[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: VIRT_ADDR_VARIES
From: |
Paul Eggert |
Subject: |
Re: VIRT_ADDR_VARIES |
Date: |
Thu, 10 Nov 2011 08:50:39 -0800 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:7.0) Gecko/20110927 Thunderbird/7.0 |
On 11/10/11 08:43, Eli Zaretskii wrote:
> But why would someone do such a thing?
Performance.
Comparing against my_edata takes just one instruction.
Comparing against both the start and the end of 'pure'
takes two comparisons and some glue instructions and
typically includes a conditional branch. The optimization
I've suggested in this thread prunes away most of this
underbrush, but it's still one more instruction
(a 'subtract') over the my_edata approach.
- Re: VIRT_ADDR_VARIES, (continued)
- Re: VIRT_ADDR_VARIES, Andreas Schwab, 2011/11/08
- Re: VIRT_ADDR_VARIES, Paul Eggert, 2011/11/09
- Re: VIRT_ADDR_VARIES, Andreas Schwab, 2011/11/09
- Re: VIRT_ADDR_VARIES, Paul Eggert, 2011/11/10
- Re: VIRT_ADDR_VARIES, Andreas Schwab, 2011/11/10
- Re: VIRT_ADDR_VARIES, Paul Eggert, 2011/11/10
- Re: VIRT_ADDR_VARIES, Eli Zaretskii, 2011/11/10
- Re: VIRT_ADDR_VARIES, Andreas Schwab, 2011/11/10
- Re: VIRT_ADDR_VARIES, Paul Eggert, 2011/11/10
- Re: VIRT_ADDR_VARIES, Eli Zaretskii, 2011/11/10
- Re: VIRT_ADDR_VARIES,
Paul Eggert <=
- Re: VIRT_ADDR_VARIES, Eli Zaretskii, 2011/11/10
- Re: VIRT_ADDR_VARIES, Paul Eggert, 2011/11/13
- Re: VIRT_ADDR_VARIES, Stephen J. Turnbull, 2011/11/09
- Re: VIRT_ADDR_VARIES, Paul Eggert, 2011/11/10
- Re: VIRT_ADDR_VARIES, Andreas Schwab, 2011/11/10
- Re: VIRT_ADDR_VARIES, Paul Eggert, 2011/11/10