The main assumption of the async request design is that the file system adds block requests to the block device queue asynchronously without waiting for completion (see the Rationale section of https://wiki.linaro.org/WorkingGroups/Kernel/Specs/StoragePerfMMC-async-req). We found out that in case of sequential read operations this is not the case, due to the read ahead mechanism. When mmcqt reports on completion of a request there should be a context switch to allow the insertion of the next read ahead BIOs to the block layer. Since the mmcqd tries to fetch another request immediately after the completion of the previous request it gets NULL and starts waiting for the completion of the previous request. This wait on completion gives the FS the opportunity to insert the next request but the MMC layer is already blocked on the previous request completion and is not aware of the new request waiting to be fetched. This patch fixes the above behavior and allows the async request mechanism to work in case of sequential read scenarios. The main idea is to replace the blocking wait for a completion with an event driven mechanism and add a new event of new_request. When the block layer notifies the MMC layer on a new request, we check for the above case where MMC layer is waiting on a previous request completion and the current request is NULL. In such a case the new_request event will be triggered to wakeup the waiting thread. MMC layer will then fetch the new request and after its preparation will go back to waiting on the completion. Our tests showed that this fix improves the read sequential throughput by 16%. Konstantin Dorfman (1): mmc: fix async request mechanism for sequential read scenarios drivers/mmc/card/block.c | 30 ++++----- drivers/mmc/card/queue.c | 34 ++++++++- drivers/mmc/card/queue.h | 1 + drivers/mmc/core/core.c | 165 ++++++++++++++++++++++++++++++++++++++++++++++ include/linux/mmc/card.h | 12 ++++ include/linux/mmc/core.h | 21 ++++++ 6 files changed, 243 insertions(+), 20 deletions(-) -- 1.7.6 -- Konstantin Dorfman, QUALCOMM ISRAEL, on behalf of Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation -- To unsubscribe from this list: send the line "unsubscribe linux-mmc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
