Bug 87071

Summary: [clover/redwood] John the Ripper: LLVM triggered Diagnostic Handler: unsupported call to function S2KItSaltedSHA1Generator in gpg
Product: Mesa Reporter: Dennis Schridde <devurandom>
Component: Drivers/Gallium/r600Assignee: Default DRI bug account <dri-devel>
Status: RESOLVED MOVED QA Contact:
Severity: normal    
Priority: medium    
Version: 10.3   
Hardware: x86-64 (AMD64)   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:
Bug Depends on:    
Bug Blocks: 99553    

Description Dennis Schridde 2014-12-07 13:03:05 UTC
Mesa 10.3.4 / LLVM 3.5.0 / libclc 0.0.1_pre20141027 seems to be unable to compile the OpenCL function S2KItSaltedSHA1Generator from current John-The-Ripper Git [1].

The function is too long to quote here, but the signature is:
inline void S2KItSaltedSHA1Generator(__global const uchar *password, int password_length, __global const uchar *salt, int count, __global uchar *key, int length)

My hardware as reported by JTR is:
# /usr/sbin/john -list=opencl-devices
Platform #0 name: Default
Platform version: OpenCL 1.1 MESA 10.3.4
        Device #0 (0) name:     AMD REDWOOD
        Device vendor:          X.Org
        Device type:            GPU (LE)
        Device version:         OpenCL 1.1 MESA 10.3.4
        Driver version:         10.3.4
        Native vector widths:   char 16, short 8, int 4, long 2
        Preferred vector width: char 16, short 8, int 4, long 2
        Global Memory:          1024.0 MB
        Local Memory:           32.0 KB (Local)
        Max memory alloc. size: 256.2 MB
        Max clock (MHz):        700
        Max Work Group Size:    256
        Parallel compute cores: 5

[1] https://github.com/magnumripper/JohnTheRipper/blob/bleeding-jumbo/src/opencl/gpg_kernel.cl#L369
Comment 1 Dennis Schridde 2014-12-07 13:04:44 UTC
P.S: John's complete output is:
# /usr/sbin/john --format=gpg-opencl […]
Device 0: AMD REDWOOD
Build log: �
Local worksize (LWS) 64, Global worksize (GWS) 8192
Loaded 1 password hash (gpg-opencl, OpenPGP / GnuPG Secret Key [SHA1 OpenCL])
Will run 2 OpenMP threads
LLVM triggered Diagnostic Handler: unsupported call to function S2KItSaltedSHA1Generator in gpg
Segmentation fault
Comment 2 Wade Cline 2016-08-25 02:16:01 UTC
I've also run into this issue, although a workaround exists (see: https://github.com/magnumripper/JohnTheRipper/issues/2204).  The problem appears to be with the declaration of 'inline'; removing the 'inline' declaration fixes the compilation issue.

Hardware/software information as follows is:
frostsnow@seneca ~/software/JohnTheRipper/run $ ./john --list=OpenCL-devices
Platform #0 name: Clover, version: OpenCL 1.1 Mesa 12.0.1
    Device #0 (0) name:     AMD HAWAII (DRM 2.45.0 / 4.7.0, LLVM 3.8.1)
    Device vendor:          AMD
    Device type:            GPU (LE)
    Device version:         OpenCL 1.1 Mesa 12.0.1
    Driver version:         12.0.1 - Catalyst  
    Native vector widths:   char 16, short 8, int 4, long 2
    Preferred vector width: char 16, short 8, int 4, long 2
    Global Memory:          1024.0 MB
    Local Memory:           32.0 KB (Local)
    Max memory alloc. size: 256.2 MB
    Max clock (MHz):        1040
    Max Work Group Size:    256
    Parallel compute cores: 44
    Stream processors:      3520  (44 x 80)
    Speed index:            3660800
Comment 3 GitLab Migration User 2019-09-18 19:18:46 UTC
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/mesa/mesa/issues/536.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.