map kernel memory to user space and use memory for dma operations

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]


Hi,
  We need to get large chunk of contiguous memory in user space so that an application can directly write to it. The memory will later be used for DMA operations.

ARM uses virtual address to access the cache, now dual mapping, one in kernel and one in user space, will cause coherency problem.

Assume we can take care of coherency issue in software, still TLBs will
have more than one mapping for the same physical page. Is it an issue on ARM? On certain architecture we need to evict the previous TLB entry to avoid machine error.


Kallol 

-------------------------------------------------------------------
List admin: http://lists.arm.linux.org.uk/mailman/listinfo/linux-arm
FAQ:        http://www.arm.linux.org.uk/mailinglists/faq.php
Etiquette:  http://www.arm.linux.org.uk/mailinglists/etiquette.php

[Linux ARM]     [Linux ARM MSM]     [Linux ARM Kernel]     [Fedora ARM]     [IETF Annouce]     [Security]     [Bugtraq]     [Linux]     [Linux OMAP]     [Linux MIPS]     [ECOS]     [Asterisk Internet PBX]     [Linux API]

Add to Google Follow linuxarm on Twitter