[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [f-cpu] TLB resume
cedric a écrit :
>
> Here is an extract from the TLB discussion that append one month ago.
>
> Michael propose this :
> "
> I've been thinking about the TLB before; IMHO, we need at least the
> following (assuming <n> bits for the page offset):
>
> - virtual address (64-<n> bits)
> - physical address (64-<n> bits)
> - address space identifier (ASI; 8 bits was suggested)
> - supervisor access rights (RWX, 3 bits)
> - user access rights (RWX, 3 bits)
> - valid bit (indicating that the entry is valid)
> - dirty bit (indicating that the page has been written to)
> - used bit (indicating that the page has been accessed)
>
> - page size (4K << size, 6 bits)
-> too much ?
> "
>
> "present bits" has been removed because it's only needed by HW TLB after the
> discussion. I hope I have maid a good resume, but if someone want to add
> something, before I add this to the manual (perhaps we must clarify how to
> access it before).
>
> Cedric
So to resume one more time every one is ok with that.Christophe think
it's better to use supervisor right, so why not ! (just one question :
we want to avoid that user could give kernel adresse to OS call, so we
want to put no-read bit on kernel data page, so how the kernel could
read it's own pages ?)
For the page size, 4k << size is a little bit too much. I prefer
something like previously said:
- tiny for message passing ~4Ko
- usual page for program 64~256 Kb
- For kernel, frame buffer 4-64 Mb
- udge for data base,... 256-1024 Mb
I propose to not fixed the size now but let it to be defined "later".
Could Cédric put that in the manual ?
nicO
*************************************************************
To unsubscribe, send an e-mail to majordomo@seul.org with
unsubscribe f-cpu in the body. http://f-cpu.seul.org/