![]() |
amd-gpu-pro update - Printable Version +- hashcat Forum (https://hashcat.net/forum) +-- Forum: Misc (https://hashcat.net/forum/forum-15.html) +--- Forum: General Talk (https://hashcat.net/forum/forum-33.html) +--- Thread: amd-gpu-pro update (/thread-5959.html) |
RE: amd-gpu-pro update - NoInstall - 12-14-2016 awkward moment ... I have a TAHITI Radeon HD 7900 series thanks Johnya. I will have a look at drivers for the card unless you have a thought on this??? RE: amd-gpu-pro update - johnya - 12-14-2016 No problem! (12-14-2016, 11:19 PM)NoInstall Wrote: awkward moment ... I have a TAHITI Radeon HD 7900 series I think, you have to install fglrx or buy a new card. (12-14-2016, 11:19 PM)NoInstall Wrote: I will have a look at drivers for the card unless you have a thought on this?? RE: amd-gpu-pro update - NoInstall - 12-25-2016 I am almost there ... I installed the fglrx drivers on ubuntu14.04.05 following atom's instructions. It did not work if I compile the hashcat source code or use v3.20 however works (half works) on v3.10. But on 3.10 it half works, as in if I can only use 1 of the 2 GPUs of the HD7990 card or it freezes/crashes ubuntu ??? For example this takes a few seconds to return the benchmark info and works: ************************************************************ ./hashcat64.bin -b -m 2500 -d2 hashcat (v3.10) starting in benchmark-mode... OpenCL Platform #1: Advanced Micro Devices, Inc. ================================================ - Device #1: Tahiti, skipped - Device #2: Tahiti, 2101/3017 MB allocatable, 32MCU - Device #3: AMD FX(tm)-8320 Eight-Core Processor, skipped Hashtype: WPA/WPA2 Speed.Dev.#2.: 145.3 kH/s (95.25ms) Or this also takes a few seconds to return the benchmark info and works: ************************************************************ ./hashcat64.bin -b -m 2500 -d1 hashcat (v3.10) starting in benchmark-mode... OpenCL Platform #1: Advanced Micro Devices, Inc. ================================================ - Device #1: Tahiti, 2101/3017 MB allocatable, 32MCU - Device #2: Tahiti, skipped Hashtype: WPA/WPA2 Speed.Dev.#1.: 145.3 kH/s (95.26ms) But this does not work and freezes Ubuntu console and crashes after a few minutes: ************************************************************************ ./hashcat64.bin -b -m 2500 -d1,2 OR ./hashcat64.bin -b -m 2500 hashcat (v3.10) starting in benchmark-mode... OpenCL Platform #1: Advanced Micro Devices, Inc. ================================================ - Device #1: Tahiti, 2101/3017 MB allocatable, 32MCU - Device #2: Tahiti, 2101/3017 MB allocatable, 32MCU Hashtype: WPA/WPA2 And judging from the temperature readings it is only loading the first GPU ******************************************************** Adapter 0 - AMD Radeon HD 7900 Series Sensor 0: Temperature - 64.00 C Adapter 1 - AMD Radeon HD 7900 Series Sensor 0: Temperature - 37.00 C RE: amd-gpu-pro update - atom - 12-26-2016 Did you run sudo amdconfig --initial -f --adapter=all ? RE: amd-gpu-pro update - Si2006 - 01-01-2017 These drivers much much slower running my 290x dict + rules. A single md5 is about 350MH/s in comparsion with my 1070 (3100MH/s) whats the problems with these drivers? v15.12 are like 5x + faster with the fast hashes. Ubuntu Server 16.04 / amdgpu-pro 16.50 / nvidia 375.26 \ hashcat 3.20+ Session..........: hashcat Status...........: Quit Hash.Type........: MD5 Hash.Target......: 00000000000000000000000000000000 Time.Started.....: Sun Jan 1 14:31:23 2017 (12 secs) Time.Estimated...: Sun Jan 1 16:48:11 2017 (2 hours, 16 mins) Input.Base.......: File (/media/found.2015.txt) Input.Mod........: Rules (rules/dive.rule) Speed.Dev.#1.....: 335.2 MH/s (348.64ms) Speed.Dev.#2.....: 310.2 MH/s (376.71ms) Speed.Dev.#3.....: 312.8 MH/s (373.52ms) Speed.Dev.#5.....: 3184.4 MH/s (316.01ms) Speed.Dev.#*.....: 4142.6 MH/s Recovered........: 0/1 (0.00%) Digests, 0/1 (0.00%) Salts Progress.........: 42147052362/33996704056172 (0.12%) Rejected.........: 62126922/42147052362 (0.15%) Restore.Point....: 0/343103002 (0.00%) Candidates.#1....: -> 0704070614 Candidates.#2....: 0545467 -> 0704144479 Candidates.#3....: *-<>& -> 07/04/099 Candidates.#5....: 00912amasoso0 -> 07041930 HWMon.Dev.#1.....: Temp: 70c Fan: 30% Core:1000Mhz Mem:1250Mhz Lanes:16 HWMon.Dev.#2.....: Temp: 57c Fan: 30% Core: 483Mhz Mem: 150Mhz Lanes:16 HWMon.Dev.#3.....: Temp: 54c Fan: 30% Core: 662Mhz Mem:1250Mhz Lanes:1 HWMon.Dev.#5.....: Temp: 45c Fan: 30% Util: 29% Core:1873Mhz Mem:3802Mhz Lanes:1 RE: amd-gpu-pro update - atom - 01-01-2017 16.50 are known to be broken (because of speed), use 16.40. RE: amd-gpu-pro update - Si2006 - 01-01-2017 (01-01-2017, 07:59 PM)atom Wrote: 16.50 are known to be broken (because of speed), use 16.40.I've been using 16.40-348864. I only updated to 16.50 today hoping for an improvement but there isn't any noticeable change between both drivers. Edit: Heres the same test with 16.40, Ok very slight speed increase although still nowhere near the 1070 speed as expected. OpenCL Platform #1: Advanced Micro Devices, Inc. ================================================ * Device #1: Hawaii, 2047/4054 MB allocatable, 14MCU * Device #2: Hawaii, 2047/4066 MB allocatable, 14MCU * Device #3: Hawaii, 2047/4066 MB allocatable, 14MCU * Device #4: AMD FX(tm)-6300 Six-Core Processor, skipped OpenCL Platform #2: NVIDIA Corporation ====================================== * Device #5: GeForce GTX 1070, 2028/8113 MB allocatable, 15MCU Hashes: 1 digests; 1 unique digests, 1 unique salts Bitmaps: 16 bits, 65536 entries, 0x0000ffff mask, 262144 bytes, 5/13 rotates Rules: 99086 Applicable Optimizers: * Zero-Byte * Precompute-Init * Precompute-Merkle-Demgard * Meet-In-The-Middle * Early-Skip * Not-Salted * Not-Iterated * Single-Hash * Single-Salt * Raw-Hash Watchdog: Temperature abort trigger set to 90c Watchdog: Temperature retain trigger set to 75c Cache-hit dictionary stats /media/found.2015.txt: 3252675084 bytes, 343103002 words, 33996704056 [s]tatus [p]ause [r]esume [b]ypass [c]heckpoint [q]uit => q Session..........: hashcat Status...........: Quit Hash.Type........: MD5 Hash.Target......: 00000000000000000000000000000000 Time.Started.....: Sun Jan 1 19:28:33 2017 (11 secs) Time.Estimated...: Sun Jan 1 21:17:41 2017 (1 hour, 48 mins) Input.Base.......: File (/media/found.2015.txt) Input.Mod........: Rules (rules/dive.rule) Speed.Dev.#1.....: 718.4 MH/s (326.53ms) Speed.Dev.#2.....: 632.6 MH/s (370.88ms) Speed.Dev.#3.....: 630.7 MH/s (371.97ms) Speed.Dev.#5.....: 3210.5 MH/s (313.45ms) Speed.Dev.#*.....: 5192.3 MH/s Recovered........: 0/1 (0.00%) Digests, 0/1 (0.00%) Salts Progress.........: 52432902288/33996704056172 (0.15%) Rejected.........: 87988368/52432902288 (0.17%) Restore.Point....: 0/343103002 (0.00%) Candidates.#1....: -> 090129552130 Candidates.#2....: 07318B72035 -> 1061ge Candidates.#3....: $HEX[18303232353135] -> $HEX[37323730393031333330] Candidates.#5....: 01txq -> 0201518898 HWMon.Dev.#1.....: Temp: 60c Fan: 30% Core: 840Mhz Mem:1250Mhz Lanes:16 HWMon.Dev.#2.....: Temp: 54c Fan: 30% Core: 947Mhz Mem:1250Mhz Lanes:16 HWMon.Dev.#3.....: Temp: 49c Fan: 30% Core: 914Mhz Mem:1250Mhz Lanes:1 HWMon.Dev.#5.....: Temp: 43c Fan: 28% Util: 67% Core:1885Mhz Mem:3802Mhz Lanes:1 Started: Sun Jan 1 19:28:18 2017 Stopped: Sun Jan 1 19:28:44 2017 RE: amd-gpu-pro update - atom - 01-02-2017 I disagree. I'd say 350MH/s -> 720MH/s is a noticeable change. Anyway, we can only try to workaround the mess up created by the latest drivers. We can't change the driver itself. Note that for new Linux kernel and new GPU you can't use older Drivers, therefore it makes no sense in "ignoring" the progress of them. We're forced to move with it. Complain at AMD if you want, but they wont listen. RE: amd-gpu-pro update - NoInstall - 01-02-2017 (12-26-2016, 12:48 PM)atom Wrote: Did you run sudo amdconfig --initial -f --adapter=all ? Hi atom, I did during the setup and tried again just in case but no change. I still have the same issue. I also tried to run 2 separate instance using d1 and d2 and it does not work with both GPU are on at the same time bt each work individually ??? Any thoughts RE: amd-gpu-pro update - atom - 01-03-2017 Try again with latest beta from https://hashcat.net/beta/ |