Want to Join Us ?

you'll be able to discuss, share and send private messages.

Eboot resigner

Discussion in 'Homebrew' started by Rip Cord, Sep 11, 2014.

Share This Page

  1. Rip Cord

    Administrator Staff Member Admin Developer

    not sure what you meant exactly.

    here is resign_np_free http://techbliss.org/threads/eboot-resigner.585/#post-1775
    resigns np type 3 [free]
    here is resign_rap_to_free:
    http://techbliss.org/threads/eboot-resigner.585/page-3#post-1809
    resigns and converts np with rap to np free.

    these of course won't play on official firmware because of np signature check. this is different from np hash checks and regular signature check.

    for sprx and self that use developer klics it would have to patch the calling functions in the eboot. this is not possible in a simple resigner because the klic used for loading the sprx/self is not always stored in plain text in the eboot. if the developer klic is already known then of course it can be used to resign anyway.

    I probably misunderstood your question.
    regards
     
    Last edited: Dec 27, 2014
    Rudi Rastelli and storm shadow like this.
  2. Rudi Rastelli

    Well-Known Member Developer

    You didn't missunderstand my question... i was just not aware of these resigners. :rolleyes:
    Thx
     
    Last edited: Dec 27, 2014
    Rip Cord likes this.
  3. jo3tomas

    Well-Known Member Developer

    nothing important but if you find some time you could add same option to

    resign_np_free
    resign_rap_to_free
    resign_to_disc

    regards
     
    storm shadow and Rip Cord like this.
  4. Rip Cord

    Administrator Staff Member Admin Developer

    The eboot and other self that was corrupt at compression level 5 had the same problem.
    the section data for the original eboot:
    Code (Text):

    [*] Metadata Section Headers:
    Idx Offset   Size    Type Index Hashed SHA1 Encrypted Key IV Compressed
    000 00000980 01087126 02   00   [YES]  00   [YES]    06  07 [YES]
    001 0108AD40 00202137 02   01   [YES]  08   [YES]    0E  0F [YES]
     
    the section data for the resigned eboot using compression level 5:
    Code (Text):

    Idx Offset   Size    Type Index Hashed SHA1 Encrypted Key IV Compressed
    000 00000980 010B2E3D 02   00   [YES]  00   [YES]    06  07 [YES]
    001 0108AD40 00202137 02   01   [YES]  08   [YES]    0E  0F [YES]
     
    due to lower compression level, the size of section 0 has increased so much that it has overwritten the beginning of section 1 (980 + 10B2E3D) > 108AD40

    resigned at level 6 (same compression level as original games):
    Code (Text):

    Idx Offset   Size    Type Index Hashed SHA1 Encrypted Key IV Compressed
    000 00000980 01087126 02   00   [YES]  00   [YES]    06  07 [YES]
    001 0108AD40 00202137 02   01   [YES]  08   [YES]    0E  0F [YES]
     
    though all that was edited was sdk version. I wouldn't try level 5 unless the section had turned out smaller than I like with level 6.

    I'll probably be able to add the pause for edit in the other tools this next week. Thanks for reporting the problem with the selfs.
     
    jo3tomas likes this.
  5. jo3tomas

    Well-Known Member Developer

    you could remove/disable compression level 5 and leave only 6-9 to prevent issues
     
  6. Rip Cord

    Administrator Staff Member Admin Developer

    here is resign_eboot 0.2.4 private build 2, removes option of compression level 5
    valid range 6-9, default still 6
     

    Attached Files:

    jo3tomas likes this.
  7. Rip Cord

    Administrator Staff Member Admin Developer

    resign_np_free
    resign_rap_to_free
    added pause for edit, compression range 6-9 default 6; same as for resign_eboot in previous post #129
    let me know if anyone needs any different options
     

    Attached Files:

    Me_TheKing, storm shadow and jo3tomas like this.
  8. jo3tomas

    Well-Known Member Developer

    resign_eboot - original files eboot and self/sprx - working, original self spu - working, non original files (encrypted by scetool) random zeroed last part of elf. on 5 eboots in 2 i have that bug

    all tested original untouched files working fine

    i notice, that bug in resign_eboot from post #118 is present with other compression level too, but only if i use non original eboot (encrypted by scetool, not all but 2 tested eboots from 5 have that bug) if i remember correctly scetool skip some sections maybe that cause some issues

    npdrm tools i check later

    regards
     
    Rip Cord likes this.
  9. Rip Cord

    Administrator Staff Member Admin Developer

    Thank you for all your testing; that's really great.
    I never used these tools to resign non-original files. I'll take a look; maybe one here will have the same problem. Thanks.
    regards.
     
  10. jo3tomas

    Well-Known Member Developer

    I don't even notice that earlier, i download clean release and i assume files should be original, but i checked signatures - FAILED, that EBOOT from post #121 is non original too.
     
    Rip Cord likes this.
  11. Rudi Rastelli

    Well-Known Member Developer

    Hi there... i'm working on a new tool to unlock psn games(and include it's updates) and i'm using @RipCord's resigning tools once again.
    So far i can admit all works very very well. Every game i've tried works.

    But now i'm somehow stucked on one (small) problem, which is may be somewhat off-topic, but u guys are so amazing that i hope to find some help here:

    I try to make a PKG that installs a license-edat(for 'C00'-Demos) to exdata-folder. But whatever i try, and i've tried a lot of things, while making the PKG the edat gets corrupted/messed-up somehow. PKG installs fine but the installed edat looks totaly different to the orignial one. Copying manually the original edat to exdata-folder works fine.. no prob.

    But there must be a PKG-way as well... because there are PKGs out there with working license-edats.

    Hope someone here knows about this problem and may be it's solution !?

    Regards
    Rudi
     
    Last edited: Jan 5, 2015
  12. jo3tomas

    Well-Known Member Developer

    you need modded version of make_package_npdrm, atm public there is available only one version which can create license package but there are useless because instead add original edat files from source, tool create new debug edat
     
  13. Rudi Rastelli

    Well-Known Member Developer

    Thx for the anser... even if it's bad news.
    I've found a patcher for "make_package_npdrm_rev1732" but unfortuanally not the original "make_package_npdrm_rev1732" itself.
    So obtaining original "make_package_npdrm_rev1732" and patching it could this be the way to go ?

    Regards
    Rudi
     
  14. jo3tomas

    Well-Known Member Developer

    no, this patch allow only to make licensed package (and regular with finalized files) like i describe above, in newer version make_package_npdrm that option is unlocked but still tool create only debug edat
     
    Rudi Rastelli likes this.
  15. Rudi Rastelli

    Well-Known Member Developer

    @aldostools was so gentle to share an old python script with me. I've changed it a bit and compiled it (using pyinstaller) and here's the result:
    make_package_license (source is included)
    It was developed by @CaptainCPS-X back in 2012 i guess... so credits go to him... THX

    Regards
    Rudi
     
    jo3tomas likes this.
  16. jo3tomas

    Well-Known Member Developer

    thanks, missing only sha1, because that part is zeroed. i compare pkg with original and missing some values between 00000100-0000013F

    pkgs (original and custom - created by make_package_license)
    Code (Text):
    http://www57.zippyshare.com/v/94282789/file.html
    regards

    edit:

    and there's small problem with patch - EDAT(s)-directory (win \ vs unix /)

    must be EDAT/ or "EDAT/" if your folder with edats have space in name

    example EDAT(s)-directory patch:

    EDAT - not working - tool crash
    EDAT\ - working
    "EDAT\" - not working - package is corrupted

    edit2:

    Unknown1 should be 8 not 5

    and

    header size 000000C0 not 00000080
     
    Last edited: Jan 6, 2015
  17. Rudi Rastelli

    Well-Known Member Developer

    thx... i informed @aldostools about that edat-folder-glitch(i noticed this as well, but it's easy to workaround)... guess he will look at it.

    Regarding the missing values: Created packages install fine here, so this might be a minor problem isn't it !?

    Regards
    Rudi
     
  18. jo3tomas

    Well-Known Member Developer

    i'm not sure what is included in 00000110-0000011F rest of missing values probably are constant but i checked only 2 pkgs

    edit:

    i checked older license pkg (revision 1500) and i notice there is only half these missing values (constant part) and Unknown1 and header size depends by revision
     
    Last edited: Jan 6, 2015
  19. Rip Cord

    Administrator Staff Member Admin Developer

    here is resign_to_disc 0.1.4 private build 1
    added pause for manual edits of decrypted section, enabled decrypting section of spu and sprx (was disabled before because they don't have sysprocparam)
     

    Attached Files:

    Me_TheKing and jo3tomas like this.
  20. jo3tomas

    Well-Known Member Developer

    4 updates tested and seems too everything working fine

    regards and thanks
     
    Rip Cord likes this.
Top