Commit 6007b1bd authored by Alessandro Rubini's avatar Alessandro Rubini Committed by Greg Kroah-Hartman

FMC: add a driver to write mezzanine EEPROM

This driver allows to reprogram the EEPROM in a mezzanine, to store
its own identifiers during manufacturing or to save other useful data.
Signed-off-by: default avatarAlessandro Rubini <>
Acked-by: default avatarJuan David Gonzalez Cobas <>
Acked-by: default avatarEmilio G. Cota <>
Acked-by: default avatarSamuel Iglesias Gonsalvez <>
Signed-off-by: default avatarGreg Kroah-Hartman <>
parent 056d83f3
......@@ -30,3 +30,6 @@ fmc-fakedev.txt
- about drivers/fmc/fmc-trivial.ko
- about drivers/fmc/fmc-write-eeprom.ko
This module is designed to load a binary file from /lib/firmware and to
write it to the internal EEPROM of the mezzanine card. This driver uses
the `busid' generic parameter.
Overwriting the EEPROM is not something you should do daily, and it is
expected to only happen during manufacturing. For this reason, the
module makes it unlikely for the random user to change a working EEPROM.
The module takes the following measures:
* It accepts a `file=' argument (within /lib/firmware) and if no
such argument is received, it doesn't write anything to EEPROM
(i.e. there is no default file name).
* If the file name ends with `.bin' it is written verbatim starting
at offset 0.
* If the file name ends with `.tlv' it is interpreted as
type-length-value (i.e., it allows writev(2)-like operation).
* If the file name doesn't match any of the patterns above, it is
ignored and no write is performed.
* Only cards listed with `busid=' are written to. If no busid is
specified, no programming is done (and the probe function of the
driver will fail).
Each TLV tuple is formatted in this way: the header is 5 bytes,
followed by data. The first byte is `w' for write, the next two bytes
represent the address, in little-endian byte order, and the next two
represent the data length, in little-endian order. The length does not
include the header (it is the actual number of bytes to be written).
This is a real example: that writes 5 bytes at position 0x110:
spusa.root# od -t x1 -Ax /lib/firmware/try.tlv
000000 77 10 01 05 00 30 31 32 33 34
spusa.root# insmod /tmp/fmc-write-eeprom.ko busid=0x0200 file=try.tlv
[19983.391498] spec 0000:03:00.0: write 5 bytes at 0x0110
[19983.414615] spec 0000:03:00.0: write_eeprom: success
Please note that you'll most likely want to use SDBFS to build your
EEPROM image, at least if your mezzanines are being used in the White
Rabbit environment. For this reason the TLV format is not expected to
be used much and is not expected to be developed further.
If you want to try reflashing fake EEPROM devices, you can use the
fmc-fakedev.ko module (see *note fmc-fakedev::). Whenever you change
the image starting at offset 0, it will deregister and register again
after two seconds. Please note, however, that if fmc-write-eeprom is
still loaded, the system will associate it to the new device, which
will be reprogrammed and thus will be unloaded after two seconds. The
following example removes the module after it reflashed fakedev the
first time.
spusa.root# insmod fmc-fakedev.ko
[ 72.984733] fake-fmc: Manufacturer: fake-vendor
[ 72.989434] fake-fmc: Product name: fake-design-for-testing
spusa.root# insmod fmc-write-eeprom.ko busid=0 file=fdelay-eeprom.bin; \
rmmod fmc-write-eeprom
[ 130.874098] fake-fmc: Matching a generic driver (no ID)
[ 130.887845] fake-fmc: programming 6155 bytes
[ 130.894567] fake-fmc: write_eeprom: success
[ 132.895794] fake-fmc: Manufacturer: CERN
[ 132.899872] fake-fmc: Product name: FmcDelay1ns4cha
Writing to the EEPROM
Once you have created a binary file for your EEPROM, you can write it
to the storage medium using the fmc-write-eeprom (See *note
fmc-write-eeprom::, while relying on a carrier driver. The procedure
here shown here uses the SPEC driver
The example assumes no driver is already loaded (actually, I unloaded
them by hand as everything loads automatically at boot time after you
installed the modules), and shows kernel messages together with
commands. Here the prompt is spusa.root# and two SPEC cards are plugged
in the system.
spusa.root# insmod fmc.ko
spusa.root# insmod spec.ko
[13972.382818] spec 0000:02:00.0: probe for device 0002:0000
[13972.392773] spec 0000:02:00.0: got file "fmc/spec-init.bin", 1484404 (0x16a674) bytes
[13972.591388] spec 0000:02:00.0: FPGA programming successful
[13972.883011] spec 0000:02:00.0: EEPROM has no FRU information
[13972.888719] spec 0000:02:00.0: No device_id filled, using index
[13972.894676] spec 0000:02:00.0: No mezzanine_name found
[13972.899863] /home/rubini/wip/spec-sw/kernel/spec-gpio.c - spec_gpio_init
[13972.906578] spec 0000:04:00.0: probe for device 0004:0000
[13972.916509] spec 0000:04:00.0: got file "fmc/spec-init.bin", 1484404 (0x16a674) bytes
[13973.115096] spec 0000:04:00.0: FPGA programming successful
[13973.401798] spec 0000:04:00.0: EEPROM has no FRU information
[13973.407474] spec 0000:04:00.0: No device_id filled, using index
[13973.413417] spec 0000:04:00.0: No mezzanine_name found
[13973.418600] /home/rubini/wip/spec-sw/kernel/spec-gpio.c - spec_gpio_init
spusa.root# ls /sys/bus/fmc/devices
fmc-0000 fmc-0001
spusa.root# insmod fmc-write-eeprom.ko busid=0x0200 file=fdelay-eeprom.bin
[14103.966259] spec 0000:02:00.0: Matching an generic driver (no ID)
[14103.975519] spec 0000:02:00.0: programming 6155 bytes
[14126.373762] spec 0000:02:00.0: write_eeprom: success
[14126.378770] spec 0000:04:00.0: Matching an generic driver (no ID)
[14126.384903] spec 0000:04:00.0: fmc_write_eeprom: no filename given: not programming
[14126.392600] fmc_write_eeprom: probe of fmc-0001 failed with error -2
Reading back the EEPROM
In order to read back the binary content of the EEPROM of your
mezzanine device, the bus creates a read-only sysfs file called eeprom
for each mezzanine it knows about:
spusa.root# cd /sys/bus/fmc/devices; ls -l */eeprom
-r--r--r-- 1 root root 8192 Apr 9 16:53 FmcDelay1ns4cha-f001/eeprom
-r--r--r-- 1 root root 8192 Apr 9 17:19 fake-design-for-testing-f002/eeprom
-r--r--r-- 1 root root 8192 Apr 9 17:19 fake-design-for-testing-f003/eeprom
-r--r--r-- 1 root root 8192 Apr 9 17:19 fmc-f004/eeprom
......@@ -32,4 +32,12 @@ config FMC_TRIVIAL
The driver also handles interrupts (we used it with a real carrier
before the mezzanines were produced)
tristate "FMC mezzanine driver to write I2C EEPROM"
This driver matches every mezzanine device and can write the
internal EEPROM of the PCB, using the firmware loader to get
its binary and the function carrier->reprogram to actually do it.
It is useful when the mezzanines are produced.
endif # FMC
......@@ -9,3 +9,4 @@ fmc-y += fmc-dump.o
obj-$(CONFIG_FMC_FAKEDEV) += fmc-fakedev.o
obj-$(CONFIG_FMC_TRIVIAL) += fmc-trivial.o
obj-$(CONFIG_FMC_WRITE_EEPROM) += fmc-write-eeprom.o
* Copyright (C) 2012 CERN (
* Author: Alessandro Rubini <>
* Released according to the GNU GPL, version 2 or any later version.
* This work is part of the White Rabbit project, a research effort led
* by CERN, the European Institute for Nuclear Research.
#include <linux/module.h>
#include <linux/string.h>
#include <linux/firmware.h>
#include <linux/init.h>
#include <linux/fmc.h>
#include <asm/unaligned.h>
* This module uses the firmware loader to program the whole or part
* of the FMC eeprom. The meat is in the _run functions. However, no
* default file name is provided, to avoid accidental mishaps. Also,
* you must pass the busid argument
static struct fmc_driver fwe_drv;
/* The "file=" is like the generic "gateware=" used elsewhere */
static char *fwe_file[FMC_MAX_CARDS];
static int fwe_file_n;
module_param_array_named(file, fwe_file, charp, &fwe_file_n, 444);
static int fwe_run_tlv(struct fmc_device *fmc, const struct firmware *fw,
int write)
const uint8_t *p = fw->data;
int len = fw->size;
uint16_t thislen, thisaddr;
int err;
/* format is: 'w' addr16 len16 data... */
while (len > 5) {
thisaddr = get_unaligned_le16(p+1);
thislen = get_unaligned_le16(p+3);
if (p[0] != 'w' || thislen + 5 > len) {
dev_err(&fmc->dev, "invalid tlv at offset %ti\n",
p - fw->data);
return -EINVAL;
err = 0;
if (write) {
dev_info(&fmc->dev, "write %i bytes at 0x%04x\n",
thislen, thisaddr);
err = fmc->op->write_ee(fmc, thisaddr, p + 5, thislen);
if (err < 0) {
dev_err(&fmc->dev, "write failure @0x%04x\n",
return err;
p += 5 + thislen;
len -= 5 + thislen;
if (write)
dev_info(&fmc->dev, "write_eeprom: success\n");
return 0;
static int fwe_run_bin(struct fmc_device *fmc, const struct firmware *fw)
int ret;
dev_info(&fmc->dev, "programming %zi bytes\n", fw->size);
ret = fmc->op->write_ee(fmc, 0, (void *)fw->data, fw->size);
if (ret < 0) {
dev_info(&fmc->dev, "write_eeprom: error %i\n", ret);
return ret;
dev_info(&fmc->dev, "write_eeprom: success\n");
return 0;
static int fwe_run(struct fmc_device *fmc, const struct firmware *fw, char *s)
char *last4 = s + strlen(s) - 4;
int err;
if (!strcmp(last4, ".bin"))
return fwe_run_bin(fmc, fw);
if (!strcmp(last4, ".tlv")) {
err = fwe_run_tlv(fmc, fw, 0);
if (!err)
err = fwe_run_tlv(fmc, fw, 1);
return err;
dev_err(&fmc->dev, "invalid file name \"%s\"\n", s);
return -EINVAL;
* Programming is done at probe time. Morever, only those listed with
* busid= are programmed.
* card is probed for, only one is programmed. Unfortunately, it's
* difficult to know in advance when probing the first card if others
* are there.
int fwe_probe(struct fmc_device *fmc)
int err, index = 0;
const struct firmware *fw;
struct device *dev = &fmc->dev;
char *s;
if (!fwe_drv.busid_n) {
dev_err(dev, "%s: no busid passed, refusing all cards\n",
return -ENODEV;
if (fmc->op->validate)
index = fmc->op->validate(fmc, &fwe_drv);
if (index < 0) {
pr_err("%s: refusing device \"%s\"\n", KBUILD_MODNAME,
return -ENODEV;
if (index >= fwe_file_n) {
pr_err("%s: no filename for device index %i\n",
return -ENODEV;
s = fwe_file[index];
if (!s) {
pr_err("%s: no filename for \"%s\" not programming\n",
KBUILD_MODNAME, dev_name(dev));
return -ENOENT;
err = request_firmware(&fw, s, dev);
if (err < 0) {
dev_err(&fmc->dev, "request firmware \"%s\": error %i\n",
s, err);
return err;
fwe_run(fmc, fw, s);
return 0;
int fwe_remove(struct fmc_device *fmc)
return 0;
static struct fmc_driver fwe_drv = {
.probe = fwe_probe,
.remove = fwe_remove,
/* no table, as the current match just matches everything */
static int fwe_init(void)
int ret;
ret = fmc_driver_register(&fwe_drv);
return ret;
static void fwe_exit(void)
Markdown is supported
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment