stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Daniel Drake <drake@endlessm.com>,
	Oleksij Rempel <linux@rempel-privat.de>,
	Ulf Hansson <ulf.hansson@linaro.org>
Subject: [PATCH 5.0 04/52] mmc: alcor: fix DMA reads
Date: Tue, 26 Mar 2019 15:29:51 +0900	[thread overview]
Message-ID: <20190326042701.190923076@linuxfoundation.org> (raw)
In-Reply-To: <20190326042700.963224437@linuxfoundation.org>

5.0-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Daniel Drake <drake@endlessm.com>

commit 5ea47691bd99e1100707ec63364aff72324e2af4 upstream.

Setting max_blk_count to 1 here was causing the mmc block layer
to always use the MMC_READ_SINGLE_BLOCK command here, which the
driver does not DMA-accelerate.

Drop the max_blk_ settings here. The mmc host defaults suffice,
along with the max_segs and max_seg_size settings, which I have
now documented in more detail.

Now each MMC command reads 4 512-byte blocks, using DMA instead of
PIO. On my SD card, this increases read performance (measured with dd)
from 167kb/sec to 4.6mb/sec.

Link: http://lkml.kernel.org/r/CAD8Lp47L5T3jnAjBiPs1cQ+yFA3L6LJtgFvMETnBrY63-Zdi2g@mail.gmail.com
Signed-off-by: Daniel Drake <drake@endlessm.com>
Reviewed-by: Oleksij Rempel <linux@rempel-privat.de>
Fixes: c5413ad815a6 ("mmc: add new Alcor Micro Cardreader SD/MMC driver")
Cc: stable@vger.kernel.org
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 drivers/mmc/host/alcor.c |   25 +++++++++++++++++++------
 1 file changed, 19 insertions(+), 6 deletions(-)

--- a/drivers/mmc/host/alcor.c
+++ b/drivers/mmc/host/alcor.c
@@ -1044,14 +1044,27 @@ static void alcor_init_mmc(struct alcor_
 	mmc->caps2 = MMC_CAP2_NO_SDIO;
 	mmc->ops = &alcor_sdc_ops;
 
-	/* Hardware cannot do scatter lists */
+	/* The hardware does DMA data transfer of 4096 bytes to/from a single
+	 * buffer address. Scatterlists are not supported, but upon DMA
+	 * completion (signalled via IRQ), the original vendor driver does
+	 * then immediately set up another DMA transfer of the next 4096
+	 * bytes.
+	 *
+	 * This means that we need to handle the I/O in 4096 byte chunks.
+	 * Lacking a way to limit the sglist entries to 4096 bytes, we instead
+	 * impose that only one segment is provided, with maximum size 4096,
+	 * which also happens to be the minimum size. This means that the
+	 * single-entry sglist handled by this driver can be handed directly
+	 * to the hardware, nice and simple.
+	 *
+	 * Unfortunately though, that means we only do 4096 bytes I/O per
+	 * MMC command. A future improvement would be to make the driver
+	 * accept sg lists and entries of any size, and simply iterate
+	 * through them 4096 bytes at a time.
+	 */
 	mmc->max_segs = AU6601_MAX_DMA_SEGMENTS;
 	mmc->max_seg_size = AU6601_MAX_DMA_BLOCK_SIZE;
-
-	mmc->max_blk_size = mmc->max_seg_size;
-	mmc->max_blk_count = mmc->max_segs;
-
-	mmc->max_req_size = mmc->max_seg_size * mmc->max_segs;
+	mmc->max_req_size = mmc->max_seg_size;
 }
 
 static int alcor_pci_sdmmc_drv_probe(struct platform_device *pdev)



  parent reply	other threads:[~2019-03-26  6:42 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-03-26  6:29 [PATCH 5.0 00/52] 5.0.5-stable review Greg Kroah-Hartman
2019-03-26  6:29 ` [PATCH 5.0 01/52] ALSA: hda - add Lenovo IdeaCentre B550 to the power_save_blacklist Greg Kroah-Hartman
2019-03-26  6:29 ` [PATCH 5.0 02/52] ALSA: firewire-motu: use version field of unit directory to identify model Greg Kroah-Hartman
2019-03-26  6:29 ` [PATCH 5.0 03/52] mmc: pxamci: fix enum type confusion Greg Kroah-Hartman
2019-03-26  6:29 ` Greg Kroah-Hartman [this message]
2019-03-26  6:29 ` [PATCH 5.0 05/52] mmc: mxcmmc: "Revert mmc: mxcmmc: handle highmem pages" Greg Kroah-Hartman
2019-03-26  6:29 ` [PATCH 5.0 06/52] mmc: renesas_sdhi: limit block count to 16 bit for old revisions Greg Kroah-Hartman
2019-03-26  6:29 ` [PATCH 5.0 07/52] drm/amdgpu: fix invalid use of change_bit Greg Kroah-Hartman
2019-03-26  6:29 ` [PATCH 5.0 08/52] drm/vmwgfx: Dont double-free the mode stored in par->set_mode Greg Kroah-Hartman
2019-03-26  6:29 ` [PATCH 5.0 09/52] drm/vmwgfx: Return 0 when gmrid::get_node runs out of IDs Greg Kroah-Hartman
2019-03-26  6:29 ` [PATCH 5.0 10/52] iommu/amd: fix sg->dma_address for sg->offset bigger than PAGE_SIZE Greg Kroah-Hartman
2019-03-26  6:29 ` [PATCH 5.0 11/52] iommu/iova: Fix tracking of recently failed iova address Greg Kroah-Hartman
2019-03-26  6:29 ` [PATCH 5.0 12/52] libceph: wait for latest osdmap in ceph_monc_blacklist_add() Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 13/52] udf: Fix crash on IO error during truncate Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 14/52] mips: loongson64: lemote-2f: Add IRQF_NO_SUSPEND to "cascade" irqaction Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 15/52] MIPS: Ensure ELF appended dtb is relocated Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 16/52] MIPS: Fix kernel crash for R6 in jump label branch function Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 17/52] powerpc/vdso64: Fix CLOCK_MONOTONIC inconsistencies across Y2038 Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 18/52] powerpc/security: Fix spectre_v2 reporting Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 19/52] net/mlx5: Fix DCT creation bad flow Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 20/52] scsi: core: Avoid that a kernel warning appears during system resume Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 21/52] scsi: qla2xxx: Fix FC-AL connection target discovery Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 22/52] scsi: ibmvscsi: Protect ibmvscsi_head from concurrent modificaiton Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 23/52] scsi: ibmvscsi: Fix empty event pool access during host removal Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 24/52] futex: Ensure that futex address is aligned in handle_futex_death() Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 25/52] cifs: allow guest mounts to work for smb3.11 Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 26/52] perf probe: Fix getting the kernel map Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 27/52] objtool: Move objtool_file struct off the stack Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 28/52] irqchip/gic-v3-its: Fix comparison logic in lpi_range_cmp Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 29/52] clocksource/drivers/riscv: Fix clocksource mask Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 30/52] SMB3: Fix SMB3.1.1 guest mounts to Samba Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 31/52] ALSA: hda - Dont trigger jackpoll_work in azx_resume Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 32/52] ALSA: ac97: Fix of-node refcount unbalance Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 33/52] ext4: fix NULL pointer dereference while journal is aborted Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 34/52] ext4: fix data corruption caused by unaligned direct AIO Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 35/52] ext4: brelse all indirect buffer in ext4_ind_remove_space() Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 36/52] media: v4l2-ctrls.c/uvc: zero v4l2_event Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 37/52] Bluetooth: hci_uart: Check if socket buffer is ERR_PTR in h4_recv_buf() Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 38/52] Bluetooth: Fix decrementing reference count twice in releasing socket Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 39/52] Bluetooth: hci_ldisc: Initialize hci_dev before open() Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 40/52] Bluetooth: hci_ldisc: Postpone HCI_UART_PROTO_READY bit set in hci_uart_set_proto() Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 41/52] drm/vkms: Fix flush_work() without INIT_WORK() Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 42/52] RDMA/cma: Rollback source IP address if failing to acquire device Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 43/52] f2fs: fix to avoid deadlock of atomic file operations Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 44/52] aio: simplify - and fix - fget/fput for io_submit() Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 45/52] netfilter: ebtables: remove BUGPRINT messages Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 46/52] loop: access lo_backing_file only when the loop device is Lo_bound Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 47/52] x86/unwind: Handle NULL pointer calls better in frame unwinder Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 48/52] x86/unwind: Add hardcoded ORC entry for NULL Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 49/52] locking/lockdep: Add debug_locks check in __lock_downgrade() Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 50/52] mm, mempolicy: fix uninit memory access Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 51/52] ALSA: hda - Record the current power state before suspend/resume calls Greg Kroah-Hartman
2019-03-26  6:30 ` [PATCH 5.0 52/52] ALSA: hda - Enforces runtime_resume after S3 and S4 for each codec Greg Kroah-Hartman
2019-03-26 15:20 ` [PATCH 5.0 00/52] 5.0.5-stable review Jon Hunter
2019-03-27  0:56   ` Greg Kroah-Hartman
2019-03-26 17:50 ` Guenter Roeck
2019-03-27  0:59   ` Greg Kroah-Hartman
2019-03-26 23:18 ` shuah
2019-03-27  0:55   ` Greg Kroah-Hartman
2019-03-27  4:06 ` Naresh Kamboju
2019-03-27  5:06   ` Greg Kroah-Hartman

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20190326042701.190923076@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=drake@endlessm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux@rempel-privat.de \
    --cc=stable@vger.kernel.org \
    --cc=ulf.hansson@linaro.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).