hashcat
advanced password recovery

chrisanag · (This post was last modified: 12-27-2016, 01:33 AM by chrisanag.)

Hi ... i recently installed my msi rx470/8GB after i installed the AMDGPU-PRO 16.50 (dkms for some reason is always fail to build ) i have saw that when i run hashcat it tells me that my GPU has only 14 Compute Units(MCU) but as i have read it has 32 MCU.... Thnx in advance

inxi-Gx
Graphics: Card: Advanced Micro Devices [AMD/ATI] Ellesmere [Radeon RX 470/480] bus-ID: 01:00.0
Display Server: X.Org 1.18.4 drivers: ati,amdgpu (unloaded: fbdev,vesa,radeon)
Resolution:1920x1080@60.00hz                                                                                                                                     GLX Renderer: AMD Radeon RX 470 Graphics GLX Version: 4.5.13462 - CP

hashcat -I

hashcat (v3.20-55-gd259f96) starting...

OpenCL Info:

Platform ID #1
Vendor : Advanced Micro Devices, Inc.
Name    : AMD Accelerated Parallel Processing
Version : OpenCL 2.0 AMD-APP (2236.5)

Device ID #1
    Type           : GPU
    Vendor ID      : 1
    Vendor         : Advanced Micro Devices, Inc.
    Name           : Ellesmere
    Version        : OpenCL 1.2 AMD-APP (2236.5)
    Processor(s)   : 14
    Clock          : 555
    Memory         : 2047/7873 MB allocatable
    OpenCL Version : OpenCL C 1.2
    Driver Version : 2236.5

Device ID #2
    Type           : CPU
    Vendor ID      : 1
    Vendor         : AuthenticAMD
    Name           : AMD FX-8370 Eight-Core Processor
    Version        : OpenCL 1.2 AMD-APP (2236.5)
    Processor(s)   : 8
    Clock          : 2800
    Memory         : 2047/15948 MB allocatable
    OpenCL Version : OpenCL C 1.2
    Driver Version : 2236.5 (sse2,avx,fma4)

uname -a
Linux 4.8.0-30-generic #32-Ubuntu SMP Fri Dec 2 03:43:27 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux

chris@nemesis:~/hashcat$ /opt/amdgpu-pro/bin/clinfo
Number of platforms:                             1
Platform Profile:                              FULL_PROFILE
Platform Version:                              OpenCL 2.0 AMD-APP (2236.5)
Platform Name:                                 AMD Accelerated Parallel Processing
Platform Vendor:                               Advanced Micro Devices, Inc.
Platform Extensions:                           cl_khr_icd cl_amd_event_callback cl_amd_offline_devices

Platform Name:                                 AMD Accelerated Parallel Processing
Number of devices:                               2
Device Type:                                   CL_DEVICE_TYPE_GPU
Vendor ID:                                     1002h
Board name:                                    AMD Radeon (TM) RX 470 Graphics
Device Topology:                               PCI[ B#1, D#0, F#0 ]
Max compute units:                             14
Max work items dimensions:                     3
    Max work items[0]:                           256
    Max work items[1]:                           256
    Max work items[2]:                           256
Max work group size:                           256
Preferred vector width char:                   4
Preferred vector width short:                  2
Preferred vector width int:                    1
Preferred vector width long:                   1
Preferred vector width float:                  1
Preferred vector width double:                 1
Native vector width char:                      4
Native vector width short:                     2
Native vector width int:                       1
Native vector width long:                      1
Native vector width float:                     1
Native vector width double:                    1
Max clock frequency:                           555Mhz
Address bits:                                  64
Max memory allocation:                         4244635648
Image support:                                 Yes
Max number of images read arguments:           128
Max number of images write arguments:          8
Max image 2D width:                            16384
Max image 2D height:                           16384
Max image 3D width:                            2048
Max image 3D height:                           2048
Max image 3D depth:                            2048
Max samplers within kernel:                    16
Max size of kernel argument:                   1024
Alignment (bits) of base address:              2048
Minimum alignment (bytes) for any datatype:    128
Single precision floating point capability
    Denorms:                                     No
    Quiet NaNs:                                  Yes
    Round to nearest even:                       Yes
    Round to zero:                               Yes
    Round to +ve and infinity:                   Yes
    IEEE754-2008 fused multiply-add:             Yes
Cache type:                                    Read/Write
Cache line size:                               64
Cache size:                                    16384
Global memory size:                            8256319488
Constant buffer size:                          65536
Max number of constant args:                   8
Local memory type:                             Scratchpad
Local memory size:                             32768
Max pipe arguments:                            0
Max pipe active reservations:                  0
Max pipe packet size:                          0
Max global variable size:                      0
Max global variable preferred total size:      0
Max read/write image args:                     0
Max on device events:                          0
Queue on device max size:                      0
Max on device queues:                          0
Queue on device preferred size:                0
SVM capabilities:
    Coarse grain buffer:                         No
    Fine grain buffer:                           No
    Fine grain system:                           No
    Atomics:                                     No
Preferred platform atomic alignment:           0
Preferred global atomic alignment:             0
Preferred local atomic alignment:              0
Kernel Preferred work group size multiple:     64
Error correction support:                      0
Unified memory for Host and Device:            0
Profiling timer resolution:                    1
Device endianess:                              Little
Available:                                     Yes
Compiler available:                            Yes
Execution capabilities:
    Execute OpenCL kernels:                      Yes
    Execute native function:                     No
Queue on Host properties:
    Out-of-Order:                                No
    Profiling :                                  Yes
Queue on Device properties:
    Out-of-Order:                                No
    Profiling :                                  No
Platform ID:                                   0x7f62c9e59db8
Name:                                          Ellesmere
Vendor:                                        Advanced Micro Devices, Inc.
Device OpenCL C version:                       OpenCL C 1.2
Driver version:                                2236.5
Profile:                                       FULL_PROFILE
Version:                                       OpenCL 1.2 AMD-APP (2236.5)
Extensions:                                    cl_khr_fp64 cl_amd_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_fp16 cl_khr_gl_sharing cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_spir cl_khr_gl_event

**epixoip** · (This post was last modified: 05-07-2017, 07:09 AM by epixoip.)

RX470 is essentially just a 7970 at heart, and does indeed have 32 MCU. I'm not sure why clinfo is telling you that you have 14 MCU, which would make it an HD7790/RX460. Probably a driver issue.

chrisanag · 12-27-2016, 04:27 PM

At last i found this thread :
https://community.amd.com/thread/207051

It is a bug from 16.30 amdgpu-pro and still it is not solved... Tongue

Tnx epixoip for the fast reply...

Lets see ...

Still my performance on hashcat is

Hashtype: WPA/WPA2

Speed.Dev.#1.....: 182.4 kH/s (78.00ms)

which i think is the same with others as i saw on internet benchmarks...

BOuRNe_AGAIN · (This post was last modified: 05-07-2017, 07:11 AM by epixoip.)

(12-27-2016, 07:07 AM)epixoip Wrote: RX470 is essentially just a 7970 at heart, and does indeed have 32 MCU. I'm not sure why clinfo is telling you that you have 14 MCU, which would make it an HD7790/RX460. Probably a driver issue.

Isn't the 470 based on the Polaris architecture while the 7970 is based on a different one? (GCN maybe?) Also, I believe the R9 280x is a rebranded 7970.

**epixoip** · (This post was last modified: 05-07-2017, 07:12 AM by epixoip.)

Yes, the R9 280X is a rebranded 7970. And the RX470 is essentially just a 7970 with a die shrink. It's nothing special.

Flomac · 05-05-2017, 07:21 PM

Sorry, but that's wrong. The Polaris 10 tops out at 2304 cores (RX480) and comes with 2048 in the RX470, while the Tahiti chip had only 2048 cores at all. I don't says the architecture changed much, but it got 2.6 times more cache and some goodies. Does this matter for Hashcat? Probably not much. It's just definitly not the same chip rebaked in a die shrink.

**epixoip** · 05-06-2017, 06:41 AM

Bro, seriously, lift the veil. Look with your soul, not your eyes. *Everything* that AMD has released since the dawn of GCN has been based on the 7970. Every new GPU they've released has either been a 7970, or a 7970 with fewer MCUs, or a 7970 with more MCUs. And sure, a die shrink and maybe some extra cache, whatever. But it's still a fucking 7970. If you don't believe me, just compare the block diagrams for any two GCN GPUs from any generation.

Here's AMD's formula for chip engineering:

Code:
HOW AMD DESIGNS ITS "NEW" GPU ARCHITECTURES

+--------------+

| Need new GPU |

+-------+------+

        |

+-------v------+

|   Haz 7970!  |

+-------+------+

        |

+-------v------+

|  Haz money?  |-----> LOL NO.

+-------+------+          |       +-------------+   +---------------+

        |                 |       | PASTE MCUs! |-->| Can power it? |

  THANKS XBOX!            |       +------^------+   +---------------+

        |                 |              |                   |

+-------v------+   +------v-------+      |                LOL NO.

| DIE SHRINK!! |-->|  Moar cores? |--> MOAR!                 |

+--------------+   +------+-------+                 +--------v------+   +----------------+

                          |        \                |   POWERTUNE!  |-->| Can power now? |--> PRAISE SATAN!

                        LESS!       `SAMESIES!      +---------------+   +--------+-------+          |

                          |              |                                       |                  |

                   +------v------+  +----v------------+                          v                  |

                   |  CUT MCUs!  |  | MAKE 7970 AGAIN |<---------------------- FUCK.                |

                   +------+------+  +--------+--------+                                             |

                          |                  |                                                      |

                          |                  |                                                      |

                          |                  |    +----------------------+                          |

                          |                  +---->                      |                          |

                          +----------------------->  NEW ARCHITECTURE!!  <--------------------------+

                                                  |                      |

                                                  +----------------------+

© 2017, epixoip. All rights reserved.

**undeath** · 05-06-2017, 06:58 PM

I find the flowchart above very offensive and not true. It's completely ignoring the part where AMD increases the clock rate that subsequently needs to be throttled by powertune.

Flomac · (This post was last modified: 05-06-2017, 08:18 PM by Flomac.)

(05-06-2017, 06:41 AM)epixoip Wrote: Bro, seriously, lift the veil. Look with your soul, not your eyes. *Everything* that AMD has released since the dawn of GCN has been based on the 7970. Every new GPU they've released has either been a 7970, or a 7970 with fewer MCUs, or a 7970 with more MCUs. And sure, a die shrink and maybe some extra cache, whatever. But it's still a fucking 7970. If you don't believe me, just compare the block diagrams for any two GCN GPUs from any generation.

Chill down, Bro. I just wanted to make a pedantic comment, cause some might come to the idea AMD is reproducing _exactly_ the same chip for over six years now.

Is it basically the same architecture since half a decade? Definitly yes.
Is it the same chip shrinked down over and over again? No.
Is it important for hashcat anyway? Not really.
Any reason buying it for hashcat? A big NO.

'Nuff said.

EDIT: nice chart though, you should sell it to AMD and save'em some time Wink

**epixoip** · 05-07-2017, 07:27 AM

Undeath: I think this is mostly covered by "Can power it?" Smile

Flomac: I <3 you bro, but your lack of cynicism towards AMD displeases me. Anyway, yes, pedantically speaking, there are lots of little differences, but none of them ultimately matter for compute. Are they stamping the exact same physical chip? No. But they're pretty goddamned close Smile

Although in all fairness, GCN2 did add more ACE's and organized the MCUs into Shader Engines. That was a fairly significant change, I'll admit. But it didn't really do anything for compute performance. BUT they did cut the number of ACE's in half on Polaris, so it's a lot closer to a 7970 than a 290X. So, comparison is still valid Tongue

Login
Username/Email:
Password:	Lost Password?
	Remember me

hashcat advanced password recovery

hashcat
advanced password recovery