audidude

28 ms·

271.
by audidude 17y ago
if you work for a company where writing extensions to the server is common-place, i would say the code quality of cherokee is a strong feature as well.
272.
by audidude 17y ago
http://www.cherokee-project.com/benchmarks.html pretty easy to find them.
273.
by audidude 17y ago
keep a journal. whether or not you post that online is up to you.
274.
by audidude 17y ago
Start reading code. Every day, every free moment. Find an app in Linux which you enjoy using or think would be hard to write. Read it from main() to exit().
275.
by audidude 17y ago
Neat hack, but I far more enjoy glib's g_test_*() infrastructure. It also integrates well with `make' and can print out reports or store them for future comparisons.
276.
by audidude 17y ago
ip addresses are always stored in network-byte-order
277.
by audidude 17y ago
Something has to load it from disk into the cache line ..
278.
by audidude 17y ago
Thats fair. I was a bit narrow sighted in my vision of high-performance scenarios.
279.
by audidude 17y ago
You will often have to do a page-fault for both, so loading wont be too much an issue. However, you have to get it into an integer to compare it to begin with. So the string method is purely overhead.
280.
by audidude 17y ago
Read my post again. I said store . There is no reason to store in a compiled format for humans.
281.
by audidude 17y ago
Your hunch would be wrong. #include <stdio.h> int main (int argc, char *argv[]) { unsigned int ip = 3232235777u; printf ("%d.%d.%d.%d\n", (ip & 0xFF000000) >> 24, (
282.
by audidude 17y ago
why would you ever store a 32-bit ip address on disk or memory in anything other than a 32-bit word
283.
by audidude 17y ago
Don't get me wrong, i love mono and vala and use them both regularly. But vala doesn't have stack protection, etc which is quite useful for certain situations.
284.
by audidude 17y ago
If you don't share the client sockets at a level that can distribute work to all of the processes you will get an unbalanced number of connection:keep-alives amongst the workers.
285.
by audidude 17y ago
That and IIS7's integrated pipeline.
286.
by audidude 17y ago
I tried to argue that was an implementation detail with sacrifices no matter which direction you go. One could also argue that its problematic to deal with run-away runtimes with memory issues. The GC does provide an interesting problem I
287.
by audidude 17y ago
Copying my comment from my comment page. Well like I mentioned I believe that is an implementation detail. For example, if you have the subprocess you have two fundamental designs. The first being where the master process still manages the
288.
by audidude 17y ago
Quit my job, bootstrapping, starting today :-)
289.
by audidude 17y ago
and if i wasn't clear, the reason this is bad is because you wake up the cpu every 10ms to check. this is quite bad for power consumption, especially if it is a critical path in your framework.
290.
by audidude 17y ago
Instead, use a "message reply" concept. For example, for your data you could send a tuple with a callback or a condition. Then block on the condition, and have the lambda/delegate in the idle thread signal the condition. Make sure you hold
291.
by audidude 17y ago
Neat idea, but checkout that horrible horrible synchronous message pass to the gtk idle thread ... def execute((R, function, args, kwargs)): R.result = apply(function, args, kwargs) while R.result is NoResult:
292.
by audidude 18y ago
I should also add that messages are ref-counted. The only time they will be serialized is when passing between processes or hosts (which I haven't written yet but is a goal).
293.
by audidude 18y ago
There is nothing preventing Iris from working with libev or libevent. It's primary focus is how to handle work after your event loop. For example, you can saturate your network card on very little use of a single cpu. Iris focuses on how to
294.
Iris, concurrency library for C (NSOperation, Twisted deffereds, and CCR in one)(github.com)
13 points by audidude 18y ago|5 comments