On Sun, 2015-07-26 at 17:47 +0100, Andrew Cooper wrote:
> For 4.7 (which happens to coincide with the splitting up of libxc), I
> recommend introducing xc_unmap()
I thought I'd done that in my latest series, but looking for the
precise name now I see that there is no such function and
xenforeignme
On 24/07/2015 23:40, Wei Liu wrote:
> Originally there was only one counter to keep track of pages. It was
> used erroneously to keep track of how many pages were mapped and how
> many pages needed to be send. In the end munmap(2) always has 0 as the
> length argument, which resulted in leaking the
On July 24, 2015 6:40:27 PM EDT, Wei Liu wrote:
>Originally there was only one counter to keep track of pages. It was
>used erroneously to keep track of how many pages were mapped and how
>many pages needed to be send. In the
sent
> end munmap(2) always has 0 as the
>length argument, which resu
Originally there was only one counter to keep track of pages. It was
used erroneously to keep track of how many pages were mapped and how
many pages needed to be send. In the end munmap(2) always has 0 as the
length argument, which resulted in leaking the mapping.
This problem is discovered on 32b
