Databases
StackOverflow2013
Postsdbo
POHow to implement memmove in standard C without an intermediate copy?
 

Note that there are some explanatory texts on larger screens.

plurals
  1. POHow to implement memmove in standard C without an intermediate copy?
    primarykey
    data
    text
    <p>From the man page on my system:</p> <blockquote> <p>void *memmove(void *dst, const void *src, size_t len); </p> <p>DESCRIPTION<br> The memmove() function copies len bytes from string src to string dst.<br> <strong>The two strings may overlap</strong>; the copy is always done in a non-destructive<br> manner. </p> </blockquote> <p>From the C99 standard:</p> <blockquote> <p>6.5.8.5 When two pointers are compared, the result depends on the relative locations in the address space of the objects pointed to. If two pointers to object or incomplete types both point to the same object, or both point one past the last element of the same array object, theycompare equal. If the objects pointed to are members of the same aggregate object, pointers to structure members declared later compare greater than pointers to members declared earlier in the structure, and pointers to array elements with larger subscript values compare greater than pointers to elements of the same array with lower subscript values. All pointers to members of the same union object compare equal. If the expression <code>P</code> points to an element of an array object and the expression Q points to the last element of the same array object, the pointer expression <code>Q+1</code> compares greater than <code>P</code>. In all other cases, the behavior is <strong>undefined</strong>.</p> </blockquote> <p>The emphasis is mine.</p> <p>The arguments <code>dst</code> and <code>src</code> can be converted to pointers to <code>char</code> so as to alleviate strict aliasing problems, but is it possible to compare two pointers that may point inside different blocks, so as to do the copy in the correct order in case they point inside the same block?</p> <p>The obvious solution is <code>if (src &lt; dst)</code>, but that is undefined if <code>src</code> and <code>dst</code> point to different blocks. "Undefined" means you should not even assume that the condition returns 0 or 1 (this would have been called "unspecified" in the standard's vocabulary).</p> <p>An alternative is <code>if ((uintptr_t)src &lt; (uintptr_t)dst)</code>, which is at least unspecified, but I am not sure that the standard guarantees that when <code>src &lt; dst</code> is defined, it is equivalent to <code>(uintptr_t)src &lt; (uintptr_t)dst)</code>. Pointer comparison is defined from pointer arithmetic. For instance, when I read section 6.5.6 on addition, it seems to me that pointer arithmetic could go in the direction opposite to <code>uintptr_t</code> arithmetic, that is, that a compliant compiler might have, when <code>p</code> is of type <code>char*</code>:</p> <pre><code>((uintptr_t)p)+1==((uintptr_t)(p-1) </code></pre> <p>This is only an example. Generally speaking very little seems to be guaranteed when converting pointers to integers.</p> <p>This is a purely academic question, because <code>memmove</code> is provided together with the compiler. In practice, the compiler authors can simply promote undefined pointer comparison to unspecified behavior, or use the relevant pragma to force their compiler to compile their <code>memmove</code> correctly. For instance, <a href="http://www.student.cs.uwaterloo.ca/~cs350/common/os161-src-html/memmove_8c-source.html" rel="nofollow noreferrer">this implementation</a> has this snippet:</p> <pre><code>if ((uintptr_t)dst &lt; (uintptr_t)src) { /* * As author/maintainer of libc, take advantage of the * fact that we know memcpy copies forwards. */ return memcpy(dst, src, len); } </code></pre> <p>I would still like to use this example as proof that the standard goes too far with undefined behaviors, if it is true that <code>memmove</code> cannot be implemented efficiently in standard C. For instance, no-one ticked when answering <a href="https://stackoverflow.com/questions/3572309/memmove-implementation-in-c">this SO question</a>.</p>
    singulars
    1. PO
      singulars
      1. PTAnswer
    1. This table or related slice is empty.
    1. PTQuestion
    1. USCommunity
    1. USPascal Cuoq
    plurals
    1. PL
      singulars
      1. LTLinked
    1. PL
      singulars
      1. LTLinked
    2. PL
      singulars
      1. LTLinked
    3. PL
      singulars
      1. LTLinked
    1. This table or related slice is empty.
    1. PO
      singulars
      1. PTAnswer
    2. PO
      singulars
      1. PTAnswer
    3. PO
      singulars
      1. PTAnswer
    1. VO
      singulars
      1. POHow to implement memmove in standard C without an intermediate copy?
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. POHow to implement memmove in standard C without an intermediate copy?
      1. This table or related slice is empty.
      1. VTUpMod
    3. VO
      singulars
      1. POHow to implement memmove in standard C without an intermediate copy?
      1. This table or related slice is empty.
      1. VTUpMod
    1. CO6.5.8.5 seems very badly worded, one way or another. It starts with a simple statement: "he result depends on the relative locations in the address space of the objects pointed to." THEN it runs off all prescriptive without being at all clear what happens if one of the prescriptive rules conflicts with the relative-locations-in-address-space rule. Is it trying to prescribe how structs should be laid out, or how pointers should compare?
      singulars
      1. POHow to implement memmove in standard C without an intermediate copy?
      1. USChris Becke
    2. CO+1 for well-explained and interesting question.
      singulars
      1. POHow to implement memmove in standard C without an intermediate copy?
      1. USR..
    3. COThis is not only academic, a blitter with support for overlapping regions has exactly this issue, see for example my patch at http://www.winehq.org/pipermail/wine-patches/2008-March/051766.html (it uses memmove for each line, and an outer for loop with step and starting lines selected by a pointer comparison).
      singulars
      1. POHow to implement memmove in standard C without an intermediate copy?
      1. USninjalj
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload