Hello,
On Tue, Jun 14, 2011 at 10:17:10AM +0300, Péter Ujfalusi wrote:
The latency in most cases comes from the fact, that we are running an embedded system. Number of peripherals are connected via I2C, these drivers are using workqueues to communicate with the IC. Since only one device can communicate through I2C at the time. This is basically the source of the latency. It does not really matter, if the devices are on the same I2C bus or not, it is enough if two work belonging to device, which happens to be on the same I2C bus, and the first work in the queue takes long time to complete (reading back bigger chunk of info, configuring, etc). Even if we could schedule the second work on the other CPU, it will be put waiting till the I2C bus is free, so both CPU core has work assigned, the first is keeping the I2C bus, the other waits for the I2C bus, and the third is waiting to be scheduled (which will be happening, when the first work finished). IMHO the tactile feedback (vibra) should have an excuse to have separate WQ to avoid latency spikes. I agree, that most cases we can use the global wq.
Thanks for the explanation. I have a couple more questions.
* While transferring data from I2C, I suppose the work item is fully occupying the CPU? If so, how long delay are we talking about? Millisecs?
* You said that the if one task is accessing I2C bus, the other would wait even if scheduled on a different CPU. Is access to I2C bus protected with a spinlock?
Also, as it's currently implemented, single threaded wq's effectively bypass concurrency level control. This is an implementation detail which may change in the future, so even if you're seeing lower latency by using a separate single threaded wq, it's an accident and if you require lower latency you should be expressing the requirement explicitly.
Thank you.