|
Namazu for hns による簡易全文検索 詳しくは 詳細指定/ヘルプを参照して下さい |
|||||||||||||||||||||||||||||||||||||||||||||||||
2013年08月30日(金) 旧暦 [n年日記] [更新:"2013/08/30 20:33:49"]#1 [assembler] argument of .align
In the assembly language, we have .align directive.
.align 2This will advance PC location counter to some boundary. The problem is: the semantics of argument. Some assemblers take this value as number of bytes, and implying the number must be a power of 2. Say, .align 5is invalid syntax. 5 is not the power of 2. The other assembler may take the number of bits. If is n, it aligns to 2^n. ＠ Assembler for PReP interprets bits of power of 2.:.align 2will align to 'x xx00' in bit representation. So if we wants to be 'x 0000', .align 4is what we need. ＠ Assember of ofppc interprets bytes of power of 2.:.align 2will align 'x 0000', meaning 2^2 bytes. ＠ Do you believe that ?:
I don't believe above explanation, but my observation of
port-prep follows above rule.
＠ the patch makes PReP to work, ofppc to invalid syntax.:( つっこみ )
|
最近の日記 2024年05月08日 ・comparison on ./buildsh tools 2024年05月06日 ・py-setuptools (python 3.11.9) ・make release took 1 hours and 10 min ・qemu invocation for 10.99.10 2024年05月05日 ・Windows 10 version ・serial connection ・bc bench 2024年05月04日 ・Trial on 10.99.10 ・another version (later trial) to succeed 2024年04月29日 ・dkim | ||