Skip to contentSkip to conversation

size-in-bytes

PATCH [email protected] 2 participants 2 messages Feb 16, 2002

lore
Share & tools
lore
#0 lore raw

From [email protected] Sat Feb 16 23:05:50 2002

Well I see that you have killed all places where it get's used - that's fine!
But why don't just kill (fix) the initializations as well. You have already:

g->part[minor(dev)].nr_sects;

and friends right there at your hands ;-).

Yes, of course there is a reason this patch is called 02*.
I planned a road to happiness in 92 steps, where each of the steps does something clear and simple, simplifies the tree, beautifies the code, restructures in a clearly necessary way.
Indeed, your suggested steps are also there.
Five or six of these steps found their way into the kernel, (some thanks to Christoph Hellwig) but there is still a long way to go. The present patch is just a rediff of step 02.

I do not know what the best strategy is, these times. I see you and Vojtech do good things to the IDE code, but would myself prefer to do such things in a series of really small steps. That way it is also very clear for Andre what happens.

Andries

#1 lore raw
Quoted 2 lines · [email protected] wrote:

I planned a road to happiness in 92 steps, where each of the steps

Beware I may drive you to ruin if there will be less or more then 92 ;-)...

Quoted 7 lines

does something clear and simple, simplifies the tree, beautifies the code, restructures in a clearly necessary way.
Indeed, your suggested steps are also there.
Five or six of these steps found their way into the kernel, (some thanks to Christoph Hellwig) but there is still a long way to go. The present patch is just a rediff of step 02.

My main concern is simple the following: The broken design was there for already 8 years and the cleanup just goes far too slow for my personal taste if it goes at the same peace of speed as of now...

Quoted 6 lines

I do not know what the best strategy is, these times. I see you and Vojtech do good things to the IDE code, but would myself prefer to do such things in a series of really small steps. That way it is also very clear for Andre what happens.

Well I have no problems to synchronize in small steps. However I have problems with synchronization in microsteps, becouse due to my daily dueties, which btw. have *nothing* with kernel hacking to do, I can hardly go below the resolution of a day.... Or more precisely: some time at afternoon which can vary between:
- nothing
- a whole night
- a whole weekend

depending on:
- my personal mood,
- my druglevel,
- blood preassure,
- my world domination dreams (Andre thinks so apparently :-),
- the current conjugation between saturn and venus,

and so on...

But in fact: I'm flexible and even more happy if somebody, who can certainly devote more time to it then me, just picks the shit^Wsuggestions I call patches up...

In regards of Andre, well unfortunately, after reading his last mail about my employer and what-a-not, well I have some... please excuse me... concerns about his mind-wellth ;-)