPI, + .manufacture_id = GENERIC_MANUF_ID, + .model_id = GENERIC_DEVICE_ID, + .total_size = 0, + .page_size = 256, + .tested = TEST_BAD_PREW, + .probe = probe_spi_res2, + .write = NULL, + }, + { NULL } }; diff --git a/flashrom.c b/flashrom.c index fdc5412..148f969 100644 --- a/flashrom.c +++ b/flashrom.c @@ -1028,7 +1028,9 @@ int probe_flash(struct registered_programmer *pgm, int startchip, struct flashct if (startchip == 0) break; /* Not the first flash chip detected on this bus, but not a generic match either. */ - if ((flash->chip->model_id != GENERIC_DEVICE_ID) && (flash->chip->model_id != SFDP_DEVICE_ID)) + if ((flash->chip->manufacture_id != GENERIC_MANUF_ID) && + (flash->chip->model_id != GENERIC_DEVICE_ID) && + (flash->chip->model_id != SFDP_DEVICE_ID)) break; /* Not the first flash chip detected on this bus, and it's just a generic match. Ignore it. */ notfound: diff --git a/spi25.c b/spi25.c index 914b821..7d9e948 100644 --- a/spi25.c +++ b/spi25.c @@ -232,6 +232,7 @@ int probe_spi_rems(struct flashctx *flash) int probe_spi_res1(struct flashctx *flash) { + const struct flashchip *chip = flash->chip; static const unsigned char allff[] = {0xff, 0xff, 0xff}; static const unsigned char all00[] = {0x00, 0x00, 0x00}; unsigned char readarr[3]; @@ -265,18 +266,23 @@ int probe_spi_res1(struct flashctx *flash) msg_cdbg("%s: id 0x%x\n", __func__, id2); - if (id2 != flash->chip->model_id) - return 0; + if (id2 == flash->chip->model_id) { + /* Print the status register to tell the user about possible write protection. */ + spi_prettyprint_status_register(flash); - /* Print the status register to tell the - * user about possible write protection. - */ - spi_prettyprint_status_register(flash); - return 1; + return 1; + } + + /* Test if there is any device ID. */ + if (GENERIC_MANUF_ID == chip->manufacture_id && id2 == chip->model_id) + return 1; + + return 0; } int probe_spi_res2(struct flashctx *flash) { + const struct flashchip *chip = flash->chip; unsigned char readarr[2]; uint32_t id1, id2; @@ -289,13 +295,44 @@ int probe_spi_res2(struct flashctx *flash) msg_cdbg("%s: id1 0x%x, id2 0x%x\n", __func__, id1, id2); - if (id1 != flash->chip->manufacture_id || id2 != flash->chip->model_id) + if (id1 == chip->manufacture_id && id2 == chip->model_id) { + /* Print the status register to tell the + * user about possible write protection. + */ + spi_prettyprint_status_register(flash); + + return 1; + } + + /* Test if this is a pure vendor match. */ + if (id1 == chip->manufacture_id && GENERIC_DEVICE_ID == chip->model_id) + return 1; + + /* Test if there is any vendor ID and if this is indeed a two-byte RES. */ + if (GENERIC_MANUF_ID == chip->manufacture_id && id1 != 0xff && id1 != id2) + return 1; + + return 0; +} + +int probe_spi_automagic_res1(struct flashctx *flash) +{ + unsigned char readarr[1]; + unsigned int total_size; + uint32_t id1; + + if (spi_res(flash, readarr, 1)) return 0; - /* Print the status register to tell the - * user about possible write protection. - */ - spi_prettyprint_status_register(flash); + id1 = readarr[0]; + msg_cdbg("%s: id1 0x%02x\n", __func__, id1); + if (id1 < 0x10 || id1 > 0x17) { + msg_cdbg("RES ID out of bounds!\n"); + return 0; + } + total_size = (2 << id1) / 1024; + flash->chip->total_size = total_size; + flash->chip->tested = TEST_OK_PR | TEST_BAD_ERASE | TEST_BAD_WRITE; return 1; } -- Kind regards, Stefan Tauner

1 0

Re: [flashrom] [PATCH 2/3] Add RES/REMS support to all dummyflasher emulated chips.
by Carl-Daniel Hailfinger 30 Aug '12

30 Aug '12

Am 25.08.2012 22:35 schrieb Stefan Tauner: > Also, fix a few odd corner cases in RES/REMS support in dummyflasher emulation. > > Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> and committed in r1589. Regards, Carl-Daniel -- http://www.hailfinger.org/

1 0

← Newer
1
2
3
4
5
6
7
...
16
Older →

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

flashrom August 2012