Enable indirect segment I/Os by default in the Xen blkfront driver when

running on EC2.  Due to improvements in EC2, the performance penalty which
was present on some EC2 instances no longer exists, and enabling this
feature now consistently yields ~20% higher throughput with equal or lower
latency.

Reverts:	r286063
Approved by:	re (gjb)
MFC after:	2 weeks
Relnotes:	Improved disk throughput on EC2
This commit is contained in:
Colin Percival 2016-06-29 23:56:24 +00:00
parent 4b61754c68
commit 4d1afa14d5
Notes: svn2git 2020-12-20 02:59:44 +00:00
svn path=/head/; revision=302288

View file

@ -72,11 +72,6 @@ vm_extra_pre_umount() {
# nodes, but apply the workaround just in case.
echo 'hw.broken_txfifo="1"' >> ${DESTDIR}/boot/loader.conf
# Some EC2 instances suffer a significant (~40%) reduction in
# throughput when using blkif indirect segment I/Os. Disable this
# by default for now.
echo 'hw.xbd.xbd_enable_indirect="0"' >> ${DESTDIR}/boot/loader.conf
# Some EC2 instances run on a version of Xen which has a bug relating
# to the migration of MSI-X interrupts; this is visible as SR-IOV
# networking (aka. "EC2 Enhanced Networking") not being able to pass