On Fri, Oct 30, 2015 at 06:58:45PM -0400, ira.weiny@xxxxxxxxx wrote: > From: Mitko Haralanov <mitko.haralanov@xxxxxxxxx> > > Expected receives work by user-space libraries (PSM) calling into the driver > with information about the user's receive buffer and have the driver DMA-map > that buffer and program the HFI to receive data directly into it. > > This is an expensive operation as it requires the driver to pin the pages which > the user's buffer maps to, DMA-map them, and then program the HFI. > > When the receive is complete, user-space libraries have to call into the driver > again so the buffer is removed from the HFI, un-mapped, and the pages unpinned. > > All of these operations are expensive, considering that a lot of applications > (especially micro-benchmarks) use the same buffer over and over. > > In order to get better performance for user-space applications, it is highly > beneficial that they don't continuously call into the driver to register and > unregister the same buffer. Rather, they can register the buffer and cache it > for future work. The buffer can be unregistered when it is freed by the user. > > This change implements such buffer caching by making use of the kernel's MMU > notifier API. User-space libraries call into the driver only when the need to > register a new buffer. > > Once a buffer is registered, it stays programmed into the HFI until the kernel > notifies the driver that the buffer has been freed by the user. At that time, > the user-space library is notified and it can do the necessary work to remove > the buffer from its cache. > > Buffers which have been invalidated by the kernel are not automatically removed > from the HFI and do not have their pages unpinned. Buffers are only completely > removed when the user-space libraries call into the driver to free them. This > is done to ensure that any ongoing transfers into that buffer are complete. > This is important when a buffer is not completely freed but rather it is > shrunk. The user-space library could still have uncompleted transfers into the > remaining buffer. > > With this feature, it is important that systems are setup with reasonable > limits for the amount of lockable memory. Keeping the limit at "unlimited" (as > we've done up to this point), may result in jobs being killed by the kernel's > OOM due to them taking up excessive amounts of memory. > > Reviewed-by: Arthur Kepner <arthur.kepner@xxxxxxxxx> > Reviewed-by: Dennis Dalessandro <dennis.dalessandro@xxxxxxxxx> > Signed-off-by: Mitko Haralanov <mitko.haralanov@xxxxxxxxx> > Signed-off-by: Ira Weiny <ira.weiny@xxxxxxxxx> > > --- > Changes from V3: > Reworked based on the removal of the file pointer macros > Split out some prep patches and code clean up > > Changes from V2: > Fix random Kconfig 0-day build error > Fix leak of random memory to user space caught by Dan Carpenter > Separate out pointer bug fix into a previous patch > Change error checks in case statement per Dan's comments > > drivers/staging/rdma/hfi1/file_ops.c | 469 ++--------------- > drivers/staging/rdma/hfi1/hfi.h | 43 +- > drivers/staging/rdma/hfi1/init.c | 5 +- > drivers/staging/rdma/hfi1/trace.h | 132 +++-- > drivers/staging/rdma/hfi1/user_exp_rcv.c | 874 ++++++++++++++++++++++++++++++- > drivers/staging/rdma/hfi1/user_pages.c | 110 +--- > drivers/staging/rdma/hfi1/user_sdma.c | 13 + > include/uapi/rdma/hfi/hfi1_user.h | 14 +- > 8 files changed, 1069 insertions(+), 591 deletions(-) This is still a really big patch, any chance you can break it up into smaller, reviewable parts? I see you add different operations, perhaps break it up into one patch per logical thing? thanks, greg k-h -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
