• Messages sorted by: [ date ][ thread ][ subject ][ author ]
• [ Next ][ Previous ] Maybe in reply to: Jonathan M. Smith

If EROS allows it, the copy of A's data must be restarted from the beginning after the page fault is resolved. Otherwise what B sees is a combination of the part of C's data in the buffer followed by some of A's data. Not good.

Of course, except for the caveat that if C made it to byte 4098 before faulting, A resumed, and A is only sending 4096, bytes 4096 and 4097
(from zero) will contain residual data from C. The length reported to
the recipient will be the length of the completed message: 4096.

shap

• [ Next ] Maybe in reply to: Jonathan M. Smith