This series redoes the way time spent waiting for I/O is accounted to the vm_clock.
The current code is advancing qemu_icount before waiting for I/O. Instead, after the patch qemu_icount is left aside (it is a pure instruction counter) and qemu_icount_bias is changed according to the actual amount of time spent in the wait. This is more accurate, and actually works in the iothread case as well. (I started this as an experiment while trying to understand what was going on. But it fixes the bug and does not break the no-iothread case, so hey...). Patch 1 is a cleanup to Edgar's commit 225d02c (Avoid deadlock whith iothread and icount, 2011-01-23). Patch 2 fixes another misunderstanding in the role of qemu_next_deadline. Patches 3 and 4 implement the actual new accounting algorithm. With these patches, iothread "-icount N" doesn't work when the actual execution speed cannot keep up with the requested speed; the execution in that case is not deterministic. It works when the requested speed is slow enough. (side note: all occurrences of wrong braces are in code that is purely moved and/or reindented, _and_ that in fact disappears in subsequent patches. So, I plead myself innocent). Paolo Bonzini (4): do not use qemu_icount_delta in the !use_icount case qemu_next_deadline should not consider host-time timers rewrite accounting of wait time to the vm_clock inline qemu_icount_delta qemu-timer.c | 114 ++++++++++++++++++++++++--------------------------------- 1 files changed, 48 insertions(+), 66 deletions(-) -- 1.7.3.5
