Linux hacker and outdoor explorer - particularly of obscure islets and holes in the ground.
|
2d
|
|
awarded | Revival |
|
Jul
7 |
|
comment |
Demonstrate effects of summation order @chux, you're right: these values are all positive. Your observation could form the basis of a review. But remember that the purpose of the code was to make the accumulation of error visible, rather than to present a fully robust summation. The description even says, "the adaptive pairwise version only works when all inputs are positive." |
|
Jul
5 |
|
awarded | Yearling |
|
Jul
5 |
|
reviewed | Approve suggested edit on Excel sheet manipulation program |
|
Jul
4 |
|
revised |
Error handling for singly linked list in C Spelling and grammar |
|
Jul
1 |
|
awarded | Necromancer |
|
Jun
28 |
|
revised |
What is the purpose of the M1 pin on a Z80 Spelling and grammar |
|
Jun
28 |
|
revised |
Do IDE data lines need pull-up resistors? Spelling and grammar |
|
Jun
28 |
|
revised |
Verbose sleep command that displays pending time seconds/minutes? Typo fix |
|
Jun
28 |
|
awarded | Copy Editor |
|
Jun
27 |
|
comment |
Writing a generic makefile for C projects Sorry for the very short review - I have a lot going on until late July. I hope you get a more thorough answer as well! |
|
Jun
27 |
|
reviewed | Leave Open A thread-safe performant Money Transfer API in Java |
|
Jun
27 |
|
revised |
What actual purpose do accent characters in ISO-8859-1 and Windows 1252 serve? Spelling and grammar |
|
Jun
27 |
|
comment |
Writing a generic makefile for C projects That can be a problem, as they share the same names and we can end up linking a mismatching set of objects. I guess it depends whether their public interfaces depend on NDEBUG etc. (when we add that to CFLAGS release builds). Instead of rewriting %.o: %c rules, I find it better to start a sub-make in the output directory (something like +$(MAKE) $(builddir) -f ../Makefile $@ VPATH=$(PWD)). Then you can build as many configurations as you want, from a single source directory which doesn't need to be writeable by the user doing the build.
|
|
Jun
27 |
|
comment |
Sorting and filtering using dynamic parameters To help reviewers give you better answers, we need to know what the code is intended to achieve. Please add sufficient context to your question to describe the purpose of the code. We want to know why much more than how. The more you tell us about what your code is for, the easier it will be for reviewers to help you. Also, [edit] the title to simply summarise the task, rather than your concerns about the code. |
|
Jun
27 |
|
answered | Writing a generic makefile for C projects |
|
Jun
27 |
|
comment |
An Attempt at Creating a Type-Safe, Generic Swap Macro With Compile Time Error Checking ... The buffer size would need tuning from that 16 according to profile results; I don't know whether uintmax_t is a good way to get it aligned; for small transfers it's probably faster to use the existing char-by-char implementation (again, benchmark to determine the cut-over point, if any - that's probably very platform-dependent).
|
|
Jun
27 |
|
comment |
An Attempt at Creating a Type-Safe, Generic Swap Macro With Compile Time Error Checking No, library memcpy() and family tend to be targeted towards specific compilers and specific processors (often not even written in C) rather than being portable code. To be honest, I wouldn't attempt one of those. You might get a performance boost by using memcpy() to swap a substantial chunk at a time via a buffer that's aligned (e.g. uintmax_t buf[16]; while (psize) { size_t size = min(sizeof buf, psize); memcpy(&buf, a, size); memcpy(a, b, size); memcpy(b, &buf, size); a += size; b += size; psize -= size); }. Obviously there's still work to do there:...
|
|
Jun
21 |
|
revised |
Calculate overlapping time periods in PowerQuery Improve description of purpose in title |
|
Jun
21 |
|
revised |
Project Euler 127 - abc-hits edited tags |