aliguori

12 ms·

61.
by aliguori 15y ago
I'd be surprised if it was 64-bit. 64-bit PV Xen is awfully slow because of exactly what you reference. HVM would be quite a bit better.
62.
by aliguori 15y ago
I think netfront still uses a single transmit queue so you'll only ever see one VCPU cpu bound for intensive I/O.
63.
by aliguori 15y ago
This is a very well known issue (at least among virtualization developers :-)). ESX does a surprisingly good job handling this but both Xen and KVM are still trying to catch up here. The issue is small packet performance. You can isolate
64.
by aliguori 15y ago
You would be surprised at how many software developers have hobbies that involve working with their hands. I spend much of my free time doing wood working and am sure it's something that I would introduce my children to at a young age (as
65.
by aliguori 15y ago
Developer Works is basically just an electronic magazine. Anyone can submit an articles. There are plenty of positive node.js articles such as http://www.ibm.com/developerworks/cloud/library/cl-nodejsclo...
66.
by aliguori 15y ago
It's not so easy because what happens if you have code like this: static File globalFile; void register_file(File aFile) { globalFile = aFile; } void myMethod() { disposable File myFile = new File(somePath);
67.
by aliguori 15y ago
Even within a single culture, it varies widely. Just in the US, look at the opposing views toward capital punishment.
68.
by aliguori 15y ago
Most C++ implementations throw their hands up at setjmp/longjmp/setcontext/makecontext. Even if it appears to work, it's a dangerous set of routines to use in C++.
69.
by aliguori 16y ago
Exokernel was basically a very early form of a hypervisor. It's not totally novel as hypervisors have been around for a very long time but what makes the Exokernel interesting is that it doesn't try to run unmodified guests but rather runs
70.
by aliguori 16y ago
This is where recursive decent parsing really shines. When an invalid sequence is detected deep in the recursion, you don't have enough context to generate the best error message but as you unwind, you can add additional context such that
71.
by aliguori 16y ago
Because there was still a slim chance that it thought it was right. If it's well ahead, it shouldn't bet $0 unless it knows with absolutely confidence that it's wrong.
72.
by aliguori 16y ago
I'm a graduate of UTCS. I only had one class that used Java. Different classes were taught in C, Lisp, Scheme, ML, Haskell, etc. If anything, most students complain about not enough Java. It's certainly the case that startups aren't empha
73.
by aliguori 16y ago
Considering that their "core" is a bunch of code that was ripped off of QEMU, I don't see the creation of "extension packs" as a mechanism to be more attractive to open source developers. What would be attractive is contributing their chan
74.
by aliguori 16y ago
Which is really the odd ball. In a left-to-right language, it's sort of bizarre that we don't write `4 = x' since the rvalue tends to be the more complicated part of the expression. The curious bit is, AT&T assembly syntax does follow
75.
by aliguori 16y ago
memcpy(lvalue, rvalue, size) lvalue == write location rvalue == read location It comes from assignment syntax where the left hand side is the target of the assignment and the right hand side is the source. So this makes a ton of sense in C
76.
by aliguori 16y ago
> * Sexism and unprofessionalism can be avoided without being prudish about the content of your presentation. This is a slide I used in one of my Italian presentations. Actually, even something like this is still considered offensive.
77.
by aliguori 16y ago
This isn't terribly portable FWIW. Not all calling conventions massage data into either a 32-bit or 64-bit argument. In particular, passing structs by value has some interesting rules even in the common calling conventions. It will certai
78.
by aliguori 16y ago
I've found few projects that are even close to schizophrenic. I think the thing that people often miss is that getting a patch that works is really only 50% of the process of getting a bug fix/feature merged into an Open Source project. In
79.
by aliguori 16y ago
Well, it's more like: Binary numbers are a countably infinite set. Decimal numbers are a countably infinite set. You can therefore map binary representation to decimal representation 1-1 with a simple mapping function. OTOH, floating poin
80.
by aliguori 16y ago
Heh, I'm not a suit but I do appreciate being able to connect to a projector. Since the X201 doesn't have a CD-ROM, a docking station is basically a prerequisite IMHO. I love my X201. It's very light but has a full size keyboard. I tend
81.
by aliguori 16y ago
All recent IBM x86 servers use UEFI.
82.
by aliguori 16y ago
No, TCP ACK has nothing to do with this. All that matters is when the NIC has completed DMA'ing the packet. At that point, userspace can be unblocked. Non-blocking sockets only make sense when there's an intermediate buffer.
83.
by aliguori 16y ago
Problem is you don't know which recv() a packet is destined for until you actually receive it and can parse the header and apply all of the filtering rules. But what you describe is exactly how O_DIRECT achieves zero/copy for read/write. Of
84.
by aliguori 16y ago
No. If the act of write() is to construct and transmit and SKB, and SKB can be created from a scatter/gather list of pages plus an offset into the first page. But this implies that you can track the SKBs lifecycle in the kernel such that y
85.
by aliguori 16y ago
I'm assuming that the real objection to an interface like sendfile() is that you shouldn't need to copy_from_user() just to construct an SKB if the user address being referred to is pointing at a file backed VMA. If we could do a zero-copy
86.
by aliguori 16y ago
No, this is not correct. The idea of sendfile() is that you can send a physical address (probably something in the buffer cache) directly to a network adapter in a zero copy manner. That's current impossible with a send()/write() to a fil
87.
by aliguori 16y ago
You typically want to enable the post-update hook on the server by mv '${GITDIR}/hooks/post-update.sample' '${GITDIR}/hooks/post-update'. This makes sure that the repository works with dumb servers which is required for things like http/cgi
88.
by aliguori 16y ago
Right. You have to be very careful to always yield to another function that uses yield. It can become very hairy with deep call chains. The limitation is that Python has to be statically prove that a function is really a generator for yie
89.
by aliguori 16y ago
The analysis is flawed because it merely takes into the account the fact that ebp is callee saved verses using a different register that is caller saved. But ultimately, the register still needs to be spilled somewhere so you're not changi