As a reaction, Openssh will deprecate SHA-1 support in a future release: http://www.openssh.com/txt/release-8.3
The test fails on my OpenWRT 19.07.03 router's Dropbear installation.
SHA-1 collisions now cost $45k [pdf]
11–20 of 64 posts
Re: SHA-1 collisions now cost $45k [pdf]
#12Earlier quoted context omitted.
It would scale linearly right? Double the amount of computing thrown at it, you'd half the time on average. Or am I mistaken?
Depending on the algorithm, you might be able to skip from GPGPU straight to ASIC.
At this scale - I’m not convinced that $5,000 worth of ASICs would have the same performance as $5,000 worth of GeForce cards - especially after factoring in lead-time to design the ASIC.
As for FPGAs - forgive my ignorance - but can they even handle that kind of load? In my head I just see them breaking-down under the sheer thermal load.
Re: SHA-1 collisions now cost $45k [pdf]
#13Not knowing much about binary executable formats, what does this mean for binary executables or libraries? How easy is it to insert e.g. a remote shell or keystroke logger into an executable or library using only a prefix? Or would that require arbitrary in-place edits to a file?
Re: SHA-1 collisions now cost $45k [pdf]
#14Earlier quoted context omitted.
Depending on the algorithm, you might be able to skip from GPGPU straight to ASIC.
At this scale - I’m not convinced that $5,000 worth of ASICs would have the same performance as $5,000 worth of GeForce cards - especially after factoring in lead-time to design the ASIC. As for FPGAs - forgive my ignorance - but can they even handle that kind of load? In my head I just see them breaking-down under the sheer thermal load.
> As for FPGAs - forgive my ignorance - but can they even handle that kind of load? In my head I just see them breaking-down under the sheer thermal load.
They worked for Bitcoin before miners switched to ASICS; I imagine they'd be fine in this application as well.
Re: SHA-1 collisions now cost $45k [pdf]
#15It also cost 2 months (FTA)... So the ever returning question is : how long does it take when NSA/military/government-levle funding is applied ?
It would scale linearly right? Double the amount of computing thrown at it, you'd half the time on average. Or am I mistaken?
And of course double the compute for half the time = same cost (more or less). I would imagine the NSA has sunk enough into hardware to do this fairly cheaply per-pass, and very quickly (hours not weeks).
Re: SHA-1 collisions now cost $45k [pdf]
#16One of the best things to come out of protocol labs is https://multiformats.io/
Really simple mechanisms for things like identifying the hash algorithm and gives you a programmatic way of supporting new hash algorithms without breaking or changing anything that depends on the old.
Re: SHA-1 collisions now cost $45k [pdf]
#17Not a crypto expert, how easy/hard it is with this (or other techniques) at the moment to generate a random file which matches a given SHA1 hash? Can have totally random bits lets say.
Re: SHA-1 collisions now cost $45k [pdf]
#18One of the best things to come out of protocol labs is https://multiformats.io/ Really simple mechanisms for things like identifying the hash algorithm and gives you a programmatic way of supporting new hash algorithms without breaking or changing anything that depends on the old.
The first example [1] doesn't have the hash-func-type prefix
[1] https://multiformats.io/multihash/#the-multihash-format
Edit: someone already detected this: https://github.com/multiformats/website/pull/62
Re: SHA-1 collisions now cost $45k [pdf]
#19Not knowing much about binary executable formats, what does this mean for binary executables or libraries? How easy is it to insert e.g. a remote shell or keystroke logger into an executable or library using only a prefix? Or would that require arbitrary in-place edits to a file?
If you control the executable then it's easy. If you can only append then it's harder.
Re: SHA-1 collisions now cost $45k [pdf]
#20How safe is SHA-256 now? Or Is SHA-512 needed in the near future?