linux-mmc.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Doug Anderson <dianders@chromium.org>
To: Chris Ball <cjb@laptop.org>
Cc: Jaehoon Chung <jh80.chung@samsung.com>,
	Seungwon Jeon <tgih.jun@samsung.com>,
	James Hogan <james.hogan@imgtec.com>,
	Grant Grundler <grundler@chromium.org>,
	Alim Akhtar <alim.akhtar@samsung.com>,
	Abhilash Kesavan <a.kesavan@samsung.com>,
	Tomasz Figa <tomasz.figa@gmail.com>,
	Olof Johansson <olof@lixom.net>,
	Doug Anderson <dianders@chromium.org>,
	linux-mmc@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH v5 1/4] mmc: dw_mmc: Invalidate cache of current_speed after suspend/resume
Date: Fri,  9 Aug 2013 09:33:17 -0700	[thread overview]
Message-ID: <1376066000-5495-2-git-send-email-dianders@chromium.org> (raw)
In-Reply-To: <1376066000-5495-1-git-send-email-dianders@chromium.org>

The dw_mmc driver keeps a cache of the current slot->clock in order to
avoid doing a whole lot of work every time set_ios() is called.
However, after suspend/resume the register values are bogus so we need
to ensure that the cached value is invalidated.

Specifically I saw problems with the SD Card slot, which doesn't have
MMC_KEEP_POWER.  Problems showed up when no card was inserted across
suspend/resume.  In other words:

1. At boot time, slot is all setup and configured to 400kHz.
2. Suspend
3. Resume; clock registers are reset (by suspend/resume) and not
   restored since dw_mmc still thinks slot is configured for 400kHz
   due to host->current_speed cache.
4. Insert card.
5. No code sees any need to change the clock for detecting the card,
   since everyone thinks it's at 400kHz.  ...but it's not.

Invalidating the current_speed also means that we don't need to call:
  dw_mci_setup_bus(slot, true);
...to force an update of the clock in the case when the slot was left
powered.

Before this patch, many scenarios worked fine across suspend/resume
since the core mmc code ends up adjusting the clock quite a bit
during/suspend resume (and this ended up invalidating the cache for
us).

Signed-off-by: Doug Anderson <dianders@chromium.org>
Reviewed-by: Tomasz Figa <t.figa@samsung.com>
---
Changes in v5:
- Remove force_clkinit as per Jaehoon
- Update commit message to (hopefully) be clearer

Changes in v4: None
Changes in v3: None
Changes in v2:
- Fix typo (some -> come)
- Use ~0 instead of 0xFFFFFFFF; add comment about value

 drivers/mmc/host/dw_mmc.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
index ee5f167..e614b03 100644
--- a/drivers/mmc/host/dw_mmc.c
+++ b/drivers/mmc/host/dw_mmc.c
@@ -629,13 +629,13 @@ static void mci_send_cmd(struct dw_mci_slot *slot, u32 cmd, u32 arg)
 		cmd, arg, cmd_status);
 }
 
-static void dw_mci_setup_bus(struct dw_mci_slot *slot, bool force_clkinit)
+static void dw_mci_setup_bus(struct dw_mci_slot *slot)
 {
 	struct dw_mci *host = slot->host;
 	u32 div;
 	u32 clk_en_a;
 
-	if (slot->clock != host->current_speed || force_clkinit) {
+	if (slot->clock != host->current_speed) {
 		div = host->bus_hz / slot->clock;
 		if (host->bus_hz % slot->clock && host->bus_hz > slot->clock)
 			/*
@@ -819,7 +819,7 @@ static void dw_mci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 		drv_data->set_ios(slot->host, ios);
 
 	/* Slot specific timing and width adjustment */
-	dw_mci_setup_bus(slot, false);
+	dw_mci_setup_bus(slot);
 
 	switch (ios->power_mode) {
 	case MMC_POWER_UP:
@@ -2511,13 +2511,19 @@ int dw_mci_resume(struct dw_mci *host)
 		   DW_MCI_ERROR_FLAGS | SDMMC_INT_CD);
 	mci_writel(host, CTRL, SDMMC_CTRL_INT_ENABLE);
 
+	/*
+	 * Invalidate the 'current_speed' value since CLKDIV has come up in
+	 * default state and our cache is incorrect; set to something we know
+	 * slot->clock won't be.
+	 */
+	host->current_speed = ~0;
+
 	for (i = 0; i < host->num_slots; i++) {
 		struct dw_mci_slot *slot = host->slot[i];
 		if (!slot)
 			continue;
 		if (slot->mmc->pm_flags & MMC_PM_KEEP_POWER) {
 			dw_mci_set_ios(slot->mmc, &slot->mmc->ios);
-			dw_mci_setup_bus(slot, true);
 		}
 
 		ret = mmc_resume_host(host->slot[i]->mmc);
-- 
1.8.3


  reply	other threads:[~2013-08-09 16:33 UTC|newest]

Thread overview: 65+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-07-09 17:31 [PATCH 0/5] mmc: dw_mmc: fixes for suspend/resume on exynos Doug Anderson
2013-07-09 17:31 ` [PATCH 1/5] mmc: dw_mmc: Invalidate cache of current_speed after suspend/resume Doug Anderson
2013-07-09 17:31 ` [PATCH 2/5] mmc: dw_mmc: Add suspend/resume callbacks; disable irq during suspend Doug Anderson
2013-07-09 21:17   ` James Hogan
2013-07-09 21:31     ` Doug Anderson
2013-07-09 17:31 ` [PATCH 3/5] mmc: dw_mmc: Add exynos resume callback to clear WAKEUP_INT Doug Anderson
2013-07-09 19:09   ` Doug Anderson
2013-07-11  0:43     ` Grant Grundler
2013-07-09 17:31 ` [PATCH 4/5] mmc: dw_mmc: Always setup the bus after suspend/resume Doug Anderson
2013-07-09 17:31 ` [PATCH 5/5] mmc: dw_mmc: Set timeout to max upon resume Doug Anderson
2013-07-09 23:19 ` [PATCH v2 0/5] mmc: dw_mmc: fixes for suspend/resume on exynos Doug Anderson
2013-07-09 23:19   ` [PATCH v2 1/5] mmc: dw_mmc: Invalidate cache of current_speed after suspend/resume Doug Anderson
2013-07-09 23:19   ` [PATCH v2 2/5] mmc: dw_mmc: Add suspend_noirq/resume_noirq callbacks for dw_mmc-pltfm Doug Anderson
2013-07-10  8:37     ` James Hogan
2013-07-10 15:08       ` Doug Anderson
2013-07-09 23:19   ` [PATCH v2 3/5] mmc: dw_mmc: Add exynos resume_noirq callback to clear WAKEUP_INT Doug Anderson
2013-07-10 14:54     ` Seungwon Jeon
2013-07-10 15:05       ` Doug Anderson
2013-07-15 12:09         ` Seungwon Jeon
2013-07-31 16:18           ` Doug Anderson
2013-08-06 21:36             ` Doug Anderson
2013-07-09 23:19   ` [PATCH v2 4/5] mmc: dw_mmc: Always setup the bus after suspend/resume Doug Anderson
2013-07-09 23:19   ` [PATCH v2 5/5] mmc: dw_mmc: Set timeout to max upon resume Doug Anderson
2013-07-10 14:54     ` Seungwon Jeon
2013-07-10 15:42   ` [PATCH v3 0/5] mmc: dw_mmc: fixes for suspend/resume on exynos Doug Anderson
2013-07-10 15:42     ` [PATCH v3 1/5] mmc: dw_mmc: Invalidate cache of current_speed after suspend/resume Doug Anderson
2013-07-10 15:42     ` [PATCH v3 2/5] mmc: dw_mmc: Add suspend_noirq/resume_noirq callbacks for dw_mmc-pltfm Doug Anderson
2013-07-15 12:09       ` Seungwon Jeon
2013-08-06 21:32         ` Doug Anderson
2013-07-10 15:42     ` [PATCH v3 3/5] mmc: dw_mmc: Add exynos resume_noirq callback to clear WAKEUP_INT Doug Anderson
2013-07-16  1:36       ` Jaehoon Chung
2013-07-10 15:42     ` [PATCH v3 4/5] mmc: dw_mmc: Always setup the bus after suspend/resume Doug Anderson
2013-07-10 15:42     ` [PATCH v3 5/5] mmc: dw_mmc: Set timeout to max upon resume Doug Anderson
2013-08-06 21:37     ` [PATCH v4 0/4] mmc: dw_mmc: fixes for suspend/resume on exynos Doug Anderson
2013-08-06 21:37       ` [PATCH v4 1/4] mmc: dw_mmc: Invalidate cache of current_speed after suspend/resume Doug Anderson
2013-08-06 21:58         ` Tomasz Figa
2013-08-08  5:14         ` Jaehoon Chung
2013-08-09 13:32         ` Seungwon Jeon
2013-08-09 15:22           ` Doug Anderson
2013-08-12  7:14             ` Seungwon Jeon
2013-08-22  0:54               ` Doug Anderson
2013-08-22 16:25                 ` Doug Anderson
2013-08-06 21:37       ` [PATCH v4 2/4] mmc: dw_mmc: Add exynos resume_noirq callback to clear WAKEUP_INT Doug Anderson
2013-08-06 21:58         ` Tomasz Figa
2013-08-06 22:09           ` Doug Anderson
2013-08-06 22:20             ` Tomasz Figa
2013-08-09 13:33         ` Seungwon Jeon
2013-08-09 15:05           ` Doug Anderson
2013-08-06 21:37       ` [PATCH v4 3/4] mmc: dw_mmc: Always setup the bus after suspend/resume Doug Anderson
2013-08-06 22:01         ` Tomasz Figa
2013-08-09 13:35         ` Seungwon Jeon
2013-08-09 15:43           ` Doug Anderson
2013-08-12  7:20             ` Seungwon Jeon
2013-08-06 21:37       ` [PATCH v4 4/4] mmc: dw_mmc: Set timeout to max upon resume Doug Anderson
2013-08-06 22:02         ` Tomasz Figa
2013-08-09 16:33       ` [PATCH v5 0/4] mmc: dw_mmc: fixes for suspend/resume on exynos Doug Anderson
2013-08-09 16:33         ` Doug Anderson [this message]
2013-08-09 16:33         ` [PATCH v5 2/4] mmc: dw_mmc: Add exynos resume_noirq callback to clear WAKEUP_INT Doug Anderson
2013-08-09 16:41           ` Fabio Estevam
2013-08-09 16:48             ` Doug Anderson
2013-08-12  7:21           ` Seungwon Jeon
2013-08-09 16:33         ` [PATCH v5 3/4] mmc: dw_mmc: Always setup the bus after suspend/resume Doug Anderson
2013-08-09 16:33         ` [PATCH v5 4/4] mmc: dw_mmc: Set timeout to max upon resume Doug Anderson
2013-08-21 11:48         ` [PATCH v5 0/4] mmc: dw_mmc: fixes for suspend/resume on exynos Seungwon Jeon
2013-08-21 15:13           ` Doug Anderson

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=1376066000-5495-2-git-send-email-dianders@chromium.org \
    --to=dianders@chromium.org \
    --cc=a.kesavan@samsung.com \
    --cc=alim.akhtar@samsung.com \
    --cc=cjb@laptop.org \
    --cc=grundler@chromium.org \
    --cc=james.hogan@imgtec.com \
    --cc=jh80.chung@samsung.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mmc@vger.kernel.org \
    --cc=olof@lixom.net \
    --cc=tgih.jun@samsung.com \
    --cc=tomasz.figa@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).