Is it okay to add INTR_MPSAFE for all INTR_TYPE_NET drivers?
mbuf and bpf routines are all mp-safe, so it seems that
it is safe to make network device drivers out of Giant lock.
Or is there any unresolved related issues?
--
Kyunghwan Kim
[EMAIL PROTECTED]
To Unsubscribe: send mail to [EMAIL
Kyunghwan Kim writes:
Is it okay to add INTR_MPSAFE for all INTR_TYPE_NET drivers?
NO!
mbuf and bpf routines are all mp-safe, so it seems that
it is safe to make network device drivers out of Giant lock.
Or is there any unresolved related issues?
Yes, the mbuf allocator must
On Tue, Dec 17, 2002 at 02:31:31PM -0500, Andrew Gallatin wrote:
mbuf and bpf routines are all mp-safe, so it seems that
it is safe to make network device drivers out of Giant lock.
Or is there any unresolved related issues?
Yes, the mbuf allocator must occasionally call kmem_malloc(),
On Wed, Dec 18, 2002 at 04:53:00AM +0900, Kyunghwan Kim wrote:
On Tue, Dec 17, 2002 at 02:31:31PM -0500, Andrew Gallatin wrote:
mbuf and bpf routines are all mp-safe, so it seems that
it is safe to make network device drivers out of Giant lock.
Or is there any unresolved related
Kyunghwan Kim writes:
On Wed, Dec 18, 2002 at 04:53:00AM +0900, Kyunghwan Kim wrote:
On Tue, Dec 17, 2002 at 02:31:31PM -0500, Andrew Gallatin wrote:
mbuf and bpf routines are all mp-safe, so it seems that
it is safe to make network device drivers out of Giant lock.
Or is