The battle againt the BMC – Part 2

Update 4/11/2011: I managed to find out some more info about the packaging scheme Dell uses for their BMC firmware files. I deciphered most of the container format. I am in the process of testing modifications right now, but for the moment I updated the version of the tool below with a new version. You can also download the program directly here: dell-extract-bmc-firmware.tar.gz.

Firmware header

Deciphering the firmware header.

As mentioned earlier, I started to look into hacking the BMC firmware in order to solve my problem of the hard-coded failure thresholds of my PowerEdge 2800.

I had a look into the firmware flash file, and noticed that it seems to consist of several files (as usual for BIOS/firmwares). As this might increase my chances not to brick my BMC, I decided to I separate the individual files for starters. I couldn’t find a program that does that (the firmware tools of the Dell linux community are closed-source, unfortunately), so I grabbed a hex-editor and deciphered (more or less) the firmware’s header. Here’s the corresponding C program:

// vim: ts=4 ai noexpandtab nopaste
 * This program can extract and check the different files contained in a firmware file
 * for a Dell PowerEdge BMC.

#include <stdlib.h>
#include <stdio.h>
#include <stdint.h>
#include <string.h>
#include <unistd.h>
#include <arpa/inet.h>

typedef struct
    uint8_t     hex02;
    uint8_t     numBlocks;  // number of subfiles in system
    uint32_t    filesize;
    uint16_t    zero;
    char        dellHeaderStr[9];
} header_t;

typedef struct
    uint8_t     zero1;
    uint8_t     type;       // 0x000b -> SD_${system}.FLC
    uint8_t     zero2;
    uint8_t     system;     // 0, 1, 2
    uint8_t     zeros[3];
    uint16_t    unknownFixedData;
    uint16_t    crc16;
    uint32_t    length;
    uint32_t    offset;
    char        filename[32];
} flc_block_t;
// 4x1+3x1+2+2+4+4+32=51

uint16_t endian_swap16(uint16_t x)
    return (x>>8) |

uint32_t endian_swap32(uint32_t x)
    return (x>>24) |
            ((x<<8) & 0x00FF0000) |
            ((x>>8) & 0x0000FF00) |

/** CRC table for the CRC-16. The poly is 0x8005 (x^16 + x^15 + x^2 + 1) */
uint16_t const crc16_table[256] = {
        0x0000, 0xC0C1, 0xC181, 0x0140, 0xC301, 0x03C0, 0x0280, 0xC241,
        0xC601, 0x06C0, 0x0780, 0xC741, 0x0500, 0xC5C1, 0xC481, 0x0440,
        0xCC01, 0x0CC0, 0x0D80, 0xCD41, 0x0F00, 0xCFC1, 0xCE81, 0x0E40,
        0x0A00, 0xCAC1, 0xCB81, 0x0B40, 0xC901, 0x09C0, 0x0880, 0xC841,
        0xD801, 0x18C0, 0x1980, 0xD941, 0x1B00, 0xDBC1, 0xDA81, 0x1A40,
        0x1E00, 0xDEC1, 0xDF81, 0x1F40, 0xDD01, 0x1DC0, 0x1C80, 0xDC41,
        0x1400, 0xD4C1, 0xD581, 0x1540, 0xD701, 0x17C0, 0x1680, 0xD641,
        0xD201, 0x12C0, 0x1380, 0xD341, 0x1100, 0xD1C1, 0xD081, 0x1040,
        0xF001, 0x30C0, 0x3180, 0xF141, 0x3300, 0xF3C1, 0xF281, 0x3240,
        0x3600, 0xF6C1, 0xF781, 0x3740, 0xF501, 0x35C0, 0x3480, 0xF441,
        0x3C00, 0xFCC1, 0xFD81, 0x3D40, 0xFF01, 0x3FC0, 0x3E80, 0xFE41,
        0xFA01, 0x3AC0, 0x3B80, 0xFB41, 0x3900, 0xF9C1, 0xF881, 0x3840,
        0x2800, 0xE8C1, 0xE981, 0x2940, 0xEB01, 0x2BC0, 0x2A80, 0xEA41,
        0xEE01, 0x2EC0, 0x2F80, 0xEF41, 0x2D00, 0xEDC1, 0xEC81, 0x2C40,
        0xE401, 0x24C0, 0x2580, 0xE541, 0x2700, 0xE7C1, 0xE681, 0x2640,
        0x2200, 0xE2C1, 0xE381, 0x2340, 0xE101, 0x21C0, 0x2080, 0xE041,
        0xA001, 0x60C0, 0x6180, 0xA141, 0x6300, 0xA3C1, 0xA281, 0x6240,
        0x6600, 0xA6C1, 0xA781, 0x6740, 0xA501, 0x65C0, 0x6480, 0xA441,
        0x6C00, 0xACC1, 0xAD81, 0x6D40, 0xAF01, 0x6FC0, 0x6E80, 0xAE41,
        0xAA01, 0x6AC0, 0x6B80, 0xAB41, 0x6900, 0xA9C1, 0xA881, 0x6840,
        0x7800, 0xB8C1, 0xB981, 0x7940, 0xBB01, 0x7BC0, 0x7A80, 0xBA41,
        0xBE01, 0x7EC0, 0x7F80, 0xBF41, 0x7D00, 0xBDC1, 0xBC81, 0x7C40,
        0xB401, 0x74C0, 0x7580, 0xB541, 0x7700, 0xB7C1, 0xB681, 0x7640,
        0x7200, 0xB2C1, 0xB381, 0x7340, 0xB101, 0x71C0, 0x7080, 0xB041,
        0x5000, 0x90C1, 0x9181, 0x5140, 0x9301, 0x53C0, 0x5280, 0x9241,
        0x9601, 0x56C0, 0x5780, 0x9741, 0x5500, 0x95C1, 0x9481, 0x5440,
        0x9C01, 0x5CC0, 0x5D80, 0x9D41, 0x5F00, 0x9FC1, 0x9E81, 0x5E40,
        0x5A00, 0x9AC1, 0x9B81, 0x5B40, 0x9901, 0x59C0, 0x5880, 0x9841,
        0x8801, 0x48C0, 0x4980, 0x8941, 0x4B00, 0x8BC1, 0x8A81, 0x4A40,
        0x4E00, 0x8EC1, 0x8F81, 0x4F40, 0x8D01, 0x4DC0, 0x4C80, 0x8C41,
        0x4400, 0x84C1, 0x8581, 0x4540, 0x8701, 0x47C0, 0x4680, 0x8641,
        0x8201, 0x42C0, 0x4380, 0x8341, 0x4100, 0x81C1, 0x8081, 0x4040

static inline uint16_t crc16_byte(uint16_t crc, const uint8_t data)
    return (crc >> 8) ^ crc16_table[(crc ^ data) & 0xff];

uint16_t calccrc16(uint8_t const *buffer, size_t len)
    uint16_t crc = 0x0000;

    while (len--)
        crc = crc16_byte(crc, *buffer++);
    return crc;

int main(int argc, char *argv[])
    if (argc != 2)
        fprintf(stderr, "Usage: %s <firmware>\n", argv[0]);
    FILE* flashFile = fopen(argv[1], "r");

    // get filesize
    fseek(flashFile, 0, SEEK_END);
    uint32_t filesize = ftell(flashFile);
    fseek(flashFile, 0, SEEK_SET);

    // read the header
    header_t header;
    if (fread(&header, sizeof(header_t), 1, flashFile) == 0)
        fprintf(stderr, "Error: Can't read header.\n");

    // check that it's a valid header as far as we know
    if (header.hex02 != 0x02 || != 0 ||
        filesize != header.filesize ||
        strncmp(header.dellHeaderStr, "DELL_INC", 8) != 0)
        fprintf(stderr, "Error: Header not valid.\n");

    // calculate header crc
    fseek(flashFile, 0, 0);
    uint16_t totalHeaderSize = sizeof(flc_block_t) * header.numBlocks + sizeof(header_t);
    uint8_t headerBuf[totalHeaderSize];
    fread(&headerBuf, totalHeaderSize, 1, flashFile);
    uint16_t headerCRC16 = calccrc16(headerBuf, totalHeaderSize);

    // calculate total file crc
    fseek(flashFile, 0, 0);
    uint8_t fileBuf[header.filesize-2];
    fread(&fileBuf, header.filesize-2, 1, flashFile);
    uint16_t fileCRC16 = calccrc16(fileBuf, header.filesize-2);
    uint16_t fileCRC16Dell;
    fread(&fileCRC16Dell, 2, 1, flashFile);
    printf("Valid Dell PowerEdge BMC firmware header found:\n\n");
    printf("  - number of blocks : %d\n",   header.numBlocks);
    printf("  - oemstr (fixed)   : %s\n",   header.dellHeaderStr);
    printf("  - total file size  : %d\n",   header.filesize);
    printf("  - total header size: %d\n",   totalHeaderSize);
    printf("  - header CRC16     : 0x%04x\n",   headerCRC16);
    printf("  - total file CRC16 : 0x%04x\n\n", fileCRC16);
    if (fileCRC16 == fileCRC16Dell)
       printf("  * CRC16 check OK\n");
       printf("  * CRC16 check FAILED, actual CRC16 is 0x%04x instead of 0x%04x\n", fileCRC16, fileCRC16Dell);


    // read all blocks
    fseek(flashFile, sizeof(header_t), 0);
    flc_block_t flcBlock[header.numBlocks];
    fread(&flcBlock, sizeof(flc_block_t), header.numBlocks, flashFile);

    uint8_t i;
    for (i = 0; i < header.numBlocks; i++)
        // check if our understanding of format is correct
        if (flcBlock[i].zero1 != 0 || flcBlock[i].zero2 != 0 || flcBlock[i].zeros[0] != 0 ||
            flcBlock[i].zeros[1] != 0 || flcBlock[i].zeros[2] != 0)
            fprintf(stderr, "Error: Block %d not valid.\n", i);

        printf("Block %d:\n\n", i);
        printf("  - type     : %d/0x%02x (defines block type, 0x0b is sensor data table)\n", flcBlock[i].type, flcBlock[i].type);
        printf("  - system # : %d/0x%02x (running number for systems in this firmware file)\n", flcBlock[i].system, flcBlock[i].system);
        printf("  - unknown  : %d/0x%04x (always same for all blocks in a single firmware file)\n", flcBlock[i].unknownFixedData, flcBlock[i].unknownFixedData);
        printf("  - offset   : %d\n", flcBlock[i].offset);
        printf("  - length   : %d\n", flcBlock[i].length);
        printf("  - filename : %s\n\n", flcBlock[i].filename);

        // extract the block according to the offset and length given in the block desc.
        printf("  * extracting block...");
        char* blockData = (char*) malloc(flcBlock[i].length);

        fseek(flashFile, flcBlock[i].offset, 0);
        fread(blockData, flcBlock[i].length, 1, flashFile);

        FILE* blockFile = fopen(flcBlock[i].filename, "w");
        fwrite(blockData, flcBlock[i].length, 1, blockFile);
        uint16_t blockCRC16 = calccrc16(blockData, flcBlock[i].length);

        if (blockCRC16 == flcBlock[i].crc16)
          printf("  * CRC16 check OK\n");
          printf("  * CRC16 check FAILED, actual CRC16 is 0x%04x instead of 0x%04x\n", blockCRC16, flcBlock[i].crc16);



The names of the individual files are listed below. They are organized in blocks (that’s what I call them), and apparently by function. Get the latest BMC firmware (30/6/2009, v1.83, A10) and apply my program to retrieve the individual files.

  • block 0 (code, big files):
    • BB.FLC
    • OB.FLC
    • ID.FLC
  • block 1 (*_BB files):
    • SD_BB.FLC
    • FI_BB.FLC
    • TOC_BB.FLC
    • IO_BB.FLC
    • IS_BB.FLC
    • OEM_BB.FLC
  • block 2 (*_K_C files):
    • SD_K_C.FLC
    • FI_K_C.FLC
    • TOC_K_C.FLC
    • IO_K_C.FLC
    • IS_K_C.FLC
    • OEM_K_C.FLC

The BMC seems to be little-endian (makes only sense I guess). I’ve scanned the different files for appearances of the threshold values (900 and 2025/0x07e9 and 0x0384 in int and 0x6144000 in float). No avail. Darn. Either I am doing something wrong or the thesholds are not hard-coded in the firmware (I had my hopes up when I saw the OEM_DEF.FLC file, which actually contains the default BMC password and the like). Maybe the thresholds are stored in the configuration flash after all – only how can we access it?


I finally managed to adjust the critical fan thresholds by patching the BMC firmware! Here’s the howto. Additionally, I created a project page for my server.

4 thoughts on “The battle againt the BMC – Part 2

    • Thanks! Quite motivating a comment like this… I trashed the firmware project as I figured it might be too much work–but then, if it would be of use to other people as well, I might well invest the time, so I took it up again. I can’t promise anything though, the stuff’s rather obscure (I don’t even know the processor architecture of the BMC).

      Another option would be to design a hardware filter (a microcontroller that monitors the fans and supplies the right signals to the BMC). Would you also be interested in a solution like this? It would certainly be less dangerous and more flexible, but would require and additional board in the server (and more cables)…

  1. Pingback: How-to adjust the fan thresholds of a Dell PowerEdge | tüfteln

Leave a Reply

Your email address will not be published. Required fields are marked *