Hello, Rajat!
You wrote to "Roman Mashak" <mrv@xxxxxxxxxxxxx> on Wed, 31 Jan 2007 13:30:32
+0530:
RJ> Yes you are right about the firware. Lets talk about IRQ. The firmware
RJ> only finds out which irq lines a device can use and sets in its config
RJ> space, an IRQ number that this device will use i.e. the IRQ that the
RJ> device should generate when it wants to interrupt the CPU.
Thank you very much, it's much clear now. I have one more question about the
other hardware relevant fields in 'struct net_device': 'base_addr' and
'[r]mem_start/end'. Why is 'base_addr' recommended to be assigned during
device probe, and not later when we can be sure we do need this resource?
And I didn't notice '8139too' driver uses '[r]mem_start', '[r]mem_end',
isn't it mandatory?
RJ> So, from device's perspective it is set which IRQ should it generate,
RJ> but the kernel is yet not aware of it. In fact, the kernel only
RJ> allocates this IRQ to this device (or rather hooks the ISR to this IRQ
RJ> number) when you call request_irq().
RJ> In a scenario where the IRQs are not shared, that IRQ could be used by
RJ> another device till your device needs it. And hence the need for delay
RJ> in IRQ allocation from kernel's perspecive.
---
Best regards, Roman Mashak
--
Kernelnewbies: Help each other learn about the Linux kernel.
Archive: http://mail.nl.linux.org/kernelnewbies/
FAQ: http://kernelnewbies.org/faq/
