wireguard-linux/drivers/mmc/core/block.c, branch jd/unified-crypt-queue

mmc: core: Re-work the code for eMMC sanitize

2020-03-26T13:45:31Z

The error path for sanitize operations that completes with -ETIMEDOUT, is tightly coupled with the internal request handling code of the core. More precisely, mmc_wait_for_req_done() checks for specific sanitize errors. This is not only inefficient as it affects all types of requests, but also hackish. Therefore, let's improve the behaviour by moving the error path out of the mmc core. To do that, retuning needs to be held while running the sanitize operation. Moreover, to avoid exporting unnecessary symbols to the mmc block module, let's move the code into the mmc_ops.c file. While updating the actual code, let's also take the opportunity to clean up some of the mess around it. Signed-off-by: Ulf Hansson Link: https://lore.kernel.org/r/20200316152152.15122-1-ulf.hansson@linaro.org

mmc: core: Enable re-use of mmc_blk_in_tran_state()

2020-03-24T13:35:39Z

To allow subsequent changes to re-use the code from the static function mmc_blk_in_tran_state(), let's move it to a public header. While at it, let's also rename it to mmc_ready_for_data(), as to try to better describe its purpose. Signed-off-by: Ulf Hansson Tested-by: Baolin Wang Tested-by: Ludovic Barre Reviewed-by: Ludovic Barre Link: https://lore.kernel.org/r/20200204085449.32585-7-ulf.hansson@linaro.org

mmc: Add MMC host software queue support

2020-03-24T13:35:39Z

Now the MMC read/write stack will always wait for previous request is completed by mmc_blk_rw_wait(), before sending a new request to hardware, or queue a work to complete request, that will bring context switching overhead and spend some extra time to poll the card for busy completion for I/O writes via sending CMD13, especially for high I/O per second rates, to affect the IO performance. Thus this patch introduces MMC software queue interface based on the hardware command queue engine's interfaces, which is similar with the hardware command queue engine's idea, that can remove the context switching. Moreover we set the default queue depth as 64 for software queue, which allows more requests to be prepared, merged and inserted into IO scheduler to improve performance, but we only allow 2 requests in flight, that is enough to let the irq handler always trigger the next request without a context switch, as well as avoiding a long latency. Moreover the host controller should support HW busy detection for I/O operations when enabling the host software queue. That means, the host controller must not complete a data transfer request, until after the card stops signals busy. From the fio testing data in cover letter, we can see the software queue can improve some performance with 4K block size, increasing about 16% for random read, increasing about 90% for random write, though no obvious improvement for sequential read and write. Moreover we can expand the software queue interface to support MMC packed request or packed command in future. Reviewed-by: Arnd Bergmann Signed-off-by: Baolin Wang Signed-off-by: Baolin Wang Link: https://lore.kernel.org/r/4409c1586a9b3ed20d57ad2faf6c262fc3ccb6e2.1581478568.git.baolin.wang7@gmail.com Signed-off-by: Ulf Hansson

mmc: block: Use generic_cmd6_time when modifying INAND_CMD38_ARG_EXT_CSD

2020-01-24T11:16:05Z

The INAND_CMD38_ARG_EXT_CSD is a vendor specific EXT_CSD register, which is used to prepare an erase/trim operation. However, it doesn't make sense to use a timeout of 10 minutes while updating the register, which becomes the case when the timeout_ms argument for mmc_switch() is set to zero. Instead, let's use the generic_cmd6_time, as that seems like a reasonable timeout to use for these cases. Signed-off-by: Ulf Hansson Link: https://lore.kernel.org/r/20200122142747.5690-3-ulf.hansson@linaro.org

mmc: block: Add CMD13 polling for MMC IOCTLS with R1B response

2019-11-13T15:10:16Z

MMC IOCTLS with R1B responses may cause the card to enter the busy state, which means it's not ready to receive a new request. To prevent new requests from being sent to the card, use a CMD13 polling loop to verify that the card returns to the transfer state, before completing the request. Signed-off-by: Chaotian Jing Reviewed-by: Avri Altman Cc: stable@vger.kernel.org Signed-off-by: Ulf Hansson

mmc: block: Make card_busy_detect() a bit more generic

2019-11-13T15:10:16Z

To prepare for more users of card_busy_detect(), let's drop the struct request * as an in-parameter and convert to log the error message via dev_err() instead of pr_err(). Signed-off-by: Chaotian Jing Reviewed-by: Avri Altman Cc: stable@vger.kernel.org Signed-off-by: Ulf Hansson

Merge branch 'fixes' into next

2019-02-28T08:16:18Z

mmc: core: align max segment size with logical block size

2019-02-27T17:01:25Z

Logical block size is the lowest possible block size that the storage device can address. Max segment size is often related with controller's DMA capability. And it is reasonable to align max segment size with logical block size. SDHCI sets un-aligned max segment size, and causes ADMA error, so fix it by aligning max segment size with logical block size. Reported-by: Naresh Kamboju Cc: Christoph Hellwig Cc: Naresh Kamboju Cc: Faiz Abbas Cc: linux-block@vger.kernel.org Signed-off-by: Ming Lei Signed-off-by: Ulf Hansson

mmc: core: Calculate the discard arg only once

2019-02-25T07:40:58Z

In MMC, the discard arg is a read-only ext_csd parameter - set it once on card init. To be consistent, do that for SD as well even though its discard arg is always 0x0. Signed-off-by: Avri Altman Signed-off-by: Ulf Hansson

mmc: block: fix debugfs_simple_attr.cocci warnings

2019-02-25T07:40:58Z

Use DEFINE_DEBUGFS_ATTRIBUTE rather than DEFINE_SIMPLE_ATTRIBUTE for debugfs files. Semantic patch information: Rationale: DEFINE_SIMPLE_ATTRIBUTE + debugfs_create_file() imposes some significant overhead as compared to DEFINE_DEBUGFS_ATTRIBUTE + debugfs_create_file_unsafe(). Generated by: scripts/coccinelle/api/debugfs/debugfs_simple_attr.cocci Signed-off-by: YueHaibing Signed-off-by: Ulf Hansson