On 14 Aug 2015, at 02:44, Theo de Raadt dera...@cvs.openbsd.org wrote:
Regarding other strict-alignment architectures, em(4) is probably one
of the more popular gigabit ethernet options for those architectures
that have PCI slots. I don't think any of these machines are severely
memory
On 11/08/15(Tue) 16:53, Mark Kettenis wrote:
Like ix(4), em(4) hardware doesn't provide an easy/efficient way to
guarantee alignment of protocol headers for received mbufs. The
current code makes an attempt to m_adj() the mbuf if the maximum
hardware frame size is smaller than the cluster
Martin Pieuchot wrote:
How many sparc64 come with em(4)? Can we assume that the amount of
wasted memory on such system is acceptable? What about other strict-
alignment architectures?
just(? mostly?) t5120. mine has 32gb in it. it is, or could be, a popular
openbsd machine. it's also new
From: Ted Unangst t...@tedunangst.com
Date: Thu, 13 Aug 2015 10:04:55 -0400
Martin Pieuchot wrote:
How many sparc64 come with em(4)? Can we assume that the amount of
wasted memory on such system is acceptable? What about other strict-
alignment architectures?
just(? mostly?)
On 13/08/15(Thu) 16:43, Mark Kettenis wrote:
From: Ted Unangst t...@tedunangst.com
Date: Thu, 13 Aug 2015 10:04:55 -0400
Martin Pieuchot wrote:
How many sparc64 come with em(4)? Can we assume that the amount of
wasted memory on such system is acceptable? What about other
Regarding other strict-alignment architectures, em(4) is probably one
of the more popular gigabit ethernet options for those architectures
that have PCI slots. I don't think any of these machines are severely
memory starved, but memory might be limited to something like 256MB of
physical
Like ix(4), em(4) hardware doesn't provide an easy/efficient way to
guarantee alignment of protocol headers for received mbufs. The
current code makes an attempt to m_adj() the mbuf if the maximum
hardware frame size is smaller than the cluster size. But ever since
we changed our drivers to