| www.delorie.com/archives/browse.cgi | search |
| Sender: | nate AT cartsys DOT com |
| Message-ID: | <361979EC.B55399B2@cartsys.com> |
| Date: | Mon, 05 Oct 1998 19:01:16 -0700 |
| From: | Nate Eldredge <nate AT cartsys DOT com> |
| X-Mailer: | Mozilla 4.05 [en] (X11; I; Linux 2.0.35 i486) |
| MIME-Version: | 1.0 |
| To: | Omar Hawk <omarhawk AT t-online DOT de> |
| CC: | djgpp AT delorie DOT com |
| Subject: | Re: allocating dword aligned memory with __dpmi_allocate_memory |
| References: | <6v8ag4$922$1 AT news02 DOT btx DOT dtag DOT de> |
| Reply-To: | djgpp AT delorie DOT com |
Omar Hawk wrote: > > Hello there, > > I want to write an assembler function that copies mem using the string > instructions MOVSD & MOVSW. I'd like to use MOVSD to store 2 words a > time. How can I allocate DWORD ALIGNED memory using > __dpmi_allocate_memory? This would be very helpful if one is using a > 16bit vesa mode: I hope to double the speed by a factor of 2. Why can't you just use `malloc'? The memory block it returns will be at least dword aligned. (If you don't trust it, you could always do: void *p = malloc(size + 3); p = (void *)((((unsigned long)p)+3)&~4); ) -- Nate Eldredge nate AT cartsys DOT com
| webmaster | delorie software privacy |
| Copyright © 2019 by DJ Delorie | Updated Jul 2019 |