Join Us and become a Member for a Verified Badge to access private areas with the latest PS4 PKGs.
Status
Not open for further replies.
Previously we've seen guides on How to Obtain Your PS4 SFlash alongside some example file dumps, and today developer @zecoxao shared on Twitter SFlash0Unpack which is a script to unpack sflash0 files from PS4 flash dumps for PlayStation 4 scene devs to examine followed by a PS4 SFlash0 Tool by SocraticBliss.

In related news, pearlxcore (Twitter) also made available a PS4 Dump Extractor utilized to extract PlayStation 4 dumps including sflash0 files. :tup:

Download: sflash0unpack-master.zip / GIT / ps4_sflash0_pack_tool.py / GIT / ps4_sflash0_tool.py / ps4_sflash0_tool / PS4.Dump.Extractor.exe / PS4 Dump Extractor GIT

Below is main.c from Github followed by the makefile, with a Python version of the PS4 Sflash0 Pack Tool from @SocraticBliss (Twitter) on Github for those interested and also a extractor.pl Perl version from BwE! :ninja:

Main.c:
Code:
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include <inttypes.h>
#include <sys/stat.h>

/*
   0x0 <- Header (0x1000)
   0x1000 <- Unk (0x1000)
   0x2000 <- MBR1 (0x1000) (for sflash0s1.cryptx3b)
   0x3000 <- MBR2 (0x1000) (for sflash0s1.cryptx3)
   0x4000 <- sflash0s0x32b (emc_ipl) (0x60000)
   0x64000 <- sflash0s0x32 (emc_ipl) (0x60000)
   0xC4000 <- sflash0s0x33 (eap_kbl) (0x80000)
   0x144000 <- sflash0s0x34 (wifi fw) (0x80000)
   0x1C4000 <- sflash0s0x38 (nvs) (0xC000)
   0x1D0000 <- sflash0s0x0 (blank1) (0x30000)
   0x200000 <- Header2 (0x1000)
   0x201000 <- Unk 2(0x1000)
   0x202000 <- MBR3(0x1000) (for sflash0s1.cryptx2b)
   0x203000 <- MBR4(0x1000) (for sflash0s1.cryptx2)
   0x204000 <- sflash0s1.cryptx2b (sam_ipl/secure loader) (0x3E000)
   0x242000 <- sflash0s1.cryptx2 (sam_ipl/secure loader) (0x3E000)
   0x280000 <- sflash0s1.cryptx1 (idata) (0x80000)
   0x300000 <- sflash0s1.cryptx39 (bd_hrl?) (0x80000)
   0x380000 <- sflash0s1.cryptx6 (Virtual TRM) (0x40000)
   0x3C0000 <- sflash0s1.cryptx3b (secure kernel, secure modules) (0xCC0000)
   0x1080000 <- sflash0s1.cryptx3 (secure kernel, secure modules) (0xCC0000)
   0x1D40000 <- sflash0s1.cryptx40 (blank2) (0x2C0000)
*/

typedef struct
{
   unsigned char header[0x1000];
   unsigned char unk[0x1000];
   unsigned char mbr1[0x1000];
   unsigned char mbr2[0x1000];
   unsigned char emc_iplb[0x60000];
   unsigned char emc_ipl[0x60000];
   unsigned char eap_kbl[0x80000];
   unsigned char wifi_fw[0x80000];
   unsigned char nvs[0xC000];
   unsigned char blank[0x30000];
   unsigned char header2[0x1000];
   unsigned char unk2[0x1000];
   unsigned char mbr3[0x1000];
   unsigned char mbr4[0x1000];
   unsigned char sam_iplb[0x3E000];
   unsigned char sam_ipl[0x3E000];
   unsigned char idata[0x80000];
   unsigned char bd_hrl[0x80000];
   unsigned char vtrm[0x40000];
   unsigned char secureb[0xCC0000];
   unsigned char secure[0xCC0000];
   unsigned char blank2[0x2C0000];
} SFLASH0;

int main(int argc, char **argv){

   if(argc != 3){
       printf ("\nusage: sflash0unpack [sflash0] [outdir]  \n");
       return -1;
   }

   unsigned char out[256];

   mkdir(argv[2],0777);

   FILE *fp = fopen(argv[1],"rb");

   unsigned char *buf = (unsigned char*) malloc (0x2000000);

   fread(buf,0x2000000,1,fp);

   SFLASH0* entries = (SFLASH0*)buf;

   sprintf(out,"%s/header.bin",argv[2]);

   FILE *fl = fopen(out,"wb");

   fwrite(entries->header,sizeof(entries->header),1,fl);

   fclose(fl);

   sprintf(out,"%s/emc_ipl.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->emc_ipl,sizeof(entries->emc_ipl),1,fl);

   fclose(fl);

   sprintf(out,"%s/emc_iplb.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->emc_iplb,sizeof(entries->emc_iplb),1,fl);

   fclose(fl);

   sprintf(out,"%s/eap_kbl.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->eap_kbl,sizeof(entries->eap_kbl),1,fl);

   fclose(fl);

   sprintf(out,"%s/wifi_fw.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->wifi_fw,sizeof(entries->wifi_fw),1,fl);

   fclose(fl);

   sprintf(out,"%s/sam_ipl.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->sam_ipl,sizeof(entries->sam_ipl),1,fl);

   fclose(fl);

   sprintf(out,"%s/sam_iplb.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->sam_iplb,sizeof(entries->sam_iplb),1,fl);

   fclose(fl);

   sprintf(out,"%s/idata.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->idata,sizeof(entries->idata),1,fl);

   fclose(fl);

   sprintf(out,"%s/bd_hrl.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->bd_hrl,sizeof(entries->bd_hrl),1,fl);

   fclose(fl);

   sprintf(out,"%s/vtrm.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->vtrm,sizeof(entries->vtrm),1,fl);

   fclose(fl);

   sprintf(out,"%s/secure.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->secure,sizeof(entries->secure),1,fl);

   fclose(fl);

   sprintf(out,"%s/secureb.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->secureb,sizeof(entries->secureb),1,fl);

   fclose(fl);

   sprintf(out,"%s/blank.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->blank,sizeof(entries->blank),1,fl);

   fclose(fl);

   sprintf(out,"%s/nvs.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->nvs,sizeof(entries->nvs),1,fl);

   fclose(fl);

   sprintf(out,"%s/unk.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->unk,sizeof(entries->unk),1,fl);

   fclose(fl);

   sprintf(out,"%s/mbr1.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->mbr1,sizeof(entries->mbr1),1,fl);

   fclose(fl);

   sprintf(out,"%s/mbr2.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->mbr2,sizeof(entries->mbr2),1,fl);

   fclose(fl);

   sprintf(out,"%s/unk2.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->unk2,sizeof(entries->unk2),1,fl);

   fclose(fl);

   sprintf(out,"%s/mbr3.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->mbr3,sizeof(entries->mbr3),1,fl);

   fclose(fl);

   sprintf(out,"%s/mbr4.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->mbr4,sizeof(entries->mbr4),1,fl);

   fclose(fl);

   sprintf(out,"%s/header2.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->header2,sizeof(entries->header2),1,fl);

   fclose(fl);

   sprintf(out,"%s/blank2.bin",argv[2]);

   fl = fopen(out,"wb");

   fwrite(entries->blank2,sizeof(entries->blank2),1,fl);

   fclose(fl);

   fclose(fp);


   return 0;
}
:idea: To find out (and document) the minimum version of your PS4, download with ftp buf 0x200 payload the file /dev/sflash0s1.cryptx1 (size 512KB) at 0x10008 you can find the minver (endian swapped) next to the SMI header
ps4_sflash0_pack_tool.py
Code:
# PS4 Sflash0 Pack Tool
# SocraticBliss (R)
# Thanks to zecoxao <3

import os
import sys

'''
   Offsets

   0x0       <- Header (0x1000)
   0x1000    <- Unk    (0x1000)
   0x2000    <- MBR1   (for sflash0s1.cryptx3b) (0x1000)
   0x3000    <- MBR2   (for sflash0s1.cryptx3) (0x1000)
   0x4000    <- sflash0s0x32b (emc_ipl) (0x60000)
   0x64000   <- sflash0s0x32  (emc_ipl) (0x60000)
   0xC4000   <- sflash0s0x33  (eap_kbl) (0x80000)
   0x144000  <- sflash0s0x34  (wifi fw) (0x80000)
   0x1C4000  <- sflash0s0x38  (nvs) (0xC000)
   0x1D0000  <- sflash0s0x0   (blank) (0x30000)
   0x200000  <- Header2 (0x1000)
   0x201000  <- Unk2    (0x1000)
   0x202000  <- MBR3    (for sflash0s1.cryptx2b) (0x1000)
   0x203000  <- MBR4    (for sflash0s1.cryptx2) (0x1000)
   0x204000  <- sflash0s1.cryptx2b (sam_ipl/secure loader) (0x3E000)
   0x242000  <- sflash0s1.cryptx2  (sam_ipl/secure loader) (0x3E000)
   0x280000  <- sflash0s1.cryptx1  (idata) (0x80000)
   0x300000  <- sflash0s1.cryptx39 (bd_hrl?) (0x80000)
   0x380000  <- sflash0s1.cryptx6  (Virtual TRM) (0x40000)
   0x3C0000  <- sflash0s1.cryptx3b (secure kernel, secure modules) (0xCC0000)
   0x1080000 <- sflash0s1.cryptx3  (secure kernel, secure modules) (0xCC0000)
   0x1D40000 <- sflash0s1.cryptx40 (blank2) (0x2C0000)
'''

SFLASH0 = [
   ('header.bin',   0x0,       0x1000),
   ('unknown.bin',  0x1000,    0x1000),
   ('mbr1.bin',     0x2000,    0x1000),
   ('mbr2.bin',     0x3000,    0x1000),
   ('emc_iplb.bin', 0x4000,    0x60000),
   ('emc_ipl.bin',  0x64000,   0x60000),
   ('eap_kbl.bin',  0xC4000,   0x80000),
   ('wifi_fw.bin',  0x144000,  0x80000),
   ('nvs.bin',      0x1C4000,  0xC000),
   ('blank.bin',    0x1D0000,  0x30000),
   ('header2.bin',  0x200000,  0x1000),
   ('unknown2.bin', 0x201000,  0x1000),
   ('mbr3.bin',     0x202000,  0x1000),
   ('mbr4.bin',     0x203000,  0x1000),
   ('sam_iplb.bin', 0x204000,  0x3E000),
   ('sam_ipl.bin',  0x242000,  0x3E000),
   ('idata.bin',    0x280000,  0x80000),
   ('bd_hrl.bin',   0x300000,  0x80000),
   ('vtrm.bin',     0x380000,  0x40000),
   ('secureb.bin',  0x3C0000,  0xCC0000),
   ('secure.bin',   0x1080000, 0xCC0000),
   ('blank2.bin',   0x1D40000, 0x2C0000),
]

# Unpack entries from a Sflash0 binary...
def unpack(file, dir):

   with open(file, 'rb') as input:
       sflash0 = input.read()

       # Validate input file...
       if sflash0[:0x4] != 'SONY':
           raise SystemExit('\nInvalid PS4 Sflash0 binary!')

       for num, entry in enumerate(SFLASH0):
           with open('%s/%s' % (dir, SFLASH0[num][0]), 'wb') as output:
               begin = SFLASH0[num][1]
               end = begin + SFLASH0[num][2]
  
               output.write(sflash0[begin:end])
               print('Unpacked %s' % SFLASH0[num][0])

# Pack entries into a Sflash0 binary...
def pack(dir, file):

   with open(file, 'wb') as output:
       try:
           for num, entry in enumerate(SFLASH0):
               with open('%s/%s' % (dir, SFLASH0[num][0]), 'rb') as input:
                   output.write(input.read())

       except IOError as error:
           raise SystemExit('\n%s' % error)


def main(argc, argv):

   # Print Usage Statement...
   if argc not in [2, 3]:
       raise SystemExit('\nUsage: python %s <input> [output]' % argv[0])

   # File Input -> Unpack
   if os.path.isfile(argv[1]):

       # Create a custom directory...
       if argc == 3:
           try:
               os.makedirs(argv[2])
           except:
               pass

       unpack(argv[1], argv[2] if argc == 3 else '.')

   # Directory Input -> Pack
   elif os.path.isdir(argv[1]):
       pack(argv[1], argv[2] if argc == 3 else 'sflash0.bin')

   else:
       raise SystemExit('\nUsage: python %s <input> [output]' % argv[0])

   print('\nDone!')

if __name__ == '__main__':
   main(len(sys.argv), sys.argv)
And From the ps4_sflash0_tool README.md: PS4 SFlash0 Tool

SocraticBliss (R)

ps4_sflash_tool.py: Python script for [un]packing your PS4's SFlash0

Usage

Unpacking SFlash0

Code:
python ps4_sflash_pack_tool.py <sflash0.bin> [Optional Output Directory]
Packing SFlash0
Code:
python ps4_sflash_pack_tool.py <Directory> [Optional Output SFlash0 Name]
If you have any suggestions or ideas, please feel free to create pull requests!
To make the most out of this, we have to work together! :lovewins:
decided to release a tool to unpack the sflash0 plaintext partition and the ciphertext counterpart you obtain from PS4 called ps4norunpack. you can find it here:
  • ps4norextract.7z (1.58 KB) / PS4NORUnpack GIT - PS4 NOR Unpack parses sflash0/sflash0s0 and sflash0s1.crypt/sflash0s1 (with 0x4000 bytes header from sflash0s1.crypt)
PS4 Dump Extractor.png

small python program that calculates the entropy of the file "sflash0":
One could adapt this to use direct data from sflash0 instead of the whole flash as a means of validating it. i'm too lazy to do this so i'll let you guys use your brains.
Code:
#!/usr/bin/env python3

import math

def entropy(filename):
    with open(filename, "rb") as file:
        counters = {byte: 0 for byte in range(2 ** 8)}  # start all counters with zeros

        for byte in file.read():  # read in chunks for large files
            counters[byte] += 1  # increase counter for specified byte

        filesize = file.tell()  # we can get file size by reading current position

        probabilities = [counter / filesize for counter in counters.values()]  # calculate probabilities for each byte

        entropy = -sum(probability * math.log2(probability) for probability in probabilities if probability > 0)  # final sum

        print(entropy)
      
entropy("sflash0")
for those curious about why the revert requires the values at 0x201000 to be of a specific type. the logic behind it is that the bank type is stored there. Sony checks for a 00 or 80 in the first byte for bank 00 or 01 respectively
the reason why option 7 worked was because, when decrypting those values, the first byte happened to match 00 or 80, thus making the console believe it was on firmware 9.00 (which was located in backup bank)
fun fact, all those 8 options could have not worked. there is a 1/128 or 2/256 possibility of the person changing the values gets it right, because one byte has 256 possibilities and we're looking for one out of two
yes. you ftp from your hacked console the file sflash0s1.crypt
you can find the previous firmware *** version in COREOS

SFlash0Unpack Unpack SFlash0 Files From PS4 Flash Dumps by Zecoxao.jpg
 

Comments

It's not reprogramming a bios it's writing a reverse engineered bios with patches to allow for downgrade or write original bios back to fix a corrupted bios. As long as you have a bios dump of your system saved
 
@Chaos Kid
To allow for downgrade? Are you sure? For layout we mean the exact number of files that make up the dump of the Nor chip, their names, the dimensions and the offsets of their exact position in the dump.

For layout we mean the exact number of files that make up the dump of Nor, their names, the dimensions and the offsets of their exact position in the dump, for now they are useless information for normal users but they are proof of the fact that contrary to what someone say, devs are always at work, what normal users could do to help, would be to dump the Nor and share it with BwE to improve its validator in order to identify many more possible corruptions that often they make it unusable our consoles.
 
@XfactorX

Yes I am sure what can be done with these when things are R.E it's from my experience in the past and knowing how it works. Just member dumps are good when validated correctly but never share all the dump they contain personal info that could be bad when shared.

Dumps in the ps3 scene were used to flash back and allow for downgrade when the correct patch is applied to the dump itself with offsets and using a jb device to put in jig mode. While I get what your saying about correct sizes people don't always understand what a bios dump could be used for which is what I explained for on page one so people are aware.
 
@Chaos Kid

Undoubtedly, I must have expressed myself badly, so I apologize for my bad english. The part related to you was just my question about the downgrade because as you wrote it seemed that we have already come to understand how to apply the patches to perform the downgrade.

All the rest of what I wrote, apart from the question, was referred to those who ask what it was zecoxao announcement. Returning to you, I've known you for many years and I read what you write in the various places, so I assure you that I did not want to question your knowledge, in fact, when I read what you wrote on the downgrade, I had a stroke.

Regarding the personal data contained in the Nor chip, it must be said that the MAC and serial numbers can be removed, and I believe that those who are able to do the dump also has the ability to remove those data.

As far as IDPS is concerned, there is no danger since it is not present in NOR.
 
@XfactorX

At this time there is no downgrade features available but the intention is there for downgrade it's just understanding how it works and there may be possible someone who already has these patches ready for it just has not released them for the fact of 6.20 release as the system fw has loop holes in it and is pointless to release now when a few months something pops up.

I wasn't doubting your skills either as I knew what you were getting at with zecoaxo and needing all requirements but those aren't all that will be needed to keep a valid flash dump and for flashing back.

Even an analyzer on the flash can give you tons of data on fw update where you can see everything that's going on through the line but I'm glad to see your helping others.
 
Status
Not open for further replies.
Back
Top