1 |
>>>>> "MH" == Michael Haubenwallner <haubi@g.o> writes: |
2 |
|
3 |
MH> I've heard a colleague of mine debugged for 50(!) hours after moving |
4 |
MH> some quite old application to some recent Linux before he replaced a |
5 |
MH> memcpy by memmove, so this did ring some bells. |
6 |
|
7 |
MH> However, now he said this was on Ubuntu 10.04.1 LTS, having |
8 |
MH> glibc-2.11, so this might have been unrelated indeed. |
9 |
|
10 |
Check the archives of the glibc lists, as well as its bug db. |
11 |
|
12 |
There has been quite a bit of discussion there on that issue. |
13 |
|
14 |
It was added for sse3 some time ago; I beleive it was Intel engineers |
15 |
who contributed it for sse3, showing that it was inedeed faster on their |
16 |
chips to start at the high point and decrement the counter rather than |
17 |
starting at the low point and incrementing. |
18 |
|
19 |
The discussion in their lists does a better job of documenting the issue. |
20 |
|
21 |
-JimC |
22 |
-- |
23 |
James Cloos <cloos@×××××××.com> OpenPGP: 1024D/ED7DAEA6 |