Messages in this thread | | | Date | Tue, 17 Oct 2000 22:06:35 +0200 | From | Andrea Arcangeli <> | Subject | Re: mapping user space buffer to kernel address space |
| |
On Tue, Oct 17, 2000 at 12:27:30PM -0700, David S. Miller wrote: > Hint: smp_flush_tlb_page() > > Current kiobufs never need to do that, under any circumstances. > This is not by accident.
I don't understand. flush_tlb_page() done in the context of a thread won't care about the state of the physical page. I don't see how it's related to kiobufs.
> I don't see what the big deal is in requiring that user applications > do their own locking to protect page contents being modified in one > thread while a direct I/O from it is happening in another thread. I
We never talked about this issue in the previous emails. That is required regardless of how we solve the MM corruption.
> also don't see why any bug with kiobufs can't be fixed without the > expensive and complex pinning.
IMHO pinning the page in the pte is less expensive and less complex than making rawio and the VM aware of those issues. (remap_page_range is so clean implementation exactly because it pins the page into the pte)
Andrea - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
| |