flashrom

flashrom@flashrom.org

9164 discussions

Re: [flashrom] Added chipsets
by Stefan Tauner 22 Feb '13

22 Feb '13

On Fri, 22 Feb 2013 08:07:33 +0300 Peter Lemenkov <lemenkov(a)gmail.com> wrote: > Hello! > > 2013/2/22 Stefan Tauner <stefan.tauner(a)student.tuwien.ac.at>: > > > It would also make sense to split your huge patch into smaller pieces > > (e.g. one patch per manufacturer) so that we can review it easier. > > Maybe it's time to consider switching to Git entirely? Git is much > more better suited for this kind of task. You need to convince Carl-Daniel not me :) I have never checked out flashrom with subversion directly, only with git-svn and that works very nicely and there is no reason why submitting patches would be easier due to a switch to git alone (i.e. when we still demand patches to be sent to the mailing list instead of pull requests on github or something similar). As said, I use it all the time and there has been no problems with the patches generated by me or anybody else using git AFAIK. Also, Carl-Daniel repeatedly said there is no way he will switch to git and since there is no immediate need to do so this will not happen in the near future. If anybody has major problems with the git-svn/git send-email workflow he can also send me pull requests on github https://github.com/stefanct/flashrom But I don't make any promises that I will merge them in a timely manner *or at all* and this should be regarded as the last resort. -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

1 0

[PATCH] dediprog: Use asynchronous bulk transfers for reading
by Nico Huber 22 Feb '13

22 Feb '13

This utilizes libusb-1.0's interface for asynchronous transfers. A read is split into 512 byte chunks as required by the dediprog hardware. Up to eight chunks are scheduled for asynchronous transfer which results in a viable speed-up. First tests have shown a three times speed increase. Signed-off-by: Nico Huber <nico.huber(a)secunet.com> --- dediprog.c | 92 ++++++++++++++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 83 insertions(+), 9 deletions(-) diff --git a/dediprog.c b/dediprog.c index 93c4072..3cd7b01 100644 --- a/dediprog.c +++ b/dediprog.c @@ -31,6 +31,7 @@ #define FIRMWARE_VERSION(x,y,z) ((x << 16) | (y << 8) | z) #define DEFAULT_TIMEOUT 3000 +#define DEDIPROG_ASYNC_TRANSFERS 8 /* at most 8 asynchronous transfers */ static libusb_context *usb_ctx; static libusb_device_handle *dediprog_handle; static int dediprog_firmwareversion; @@ -235,6 +236,38 @@ static int dediprog_set_spi_speed(unsigned int spispeed_idx) return 0; } +struct dediprog_transfer_status { + int error; /* OK if 0, ERROR else */ + unsigned int queued_idx; + unsigned int finished_idx; +}; + +static void dediprog_bulk_read_cb(struct libusb_transfer *const transfer) +{ + struct dediprog_transfer_status *const status = (struct dediprog_transfer_status *)transfer->user_data; + if (transfer->status != LIBUSB_TRANSFER_COMPLETED) { + status->error = 1; + msg_perr("SPI bulk read failed!\n"); + } + ++status->finished_idx; +} + +static int dediprog_bulk_read_poll(const struct dediprog_transfer_status *const status, const int finish) +{ + if (status->finished_idx >= status->queued_idx) + return 0; + + do { + struct timeval timeout = { 10, 0 }; + const int ret = libusb_handle_events_timeout_completed(usb_ctx, &timeout, NULL); + if (ret < 0) { + msg_perr("Polling read events failed: %i %s!\n", ret, libusb_error_name(ret)); + return 1; + } + } while (finish && (status->finished_idx < status->queued_idx)); + return 0; +} + /* Bulk read interface, will read multiple 512 byte chunks aligned to 512 bytes. * @start start address * @len length @@ -243,7 +276,7 @@ static int dediprog_set_spi_speed(unsigned int spispeed_idx) static int dediprog_spi_bulk_read(struct flashctx *flash, uint8_t *buf, unsigned int start, unsigned int len) { - int ret, transferred; + int ret, err = 1; unsigned int i; /* chunksize must be 512, other sizes will NOT work at all. */ const unsigned int chunksize = 0x200; @@ -253,6 +286,10 @@ static int dediprog_spi_bulk_read(struct flashctx *flash, uint8_t *buf, chunksize & 0xff, (chunksize >> 8) & 0xff }; + struct dediprog_transfer_status status = { 0, 0, 0 }; + struct libusb_transfer *transfers[DEDIPROG_ASYNC_TRANSFERS] = { NULL, }; + struct libusb_transfer *transfer; + if ((start % chunksize) || (len % chunksize)) { msg_perr("%s: Unaligned start=%i, len=%i! Please report a bug " "at flashrom(a)flashrom.org\n", __func__, start, len); @@ -273,17 +310,54 @@ static int dediprog_spi_bulk_read(struct flashctx *flash, uint8_t *buf, return 1; } - for (i = 0; i < count; i++) { - ret = libusb_bulk_transfer(dediprog_handle, 0x80 | dediprog_endpoint, - buf + i * chunksize, chunksize, &transferred, DEFAULT_TIMEOUT); - if ((ret < 0) || (transferred != chunksize)) { - msg_perr("SPI bulk read %i failed, expected %i, got %i %s!\n", - i, chunksize, ret, libusb_error_name(ret)); - return 1; + /* + * Ring buffer of bulk transfers. + * Poll until at least one transfer is ready, + * schedule next transfers until buffer is full. + */ + + /* Allocate bulk transfers. */ + for (i = 0; i < min(DEDIPROG_ASYNC_TRANSFERS, count); ++i) { + transfers[i] = libusb_alloc_transfer(0); + if (!transfers[i]) { + msg_perr("Allocating libusb transfer %i failed: %s!\n", i, libusb_error_name(ret)); + goto _err_free; } } - return 0; + /* Now transfer requested chunks using libusb's asynchronous interface. */ + while (!status.error && (status.queued_idx < count)) { + while ((status.queued_idx - status.finished_idx) < DEDIPROG_ASYNC_TRANSFERS) { + transfer = transfers[status.queued_idx % DEDIPROG_ASYNC_TRANSFERS]; + libusb_fill_bulk_transfer(transfer, dediprog_handle, 0x80 | dediprog_endpoint, + (unsigned char *)buf + status.queued_idx * chunksize, chunksize, + dediprog_bulk_read_cb, &status, DEFAULT_TIMEOUT); + transfer->flags |= LIBUSB_TRANSFER_SHORT_NOT_OK; + ret = libusb_submit_transfer(transfer); + if (ret < 0) { + msg_perr("Submitting SPI bulk read %i failed: %i %s!\n", + status.queued_idx, ret, libusb_error_name(ret)); + goto _err_free; + } + ++status.queued_idx; + } + if (dediprog_bulk_read_poll(&status, 0)) + goto _err_free; + } + /* Wait for transfers to finish. */ + if (dediprog_bulk_read_poll(&status, 1)) + goto _err_free; + /* Check if everything has been transmitted. */ + if ((status.finished_idx < count) || status.error) + goto _err_free; + + err = 0; + +_err_free: + dediprog_bulk_read_poll(&status, 1); + for (i = 0; i < DEDIPROG_ASYNC_TRANSFERS; ++i) + if (transfers[i]) libusb_free_transfer(transfers[i]); + return err; } static int dediprog_spi_read(struct flashctx *flash, uint8_t *buf, -- 1.7.9.5

1 0

Re: [flashrom] Added chipsets
by Stefan Tauner 22 Feb '13

22 Feb '13

Hello Николай, that's quite some work you have done, thank you. There is one very important bit missing though: the signed-off-by line that certifies that you are legally allowed to release these changes under a free license. Please see http://flashrom.org/Development_Guidelines and http://www.coreboot.org/Development_Guidelines#Sign-off_Procedure for details. We can not commit your changes without this. It would also make sense to split your huge patch into smaller pieces (e.g. one patch per manufacturer) so that we can review it easier. I have not looked at the patch in detail, but some naming will have to change probably. > +#define SPANSION_S25FL128 0x2018 > +#define SPANSION_S25FL256 0x0219 > +#define SPANSION_S25FL512 0x0220 these look like typos? -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

3 3

EN25Q32(A/B) fully tested
by Lucian Langa 22 Feb '13

22 Feb '13

Hi, I just tested EN25Q32 chip on a raspberry spi programmer. The memory chip comes from a TP-LINK router (mr3220v2) that contains this chips. I used flashrom+rpi(spi) to reflash router firmware, as I previously bricked the router beyond the recovery. I have desoldered the chip from the router and connected it directly to the raspberry headers. All in all it went fine I managed to reflash the chip and router works fine now. See attached logs from all operation I used (READ/ERASE/WRITE) Cheers, --lucian

2 1

Logs for ASRock G31M-S rev 2.0, Gigabyte GA-H77M-D3H rev. 1.1
by Александр Владимирович 22 Feb '13

22 Feb '13

Logs for ASRock G31M-S rev 2.0, Gigabyte GA-H77M-D3H rev. 1.1

2 1

[commit] r1649 - trunk
by repository service 20 Feb '13

20 Feb '13

Author: hailfinger Date: Wed Feb 20 19:03:36 2013 New Revision: 1649 URL: http://flashrom.org/trac/flashrom/changeset/1649 Log: Adds a programmer parameter 'spispeed' to the dediprog driver to control the transfer rate on the spi bus. The following rates are available (in Hz): 375k, 750k, 1.5M, 2.18M, 3M, 8M, 12M and 24M The original driver reinitializes the programmer after setting the speed, so the initialization calls have moved into a new function dediprog_setup() which is called twice. Signed-off-by: Nico Huber <nico.huber(a)secunet.com> Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Modified: trunk/dediprog.c trunk/flashrom.8 Modified: trunk/dediprog.c ============================================================================== --- trunk/dediprog.c Mon Feb 4 05:38:42 2013 (r1648) +++ trunk/dediprog.c Wed Feb 20 19:03:36 2013 (r1649) @@ -158,7 +158,23 @@ return 0; } -#if 0 +struct dediprog_spispeeds { + const char *const name; + const int speed; +}; + +static const struct dediprog_spispeeds spispeeds[] = { + { "24M", 0x0 }, + { "12M", 0x2 }, + { "8M", 0x1 }, + { "3M", 0x3 }, + { "2.18M", 0x4 }, + { "1.5M", 0x5 }, + { "750k", 0x6 }, + { "375k", 0x7 }, + { NULL, 0x0 }, +}; + /* After dediprog_set_spi_speed, the original app always calls * dediprog_set_spi_voltage(0) and then * dediprog_check_devicestring() four times in a row. @@ -166,56 +182,20 @@ * This looks suspiciously like the microprocessor in the SF100 has to be * restarted/reinitialized in case the speed changes. */ -static int dediprog_set_spi_speed(uint16_t speed) +static int dediprog_set_spi_speed(unsigned int spispeed_idx) { int ret; - unsigned int khz; - /* Case 1 and 2 are in weird order. Probably an organically "grown" - * interface. - * Base frequency is 24000 kHz, divisors are (in order) - * 1, 3, 2, 8, 11, 16, 32, 64. - */ - switch (speed) { - case 0x0: - khz = 24000; - break; - case 0x1: - khz = 8000; - break; - case 0x2: - khz = 12000; - break; - case 0x3: - khz = 3000; - break; - case 0x4: - khz = 2180; - break; - case 0x5: - khz = 1500; - break; - case 0x6: - khz = 750; - break; - case 0x7: - khz = 375; - break; - default: - msg_perr("Unknown frequency selector 0x%x! Aborting.\n", speed); - return 1; - } - msg_pdbg("Setting SPI speed to %u kHz\n", khz); + msg_pdbg("SPI speed is %sHz\n", spispeeds[spispeed_idx].name); - ret = usb_control_msg(dediprog_handle, 0x42, 0x61, speed, 0xff, NULL, - 0x0, DEFAULT_TIMEOUT); + ret = usb_control_msg(dediprog_handle, 0x42, 0x61, spispeeds[spispeed_idx].speed, 0xff, + NULL, 0x0, DEFAULT_TIMEOUT); if (ret != 0x0) { - msg_perr("Command Set SPI Speed 0x%x failed!\n", speed); + msg_perr("Command Set SPI Speed 0x%x failed!\n", spispeeds[spispeed_idx].speed); return 1; } return 0; } -#endif /* Bulk read interface, will read multiple 512 byte chunks aligned to 512 bytes. * @start start address @@ -742,6 +722,28 @@ return millivolt; } +static int dediprog_setup(void) +{ + /* URB 6. Command A. */ + if (dediprog_command_a()) { + return 1; + } + /* URB 7. Command A. */ + if (dediprog_command_a()) { + return 1; + } + /* URB 8. Command Prepare Receive Device String. */ + /* URB 9. Command Receive Device String. */ + if (dediprog_check_devicestring()) { + return 1; + } + /* URB 10. Command C. */ + if (dediprog_command_c()) { + return 1; + } + return 0; +} + static const struct spi_programmer spi_programmer_dediprog = { .type = SPI_CONTROLLER_DEDIPROG, .max_data_read = MAX_DATA_UNSPECIFIED, @@ -783,13 +785,29 @@ int dediprog_init(void) { struct usb_device *dev; - char *voltage, *device; + char *voltage, *device, *spispeed; + int spispeed_idx = 2; int millivolt = 3500; long usedevice = 0; - int ret; + int i, ret; msg_pspew("%s\n", __func__); + spispeed = extract_programmer_param("spispeed"); + if (spispeed) { + for (i = 0; spispeeds[i].name; ++i) { + if (!strcasecmp(spispeeds[i].name, spispeed)) { + spispeed_idx = i; + break; + } + } + if (!spispeeds[i].name) { + msg_perr("Error: Invalid 'spispeed' value.\n"); + free(spispeed); + return 1; + } + free(spispeed); + } voltage = extract_programmer_param("voltage"); if (voltage) { millivolt = parse_voltage(voltage); @@ -852,33 +870,24 @@ return 1; } dediprog_endpoint = 2; - + if (register_shutdown(dediprog_shutdown, NULL)) return 1; dediprog_set_leds(PASS_ON|BUSY_ON|ERROR_ON); - /* URB 6. Command A. */ - if (dediprog_command_a()) { + /* Perform basic setup. */ + if (dediprog_setup()) { dediprog_set_leds(PASS_OFF|BUSY_OFF|ERROR_ON); return 1; } - /* URB 7. Command A. */ - if (dediprog_command_a()) { - dediprog_set_leds(PASS_OFF|BUSY_OFF|ERROR_ON); - return 1; - } - /* URB 8. Command Prepare Receive Device String. */ - /* URB 9. Command Receive Device String. */ - if (dediprog_check_devicestring()) { - dediprog_set_leds(PASS_OFF|BUSY_OFF|ERROR_ON); - return 1; - } - /* URB 10. Command C. */ - if (dediprog_command_c()) { + + /* After setting voltage and speed, perform setup again. */ + if (dediprog_set_spi_voltage(0) || dediprog_set_spi_speed(spispeed_idx) || dediprog_setup()) { dediprog_set_leds(PASS_OFF|BUSY_OFF|ERROR_ON); return 1; } + /* URB 11. Command Set SPI Voltage. */ if (dediprog_set_spi_voltage(millivolt)) { dediprog_set_leds(PASS_OFF|BUSY_OFF|ERROR_ON); Modified: trunk/flashrom.8 ============================================================================== --- trunk/flashrom.8 Mon Feb 4 05:38:42 2013 (r1648) +++ trunk/flashrom.8 Wed Feb 20 19:03:36 2013 (r1649) @@ -676,6 +676,18 @@ Usage example to select the second device: .sp .B " flashrom \-p dediprog:device=1" +.sp +An optional +.B spispeed +parameter specifies the frequency of the SPI bus. Syntax is +.sp +.B " flashrom \-p dediprog:spispeed=frequency" +.sp +where +.B frequency +can be +.BR 375k ", " 750k ", " 1.5M ", " 2.18M ", " 3M ", " 8M ", " 12M " or " 24M +(in Hz). The default is a frequency of 12 MHz. .SS .BR "rayer_spi " programmer The default I/O base address used for the parallel port is 0x378 and you can use

1 0

flashrom can never write if the flash chip isn't found automatically
by 123 123 19 Feb '13

19 Feb '13

1 0

W836xx: 03RG89 Ver FAB1: flashrom -V
by Don Alexander 19 Feb '13

19 Feb '13

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 flashrom -V flashrom v0.9.5.2-r1546 on Linux 3.2.0-4-amd64 (x86_64) flashrom is free software, get the source code at http://www.flashrom.org flashrom was built with libpci 3.1.7, GCC 4.6.2, little endian Command line (1 args): flashrom -V Calibrating delay loop... OS timer resolution is 1 usecs, 3013M loops per second, 10 myus = 10 us, 100 myus = 115 us, 1000 myus = 991 us, 10000 myus = 10179 us, 4 myus = 12 us, OK. Initializing internal programmer No coreboot table found. DMI string system-manufacturer: "Dell Inc. " DMI string system-product-name: "Dell System XPS L702X" DMI string system-version: "Not Specified" DMI string baseboard-manufacturer: "Dell Inc. " DMI string baseboard-product-name: "03RG89" DMI string baseboard-version: "FAB1" DMI string chassis-type: "Portable" Laptop detected via DMI. W836xx enter config mode worked or we were already in config mode. W836xx leave config mode had no effect. Active config mode, unknown reg 0x20 ID: 85. Please send the output of "flashrom -V" to flashrom(a)flashrom.org with W836xx: your board name: flashrom -V as the subject to help us finish support for your Super I/O. Thanks. Found ITE EC, ID 0x8518,Rev 0x02 on port 0x4e. ======================================================================== WARNING! You seem to be running flashrom on an unsupported laptop. Laptops, notebooks and netbooks are difficult to support and we recommend to use the vendor flashing utility. The embedded controller (EC) in these machines often interacts badly with flashing. See http://www.flashrom.org/Laptops for details. If flash is shared with the EC, erase is guaranteed to brick your laptop and write may brick your laptop. Read and probe may irritate your EC and cause fan failure, backlight failure and sudden poweroff. You have been warned. ======================================================================== Aborting. - -- == Don Alexander It's a tough job, but some mug has to do it... RooSoft Ltd -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iEYEARECAAYFAlEi94cACgkQuipFNInZ6evyAACfWpcUW59YjP9xWE8uAttNQvw/ HP0AnRo5JGXUL4+mCVBRea2W861GexUh =H5wR -----END PGP SIGNATURE-----

1 0

Re: [flashrom] Please send a verbose log
by Eric 18 Feb '13

18 Feb '13

Hi, Here is a verbose log for this : http://fr.asus.com/Motherboards/Intel_Socket_1155/P8P67M_PRO -- Si tu as une pomme, que j'ai une pomme, et que l'on échange nos pommes, nous aurons chacun une pomme. Mais si tu as une idée, que j'ai une idée, et que l'on échange nos idées, nous aurons chacun deux idées. La différence entre Linux et Windows ,c'est que Linux c'est comme une belle fille , il faut l'apprivoiser , la comprendre , la séduire tandis que Windows , c'est comme une prostitué , il faut payer et surtout bien se protéger. Pour remédier à ce problème, la solution c'est Ubuntu ! C'est une distribution GNU/Linux qui remplace avantageusement Ouinedoze. Vous êtes libre d'en obtenir une copie pour installation et utilisation en la téléchargeant gratuitement et légalement sur le lien ci-dessous. http://www.ubuntu-fr.org/telechargement

1 0

Re: [flashrom] Bypassing the size check
by Stefan Tauner 15 Feb '13

15 Feb '13

On Thu, 14 Feb 2013 21:55:50 -0800 Jason Craik <jason.c7(a)gmail.com> wrote: > Hello, > > I'm trying to downgrade the bios on my Asus P8Z77-v LK motherboard and am > using a dos version of your program to try and do it. When I try and > reflash with the old .rom file it gives me an error about having two > different size files. I'm wondering if theres a way to just bypass that > check and force write the old .rom file onto the chip. Hello Jason, there is no command line switch to do this and there is no reason to do this in any sane use case either. How did you obtain the "old .rom file"? Obviously not by reading it with flashrom because then it would have the correct size. Also, P8Z77 sounds like a locked down board... please send a verbose log. I assume you have obtained an image of *the BIOS region only* by other means and want to write it to *the BIOS region only*... if you really know what you are doing, please take a look at the documentation of the layout file in the manpage and think about how you want to recover if something goes wrong. ;) -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

flashrom