On 2017-02-26 23:25, Jakub Jermář wrote:
Btw, it would be interesting to see how these systems compare in an IPC
ping-pong test, such as when you type tester ping_pong in HelenOS.
I find hard to make a meaningful apples-to-apples comparison with bare
IPC tests, as semantics would be lost but can't be ignored.
I remember, back in the day, L4Ka was the champion of ping-pong tests.
That was mainly due to an aggressive use of CPU registers for carrying
the payload. But the moment you tried to send a significant amount of
data (something that wouldn't fit into the registers), its advantage
quickly dissipated.
I integrated a fix for this in mainline revision 2585. I streamlined
your patch a little bit by adding the detection of a too small area and
an attempt to grow it before the actual allocation code. This allowed
me
to get rid of the retry code. Additionally, the area is only grown when
realloc() is sure there are no used blocks that would prevent growing
the block in-place.
Nice! Glad to help.
Sergio.
_______________________________________________
HelenOS-devel mailing list
[email protected]
http://lists.modry.cz/listinfo/helenos-devel
