I'm writing Data Encryption Standard "cracker" using C++ and CUDA. It was going to be simple brute-force - trying all possible keys to decrypt encrypted data and check if result is equal to initial plain-text message.
The problem is that generation of 2^56 keys takes time (and memory). My first approach was to generate keys recursively and save them to file.
Do you have any suggestions how to improve this?
You don' really need recursion, neither you need storing your keys.
All space of DES keys (if we don't count 12 or so weak keys, which won't change anything for your purposes) is a space of 56-bit-long numbers (which BTW fit into standard uint64_t), and you can just iterate through numbers from 0 to 2^56-1, feeding the next number as a 56-bit number to your CUDA core whenever the core reports that it is done with the previous key.
If not for cores, the code could look such as:
for(uint64_t i=0;i<0xFFFFFFFFFFFFFFULL /* double-check number of F's so the number is 2^56-1 */;++i) {
uint8_t key[7];
//below is endianness-agnostic conversion
key[0] = (uint8_t)i;
key[1] = (uint8_t)(i>>8);
key[2] = (uint8_t)(i>>16);
key[3] = (uint8_t)(i>>24);
key[4] = (uint8_t)(i>>32);
key[5] = (uint8_t)(i>>40);
key[6] = (uint8_t)(i>>48);
bool found = try_your_des_code(key,data_to_decrypt);
if(found) printf("Eureka!\n");
}
To allow restarting your program in case if anything goes wrong, you need to store (in persistent storage, such as file) only this number i (with cores, strictly speaking - the number i should be written to persistent storage only after all the numbers before it has already been processed by CUDA cores, but generally the difference of 2000 or so keys won't make any difference performance-wise).
Related
I use SEAL 2.3.1 and this is my parameter setting:
seal::EncryptionParameters parms;
parms.set_poly_modulus("1x^2048 + 1"); // n = 2048
parms.set_coeff_modulus(coeff_modulus_128(2048)); // q = 54-bit prime
parms.set_plain_modulus(1 << 8); // t = 256
seal::SEALContext context(parms);
And some Ciphertext encrypted1; holding the number 5. The manual say that one can use the seal::Simulator class for reading the noise budget without the secret key. The only thing that I've found was this in the simulator.h file.
/**
Creates a simulation of a ciphertext encrypted with the specified encryption
parameters and given invariant noise budget. The given noise budget must be
at least zero, and at most the significant bit count of the coefficient
modulus minus two.
#param[in] parms The encryption parameters
#param[in] noise_budget The invariant noise budget of the created ciphertext
#param[in] ciphertext_size The size of the created ciphertext
#throws std::invalid_argument if ciphertext_size is less than 2
#throws std::invalid_argument if noise_budget is not in the valid range
*/
Simulation(const EncryptionParameters &parms, int ciphertext_size,
int noise_budget);
I can set it with some other Ciphertext encrypted2:
seal::Simulation(parms, encrypted2.size(), (context.total_coeff_modulus().significant_bit_count() - log2(context.poly_modulus().coeff_count() - 1) - log2(context.plain_modulus().value()));
But using this will only create a simulated Ciphertext without any real connection to the actual encrypted1 Ciphertext noise budget.
Is there a way to approximate the noise budget of encrypted1 without the secret key? This situations is important when I or someone else does some computation on externally stored Ciphertexts, e.g. in a cloud database and needs to check the noise budget server side without revealing the secret key.
The Simulation class is meant to estimate the noise budget consumption in various operations so that those operations don't actually have to be executed on real data. Moreover, it uses heuristic upper bounds estimate for the noise consumption, i.e. most likely it overestimates the noise consumption and this effect becomes more pronounced when the computation is more complicated, sometimes resulting in huge overestimates of the noise consumption. Of course, the idea is that the computation is guaranteed to work if it works according to the simulator. A typical use of Simulation would be through the ChooserPoly (and related) classes; this is demonstrated in one of the examples in SEALExamples/main.cpp for SEAL versions < 3.0.
It is impossible to know or estimate the noise in a ciphertext without knowing how that ciphertext was produced. So if I give you a ciphertext without telling you anything else (except encryption parameter), then you should not be able to know anything about the noise budget unless you know the secret key. I agree that in some cases it could be important for someone to know right away if a ciphertext is still valid for further computations, but it's not possible without some external mechanism.
I'm new to Qt's multimedia library and in my application I want to mix audio from multiple input devices (e.g. microphone), in order to stream it via TCP.
As far as I know I have to obtain the specific QAudioDeviceInfo for all needed devices first - together with an according QAudioFormat object - and use this with QAudioInput. Then I simply call start() for every created QAudioInput object and read out pending bytes with readLine().
But how can I mix audio data of multiple devices to one buffer?
I am not sure if there is any Qt specific method / class to do this. However it's pretty simple to do it yourself.
The most basic way (assuming you are using PCM), you can simply add the two streams/buffers together word by word (if I recall they are 16-bit PCM words).
So if you have two input buffers:
int16 buff1[10];
int16 buff2[10];
int16 mixBuff[10];
// Fill them...
//... code goes here to read from the buffers ....
// Add them (effectively mix them)
for (int i = 0; i < 10; i++)
{
mixBuff[i] = buff1[i] + buff2[i];
}
Now, this is very crude and does not take any scaling into consideration. So imagine buff1 and buff2 both use 80% of the dynamic range (call this full volume, beyond which you get distortion), then when you add them together you will get number overrun (i.e. 16-bit max is 65535 so 50000 + 50000 will be a over run).
Each time you mix you effectively need half the two inputs (so 65535 / 2 + 65535 / 2 = 65535... i.e. when you add them up you can't overrun). So your mix code is like this:
for (int i = 0; i < 10; i++)
{
mixBuff[i] = (buff1[i] >> 1) + (buff2[i] >> 1);
}
There is much more you can do (noise removal etc...) but then the maths start getting a bit hairy. This is very simple. You can use the shift afterwards to increase / decrease volume as simple volume control if you want.
EDIT
One thing to note... you are using readline() (which the docs say reads the data out as ASCII). I always use read() which it does not state the "format" it is read out, but I am assuming binary. So this code may not work if you use readline() but I have never tried it. It works well for read(), you don't really want to be working in ASCII if you want to manipulate the data.
I'm working on software for processing audio in real time in C++ with Qt. I need that requirements are minimized.
Defining a temporary buffer 40ms, launching our device with a sampling frequency Fs = 8000Hz, every 320 samples entered a feature called Data Processing ().
The idea is to have a global buffer that stores the 10s last recorded, 80000 samples.
This Buffer in each iteration eliminates the initial 320 samples and looped at the end, 320 new samples. Thus the buffer is updated and the user can observe the real-time graphical representation of the recorded signal.
At first I thought of using QVector (equivalent to std::vector but for Qt) for this deployment, thus we reduce the process a few lines of code
int NUM_POINTS=320;
DatosTemporales.erase(DatosTemporales.begin(),DatosTemporales.begin()+NUM_POINTS);
DatosTemporales+= (DatosNuevos); // Datos Nuevos con un tamaño de NUM_POINTS
In each iteration we create a vector of 80000 samples in addition to free some positions so requires some processing time. An alternative for opting was the use of * double, and iterations a loop:
for(int i=0;i<80000;i++){
if(i<80000-NUM_POINTS){
aux=DatosTemporales[i];
DatosTemporales[i+NUM_POINTS]=aux;
}else{
DatosTemporales[i]=DatosNuevos[i-NUN_POINTS];
}
}
Does fails. I think the best way is to use dynamic memory. Implementing this process by pointers. Could anyone give me some idea how to implement it?
It sounds like what you are looking for is a circular buffer.
https://www.google.com/search?q=qcircularbuffer
https://qt.gitorious.org/qt/qtbase/merge_requests/60
And it looks like you only need the header file and you should be good to go.
A similar tool that is already in the Qt data set is found here:
http://doc.qt.io/qt-5/qcontiguouscache.html#details
The advantage of using a system like these presented, is that they don't need to have dynamic memory, it just needs to move the head and the tail pointers.
Hope that helps.
Hoping someone more experienced in OpenCL usage may be able to help me here! I'm doing a project (to help me learn a bit more crypto and to try my hand at GPGPU programming) where I'm trying to implement my own SHA-1 algorighm.
Ultimately my question is about maximizing my throughput rates. At present I'm seeing something like 56.1 MH/sec, which compares very badly to open source programs I've looked at, such as John the Ripper and OCLHashcat, which are giving 1,000 and 1,500 MH/sec respectively (heck, I'd be well-chuffed with a 3rd of that!).
So, what I'm doing
I've written a SHA-1 implementation in an OpenCL kernel and a C++ host application to load data to the GPU (using CL 1.2 C++ wrapper). I'm generating blocks of candidate data to hash in a threaded fashion on the CPU and loading this data onto the global GPU memory using the CL C++ call to enqueueWriteBuffer (using uchars to represent the bytes to hash):
errorCode = dispatchQueue->enqueueWriteBuffer(
inputBuffer,
CL_FALSE,//CL_TRUE,
0,
sizeof(cl_uchar) * inputBufferSize,
passwordBuffer,
NULL,
&dispatchDelegate);
I'm en-queuing data using enqueueNDRangeKernel in the following manner (where global worksize is a user-defined variable, at present I've set this to my GPUs maximum flattened global worksize of 16.777 million per run):
errorCode = dispatchQueue->enqueueNDRangeKernel(
*kernel,
NullRange,
NDRange(globalWorkgroupSize, 1),
NullRange,
NULL,
NULL);
This means that (per dispatch) I load 16.777 million items in a 1D array and index from my kernel into this using get_global_offset(0).
My Kernel signature:
__kernel void sha1Crack(__global uchar* out, __global uchar* in,
__constant int* passLen, __constant int* targetHash,
__global bool* collisionFound)
{
//Kernel Instance Global GPU Mem IO Mapping:
__private int id = get_global_id(0);
__private int inputIndexStart = id * passwordLen;
//Select Password input key space:
#pragma unroll
for (i = 0; i < passwordLen; i++)
{
inputMem[i] = in[inputIndexStart + i];
}
//SHA1 Code omitted for brevity...
}
So, given all this: am I doing something fundamentally wrong in the way I'm loading data? I.e. 1 call to enqueueNDrange for 16.7 million kernel executions over a 1D input vector? Should I be using a 2-D space and sub-dividing into localworkgroup ranges? I tried playing with this but it didn't seem quicker.
Or, perhaps as likely is my algorithm itself the source of slowness? I've spent a good while optimizing it and manually unrolling all of the loop stages using pre-processor directives.
I've read about memory coalescing on the hardware. Could that be my issue? :S
Any advice at all appreciated! If I've missed anything important please let me know and I'll update.
Thanks in advance! ;)
Update: 16,777,216 is the device maximum reported workgroup size; 256**3. The global array of boolean values is one boolean. It's set to false at the start of the kernel enqueue, then a branching statement sets this to true if a collision is found only - will that force a convergence? passwordLen is the length of the current input value and target hash is an int[4] encoded hash to check against.
Your 'maximum flattened global worksize' should be multiplied by passwordLen. It is the number of kernels you can run, not the maximal length of an input array. You can most likely send much more data than this to the GPU.
Other potential issues: the 'generating blocks of candidate data to hash in a threaded fashion on the CPU', try doing this in advance of the kernel iterations to see whether the delay is in the generation of the data blocks or in the processing of the kernels; your sha1 algorithm is the other obvious potential issue. I'm not sure how much you've really optimised it by 'unrolling' the loops, usually the bigger optimisation issue is 'if' statements (if a single kernel instance within a workgroup tests to true then all of the lockstepped workgroup instances must follow that branch in parallel).
And DarkZeros is correct, you should manually play with the local workgroup size making it the highest common multiple of the global size and the number of kernels which can be run at once on the card. The easiest way to do this is to round up the global work group size to the next multiple of the card capacity and use an external if{} statement in the kernel only running the kernel for global_id less than the actual number of kernels you want to run.
Dave.
I have a 'server' program that updates many linked lists in shared memory in response to external events. I want client programs to notice an update on any of the lists as quickly as possible (lowest latency). The server marks a linked list's node's state_ as FILLED once its data is filled in and its next pointer has been set to a valid location. Until then, its state_ is NOT_FILLED_YET. I am using memory barriers to make sure that clients don't see the state_ as FILLED before the data within is actually ready (and it seems to work, I never see corrupt data). Also, state_ is volatile to be sure the compiler doesn't lift the client's checking of it out of loops.
Keeping the server code exactly the same, I've come up with 3 different methods for the client to scan the linked lists for changes. The question is: Why is the 3rd method fastest?
Method 1: Round robin over all the linked lists (called 'channels') continuously, looking to see if any nodes have changed to 'FILLED':
void method_one()
{
std::vector<Data*> channel_cursors;
for(ChannelList::iterator i = channel_list.begin(); i != channel_list.end(); ++i)
{
Data* current_item = static_cast<Data*>(i->get(segment)->tail_.get(segment));
channel_cursors.push_back(current_item);
}
while(true)
{
for(std::size_t i = 0; i < channel_list.size(); ++i)
{
Data* current_item = channel_cursors[i];
ACQUIRE_MEMORY_BARRIER;
if(current_item->state_ == NOT_FILLED_YET) {
continue;
}
log_latency(current_item->tv_sec_, current_item->tv_usec_);
channel_cursors[i] = static_cast<Data*>(current_item->next_.get(segment));
}
}
}
Method 1 gave very low latency when then number of channels was small. But when the number of channels grew (250K+) it became very slow because of looping over all the channels. So I tried...
Method 2: Give each linked list an ID. Keep a separate 'update list' to the side. Every time one of the linked lists is updated, push its ID on to the update list. Now we just need to monitor the single update list, and check the IDs we get from it.
void method_two()
{
std::vector<Data*> channel_cursors;
for(ChannelList::iterator i = channel_list.begin(); i != channel_list.end(); ++i)
{
Data* current_item = static_cast<Data*>(i->get(segment)->tail_.get(segment));
channel_cursors.push_back(current_item);
}
UpdateID* update_cursor = static_cast<UpdateID*>(update_channel.tail_.get(segment));
while(true)
{
ACQUIRE_MEMORY_BARRIER;
if(update_cursor->state_ == NOT_FILLED_YET) {
continue;
}
::uint32_t update_id = update_cursor->list_id_;
Data* current_item = channel_cursors[update_id];
if(current_item->state_ == NOT_FILLED_YET) {
std::cerr << "This should never print." << std::endl; // it doesn't
continue;
}
log_latency(current_item->tv_sec_, current_item->tv_usec_);
channel_cursors[update_id] = static_cast<Data*>(current_item->next_.get(segment));
update_cursor = static_cast<UpdateID*>(update_cursor->next_.get(segment));
}
}
Method 2 gave TERRIBLE latency. Whereas Method 1 might give under 10us latency, Method 2 would inexplicably often given 8ms latency! Using gettimeofday it appears that the change in update_cursor->state_ was very slow to propogate from the server's view to the client's (I'm on a multicore box, so I assume the delay is due to cache). So I tried a hybrid approach...
Method 3: Keep the update list. But loop over all the channels continuously, and within each iteration check if the update list has updated. If it has, go with the number pushed onto it. If it hasn't, check the channel we've currently iterated to.
void method_three()
{
std::vector<Data*> channel_cursors;
for(ChannelList::iterator i = channel_list.begin(); i != channel_list.end(); ++i)
{
Data* current_item = static_cast<Data*>(i->get(segment)->tail_.get(segment));
channel_cursors.push_back(current_item);
}
UpdateID* update_cursor = static_cast<UpdateID*>(update_channel.tail_.get(segment));
while(true)
{
for(std::size_t i = 0; i < channel_list.size(); ++i)
{
std::size_t idx = i;
ACQUIRE_MEMORY_BARRIER;
if(update_cursor->state_ != NOT_FILLED_YET) {
//std::cerr << "Found via update" << std::endl;
i--;
idx = update_cursor->list_id_;
update_cursor = static_cast<UpdateID*>(update_cursor->next_.get(segment));
}
Data* current_item = channel_cursors[idx];
ACQUIRE_MEMORY_BARRIER;
if(current_item->state_ == NOT_FILLED_YET) {
continue;
}
found_an_update = true;
log_latency(current_item->tv_sec_, current_item->tv_usec_);
channel_cursors[idx] = static_cast<Data*>(current_item->next_.get(segment));
}
}
}
The latency of this method was as good as Method 1, but scaled to large numbers of channels. The problem is, I have no clue why. Just to throw a wrench in things: if I uncomment the 'found via update' part, it prints between EVERY LATENCY LOG MESSAGE. Which means things are only ever found on the update list! So I don't understand how this method can be faster than method 2.
The full, compilable code (requires GCC and boost-1.41) that generates random strings as test data is at: http://pastebin.com/0kuzm3Uf
Update: All 3 methods are effectively spinlocking until an update occurs. The difference is in how long it takes them to notice the update has occurred. They all continuously tax the processor, so that doesn't explain the speed difference. I'm testing on a 4-core machine with nothing else running, so the server and the client have nothing to compete with. I've even made a version of the code where updates signal a condition and have clients wait on the condition -- it didn't help the latency of any of the methods.
Update2: Despite there being 3 methods, I've only tried 1 at a time, so only 1 server and 1 client are competing for the state_ member.
Hypothesis: Method 2 is somehow blocking the update from getting written by the server.
One of the things you can hammer, besides the processor cores themselves, is your coherent cache. When you read a value on a given core, the L1 cache on that core has to acquire read access to that cache line, which means it needs to invalidate the write access to that line that any other cache has. And vice versa to write a value. So this means that you're continually ping-ponging the cache line back and forth between a "write" state (on the server-core's cache) and a "read" state (in the caches of all the client cores).
The intricacies of x86 cache performance are not something I am entirely familiar with, but it seems entirely plausible (at least in theory) that what you're doing by having three different threads hammering this one memory location as hard as they can with read-access requests is approximately creating a denial-of-service attack on the server preventing it from writing to that cache line for a few milliseconds on occasion.
You may be able to do an experiment to detect this by looking at how long it takes for the server to actually write the value into the update list, and see if there's a delay there corresponding to the latency.
You might also be able to try an experiment of removing cache from the equation, by running everything on a single core so the client and server threads are pulling things out of the same L1 cache.
I don't know if you have ever read the Concurrency columns from Herb Sutter. They are quite interesting, especially when you get into the cache issues.
Indeed the Method2 seems better here because the id being smaller than the data in general would mean that you don't have to do round-trips to the main memory too often (which is taxing).
However, what can actually happen is that you have such a line of cache:
Line of cache = [ID1, ID2, ID3, ID4, ...]
^ ^
client server
Which then creates contention.
Here is Herb Sutter's article: Eliminate False Sharing. The basic idea is simply to artificially inflate your ID in the list so that it occupies one line of cache entirely.
Check out the other articles in the serie while you're at it. Perhaps you'll get some ideas. There's a nice lock-free circular buffer I think that could help for your update list :)
I've noticed in both method 1 and method 3 you have a line, ACQUIRE_MEMORY_BARRIER, which I assume has something to do with multi-threading/race conditions?
Either way, method 2 doesn't have any sleeps which means the following code...
while(true)
{
if(update_cursor->state_ == NOT_FILLED_YET) {
continue;
}
is going to hammer the processor. The typical way to do this kind of producer/consumer task is to use some kind of semaphore to signal to the reader that the update list has changed. A search for producer/consumer multi threading should give you a large number of examples. The main idea here is that this allows the thread to go to sleep while it's waiting for the update_cursor->state to change. This prevents this thread from stealing all the cpu cycles.
The answer was tricky to figure out, and to be fair would be hard with the information I presented though if anyone actually compiled the source code I provided they'd have a fighting chance ;) I said that "found via update list" was printed after every latency log message, but this wasn't actually true -- it was only true for as far as I could scrollback in my terminal. At the very beginning there were a slew of updates found without using the update list.
The issue is that between the time when I set my starting point in the update list and my starting point in each of the data lists, there is going to be some lag because these operations take time. Remember, the lists are growing the whole time this is going on. Consider the simplest case where I have 2 data lists, A and B. When I set my starting point in the update list there happen to be 60 elements in it, due to 30 updates on list A and 30 updates on list B. Say they've alternated:
A
B
A
B
A // and I start looking at the list here
B
But then after I set the update list to there, there are a slew of updates to B and no updates to A. Then I set my starting places in each of the data lists. My starting points for the data lists are going to be after that surge of updates, but my starting point in the update list is before that surge, so now I'm going to check for a bunch of updates without finding them. The mixed approach above works best because by iterating over all the elements when it can't find an update, it quickly closes the temporal gap between where the update list is and where the data lists are.