| Message ID | 20190227130237.7725-1-ming.lei@redhat.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | mmc: align max segment size with logical block size | expand |
+ Christoph On Wed, 27 Feb 2019 at 14:02, Ming Lei <ming.lei@redhat.com> wrote: > > 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. Seems reasonable. However, is this specific to MMC or should it be managed in block layer? > > Reported-by: Naresh Kamboju <naresh.kamboju@linaro.org> > Cc: Naresh Kamboju <naresh.kamboju@linaro.org> > Cc: Faiz Abbas <faiz_abbas@ti.com> > Cc: linux-block@vger.kernel.org > Signed-off-by: Ming Lei <ming.lei@redhat.com> > --- > drivers/mmc/core/block.c | 6 ------ > drivers/mmc/core/queue.c | 9 ++++++++- > 2 files changed, 8 insertions(+), 7 deletions(-) > > diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c > index 14f3fdb8c6bb..9ce8eb51a60f 100644 > --- a/drivers/mmc/core/block.c > +++ b/drivers/mmc/core/block.c > @@ -2380,12 +2380,6 @@ static struct mmc_blk_data *mmc_blk_alloc_req(struct mmc_card *card, > snprintf(md->disk->disk_name, sizeof(md->disk->disk_name), > "mmcblk%u%s", card->host->index, subname ? subname : ""); > > - if (mmc_card_mmc(card)) > - blk_queue_logical_block_size(md->queue.queue, > - card->ext_csd.data_sector_size); > - else > - blk_queue_logical_block_size(md->queue.queue, 512); > - > set_capacity(md->disk, size); > > if (mmc_host_cmd23(card->host)) { > diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c > index cc19e71c71d4..891bdbe14192 100644 > --- a/drivers/mmc/core/queue.c > +++ b/drivers/mmc/core/queue.c > @@ -355,6 +355,7 @@ static void mmc_setup_queue(struct mmc_queue *mq, struct mmc_card *card) > { > struct mmc_host *host = card->host; > u64 limit = BLK_BOUNCE_HIGH; > + unsigned block_size = 512; > > if (mmc_dev(host)->dma_mask && *mmc_dev(host)->dma_mask) > limit = (u64)dma_max_pfn(mmc_dev(host)) << PAGE_SHIFT; > @@ -368,7 +369,13 @@ static void mmc_setup_queue(struct mmc_queue *mq, struct mmc_card *card) > blk_queue_max_hw_sectors(mq->queue, > min(host->max_blk_count, host->max_req_size / 512)); > blk_queue_max_segments(mq->queue, host->max_segs); > - blk_queue_max_segment_size(mq->queue, host->max_seg_size); > + > + if (mmc_card_mmc(card)) > + block_size = card->ext_csd.data_sector_size; > + > + blk_queue_logical_block_size(mq->queue, block_size); > + blk_queue_max_segment_size(mq->queue, host->max_seg_size & > + ~(block_size - 1)); Can't we use round_down() instead? No? > > INIT_WORK(&mq->recovery_work, mmc_mq_recovery_handler); > INIT_WORK(&mq->complete_work, mmc_blk_mq_complete_work); > -- > 2.9.5 > Kind regards Uffe
On Wed, Feb 27, 2019 at 02:57:04PM +0100, Ulf Hansson wrote: > + Christoph > > On Wed, 27 Feb 2019 at 14:02, Ming Lei <ming.lei@redhat.com> wrote: > > > > 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. > > Seems reasonable. However, is this specific to MMC or should it be > managed in block layer? So far, only see such broken max segment size on MMC, so suggest to fix in MMC, otherwise we may have to align the limit in fast path. > > > > > Reported-by: Naresh Kamboju <naresh.kamboju@linaro.org> > > Cc: Naresh Kamboju <naresh.kamboju@linaro.org> > > Cc: Faiz Abbas <faiz_abbas@ti.com> > > Cc: linux-block@vger.kernel.org > > Signed-off-by: Ming Lei <ming.lei@redhat.com> > > --- > > drivers/mmc/core/block.c | 6 ------ > > drivers/mmc/core/queue.c | 9 ++++++++- > > 2 files changed, 8 insertions(+), 7 deletions(-) > > > > diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c > > index 14f3fdb8c6bb..9ce8eb51a60f 100644 > > --- a/drivers/mmc/core/block.c > > +++ b/drivers/mmc/core/block.c > > @@ -2380,12 +2380,6 @@ static struct mmc_blk_data *mmc_blk_alloc_req(struct mmc_card *card, > > snprintf(md->disk->disk_name, sizeof(md->disk->disk_name), > > "mmcblk%u%s", card->host->index, subname ? subname : ""); > > > > - if (mmc_card_mmc(card)) > > - blk_queue_logical_block_size(md->queue.queue, > > - card->ext_csd.data_sector_size); > > - else > > - blk_queue_logical_block_size(md->queue.queue, 512); > > - > > set_capacity(md->disk, size); > > > > if (mmc_host_cmd23(card->host)) { > > diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c > > index cc19e71c71d4..891bdbe14192 100644 > > --- a/drivers/mmc/core/queue.c > > +++ b/drivers/mmc/core/queue.c > > @@ -355,6 +355,7 @@ static void mmc_setup_queue(struct mmc_queue *mq, struct mmc_card *card) > > { > > struct mmc_host *host = card->host; > > u64 limit = BLK_BOUNCE_HIGH; > > + unsigned block_size = 512; > > > > if (mmc_dev(host)->dma_mask && *mmc_dev(host)->dma_mask) > > limit = (u64)dma_max_pfn(mmc_dev(host)) << PAGE_SHIFT; > > @@ -368,7 +369,13 @@ static void mmc_setup_queue(struct mmc_queue *mq, struct mmc_card *card) > > blk_queue_max_hw_sectors(mq->queue, > > min(host->max_blk_count, host->max_req_size / 512)); > > blk_queue_max_segments(mq->queue, host->max_segs); > > - blk_queue_max_segment_size(mq->queue, host->max_seg_size); > > + > > + if (mmc_card_mmc(card)) > > + block_size = card->ext_csd.data_sector_size; > > + > > + blk_queue_logical_block_size(mq->queue, block_size); > > + blk_queue_max_segment_size(mq->queue, host->max_seg_size & > > + ~(block_size - 1)); > > Can't we use round_down() instead? No? Sure, use round_down() in V2. thanks, Ming
diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c index 14f3fdb8c6bb..9ce8eb51a60f 100644 --- a/drivers/mmc/core/block.c +++ b/drivers/mmc/core/block.c @@ -2380,12 +2380,6 @@ static struct mmc_blk_data *mmc_blk_alloc_req(struct mmc_card *card, snprintf(md->disk->disk_name, sizeof(md->disk->disk_name), "mmcblk%u%s", card->host->index, subname ? subname : ""); - if (mmc_card_mmc(card)) - blk_queue_logical_block_size(md->queue.queue, - card->ext_csd.data_sector_size); - else - blk_queue_logical_block_size(md->queue.queue, 512); - set_capacity(md->disk, size); if (mmc_host_cmd23(card->host)) { diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c index cc19e71c71d4..891bdbe14192 100644 --- a/drivers/mmc/core/queue.c +++ b/drivers/mmc/core/queue.c @@ -355,6 +355,7 @@ static void mmc_setup_queue(struct mmc_queue *mq, struct mmc_card *card) { struct mmc_host *host = card->host; u64 limit = BLK_BOUNCE_HIGH; + unsigned block_size = 512; if (mmc_dev(host)->dma_mask && *mmc_dev(host)->dma_mask) limit = (u64)dma_max_pfn(mmc_dev(host)) << PAGE_SHIFT; @@ -368,7 +369,13 @@ static void mmc_setup_queue(struct mmc_queue *mq, struct mmc_card *card) blk_queue_max_hw_sectors(mq->queue, min(host->max_blk_count, host->max_req_size / 512)); blk_queue_max_segments(mq->queue, host->max_segs); - blk_queue_max_segment_size(mq->queue, host->max_seg_size); + + if (mmc_card_mmc(card)) + block_size = card->ext_csd.data_sector_size; + + blk_queue_logical_block_size(mq->queue, block_size); + blk_queue_max_segment_size(mq->queue, host->max_seg_size & + ~(block_size - 1)); INIT_WORK(&mq->recovery_work, mmc_mq_recovery_handler); INIT_WORK(&mq->complete_work, mmc_blk_mq_complete_work);
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 <naresh.kamboju@linaro.org> Cc: Naresh Kamboju <naresh.kamboju@linaro.org> Cc: Faiz Abbas <faiz_abbas@ti.com> Cc: linux-block@vger.kernel.org Signed-off-by: Ming Lei <ming.lei@redhat.com> --- drivers/mmc/core/block.c | 6 ------ drivers/mmc/core/queue.c | 9 ++++++++- 2 files changed, 8 insertions(+), 7 deletions(-)