On Mon, Oct 12, 2015 at 10:50:07AM +0200, Karlsson & Wang wrote:
> Yes it works great! Do you think this should be part of standard behaviour?
I talked to Seb about this change and we are uncomfortable putting it in
2.7 since it is an unnecessary change for Mesa boards.
However, I've prepared a branch
"(origin/)jepler/hm2-explicit-queue-write" which should be semantically
the same as the patch I offered earlier in this thread, which *you*
could merge to your own 2.7ish branch for your use and which I will
merge to our master branch if your testing produces a positive result.
I reworked it slightly because the first version created *two* ways to
gather up reads (explicit and implicit); the new version gets rid of the
old implicit way. I did not test it on real hardware, I only
compile-tested it.
Jeff
The following changes since commit 71b31967e84d5e3306d447073d7a1843436ac1b3:
LinuxCNC 2.7.1 release (2015-10-17 21:07:59 -0600)
are available in the git repository at:
git://git.linuxcnc.org/git/linuxcnc.git jepler/hm2-explicit-queue-write
for you to fetch changes up to 66ca88f741a74a0e43f4bcf627c8dece96ef2d6d:
hm2_eth: explicitly queue up writes from force_write (2015-10-27 20:27:05
-0500)
----------------------------------------------------------------
Jeff Epler (2):
hm2_eth: make write queueing explicit, not implicit
hm2_eth: explicitly queue up writes from force_write
src/hal/drivers/mesa-hostmot2/hm2_eth.c | 2 +-
src/hal/drivers/mesa-hostmot2/hostmot2-lowlevel.h | 2 ++
src/hal/drivers/mesa-hostmot2/hostmot2.c | 6 ++++++
3 files changed, 9 insertions(+), 1 deletion(-)
------------------------------------------------------------------------------
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users
