Systems software technology and business
Jamie Kitman’s look at the twisted path Toyota followed to it’s current difficulties inspired me to think about software and money – two topics I spend way too much time thinking about. As a purely disinterested observer (ahem) it has come to my attention, repeatedly, that manufacturing companies undervalue, underinvest in, and undertest software. On [...]
Re: Schedule idle
MOLNAR Ingo (mingo@chiara.csoma.elte.hu)
Wed, 11 Nov 1998 04:09:32 +0100 (CET)
[...]
> _please_ We can do better than this. Only semaphores (not spinlocks) need
> to have the priority inheritance. [...]
nope there are _not_ only semaphores, but many other types of locks.
> [...] This can be done with lists off the
> semaphore and tasks… [...]
it’s _not_ easy [...]
An interesting paper appearing in ASPLOS proceedings provides a “deterministic” locking method
Kendo enforces a deterministic interleaving of lock acquisitions and specially declared non-protected reads through a novel dynamically load-balanced deterministic scheduling algorithm. The algorithm tracks the progress of each thread using performance counters to construct a deterministic logical time that is used to compute an [...]
Mr. Myhrvold: All of this fear is from people who have guilty knowledge of their own actions. There are lots of major tech companies that grew from zero to gigantically successful in a very short period of time without investing in their own inventions. They got there by using other people’s inventions.
[cite]
No kidding.
Most of the new draft of the Concurrent Programs paper has to do with trying to specify problems and solutions in synchronization via an atomic “compare and swap” operation. Even these operations are surprisingly complicated once put under the microscope – or not so surprisingly complicated if you think about the details of using or [...]