Today, Rich Felker has published the next release of musl the lightweight, standard conforming C library. He says:
October 14, 2014
September 8, 2014
In two ancient posts I have talked about arrays in modern C, “don’t be afraid of variably modified types” and “VLA as function arguments”. Still there seem to be a lot of people that, perhaps just by bad habit, that prefer to use “fake matrices” instead of real matrices in C. Unfortunately among these people are a lot of university teachers that preach that bad parole to their students. I just try to make a list of the advantages of real matrices, here.
April 2, 2014
I recently reviewed some document on security recommendations where I was baffled by the fact that the code examples were sprinkled with casts all over the place. I had thought that people that are concerned with software security in mind would adhere to one of the most important rules in C programming:
Casts are harmful and evil.
The “evil” here is to be read as reference to black magic. Most uses of cast are merely done in the spirit of “casting a spell” by people that try to quieten their compiler. The sorcerer’s apprentice approach: if I don’t see the evil, it isn’t there.
For me it is evident that every cast punches a hole in C’s type system. So, concerned with code security, we should avoid them as much as possible. But this evidence doesn’t yet seem shared (meaning that it is not so evident 🙂 and I decided to explain things here in more detail.
Casts (explicit conversions) in C come with three different flavors, depending on the cast-to and cast-from type
- pointer to pointer
- pointer to integer or vice versa
- integer to integer
December 18, 2013
As I showed int this post, using > as right angle brackets was not a particularly good idea, but trying to patch this misdesign even makes it worth. After a bit of experimenting I found an expression that is in fact valid for both, C++98 and
C++11, but that has a different interpretation in both languages:
fon< fun< 1 >>::three >::two >::one
So if you have to maintain a large code base with templates that depend on integers that are perhaps produced automatically by some tools, be happy, you will not be out of work for a while: changing your compiler to
C++11 might change the semantics of your code.
December 15, 2013
It is long time that I didn’t look into C++, I have to admit. By coincidence I recently unearthed a hilarious example that I had once written that shows the difficulty of parsing some C++ code, as well as for compilers as for us poor humans. It all starts with the
>> operator that (supposedly until C++11) could cause problems as in the following:
toto< tutu< 3 >> A;
Here the >> is (was) interpreted as `right shift’ operator and thus this code would create a compile time error. C++11 changed this by introducing the possibility that in that case the right-shift-operator-token closes the two template angle brackets. The argument is that shift operators in template arguments are rare (which is probably true) and so this sacrifices some valid uses of that operator for the sake of causing less brain damage to C++ newbies.
October 28, 2013
Let’s take the occasion of the change back from DST here in Europe, not in the US, yet, to look how times are handled in C.
The C standard proposes a large variety of types for representing times:
double and textual representations as
char. It is a bit complicated to find out what the proper type for a particular purpose is, so let me try to explain this.
The first class of “times” can be classified as calendar times, times with a granularity and range as it would typically appear in a human calendar, as for appointments, birthdays and so on. Some of the functions that manipulate these in C99 are a bit dangerous, they operate on global state. Let us have a look how these interact:
August 22, 2013
Sometimes in C it is useful to distinguish if an expression is an “integral constant expression” or a “null pointer constant”. E.g for an object that is allocated statically, only such expressions are valid initializers. Usually we are able to determine that directly when writing an initializer, but if we want to initialize a more complicated
struct with a function like initializer macro, with earlier versions of C we have the choice:
- Use a compiler extension such as gcc’s
- We’d have to write two different versions of such a macro, one for static allocation and one for automatic.
In the following I will explain how to achieve such a goal with C11’s
_Generic feature. I am not aware of a C++ feature that provides the same possibilities. Also, this uses the ternary operator (notably different in C and C++), so readers that merely come from that community should read the following with precaution.
July 15, 2013
Again I had a discussion with someone from a C++ background who claimed that one should use signed integer types where possible, and who also claimed that the unsignedness of
size_t is merely a historical accident and would never be defined as such nowadays. I strongly disagree with that, so I decided to write this up, for once.
What I write here will only work with C, and can possibly extended to C++ and other languages that implement unsigned integer types, e.g good old Pascal had a
February 4, 2013
Somewhat hidden in Annex K, C11 introduces a new term into the C standard, namely runtime-constraint violations. They offer an important change of concept for the functions that are defined in that annex: if such a function is e.g called with invalid parameters, a specific function (called runtime-constraint handler) is called, that could e.g abort the program, or just issue an error message. This is in sharp contrast to the runtime error handling in the rest of the C standard, where the behavior under such errors is mostly undefined (anything may happen then) or sometimes reported to implementation defined behavior (and thus poorly portable and predictable).
Annex K, obscurely coined “Bounds checking interfaces“, introduces some
typedef and a series of replacement functions for many C library functions. The function names in this series are usually derived from the name of the function they replace and by adding the suffix
_s to the function name, e.g the function
qsort gets a “secure” twin interface called
qsort_s, as we have seen in an earlier post.
December 4, 2012
I recently started to implement parts of the “Bounds checking interfaces” of C11 (Annex K) for P99 and observed a nice property of my implementation of
qsort_s. Since for P99 basically all functions are inlined, my compilers (gcc and clang) are able to integrate the comparison functions completely into the sorting code, just as an equivalent implementation in C++ would achieve with