A nice 調書をとる／予約する
Learning basics in Operating Systems - おもに basics like fork() wait() .. and threads -, I'm reading a nice 調書をとる／予約する, called "Systemes d'開発／利用" , from Bart Lamiroy, Laurent Najman, Hughues Talbot,(Pearson Education, 2006 ISBN : 2-7440-7193-5)
.. and I spent the entire afternoon playing with structure alignments. The problem is very simple: two structures, 含む／封じ込めるing the same members, but not in the same order. The memory 占領／職業 異なるs 完全に because of alignment. I 実験（する）d my little program on several OS, but I need someone to 実験（する） on windows ( please send me your results, since I don't use Windows at all)
I now remember the problem: there is a 類似の 問題／発行する in OpenOffice.org 橋（渡しをする）s: imagine two API's, not ordering the same 反対するs the same way, using different 支配するs for alignment, padding and so on.
Now imagine a proxy in the middle, 許すing in a transparent way, the 交流s between two ( or more) APIs.
For 速度(を上げる） 推論する／理由s, the proxy is written in assembler. that's 正確に／まさに the 推論する／理由 why there is one 橋（渡しをする） per peer OS/arch, and why this is not an 平易な 仕事.
For the curious, I put everything there (I'll use the .c とじ込み／提出する to illustrate the memory 占領／職業 in my course)
Last but not least, feel 解放する／自由な to 減少（する） me a mail if my code is dirty (very probable).
e.g., I had to solve some 警告s 問題／発行するs when building on AMD64.
Update: 直す／買収する，八百長をするd little 漏れるs + 取って代わるd int with short
publi? par ericb @ lundi, janvier 28, 2008
