Right, there is no concern about overflow. I believe that using a native word is usually more efficient in terms of both performance and code size. As long as the variable is on the stack, it is using up a machine-word-sized stack slot anyway. -----Original Message----- From: Len Brown [mailto:lenb@xxxxxxxxxx] Sent: Tuesday, April 22, 2008 4:23 PM To: Moore, Robert Cc: Alexey Starikovskiy; linux-acpi@xxxxxxxxxxxxxxx Subject: Re: [PATCH 65/73] ACPICA: Fix for extraneous debug message for packages On Tuesday 22 April 2008, Moore, Robert wrote: > >would probably work just as well. Certainly acpi_native_uint was > overkill. > > I disagree. Often more efficient to use machine native word rather than > fooling around with a byte load/clear and store perhaps you mis-read my message. I went with "unsigned", which is 32-bits on both 32 and 64 bit x86. I shouldn't have mentioned that a byte was sufficient -- my point was simply that it is darn unlikely that we'd overflow a capacity of 2^8. There is zero chance we'd overflow 2^32; and thus no practical utility for 2^64 index capacitiy. I don't think that using a 32-bit index is a performance penalty on the hardware we care about. I also don't think that performance is important here. -Len -- To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
