From: Adrian Hunter <adrian.hunter@intel.com>
To: Ulf Hansson <ulf.hansson@linaro.org>
Cc: linux-mmc <linux-mmc@vger.kernel.org>,
Alex Lemberg <alex.lemberg@sandisk.com>,
Mateusz Nowak <mateusz.nowak@intel.com>,
Yuliy Izrailov <Yuliy.Izrailov@sandisk.com>,
Jaehoon Chung <jh80.chung@samsung.com>,
Dong Aisheng <dongas86@gmail.com>,
Das Asutosh <asutoshd@codeaurora.org>,
Zhangfei Gao <zhangfei.gao@gmail.com>,
Sujit Reddy Thumma <sthumma@codeaurora.org>,
Dorfman Konstantin <kdorfman@codeaurora.org>,
David Griego <david.griego@linaro.org>,
Sahitya Tummala <stummala@codeaurora.org>,
Harjani Ritesh <riteshh@codeaurora.org>
Subject: [PATCH RFC 03/46] mmc: sdhci: Move busy signal handling into sdhci_finish_cmd()
Date: Thu, 9 Jun 2016 14:52:03 +0300 [thread overview]
Message-ID: <1465473166-22532-4-git-send-email-adrian.hunter@intel.com> (raw)
In-Reply-To: <1465473166-22532-1-git-send-email-adrian.hunter@intel.com>
In order to support commands during data transfer, command and data
handling needs to be untangled.
That means sdhci_finish_cmd() must not be called from the data IRQ
handler. It is being called because of busy signal handling, which
is treating the command as not finished until the busy signal is
released.
Instead, move busy signal handling from sdhci_cmd_irq() into
sdhci_finish_cmd(). Then the data IRQ handler does not need to call
sdhci_finish_cmd() and can instead finish the request.
What this means in practice for a command with busy signaling, is that
the command response is read from the host controller when the command
complete interrupt is received, thus freeing up the command circuit for
other commands.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
drivers/mmc/host/sdhci.c | 54 +++++++++++++++++++++++-------------------------
1 file changed, 26 insertions(+), 28 deletions(-)
diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index 040af1b6b412..fc4301f68a93 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -45,7 +45,6 @@ static unsigned int debug_quirks2;
static void sdhci_finish_data(struct sdhci_host *);
-static void sdhci_finish_command(struct sdhci_host *);
static int sdhci_execute_tuning(struct mmc_host *mmc, u32 opcode);
static void sdhci_enable_preset_value(struct sdhci_host *host, bool enable);
static int sdhci_get_cd(struct mmc_host *mmc);
@@ -1079,6 +1078,28 @@ static void sdhci_finish_command(struct sdhci_host *host)
}
}
+ /*
+ * The host can send and interrupt when the busy state has
+ * ended, allowing us to wait without wasting CPU cycles.
+ * The busy signal uses DAT0 so this is similar to waiting
+ * for data to complete.
+ *
+ * Note: The 1.0 specification is a bit ambiguous about this
+ * feature so there might be some problems with older
+ * controllers.
+ */
+ if (host->cmd->flags & MMC_RSP_BUSY) {
+ if (host->cmd->data) {
+ DBG("Cannot wait for busy signal when also doing a data transfer");
+ } else if (!(host->quirks & SDHCI_QUIRK_NO_BUSY_IRQ) &&
+ !host->busy_handle) {
+ /* Mark that command complete before busy is ended */
+ host->busy_handle = 1;
+ host->cmd = NULL;
+ return;
+ }
+ }
+
/* Finished CMD23, now send actual command. */
if (host->cmd == host->mrq->sbc) {
host->cmd = NULL;
@@ -2289,33 +2310,10 @@ static void sdhci_cmd_irq(struct sdhci_host *host, u32 intmask, u32 *mask)
return;
}
- /*
- * The host can send and interrupt when the busy state has
- * ended, allowing us to wait without wasting CPU cycles.
- * Unfortunately this is overloaded on the "data complete"
- * interrupt, so we need to take some care when handling
- * it.
- *
- * Note: The 1.0 specification is a bit ambiguous about this
- * feature so there might be some problems with older
- * controllers.
- */
- if (host->cmd->flags & MMC_RSP_BUSY) {
- if (host->cmd->data)
- DBG("Cannot wait for busy signal when also doing a data transfer");
- else if (!(host->quirks & SDHCI_QUIRK_NO_BUSY_IRQ)
- && !host->busy_handle) {
- /* Mark that command complete before busy is ended */
- host->busy_handle = 1;
- return;
- }
-
- /* The controller does not support the end-of-busy IRQ,
- * fall through and take the SDHCI_INT_RESPONSE */
- } else if ((host->quirks2 & SDHCI_QUIRK2_STOP_WITH_TC) &&
- host->cmd->opcode == MMC_STOP_TRANSMISSION && !host->data) {
+ if ((host->quirks2 & SDHCI_QUIRK2_STOP_WITH_TC) &&
+ !(host->cmd->flags & MMC_RSP_BUSY) && !host->data &&
+ host->cmd->opcode == MMC_STOP_TRANSMISSION)
*mask &= ~SDHCI_INT_DATA_END;
- }
if (intmask & SDHCI_INT_RESPONSE)
sdhci_finish_command(host);
@@ -2389,7 +2387,7 @@ static void sdhci_data_irq(struct sdhci_host *host, u32 intmask)
* sure we do things in the proper order.
*/
if (host->busy_handle)
- sdhci_finish_command(host);
+ tasklet_schedule(&host->finish_tasklet);
else
host->busy_handle = 1;
return;
--
1.9.1
next prev parent reply other threads:[~2016-06-09 11:57 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-09 11:52 [PATCH RFC 00/46] mmc: mmc: Add Software Command Queuing Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 01/46] mmc: core: Add support for sending commands during data transfer Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 02/46] mmc: mmc_test: Add tests for sending commands during transfer Adrian Hunter
2016-06-09 11:52 ` Adrian Hunter [this message]
2016-06-09 11:52 ` [PATCH RFC 04/46] mmc: sdhci: Get rid of redundant BUG_ONs Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 05/46] mmc: sdhci: Simplify sdhci_finish_command() by clearing host->cmd at the start Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 06/46] mmc: sdhci: Record what command is using the data lines Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 07/46] mmc: sdhci: Get rid of host->busy_handle Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 08/46] mmc: sdhci: Reduce the use of host->mrq Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 09/46] mmc: sdhci: Move host->data warning Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 10/46] mmc: sdhci: Factor out sdhci_finish_mrq() Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 11/46] mmc: sdhci: Factor out sdhci_needs_reset() Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 12/46] mmc: sdhci: Track whether a reset is pending Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 13/46] mmc: sdhci: Clear pointers when a request finishes Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 14/46] mmc: sdhci: Ensure all requests get errored out Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 15/46] mmc: sdhci: Factor out sdhci_data_line_cmd() Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 16/46] mmc: sdhci: Separate timer timeout for command and data requests Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 17/46] mmc: sdhci: Allow for finishing multiple requests Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 18/46] mmc: sdhci: Factor out sdhci_auto_cmd12() Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 19/46] mmc: sdhci: Do not reset cmd or data circuits that are in use Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 20/46] mmc: sdhci: Support cap_cmd_during_tfr requests Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 21/46] mmc: sdhci-pci: Set MMC_CAP_CMD_DURING_TFR for Intel eMMC controllers Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 22/46] mmc: sdhci-acpi: " Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 23/46] mmc: queue: Fix queue thread wake-up Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 24/46] mmc: queue: Factor out mmc_queue_alloc_bounce_bufs() Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 25/46] mmc: queue: Factor out mmc_queue_alloc_bounce_sgs() Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 26/46] mmc: queue: Factor out mmc_queue_alloc_sgs() Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 27/46] mmc: queue: Factor out mmc_queue_reqs_free_bufs() Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 28/46] mmc: queue: Introduce queue depth Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 29/46] mmc: queue: Use queue depth to allocate and free Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 30/46] mmc: queue: Allocate queue of size qdepth Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 31/46] mmc: mmc: Add Command Queue definitions Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 32/46] mmc: mmc: Add functions to enable / disable the Command Queue Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 33/46] mmc: mmc_test: Disable Command Queue while mmc_test is used Adrian Hunter
2016-06-10 10:44 ` [PATCH RFC V2 " Adrian Hunter
2016-06-10 10:59 ` Venu Byravarasu
2016-06-10 11:36 ` Adrian Hunter
2016-06-10 12:03 ` Venu Byravarasu
2016-06-09 11:52 ` [PATCH RFC 34/46] mmc: block: Disable Command Queue while RPMB " Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 35/46] mmc: core: Do not prepare a new request twice Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 36/46] mmc: core: Export mmc_retune_hold() and mmc_retune_release() Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 37/46] mmc: block: Factor out mmc_blk_requeue() Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 38/46] mmc: block: Fix 4K native sector check Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 39/46] mmc: block: Use local var for mqrq_cur Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 40/46] mmc: block: Pass mqrq to mmc_blk_prep_packed_list() Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 41/46] mmc: block: Introduce queue semantics Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 42/46] mmc: queue: Add a function to control wake-up on new requests Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 43/46] mmc: block: Add Software Command Queuing Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 44/46] mmc: mmc: Enable " Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 45/46] mmc: sdhci-pci: Enable Software Command Queuing for some Intel controllers Adrian Hunter
2016-06-09 11:52 ` [PATCH RFC 46/46] mmc: sdhci-acpi: " Adrian Hunter
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=1465473166-22532-4-git-send-email-adrian.hunter@intel.com \
--to=adrian.hunter@intel.com \
--cc=Yuliy.Izrailov@sandisk.com \
--cc=alex.lemberg@sandisk.com \
--cc=asutoshd@codeaurora.org \
--cc=david.griego@linaro.org \
--cc=dongas86@gmail.com \
--cc=jh80.chung@samsung.com \
--cc=kdorfman@codeaurora.org \
--cc=linux-mmc@vger.kernel.org \
--cc=mateusz.nowak@intel.com \
--cc=riteshh@codeaurora.org \
--cc=sthumma@codeaurora.org \
--cc=stummala@codeaurora.org \
--cc=ulf.hansson@linaro.org \
--cc=zhangfei.gao@gmail.com \
/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).