Merge branch 'master' into for-next

Fast-forwarded to current state of Linus' tree as there are patches to be
applied for files that didn't exist on the old branch.

47 files changed, 3375 insertions, 1526 deletions

diff --git a/drivers/mmc/card/Kconfig b/drivers/mmc/card/Kconfig
index 2a876c4099c..3b1f783bf92 100644
--- a/drivers/mmc/card/Kconfig
+++ b/drivers/mmc/card/Kconfig
@@ -58,12 +58,11 @@ config SDIO_UART
 
 config MMC_TEST
 	tristate "MMC host test driver"
-	default n
 	help
 	  Development driver that performs a series of reads and writes
 	  to a memory card in order to expose certain well known bugs
 	  in host controllers. The tests are executed by writing to the
-	  "test" file in sysfs under each card. Note that whatever is
+	  "test" file in debugfs under each card. Note that whatever is
 	  on your card will be overwritten by these tests.
 
 	  This driver is only of interest to those developing or
diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c
index bfc8a8ae55d..61d233a7c11 100644
--- a/drivers/mmc/card/block.c
+++ b/drivers/mmc/card/block.c
@@ -621,6 +621,7 @@ static struct mmc_blk_data *mmc_blk_alloc(struct mmc_card *card)
 	md->disk->private_data = md;
 	md->disk->queue = md->queue.queue;
 	md->disk->driverfs_dev = &card->dev;
+	set_disk_ro(md->disk, md->read_only);
 
 	/*
 	 * As discussed on lkml, GENHD_FL_REMOVABLE should:
diff --git a/drivers/mmc/card/mmc_test.c b/drivers/mmc/card/mmc_test.c
index 21adc27f413..abc1a63bcc5 100644
--- a/drivers/mmc/card/mmc_test.c
+++ b/drivers/mmc/card/mmc_test.c
@@ -88,6 +88,7 @@ struct mmc_test_area {
  * @sectors: amount of sectors to check in one group
  * @ts: time values of transfer
  * @rate: calculated transfer rate
+ * @iops: I/O operations per second (times 100)
  */
 struct mmc_test_transfer_result {
 	struct list_head link;
@@ -95,6 +96,7 @@ struct mmc_test_transfer_result {
 	unsigned int sectors;
 	struct timespec ts;
 	unsigned int rate;
+	unsigned int iops;
 };
 
 /**
@@ -226,9 +228,10 @@ static int mmc_test_wait_busy(struct mmc_test_card *test)
 
 		if (!busy && mmc_test_busy(&cmd)) {
 			busy = 1;
-			printk(KERN_INFO "%s: Warning: Host did not "
-				"wait for busy state to end.\n",
-				mmc_hostname(test->card->host));
+			if (test->card->host->caps & MMC_CAP_WAIT_WHILE_BUSY)
+				printk(KERN_INFO "%s: Warning: Host did not "
+					"wait for busy state to end.\n",
+					mmc_hostname(test->card->host));
 		}
 	} while (mmc_test_busy(&cmd));
 
@@ -289,7 +292,7 @@ static void mmc_test_free_mem(struct mmc_test_mem *mem)
 }
 
 /*
- * Allocate a lot of memory, preferrably max_sz but at least min_sz.  In case
+ * Allocate a lot of memory, preferably max_sz but at least min_sz.  In case
  * there isn't much memory do not exceed 1/16th total lowmem pages.  Also do
  * not exceed a maximum number of segments and try not to make segments much
  * bigger than maximum segment size.
@@ -494,7 +497,7 @@ static unsigned int mmc_test_rate(uint64_t bytes, struct timespec *ts)
  */
 static void mmc_test_save_transfer_result(struct mmc_test_card *test,
 	unsigned int count, unsigned int sectors, struct timespec ts,
-	unsigned int rate)
+	unsigned int rate, unsigned int iops)
 {
 	struct mmc_test_transfer_result *tr;
 
@@ -509,6 +512,7 @@ static void mmc_test_save_transfer_result(struct mmc_test_card *test,
 	tr->sectors = sectors;
 	tr->ts = ts;
 	tr->rate = rate;
+	tr->iops = iops;
 
 	list_add_tail(&tr->link, &test->gr->tr_lst);
 }
@@ -519,20 +523,22 @@ static void mmc_test_save_transfer_result(struct mmc_test_card *test,
 static void mmc_test_print_rate(struct mmc_test_card *test, uint64_t bytes,
 				struct timespec *ts1, struct timespec *ts2)
 {
-	unsigned int rate, sectors = bytes >> 9;
+	unsigned int rate, iops, sectors = bytes >> 9;
 	struct timespec ts;
 
 	ts = timespec_sub(*ts2, *ts1);
 
 	rate = mmc_test_rate(bytes, &ts);
+	iops = mmc_test_rate(100, &ts); /* I/O ops per sec x 100 */
 
 	printk(KERN_INFO "%s: Transfer of %u sectors (%u%s KiB) took %lu.%09lu "
-			 "seconds (%u kB/s, %u KiB/s)\n",
+			 "seconds (%u kB/s, %u KiB/s, %u.%02u IOPS)\n",
 			 mmc_hostname(test->card->host), sectors, sectors >> 1,
 			 (sectors & 1 ? ".5" : ""), (unsigned long)ts.tv_sec,
-			 (unsigned long)ts.tv_nsec, rate / 1000, rate / 1024);
+			 (unsigned long)ts.tv_nsec, rate / 1000, rate / 1024,
+			 iops / 100, iops % 100);
 
-	mmc_test_save_transfer_result(test, 1, sectors, ts, rate);
+	mmc_test_save_transfer_result(test, 1, sectors, ts, rate, iops);
 }
 
 /*
@@ -542,22 +548,24 @@ static void mmc_test_print_avg_rate(struct mmc_test_card *test, uint64_t bytes,
 				    unsigned int count, struct timespec *ts1,
 				    struct timespec *ts2)
 {
-	unsigned int rate, sectors = bytes >> 9;
+	unsigned int rate, iops, sectors = bytes >> 9;
 	uint64_t tot = bytes * count;
 	struct timespec ts;
 
 	ts = timespec_sub(*ts2, *ts1);
 
 	rate = mmc_test_rate(tot, &ts);
+	iops = mmc_test_rate(count * 100, &ts); /* I/O ops per sec x 100 */
 
 	printk(KERN_INFO "%s: Transfer of %u x %u sectors (%u x %u%s KiB) took "
-			 "%lu.%09lu seconds (%u kB/s, %u KiB/s)\n",
+			 "%lu.%09lu seconds (%u kB/s, %u KiB/s, "
+			 "%u.%02u IOPS)\n",
 			 mmc_hostname(test->card->host), count, sectors, count,
 			 sectors >> 1, (sectors & 1 ? ".5" : ""),
 			 (unsigned long)ts.tv_sec, (unsigned long)ts.tv_nsec,
-			 rate / 1000, rate / 1024);
+			 rate / 1000, rate / 1024, iops / 100, iops % 100);
 
-	mmc_test_save_transfer_result(test, count, sectors, ts, rate);
+	mmc_test_save_transfer_result(test, count, sectors, ts, rate, iops);
 }
 
 /*
@@ -1425,28 +1433,29 @@ static int mmc_test_area_cleanup(struct mmc_test_card *test)
 }
 
 /*
- * Initialize an area for testing large transfers.  The size of the area is the
- * preferred erase size which is a good size for optimal transfer speed.  Note
- * that is typically 4MiB for modern cards.  The test area is set to the middle
- * of the card because cards may have different charateristics at the front
- * (for FAT file system optimization).  Optionally, the area is erased (if the
- * card supports it) which may improve write performance.  Optionally, the area
- * is filled with data for subsequent read tests.
+ * Initialize an area for testing large transfers.  The test area is set to the
+ * middle of the card because cards may have different charateristics at the
+ * front (for FAT file system optimization).  Optionally, the area is erased
+ * (if the card supports it) which may improve write performance.  Optionally,
+ * the area is filled with data for subsequent read tests.
  */
 static int mmc_test_area_init(struct mmc_test_card *test, int erase, int fill)
 {
 	struct mmc_test_area *t = &test->area;
-	unsigned long min_sz = 64 * 1024;
+	unsigned long min_sz = 64 * 1024, sz;
 	int ret;
 
 	ret = mmc_test_set_blksize(test, 512);
 	if (ret)
 		return ret;
 
-	if (test->card->pref_erase > TEST_AREA_MAX_SIZE >> 9)
-		t->max_sz = TEST_AREA_MAX_SIZE;
-	else
-		t->max_sz = (unsigned long)test->card->pref_erase << 9;
+	/* Make the test area size about 4MiB */
+	sz = (unsigned long)test->card->pref_erase << 9;
+	t->max_sz = sz;
+	while (t->max_sz < 4 * 1024 * 1024)
+		t->max_sz += sz;
+	while (t->max_sz > TEST_AREA_MAX_SIZE && t->max_sz > sz)
+		t->max_sz -= sz;
 
 	t->max_segs = test->card->host->max_segs;
 	t->max_seg_sz = test->card->host->max_seg_size;
@@ -1766,6 +1775,187 @@ static int mmc_test_profile_seq_trim_perf(struct mmc_test_card *test)
 	return 0;
 }
 
+static unsigned int rnd_next = 1;
+
+static unsigned int mmc_test_rnd_num(unsigned int rnd_cnt)
+{
+	uint64_t r;
+
+	rnd_next = rnd_next * 1103515245 + 12345;
+	r = (rnd_next >> 16) & 0x7fff;
+	return (r * rnd_cnt) >> 15;
+}
+
+static int mmc_test_rnd_perf(struct mmc_test_card *test, int write, int print,
+			     unsigned long sz)
+{
+	unsigned int dev_addr, cnt, rnd_addr, range1, range2, last_ea = 0, ea;
+	unsigned int ssz;
+	struct timespec ts1, ts2, ts;
+	int ret;
+
+	ssz = sz >> 9;
+
+	rnd_addr = mmc_test_capacity(test->card) / 4;
+	range1 = rnd_addr / test->card->pref_erase;
+	range2 = range1 / ssz;
+
+	getnstimeofday(&ts1);
+	for (cnt = 0; cnt < UINT_MAX; cnt++) {
+		getnstimeofday(&ts2);
+		ts = timespec_sub(ts2, ts1);
+		if (ts.tv_sec >= 10)
+			break;
+		ea = mmc_test_rnd_num(range1);
+		if (ea == last_ea)
+			ea -= 1;
+		last_ea = ea;
+		dev_addr = rnd_addr + test->card->pref_erase * ea +
+			   ssz * mmc_test_rnd_num(range2);
+		ret = mmc_test_area_io(test, sz, dev_addr, write, 0, 0);
+		if (ret)
+			return ret;
+	}
+	if (print)
+		mmc_test_print_avg_rate(test, sz, cnt, &ts1, &ts2);
+	return 0;
+}
+
+static int mmc_test_random_perf(struct mmc_test_card *test, int write)
+{
+	unsigned int next;
+	unsigned long sz;
+	int ret;
+
+	for (sz = 512; sz < test->area.max_tfr; sz <<= 1) {
+		/*
+		 * When writing, try to get more consistent results by running
+		 * the test twice with exactly the same I/O but outputting the
+		 * results only for the 2nd run.
+		 */
+		if (write) {
+			next = rnd_next;
+			ret = mmc_test_rnd_perf(test, write, 0, sz);
+			if (ret)
+				return ret;
+			rnd_next = next;
+		}
+		ret = mmc_test_rnd_perf(test, write, 1, sz);
+		if (ret)
+			return ret;
+	}
+	sz = test->area.max_tfr;
+	if (write) {
+		next = rnd_next;
+		ret = mmc_test_rnd_perf(test, write, 0, sz);
+		if (ret)
+			return ret;
+		rnd_next = next;
+	}
+	return mmc_test_rnd_perf(test, write, 1, sz);
+}
+
+/*
+ * Random read performance by transfer size.
+ */
+static int mmc_test_random_read_perf(struct mmc_test_card *test)
+{
+	return mmc_test_random_perf(test, 0);
+}
+
+/*
+ * Random write performance by transfer size.
+ */
+static int mmc_test_random_write_perf(struct mmc_test_card *test)
+{
+	return mmc_test_random_perf(test, 1);
+}
+
+static int mmc_test_seq_perf(struct mmc_test_card *test, int write,
+			     unsigned int tot_sz, int max_scatter)
+{
+	unsigned int dev_addr, i, cnt, sz, ssz;
+	struct timespec ts1, ts2;
+	int ret;
+
+	sz = test->area.max_tfr;
+	/*
+	 * In the case of a maximally scattered transfer, the maximum transfer
+	 * size is further limited by using PAGE_SIZE segments.
+	 */
+	if (max_scatter) {
+		struct mmc_test_area *t = &test->area;
+		unsigned long max_tfr;
+
+		if (t->max_seg_sz >= PAGE_SIZE)
+			max_tfr = t->max_segs * PAGE_SIZE;
+		else
+			max_tfr = t->max_segs * t->max_seg_sz;
+		if (sz > max_tfr)
+			sz = max_tfr;
+	}
+
+	ssz = sz >> 9;
+	dev_addr = mmc_test_capacity(test->card) / 4;
+	if (tot_sz > dev_addr << 9)
+		tot_sz = dev_addr << 9;
+	cnt = tot_sz / sz;
+	dev_addr &= 0xffff0000; /* Round to 64MiB boundary */
+
+	getnstimeofday(&ts1);
+	for (i = 0; i < cnt; i++) {
+		ret = mmc_test_area_io(test, sz, dev_addr, write,
+				       max_scatter, 0);
+		if (ret)
+			return ret;
+		dev_addr += ssz;
+	}
+	getnstimeofday(&ts2);
+
+	mmc_test_print_avg_rate(test, sz, cnt, &ts1, &ts2);
+
+	return 0;
+}
+
+static int mmc_test_large_seq_perf(struct mmc_test_card *test, int write)
+{
+	int ret, i;
+
+	for (i = 0; i < 10; i++) {
+		ret = mmc_test_seq_perf(test, write, 10 * 1024 * 1024, 1);
+		if (ret)
+			return ret;
+	}
+	for (i = 0; i < 5; i++) {
+		ret = mmc_test_seq_perf(test, write, 100 * 1024 * 1024, 1);
+		if (ret)
+			return ret;
+	}
+	for (i = 0; i < 3; i++) {
+		ret = mmc_test_seq_perf(test, write, 1000 * 1024 * 1024, 1);
+		if (ret)
+			return ret;
+	}
+
+	return ret;
+}
+
+/*
+ * Large sequential read performance.
+ */
+static int mmc_test_large_seq_read_perf(struct mmc_test_card *test)
+{
+	return mmc_test_large_seq_perf(test, 0);
+}
+
+/*
+ * Large sequential write performance.
+ */
+static int mmc_test_large_seq_write_perf(struct mmc_test_card *test)
+{
+	return mmc_test_large_seq_perf(test, 1);
+}
+
 static const struct mmc_test_case mmc_test_cases[] = {
 	{
 		.name = "Basic write (no data verification)",
@@ -2005,6 +2195,34 @@ static const struct mmc_test_case mmc_test_cases[] = {
 		.cleanup = mmc_test_area_cleanup,
 	},
 
+	{
+		.name = "Random read performance by transfer size",
+		.prepare = mmc_test_area_prepare,
+		.run = mmc_test_random_read_perf,
+		.cleanup = mmc_test_area_cleanup,
+	},
+
+	{
+		.name = "Random write performance by transfer size",
+		.prepare = mmc_test_area_prepare,
+		.run = mmc_test_random_write_perf,
+		.cleanup = mmc_test_area_cleanup,
+	},
+
+	{
+		.name = "Large sequential read into scattered pages",
+		.prepare = mmc_test_area_prepare,
+		.run = mmc_test_large_seq_read_perf,
+		.cleanup = mmc_test_area_cleanup,
+	},
+
+	{
+		.name = "Large sequential write from scattered pages",
+		.prepare = mmc_test_area_prepare,
+		.run = mmc_test_large_seq_write_perf,
+		.cleanup = mmc_test_area_cleanup,
+	},
+
 };
 
 static DEFINE_MUTEX(mmc_test_lock);
@@ -2148,11 +2366,11 @@ static int mtf_test_show(struct seq_file *sf, void *data)
 		seq_printf(sf, "Test %d: %d\n", gr->testcase + 1, gr->result);
 
 		list_for_each_entry(tr, &gr->tr_lst, link) {
-			seq_printf(sf, "%u %d %lu.%09lu %u\n",
+			seq_printf(sf, "%u %d %lu.%09lu %u %u.%02u\n",
 				tr->count, tr->sectors,
 				(unsigned long)tr->ts.tv_sec,
 				(unsigned long)tr->ts.tv_nsec,
-				tr->rate);
+				tr->rate, tr->iops / 100, tr->iops % 100);
 		}
 	}
 
diff --git a/drivers/mmc/card/queue.c b/drivers/mmc/card/queue.c
index 4e42d030e09..2ae727568df 100644
--- a/drivers/mmc/card/queue.c
+++ b/drivers/mmc/card/queue.c
@@ -55,8 +55,7 @@ static int mmc_queue_thread(void *d)
 
 		spin_lock_irq(q->queue_lock);
 		set_current_state(TASK_INTERRUPTIBLE);
-		if (!blk_queue_plugged(q))
-			req = blk_fetch_request(q);
+		req = blk_fetch_request(q);
 		mq->req = req;
 		spin_unlock_irq(q->queue_lock);
 
diff --git a/drivers/mmc/core/Makefile b/drivers/mmc/core/Makefile
index 86b47911933..639501970b4 100644
--- a/drivers/mmc/core/Makefile
+++ b/drivers/mmc/core/Makefile
@@ -6,6 +6,7 @@ obj-$(CONFIG_MMC)		+= mmc_core.o
 mmc_core-y			:= core.o bus.o host.o \
 				   mmc.o mmc_ops.o sd.o sd_ops.o \
 				   sdio.o sdio_ops.o sdio_bus.o \
-				   sdio_cis.o sdio_io.o sdio_irq.o
+				   sdio_cis.o sdio_io.o sdio_irq.o \
+				   quirks.o
 
 mmc_core-$(CONFIG_DEBUG_FS)	+= debugfs.o
diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
index 150b5f3cd40..1f453acc868 100644
--- a/drivers/mmc/core/core.c
+++ b/drivers/mmc/core/core.c
@@ -167,8 +167,6 @@ mmc_start_request(struct mmc_host *host, struct mmc_request *mrq)
 
 	WARN_ON(!host->claimed);
 
-	led_trigger_event(host->led, LED_FULL);
-
 	mrq->cmd->error = 0;
 	mrq->cmd->mrq = mrq;
 	if (mrq->data) {
@@ -194,6 +192,7 @@ mmc_start_request(struct mmc_host *host, struct mmc_request *mrq)
 		}
 	}
 	mmc_host_clk_ungate(host);
+	led_trigger_event(host->led, LED_FULL);
 	host->ops->request(host, mrq);
 }
 
@@ -528,7 +527,14 @@ int mmc_try_claim_host(struct mmc_host *host)
 }
 EXPORT_SYMBOL(mmc_try_claim_host);
 
-static void mmc_do_release_host(struct mmc_host *host)
+/**
+ *	mmc_do_release_host - release a claimed host
+ *	@host: mmc host to release
+ *
+ *	If you successfully claimed a host, this function will
+ *	release it again.
+ */
+void mmc_do_release_host(struct mmc_host *host)
 {
 	unsigned long flags;
 
@@ -543,6 +549,7 @@ static void mmc_do_release_host(struct mmc_host *host)
 		wake_up(&host->wq);
 	}
 }
+EXPORT_SYMBOL(mmc_do_release_host);
 
 void mmc_host_deeper_disable(struct work_struct *work)
 {
@@ -1002,6 +1009,13 @@ static void mmc_power_off(struct mmc_host *host)
 {
 	host->ios.clock = 0;
 	host->ios.vdd = 0;
+
+	/*
+	 * Reset ocr mask to be the highest possible voltage supported for
+	 * this mmc host. This value will be used at next power up.
+	 */
+	host->ocr = 1 << (fls(host->ocr_avail) - 1);
+
 	if (!mmc_host_is_spi(host)) {
 		host->ios.bus_mode = MMC_BUSMODE_OPENDRAIN;
 		host->ios.chip_select = MMC_CS_DONTCARE;
@@ -1495,6 +1509,12 @@ static int mmc_rescan_try_freq(struct mmc_host *host, unsigned freq)
 		mmc_hostname(host), __func__, host->f_init);
 #endif
 	mmc_power_up(host);
+
+	/*
+	 * sdio_reset sends CMD52 to reset card.  Since we do not know
+	 * if the card is being re-initialized, just send it.  CMD52
+	 * should be ignored by SD/eMMC cards.
+	 */
 	sdio_reset(host);
 	mmc_go_idle(host);
 
diff --git a/drivers/mmc/core/core.h b/drivers/mmc/core/core.h
index ca1fdde29df..20b1c0831ea 100644
--- a/drivers/mmc/core/core.h
+++ b/drivers/mmc/core/core.h
@@ -61,6 +61,8 @@ int mmc_attach_mmc(struct mmc_host *host);
 int mmc_attach_sd(struct mmc_host *host);
 int mmc_attach_sdio(struct mmc_host *host);
 
+void mmc_fixup_device(struct mmc_card *card);
+
 /* Module parameters */
 extern int use_spi_crc;
 
diff --git a/drivers/mmc/core/host.c b/drivers/mmc/core/host.c
index b3ac6c5bc5c..461e6a17fb9 100644
--- a/drivers/mmc/core/host.c
+++ b/drivers/mmc/core/host.c
@@ -160,10 +160,7 @@ static bool mmc_host_may_gate_card(struct mmc_card *card)
 	 * gate the clock, because there is somebody out there that may still
 	 * be using it.
 	 */
-	if (mmc_card_sdio(card))
-		return false;
-
-	return true;
+	return !(card->quirks & MMC_QUIRK_BROKEN_CLK_GATING);
 }
 
 /**
diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c
index 16006ef153f..772d0d0a541 100644
--- a/drivers/mmc/core/mmc.c
+++ b/drivers/mmc/core/mmc.c
@@ -302,6 +302,44 @@ static int mmc_read_ext_csd(struct mmc_card *card)
 	}
 
 	if (card->ext_csd.rev >= 4) {
+		/*
+		 * Enhanced area feature support -- check whether the eMMC
+		 * card has the Enhanced area enabled.  If so, export enhanced
+		 * area offset and size to user by adding sysfs interface.
+		 */
+		if ((ext_csd[EXT_CSD_PARTITION_SUPPORT] & 0x2) &&
+				(ext_csd[EXT_CSD_PARTITION_ATTRIBUTE] & 0x1)) {
+			u8 hc_erase_grp_sz =
+				ext_csd[EXT_CSD_HC_ERASE_GRP_SIZE];
+			u8 hc_wp_grp_sz =
+				ext_csd[EXT_CSD_HC_WP_GRP_SIZE];
+
+			card->ext_csd.enhanced_area_en = 1;
+			/*
+			 * calculate the enhanced data area offset, in bytes
+			 */
+			card->ext_csd.enhanced_area_offset =
+				(ext_csd[139] << 24) + (ext_csd[138] << 16) +
+				(ext_csd[137] << 8) + ext_csd[136];
+			if (mmc_card_blockaddr(card))
+				card->ext_csd.enhanced_area_offset <<= 9;
+			/*
+			 * calculate the enhanced data area size, in kilobytes
+			 */
+			card->ext_csd.enhanced_area_size =
+				(ext_csd[142] << 16) + (ext_csd[141] << 8) +
+				ext_csd[140];
+			card->ext_csd.enhanced_area_size *=
+				(size_t)(hc_erase_grp_sz * hc_wp_grp_sz);
+			card->ext_csd.enhanced_area_size <<= 9;
+		} else {
+			/*
+			 * If the enhanced area is not enabled, disable these
+			 * device attributes.
+			 */
+			card->ext_csd.enhanced_area_offset = -EINVAL;
+			card->ext_csd.enhanced_area_size = -EINVAL;
+		}
 		card->ext_csd.sec_trim_mult =
 			ext_csd[EXT_CSD_SEC_TRIM_MULT];
 		card->ext_csd.sec_erase_mult =
@@ -336,6 +374,9 @@ MMC_DEV_ATTR(manfid, "0x%06x\n", card->cid.manfid);
 MMC_DEV_ATTR(name, "%s\n", card->cid.prod_name);
 MMC_DEV_ATTR(oemid, "0x%04x\n", card->cid.oemid);
 MMC_DEV_ATTR(serial, "0x%08x\n", card->cid.serial);
+MMC_DEV_ATTR(enhanced_area_offset, "%llu\n",
+		card->ext_csd.enhanced_area_offset);
+MMC_DEV_ATTR(enhanced_area_size, "%u\n", card->ext_csd.enhanced_area_size);
 
 static struct attribute *mmc_std_attrs[] = {
 	&dev_attr_cid.attr,
@@ -349,6 +390,8 @@ static struct attribute *mmc_std_attrs[] = {
 	&dev_attr_name.attr,
 	&dev_attr_oemid.attr,
 	&dev_attr_serial.attr,
+	&dev_attr_enhanced_area_offset.attr,
+	&dev_attr_enhanced_area_size.attr,
 	NULL,
 };
 
@@ -378,6 +421,7 @@ static int mmc_init_card(struct mmc_host *host, u32 ocr,
 	int err, ddr = 0;
 	u32 cid[4];
 	unsigned int max_dtr;
+	u32 rocr;
 
 	BUG_ON(!host);
 	WARN_ON(!host->claimed);
@@ -391,7 +435,7 @@ static int mmc_init_card(struct mmc_host *host, u32 ocr,
 	mmc_go_idle(host);
 
 	/* The extra bit indicates that we support high capacity */
-	err = mmc_send_op_cond(host, ocr | (1 << 30), NULL);
+	err = mmc_send_op_cond(host, ocr | (1 << 30), &rocr);
 	if (err)
 		goto err;
 
@@ -479,11 +523,51 @@ static int mmc_init_card(struct mmc_host *host, u32 ocr,
 		err = mmc_read_ext_csd(card);
 		if (err)
 			goto free_card;
+
+		/* If doing byte addressing, check if required to do sector
+		 * addressing.  Handle the case of <2GB cards needing sector
+		 * addressing.  See section 8.1 JEDEC Standard JED84-A441;
+		 * ocr register has bit 30 set for sector addressing.
+		 */
+		if (!(mmc_card_blockaddr(card)) && (rocr & (1<<30)))
+			mmc_card_set_blockaddr(card);
+
 		/* Erase size depends on CSD and Extended CSD */
 		mmc_set_erase_size(card);
 	}
 
 	/*
+	 * If enhanced_area_en is TRUE, host needs to enable ERASE_GRP_DEF
+	 * bit.  This bit will be lost every time after a reset or power off.
+	 */
+	if (card->ext_csd.enhanced_area_en) {
+		err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
+				EXT_CSD_ERASE_GROUP_DEF, 1);
+
+		if (err && err != -EBADMSG)
+			goto free_card;
+
+		if (err) {
+			err = 0;
+			/*
+			 * Just disable enhanced area off & sz
+			 * will try to enable ERASE_GROUP_DEF
+			 * during next time reinit
+			 */
+			card->ext_csd.enhanced_area_offset = -EINVAL;
+			card->ext_csd.enhanced_area_size = -EINVAL;
+		} else {
+			card->ext_csd.erase_group_def = 1;
+			/*
+			 * enable ERASE_GRP_DEF successfully.
+			 * This will affect the erase size, so
+			 * here need to reset erase size
+			 */
+			mmc_set_erase_size(card);
+		}
+	}
+
+	/*
 	 * Activate high speed (if supported)
 	 */
 	if ((card->ext_csd.hs_max_dtr != 0) &&
diff --git a/drivers/mmc/core/mmc_ops.c b/drivers/mmc/core/mmc_ops.c
index 60842f878de..f3b22bf89cc 100644
--- a/drivers/mmc/core/mmc_ops.c
+++ b/drivers/mmc/core/mmc_ops.c
@@ -105,7 +105,7 @@ int mmc_go_idle(struct mmc_host *host)
 	 * that in case of hardware that won't pull up DAT3/nCS otherwise.
 	 *
 	 * SPI hosts ignore ios.chip_select; it's managed according to
-	 * rules that must accomodate non-MMC slaves which this layer
+	 * rules that must accommodate non-MMC slaves which this layer
 	 * won't even know about.
 	 */
 	if (!mmc_host_is_spi(host)) {
diff --git a/drivers/mmc/core/quirks.c b/drivers/mmc/core/quirks.c
new file mode 100644
index 00000000000..11118b74eb2
--- /dev/null
+++ b/drivers/mmc/core/quirks.c
@@ -0,0 +1,84 @@
+/*
+ *  This file contains work-arounds for many known sdio hardware
+ *  bugs.
+ *
+ *  Copyright (c) 2011 Pierre Tardy <tardyp@gmail.com>
+ *  Inspired from pci fixup code:
+ *  Copyright (c) 1999 Martin Mares <mj@ucw.cz>
+ *
+ */
+
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/mmc/card.h>
+#include <linux/mod_devicetable.h>
+
+/*
+ *  The world is not perfect and supplies us with broken mmc/sdio devices.
+ *  For at least a part of these bugs we need a work-around
+ */
+
+struct mmc_fixup {
+	u16 vendor, device;	/* You can use SDIO_ANY_ID here of course */
+	void (*vendor_fixup)(struct mmc_card *card, int data);
+	int data;
+};
+
+/*
+ * This hook just adds a quirk unconditionnally
+ */
+static void __maybe_unused add_quirk(struct mmc_card *card, int data)
+{
+	card->quirks |= data;
+}
+
+/*
+ * This hook just removes a quirk unconditionnally
+ */
+static void __maybe_unused remove_quirk(struct mmc_card *card, int data)
+{
+	card->quirks &= ~data;
+}
+
+/*
+ * This hook just adds a quirk for all sdio devices
+ */
+static void add_quirk_for_sdio_devices(struct mmc_card *card, int data)
+{
+	if (mmc_card_sdio(card))
+		card->quirks |= data;
+}
+
+#ifndef SDIO_VENDOR_ID_TI
+#define SDIO_VENDOR_ID_TI		0x0097
+#endif
+
+#ifndef SDIO_DEVICE_ID_TI_WL1271
+#define SDIO_DEVICE_ID_TI_WL1271	0x4076
+#endif
+
+static const struct mmc_fixup mmc_fixup_methods[] = {
+	/* by default sdio devices are considered CLK_GATING broken */
+	/* good cards will be whitelisted as they are tested */
+	{ SDIO_ANY_ID, SDIO_ANY_ID,
+		add_quirk_for_sdio_devices, MMC_QUIRK_BROKEN_CLK_GATING },
+	{ SDIO_VENDOR_ID_TI, SDIO_DEVICE_ID_TI_WL1271,
+		remove_quirk, MMC_QUIRK_BROKEN_CLK_GATING },
+	{ 0 }
+};
+
+void mmc_fixup_device(struct mmc_card *card)
+{
+	const struct mmc_fixup *f;
+
+	for (f = mmc_fixup_methods; f->vendor_fixup; f++) {
+		if ((f->vendor == card->cis.vendor
+		     || f->vendor == (u16) SDIO_ANY_ID) &&
+		    (f->device == card->cis.device
+		     || f->device == (u16) SDIO_ANY_ID)) {
+			dev_dbg(&card->dev, "calling %pF\n", f->vendor_fixup);
+			f->vendor_fixup(card, f->data);
+		}
+	}
+}
+EXPORT_SYMBOL(mmc_fixup_device);
diff --git a/drivers/mmc/core/sd.c b/drivers/mmc/core/sd.c
index d18c32bca99..6dac89fe053 100644
--- a/drivers/mmc/core/sd.c
+++ b/drivers/mmc/core/sd.c
@@ -21,6 +21,7 @@
 #include "core.h"
 #include "bus.h"
 #include "mmc_ops.h"
+#include "sd.h"
 #include "sd_ops.h"
 
 static const unsigned int tran_exp[] = {
diff --git a/drivers/mmc/core/sd_ops.c b/drivers/mmc/core/sd_ops.c
index 797cdb5887f..76af349c14b 100644
--- a/drivers/mmc/core/sd_ops.c
+++ b/drivers/mmc/core/sd_ops.c
@@ -9,6 +9,7 @@
  * your option) any later version.
  */
 
+#include <linux/slab.h>
 #include <linux/types.h>
 #include <linux/scatterlist.h>
 
@@ -252,6 +253,7 @@ int mmc_app_send_scr(struct mmc_card *card, u32 *scr)
 	struct mmc_command cmd;
 	struct mmc_data data;
 	struct scatterlist sg;
+	void *data_buf;
 
 	BUG_ON(!card);
 	BUG_ON(!card->host);
@@ -263,6 +265,13 @@ int mmc_app_send_scr(struct mmc_card *card, u32 *scr)
 	if (err)
 		return err;
 
+	/* dma onto stack is unsafe/nonportable, but callers to this
+	 * routine normally provide temporary on-stack buffers ...
+	 */
+	data_buf = kmalloc(sizeof(card->raw_scr), GFP_KERNEL);
+	if (data_buf == NULL)
+		return -ENOMEM;
+
 	memset(&mrq, 0, sizeof(struct mmc_request));
 	memset(&cmd, 0, sizeof(struct mmc_command));
 	memset(&data, 0, sizeof(struct mmc_data));
@@ -280,12 +289,15 @@ int mmc_app_send_scr(struct mmc_card *card, u32 *scr)
 	data.sg = &sg;
 	data.sg_len = 1;
 
-	sg_init_one(&sg, scr, 8);
+	sg_init_one(&sg, data_buf, 8);
 
 	mmc_set_data_timeout(&data, card);
 
 	mmc_wait_for_req(card->host, &mrq);
 
+	memcpy(scr, data_buf, sizeof(card->raw_scr));
+	kfree(data_buf);
+
 	if (cmd.error)
 		return cmd.error;
 	if (data.error)
diff --git a/drivers/mmc/core/sdio.c b/drivers/mmc/core/sdio.c
index ebc62ad4cc5..db0f0b44d68 100644
--- a/drivers/mmc/core/sdio.c
+++ b/drivers/mmc/core/sdio.c
@@ -395,6 +395,14 @@ static int mmc_sdio_init_card(struct mmc_host *host, u32 ocr,
 		if (err)
 			goto remove;
 
+		/*
+		 * Update oldcard with the new RCA received from the SDIO
+		 * device -- we're doing this so that it's updated in the
+		 * "card" struct when oldcard overwrites that later.
+		 */
+		if (oldcard)
+			oldcard->rca = card->rca;
+
 		mmc_set_bus_mode(host, MMC_BUSMODE_PUSHPULL);
 	}
 
@@ -458,6 +466,7 @@ static int mmc_sdio_init_card(struct mmc_host *host, u32 ocr,
 
 		card = oldcard;
 	}
+	mmc_fixup_device(card);
 
 	if (card->type == MMC_TYPE_SD_COMBO) {
 		err = mmc_sd_setup_card(host, card, oldcard != NULL);
diff --git a/drivers/mmc/core/sdio_irq.c b/drivers/mmc/core/sdio_irq.c
index bb192f90e8e..b3001617e67 100644
--- a/drivers/mmc/core/sdio_irq.c
+++ b/drivers/mmc/core/sdio_irq.c
@@ -45,7 +45,7 @@ static int process_sdio_pending_irqs(struct mmc_card *card)
 			struct sdio_func *func = card->sdio_func[i - 1];
 			if (!func) {
 				printk(KERN_WARNING "%s: pending IRQ for "
-					"non-existant function\n",
+					"non-existent function\n",
 					mmc_card_id(card));
 				ret = -EINVAL;
 			} else if (func->irq_handler) {
diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig
index 54f91321749..94df40531c3 100644
--- a/drivers/mmc/host/Kconfig
+++ b/drivers/mmc/host/Kconfig
@@ -311,7 +311,7 @@ config MMC_MSM
 
 config MMC_MXC
 	tristate "Freescale i.MX2/3 Multimedia Card Interface support"
-	depends on ARCH_MXC
+	depends on MACH_MX21 || MACH_MX27 || ARCH_MX31
 	help
 	  This selects the Freescale i.MX2/3 Multimedia card Interface.
 	  If you have a i.MX platform with a Multimedia Card slot,
@@ -319,6 +319,15 @@ config MMC_MXC
 
 	  If unsure, say N.
 
+config MMC_MXS
+	tristate "Freescale MXS Multimedia Card Interface support"
+	depends on ARCH_MXS && MXS_DMA
+	help
+	  This selects the Freescale SSP MMC controller found on MXS based
+	  platforms like mx23/28.
+
+	  If unsure, say N.
+
 config MMC_TIFM_SD
 	tristate "TI Flash Media MMC/SD Interface support  (EXPERIMENTAL)"
 	depends on EXPERIMENTAL && PCI
@@ -430,13 +439,25 @@ config MMC_SDRICOH_CS
 	  To compile this driver as a module, choose M here: the
 	  module will be called sdricoh_cs.
 
+config MMC_TMIO_CORE
+	tristate
+
 config MMC_TMIO
 	tristate "Toshiba Mobile IO Controller (TMIO) MMC/SD function support"
-	depends on MFD_TMIO || MFD_ASIC3 || MFD_SH_MOBILE_SDHI
+	depends on MFD_TMIO || MFD_ASIC3
+	select MMC_TMIO_CORE
 	help
 	  This provides support for the SD/MMC cell found in TC6393XB,
 	  T7L66XB and also HTC ASIC3
 
+config MMC_SDHI
+	tristate "SH-Mobile SDHI SD/SDIO controller support"
+	depends on SUPERH || ARCH_SHMOBILE
+	select MMC_TMIO_CORE
+	help
+	  This provides support for the SDHI SD/SDIO controller found in
+	  SuperH and ARM SH-Mobile SoCs
+
 config MMC_CB710
 	tristate "ENE CB710 MMC/SD Interface support"
 	depends on PCI
diff --git a/drivers/mmc/host/Makefile b/drivers/mmc/host/Makefile
index e834fb223e9..4f1df0aae57 100644
--- a/drivers/mmc/host/Makefile
+++ b/drivers/mmc/host/Makefile
@@ -6,6 +6,7 @@ obj-$(CONFIG_MMC_ARMMMCI)	+= mmci.o
 obj-$(CONFIG_MMC_PXA)		+= pxamci.o
 obj-$(CONFIG_MMC_IMX)		+= imxmmc.o
 obj-$(CONFIG_MMC_MXC)		+= mxcmmc.o
+obj-$(CONFIG_MMC_MXS)		+= mxs-mmc.o
 obj-$(CONFIG_MMC_SDHCI)		+= sdhci.o
 obj-$(CONFIG_MMC_SDHCI_PCI)	+= sdhci-pci.o
 obj-$(CONFIG_MMC_SDHCI_PXA)	+= sdhci-pxa.o
@@ -28,7 +29,13 @@ endif
 obj-$(CONFIG_MMC_S3C)   	+= s3cmci.o
 obj-$(CONFIG_MMC_SDRICOH_CS)	+= sdricoh_cs.o
 obj-$(CONFIG_MMC_TMIO)		+= tmio_mmc.o
-obj-$(CONFIG_MMC_CB710)	+= cb710-mmc.o
+obj-$(CONFIG_MMC_TMIO_CORE)	+= tmio_mmc_core.o
+tmio_mmc_core-y			:= tmio_mmc_pio.o
+ifneq ($(CONFIG_MMC_SDHI),n)
+tmio_mmc_core-y			+= tmio_mmc_dma.o
+endif
+obj-$(CONFIG_MMC_SDHI)		+= sh_mobile_sdhi.o
+obj-$(CONFIG_MMC_CB710)		+= cb710-mmc.o
 obj-$(CONFIG_MMC_VIA_SDMMC)	+= via-sdmmc.o
 obj-$(CONFIG_SDH_BFIN)		+= bfin_sdh.o
 obj-$(CONFIG_MMC_DW)		+= dw_mmc.o
diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c
index ad2a7a032cd..ea3888b65d5 100644
--- a/drivers/mmc/host/atmel-mci.c
+++ b/drivers/mmc/host/atmel-mci.c
@@ -127,7 +127,7 @@ struct atmel_mci_dma {
  * EVENT_DATA_COMPLETE is set in @pending_events, all data-related
  * interrupts must be disabled and @data_status updated with a
  * snapshot of SR. Similarly, before EVENT_CMD_COMPLETE is set, the
- * CMDRDY interupt must be disabled and @cmd_status updated with a
+ * CMDRDY interrupt must be disabled and @cmd_status updated with a
  * snapshot of SR, and before EVENT_XFER_COMPLETE can be set, the
  * bytes_xfered field of @data must be written. This is ensured by
  * using barriers.
@@ -578,7 +578,8 @@ static void atmci_dma_cleanup(struct atmel_mci *host)
 	struct mmc_data			*data = host->data;
 
 	if (data)
-		dma_unmap_sg(&host->pdev->dev, data->sg, data->sg_len,
+		dma_unmap_sg(host->dma.chan->device->dev,
+			     data->sg, data->sg_len,
 			     ((data->flags & MMC_DATA_WRITE)
 			      ? DMA_TO_DEVICE : DMA_FROM_DEVICE));
 }
@@ -588,7 +589,7 @@ static void atmci_stop_dma(struct atmel_mci *host)
 	struct dma_chan *chan = host->data_chan;
 
 	if (chan) {
-	  chan->device->device_control(chan, DMA_TERMINATE_ALL, 0);
+		dmaengine_terminate_all(chan);
 		atmci_dma_cleanup(host);
 	} else {
 		/* Data transfer was stopped by the interrupt handler */
@@ -684,11 +685,11 @@ atmci_prepare_data_dma(struct atmel_mci *host, struct mmc_data *data)
 	else
 		direction = DMA_TO_DEVICE;
 
-	sglen = dma_map_sg(&host->pdev->dev, data->sg, data->sg_len, direction);
-	if (sglen != data->sg_len)
-		goto unmap_exit;
+	sglen = dma_map_sg(chan->device->dev, data->sg,
+			   data->sg_len, direction);
+
 	desc = chan->device->device_prep_slave_sg(chan,
-			data->sg, data->sg_len, direction,
+			data->sg, sglen, direction,
 			DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
 	if (!desc)
 		goto unmap_exit;
@@ -699,7 +700,7 @@ atmci_prepare_data_dma(struct atmel_mci *host, struct mmc_data *data)
 
 	return 0;
 unmap_exit:
-	dma_unmap_sg(&host->pdev->dev, data->sg, sglen, direction);
+	dma_unmap_sg(chan->device->dev, data->sg, data->sg_len, direction);
 	return -ENOMEM;
 }
 
@@ -709,8 +710,8 @@ static void atmci_submit_data(struct atmel_mci *host)
 	struct dma_async_tx_descriptor	*desc = host->dma.data_desc;
 
 	if (chan) {
-		desc->tx_submit(desc);
-		chan->device->device_issue_pending(chan);
+		dmaengine_submit(desc);
+		dma_async_issue_pending(chan);
 	}
 }
 
@@ -1081,7 +1082,7 @@ static void atmci_request_end(struct atmel_mci *host, struct mmc_request *mrq)
 	/*
 	 * Update the MMC clock rate if necessary. This may be
 	 * necessary if set_ios() is called when a different slot is
-	 * busy transfering data.
+	 * busy transferring data.
 	 */
 	if (host->need_clock_update) {
 		mci_writel(host, MR, host->mode_reg);
diff --git a/drivers/mmc/host/cb710-mmc.c b/drivers/mmc/host/cb710-mmc.c
index 66b4ce587f4..ce2a47b71dd 100644
--- a/drivers/mmc/host/cb710-mmc.c
+++ b/drivers/mmc/host/cb710-mmc.c
@@ -205,7 +205,7 @@ static int cb710_wait_while_busy(struct cb710_slot *slot, uint8_t mask)
 			"WAIT12: waited %d loops, mask %02X, entry val %08X, exit val %08X\n",
 			limit, mask, e, x);
 #endif
-	return 0;
+	return err;
 }
 
 static void cb710_mmc_set_transfer_size(struct cb710_slot *slot,
diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
index 2fcc82577c1..87e1f57ec9b 100644
--- a/drivers/mmc/host/dw_mmc.c
+++ b/drivers/mmc/host/dw_mmc.c
@@ -32,6 +32,7 @@
 #include <linux/mmc/mmc.h>
 #include <linux/mmc/dw_mmc.h>
 #include <linux/bitops.h>
+#include <linux/regulator/consumer.h>
 
 #include "dw_mmc.h"
 
@@ -315,7 +316,7 @@ static void dw_mci_idmac_stop_dma(struct dw_mci *host)
 
 	/* Stop the IDMAC running */
 	temp = mci_readl(host, BMOD);
-	temp &= ~SDMMC_IDMAC_ENABLE;
+	temp &= ~(SDMMC_IDMAC_ENABLE | SDMMC_IDMAC_FB);
 	mci_writel(host, BMOD, temp);
 }
 
@@ -384,7 +385,7 @@ static void dw_mci_idmac_start_dma(struct dw_mci *host, unsigned int sg_len)
 
 	/* Enable the IDMAC */
 	temp = mci_readl(host, BMOD);
-	temp |= SDMMC_IDMAC_ENABLE;
+	temp |= SDMMC_IDMAC_ENABLE | SDMMC_IDMAC_FB;
 	mci_writel(host, BMOD, temp);
 
 	/* Start it running */
@@ -562,7 +563,8 @@ static void dw_mci_setup_bus(struct dw_mci_slot *slot)
 			     SDMMC_CMD_UPD_CLK | SDMMC_CMD_PRV_DAT_WAIT, 0);
 
 		/* enable clock */
-		mci_writel(host, CLKENA, SDMMC_CLKEN_ENABLE);
+		mci_writel(host, CLKENA, SDMMC_CLKEN_ENABLE |
+			   SDMMC_CLKEN_LOW_PWR);
 
 		/* inform CIU */
 		mci_send_cmd(slot,
@@ -661,6 +663,7 @@ static void dw_mci_request(struct mmc_host *mmc, struct mmc_request *mrq)
 static void dw_mci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 {
 	struct dw_mci_slot *slot = mmc_priv(mmc);
+	u32 regs;
 
 	/* set default 1 bit mode */
 	slot->ctype = SDMMC_CTYPE_1BIT;
@@ -672,6 +675,16 @@ static void dw_mci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 	case MMC_BUS_WIDTH_4:
 		slot->ctype = SDMMC_CTYPE_4BIT;
 		break;
+	case MMC_BUS_WIDTH_8:
+		slot->ctype = SDMMC_CTYPE_8BIT;
+		break;
+	}
+
+	/* DDR mode set */
+	if (ios->ddr) {
+		regs = mci_readl(slot->host, UHS_REG);
+		regs |= (0x1 << slot->id) << 16;
+		mci_writel(slot->host, UHS_REG, regs);
 	}
 
 	if (ios->clock) {
@@ -717,7 +730,9 @@ static int dw_mci_get_cd(struct mmc_host *mmc)
 	struct dw_mci_board *brd = slot->host->pdata;
 
 	/* Use platform get_cd function, else try onboard card detect */
-	if (brd->get_cd)
+	if (brd->quirks & DW_MCI_QUIRK_BROKEN_CARD_DETECTION)
+		present = 1;
+	else if (brd->get_cd)
 		present = !brd->get_cd(slot->id);
 	else
 		present = (mci_readl(slot->host, CDETECT) & (1 << slot->id))
@@ -1019,13 +1034,10 @@ static void dw_mci_read_data_pio(struct dw_mci *host)
 	struct mmc_data	*data = host->data;
 	int shift = host->data_shift;
 	u32 status;
-	unsigned int nbytes = 0, len, old_len, count = 0;
+	unsigned int nbytes = 0, len;
 
 	do {
 		len = SDMMC_GET_FCNT(mci_readl(host, STATUS)) << shift;
-		if (count == 0)
-			old_len = len;
-
 		if (offset + len <= sg->length) {
 			host->pull_data(host, (void *)(buf + offset), len);
 
@@ -1070,7 +1082,6 @@ static void dw_mci_read_data_pio(struct dw_mci *host)
 			tasklet_schedule(&host->tasklet);
 			return;
 		}
-		count++;
 	} while (status & SDMMC_INT_RXDR); /*if the RXDR is ready read again*/
 	len = SDMMC_GET_FCNT(mci_readl(host, STATUS));
 	host->pio_offset = offset;
@@ -1395,7 +1406,11 @@ static int __init dw_mci_init_slot(struct dw_mci *host, unsigned int id)
 	if (host->pdata->setpower)
 		host->pdata->setpower(id, 0);
 
-	mmc->caps = 0;
+	if (host->pdata->caps)
+		mmc->caps = host->pdata->caps;
+	else
+		mmc->caps = 0;
+
 	if (host->pdata->get_bus_wd)
 		if (host->pdata->get_bus_wd(slot->id) >= 4)
 			mmc->caps |= MMC_CAP_4_BIT_DATA;
@@ -1426,6 +1441,13 @@ static int __init dw_mci_init_slot(struct dw_mci *host, unsigned int id)
 	}
 #endif /* CONFIG_MMC_DW_IDMAC */
 
+	host->vmmc = regulator_get(mmc_dev(mmc), "vmmc");
+	if (IS_ERR(host->vmmc)) {
+		printk(KERN_INFO "%s: no vmmc regulator found\n", mmc_hostname(mmc));
+		host->vmmc = NULL;
+	} else
+		regulator_enable(host->vmmc);
+
 	if (dw_mci_get_cd(mmc))
 		set_bit(DW_MMC_CARD_PRESENT, &slot->flags);
 	else
@@ -1441,6 +1463,12 @@ static int __init dw_mci_init_slot(struct dw_mci *host, unsigned int id)
 	/* Card initially undetected */
 	slot->last_detect_state = 0;
 
+	/*
+	 * Card may have been plugged in prior to boot so we
+	 * need to run the detect tasklet
+	 */
+	tasklet_schedule(&host->card_tasklet);
+
 	return 0;
 }
 
@@ -1619,8 +1647,9 @@ static int dw_mci_probe(struct platform_device *pdev)
 	 */
 	fifo_size = mci_readl(host, FIFOTH);
 	fifo_size = (fifo_size >> 16) & 0x7ff;
-	mci_writel(host, FIFOTH, ((0x2 << 28) | ((fifo_size/2 - 1) << 16) |
-				  ((fifo_size/2) << 0)));
+	host->fifoth_val = ((0x2 << 28) | ((fifo_size/2 - 1) << 16) |
+			((fifo_size/2) << 0));
+	mci_writel(host, FIFOTH, host->fifoth_val);
 
 	/* disable clock to CIU */
 	mci_writel(host, CLKENA, 0);
@@ -1683,6 +1712,12 @@ err_dmaunmap:
 			  host->sg_cpu, host->sg_dma);
 	iounmap(host->regs);
 
+	if (host->vmmc) {
+		regulator_disable(host->vmmc);
+		regulator_put(host->vmmc);
+	}
+
+
 err_freehost:
 	kfree(host);
 	return ret;
@@ -1714,6 +1749,11 @@ static int __exit dw_mci_remove(struct platform_device *pdev)
 	if (host->use_dma && host->dma_ops->exit)
 		host->dma_ops->exit(host);
 
+	if (host->vmmc) {
+		regulator_disable(host->vmmc);
+		regulator_put(host->vmmc);
+	}
+
 	iounmap(host->regs);
 
 	kfree(host);
@@ -1729,6 +1769,9 @@ static int dw_mci_suspend(struct platform_device *pdev, pm_message_t mesg)
 	int i, ret;
 	struct dw_mci *host = platform_get_drvdata(pdev);
 
+	if (host->vmmc)
+		regulator_enable(host->vmmc);
+
 	for (i = 0; i < host->num_slots; i++) {
 		struct dw_mci_slot *slot = host->slot[i];
 		if (!slot)
@@ -1744,6 +1787,9 @@ static int dw_mci_suspend(struct platform_device *pdev, pm_message_t mesg)
 		}
 	}
 
+	if (host->vmmc)
+		regulator_disable(host->vmmc);
+
 	return 0;
 }
 
@@ -1752,6 +1798,23 @@ static int dw_mci_resume(struct platform_device *pdev)
 	int i, ret;
 	struct dw_mci *host = platform_get_drvdata(pdev);
 
+	if (host->dma_ops->init)
+		host->dma_ops->init(host);
+
+	if (!mci_wait_reset(&pdev->dev, host)) {
+		ret = -ENODEV;
+		return ret;
+	}
+
+	/* Restore the old value at FIFOTH register */
+	mci_writel(host, FIFOTH, host->fifoth_val);
+
+	mci_writel(host, RINTSTS, 0xFFFFFFFF);
+	mci_writel(host, INTMASK, SDMMC_INT_CMD_DONE | SDMMC_INT_DATA_OVER |
+		   SDMMC_INT_TXDR | SDMMC_INT_RXDR |
+		   DW_MCI_ERROR_FLAGS | SDMMC_INT_CD);
+	mci_writel(host, CTRL, SDMMC_CTRL_INT_ENABLE);
+
 	for (i = 0; i < host->num_slots; i++) {
 		struct dw_mci_slot *slot = host->slot[i];
 		if (!slot)
diff --git a/drivers/mmc/host/dw_mmc.h b/drivers/mmc/host/dw_mmc.h
index 5dd55a75233..23c662af561 100644
--- a/drivers/mmc/host/dw_mmc.h
+++ b/drivers/mmc/host/dw_mmc.h
@@ -43,6 +43,7 @@
 #define SDMMC_USRID		0x068
 #define SDMMC_VERID		0x06c
 #define SDMMC_HCON		0x070
+#define SDMMC_UHS_REG		0x074
 #define SDMMC_BMOD		0x080
 #define SDMMC_PLDMND		0x084
 #define SDMMC_DBADDR		0x088
@@ -51,7 +52,6 @@
 #define SDMMC_DSCADDR		0x094
 #define SDMMC_BUFADDR		0x098
 #define SDMMC_DATA		0x100
-#define SDMMC_DATA_ADR		0x100
 
 /* shift bit field */
 #define _SBF(f, v)		((v) << (f))
diff --git a/drivers/mmc/host/mmc_spi.c b/drivers/mmc/host/mmc_spi.c
index 2f7fc0c5146..7c1e16aaf17 100644
--- a/drivers/mmc/host/mmc_spi.c
+++ b/drivers/mmc/host/mmc_spi.c
@@ -99,7 +99,7 @@
 #define r1b_timeout		(HZ * 3)
 
 /* One of the critical speed parameters is the amount of data which may
- * be transfered in one command. If this value is too low, the SD card
+ * be transferred in one command. If this value is too low, the SD card
  * controller has to do multiple partial block writes (argggh!). With
  * today (2008) SD cards there is little speed gain if we transfer more
  * than 64 KBytes at a time. So use this value until there is any indication
diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
index 5bbb87d1025..b4a7e4fba90 100644
--- a/drivers/mmc/host/mmci.c
+++ b/drivers/mmc/host/mmci.c
@@ -68,6 +68,12 @@ static struct variant_data variant_arm = {
 	.datalength_bits	= 16,
 };
 
+static struct variant_data variant_arm_extended_fifo = {
+	.fifosize		= 128 * 4,
+	.fifohalfsize		= 64 * 4,
+	.datalength_bits	= 16,
+};
+
 static struct variant_data variant_u300 = {
 	.fifosize		= 16 * 4,
 	.fifohalfsize		= 8 * 4,
@@ -1277,10 +1283,15 @@ static int mmci_resume(struct amba_device *dev)
 static struct amba_id mmci_ids[] = {
 	{
 		.id	= 0x00041180,
-		.mask	= 0x000fffff,
+		.mask	= 0xff0fffff,
 		.data	= &variant_arm,
 	},
 	{
+		.id	= 0x01041180,
+		.mask	= 0xff0fffff,
+		.data	= &variant_arm_extended_fifo,
+	},
+	{
 		.id	= 0x00041181,
 		.mask	= 0x000fffff,
 		.data	= &variant_arm,
diff --git a/drivers/mmc/host/msm_sdcc.c b/drivers/mmc/host/msm_sdcc.c
index 97c9b3638d5..a4c865a5286 100644
--- a/drivers/mmc/host/msm_sdcc.c
+++ b/drivers/mmc/host/msm_sdcc.c
@@ -267,14 +267,6 @@ msmsdcc_dma_complete_tlet(unsigned long data)
 	dma_unmap_sg(mmc_dev(host->mmc), host->dma.sg, host->dma.num_ents,
 		     host->dma.dir);
 
-	if (host->curr.user_pages) {
-		struct scatterlist *sg = host->dma.sg;
-		int i;
-
-		for (i = 0; i < host->dma.num_ents; i++)
-			flush_dcache_page(sg_page(sg++));
-	}
-
 	host->dma.sg = NULL;
 	host->dma.busy = 0;
 
diff --git a/drivers/mmc/host/mxcmmc.c b/drivers/mmc/host/mxcmmc.c
index 4428594261c..cc20e025932 100644
--- a/drivers/mmc/host/mxcmmc.c
+++ b/drivers/mmc/host/mxcmmc.c
@@ -32,16 +32,14 @@
 #include <linux/io.h>
 #include <linux/gpio.h>
 #include <linux/regulator/consumer.h>
+#include <linux/dmaengine.h>
 
 #include <asm/dma.h>
 #include <asm/irq.h>
 #include <asm/sizes.h>
 #include <mach/mmc.h>
 
-#ifdef CONFIG_ARCH_MX2
-#include <mach/dma-mx1-mx2.h>
-#define HAS_DMA
-#endif
+#include <mach/dma.h>
 
 #define DRIVER_NAME "mxc-mmc"
 
@@ -118,7 +116,8 @@ struct mxcmci_host {
 	void __iomem		*base;
 	int			irq;
 	int			detect_irq;
-	int			dma;
+	struct dma_chan		*dma;
+	struct dma_async_tx_descriptor *desc;
 	int			do_dma;
 	int			default_irq_mask;
 	int			use_sdio;
@@ -129,7 +128,6 @@ struct mxcmci_host {
 	struct mmc_command	*cmd;
 	struct mmc_data		*data;
 
-	unsigned int		dma_nents;
 	unsigned int		datasize;
 	unsigned int		dma_dir;
 
@@ -144,6 +142,11 @@ struct mxcmci_host {
 	spinlock_t		lock;
 
 	struct regulator	*vcc;
+
+	int			burstlen;
+	int			dmareq;
+	struct dma_slave_config dma_slave_config;
+	struct imx_dma_data	dma_data;
 };
 
 static void mxcmci_set_clk_rate(struct mxcmci_host *host, unsigned int clk_ios);
@@ -206,17 +209,16 @@ static void mxcmci_softreset(struct mxcmci_host *host)
 
 	writew(0xff, host->base + MMC_REG_RES_TO);
 }
+static int mxcmci_setup_dma(struct mmc_host *mmc);
 
 static int mxcmci_setup_data(struct mxcmci_host *host, struct mmc_data *data)
 {
 	unsigned int nob = data->blocks;
 	unsigned int blksz = data->blksz;
 	unsigned int datasize = nob * blksz;
-#ifdef HAS_DMA
 	struct scatterlist *sg;
-	int i;
-	int ret;
-#endif
+	int i, nents;
+
 	if (data->flags & MMC_DATA_STREAM)
 		nob = 0xffff;
 
@@ -227,7 +229,9 @@ static int mxcmci_setup_data(struct mxcmci_host *host, struct mmc_data *data)
 	writew(blksz, host->base + MMC_REG_BLK_LEN);
 	host->datasize = datasize;
 
-#ifdef HAS_DMA
+	if (!mxcmci_use_dma(host))
+		return 0;
+
 	for_each_sg(data->sg, sg, data->sg_len, i) {
 		if (sg->offset & 3 || sg->length & 3) {
 			host->do_dma = 0;
@@ -235,34 +239,30 @@ static int mxcmci_setup_data(struct mxcmci_host *host, struct mmc_data *data)
 		}
 	}
 
-	if (data->flags & MMC_DATA_READ) {
+	if (data->flags & MMC_DATA_READ)
 		host->dma_dir = DMA_FROM_DEVICE;
-		host->dma_nents = dma_map_sg(mmc_dev(host->mmc), data->sg,
-					     data->sg_len,  host->dma_dir);
-
-		ret = imx_dma_setup_sg(host->dma, data->sg, host->dma_nents,
-				datasize,
-				host->res->start + MMC_REG_BUFFER_ACCESS,
-				DMA_MODE_READ);
-	} else {
+	else
 		host->dma_dir = DMA_TO_DEVICE;
-		host->dma_nents = dma_map_sg(mmc_dev(host->mmc), data->sg,
-					     data->sg_len,  host->dma_dir);
 
-		ret = imx_dma_setup_sg(host->dma, data->sg, host->dma_nents,
-				datasize,
-				host->res->start + MMC_REG_BUFFER_ACCESS,
-				DMA_MODE_WRITE);
-	}
+	nents = dma_map_sg(host->dma->device->dev, data->sg,
+				     data->sg_len,  host->dma_dir);
+	if (nents != data->sg_len)
+		return -EINVAL;
+
+	host->desc = host->dma->device->device_prep_slave_sg(host->dma,
+		data->sg, data->sg_len, host->dma_dir,
+		DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
 
-	if (ret) {
-		dev_err(mmc_dev(host->mmc), "failed to setup DMA : %d\n", ret);
-		return ret;
+	if (!host->desc) {
+		dma_unmap_sg(host->dma->device->dev, data->sg, data->sg_len,
+				host->dma_dir);
+		host->do_dma = 0;
+		return 0; /* Fall back to PIO */
 	}
 	wmb();
 
-	imx_dma_enable(host->dma);
-#endif /* HAS_DMA */
+	dmaengine_submit(host->desc);
+
 	return 0;
 }
 
@@ -337,13 +337,11 @@ static int mxcmci_finish_data(struct mxcmci_host *host, unsigned int stat)
 	struct mmc_data *data = host->data;
 	int data_error;
 
-#ifdef HAS_DMA
 	if (mxcmci_use_dma(host)) {
-		imx_dma_disable(host->dma);
-		dma_unmap_sg(mmc_dev(host->mmc), data->sg, host->dma_nents,
+		dmaengine_terminate_all(host->dma);
+		dma_unmap_sg(host->dma->device->dev, data->sg, data->sg_len,
 				host->dma_dir);
 	}
-#endif
 
 	if (stat & STATUS_ERR_MASK) {
 		dev_dbg(mmc_dev(host->mmc), "request failed. status: 0x%08x\n",
@@ -545,7 +543,6 @@ static void mxcmci_datawork(struct work_struct *work)
 	}
 }
 
-#ifdef HAS_DMA
 static void mxcmci_data_done(struct mxcmci_host *host, unsigned int stat)
 {
 	struct mmc_data *data = host->data;
@@ -568,7 +565,6 @@ static void mxcmci_data_done(struct mxcmci_host *host, unsigned int stat)
 		mxcmci_finish_request(host, host->req);
 	}
 }
-#endif /* HAS_DMA */
 
 static void mxcmci_cmd_done(struct mxcmci_host *host, unsigned int stat)
 {
@@ -606,12 +602,10 @@ static irqreturn_t mxcmci_irq(int irq, void *devid)
 	sdio_irq = (stat & STATUS_SDIO_INT_ACTIVE) && host->use_sdio;
 	spin_unlock_irqrestore(&host->lock, flags);
 
-#ifdef HAS_DMA
 	if (mxcmci_use_dma(host) &&
 	    (stat & (STATUS_READ_OP_DONE | STATUS_WRITE_OP_DONE)))
 		writel(STATUS_READ_OP_DONE | STATUS_WRITE_OP_DONE,
 			host->base + MMC_REG_STATUS);
-#endif
 
 	if (sdio_irq) {
 		writel(STATUS_SDIO_INT_ACTIVE, host->base + MMC_REG_STATUS);
@@ -621,14 +615,14 @@ static irqreturn_t mxcmci_irq(int irq, void *devid)
 	if (stat & STATUS_END_CMD_RESP)
 		mxcmci_cmd_done(host, stat);
 
-#ifdef HAS_DMA
 	if (mxcmci_use_dma(host) &&
 		  (stat & (STATUS_DATA_TRANS_DONE | STATUS_WRITE_OP_DONE)))
 		mxcmci_data_done(host, stat);
-#endif
+
 	if (host->default_irq_mask &&
 		  (stat & (STATUS_CARD_INSERTION | STATUS_CARD_REMOVAL)))
 		mmc_detect_change(host->mmc, msecs_to_jiffies(200));
+
 	return IRQ_HANDLED;
 }
 
@@ -642,9 +636,10 @@ static void mxcmci_request(struct mmc_host *mmc, struct mmc_request *req)
 
 	host->req = req;
 	host->cmdat &= ~CMD_DAT_CONT_INIT;
-#ifdef HAS_DMA
-	host->do_dma = 1;
-#endif
+
+	if (host->dma)
+		host->do_dma = 1;
+
 	if (req->data) {
 		error = mxcmci_setup_data(host, req->data);
 		if (error) {
@@ -660,6 +655,7 @@ static void mxcmci_request(struct mmc_host *mmc, struct mmc_request *req)
 	}
 
 	error = mxcmci_start_cmd(host, req->cmd, cmdat);
+
 out:
 	if (error)
 		mxcmci_finish_request(host, req);
@@ -698,22 +694,46 @@ static void mxcmci_set_clk_rate(struct mxcmci_host *host, unsigned int clk_ios)
 			prescaler, divider, clk_in, clk_ios);
 }
 
+static int mxcmci_setup_dma(struct mmc_host *mmc)
+{
+	struct mxcmci_host *host = mmc_priv(mmc);
+	struct dma_slave_config *config = &host->dma_slave_config;
+
+	config->dst_addr = host->res->start + MMC_REG_BUFFER_ACCESS;
+	config->src_addr = host->res->start + MMC_REG_BUFFER_ACCESS;
+	config->dst_addr_width = 4;
+	config->src_addr_width = 4;
+	config->dst_maxburst = host->burstlen;
+	config->src_maxburst = host->burstlen;
+
+	return dmaengine_slave_config(host->dma, config);
+}
+
 static void mxcmci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 {
 	struct mxcmci_host *host = mmc_priv(mmc);
-#ifdef HAS_DMA
-	unsigned int blen;
+	int burstlen, ret;
+
 	/*
 	 * use burstlen of 64 in 4 bit mode (--> reg value  0)
 	 * use burstlen of 16 in 1 bit mode (--> reg value 16)
 	 */
 	if (ios->bus_width == MMC_BUS_WIDTH_4)
-		blen = 0;
+		burstlen = 64;
 	else
-		blen = 16;
+		burstlen = 16;
+
+	if (mxcmci_use_dma(host) && burstlen != host->burstlen) {
+		host->burstlen = burstlen;
+		ret = mxcmci_setup_dma(mmc);
+		if (ret) {
+			dev_err(mmc_dev(host->mmc),
+				"failed to config DMA channel. Falling back to PIO\n");
+			dma_release_channel(host->dma);
+			host->do_dma = 0;
+		}
+	}
 
-	imx_dma_config_burstlen(host->dma, blen);
-#endif
 	if (ios->bus_width == MMC_BUS_WIDTH_4)
 		host->cmdat |= CMD_DAT_CONT_BUS_WIDTH_4;
 	else
@@ -794,6 +814,18 @@ static void mxcmci_init_card(struct mmc_host *host, struct mmc_card *card)
 		host->caps |= MMC_CAP_4_BIT_DATA;
 }
 
+static bool filter(struct dma_chan *chan, void *param)
+{
+	struct mxcmci_host *host = param;
+
+	if (!imx_dma_is_general_purpose(chan))
+		return false;
+
+	chan->private = &host->dma_data;
+
+	return true;
+}
+
 static const struct mmc_host_ops mxcmci_ops = {
 	.request		= mxcmci_request,
 	.set_ios		= mxcmci_set_ios,
@@ -808,6 +840,7 @@ static int mxcmci_probe(struct platform_device *pdev)
 	struct mxcmci_host *host = NULL;
 	struct resource *iores, *r;
 	int ret = 0, irq;
+	dma_cap_mask_t mask;
 
 	printk(KERN_INFO "i.MX SDHC driver\n");
 
@@ -883,29 +916,23 @@ static int mxcmci_probe(struct platform_device *pdev)
 
 	writel(host->default_irq_mask, host->base + MMC_REG_INT_CNTR);
 
-#ifdef HAS_DMA
-	host->dma = imx_dma_request_by_prio(DRIVER_NAME, DMA_PRIO_LOW);
-	if (host->dma < 0) {
-		dev_err(mmc_dev(host->mmc), "imx_dma_request_by_prio failed\n");
-		ret = -EBUSY;
-		goto out_clk_put;
-	}
-
 	r = platform_get_resource(pdev, IORESOURCE_DMA, 0);
-	if (!r) {
-		ret = -EINVAL;
-		goto out_free_dma;
+	if (r) {
+		host->dmareq = r->start;
+		host->dma_data.peripheral_type = IMX_DMATYPE_SDHC;
+		host->dma_data.priority = DMA_PRIO_LOW;
+		host->dma_data.dma_request = host->dmareq;
+		dma_cap_zero(mask);
+		dma_cap_set(DMA_SLAVE, mask);
+		host->dma = dma_request_channel(mask, filter, host);
+		if (host->dma)
+			mmc->max_seg_size = dma_get_max_seg_size(
+					host->dma->device->dev);
 	}
 
-	ret = imx_dma_config_channel(host->dma,
-				     IMX_DMA_MEMSIZE_32 | IMX_DMA_TYPE_FIFO,
-				     IMX_DMA_MEMSIZE_32 | IMX_DMA_TYPE_LINEAR,
-				     r->start, 0);
-	if (ret) {
-		dev_err(mmc_dev(host->mmc), "failed to config DMA channel\n");
-		goto out_free_dma;
-	}
-#endif
+	if (!host->dma)
+		dev_info(mmc_dev(host->mmc), "dma not available. Using PIO\n");
+
 	INIT_WORK(&host->datawork, mxcmci_datawork);
 
 	ret = request_irq(host->irq, mxcmci_irq, 0, DRIVER_NAME, host);
@@ -928,9 +955,8 @@ static int mxcmci_probe(struct platform_device *pdev)
 out_free_irq:
 	free_irq(host->irq, host);
 out_free_dma:
-#ifdef HAS_DMA
-	imx_dma_free(host->dma);
-#endif
+	if (host->dma)
+		dma_release_channel(host->dma);
 out_clk_put:
 	clk_disable(host->clk);
 	clk_put(host->clk);
@@ -960,9 +986,10 @@ static int mxcmci_remove(struct platform_device *pdev)
 
 	free_irq(host->irq, host);
 	iounmap(host->base);
-#ifdef HAS_DMA
-	imx_dma_free(host->dma);
-#endif
+
+	if (host->dma)
+		dma_release_channel(host->dma);
+
 	clk_disable(host->clk);
 	clk_put(host->clk);
 
diff --git a/drivers/mmc/host/mxs-mmc.c b/drivers/mmc/host/mxs-mmc.c
new file mode 100644
index 00000000000..99d39a6a103
--- /dev/null
+++ b/drivers/mmc/host/mxs-mmc.c
@@ -0,0 +1,874 @@
+/*
+ * Portions copyright (C) 2003 Russell King, PXA MMCI Driver
+ * Portions copyright (C) 2004-2005 Pierre Ossman, W83L51xD SD/MMC driver
+ *
+ * Copyright 2008 Embedded Alley Solutions, Inc.
+ * Copyright 2009-2011 Freescale Semiconductor, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/ioport.h>
+#include <linux/platform_device.h>
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/dma-mapping.h>
+#include <linux/dmaengine.h>
+#include <linux/highmem.h>
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/completion.h>
+#include <linux/mmc/host.h>
+#include <linux/mmc/mmc.h>
+#include <linux/mmc/sdio.h>
+#include <linux/gpio.h>
+#include <linux/regulator/consumer.h>
+
+#include <mach/mxs.h>
+#include <mach/common.h>
+#include <mach/dma.h>
+#include <mach/mmc.h>
+
+#define DRIVER_NAME	"mxs-mmc"
+
+/* card detect polling timeout */
+#define MXS_MMC_DETECT_TIMEOUT			(HZ/2)
+
+#define SSP_VERSION_LATEST	4
+#define ssp_is_old()		(host->version < SSP_VERSION_LATEST)
+
+/* SSP registers */
+#define HW_SSP_CTRL0				0x000
+#define  BM_SSP_CTRL0_RUN			(1 << 29)
+#define  BM_SSP_CTRL0_SDIO_IRQ_CHECK		(1 << 28)
+#define  BM_SSP_CTRL0_IGNORE_CRC		(1 << 26)
+#define  BM_SSP_CTRL0_READ			(1 << 25)
+#define  BM_SSP_CTRL0_DATA_XFER			(1 << 24)
+#define  BP_SSP_CTRL0_BUS_WIDTH			(22)
+#define  BM_SSP_CTRL0_BUS_WIDTH			(0x3 << 22)
+#define  BM_SSP_CTRL0_WAIT_FOR_IRQ		(1 << 21)
+#define  BM_SSP_CTRL0_LONG_RESP			(1 << 19)
+#define  BM_SSP_CTRL0_GET_RESP			(1 << 17)
+#define  BM_SSP_CTRL0_ENABLE			(1 << 16)
+#define  BP_SSP_CTRL0_XFER_COUNT		(0)
+#define  BM_SSP_CTRL0_XFER_COUNT		(0xffff)
+#define HW_SSP_CMD0				0x010
+#define  BM_SSP_CMD0_DBL_DATA_RATE_EN		(1 << 25)
+#define  BM_SSP_CMD0_SLOW_CLKING_EN		(1 << 22)
+#define  BM_SSP_CMD0_CONT_CLKING_EN		(1 << 21)
+#define  BM_SSP_CMD0_APPEND_8CYC		(1 << 20)
+#define  BP_SSP_CMD0_BLOCK_SIZE			(16)
+#define  BM_SSP_CMD0_BLOCK_SIZE			(0xf << 16)
+#define  BP_SSP_CMD0_BLOCK_COUNT		(8)
+#define  BM_SSP_CMD0_BLOCK_COUNT		(0xff << 8)
+#define  BP_SSP_CMD0_CMD			(0)
+#define  BM_SSP_CMD0_CMD			(0xff)
+#define HW_SSP_CMD1				0x020
+#define HW_SSP_XFER_SIZE			0x030
+#define HW_SSP_BLOCK_SIZE			0x040
+#define  BP_SSP_BLOCK_SIZE_BLOCK_COUNT		(4)
+#define  BM_SSP_BLOCK_SIZE_BLOCK_COUNT		(0xffffff << 4)
+#define  BP_SSP_BLOCK_SIZE_BLOCK_SIZE		(0)
+#define  BM_SSP_BLOCK_SIZE_BLOCK_SIZE		(0xf)
+#define HW_SSP_TIMING				(ssp_is_old() ? 0x050 : 0x070)
+#define  BP_SSP_TIMING_TIMEOUT			(16)
+#define  BM_SSP_TIMING_TIMEOUT			(0xffff << 16)
+#define  BP_SSP_TIMING_CLOCK_DIVIDE		(8)
+#define  BM_SSP_TIMING_CLOCK_DIVIDE		(0xff << 8)
+#define  BP_SSP_TIMING_CLOCK_RATE		(0)
+#define  BM_SSP_TIMING_CLOCK_RATE		(0xff)
+#define HW_SSP_CTRL1				(ssp_is_old() ? 0x060 : 0x080)
+#define  BM_SSP_CTRL1_SDIO_IRQ			(1 << 31)
+#define  BM_SSP_CTRL1_SDIO_IRQ_EN		(1 << 30)
+#define  BM_SSP_CTRL1_RESP_ERR_IRQ		(1 << 29)
+#define  BM_SSP_CTRL1_RESP_ERR_IRQ_EN		(1 << 28)
+#define  BM_SSP_CTRL1_RESP_TIMEOUT_IRQ		(1 << 27)
+#define  BM_SSP_CTRL1_RESP_TIMEOUT_IRQ_EN	(1 << 26)
+#define  BM_SSP_CTRL1_DATA_TIMEOUT_IRQ		(1 << 25)
+#define  BM_SSP_CTRL1_DATA_TIMEOUT_IRQ_EN	(1 << 24)
+#define  BM_SSP_CTRL1_DATA_CRC_IRQ		(1 << 23)
+#define  BM_SSP_CTRL1_DATA_CRC_IRQ_EN		(1 << 22)
+#define  BM_SSP_CTRL1_FIFO_UNDERRUN_IRQ		(1 << 21)
+#define  BM_SSP_CTRL1_FIFO_UNDERRUN_IRQ_EN	(1 << 20)
+#define  BM_SSP_CTRL1_RECV_TIMEOUT_IRQ		(1 << 17)
+#define  BM_SSP_CTRL1_RECV_TIMEOUT_IRQ_EN	(1 << 16)
+#define  BM_SSP_CTRL1_FIFO_OVERRUN_IRQ		(1 << 15)
+#define  BM_SSP_CTRL1_FIFO_OVERRUN_IRQ_EN	(1 << 14)
+#define  BM_SSP_CTRL1_DMA_ENABLE		(1 << 13)
+#define  BM_SSP_CTRL1_POLARITY			(1 << 9)
+#define  BP_SSP_CTRL1_WORD_LENGTH		(4)
+#define  BM_SSP_CTRL1_WORD_LENGTH		(0xf << 4)
+#define  BP_SSP_CTRL1_SSP_MODE			(0)
+#define  BM_SSP_CTRL1_SSP_MODE			(0xf)
+#define HW_SSP_SDRESP0				(ssp_is_old() ? 0x080 : 0x0a0)
+#define HW_SSP_SDRESP1				(ssp_is_old() ? 0x090 : 0x0b0)
+#define HW_SSP_SDRESP2				(ssp_is_old() ? 0x0a0 : 0x0c0)
+#define HW_SSP_SDRESP3				(ssp_is_old() ? 0x0b0 : 0x0d0)
+#define HW_SSP_STATUS				(ssp_is_old() ? 0x0c0 : 0x100)
+#define  BM_SSP_STATUS_CARD_DETECT		(1 << 28)
+#define  BM_SSP_STATUS_SDIO_IRQ			(1 << 17)
+#define HW_SSP_VERSION				(cpu_is_mx23() ? 0x110 : 0x130)
+#define  BP_SSP_VERSION_MAJOR			(24)
+
+#define BF_SSP(value, field)	(((value) << BP_SSP_##field) & BM_SSP_##field)
+
+#define MXS_MMC_IRQ_BITS	(BM_SSP_CTRL1_SDIO_IRQ		| \
+				 BM_SSP_CTRL1_RESP_ERR_IRQ	| \
+				 BM_SSP_CTRL1_RESP_TIMEOUT_IRQ	| \
+				 BM_SSP_CTRL1_DATA_TIMEOUT_IRQ	| \
+				 BM_SSP_CTRL1_DATA_CRC_IRQ	| \
+				 BM_SSP_CTRL1_FIFO_UNDERRUN_IRQ	| \
+				 BM_SSP_CTRL1_RECV_TIMEOUT_IRQ  | \
+				 BM_SSP_CTRL1_FIFO_OVERRUN_IRQ)
+
+#define SSP_PIO_NUM	3
+
+struct mxs_mmc_host {
+	struct mmc_host			*mmc;
+	struct mmc_request		*mrq;
+	struct mmc_command		*cmd;
+	struct mmc_data			*data;
+
+	void __iomem			*base;
+	int				irq;
+	struct resource			*res;
+	struct resource			*dma_res;
+	struct clk			*clk;
+	unsigned int			clk_rate;
+
+	struct dma_chan         	*dmach;
+	struct mxs_dma_data		dma_data;
+	unsigned int			dma_dir;
+	u32				ssp_pio_words[SSP_PIO_NUM];
+
+	unsigned int			version;
+	unsigned char			bus_width;
+	spinlock_t			lock;
+	int				sdio_irq_en;
+};
+
+static int mxs_mmc_get_ro(struct mmc_host *mmc)
+{
+	struct mxs_mmc_host *host = mmc_priv(mmc);
+	struct mxs_mmc_platform_data *pdata =
+		mmc_dev(host->mmc)->platform_data;
+
+	if (!pdata)
+		return -EFAULT;
+
+	if (!gpio_is_valid(pdata->wp_gpio))
+		return -EINVAL;
+
+	return gpio_get_value(pdata->wp_gpio);
+}
+
+static int mxs_mmc_get_cd(struct mmc_host *mmc)
+{
+	struct mxs_mmc_host *host = mmc_priv(mmc);
+
+	return !(readl(host->base + HW_SSP_STATUS) &
+		 BM_SSP_STATUS_CARD_DETECT);
+}
+
+static void mxs_mmc_reset(struct mxs_mmc_host *host)
+{
+	u32 ctrl0, ctrl1;
+
+	mxs_reset_block(host->base);
+
+	ctrl0 = BM_SSP_CTRL0_IGNORE_CRC;
+	ctrl1 = BF_SSP(0x3, CTRL1_SSP_MODE) |
+		BF_SSP(0x7, CTRL1_WORD_LENGTH) |
+		BM_SSP_CTRL1_DMA_ENABLE |
+		BM_SSP_CTRL1_POLARITY |
+		BM_SSP_CTRL1_RECV_TIMEOUT_IRQ_EN |
+		BM_SSP_CTRL1_DATA_CRC_IRQ_EN |
+		BM_SSP_CTRL1_DATA_TIMEOUT_IRQ_EN |
+		BM_SSP_CTRL1_RESP_TIMEOUT_IRQ_EN |
+		BM_SSP_CTRL1_RESP_ERR_IRQ_EN;
+
+	writel(BF_SSP(0xffff, TIMING_TIMEOUT) |
+	       BF_SSP(2, TIMING_CLOCK_DIVIDE) |
+	       BF_SSP(0, TIMING_CLOCK_RATE),
+	       host->base + HW_SSP_TIMING);
+
+	if (host->sdio_irq_en) {
+		ctrl0 |= BM_SSP_CTRL0_SDIO_IRQ_CHECK;
+		ctrl1 |= BM_SSP_CTRL1_SDIO_IRQ_EN;
+	}
+
+	writel(ctrl0, host->base + HW_SSP_CTRL0);
+	writel(ctrl1, host->base + HW_SSP_CTRL1);
+}
+
+static void mxs_mmc_start_cmd(struct mxs_mmc_host *host,
+			      struct mmc_command *cmd);
+
+static void mxs_mmc_request_done(struct mxs_mmc_host *host)
+{
+	struct mmc_command *cmd = host->cmd;
+	struct mmc_data *data = host->data;
+	struct mmc_request *mrq = host->mrq;
+
+	if (mmc_resp_type(cmd) & MMC_RSP_PRESENT) {
+		if (mmc_resp_type(cmd) & MMC_RSP_136) {
+			cmd->resp[3] = readl(host->base + HW_SSP_SDRESP0);
+			cmd->resp[2] = readl(host->base + HW_SSP_SDRESP1);
+			cmd->resp[1] = readl(host->base + HW_SSP_SDRESP2);
+			cmd->resp[0] = readl(host->base + HW_SSP_SDRESP3);
+		} else {
+			cmd->resp[0] = readl(host->base + HW_SSP_SDRESP0);
+		}
+	}
+
+	if (data) {
+		dma_unmap_sg(mmc_dev(host->mmc), data->sg,
+			     data->sg_len, host->dma_dir);
+		/*
+		 * If there was an error on any block, we mark all
+		 * data blocks as being in error.
+		 */
+		if (!data->error)
+			data->bytes_xfered = data->blocks * data->blksz;
+		else
+			data->bytes_xfered = 0;
+
+		host->data = NULL;
+		if (mrq->stop) {
+			mxs_mmc_start_cmd(host, mrq->stop);
+			return;
+		}
+	}
+
+	host->mrq = NULL;
+	mmc_request_done(host->mmc, mrq);
+}
+
+static void mxs_mmc_dma_irq_callback(void *param)
+{
+	struct mxs_mmc_host *host = param;
+
+	mxs_mmc_request_done(host);
+}
+
+static irqreturn_t mxs_mmc_irq_handler(int irq, void *dev_id)
+{
+	struct mxs_mmc_host *host = dev_id;
+	struct mmc_command *cmd = host->cmd;
+	struct mmc_data *data = host->data;
+	u32 stat;
+
+	spin_lock(&host->lock);
+
+	stat = readl(host->base + HW_SSP_CTRL1);
+	writel(stat & MXS_MMC_IRQ_BITS,
+	       host->base + HW_SSP_CTRL1 + MXS_CLR_ADDR);
+
+	if ((stat & BM_SSP_CTRL1_SDIO_IRQ) && (stat & BM_SSP_CTRL1_SDIO_IRQ_EN))
+		mmc_signal_sdio_irq(host->mmc);
+
+	spin_unlock(&host->lock);
+
+	if (stat & BM_SSP_CTRL1_RESP_TIMEOUT_IRQ)
+		cmd->error = -ETIMEDOUT;
+	else if (stat & BM_SSP_CTRL1_RESP_ERR_IRQ)
+		cmd->error = -EIO;
+
+	if (data) {
+		if (stat & (BM_SSP_CTRL1_DATA_TIMEOUT_IRQ |
+			    BM_SSP_CTRL1_RECV_TIMEOUT_IRQ))
+			data->error = -ETIMEDOUT;
+		else if (stat & BM_SSP_CTRL1_DATA_CRC_IRQ)
+			data->error = -EILSEQ;
+		else if (stat & (BM_SSP_CTRL1_FIFO_UNDERRUN_IRQ |
+				 BM_SSP_CTRL1_FIFO_OVERRUN_IRQ))
+			data->error = -EIO;
+	}
+
+	return IRQ_HANDLED;
+}
+
+static struct dma_async_tx_descriptor *mxs_mmc_prep_dma(
+	struct mxs_mmc_host *host, unsigned int append)
+{
+	struct dma_async_tx_descriptor *desc;
+	struct mmc_data *data = host->data;
+	struct scatterlist * sgl;
+	unsigned int sg_len;
+
+	if (data) {
+		/* data */
+		dma_map_sg(mmc_dev(host->mmc), data->sg,
+			   data->sg_len, host->dma_dir);
+		sgl = data->sg;
+		sg_len = data->sg_len;
+	} else {
+		/* pio */
+		sgl = (struct scatterlist *) host->ssp_pio_words;
+		sg_len = SSP_PIO_NUM;
+	}
+
+	desc = host->dmach->device->device_prep_slave_sg(host->dmach,
+				sgl, sg_len, host->dma_dir, append);
+	if (desc) {
+		desc->callback = mxs_mmc_dma_irq_callback;
+		desc->callback_param = host;
+	} else {
+		if (data)
+			dma_unmap_sg(mmc_dev(host->mmc), data->sg,
+				     data->sg_len, host->dma_dir);
+	}
+
+	return desc;
+}
+
+static void mxs_mmc_bc(struct mxs_mmc_host *host)
+{
+	struct mmc_command *cmd = host->cmd;
+	struct dma_async_tx_descriptor *desc;
+	u32 ctrl0, cmd0, cmd1;
+
+	ctrl0 = BM_SSP_CTRL0_ENABLE | BM_SSP_CTRL0_IGNORE_CRC;
+	cmd0 = BF_SSP(cmd->opcode, CMD0_CMD) | BM_SSP_CMD0_APPEND_8CYC;
+	cmd1 = cmd->arg;
+
+	if (host->sdio_irq_en) {
+		ctrl0 |= BM_SSP_CTRL0_SDIO_IRQ_CHECK;
+		cmd0 |= BM_SSP_CMD0_CONT_CLKING_EN | BM_SSP_CMD0_SLOW_CLKING_EN;
+	}
+
+	host->ssp_pio_words[0] = ctrl0;
+	host->ssp_pio_words[1] = cmd0;
+	host->ssp_pio_words[2] = cmd1;
+	host->dma_dir = DMA_NONE;
+	desc = mxs_mmc_prep_dma(host, 0);
+	if (!desc)
+		goto out;
+
+	dmaengine_submit(desc);
+	return;
+
+out:
+	dev_warn(mmc_dev(host->mmc),
+		 "%s: failed to prep dma\n", __func__);
+}
+
+static void mxs_mmc_ac(struct mxs_mmc_host *host)
+{
+	struct mmc_command *cmd = host->cmd;
+	struct dma_async_tx_descriptor *desc;
+	u32 ignore_crc, get_resp, long_resp;
+	u32 ctrl0, cmd0, cmd1;
+
+	ignore_crc = (mmc_resp_type(cmd) & MMC_RSP_CRC) ?
+			0 : BM_SSP_CTRL0_IGNORE_CRC;
+	get_resp = (mmc_resp_type(cmd) & MMC_RSP_PRESENT) ?
+			BM_SSP_CTRL0_GET_RESP : 0;
+	long_resp = (mmc_resp_type(cmd) & MMC_RSP_136) ?
+			BM_SSP_CTRL0_LONG_RESP : 0;
+
+	ctrl0 = BM_SSP_CTRL0_ENABLE | ignore_crc | get_resp | long_resp;
+	cmd0 = BF_SSP(cmd->opcode, CMD0_CMD);
+	cmd1 = cmd->arg;
+
+	if (host->sdio_irq_en) {
+		ctrl0 |= BM_SSP_CTRL0_SDIO_IRQ_CHECK;
+		cmd0 |= BM_SSP_CMD0_CONT_CLKING_EN | BM_SSP_CMD0_SLOW_CLKING_EN;
+	}
+
+	host->ssp_pio_words[0] = ctrl0;
+	host->ssp_pio_words[1] = cmd0;
+	host->ssp_pio_words[2] = cmd1;
+	host->dma_dir = DMA_NONE;
+	desc = mxs_mmc_prep_dma(host, 0);
+	if (!desc)
+		goto out;
+
+	dmaengine_submit(desc);
+	return;
+
+out:
+	dev_warn(mmc_dev(host->mmc),
+		 "%s: failed to prep dma\n", __func__);
+}
+
+static unsigned short mxs_ns_to_ssp_ticks(unsigned clock_rate, unsigned ns)
+{
+	const unsigned int ssp_timeout_mul = 4096;
+	/*
+	 * Calculate ticks in ms since ns are large numbers
+	 * and might overflow
+	 */
+	const unsigned int clock_per_ms = clock_rate / 1000;
+	const unsigned int ms = ns / 1000;
+	const unsigned int ticks = ms * clock_per_ms;
+	const unsigned int ssp_ticks = ticks / ssp_timeout_mul;
+
+	WARN_ON(ssp_ticks == 0);
+	return ssp_ticks;
+}
+
+static void mxs_mmc_adtc(struct mxs_mmc_host *host)
+{
+	struct mmc_command *cmd = host->cmd;
+	struct mmc_data *data = cmd->data;
+	struct dma_async_tx_descriptor *desc;
+	struct scatterlist *sgl = data->sg, *sg;
+	unsigned int sg_len = data->sg_len;
+	int i;
+
+	unsigned short dma_data_dir, timeout;
+	unsigned int data_size = 0, log2_blksz;
+	unsigned int blocks = data->blocks;
+
+	u32 ignore_crc, get_resp, long_resp, read;
+	u32 ctrl0, cmd0, cmd1, val;
+
+	ignore_crc = (mmc_resp_type(cmd) & MMC_RSP_CRC) ?
+			0 : BM_SSP_CTRL0_IGNORE_CRC;
+	get_resp = (mmc_resp_type(cmd) & MMC_RSP_PRESENT) ?
+			BM_SSP_CTRL0_GET_RESP : 0;
+	long_resp = (mmc_resp_type(cmd) & MMC_RSP_136) ?
+			BM_SSP_CTRL0_LONG_RESP : 0;
+
+	if (data->flags & MMC_DATA_WRITE) {
+		dma_data_dir = DMA_TO_DEVICE;
+		read = 0;
+	} else {
+		dma_data_dir = DMA_FROM_DEVICE;
+		read = BM_SSP_CTRL0_READ;
+	}
+
+	ctrl0 = BF_SSP(host->bus_width, CTRL0_BUS_WIDTH) |
+		ignore_crc | get_resp | long_resp |
+		BM_SSP_CTRL0_DATA_XFER | read |
+		BM_SSP_CTRL0_WAIT_FOR_IRQ |
+		BM_SSP_CTRL0_ENABLE;
+
+	cmd0 = BF_SSP(cmd->opcode, CMD0_CMD);
+
+	/* get logarithm to base 2 of block size for setting register */
+	log2_blksz = ilog2(data->blksz);
+
+	/*
+	 * take special care of the case that data size from data->sg
+	 * is not equal to blocks x blksz
+	 */
+	for_each_sg(sgl, sg, sg_len, i)
+		data_size += sg->length;
+
+	if (data_size != data->blocks * data->blksz)
+		blocks = 1;
+
+	/* xfer count, block size and count need to be set differently */
+	if (ssp_is_old()) {
+		ctrl0 |= BF_SSP(data_size, CTRL0_XFER_COUNT);
+		cmd0 |= BF_SSP(log2_blksz, CMD0_BLOCK_SIZE) |
+			BF_SSP(blocks - 1, CMD0_BLOCK_COUNT);
+	} else {
+		writel(data_size, host->base + HW_SSP_XFER_SIZE);
+		writel(BF_SSP(log2_blksz, BLOCK_SIZE_BLOCK_SIZE) |
+		       BF_SSP(blocks - 1, BLOCK_SIZE_BLOCK_COUNT),
+		       host->base + HW_SSP_BLOCK_SIZE);
+	}
+
+	if ((cmd->opcode == MMC_STOP_TRANSMISSION) ||
+	    (cmd->opcode == SD_IO_RW_EXTENDED))
+		cmd0 |= BM_SSP_CMD0_APPEND_8CYC;
+
+	cmd1 = cmd->arg;
+
+	if (host->sdio_irq_en) {
+		ctrl0 |= BM_SSP_CTRL0_SDIO_IRQ_CHECK;
+		cmd0 |= BM_SSP_CMD0_CONT_CLKING_EN | BM_SSP_CMD0_SLOW_CLKING_EN;
+	}
+
+	/* set the timeout count */
+	timeout = mxs_ns_to_ssp_ticks(host->clk_rate, data->timeout_ns);
+	val = readl(host->base + HW_SSP_TIMING);
+	val &= ~(BM_SSP_TIMING_TIMEOUT);
+	val |= BF_SSP(timeout, TIMING_TIMEOUT);
+	writel(val, host->base + HW_SSP_TIMING);
+
+	/* pio */
+	host->ssp_pio_words[0] = ctrl0;
+	host->ssp_pio_words[1] = cmd0;
+	host->ssp_pio_words[2] = cmd1;
+	host->dma_dir = DMA_NONE;
+	desc = mxs_mmc_prep_dma(host, 0);
+	if (!desc)
+		goto out;
+
+	/* append data sg */
+	WARN_ON(host->data != NULL);
+	host->data = data;
+	host->dma_dir = dma_data_dir;
+	desc = mxs_mmc_prep_dma(host, 1);
+	if (!desc)
+		goto out;
+
+	dmaengine_submit(desc);
+	return;
+out:
+	dev_warn(mmc_dev(host->mmc),
+		 "%s: failed to prep dma\n", __func__);
+}
+
+static void mxs_mmc_start_cmd(struct mxs_mmc_host *host,
+			      struct mmc_command *cmd)
+{
+	host->cmd = cmd;
+
+	switch (mmc_cmd_type(cmd)) {
+	case MMC_CMD_BC:
+		mxs_mmc_bc(host);
+		break;
+	case MMC_CMD_BCR:
+		mxs_mmc_ac(host);
+		break;
+	case MMC_CMD_AC:
+		mxs_mmc_ac(host);
+		break;
+	case MMC_CMD_ADTC:
+		mxs_mmc_adtc(host);
+		break;
+	default:
+		dev_warn(mmc_dev(host->mmc),
+			 "%s: unknown MMC command\n", __func__);
+		break;
+	}
+}
+
+static void mxs_mmc_request(struct mmc_host *mmc, struct mmc_request *mrq)
+{
+	struct mxs_mmc_host *host = mmc_priv(mmc);
+
+	WARN_ON(host->mrq != NULL);
+	host->mrq = mrq;
+	mxs_mmc_start_cmd(host, mrq->cmd);
+}
+
+static void mxs_mmc_set_clk_rate(struct mxs_mmc_host *host, unsigned int rate)
+{
+	unsigned int ssp_rate, bit_rate;
+	u32 div1, div2;
+	u32 val;
+
+	ssp_rate = clk_get_rate(host->clk);
+
+	for (div1 = 2; div1 < 254; div1 += 2) {
+		div2 = ssp_rate / rate / div1;
+		if (div2 < 0x100)
+			break;
+	}
+
+	if (div1 >= 254) {
+		dev_err(mmc_dev(host->mmc),
+			"%s: cannot set clock to %d\n", __func__, rate);
+		return;
+	}
+
+	if (div2 == 0)
+		bit_rate = ssp_rate / div1;
+	else
+		bit_rate = ssp_rate / div1 / div2;
+
+	val = readl(host->base + HW_SSP_TIMING);
+	val &= ~(BM_SSP_TIMING_CLOCK_DIVIDE | BM_SSP_TIMING_CLOCK_RATE);
+	val |= BF_SSP(div1, TIMING_CLOCK_DIVIDE);
+	val |= BF_SSP(div2 - 1, TIMING_CLOCK_RATE);
+	writel(val, host->base + HW_SSP_TIMING);
+
+	host->clk_rate = bit_rate;
+
+	dev_dbg(mmc_dev(host->mmc),
+		"%s: div1 %d, div2 %d, ssp %d, bit %d, rate %d\n",
+		__func__, div1, div2, ssp_rate, bit_rate, rate);
+}
+
+static void mxs_mmc_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
+{
+	struct mxs_mmc_host *host = mmc_priv(mmc);
+
+	if (ios->bus_width == MMC_BUS_WIDTH_8)
+		host->bus_width = 2;
+	else if (ios->bus_width == MMC_BUS_WIDTH_4)
+		host->bus_width = 1;
+	else
+		host->bus_width = 0;
+
+	if (ios->clock)
+		mxs_mmc_set_clk_rate(host, ios->clock);
+}
+
+static void mxs_mmc_enable_sdio_irq(struct mmc_host *mmc, int enable)
+{
+	struct mxs_mmc_host *host = mmc_priv(mmc);
+	unsigned long flags;
+
+	spin_lock_irqsave(&host->lock, flags);
+
+	host->sdio_irq_en = enable;
+
+	if (enable) {
+		writel(BM_SSP_CTRL0_SDIO_IRQ_CHECK,
+		       host->base + HW_SSP_CTRL0 + MXS_SET_ADDR);
+		writel(BM_SSP_CTRL1_SDIO_IRQ_EN,
+		       host->base + HW_SSP_CTRL1 + MXS_SET_ADDR);
+
+		if (readl(host->base + HW_SSP_STATUS) & BM_SSP_STATUS_SDIO_IRQ)
+			mmc_signal_sdio_irq(host->mmc);
+
+	} else {
+		writel(BM_SSP_CTRL0_SDIO_IRQ_CHECK,
+		       host->base + HW_SSP_CTRL0 + MXS_CLR_ADDR);
+		writel(BM_SSP_CTRL1_SDIO_IRQ_EN,
+		       host->base + HW_SSP_CTRL1 + MXS_CLR_ADDR);
+	}
+
+	spin_unlock_irqrestore(&host->lock, flags);
+}
+
+static const struct mmc_host_ops mxs_mmc_ops = {
+	.request = mxs_mmc_request,
+	.get_ro = mxs_mmc_get_ro,
+	.get_cd = mxs_mmc_get_cd,
+	.set_ios = mxs_mmc_set_ios,
+	.enable_sdio_irq = mxs_mmc_enable_sdio_irq,
+};
+
+static bool mxs_mmc_dma_filter(struct dma_chan *chan, void *param)
+{
+	struct mxs_mmc_host *host = param;
+
+	if (!mxs_dma_is_apbh(chan))
+		return false;
+
+	if (chan->chan_id != host->dma_res->start)
+		return false;
+
+	chan->private = &host->dma_data;
+
+	return true;
+}
+
+static int mxs_mmc_probe(struct platform_device *pdev)
+{
+	struct mxs_mmc_host *host;
+	struct mmc_host *mmc;
+	struct resource *iores, *dmares, *r;
+	struct mxs_mmc_platform_data *pdata;
+	int ret = 0, irq_err, irq_dma;
+	dma_cap_mask_t mask;
+
+	iores = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	dmares = platform_get_resource(pdev, IORESOURCE_DMA, 0);
+	irq_err = platform_get_irq(pdev, 0);
+	irq_dma = platform_get_irq(pdev, 1);
+	if (!iores || !dmares || irq_err < 0 || irq_dma < 0)
+		return -EINVAL;
+
+	r = request_mem_region(iores->start, resource_size(iores), pdev->name);
+	if (!r)
+		return -EBUSY;
+
+	mmc = mmc_alloc_host(sizeof(struct mxs_mmc_host), &pdev->dev);
+	if (!mmc) {
+		ret = -ENOMEM;
+		goto out_release_mem;
+	}
+
+	host = mmc_priv(mmc);
+	host->base = ioremap(r->start, resource_size(r));
+	if (!host->base) {
+		ret = -ENOMEM;
+		goto out_mmc_free;
+	}
+
+	/* only major verion does matter */
+	host->version = readl(host->base + HW_SSP_VERSION) >>
+			BP_SSP_VERSION_MAJOR;
+
+	host->mmc = mmc;
+	host->res = r;
+	host->dma_res = dmares;
+	host->irq = irq_err;
+	host->sdio_irq_en = 0;
+
+	host->clk = clk_get(&pdev->dev, NULL);
+	if (IS_ERR(host->clk)) {
+		ret = PTR_ERR(host->clk);
+		goto out_iounmap;
+	}
+	clk_enable(host->clk);
+
+	mxs_mmc_reset(host);
+
+	dma_cap_zero(mask);
+	dma_cap_set(DMA_SLAVE, mask);
+	host->dma_data.chan_irq = irq_dma;
+	host->dmach = dma_request_channel(mask, mxs_mmc_dma_filter, host);
+	if (!host->dmach) {
+		dev_err(mmc_dev(host->mmc),
+			"%s: failed to request dma\n", __func__);
+		goto out_clk_put;
+	}
+
+	/* set mmc core parameters */
+	mmc->ops = &mxs_mmc_ops;
+	mmc->caps = MMC_CAP_SD_HIGHSPEED | MMC_CAP_MMC_HIGHSPEED |
+		    MMC_CAP_SDIO_IRQ | MMC_CAP_NEEDS_POLL;
+
+	pdata =	mmc_dev(host->mmc)->platform_data;
+	if (pdata) {
+		if (pdata->flags & SLOTF_8_BIT_CAPABLE)
+			mmc->caps |= MMC_CAP_4_BIT_DATA | MMC_CAP_8_BIT_DATA;
+		if (pdata->flags & SLOTF_4_BIT_CAPABLE)
+			mmc->caps |= MMC_CAP_4_BIT_DATA;
+	}
+
+	mmc->f_min = 400000;
+	mmc->f_max = 288000000;
+	mmc->ocr_avail = MMC_VDD_32_33 | MMC_VDD_33_34;
+
+	mmc->max_segs = 52;
+	mmc->max_blk_size = 1 << 0xf;
+	mmc->max_blk_count = (ssp_is_old()) ? 0xff : 0xffffff;
+	mmc->max_req_size = (ssp_is_old()) ? 0xffff : 0xffffffff;
+	mmc->max_seg_size = dma_get_max_seg_size(host->dmach->device->dev);
+
+	platform_set_drvdata(pdev, mmc);
+
+	ret = request_irq(host->irq, mxs_mmc_irq_handler, 0, DRIVER_NAME, host);
+	if (ret)
+		goto out_free_dma;
+
+	spin_lock_init(&host->lock);
+
+	ret = mmc_add_host(mmc);
+	if (ret)
+		goto out_free_irq;
+
+	dev_info(mmc_dev(host->mmc), "initialized\n");
+
+	return 0;
+
+out_free_irq:
+	free_irq(host->irq, host);
+out_free_dma:
+	if (host->dmach)
+		dma_release_channel(host->dmach);
+out_clk_put:
+	clk_disable(host->clk);
+	clk_put(host->clk);
+out_iounmap:
+	iounmap(host->base);
+out_mmc_free:
+	mmc_free_host(mmc);
+out_release_mem:
+	release_mem_region(iores->start, resource_size(iores));
+	return ret;
+}
+
+static int mxs_mmc_remove(struct platform_device *pdev)
+{
+	struct mmc_host *mmc = platform_get_drvdata(pdev);
+	struct mxs_mmc_host *host = mmc_priv(mmc);
+	struct resource *res = host->res;
+
+	mmc_remove_host(mmc);
+
+	free_irq(host->irq, host);
+
+	platform_set_drvdata(pdev, NULL);
+
+	if (host->dmach)
+		dma_release_channel(host->dmach);
+
+	clk_disable(host->clk);
+	clk_put(host->clk);
+
+	iounmap(host->base);
+
+	mmc_free_host(mmc);
+
+	release_mem_region(res->start, resource_size(res));
+
+	return 0;
+}
+
+#ifdef CONFIG_PM
+static int mxs_mmc_suspend(struct device *dev)
+{
+	struct mmc_host *mmc = dev_get_drvdata(dev);
+	struct mxs_mmc_host *host = mmc_priv(mmc);
+	int ret = 0;
+
+	ret = mmc_suspend_host(mmc);
+
+	clk_disable(host->clk);
+
+	return ret;
+}
+
+static int mxs_mmc_resume(struct device *dev)
+{
+	struct mmc_host *mmc = dev_get_drvdata(dev);
+	struct mxs_mmc_host *host = mmc_priv(mmc);
+	int ret = 0;
+
+	clk_enable(host->clk);
+
+	ret = mmc_resume_host(mmc);
+
+	return ret;
+}
+
+static const struct dev_pm_ops mxs_mmc_pm_ops = {
+	.suspend	= mxs_mmc_suspend,
+	.resume		= mxs_mmc_resume,
+};
+#endif
+
+static struct platform_driver mxs_mmc_driver = {
+	.probe		= mxs_mmc_probe,
+	.remove		= mxs_mmc_remove,
+	.driver		= {
+		.name	= DRIVER_NAME,
+		.owner	= THIS_MODULE,
+#ifdef CONFIG_PM
+		.pm	= &mxs_mmc_pm_ops,
+#endif
+	},
+};
+
+static int __init mxs_mmc_init(void)
+{
+	return platform_driver_register(&mxs_mmc_driver);
+}
+
+static void __exit mxs_mmc_exit(void)
+{
+	platform_driver_unregister(&mxs_mmc_driver);
+}
+
+module_init(mxs_mmc_init);
+module_exit(mxs_mmc_exit);
+
+MODULE_DESCRIPTION("FREESCALE MXS MMC peripheral");
+MODULE_AUTHOR("Freescale Semiconductor");
+MODULE_LICENSE("GPL");
diff --git a/drivers/mmc/host/of_mmc_spi.c b/drivers/mmc/host/of_mmc_spi.c
index 5530def54e5..e2aecb7f1d5 100644
--- a/drivers/mmc/host/of_mmc_spi.c
+++ b/drivers/mmc/host/of_mmc_spi.c
@@ -15,9 +15,11 @@
 #include <linux/module.h>
 #include <linux/device.h>
 #include <linux/slab.h>
+#include <linux/irq.h>
 #include <linux/gpio.h>
 #include <linux/of.h>
 #include <linux/of_gpio.h>
+#include <linux/of_irq.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/mmc_spi.h>
 #include <linux/mmc/core.h>
diff --git a/drivers/mmc/host/omap.c b/drivers/mmc/host/omap.c
index 379d2ffe4c8..2e032f0e8cf 100644
--- a/drivers/mmc/host/omap.c
+++ b/drivers/mmc/host/omap.c
@@ -1417,7 +1417,7 @@ static int __init mmc_omap_probe(struct platform_device *pdev)
 	if (res == NULL || irq < 0)
 		return -ENXIO;
 
-	res = request_mem_region(res->start, res->end - res->start + 1,
+	res = request_mem_region(res->start, resource_size(res),
 				 pdev->name);
 	if (res == NULL)
 		return -EBUSY;
@@ -1457,7 +1457,7 @@ static int __init mmc_omap_probe(struct platform_device *pdev)
 
 	host->irq = irq;
 	host->phys_base = host->mem_res->start;
-	host->virt_base = ioremap(res->start, res->end - res->start + 1);
+	host->virt_base = ioremap(res->start, resource_size(res));
 	if (!host->virt_base)
 		goto err_ioremap;
 
@@ -1514,7 +1514,7 @@ err_free_mmc_host:
 err_ioremap:
 	kfree(host);
 err_free_mem_region:
-	release_mem_region(res->start, res->end - res->start + 1);
+	release_mem_region(res->start, resource_size(res));
 	return ret;
 }
 
diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c
index 158c0ee53b2..259ece047af 100644
--- a/drivers/mmc/host/omap_hsmmc.c
+++ b/drivers/mmc/host/omap_hsmmc.c
@@ -2047,8 +2047,7 @@ static int __init omap_hsmmc_probe(struct platform_device *pdev)
 
 	res->start += pdata->reg_offset;
 	res->end += pdata->reg_offset;
-	res = request_mem_region(res->start, res->end - res->start + 1,
-							pdev->name);
+	res = request_mem_region(res->start, resource_size(res), pdev->name);
 	if (res == NULL)
 		return -EBUSY;
 
@@ -2287,7 +2286,7 @@ err1:
 err_alloc:
 	omap_hsmmc_gpio_free(pdata);
 err:
-	release_mem_region(res->start, res->end - res->start + 1);
+	release_mem_region(res->start, resource_size(res));
 	return ret;
 }
 
@@ -2324,7 +2323,7 @@ static int omap_hsmmc_remove(struct platform_device *pdev)
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	if (res)
-		release_mem_region(res->start, res->end - res->start + 1);
+		release_mem_region(res->start, resource_size(res));
 	platform_set_drvdata(pdev, NULL);
 
 	return 0;
diff --git a/drivers/mmc/host/s3cmci.c b/drivers/mmc/host/s3cmci.c
index 1ccd4b256ce..a04f87d7ee3 100644
--- a/drivers/mmc/host/s3cmci.c
+++ b/drivers/mmc/host/s3cmci.c
@@ -874,7 +874,7 @@ static void finalize_request(struct s3cmci_host *host)
 	if (!mrq->data)
 		goto request_done;
 
-	/* Calulate the amout of bytes transfer if there was no error */
+	/* Calculate the amout of bytes transfer if there was no error */
 	if (mrq->data->error == 0) {
 		mrq->data->bytes_xfered =
 			(mrq->data->blocks * mrq->data->blksz);
@@ -882,7 +882,7 @@ static void finalize_request(struct s3cmci_host *host)
 		mrq->data->bytes_xfered = 0;
 	}
 
-	/* If we had an error while transfering data we flush the
+	/* If we had an error while transferring data we flush the
 	 * DMA channel and the fifo to clear out any garbage. */
 	if (mrq->data->error != 0) {
 		if (s3cmci_host_usedma(host))
@@ -980,7 +980,7 @@ static int s3cmci_setup_data(struct s3cmci_host *host, struct mmc_data *data)
 
 	if ((data->blksz & 3) != 0) {
 		/* We cannot deal with unaligned blocks with more than
-		 * one block being transfered. */
+		 * one block being transferred. */
 
 		if (data->blocks > 1) {
 			pr_warning("%s: can't do non-word sized block transfers (blksz %d)\n", __func__, data->blksz);
diff --git a/drivers/mmc/host/sdhci-esdhc-imx.c b/drivers/mmc/host/sdhci-esdhc-imx.c
index 9b82910b9db..a19967d0bfc 100644
--- a/drivers/mmc/host/sdhci-esdhc-imx.c
+++ b/drivers/mmc/host/sdhci-esdhc-imx.c
@@ -15,13 +15,41 @@
 #include <linux/delay.h>
 #include <linux/err.h>
 #include <linux/clk.h>
+#include <linux/gpio.h>
+#include <linux/slab.h>
 #include <linux/mmc/host.h>
 #include <linux/mmc/sdhci-pltfm.h>
+#include <linux/mmc/mmc.h>
+#include <linux/mmc/sdio.h>
 #include <mach/hardware.h>
+#include <mach/esdhc.h>
 #include "sdhci.h"
 #include "sdhci-pltfm.h"
 #include "sdhci-esdhc.h"
 
+/* VENDOR SPEC register */
+#define SDHCI_VENDOR_SPEC		0xC0
+#define  SDHCI_VENDOR_SPEC_SDIO_QUIRK	0x00000002
+
+#define ESDHC_FLAG_GPIO_FOR_CD_WP	(1 << 0)
+/*
+ * The CMDTYPE of the CMD register (offset 0xE) should be set to
+ * "11" when the STOP CMD12 is issued on imx53 to abort one
+ * open ended multi-blk IO. Otherwise the TC INT wouldn't
+ * be generated.
+ * In exact block transfer, the controller doesn't complete the
+ * operations automatically as required at the end of the
+ * transfer and remains on hold if the abort command is not sent.
+ * As a result, the TC flag is not asserted and SW  received timeout
+ * exeception. Bit1 of Vendor Spec registor is used to fix it.
+ */
+#define ESDHC_FLAG_MULTIBLK_NO_INT	(1 << 1)
+
+struct pltfm_imx_data {
+	int flags;
+	u32 scratchpad;
+};
+
 static inline void esdhc_clrset_le(struct sdhci_host *host, u32 mask, u32 val, int reg)
 {
 	void __iomem *base = host->ioaddr + (reg & ~0x3);
@@ -30,6 +58,56 @@ static inline void esdhc_clrset_le(struct sdhci_host *host, u32 mask, u32 val, i
 	writel(((readl(base) & ~(mask << shift)) | (val << shift)), base);
 }
 
+static u32 esdhc_readl_le(struct sdhci_host *host, int reg)
+{
+	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+	struct pltfm_imx_data *imx_data = pltfm_host->priv;
+
+	/* fake CARD_PRESENT flag on mx25/35 */
+	u32 val = readl(host->ioaddr + reg);
+
+	if (unlikely((reg == SDHCI_PRESENT_STATE)
+			&& (imx_data->flags & ESDHC_FLAG_GPIO_FOR_CD_WP))) {
+		struct esdhc_platform_data *boarddata =
+				host->mmc->parent->platform_data;
+
+		if (boarddata && gpio_is_valid(boarddata->cd_gpio)
+				&& gpio_get_value(boarddata->cd_gpio))
+			/* no card, if a valid gpio says so... */
+			val &= SDHCI_CARD_PRESENT;
+		else
+			/* ... in all other cases assume card is present */
+			val |= SDHCI_CARD_PRESENT;
+	}
+
+	return val;
+}
+
+static void esdhc_writel_le(struct sdhci_host *host, u32 val, int reg)
+{
+	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+	struct pltfm_imx_data *imx_data = pltfm_host->priv;
+
+	if (unlikely((reg == SDHCI_INT_ENABLE || reg == SDHCI_SIGNAL_ENABLE)
+			&& (imx_data->flags & ESDHC_FLAG_GPIO_FOR_CD_WP)))
+		/*
+		 * these interrupts won't work with a custom card_detect gpio
+		 * (only applied to mx25/35)
+		 */
+		val &= ~(SDHCI_INT_CARD_REMOVE | SDHCI_INT_CARD_INSERT);
+
+	if (unlikely((imx_data->flags & ESDHC_FLAG_MULTIBLK_NO_INT)
+				&& (reg == SDHCI_INT_STATUS)
+				&& (val & SDHCI_INT_DATA_END))) {
+			u32 v;
+			v = readl(host->ioaddr + SDHCI_VENDOR_SPEC);
+			v &= ~SDHCI_VENDOR_SPEC_SDIO_QUIRK;
+			writel(v, host->ioaddr + SDHCI_VENDOR_SPEC);
+	}
+
+	writel(val, host->ioaddr + reg);
+}
+
 static u16 esdhc_readw_le(struct sdhci_host *host, int reg)
 {
 	if (unlikely(reg == SDHCI_HOST_VERSION))
@@ -41,6 +119,7 @@ static u16 esdhc_readw_le(struct sdhci_host *host, int reg)
 static void esdhc_writew_le(struct sdhci_host *host, u16 val, int reg)
 {
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+	struct pltfm_imx_data *imx_data = pltfm_host->priv;
 
 	switch (reg) {
 	case SDHCI_TRANSFER_MODE:
@@ -48,10 +127,22 @@ static void esdhc_writew_le(struct sdhci_host *host, u16 val, int reg)
 		 * Postpone this write, we must do it together with a
 		 * command write that is down below.
 		 */
-		pltfm_host->scratchpad = val;
+		if ((imx_data->flags & ESDHC_FLAG_MULTIBLK_NO_INT)
+				&& (host->cmd->opcode == SD_IO_RW_EXTENDED)
+				&& (host->cmd->data->blocks > 1)
+				&& (host->cmd->data->flags & MMC_DATA_READ)) {
+			u32 v;
+			v = readl(host->ioaddr + SDHCI_VENDOR_SPEC);
+			v |= SDHCI_VENDOR_SPEC_SDIO_QUIRK;
+			writel(v, host->ioaddr + SDHCI_VENDOR_SPEC);
+		}
+		imx_data->scratchpad = val;
 		return;
 	case SDHCI_COMMAND:
-		writel(val << 16 | pltfm_host->scratchpad,
+		if ((host->cmd->opcode == MMC_STOP_TRANSMISSION)
+			&& (imx_data->flags & ESDHC_FLAG_MULTIBLK_NO_INT))
+			val |= SDHCI_CMD_ABORTCMD;
+		writel(val << 16 | imx_data->scratchpad,
 			host->ioaddr + SDHCI_TRANSFER_MODE);
 		return;
 	case SDHCI_BLOCK_SIZE:
@@ -100,10 +191,42 @@ static unsigned int esdhc_pltfm_get_min_clock(struct sdhci_host *host)
 	return clk_get_rate(pltfm_host->clk) / 256 / 16;
 }
 
+static unsigned int esdhc_pltfm_get_ro(struct sdhci_host *host)
+{
+	struct esdhc_platform_data *boarddata = host->mmc->parent->platform_data;
+
+	if (boarddata && gpio_is_valid(boarddata->wp_gpio))
+		return gpio_get_value(boarddata->wp_gpio);
+	else
+		return -ENOSYS;
+}
+
+static struct sdhci_ops sdhci_esdhc_ops = {
+	.read_l = esdhc_readl_le,
+	.read_w = esdhc_readw_le,
+	.write_l = esdhc_writel_le,
+	.write_w = esdhc_writew_le,
+	.write_b = esdhc_writeb_le,
+	.set_clock = esdhc_set_clock,
+	.get_max_clock = esdhc_pltfm_get_max_clock,
+	.get_min_clock = esdhc_pltfm_get_min_clock,
+};
+
+static irqreturn_t cd_irq(int irq, void *data)
+{
+	struct sdhci_host *sdhost = (struct sdhci_host *)data;
+
+	tasklet_schedule(&sdhost->card_tasklet);
+	return IRQ_HANDLED;
+};
+
 static int esdhc_pltfm_init(struct sdhci_host *host, struct sdhci_pltfm_data *pdata)
 {
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+	struct esdhc_platform_data *boarddata = host->mmc->parent->platform_data;
 	struct clk *clk;
+	int err;
+	struct pltfm_imx_data *imx_data;
 
 	clk = clk_get(mmc_dev(host->mmc), NULL);
 	if (IS_ERR(clk)) {
@@ -113,35 +236,94 @@ static int esdhc_pltfm_init(struct sdhci_host *host, struct sdhci_pltfm_data *pd
 	clk_enable(clk);
 	pltfm_host->clk = clk;
 
-	if (cpu_is_mx35() || cpu_is_mx51())
+	imx_data = kzalloc(sizeof(struct pltfm_imx_data), GFP_KERNEL);
+	if (!imx_data) {
+		clk_disable(pltfm_host->clk);
+		clk_put(pltfm_host->clk);
+		return -ENOMEM;
+	}
+	pltfm_host->priv = imx_data;
+
+	if (!cpu_is_mx25())
 		host->quirks |= SDHCI_QUIRK_BROKEN_TIMEOUT_VAL;
 
-	/* Fix errata ENGcm07207 which is present on i.MX25 and i.MX35 */
-	if (cpu_is_mx25() || cpu_is_mx35())
+	if (cpu_is_mx25() || cpu_is_mx35()) {
+		/* Fix errata ENGcm07207 present on i.MX25 and i.MX35 */
 		host->quirks |= SDHCI_QUIRK_NO_MULTIBLOCK;
+		/* write_protect can't be routed to controller, use gpio */
+		sdhci_esdhc_ops.get_ro = esdhc_pltfm_get_ro;
+	}
+
+	if (!(cpu_is_mx25() || cpu_is_mx35() || cpu_is_mx51()))
+		imx_data->flags |= ESDHC_FLAG_MULTIBLK_NO_INT;
+
+	if (boarddata) {
+		err = gpio_request_one(boarddata->wp_gpio, GPIOF_IN, "ESDHC_WP");
+		if (err) {
+			dev_warn(mmc_dev(host->mmc),
+				"no write-protect pin available!\n");
+			boarddata->wp_gpio = err;
+		}
 
+		err = gpio_request_one(boarddata->cd_gpio, GPIOF_IN, "ESDHC_CD");
+		if (err) {
+			dev_warn(mmc_dev(host->mmc),
+				"no card-detect pin available!\n");
+			goto no_card_detect_pin;
+		}
+
+		/* i.MX5x has issues to be researched */
+		if (!cpu_is_mx25() && !cpu_is_mx35())
+			goto not_supported;
+
+		err = request_irq(gpio_to_irq(boarddata->cd_gpio), cd_irq,
+				 IRQF_TRIGGER_FALLING | IRQF_TRIGGER_RISING,
+				 mmc_hostname(host->mmc), host);
+		if (err) {
+			dev_warn(mmc_dev(host->mmc), "request irq error\n");
+			goto no_card_detect_irq;
+		}
+
+		imx_data->flags |= ESDHC_FLAG_GPIO_FOR_CD_WP;
+		/* Now we have a working card_detect again */
+		host->quirks &= ~SDHCI_QUIRK_BROKEN_CARD_DETECTION;
+	}
+
+	return 0;
+
+ no_card_detect_irq:
+	gpio_free(boarddata->cd_gpio);
+ no_card_detect_pin:
+	boarddata->cd_gpio = err;
+ not_supported:
+	kfree(imx_data);
 	return 0;
 }
 
 static void esdhc_pltfm_exit(struct sdhci_host *host)
 {
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+	struct esdhc_platform_data *boarddata = host->mmc->parent->platform_data;
+	struct pltfm_imx_data *imx_data = pltfm_host->priv;
+
+	if (boarddata && gpio_is_valid(boarddata->wp_gpio))
+		gpio_free(boarddata->wp_gpio);
+
+	if (boarddata && gpio_is_valid(boarddata->cd_gpio)) {
+		gpio_free(boarddata->cd_gpio);
+
+		if (!(host->quirks & SDHCI_QUIRK_BROKEN_CARD_DETECTION))
+			free_irq(gpio_to_irq(boarddata->cd_gpio), host);
+	}
 
 	clk_disable(pltfm_host->clk);
 	clk_put(pltfm_host->clk);
+	kfree(imx_data);
 }
 
-static struct sdhci_ops sdhci_esdhc_ops = {
-	.read_w = esdhc_readw_le,
-	.write_w = esdhc_writew_le,
-	.write_b = esdhc_writeb_le,
-	.set_clock = esdhc_set_clock,
-	.get_max_clock = esdhc_pltfm_get_max_clock,
-	.get_min_clock = esdhc_pltfm_get_min_clock,
-};
-
 struct sdhci_pltfm_data sdhci_esdhc_imx_pdata = {
-	.quirks = ESDHC_DEFAULT_QUIRKS | SDHCI_QUIRK_BROKEN_ADMA,
+	.quirks = ESDHC_DEFAULT_QUIRKS | SDHCI_QUIRK_BROKEN_ADMA
+			| SDHCI_QUIRK_BROKEN_CARD_DETECTION,
 	/* ADMA has issues. Might be fixable */
 	.ops = &sdhci_esdhc_ops,
 	.init = esdhc_pltfm_init,
diff --git a/drivers/mmc/host/sdhci-esdhc.h b/drivers/mmc/host/sdhci-esdhc.h
index afaf1bc4913..c3b08f11194 100644
--- a/drivers/mmc/host/sdhci-esdhc.h
+++ b/drivers/mmc/host/sdhci-esdhc.h
@@ -19,13 +19,11 @@
  */
 
 #define ESDHC_DEFAULT_QUIRKS	(SDHCI_QUIRK_FORCE_BLK_SZ_2048 | \
-				SDHCI_QUIRK_BROKEN_CARD_DETECTION | \
 				SDHCI_QUIRK_NO_BUSY_IRQ | \
 				SDHCI_QUIRK_NONSTANDARD_CLOCK | \
 				SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK | \
 				SDHCI_QUIRK_PIO_NEEDS_DELAY | \
-				SDHCI_QUIRK_RESTORE_IRQS_AFTER_RESET | \
-				SDHCI_QUIRK_NO_CARD_NO_RESET)
+				SDHCI_QUIRK_RESTORE_IRQS_AFTER_RESET)
 
 #define ESDHC_SYSTEM_CONTROL	0x2c
 #define ESDHC_CLOCK_MASK	0x0000fff0
diff --git a/drivers/mmc/host/sdhci-of-esdhc.c b/drivers/mmc/host/sdhci-of-esdhc.c
index fcd0e1fcba4..ba40d6d035c 100644
--- a/drivers/mmc/host/sdhci-of-esdhc.c
+++ b/drivers/mmc/host/sdhci-of-esdhc.c
@@ -73,7 +73,9 @@ static unsigned int esdhc_of_get_min_clock(struct sdhci_host *host)
 }
 
 struct sdhci_of_data sdhci_esdhc = {
-	.quirks = ESDHC_DEFAULT_QUIRKS,
+	/* card detection could be handled via GPIO */
+	.quirks = ESDHC_DEFAULT_QUIRKS | SDHCI_QUIRK_BROKEN_CARD_DETECTION
+		| SDHCI_QUIRK_NO_CARD_NO_RESET,
 	.ops = {
 		.read_l = sdhci_be32bs_readl,
 		.read_w = esdhc_readw,
diff --git a/drivers/mmc/host/sdhci-pci.c b/drivers/mmc/host/sdhci-pci.c
index 0dc905b20ee..a136be70634 100644
--- a/drivers/mmc/host/sdhci-pci.c
+++ b/drivers/mmc/host/sdhci-pci.c
@@ -547,6 +547,14 @@ static const struct pci_device_id pci_ids[] __devinitdata = {
 	},
 
 	{
+		.vendor         = PCI_VENDOR_ID_RICOH,
+		.device         = 0xe823,
+		.subvendor      = PCI_ANY_ID,
+		.subdevice      = PCI_ANY_ID,
+		.driver_data    = (kernel_ulong_t)&sdhci_ricoh_mmc,
+	},
+
+	{
 		.vendor		= PCI_VENDOR_ID_ENE,
 		.device		= PCI_DEVICE_ID_ENE_CB712_SD,
 		.subvendor	= PCI_ANY_ID,
@@ -900,9 +908,6 @@ static struct sdhci_pci_slot * __devinit sdhci_pci_probe_slot(
 {
 	struct sdhci_pci_slot *slot;
 	struct sdhci_host *host;
-
-	resource_size_t addr;
-
 	int ret;
 
 	if (!(pci_resource_flags(pdev, bar) & IORESOURCE_MEM)) {
@@ -949,7 +954,6 @@ static struct sdhci_pci_slot * __devinit sdhci_pci_probe_slot(
 		goto free;
 	}
 
-	addr = pci_resource_start(pdev, bar);
 	host->ioaddr = pci_ioremap_bar(pdev, bar);
 	if (!host->ioaddr) {
 		dev_err(&pdev->dev, "failed to remap registers\n");
@@ -1012,16 +1016,14 @@ static int __devinit sdhci_pci_probe(struct pci_dev *pdev,
 	struct sdhci_pci_chip *chip;
 	struct sdhci_pci_slot *slot;
 
-	u8 slots, rev, first_bar;
+	u8 slots, first_bar;
 	int ret, i;
 
 	BUG_ON(pdev == NULL);
 	BUG_ON(ent == NULL);
 
-	pci_read_config_byte(pdev, PCI_CLASS_REVISION, &rev);
-
 	dev_info(&pdev->dev, "SDHCI controller found [%04x:%04x] (rev %x)\n",
-		 (int)pdev->vendor, (int)pdev->device, (int)rev);
+		 (int)pdev->vendor, (int)pdev->device, (int)pdev->revision);
 
 	ret = pci_read_config_byte(pdev, PCI_SLOT_INFO, &slots);
 	if (ret)
diff --git a/drivers/mmc/host/sdhci-pltfm.h b/drivers/mmc/host/sdhci-pltfm.h
index ea2e44d9be5..2b37016ad0a 100644
--- a/drivers/mmc/host/sdhci-pltfm.h
+++ b/drivers/mmc/host/sdhci-pltfm.h
@@ -17,7 +17,7 @@
 
 struct sdhci_pltfm_host {
 	struct clk *clk;
-	u32 scratchpad; /* to handle quirks across io-accessor calls */
+	void *priv; /* to handle quirks across io-accessor calls */
 };
 
 extern struct sdhci_pltfm_data sdhci_cns3xxx_pdata;
diff --git a/drivers/mmc/host/sdhci-s3c.c b/drivers/mmc/host/sdhci-s3c.c
index 5309ab95aad..69e3ee321eb 100644
--- a/drivers/mmc/host/sdhci-s3c.c
+++ b/drivers/mmc/host/sdhci-s3c.c
@@ -499,6 +499,9 @@ static int __devinit sdhci_s3c_probe(struct platform_device *pdev)
 	 * SDHCI block, or a missing configuration that needs to be set. */
 	host->quirks |= SDHCI_QUIRK_NO_BUSY_IRQ;
 
+	/* This host supports the Auto CMD12 */
+	host->quirks |= SDHCI_QUIRK_MULTIBLOCK_READ_ACMD12;
+
 	if (pdata->cd_type == S3C_SDHCI_CD_NONE ||
 	    pdata->cd_type == S3C_SDHCI_CD_PERMANENT)
 		host->quirks |= SDHCI_QUIRK_BROKEN_CARD_DETECTION;
diff --git a/drivers/mmc/host/sdhci-spear.c b/drivers/mmc/host/sdhci-spear.c
index d70c54c7b70..60a4c97d3d1 100644
--- a/drivers/mmc/host/sdhci-spear.c
+++ b/drivers/mmc/host/sdhci-spear.c
@@ -50,7 +50,7 @@ static irqreturn_t sdhci_gpio_irq(int irq, void *dev_id)
 	/* val == 1 -> card removed, val == 0 -> card inserted */
 	/* if card removed - set irq for low level, else vice versa */
 	gpio_irq_type = val ? IRQF_TRIGGER_LOW : IRQF_TRIGGER_HIGH;
-	set_irq_type(irq, gpio_irq_type);
+	irq_set_irq_type(irq, gpio_irq_type);
 
 	if (sdhci->data->card_power_gpio >= 0) {
 		if (!sdhci->data->power_always_enb) {
diff --git a/drivers/mmc/host/sdhci-tegra.c b/drivers/mmc/host/sdhci-tegra.c
index 4823ee94a63..f7e1f964395 100644
--- a/drivers/mmc/host/sdhci-tegra.c
+++ b/drivers/mmc/host/sdhci-tegra.c
@@ -169,7 +169,7 @@ static int tegra_sdhci_pltfm_init(struct sdhci_host *host,
 		if (rc) {
 			dev_err(mmc_dev(host->mmc),
 				"failed to allocate wp gpio\n");
-			goto out_cd;
+			goto out_irq;
 		}
 		tegra_gpio_enable(plat->wp_gpio);
 		gpio_direction_input(plat->wp_gpio);
@@ -195,6 +195,9 @@ out_wp:
 		gpio_free(plat->wp_gpio);
 	}
 
+out_irq:
+	if (gpio_is_valid(plat->cd_gpio))
+		free_irq(gpio_to_irq(plat->cd_gpio), host);
 out_cd:
 	if (gpio_is_valid(plat->cd_gpio)) {
 		tegra_gpio_disable(plat->cd_gpio);
@@ -225,6 +228,7 @@ static void tegra_sdhci_pltfm_exit(struct sdhci_host *host)
 	}
 
 	if (gpio_is_valid(plat->cd_gpio)) {
+		free_irq(gpio_to_irq(plat->cd_gpio), host);
 		tegra_gpio_disable(plat->cd_gpio);
 		gpio_free(plat->cd_gpio);
 	}
diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
index 6e0969e4065..25e8bde600d 100644
--- a/drivers/mmc/host/sdhci.h
+++ b/drivers/mmc/host/sdhci.h
@@ -45,6 +45,7 @@
 #define  SDHCI_CMD_CRC		0x08
 #define  SDHCI_CMD_INDEX	0x10
 #define  SDHCI_CMD_DATA		0x20
+#define  SDHCI_CMD_ABORTCMD	0xC0
 
 #define  SDHCI_CMD_RESP_NONE	0x00
 #define  SDHCI_CMD_RESP_LONG	0x01
diff --git a/drivers/mmc/host/sh_mmcif.c b/drivers/mmc/host/sh_mmcif.c
index 12884c27017..af97015a2fc 100644
--- a/drivers/mmc/host/sh_mmcif.c
+++ b/drivers/mmc/host/sh_mmcif.c
@@ -169,7 +169,7 @@ struct sh_mmcif_host {
 	struct dma_chan		*chan_rx;
 	struct dma_chan		*chan_tx;
 	struct completion	dma_complete;
-	unsigned int            dma_sglen;
+	bool			dma_active;
 };
 
 static inline void sh_mmcif_bitset(struct sh_mmcif_host *host,
@@ -194,10 +194,12 @@ static void mmcif_dma_complete(void *arg)
 		return;
 
 	if (host->data->flags & MMC_DATA_READ)
-		dma_unmap_sg(&host->pd->dev, host->data->sg, host->dma_sglen,
+		dma_unmap_sg(host->chan_rx->device->dev,
+			     host->data->sg, host->data->sg_len,
 			     DMA_FROM_DEVICE);
 	else
-		dma_unmap_sg(&host->pd->dev, host->data->sg, host->dma_sglen,
+		dma_unmap_sg(host->chan_tx->device->dev,
+			     host->data->sg, host->data->sg_len,
 			     DMA_TO_DEVICE);
 
 	complete(&host->dma_complete);
@@ -211,9 +213,10 @@ static void sh_mmcif_start_dma_rx(struct sh_mmcif_host *host)
 	dma_cookie_t cookie = -EINVAL;
 	int ret;
 
-	ret = dma_map_sg(&host->pd->dev, sg, host->data->sg_len, DMA_FROM_DEVICE);
+	ret = dma_map_sg(chan->device->dev, sg, host->data->sg_len,
+			 DMA_FROM_DEVICE);
 	if (ret > 0) {
-		host->dma_sglen = ret;
+		host->dma_active = true;
 		desc = chan->device->device_prep_slave_sg(chan, sg, ret,
 			DMA_FROM_DEVICE, DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
 	}
@@ -221,14 +224,9 @@ static void sh_mmcif_start_dma_rx(struct sh_mmcif_host *host)
 	if (desc) {
 		desc->callback = mmcif_dma_complete;
 		desc->callback_param = host;
-		cookie = desc->tx_submit(desc);
-		if (cookie < 0) {
-			desc = NULL;
-			ret = cookie;
-		} else {
-			sh_mmcif_bitset(host, MMCIF_CE_BUF_ACC, BUF_ACC_DMAREN);
-			chan->device->device_issue_pending(chan);
-		}
+		cookie = dmaengine_submit(desc);
+		sh_mmcif_bitset(host, MMCIF_CE_BUF_ACC, BUF_ACC_DMAREN);
+		dma_async_issue_pending(chan);
 	}
 	dev_dbg(&host->pd->dev, "%s(): mapped %d -> %d, cookie %d\n",
 		__func__, host->data->sg_len, ret, cookie);
@@ -238,7 +236,7 @@ static void sh_mmcif_start_dma_rx(struct sh_mmcif_host *host)
 		if (ret >= 0)
 			ret = -EIO;
 		host->chan_rx = NULL;
-		host->dma_sglen = 0;
+		host->dma_active = false;
 		dma_release_channel(chan);
 		/* Free the Tx channel too */
 		chan = host->chan_tx;
@@ -263,9 +261,10 @@ static void sh_mmcif_start_dma_tx(struct sh_mmcif_host *host)
 	dma_cookie_t cookie = -EINVAL;
 	int ret;
 
-	ret = dma_map_sg(&host->pd->dev, sg, host->data->sg_len, DMA_TO_DEVICE);
+	ret = dma_map_sg(chan->device->dev, sg, host->data->sg_len,
+			 DMA_TO_DEVICE);
 	if (ret > 0) {
-		host->dma_sglen = ret;
+		host->dma_active = true;
 		desc = chan->device->device_prep_slave_sg(chan, sg, ret,
 			DMA_TO_DEVICE, DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
 	}
@@ -273,14 +272,9 @@ static void sh_mmcif_start_dma_tx(struct sh_mmcif_host *host)
 	if (desc) {
 		desc->callback = mmcif_dma_complete;
 		desc->callback_param = host;
-		cookie = desc->tx_submit(desc);
-		if (cookie < 0) {
-			desc = NULL;
-			ret = cookie;
-		} else {
-			sh_mmcif_bitset(host, MMCIF_CE_BUF_ACC, BUF_ACC_DMAWEN);
-			chan->device->device_issue_pending(chan);
-		}
+		cookie = dmaengine_submit(desc);
+		sh_mmcif_bitset(host, MMCIF_CE_BUF_ACC, BUF_ACC_DMAWEN);
+		dma_async_issue_pending(chan);
 	}
 	dev_dbg(&host->pd->dev, "%s(): mapped %d -> %d, cookie %d\n",
 		__func__, host->data->sg_len, ret, cookie);
@@ -290,7 +284,7 @@ static void sh_mmcif_start_dma_tx(struct sh_mmcif_host *host)
 		if (ret >= 0)
 			ret = -EIO;
 		host->chan_tx = NULL;
-		host->dma_sglen = 0;
+		host->dma_active = false;
 		dma_release_channel(chan);
 		/* Free the Rx channel too */
 		chan = host->chan_rx;
@@ -317,7 +311,7 @@ static bool sh_mmcif_filter(struct dma_chan *chan, void *arg)
 static void sh_mmcif_request_dma(struct sh_mmcif_host *host,
 				 struct sh_mmcif_plat_data *pdata)
 {
-	host->dma_sglen = 0;
+	host->dma_active = false;
 
 	/* We can only either use DMA for both Tx and Rx or not use it at all */
 	if (pdata->dma) {
@@ -364,7 +358,7 @@ static void sh_mmcif_release_dma(struct sh_mmcif_host *host)
 		dma_release_channel(chan);
 	}
 
-	host->dma_sglen = 0;
+	host->dma_active = false;
 }
 
 static void sh_mmcif_clock_control(struct sh_mmcif_host *host, unsigned int clk)
@@ -753,7 +747,7 @@ static void sh_mmcif_start_cmd(struct sh_mmcif_host *host,
 	}
 	sh_mmcif_get_response(host, cmd);
 	if (host->data) {
-		if (!host->dma_sglen) {
+		if (!host->dma_active) {
 			ret = sh_mmcif_data_trans(host, mrq, cmd->opcode);
 		} else {
 			long time =
@@ -765,7 +759,7 @@ static void sh_mmcif_start_cmd(struct sh_mmcif_host *host,
 				ret = time;
 			sh_mmcif_bitclr(host, MMCIF_CE_BUF_ACC,
 					BUF_ACC_DMAREN | BUF_ACC_DMAWEN);
-			host->dma_sglen = 0;
+			host->dma_active = false;
 		}
 		if (ret < 0)
 			mrq->data->bytes_xfered = 0;
@@ -850,15 +844,15 @@ static void sh_mmcif_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 	struct sh_mmcif_host *host = mmc_priv(mmc);
 	struct sh_mmcif_plat_data *p = host->pd->dev.platform_data;
 
-	if (ios->power_mode == MMC_POWER_OFF) {
+	if (ios->power_mode == MMC_POWER_UP) {
+		if (p->set_pwr)
+			p->set_pwr(host->pd, ios->power_mode);
+	} else if (ios->power_mode == MMC_POWER_OFF || !ios->clock) {
 		/* clock stop */
 		sh_mmcif_clock_control(host, 0);
-		if (p->down_pwr)
+		if (ios->power_mode == MMC_POWER_OFF && p->down_pwr)
 			p->down_pwr(host->pd);
 		return;
-	} else if (ios->power_mode == MMC_POWER_UP) {
-		if (p->set_pwr)
-			p->set_pwr(host->pd, ios->power_mode);
 	}
 
 	if (ios->clock)
diff --git a/drivers/mmc/host/sh_mobile_sdhi.c b/drivers/mmc/host/sh_mobile_sdhi.c
new file mode 100644
index 00000000000..cc701236d16
--- /dev/null
+++ b/drivers/mmc/host/sh_mobile_sdhi.c
@@ -0,0 +1,171 @@
+/*
+ * SuperH Mobile SDHI
+ *
+ * Copyright (C) 2009 Magnus Damm
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Based on "Compaq ASIC3 support":
+ *
+ * Copyright 2001 Compaq Computer Corporation.
+ * Copyright 2004-2005 Phil Blundell
+ * Copyright 2007-2008 OpenedHand Ltd.
+ *
+ * Authors: Phil Blundell <pb@handhelds.org>,
+ *	    Samuel Ortiz <sameo@openedhand.com>
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/clk.h>
+#include <linux/slab.h>
+#include <linux/platform_device.h>
+#include <linux/mmc/host.h>
+#include <linux/mmc/sh_mobile_sdhi.h>
+#include <linux/mfd/tmio.h>
+#include <linux/sh_dma.h>
+
+#include "tmio_mmc.h"
+
+struct sh_mobile_sdhi {
+	struct clk *clk;
+	struct tmio_mmc_data mmc_data;
+	struct sh_dmae_slave param_tx;
+	struct sh_dmae_slave param_rx;
+	struct tmio_mmc_dma dma_priv;
+};
+
+static void sh_mobile_sdhi_set_pwr(struct platform_device *pdev, int state)
+{
+	struct sh_mobile_sdhi_info *p = pdev->dev.platform_data;
+
+	if (p && p->set_pwr)
+		p->set_pwr(pdev, state);
+}
+
+static int sh_mobile_sdhi_get_cd(struct platform_device *pdev)
+{
+	struct sh_mobile_sdhi_info *p = pdev->dev.platform_data;
+
+	if (p && p->get_cd)
+		return p->get_cd(pdev);
+	else
+		return -ENOSYS;
+}
+
+static int __devinit sh_mobile_sdhi_probe(struct platform_device *pdev)
+{
+	struct sh_mobile_sdhi *priv;
+	struct tmio_mmc_data *mmc_data;
+	struct sh_mobile_sdhi_info *p = pdev->dev.platform_data;
+	struct tmio_mmc_host *host;
+	char clk_name[8];
+	int ret;
+
+	priv = kzalloc(sizeof(struct sh_mobile_sdhi), GFP_KERNEL);
+	if (priv == NULL) {
+		dev_err(&pdev->dev, "kzalloc failed\n");
+		return -ENOMEM;
+	}
+
+	mmc_data = &priv->mmc_data;
+
+	snprintf(clk_name, sizeof(clk_name), "sdhi%d", pdev->id);
+	priv->clk = clk_get(&pdev->dev, clk_name);
+	if (IS_ERR(priv->clk)) {
+		dev_err(&pdev->dev, "cannot get clock \"%s\"\n", clk_name);
+		ret = PTR_ERR(priv->clk);
+		goto eclkget;
+	}
+
+	clk_enable(priv->clk);
+
+	mmc_data->hclk = clk_get_rate(priv->clk);
+	mmc_data->set_pwr = sh_mobile_sdhi_set_pwr;
+	mmc_data->get_cd = sh_mobile_sdhi_get_cd;
+	mmc_data->capabilities = MMC_CAP_MMC_HIGHSPEED;
+	if (p) {
+		mmc_data->flags = p->tmio_flags;
+		mmc_data->ocr_mask = p->tmio_ocr_mask;
+		mmc_data->capabilities |= p->tmio_caps;
+
+		if (p->dma_slave_tx >= 0 && p->dma_slave_rx >= 0) {
+			priv->param_tx.slave_id = p->dma_slave_tx;
+			priv->param_rx.slave_id = p->dma_slave_rx;
+			priv->dma_priv.chan_priv_tx = &priv->param_tx;
+			priv->dma_priv.chan_priv_rx = &priv->param_rx;
+			priv->dma_priv.alignment_shift = 1; /* 2-byte alignment */
+			mmc_data->dma = &priv->dma_priv;
+		}
+	}
+
+	/*
+	 * All SDHI blocks support 2-byte and larger block sizes in 4-bit
+	 * bus width mode.
+	 */
+	mmc_data->flags |= TMIO_MMC_BLKSZ_2BYTES;
+
+	/*
+	 * All SDHI blocks support SDIO IRQ signalling.
+	 */
+	mmc_data->flags |= TMIO_MMC_SDIO_IRQ;
+
+	ret = tmio_mmc_host_probe(&host, pdev, mmc_data);
+	if (ret < 0)
+		goto eprobe;
+
+	pr_info("%s at 0x%08lx irq %d\n", mmc_hostname(host->mmc),
+		(unsigned long)host->ctl, host->irq);
+
+	return ret;
+
+eprobe:
+	clk_disable(priv->clk);
+	clk_put(priv->clk);
+eclkget:
+	kfree(priv);
+	return ret;
+}
+
+static int sh_mobile_sdhi_remove(struct platform_device *pdev)
+{
+	struct mmc_host *mmc = platform_get_drvdata(pdev);
+	struct tmio_mmc_host *host = mmc_priv(mmc);
+	struct sh_mobile_sdhi *priv = container_of(host->pdata, struct sh_mobile_sdhi, mmc_data);
+
+	tmio_mmc_host_remove(host);
+	clk_disable(priv->clk);
+	clk_put(priv->clk);
+	kfree(priv);
+
+	return 0;
+}
+
+static struct platform_driver sh_mobile_sdhi_driver = {
+	.driver		= {
+		.name	= "sh_mobile_sdhi",
+		.owner	= THIS_MODULE,
+	},
+	.probe		= sh_mobile_sdhi_probe,
+	.remove		= __devexit_p(sh_mobile_sdhi_remove),
+};
+
+static int __init sh_mobile_sdhi_init(void)
+{
+	return platform_driver_register(&sh_mobile_sdhi_driver);
+}
+
+static void __exit sh_mobile_sdhi_exit(void)
+{
+	platform_driver_unregister(&sh_mobile_sdhi_driver);
+}
+
+module_init(sh_mobile_sdhi_init);
+module_exit(sh_mobile_sdhi_exit);
+
+MODULE_DESCRIPTION("SuperH Mobile SDHI driver");
+MODULE_AUTHOR("Magnus Damm");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS("platform:sh_mobile_sdhi");
diff --git a/drivers/mmc/host/tmio_mmc.c b/drivers/mmc/host/tmio_mmc.c
index e3c6ef20839..79c568461d5 100644
--- a/drivers/mmc/host/tmio_mmc.c
+++ b/drivers/mmc/host/tmio_mmc.c
@@ -1,8 +1,8 @@
 /*
- *  linux/drivers/mmc/tmio_mmc.c
+ * linux/drivers/mmc/host/tmio_mmc.c
  *
- *  Copyright (C) 2004 Ian Molton
- *  Copyright (C) 2007 Ian Molton
+ * Copyright (C) 2007 Ian Molton
+ * Copyright (C) 2004 Ian Molton
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -11,1203 +11,22 @@
  * Driver for the MMC / SD / SDIO cell found in:
  *
  * TC6393XB TC6391XB TC6387XB T7L66XB ASIC3
- *
- * This driver draws mainly on scattered spec sheets, Reverse engineering
- * of the toshiba e800  SD driver and some parts of the 2.4 ASIC3 driver (4 bit
- * support). (Further 4 bit support from a later datasheet).
- *
- * TODO:
- *   Investigate using a workqueue for PIO transfers
- *   Eliminate FIXMEs
- *   SDIO support
- *   Better Power management
- *   Handle MMC errors better
- *   double buffer support
- *
  */
 
-#include <linux/delay.h>
 #include <linux/device.h>
-#include <linux/dmaengine.h>
-#include <linux/highmem.h>
-#include <linux/interrupt.h>
-#include <linux/io.h>
-#include <linux/irq.h>
 #include <linux/mfd/core.h>
 #include <linux/mfd/tmio.h>
 #include <linux/mmc/host.h>
 #include <linux/module.h>
 #include <linux/pagemap.h>
 #include <linux/scatterlist.h>
-#include <linux/workqueue.h>
-#include <linux/spinlock.h>
-
-#define CTL_SD_CMD 0x00
-#define CTL_ARG_REG 0x04
-#define CTL_STOP_INTERNAL_ACTION 0x08
-#define CTL_XFER_BLK_COUNT 0xa
-#define CTL_RESPONSE 0x0c
-#define CTL_STATUS 0x1c
-#define CTL_IRQ_MASK 0x20
-#define CTL_SD_CARD_CLK_CTL 0x24
-#define CTL_SD_XFER_LEN 0x26
-#define CTL_SD_MEM_CARD_OPT 0x28
-#define CTL_SD_ERROR_DETAIL_STATUS 0x2c
-#define CTL_SD_DATA_PORT 0x30
-#define CTL_TRANSACTION_CTL 0x34
-#define CTL_SDIO_STATUS 0x36
-#define CTL_SDIO_IRQ_MASK 0x38
-#define CTL_RESET_SD 0xe0
-#define CTL_SDIO_REGS 0x100
-#define CTL_CLK_AND_WAIT_CTL 0x138
-#define CTL_RESET_SDIO 0x1e0
-
-/* Definitions for values the CTRL_STATUS register can take. */
-#define TMIO_STAT_CMDRESPEND    0x00000001
-#define TMIO_STAT_DATAEND       0x00000004
-#define TMIO_STAT_CARD_REMOVE   0x00000008
-#define TMIO_STAT_CARD_INSERT   0x00000010
-#define TMIO_STAT_SIGSTATE      0x00000020
-#define TMIO_STAT_WRPROTECT     0x00000080
-#define TMIO_STAT_CARD_REMOVE_A 0x00000100
-#define TMIO_STAT_CARD_INSERT_A 0x00000200
-#define TMIO_STAT_SIGSTATE_A    0x00000400
-#define TMIO_STAT_CMD_IDX_ERR   0x00010000
-#define TMIO_STAT_CRCFAIL       0x00020000
-#define TMIO_STAT_STOPBIT_ERR   0x00040000
-#define TMIO_STAT_DATATIMEOUT   0x00080000
-#define TMIO_STAT_RXOVERFLOW    0x00100000
-#define TMIO_STAT_TXUNDERRUN    0x00200000
-#define TMIO_STAT_CMDTIMEOUT    0x00400000
-#define TMIO_STAT_RXRDY         0x01000000
-#define TMIO_STAT_TXRQ          0x02000000
-#define TMIO_STAT_ILL_FUNC      0x20000000
-#define TMIO_STAT_CMD_BUSY      0x40000000
-#define TMIO_STAT_ILL_ACCESS    0x80000000
-
-/* Definitions for values the CTRL_SDIO_STATUS register can take. */
-#define TMIO_SDIO_STAT_IOIRQ	0x0001
-#define TMIO_SDIO_STAT_EXPUB52	0x4000
-#define TMIO_SDIO_STAT_EXWT	0x8000
-#define TMIO_SDIO_MASK_ALL	0xc007
-
-/* Define some IRQ masks */
-/* This is the mask used at reset by the chip */
-#define TMIO_MASK_ALL           0x837f031d
-#define TMIO_MASK_READOP  (TMIO_STAT_RXRDY | TMIO_STAT_DATAEND)
-#define TMIO_MASK_WRITEOP (TMIO_STAT_TXRQ | TMIO_STAT_DATAEND)
-#define TMIO_MASK_CMD     (TMIO_STAT_CMDRESPEND | TMIO_STAT_CMDTIMEOUT | \
-		TMIO_STAT_CARD_REMOVE | TMIO_STAT_CARD_INSERT)
-#define TMIO_MASK_IRQ     (TMIO_MASK_READOP | TMIO_MASK_WRITEOP | TMIO_MASK_CMD)
-
-#define enable_mmc_irqs(host, i) \
-	do { \
-		u32 mask;\
-		mask  = sd_ctrl_read32((host), CTL_IRQ_MASK); \
-		mask &= ~((i) & TMIO_MASK_IRQ); \
-		sd_ctrl_write32((host), CTL_IRQ_MASK, mask); \
-	} while (0)
-
-#define disable_mmc_irqs(host, i) \
-	do { \
-		u32 mask;\
-		mask  = sd_ctrl_read32((host), CTL_IRQ_MASK); \
-		mask |= ((i) & TMIO_MASK_IRQ); \
-		sd_ctrl_write32((host), CTL_IRQ_MASK, mask); \
-	} while (0)
-
-#define ack_mmc_irqs(host, i) \
-	do { \
-		sd_ctrl_write32((host), CTL_STATUS, ~(i)); \
-	} while (0)
-
-/* This is arbitrary, just noone needed any higher alignment yet */
-#define MAX_ALIGN 4
-
-struct tmio_mmc_host {
-	void __iomem *ctl;
-	unsigned long bus_shift;
-	struct mmc_command      *cmd;
-	struct mmc_request      *mrq;
-	struct mmc_data         *data;
-	struct mmc_host         *mmc;
-	int                     irq;
-	unsigned int		sdio_irq_enabled;
-
-	/* Callbacks for clock / power control */
-	void (*set_pwr)(struct platform_device *host, int state);
-	void (*set_clk_div)(struct platform_device *host, int state);
-
-	/* pio related stuff */
-	struct scatterlist      *sg_ptr;
-	struct scatterlist      *sg_orig;
-	unsigned int            sg_len;
-	unsigned int            sg_off;
-
-	struct platform_device *pdev;
-
-	/* DMA support */
-	struct dma_chan		*chan_rx;
-	struct dma_chan		*chan_tx;
-	struct tasklet_struct	dma_complete;
-	struct tasklet_struct	dma_issue;
-#ifdef CONFIG_TMIO_MMC_DMA
-	unsigned int            dma_sglen;
-	u8			bounce_buf[PAGE_CACHE_SIZE] __attribute__((aligned(MAX_ALIGN)));
-	struct scatterlist	bounce_sg;
-#endif
-
-	/* Track lost interrupts */
-	struct delayed_work	delayed_reset_work;
-	spinlock_t		lock;
-	unsigned long		last_req_ts;
-};
-
-static void tmio_check_bounce_buffer(struct tmio_mmc_host *host);
-
-static u16 sd_ctrl_read16(struct tmio_mmc_host *host, int addr)
-{
-	return readw(host->ctl + (addr << host->bus_shift));
-}
-
-static void sd_ctrl_read16_rep(struct tmio_mmc_host *host, int addr,
-		u16 *buf, int count)
-{
-	readsw(host->ctl + (addr << host->bus_shift), buf, count);
-}
-
-static u32 sd_ctrl_read32(struct tmio_mmc_host *host, int addr)
-{
-	return readw(host->ctl + (addr << host->bus_shift)) |
-	       readw(host->ctl + ((addr + 2) << host->bus_shift)) << 16;
-}
-
-static void sd_ctrl_write16(struct tmio_mmc_host *host, int addr, u16 val)
-{
-	writew(val, host->ctl + (addr << host->bus_shift));
-}
-
-static void sd_ctrl_write16_rep(struct tmio_mmc_host *host, int addr,
-		u16 *buf, int count)
-{
-	writesw(host->ctl + (addr << host->bus_shift), buf, count);
-}
-
-static void sd_ctrl_write32(struct tmio_mmc_host *host, int addr, u32 val)
-{
-	writew(val, host->ctl + (addr << host->bus_shift));
-	writew(val >> 16, host->ctl + ((addr + 2) << host->bus_shift));
-}
-
-static void tmio_mmc_init_sg(struct tmio_mmc_host *host, struct mmc_data *data)
-{
-	host->sg_len = data->sg_len;
-	host->sg_ptr = data->sg;
-	host->sg_orig = data->sg;
-	host->sg_off = 0;
-}
-
-static int tmio_mmc_next_sg(struct tmio_mmc_host *host)
-{
-	host->sg_ptr = sg_next(host->sg_ptr);
-	host->sg_off = 0;
-	return --host->sg_len;
-}
-
-static char *tmio_mmc_kmap_atomic(struct scatterlist *sg, unsigned long *flags)
-{
-	local_irq_save(*flags);
-	return kmap_atomic(sg_page(sg), KM_BIO_SRC_IRQ) + sg->offset;
-}
-
-static void tmio_mmc_kunmap_atomic(void *virt, unsigned long *flags)
-{
-	kunmap_atomic(virt, KM_BIO_SRC_IRQ);
-	local_irq_restore(*flags);
-}
-
-#ifdef CONFIG_MMC_DEBUG
-
-#define STATUS_TO_TEXT(a) \
-	do { \
-		if (status & TMIO_STAT_##a) \
-			printk(#a); \
-	} while (0)
-
-void pr_debug_status(u32 status)
-{
-	printk(KERN_DEBUG "status: %08x = ", status);
-	STATUS_TO_TEXT(CARD_REMOVE);
-	STATUS_TO_TEXT(CARD_INSERT);
-	STATUS_TO_TEXT(SIGSTATE);
-	STATUS_TO_TEXT(WRPROTECT);
-	STATUS_TO_TEXT(CARD_REMOVE_A);
-	STATUS_TO_TEXT(CARD_INSERT_A);
-	STATUS_TO_TEXT(SIGSTATE_A);
-	STATUS_TO_TEXT(CMD_IDX_ERR);
-	STATUS_TO_TEXT(STOPBIT_ERR);
-	STATUS_TO_TEXT(ILL_FUNC);
-	STATUS_TO_TEXT(CMD_BUSY);
-	STATUS_TO_TEXT(CMDRESPEND);
-	STATUS_TO_TEXT(DATAEND);
-	STATUS_TO_TEXT(CRCFAIL);
-	STATUS_TO_TEXT(DATATIMEOUT);
-	STATUS_TO_TEXT(CMDTIMEOUT);
-	STATUS_TO_TEXT(RXOVERFLOW);
-	STATUS_TO_TEXT(TXUNDERRUN);
-	STATUS_TO_TEXT(RXRDY);
-	STATUS_TO_TEXT(TXRQ);
-	STATUS_TO_TEXT(ILL_ACCESS);
-	printk("\n");
-}
-
-#else
-#define pr_debug_status(s)  do { } while (0)
-#endif
-
-static void tmio_mmc_enable_sdio_irq(struct mmc_host *mmc, int enable)
-{
-	struct tmio_mmc_host *host = mmc_priv(mmc);
-
-	if (enable) {
-		host->sdio_irq_enabled = 1;
-		sd_ctrl_write16(host, CTL_TRANSACTION_CTL, 0x0001);
-		sd_ctrl_write16(host, CTL_SDIO_IRQ_MASK,
-			(TMIO_SDIO_MASK_ALL & ~TMIO_SDIO_STAT_IOIRQ));
-	} else {
-		sd_ctrl_write16(host, CTL_SDIO_IRQ_MASK, TMIO_SDIO_MASK_ALL);
-		sd_ctrl_write16(host, CTL_TRANSACTION_CTL, 0x0000);
-		host->sdio_irq_enabled = 0;
-	}
-}
-
-static void tmio_mmc_set_clock(struct tmio_mmc_host *host, int new_clock)
-{
-	u32 clk = 0, clock;
-
-	if (new_clock) {
-		for (clock = host->mmc->f_min, clk = 0x80000080;
-			new_clock >= (clock<<1); clk >>= 1)
-			clock <<= 1;
-		clk |= 0x100;
-	}
-
-	if (host->set_clk_div)
-		host->set_clk_div(host->pdev, (clk>>22) & 1);
-
-	sd_ctrl_write16(host, CTL_SD_CARD_CLK_CTL, clk & 0x1ff);
-}
-
-static void tmio_mmc_clk_stop(struct tmio_mmc_host *host)
-{
-	struct mfd_cell *cell = host->pdev->dev.platform_data;
-	struct tmio_mmc_data *pdata = cell->driver_data;
-
-	/*
-	 * Testing on sh-mobile showed that SDIO IRQs are unmasked when
-	 * CTL_CLK_AND_WAIT_CTL gets written, so we have to disable the
-	 * device IRQ here and restore the SDIO IRQ mask before
-	 * re-enabling the device IRQ.
-	 */
-	if (pdata->flags & TMIO_MMC_SDIO_IRQ)
-		disable_irq(host->irq);
-	sd_ctrl_write16(host, CTL_CLK_AND_WAIT_CTL, 0x0000);
-	msleep(10);
-	if (pdata->flags & TMIO_MMC_SDIO_IRQ) {
-		tmio_mmc_enable_sdio_irq(host->mmc, host->sdio_irq_enabled);
-		enable_irq(host->irq);
-	}
-	sd_ctrl_write16(host, CTL_SD_CARD_CLK_CTL, ~0x0100 &
-		sd_ctrl_read16(host, CTL_SD_CARD_CLK_CTL));
-	msleep(10);
-}
-
-static void tmio_mmc_clk_start(struct tmio_mmc_host *host)
-{
-	struct mfd_cell *cell = host->pdev->dev.platform_data;
-	struct tmio_mmc_data *pdata = cell->driver_data;
-
-	sd_ctrl_write16(host, CTL_SD_CARD_CLK_CTL, 0x0100 |
-		sd_ctrl_read16(host, CTL_SD_CARD_CLK_CTL));
-	msleep(10);
-	/* see comment in tmio_mmc_clk_stop above */
-	if (pdata->flags & TMIO_MMC_SDIO_IRQ)
-		disable_irq(host->irq);
-	sd_ctrl_write16(host, CTL_CLK_AND_WAIT_CTL, 0x0100);
-	msleep(10);
-	if (pdata->flags & TMIO_MMC_SDIO_IRQ) {
-		tmio_mmc_enable_sdio_irq(host->mmc, host->sdio_irq_enabled);
-		enable_irq(host->irq);
-	}
-}
-
-static void reset(struct tmio_mmc_host *host)
-{
-	/* FIXME - should we set stop clock reg here */
-	sd_ctrl_write16(host, CTL_RESET_SD, 0x0000);
-	sd_ctrl_write16(host, CTL_RESET_SDIO, 0x0000);
-	msleep(10);
-	sd_ctrl_write16(host, CTL_RESET_SD, 0x0001);
-	sd_ctrl_write16(host, CTL_RESET_SDIO, 0x0001);
-	msleep(10);
-}
-
-static void tmio_mmc_reset_work(struct work_struct *work)
-{
-	struct tmio_mmc_host *host = container_of(work, struct tmio_mmc_host,
-						  delayed_reset_work.work);
-	struct mmc_request *mrq;
-	unsigned long flags;
-
-	spin_lock_irqsave(&host->lock, flags);
-	mrq = host->mrq;
-
-	/* request already finished */
-	if (!mrq
-	    || time_is_after_jiffies(host->last_req_ts +
-		msecs_to_jiffies(2000))) {
-		spin_unlock_irqrestore(&host->lock, flags);
-		return;
-	}
-
-	dev_warn(&host->pdev->dev,
-		"timeout waiting for hardware interrupt (CMD%u)\n",
-		mrq->cmd->opcode);
-
-	if (host->data)
-		host->data->error = -ETIMEDOUT;
-	else if (host->cmd)
-		host->cmd->error = -ETIMEDOUT;
-	else
-		mrq->cmd->error = -ETIMEDOUT;
-
-	host->cmd = NULL;
-	host->data = NULL;
-	host->mrq = NULL;
-
-	spin_unlock_irqrestore(&host->lock, flags);
-
-	reset(host);
-
-	mmc_request_done(host->mmc, mrq);
-}
-
-static void
-tmio_mmc_finish_request(struct tmio_mmc_host *host)
-{
-	struct mmc_request *mrq = host->mrq;
-
-	if (!mrq)
-		return;
-
-	host->mrq = NULL;
-	host->cmd = NULL;
-	host->data = NULL;
-
-	cancel_delayed_work(&host->delayed_reset_work);
-
-	mmc_request_done(host->mmc, mrq);
-}
-
-/* These are the bitmasks the tmio chip requires to implement the MMC response
- * types. Note that R1 and R6 are the same in this scheme. */
-#define APP_CMD        0x0040
-#define RESP_NONE      0x0300
-#define RESP_R1        0x0400
-#define RESP_R1B       0x0500
-#define RESP_R2        0x0600
-#define RESP_R3        0x0700
-#define DATA_PRESENT   0x0800
-#define TRANSFER_READ  0x1000
-#define TRANSFER_MULTI 0x2000
-#define SECURITY_CMD   0x4000
-
-static int
-tmio_mmc_start_command(struct tmio_mmc_host *host, struct mmc_command *cmd)
-{
-	struct mmc_data *data = host->data;
-	int c = cmd->opcode;
-
-	/* Command 12 is handled by hardware */
-	if (cmd->opcode == 12 && !cmd->arg) {
-		sd_ctrl_write16(host, CTL_STOP_INTERNAL_ACTION, 0x001);
-		return 0;
-	}
-
-	switch (mmc_resp_type(cmd)) {
-	case MMC_RSP_NONE: c |= RESP_NONE; break;
-	case MMC_RSP_R1:   c |= RESP_R1;   break;
-	case MMC_RSP_R1B:  c |= RESP_R1B;  break;
-	case MMC_RSP_R2:   c |= RESP_R2;   break;
-	case MMC_RSP_R3:   c |= RESP_R3;   break;
-	default:
-		pr_debug("Unknown response type %d\n", mmc_resp_type(cmd));
-		return -EINVAL;
-	}
-
-	host->cmd = cmd;
-
-/* FIXME - this seems to be ok commented out but the spec suggest this bit
- *         should be set when issuing app commands.
- *	if(cmd->flags & MMC_FLAG_ACMD)
- *		c |= APP_CMD;
- */
-	if (data) {
-		c |= DATA_PRESENT;
-		if (data->blocks > 1) {
-			sd_ctrl_write16(host, CTL_STOP_INTERNAL_ACTION, 0x100);
-			c |= TRANSFER_MULTI;
-		}
-		if (data->flags & MMC_DATA_READ)
-			c |= TRANSFER_READ;
-	}
-
-	enable_mmc_irqs(host, TMIO_MASK_CMD);
-
-	/* Fire off the command */
-	sd_ctrl_write32(host, CTL_ARG_REG, cmd->arg);
-	sd_ctrl_write16(host, CTL_SD_CMD, c);
-
-	return 0;
-}
-
-/*
- * This chip always returns (at least?) as much data as you ask for.
- * I'm unsure what happens if you ask for less than a block. This should be
- * looked into to ensure that a funny length read doesnt hose the controller.
- */
-static void tmio_mmc_pio_irq(struct tmio_mmc_host *host)
-{
-	struct mmc_data *data = host->data;
-	void *sg_virt;
-	unsigned short *buf;
-	unsigned int count;
-	unsigned long flags;
-
-	if (!data) {
-		pr_debug("Spurious PIO IRQ\n");
-		return;
-	}
-
-	sg_virt = tmio_mmc_kmap_atomic(host->sg_ptr, &flags);
-	buf = (unsigned short *)(sg_virt + host->sg_off);
-
-	count = host->sg_ptr->length - host->sg_off;
-	if (count > data->blksz)
-		count = data->blksz;
-
-	pr_debug("count: %08x offset: %08x flags %08x\n",
-		 count, host->sg_off, data->flags);
-
-	/* Transfer the data */
-	if (data->flags & MMC_DATA_READ)
-		sd_ctrl_read16_rep(host, CTL_SD_DATA_PORT, buf, count >> 1);
-	else
-		sd_ctrl_write16_rep(host, CTL_SD_DATA_PORT, buf, count >> 1);
-
-	host->sg_off += count;
-
-	tmio_mmc_kunmap_atomic(sg_virt, &flags);
-
-	if (host->sg_off == host->sg_ptr->length)
-		tmio_mmc_next_sg(host);
-
-	return;
-}
-
-/* needs to be called with host->lock held */
-static void tmio_mmc_do_data_irq(struct tmio_mmc_host *host)
-{
-	struct mmc_data *data = host->data;
-	struct mmc_command *stop;
-
-	host->data = NULL;
-
-	if (!data) {
-		dev_warn(&host->pdev->dev, "Spurious data end IRQ\n");
-		return;
-	}
-	stop = data->stop;
-
-	/* FIXME - return correct transfer count on errors */
-	if (!data->error)
-		data->bytes_xfered = data->blocks * data->blksz;
-	else
-		data->bytes_xfered = 0;
-
-	pr_debug("Completed data request\n");
-
-	/*
-	 * FIXME: other drivers allow an optional stop command of any given type
-	 *        which we dont do, as the chip can auto generate them.
-	 *        Perhaps we can be smarter about when to use auto CMD12 and
-	 *        only issue the auto request when we know this is the desired
-	 *        stop command, allowing fallback to the stop command the
-	 *        upper layers expect. For now, we do what works.
-	 */
-
-	if (data->flags & MMC_DATA_READ) {
-		if (!host->chan_rx)
-			disable_mmc_irqs(host, TMIO_MASK_READOP);
-		else
-			tmio_check_bounce_buffer(host);
-		dev_dbg(&host->pdev->dev, "Complete Rx request %p\n",
-			host->mrq);
-	} else {
-		if (!host->chan_tx)
-			disable_mmc_irqs(host, TMIO_MASK_WRITEOP);
-		dev_dbg(&host->pdev->dev, "Complete Tx request %p\n",
-			host->mrq);
-	}
-
-	if (stop) {
-		if (stop->opcode == 12 && !stop->arg)
-			sd_ctrl_write16(host, CTL_STOP_INTERNAL_ACTION, 0x000);
-		else
-			BUG();
-	}
-
-	tmio_mmc_finish_request(host);
-}
-
-static void tmio_mmc_data_irq(struct tmio_mmc_host *host)
-{
-	struct mmc_data *data;
-	spin_lock(&host->lock);
-	data = host->data;
-
-	if (!data)
-		goto out;
-
-	if (host->chan_tx && (data->flags & MMC_DATA_WRITE)) {
-		/*
-		 * Has all data been written out yet? Testing on SuperH showed,
-		 * that in most cases the first interrupt comes already with the
-		 * BUSY status bit clear, but on some operations, like mount or
-		 * in the beginning of a write / sync / umount, there is one
-		 * DATAEND interrupt with the BUSY bit set, in this cases
-		 * waiting for one more interrupt fixes the problem.
-		 */
-		if (!(sd_ctrl_read32(host, CTL_STATUS) & TMIO_STAT_CMD_BUSY)) {
-			disable_mmc_irqs(host, TMIO_STAT_DATAEND);
-			tasklet_schedule(&host->dma_complete);
-		}
-	} else if (host->chan_rx && (data->flags & MMC_DATA_READ)) {
-		disable_mmc_irqs(host, TMIO_STAT_DATAEND);
-		tasklet_schedule(&host->dma_complete);
-	} else {
-		tmio_mmc_do_data_irq(host);
-	}
-out:
-	spin_unlock(&host->lock);
-}
-
-static void tmio_mmc_cmd_irq(struct tmio_mmc_host *host,
-	unsigned int stat)
-{
-	struct mmc_command *cmd = host->cmd;
-	int i, addr;
-
-	spin_lock(&host->lock);
-
-	if (!host->cmd) {
-		pr_debug("Spurious CMD irq\n");
-		goto out;
-	}
-
-	host->cmd = NULL;
-
-	/* This controller is sicker than the PXA one. Not only do we need to
-	 * drop the top 8 bits of the first response word, we also need to
-	 * modify the order of the response for short response command types.
-	 */
-
-	for (i = 3, addr = CTL_RESPONSE ; i >= 0 ; i--, addr += 4)
-		cmd->resp[i] = sd_ctrl_read32(host, addr);
-
-	if (cmd->flags &  MMC_RSP_136) {
-		cmd->resp[0] = (cmd->resp[0] << 8) | (cmd->resp[1] >> 24);
-		cmd->resp[1] = (cmd->resp[1] << 8) | (cmd->resp[2] >> 24);
-		cmd->resp[2] = (cmd->resp[2] << 8) | (cmd->resp[3] >> 24);
-		cmd->resp[3] <<= 8;
-	} else if (cmd->flags & MMC_RSP_R3) {
-		cmd->resp[0] = cmd->resp[3];
-	}
-
-	if (stat & TMIO_STAT_CMDTIMEOUT)
-		cmd->error = -ETIMEDOUT;
-	else if (stat & TMIO_STAT_CRCFAIL && cmd->flags & MMC_RSP_CRC)
-		cmd->error = -EILSEQ;
-
-	/* If there is data to handle we enable data IRQs here, and
-	 * we will ultimatley finish the request in the data_end handler.
-	 * If theres no data or we encountered an error, finish now.
-	 */
-	if (host->data && !cmd->error) {
-		if (host->data->flags & MMC_DATA_READ) {
-			if (!host->chan_rx)
-				enable_mmc_irqs(host, TMIO_MASK_READOP);
-		} else {
-			if (!host->chan_tx)
-				enable_mmc_irqs(host, TMIO_MASK_WRITEOP);
-			else
-				tasklet_schedule(&host->dma_issue);
-		}
-	} else {
-		tmio_mmc_finish_request(host);
-	}
-
-out:
-	spin_unlock(&host->lock);
-
-	return;
-}
-
-static irqreturn_t tmio_mmc_irq(int irq, void *devid)
-{
-	struct tmio_mmc_host *host = devid;
-	struct mfd_cell	*cell = host->pdev->dev.platform_data;
-	struct tmio_mmc_data *pdata = cell->driver_data;
-	unsigned int ireg, irq_mask, status;
-	unsigned int sdio_ireg, sdio_irq_mask, sdio_status;
-
-	pr_debug("MMC IRQ begin\n");
-
-	status = sd_ctrl_read32(host, CTL_STATUS);
-	irq_mask = sd_ctrl_read32(host, CTL_IRQ_MASK);
-	ireg = status & TMIO_MASK_IRQ & ~irq_mask;
-
-	sdio_ireg = 0;
-	if (!ireg && pdata->flags & TMIO_MMC_SDIO_IRQ) {
-		sdio_status = sd_ctrl_read16(host, CTL_SDIO_STATUS);
-		sdio_irq_mask = sd_ctrl_read16(host, CTL_SDIO_IRQ_MASK);
-		sdio_ireg = sdio_status & TMIO_SDIO_MASK_ALL & ~sdio_irq_mask;
-
-		sd_ctrl_write16(host, CTL_SDIO_STATUS, sdio_status & ~TMIO_SDIO_MASK_ALL);
-
-		if (sdio_ireg && !host->sdio_irq_enabled) {
-			pr_warning("tmio_mmc: Spurious SDIO IRQ, disabling! 0x%04x 0x%04x 0x%04x\n",
-				   sdio_status, sdio_irq_mask, sdio_ireg);
-			tmio_mmc_enable_sdio_irq(host->mmc, 0);
-			goto out;
-		}
 
-		if (host->mmc->caps & MMC_CAP_SDIO_IRQ &&
-			sdio_ireg & TMIO_SDIO_STAT_IOIRQ)
-			mmc_signal_sdio_irq(host->mmc);
-
-		if (sdio_ireg)
-			goto out;
-	}
-
-	pr_debug_status(status);
-	pr_debug_status(ireg);
-
-	if (!ireg) {
-		disable_mmc_irqs(host, status & ~irq_mask);
-
-		pr_warning("tmio_mmc: Spurious irq, disabling! "
-			"0x%08x 0x%08x 0x%08x\n", status, irq_mask, ireg);
-		pr_debug_status(status);
-
-		goto out;
-	}
-
-	while (ireg) {
-		/* Card insert / remove attempts */
-		if (ireg & (TMIO_STAT_CARD_INSERT | TMIO_STAT_CARD_REMOVE)) {
-			ack_mmc_irqs(host, TMIO_STAT_CARD_INSERT |
-				TMIO_STAT_CARD_REMOVE);
-			mmc_detect_change(host->mmc, msecs_to_jiffies(100));
-		}
-
-		/* CRC and other errors */
-/*		if (ireg & TMIO_STAT_ERR_IRQ)
- *			handled |= tmio_error_irq(host, irq, stat);
- */
-
-		/* Command completion */
-		if (ireg & (TMIO_STAT_CMDRESPEND | TMIO_STAT_CMDTIMEOUT)) {
-			ack_mmc_irqs(host,
-				     TMIO_STAT_CMDRESPEND |
-				     TMIO_STAT_CMDTIMEOUT);
-			tmio_mmc_cmd_irq(host, status);
-		}
-
-		/* Data transfer */
-		if (ireg & (TMIO_STAT_RXRDY | TMIO_STAT_TXRQ)) {
-			ack_mmc_irqs(host, TMIO_STAT_RXRDY | TMIO_STAT_TXRQ);
-			tmio_mmc_pio_irq(host);
-		}
-
-		/* Data transfer completion */
-		if (ireg & TMIO_STAT_DATAEND) {
-			ack_mmc_irqs(host, TMIO_STAT_DATAEND);
-			tmio_mmc_data_irq(host);
-		}
-
-		/* Check status - keep going until we've handled it all */
-		status = sd_ctrl_read32(host, CTL_STATUS);
-		irq_mask = sd_ctrl_read32(host, CTL_IRQ_MASK);
-		ireg = status & TMIO_MASK_IRQ & ~irq_mask;
-
-		pr_debug("Status at end of loop: %08x\n", status);
-		pr_debug_status(status);
-	}
-	pr_debug("MMC IRQ end\n");
-
-out:
-	return IRQ_HANDLED;
-}
-
-#ifdef CONFIG_TMIO_MMC_DMA
-static void tmio_check_bounce_buffer(struct tmio_mmc_host *host)
-{
-	if (host->sg_ptr == &host->bounce_sg) {
-		unsigned long flags;
-		void *sg_vaddr = tmio_mmc_kmap_atomic(host->sg_orig, &flags);
-		memcpy(sg_vaddr, host->bounce_buf, host->bounce_sg.length);
-		tmio_mmc_kunmap_atomic(sg_vaddr, &flags);
-	}
-}
-
-static void tmio_mmc_enable_dma(struct tmio_mmc_host *host, bool enable)
-{
-#if defined(CONFIG_SUPERH) || defined(CONFIG_ARCH_SHMOBILE)
-	/* Switch DMA mode on or off - SuperH specific? */
-	sd_ctrl_write16(host, 0xd8, enable ? 2 : 0);
-#endif
-}
-
-static void tmio_dma_complete(void *arg)
-{
-	struct tmio_mmc_host *host = arg;
-
-	dev_dbg(&host->pdev->dev, "Command completed\n");
-
-	if (!host->data)
-		dev_warn(&host->pdev->dev, "NULL data in DMA completion!\n");
-	else
-		enable_mmc_irqs(host, TMIO_STAT_DATAEND);
-}
-
-static void tmio_mmc_start_dma_rx(struct tmio_mmc_host *host)
-{
-	struct scatterlist *sg = host->sg_ptr, *sg_tmp;
-	struct dma_async_tx_descriptor *desc = NULL;
-	struct dma_chan *chan = host->chan_rx;
-	struct mfd_cell	*cell = host->pdev->dev.platform_data;
-	struct tmio_mmc_data *pdata = cell->driver_data;
-	dma_cookie_t cookie;
-	int ret, i;
-	bool aligned = true, multiple = true;
-	unsigned int align = (1 << pdata->dma->alignment_shift) - 1;
-
-	for_each_sg(sg, sg_tmp, host->sg_len, i) {
-		if (sg_tmp->offset & align)
-			aligned = false;
-		if (sg_tmp->length & align) {
-			multiple = false;
-			break;
-		}
-	}
-
-	if ((!aligned && (host->sg_len > 1 || sg->length > PAGE_CACHE_SIZE ||
-			  align >= MAX_ALIGN)) || !multiple) {
-		ret = -EINVAL;
-		goto pio;
-	}
-
-	/* The only sg element can be unaligned, use our bounce buffer then */
-	if (!aligned) {
-		sg_init_one(&host->bounce_sg, host->bounce_buf, sg->length);
-		host->sg_ptr = &host->bounce_sg;
-		sg = host->sg_ptr;
-	}
-
-	ret = dma_map_sg(&host->pdev->dev, sg, host->sg_len, DMA_FROM_DEVICE);
-	if (ret > 0) {
-		host->dma_sglen = ret;
-		desc = chan->device->device_prep_slave_sg(chan, sg, ret,
-			DMA_FROM_DEVICE, DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
-	}
-
-	if (desc) {
-		desc->callback = tmio_dma_complete;
-		desc->callback_param = host;
-		cookie = desc->tx_submit(desc);
-		if (cookie < 0) {
-			desc = NULL;
-			ret = cookie;
-		} else {
-			chan->device->device_issue_pending(chan);
-		}
-	}
-	dev_dbg(&host->pdev->dev, "%s(): mapped %d -> %d, cookie %d, rq %p\n",
-		__func__, host->sg_len, ret, cookie, host->mrq);
-
-pio:
-	if (!desc) {
-		/* DMA failed, fall back to PIO */
-		if (ret >= 0)
-			ret = -EIO;
-		host->chan_rx = NULL;
-		dma_release_channel(chan);
-		/* Free the Tx channel too */
-		chan = host->chan_tx;
-		if (chan) {
-			host->chan_tx = NULL;
-			dma_release_channel(chan);
-		}
-		dev_warn(&host->pdev->dev,
-			 "DMA failed: %d, falling back to PIO\n", ret);
-		tmio_mmc_enable_dma(host, false);
-	}
-
-	dev_dbg(&host->pdev->dev, "%s(): desc %p, cookie %d, sg[%d]\n", __func__,
-		desc, cookie, host->sg_len);
-}
-
-static void tmio_mmc_start_dma_tx(struct tmio_mmc_host *host)
-{
-	struct scatterlist *sg = host->sg_ptr, *sg_tmp;
-	struct dma_async_tx_descriptor *desc = NULL;
-	struct dma_chan *chan = host->chan_tx;
-	struct mfd_cell	*cell = host->pdev->dev.platform_data;
-	struct tmio_mmc_data *pdata = cell->driver_data;
-	dma_cookie_t cookie;
-	int ret, i;
-	bool aligned = true, multiple = true;
-	unsigned int align = (1 << pdata->dma->alignment_shift) - 1;
-
-	for_each_sg(sg, sg_tmp, host->sg_len, i) {
-		if (sg_tmp->offset & align)
-			aligned = false;
-		if (sg_tmp->length & align) {
-			multiple = false;
-			break;
-		}
-	}
-
-	if ((!aligned && (host->sg_len > 1 || sg->length > PAGE_CACHE_SIZE ||
-			  align >= MAX_ALIGN)) || !multiple) {
-		ret = -EINVAL;
-		goto pio;
-	}
-
-	/* The only sg element can be unaligned, use our bounce buffer then */
-	if (!aligned) {
-		unsigned long flags;
-		void *sg_vaddr = tmio_mmc_kmap_atomic(sg, &flags);
-		sg_init_one(&host->bounce_sg, host->bounce_buf, sg->length);
-		memcpy(host->bounce_buf, sg_vaddr, host->bounce_sg.length);
-		tmio_mmc_kunmap_atomic(sg_vaddr, &flags);
-		host->sg_ptr = &host->bounce_sg;
-		sg = host->sg_ptr;
-	}
-
-	ret = dma_map_sg(&host->pdev->dev, sg, host->sg_len, DMA_TO_DEVICE);
-	if (ret > 0) {
-		host->dma_sglen = ret;
-		desc = chan->device->device_prep_slave_sg(chan, sg, ret,
-			DMA_TO_DEVICE, DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
-	}
-
-	if (desc) {
-		desc->callback = tmio_dma_complete;
-		desc->callback_param = host;
-		cookie = desc->tx_submit(desc);
-		if (cookie < 0) {
-			desc = NULL;
-			ret = cookie;
-		}
-	}
-	dev_dbg(&host->pdev->dev, "%s(): mapped %d -> %d, cookie %d, rq %p\n",
-		__func__, host->sg_len, ret, cookie, host->mrq);
-
-pio:
-	if (!desc) {
-		/* DMA failed, fall back to PIO */
-		if (ret >= 0)
-			ret = -EIO;
-		host->chan_tx = NULL;
-		dma_release_channel(chan);
-		/* Free the Rx channel too */
-		chan = host->chan_rx;
-		if (chan) {
-			host->chan_rx = NULL;
-			dma_release_channel(chan);
-		}
-		dev_warn(&host->pdev->dev,
-			 "DMA failed: %d, falling back to PIO\n", ret);
-		tmio_mmc_enable_dma(host, false);
-	}
-
-	dev_dbg(&host->pdev->dev, "%s(): desc %p, cookie %d\n", __func__,
-		desc, cookie);
-}
-
-static void tmio_mmc_start_dma(struct tmio_mmc_host *host,
-			       struct mmc_data *data)
-{
-	if (data->flags & MMC_DATA_READ) {
-		if (host->chan_rx)
-			tmio_mmc_start_dma_rx(host);
-	} else {
-		if (host->chan_tx)
-			tmio_mmc_start_dma_tx(host);
-	}
-}
-
-static void tmio_issue_tasklet_fn(unsigned long priv)
-{
-	struct tmio_mmc_host *host = (struct tmio_mmc_host *)priv;
-	struct dma_chan *chan = host->chan_tx;
-
-	chan->device->device_issue_pending(chan);
-}
-
-static void tmio_tasklet_fn(unsigned long arg)
-{
-	struct tmio_mmc_host *host = (struct tmio_mmc_host *)arg;
-	unsigned long flags;
-
-	spin_lock_irqsave(&host->lock, flags);
-
-	if (!host->data)
-		goto out;
-
-	if (host->data->flags & MMC_DATA_READ)
-		dma_unmap_sg(&host->pdev->dev, host->sg_ptr, host->dma_sglen,
-			     DMA_FROM_DEVICE);
-	else
-		dma_unmap_sg(&host->pdev->dev, host->sg_ptr, host->dma_sglen,
-			     DMA_TO_DEVICE);
-
-	tmio_mmc_do_data_irq(host);
-out:
-	spin_unlock_irqrestore(&host->lock, flags);
-}
-
-/* It might be necessary to make filter MFD specific */
-static bool tmio_mmc_filter(struct dma_chan *chan, void *arg)
-{
-	dev_dbg(chan->device->dev, "%s: slave data %p\n", __func__, arg);
-	chan->private = arg;
-	return true;
-}
-
-static void tmio_mmc_request_dma(struct tmio_mmc_host *host,
-				 struct tmio_mmc_data *pdata)
-{
-	/* We can only either use DMA for both Tx and Rx or not use it at all */
-	if (pdata->dma) {
-		dma_cap_mask_t mask;
-
-		dma_cap_zero(mask);
-		dma_cap_set(DMA_SLAVE, mask);
-
-		host->chan_tx = dma_request_channel(mask, tmio_mmc_filter,
-						    pdata->dma->chan_priv_tx);
-		dev_dbg(&host->pdev->dev, "%s: TX: got channel %p\n", __func__,
-			host->chan_tx);
-
-		if (!host->chan_tx)
-			return;
-
-		host->chan_rx = dma_request_channel(mask, tmio_mmc_filter,
-						    pdata->dma->chan_priv_rx);
-		dev_dbg(&host->pdev->dev, "%s: RX: got channel %p\n", __func__,
-			host->chan_rx);
-
-		if (!host->chan_rx) {
-			dma_release_channel(host->chan_tx);
-			host->chan_tx = NULL;
-			return;
-		}
-
-		tasklet_init(&host->dma_complete, tmio_tasklet_fn, (unsigned long)host);
-		tasklet_init(&host->dma_issue, tmio_issue_tasklet_fn, (unsigned long)host);
-
-		tmio_mmc_enable_dma(host, true);
-	}
-}
-
-static void tmio_mmc_release_dma(struct tmio_mmc_host *host)
-{
-	if (host->chan_tx) {
-		struct dma_chan *chan = host->chan_tx;
-		host->chan_tx = NULL;
-		dma_release_channel(chan);
-	}
-	if (host->chan_rx) {
-		struct dma_chan *chan = host->chan_rx;
-		host->chan_rx = NULL;
-		dma_release_channel(chan);
-	}
-}
-#else
-static void tmio_check_bounce_buffer(struct tmio_mmc_host *host)
-{
-}
-
-static void tmio_mmc_start_dma(struct tmio_mmc_host *host,
-			       struct mmc_data *data)
-{
-}
-
-static void tmio_mmc_request_dma(struct tmio_mmc_host *host,
-				 struct tmio_mmc_data *pdata)
-{
-	host->chan_tx = NULL;
-	host->chan_rx = NULL;
-}
-
-static void tmio_mmc_release_dma(struct tmio_mmc_host *host)
-{
-}
-#endif
-
-static int tmio_mmc_start_data(struct tmio_mmc_host *host,
-	struct mmc_data *data)
-{
-	struct mfd_cell *cell = host->pdev->dev.platform_data;
-	struct tmio_mmc_data *pdata = cell->driver_data;
-
-	pr_debug("setup data transfer: blocksize %08x  nr_blocks %d\n",
-		 data->blksz, data->blocks);
-
-	/* Some hardware cannot perform 2 byte requests in 4 bit mode */
-	if (host->mmc->ios.bus_width == MMC_BUS_WIDTH_4) {
-		int blksz_2bytes = pdata->flags & TMIO_MMC_BLKSZ_2BYTES;
-
-		if (data->blksz < 2 || (data->blksz < 4 && !blksz_2bytes)) {
-			pr_err("%s: %d byte block unsupported in 4 bit mode\n",
-			       mmc_hostname(host->mmc), data->blksz);
-			return -EINVAL;
-		}
-	}
-
-	tmio_mmc_init_sg(host, data);
-	host->data = data;
-
-	/* Set transfer length / blocksize */
-	sd_ctrl_write16(host, CTL_SD_XFER_LEN, data->blksz);
-	sd_ctrl_write16(host, CTL_XFER_BLK_COUNT, data->blocks);
-
-	tmio_mmc_start_dma(host, data);
-
-	return 0;
-}
-
-/* Process requests from the MMC layer */
-static void tmio_mmc_request(struct mmc_host *mmc, struct mmc_request *mrq)
-{
-	struct tmio_mmc_host *host = mmc_priv(mmc);
-	int ret;
-
-	if (host->mrq)
-		pr_debug("request not null\n");
-
-	host->last_req_ts = jiffies;
-	wmb();
-	host->mrq = mrq;
-
-	if (mrq->data) {
-		ret = tmio_mmc_start_data(host, mrq->data);
-		if (ret)
-			goto fail;
-	}
-
-	ret = tmio_mmc_start_command(host, mrq->cmd);
-	if (!ret) {
-		schedule_delayed_work(&host->delayed_reset_work,
-				      msecs_to_jiffies(2000));
-		return;
-	}
-
-fail:
-	host->mrq = NULL;
-	mrq->cmd->error = ret;
-	mmc_request_done(mmc, mrq);
-}
-
-/* Set MMC clock / power.
- * Note: This controller uses a simple divider scheme therefore it cannot
- * run a MMC card at full speed (20MHz). The max clock is 24MHz on SD, but as
- * MMC wont run that fast, it has to be clocked at 12MHz which is the next
- * slowest setting.
- */
-static void tmio_mmc_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
-{
-	struct tmio_mmc_host *host = mmc_priv(mmc);
-
-	if (ios->clock)
-		tmio_mmc_set_clock(host, ios->clock);
-
-	/* Power sequence - OFF -> ON -> UP */
-	switch (ios->power_mode) {
-	case MMC_POWER_OFF: /* power down SD bus */
-		if (host->set_pwr)
-			host->set_pwr(host->pdev, 0);
-		tmio_mmc_clk_stop(host);
-		break;
-	case MMC_POWER_ON: /* power up SD bus */
-		if (host->set_pwr)
-			host->set_pwr(host->pdev, 1);
-		break;
-	case MMC_POWER_UP: /* start bus clock */
-		tmio_mmc_clk_start(host);
-		break;
-	}
-
-	switch (ios->bus_width) {
-	case MMC_BUS_WIDTH_1:
-		sd_ctrl_write16(host, CTL_SD_MEM_CARD_OPT, 0x80e0);
-	break;
-	case MMC_BUS_WIDTH_4:
-		sd_ctrl_write16(host, CTL_SD_MEM_CARD_OPT, 0x00e0);
-	break;
-	}
-
-	/* Let things settle. delay taken from winCE driver */
-	udelay(140);
-}
-
-static int tmio_mmc_get_ro(struct mmc_host *mmc)
-{
-	struct tmio_mmc_host *host = mmc_priv(mmc);
-	struct mfd_cell	*cell = host->pdev->dev.platform_data;
-	struct tmio_mmc_data *pdata = cell->driver_data;
-
-	return ((pdata->flags & TMIO_MMC_WRPROTECT_DISABLE) ||
-		(sd_ctrl_read32(host, CTL_STATUS) & TMIO_STAT_WRPROTECT)) ? 0 : 1;
-}
-
-static int tmio_mmc_get_cd(struct mmc_host *mmc)
-{
-	struct tmio_mmc_host *host = mmc_priv(mmc);
-	struct mfd_cell	*cell = host->pdev->dev.platform_data;
-	struct tmio_mmc_data *pdata = cell->driver_data;
-
-	if (!pdata->get_cd)
-		return -ENOSYS;
-	else
-		return pdata->get_cd(host->pdev);
-}
-
-static const struct mmc_host_ops tmio_mmc_ops = {
-	.request	= tmio_mmc_request,
-	.set_ios	= tmio_mmc_set_ios,
-	.get_ro         = tmio_mmc_get_ro,
-	.get_cd		= tmio_mmc_get_cd,
-	.enable_sdio_irq = tmio_mmc_enable_sdio_irq,
-};
+#include "tmio_mmc.h"
 
 #ifdef CONFIG_PM
 static int tmio_mmc_suspend(struct platform_device *dev, pm_message_t state)
 {
-	struct mfd_cell	*cell = (struct mfd_cell *)dev->dev.platform_data;
+	const struct mfd_cell *cell = mfd_get_cell(dev);
 	struct mmc_host *mmc = platform_get_drvdata(dev);
 	int ret;
 
@@ -1222,7 +41,7 @@ static int tmio_mmc_suspend(struct platform_device *dev, pm_message_t state)
 
 static int tmio_mmc_resume(struct platform_device *dev)
 {
-	struct mfd_cell	*cell = (struct mfd_cell *)dev->dev.platform_data;
+	const struct mfd_cell *cell = mfd_get_cell(dev);
 	struct mmc_host *mmc = platform_get_drvdata(dev);
 	int ret = 0;
 
@@ -1243,138 +62,54 @@ out:
 #define tmio_mmc_resume NULL
 #endif
 
-static int __devinit tmio_mmc_probe(struct platform_device *dev)
+static int __devinit tmio_mmc_probe(struct platform_device *pdev)
 {
-	struct mfd_cell	*cell = (struct mfd_cell *)dev->dev.platform_data;
+	const struct mfd_cell *cell = mfd_get_cell(pdev);
 	struct tmio_mmc_data *pdata;
-	struct resource *res_ctl;
 	struct tmio_mmc_host *host;
-	struct mmc_host *mmc;
 	int ret = -EINVAL;
-	u32 irq_mask = TMIO_MASK_CMD;
 
-	if (dev->num_resources != 2)
+	if (pdev->num_resources != 2)
 		goto out;
 
-	res_ctl = platform_get_resource(dev, IORESOURCE_MEM, 0);
-	if (!res_ctl)
-		goto out;
-
-	pdata = cell->driver_data;
+	pdata = mfd_get_data(pdev);
 	if (!pdata || !pdata->hclk)
 		goto out;
 
-	ret = -ENOMEM;
-
-	mmc = mmc_alloc_host(sizeof(struct tmio_mmc_host), &dev->dev);
-	if (!mmc)
-		goto out;
-
-	host = mmc_priv(mmc);
-	host->mmc = mmc;
-	host->pdev = dev;
-	platform_set_drvdata(dev, mmc);
-
-	host->set_pwr = pdata->set_pwr;
-	host->set_clk_div = pdata->set_clk_div;
-
-	/* SD control register space size is 0x200, 0x400 for bus_shift=1 */
-	host->bus_shift = resource_size(res_ctl) >> 10;
-
-	host->ctl = ioremap(res_ctl->start, resource_size(res_ctl));
-	if (!host->ctl)
-		goto host_free;
-
-	mmc->ops = &tmio_mmc_ops;
-	mmc->caps = MMC_CAP_4_BIT_DATA | pdata->capabilities;
-	mmc->f_max = pdata->hclk;
-	mmc->f_min = mmc->f_max / 512;
-	mmc->max_segs = 32;
-	mmc->max_blk_size = 512;
-	mmc->max_blk_count = (PAGE_CACHE_SIZE / mmc->max_blk_size) *
-		mmc->max_segs;
-	mmc->max_req_size = mmc->max_blk_size * mmc->max_blk_count;
-	mmc->max_seg_size = mmc->max_req_size;
-	if (pdata->ocr_mask)
-		mmc->ocr_avail = pdata->ocr_mask;
-	else
-		mmc->ocr_avail = MMC_VDD_32_33 | MMC_VDD_33_34;
-
 	/* Tell the MFD core we are ready to be enabled */
 	if (cell->enable) {
-		ret = cell->enable(dev);
+		ret = cell->enable(pdev);
 		if (ret)
-			goto unmap_ctl;
+			goto out;
 	}
 
-	tmio_mmc_clk_stop(host);
-	reset(host);
-
-	ret = platform_get_irq(dev, 0);
-	if (ret >= 0)
-		host->irq = ret;
-	else
-		goto cell_disable;
-
-	disable_mmc_irqs(host, TMIO_MASK_ALL);
-	if (pdata->flags & TMIO_MMC_SDIO_IRQ)
-		tmio_mmc_enable_sdio_irq(mmc, 0);
-
-	ret = request_irq(host->irq, tmio_mmc_irq, IRQF_DISABLED |
-		IRQF_TRIGGER_FALLING, dev_name(&dev->dev), host);
+	ret = tmio_mmc_host_probe(&host, pdev, pdata);
 	if (ret)
 		goto cell_disable;
 
-	spin_lock_init(&host->lock);
-
-	/* Init delayed work for request timeouts */
-	INIT_DELAYED_WORK(&host->delayed_reset_work, tmio_mmc_reset_work);
-
-	/* See if we also get DMA */
-	tmio_mmc_request_dma(host, pdata);
-
-	mmc_add_host(mmc);
-
 	pr_info("%s at 0x%08lx irq %d\n", mmc_hostname(host->mmc),
 		(unsigned long)host->ctl, host->irq);
 
-	/* Unmask the IRQs we want to know about */
-	if (!host->chan_rx)
-		irq_mask |= TMIO_MASK_READOP;
-	if (!host->chan_tx)
-		irq_mask |= TMIO_MASK_WRITEOP;
-	enable_mmc_irqs(host, irq_mask);
-
 	return 0;
 
 cell_disable:
 	if (cell->disable)
-		cell->disable(dev);
-unmap_ctl:
-	iounmap(host->ctl);
-host_free:
-	mmc_free_host(mmc);
+		cell->disable(pdev);
 out:
 	return ret;
 }
 
-static int __devexit tmio_mmc_remove(struct platform_device *dev)
+static int __devexit tmio_mmc_remove(struct platform_device *pdev)
 {
-	struct mfd_cell	*cell = (struct mfd_cell *)dev->dev.platform_data;
-	struct mmc_host *mmc = platform_get_drvdata(dev);
+	const struct mfd_cell *cell = mfd_get_cell(pdev);
+	struct mmc_host *mmc = platform_get_drvdata(pdev);
 
-	platform_set_drvdata(dev, NULL);
+	platform_set_drvdata(pdev, NULL);
 
 	if (mmc) {
-		struct tmio_mmc_host *host = mmc_priv(mmc);
-		mmc_remove_host(mmc);
-		cancel_delayed_work_sync(&host->delayed_reset_work);
-		tmio_mmc_release_dma(host);
-		free_irq(host->irq, host);
+		tmio_mmc_host_remove(mmc_priv(mmc));
 		if (cell->disable)
-			cell->disable(dev);
-		iounmap(host->ctl);
-		mmc_free_host(mmc);
+			cell->disable(pdev);
 	}
 
 	return 0;
diff --git a/drivers/mmc/host/tmio_mmc.h b/drivers/mmc/host/tmio_mmc.h
new file mode 100644
index 00000000000..099ed49a259
--- /dev/null
+++ b/drivers/mmc/host/tmio_mmc.h
@@ -0,0 +1,123 @@
+/*
+ * linux/drivers/mmc/host/tmio_mmc.h
+ *
+ * Copyright (C) 2007 Ian Molton
+ * Copyright (C) 2004 Ian Molton
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Driver for the MMC / SD / SDIO cell found in:
+ *
+ * TC6393XB TC6391XB TC6387XB T7L66XB ASIC3
+ */
+
+#ifndef TMIO_MMC_H
+#define TMIO_MMC_H
+
+#include <linux/highmem.h>
+#include <linux/mmc/tmio.h>
+#include <linux/pagemap.h>
+
+/* Definitions for values the CTRL_SDIO_STATUS register can take. */
+#define TMIO_SDIO_STAT_IOIRQ	0x0001
+#define TMIO_SDIO_STAT_EXPUB52	0x4000
+#define TMIO_SDIO_STAT_EXWT	0x8000
+#define TMIO_SDIO_MASK_ALL	0xc007
+
+/* Define some IRQ masks */
+/* This is the mask used at reset by the chip */
+#define TMIO_MASK_ALL           0x837f031d
+#define TMIO_MASK_READOP  (TMIO_STAT_RXRDY | TMIO_STAT_DATAEND)
+#define TMIO_MASK_WRITEOP (TMIO_STAT_TXRQ | TMIO_STAT_DATAEND)
+#define TMIO_MASK_CMD     (TMIO_STAT_CMDRESPEND | TMIO_STAT_CMDTIMEOUT | \
+		TMIO_STAT_CARD_REMOVE | TMIO_STAT_CARD_INSERT)
+#define TMIO_MASK_IRQ     (TMIO_MASK_READOP | TMIO_MASK_WRITEOP | TMIO_MASK_CMD)
+
+struct tmio_mmc_data;
+
+struct tmio_mmc_host {
+	void __iomem *ctl;
+	unsigned long bus_shift;
+	struct mmc_command      *cmd;
+	struct mmc_request      *mrq;
+	struct mmc_data         *data;
+	struct mmc_host         *mmc;
+	int                     irq;
+	unsigned int		sdio_irq_enabled;
+
+	/* Callbacks for clock / power control */
+	void (*set_pwr)(struct platform_device *host, int state);
+	void (*set_clk_div)(struct platform_device *host, int state);
+
+	/* pio related stuff */
+	struct scatterlist      *sg_ptr;
+	struct scatterlist      *sg_orig;
+	unsigned int            sg_len;
+	unsigned int            sg_off;
+
+	struct platform_device *pdev;
+	struct tmio_mmc_data *pdata;
+
+	/* DMA support */
+	bool			force_pio;
+	struct dma_chan		*chan_rx;
+	struct dma_chan		*chan_tx;
+	struct tasklet_struct	dma_complete;
+	struct tasklet_struct	dma_issue;
+	struct scatterlist	bounce_sg;
+	u8			*bounce_buf;
+
+	/* Track lost interrupts */
+	struct delayed_work	delayed_reset_work;
+	spinlock_t		lock;
+	unsigned long		last_req_ts;
+};
+
+int tmio_mmc_host_probe(struct tmio_mmc_host **host,
+			struct platform_device *pdev,
+			struct tmio_mmc_data *pdata);
+void tmio_mmc_host_remove(struct tmio_mmc_host *host);
+void tmio_mmc_do_data_irq(struct tmio_mmc_host *host);
+
+void tmio_mmc_enable_mmc_irqs(struct tmio_mmc_host *host, u32 i);
+void tmio_mmc_disable_mmc_irqs(struct tmio_mmc_host *host, u32 i);
+
+static inline char *tmio_mmc_kmap_atomic(struct scatterlist *sg,
+					 unsigned long *flags)
+{
+	local_irq_save(*flags);
+	return kmap_atomic(sg_page(sg), KM_BIO_SRC_IRQ) + sg->offset;
+}
+
+static inline void tmio_mmc_kunmap_atomic(struct scatterlist *sg,
+					  unsigned long *flags, void *virt)
+{
+	kunmap_atomic(virt - sg->offset, KM_BIO_SRC_IRQ);
+	local_irq_restore(*flags);
+}
+
+#if defined(CONFIG_MMC_SDHI) || defined(CONFIG_MMC_SDHI_MODULE)
+void tmio_mmc_start_dma(struct tmio_mmc_host *host, struct mmc_data *data);
+void tmio_mmc_request_dma(struct tmio_mmc_host *host, struct tmio_mmc_data *pdata);
+void tmio_mmc_release_dma(struct tmio_mmc_host *host);
+#else
+static inline void tmio_mmc_start_dma(struct tmio_mmc_host *host,
+			       struct mmc_data *data)
+{
+}
+
+static inline void tmio_mmc_request_dma(struct tmio_mmc_host *host,
+				 struct tmio_mmc_data *pdata)
+{
+	host->chan_tx = NULL;
+	host->chan_rx = NULL;
+}
+
+static inline void tmio_mmc_release_dma(struct tmio_mmc_host *host)
+{
+}
+#endif
+
+#endif
diff --git a/drivers/mmc/host/tmio_mmc_dma.c b/drivers/mmc/host/tmio_mmc_dma.c
new file mode 100644
index 00000000000..d3de74ab633
--- /dev/null
+++ b/drivers/mmc/host/tmio_mmc_dma.c
@@ -0,0 +1,317 @@
+/*
+ * linux/drivers/mmc/tmio_mmc_dma.c
+ *
+ * Copyright (C) 2010-2011 Guennadi Liakhovetski
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * DMA function for TMIO MMC implementations
+ */
+
+#include <linux/device.h>
+#include <linux/dmaengine.h>
+#include <linux/mfd/tmio.h>
+#include <linux/mmc/host.h>
+#include <linux/mmc/tmio.h>
+#include <linux/pagemap.h>
+#include <linux/scatterlist.h>
+
+#include "tmio_mmc.h"
+
+#define TMIO_MMC_MIN_DMA_LEN 8
+
+static void tmio_mmc_enable_dma(struct tmio_mmc_host *host, bool enable)
+{
+#if defined(CONFIG_SUPERH) || defined(CONFIG_ARCH_SHMOBILE)
+	/* Switch DMA mode on or off - SuperH specific? */
+	writew(enable ? 2 : 0, host->ctl + (0xd8 << host->bus_shift));
+#endif
+}
+
+static void tmio_mmc_start_dma_rx(struct tmio_mmc_host *host)
+{
+	struct scatterlist *sg = host->sg_ptr, *sg_tmp;
+	struct dma_async_tx_descriptor *desc = NULL;
+	struct dma_chan *chan = host->chan_rx;
+	struct tmio_mmc_data *pdata = host->pdata;
+	dma_cookie_t cookie;
+	int ret, i;
+	bool aligned = true, multiple = true;
+	unsigned int align = (1 << pdata->dma->alignment_shift) - 1;
+
+	for_each_sg(sg, sg_tmp, host->sg_len, i) {
+		if (sg_tmp->offset & align)
+			aligned = false;
+		if (sg_tmp->length & align) {
+			multiple = false;
+			break;
+		}
+	}
+
+	if ((!aligned && (host->sg_len > 1 || sg->length > PAGE_CACHE_SIZE ||
+			  (align & PAGE_MASK))) || !multiple) {
+		ret = -EINVAL;
+		goto pio;
+	}
+
+	if (sg->length < TMIO_MMC_MIN_DMA_LEN) {
+		host->force_pio = true;
+		return;
+	}
+
+	tmio_mmc_disable_mmc_irqs(host, TMIO_STAT_RXRDY);
+
+	/* The only sg element can be unaligned, use our bounce buffer then */
+	if (!aligned) {
+		sg_init_one(&host->bounce_sg, host->bounce_buf, sg->length);
+		host->sg_ptr = &host->bounce_sg;
+		sg = host->sg_ptr;
+	}
+
+	ret = dma_map_sg(chan->device->dev, sg, host->sg_len, DMA_FROM_DEVICE);
+	if (ret > 0)
+		desc = chan->device->device_prep_slave_sg(chan, sg, ret,
+			DMA_FROM_DEVICE, DMA_CTRL_ACK);
+
+	if (desc) {
+		cookie = dmaengine_submit(desc);
+		if (cookie < 0) {
+			desc = NULL;
+			ret = cookie;
+		}
+	}
+	dev_dbg(&host->pdev->dev, "%s(): mapped %d -> %d, cookie %d, rq %p\n",
+		__func__, host->sg_len, ret, cookie, host->mrq);
+
+pio:
+	if (!desc) {
+		/* DMA failed, fall back to PIO */
+		if (ret >= 0)
+			ret = -EIO;
+		host->chan_rx = NULL;
+		dma_release_channel(chan);
+		/* Free the Tx channel too */
+		chan = host->chan_tx;
+		if (chan) {
+			host->chan_tx = NULL;
+			dma_release_channel(chan);
+		}
+		dev_warn(&host->pdev->dev,
+			 "DMA failed: %d, falling back to PIO\n", ret);
+		tmio_mmc_enable_dma(host, false);
+	}
+
+	dev_dbg(&host->pdev->dev, "%s(): desc %p, cookie %d, sg[%d]\n", __func__,
+		desc, cookie, host->sg_len);
+}
+
+static void tmio_mmc_start_dma_tx(struct tmio_mmc_host *host)
+{
+	struct scatterlist *sg = host->sg_ptr, *sg_tmp;
+	struct dma_async_tx_descriptor *desc = NULL;
+	struct dma_chan *chan = host->chan_tx;
+	struct tmio_mmc_data *pdata = host->pdata;
+	dma_cookie_t cookie;
+	int ret, i;
+	bool aligned = true, multiple = true;
+	unsigned int align = (1 << pdata->dma->alignment_shift) - 1;
+
+	for_each_sg(sg, sg_tmp, host->sg_len, i) {
+		if (sg_tmp->offset & align)
+			aligned = false;
+		if (sg_tmp->length & align) {
+			multiple = false;
+			break;
+		}
+	}
+
+	if ((!aligned && (host->sg_len > 1 || sg->length > PAGE_CACHE_SIZE ||
+			  (align & PAGE_MASK))) || !multiple) {
+		ret = -EINVAL;
+		goto pio;
+	}
+
+	if (sg->length < TMIO_MMC_MIN_DMA_LEN) {
+		host->force_pio = true;
+		return;
+	}
+
+	tmio_mmc_disable_mmc_irqs(host, TMIO_STAT_TXRQ);
+
+	/* The only sg element can be unaligned, use our bounce buffer then */
+	if (!aligned) {
+		unsigned long flags;
+		void *sg_vaddr = tmio_mmc_kmap_atomic(sg, &flags);
+		sg_init_one(&host->bounce_sg, host->bounce_buf, sg->length);
+		memcpy(host->bounce_buf, sg_vaddr, host->bounce_sg.length);
+		tmio_mmc_kunmap_atomic(sg, &flags, sg_vaddr);
+		host->sg_ptr = &host->bounce_sg;
+		sg = host->sg_ptr;
+	}
+
+	ret = dma_map_sg(chan->device->dev, sg, host->sg_len, DMA_TO_DEVICE);
+	if (ret > 0)
+		desc = chan->device->device_prep_slave_sg(chan, sg, ret,
+			DMA_TO_DEVICE, DMA_CTRL_ACK);
+
+	if (desc) {
+		cookie = dmaengine_submit(desc);
+		if (cookie < 0) {
+			desc = NULL;
+			ret = cookie;
+		}
+	}
+	dev_dbg(&host->pdev->dev, "%s(): mapped %d -> %d, cookie %d, rq %p\n",
+		__func__, host->sg_len, ret, cookie, host->mrq);
+
+pio:
+	if (!desc) {
+		/* DMA failed, fall back to PIO */
+		if (ret >= 0)
+			ret = -EIO;
+		host->chan_tx = NULL;
+		dma_release_channel(chan);
+		/* Free the Rx channel too */
+		chan = host->chan_rx;
+		if (chan) {
+			host->chan_rx = NULL;
+			dma_release_channel(chan);
+		}
+		dev_warn(&host->pdev->dev,
+			 "DMA failed: %d, falling back to PIO\n", ret);
+		tmio_mmc_enable_dma(host, false);
+	}
+
+	dev_dbg(&host->pdev->dev, "%s(): desc %p, cookie %d\n", __func__,
+		desc, cookie);
+}
+
+void tmio_mmc_start_dma(struct tmio_mmc_host *host,
+			       struct mmc_data *data)
+{
+	if (data->flags & MMC_DATA_READ) {
+		if (host->chan_rx)
+			tmio_mmc_start_dma_rx(host);
+	} else {
+		if (host->chan_tx)
+			tmio_mmc_start_dma_tx(host);
+	}
+}
+
+static void tmio_mmc_issue_tasklet_fn(unsigned long priv)
+{
+	struct tmio_mmc_host *host = (struct tmio_mmc_host *)priv;
+	struct dma_chan *chan = NULL;
+
+	spin_lock_irq(&host->lock);
+
+	if (host && host->data) {
+		if (host->data->flags & MMC_DATA_READ)
+			chan = host->chan_rx;
+		else
+			chan = host->chan_tx;
+	}
+
+	spin_unlock_irq(&host->lock);
+
+	tmio_mmc_enable_mmc_irqs(host, TMIO_STAT_DATAEND);
+
+	if (chan)
+		dma_async_issue_pending(chan);
+}
+
+static void tmio_mmc_tasklet_fn(unsigned long arg)
+{
+	struct tmio_mmc_host *host = (struct tmio_mmc_host *)arg;
+
+	spin_lock_irq(&host->lock);
+
+	if (!host->data)
+		goto out;
+
+	if (host->data->flags & MMC_DATA_READ)
+		dma_unmap_sg(host->chan_rx->device->dev,
+			     host->sg_ptr, host->sg_len,
+			     DMA_FROM_DEVICE);
+	else
+		dma_unmap_sg(host->chan_tx->device->dev,
+			     host->sg_ptr, host->sg_len,
+			     DMA_TO_DEVICE);
+
+	tmio_mmc_do_data_irq(host);
+out:
+	spin_unlock_irq(&host->lock);
+}
+
+/* It might be necessary to make filter MFD specific */
+static bool tmio_mmc_filter(struct dma_chan *chan, void *arg)
+{
+	dev_dbg(chan->device->dev, "%s: slave data %p\n", __func__, arg);
+	chan->private = arg;
+	return true;
+}
+
+void tmio_mmc_request_dma(struct tmio_mmc_host *host, struct tmio_mmc_data *pdata)
+{
+	/* We can only either use DMA for both Tx and Rx or not use it at all */
+	if (pdata->dma) {
+		dma_cap_mask_t mask;
+
+		dma_cap_zero(mask);
+		dma_cap_set(DMA_SLAVE, mask);
+
+		host->chan_tx = dma_request_channel(mask, tmio_mmc_filter,
+						    pdata->dma->chan_priv_tx);
+		dev_dbg(&host->pdev->dev, "%s: TX: got channel %p\n", __func__,
+			host->chan_tx);
+
+		if (!host->chan_tx)
+			return;
+
+		host->chan_rx = dma_request_channel(mask, tmio_mmc_filter,
+						    pdata->dma->chan_priv_rx);
+		dev_dbg(&host->pdev->dev, "%s: RX: got channel %p\n", __func__,
+			host->chan_rx);
+
+		if (!host->chan_rx)
+			goto ereqrx;
+
+		host->bounce_buf = (u8 *)__get_free_page(GFP_KERNEL | GFP_DMA);
+		if (!host->bounce_buf)
+			goto ebouncebuf;
+
+		tasklet_init(&host->dma_complete, tmio_mmc_tasklet_fn, (unsigned long)host);
+		tasklet_init(&host->dma_issue, tmio_mmc_issue_tasklet_fn, (unsigned long)host);
+
+		tmio_mmc_enable_dma(host, true);
+
+		return;
+ebouncebuf:
+		dma_release_channel(host->chan_rx);
+		host->chan_rx = NULL;
+ereqrx:
+		dma_release_channel(host->chan_tx);
+		host->chan_tx = NULL;
+		return;
+	}
+}
+
+void tmio_mmc_release_dma(struct tmio_mmc_host *host)
+{
+	if (host->chan_tx) {
+		struct dma_chan *chan = host->chan_tx;
+		host->chan_tx = NULL;
+		dma_release_channel(chan);
+	}
+	if (host->chan_rx) {
+		struct dma_chan *chan = host->chan_rx;
+		host->chan_rx = NULL;
+		dma_release_channel(chan);
+	}
+	if (host->bounce_buf) {
+		free_pages((unsigned long)host->bounce_buf, 0);
+		host->bounce_buf = NULL;
+	}
+}
diff --git a/drivers/mmc/host/tmio_mmc_pio.c b/drivers/mmc/host/tmio_mmc_pio.c
new file mode 100644
index 00000000000..62d37de6de7
--- /dev/null
+++ b/drivers/mmc/host/tmio_mmc_pio.c
@@ -0,0 +1,897 @@
+/*
+ * linux/drivers/mmc/host/tmio_mmc_pio.c
+ *
+ * Copyright (C) 2011 Guennadi Liakhovetski
+ * Copyright (C) 2007 Ian Molton
+ * Copyright (C) 2004 Ian Molton
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Driver for the MMC / SD / SDIO IP found in:
+ *
+ * TC6393XB, TC6391XB, TC6387XB, T7L66XB, ASIC3, SH-Mobile SoCs
+ *
+ * This driver draws mainly on scattered spec sheets, Reverse engineering
+ * of the toshiba e800  SD driver and some parts of the 2.4 ASIC3 driver (4 bit
+ * support). (Further 4 bit support from a later datasheet).
+ *
+ * TODO:
+ *   Investigate using a workqueue for PIO transfers
+ *   Eliminate FIXMEs
+ *   SDIO support
+ *   Better Power management
+ *   Handle MMC errors better
+ *   double buffer support
+ *
+ */
+
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <linux/highmem.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/irq.h>
+#include <linux/mfd/tmio.h>
+#include <linux/mmc/host.h>
+#include <linux/mmc/tmio.h>
+#include <linux/module.h>
+#include <linux/pagemap.h>
+#include <linux/platform_device.h>
+#include <linux/scatterlist.h>
+#include <linux/workqueue.h>
+#include <linux/spinlock.h>
+
+#include "tmio_mmc.h"
+
+static u16 sd_ctrl_read16(struct tmio_mmc_host *host, int addr)
+{
+	return readw(host->ctl + (addr << host->bus_shift));
+}
+
+static void sd_ctrl_read16_rep(struct tmio_mmc_host *host, int addr,
+		u16 *buf, int count)
+{
+	readsw(host->ctl + (addr << host->bus_shift), buf, count);
+}
+
+static u32 sd_ctrl_read32(struct tmio_mmc_host *host, int addr)
+{
+	return readw(host->ctl + (addr << host->bus_shift)) |
+	       readw(host->ctl + ((addr + 2) << host->bus_shift)) << 16;
+}
+
+static void sd_ctrl_write16(struct tmio_mmc_host *host, int addr, u16 val)
+{
+	writew(val, host->ctl + (addr << host->bus_shift));
+}
+
+static void sd_ctrl_write16_rep(struct tmio_mmc_host *host, int addr,
+		u16 *buf, int count)
+{
+	writesw(host->ctl + (addr << host->bus_shift), buf, count);
+}
+
+static void sd_ctrl_write32(struct tmio_mmc_host *host, int addr, u32 val)
+{
+	writew(val, host->ctl + (addr << host->bus_shift));
+	writew(val >> 16, host->ctl + ((addr + 2) << host->bus_shift));
+}
+
+void tmio_mmc_enable_mmc_irqs(struct tmio_mmc_host *host, u32 i)
+{
+	u32 mask = sd_ctrl_read32(host, CTL_IRQ_MASK) & ~(i & TMIO_MASK_IRQ);
+	sd_ctrl_write32(host, CTL_IRQ_MASK, mask);
+}
+
+void tmio_mmc_disable_mmc_irqs(struct tmio_mmc_host *host, u32 i)
+{
+	u32 mask = sd_ctrl_read32(host, CTL_IRQ_MASK) | (i & TMIO_MASK_IRQ);
+	sd_ctrl_write32(host, CTL_IRQ_MASK, mask);
+}
+
+static void tmio_mmc_ack_mmc_irqs(struct tmio_mmc_host *host, u32 i)
+{
+	sd_ctrl_write32(host, CTL_STATUS, ~i);
+}
+
+static void tmio_mmc_init_sg(struct tmio_mmc_host *host, struct mmc_data *data)
+{
+	host->sg_len = data->sg_len;
+	host->sg_ptr = data->sg;
+	host->sg_orig = data->sg;
+	host->sg_off = 0;
+}
+
+static int tmio_mmc_next_sg(struct tmio_mmc_host *host)
+{
+	host->sg_ptr = sg_next(host->sg_ptr);
+	host->sg_off = 0;
+	return --host->sg_len;
+}
+
+#ifdef CONFIG_MMC_DEBUG
+
+#define STATUS_TO_TEXT(a, status, i) \
+	do { \
+		if (status & TMIO_STAT_##a) { \
+			if (i++) \
+				printk(" | "); \
+			printk(#a); \
+		} \
+	} while (0)
+
+static void pr_debug_status(u32 status)
+{
+	int i = 0;
+	printk(KERN_DEBUG "status: %08x = ", status);
+	STATUS_TO_TEXT(CARD_REMOVE, status, i);
+	STATUS_TO_TEXT(CARD_INSERT, status, i);
+	STATUS_TO_TEXT(SIGSTATE, status, i);
+	STATUS_TO_TEXT(WRPROTECT, status, i);
+	STATUS_TO_TEXT(CARD_REMOVE_A, status, i);
+	STATUS_TO_TEXT(CARD_INSERT_A, status, i);
+	STATUS_TO_TEXT(SIGSTATE_A, status, i);
+	STATUS_TO_TEXT(CMD_IDX_ERR, status, i);
+	STATUS_TO_TEXT(STOPBIT_ERR, status, i);
+	STATUS_TO_TEXT(ILL_FUNC, status, i);
+	STATUS_TO_TEXT(CMD_BUSY, status, i);
+	STATUS_TO_TEXT(CMDRESPEND, status, i);
+	STATUS_TO_TEXT(DATAEND, status, i);
+	STATUS_TO_TEXT(CRCFAIL, status, i);
+	STATUS_TO_TEXT(DATATIMEOUT, status, i);
+	STATUS_TO_TEXT(CMDTIMEOUT, status, i);
+	STATUS_TO_TEXT(RXOVERFLOW, status, i);
+	STATUS_TO_TEXT(TXUNDERRUN, status, i);
+	STATUS_TO_TEXT(RXRDY, status, i);
+	STATUS_TO_TEXT(TXRQ, status, i);
+	STATUS_TO_TEXT(ILL_ACCESS, status, i);
+	printk("\n");
+}
+
+#else
+#define pr_debug_status(s)  do { } while (0)
+#endif
+
+static void tmio_mmc_enable_sdio_irq(struct mmc_host *mmc, int enable)
+{
+	struct tmio_mmc_host *host = mmc_priv(mmc);
+
+	if (enable) {
+		host->sdio_irq_enabled = 1;
+		sd_ctrl_write16(host, CTL_TRANSACTION_CTL, 0x0001);
+		sd_ctrl_write16(host, CTL_SDIO_IRQ_MASK,
+			(TMIO_SDIO_MASK_ALL & ~TMIO_SDIO_STAT_IOIRQ));
+	} else {
+		sd_ctrl_write16(host, CTL_SDIO_IRQ_MASK, TMIO_SDIO_MASK_ALL);
+		sd_ctrl_write16(host, CTL_TRANSACTION_CTL, 0x0000);
+		host->sdio_irq_enabled = 0;
+	}
+}
+
+static void tmio_mmc_set_clock(struct tmio_mmc_host *host, int new_clock)
+{
+	u32 clk = 0, clock;
+
+	if (new_clock) {
+		for (clock = host->mmc->f_min, clk = 0x80000080;
+			new_clock >= (clock<<1); clk >>= 1)
+			clock <<= 1;
+		clk |= 0x100;
+	}
+
+	if (host->set_clk_div)
+		host->set_clk_div(host->pdev, (clk>>22) & 1);
+
+	sd_ctrl_write16(host, CTL_SD_CARD_CLK_CTL, clk & 0x1ff);
+}
+
+static void tmio_mmc_clk_stop(struct tmio_mmc_host *host)
+{
+	struct resource *res = platform_get_resource(host->pdev, IORESOURCE_MEM, 0);
+
+	/* implicit BUG_ON(!res) */
+	if (resource_size(res) > 0x100) {
+		sd_ctrl_write16(host, CTL_CLK_AND_WAIT_CTL, 0x0000);
+		msleep(10);
+	}
+
+	sd_ctrl_write16(host, CTL_SD_CARD_CLK_CTL, ~0x0100 &
+		sd_ctrl_read16(host, CTL_SD_CARD_CLK_CTL));
+	msleep(10);
+}
+
+static void tmio_mmc_clk_start(struct tmio_mmc_host *host)
+{
+	struct resource *res = platform_get_resource(host->pdev, IORESOURCE_MEM, 0);
+
+	sd_ctrl_write16(host, CTL_SD_CARD_CLK_CTL, 0x0100 |
+		sd_ctrl_read16(host, CTL_SD_CARD_CLK_CTL));
+	msleep(10);
+
+	/* implicit BUG_ON(!res) */
+	if (resource_size(res) > 0x100) {
+		sd_ctrl_write16(host, CTL_CLK_AND_WAIT_CTL, 0x0100);
+		msleep(10);
+	}
+}
+
+static void tmio_mmc_reset(struct tmio_mmc_host *host)
+{
+	struct resource *res = platform_get_resource(host->pdev, IORESOURCE_MEM, 0);
+
+	/* FIXME - should we set stop clock reg here */
+	sd_ctrl_write16(host, CTL_RESET_SD, 0x0000);
+	/* implicit BUG_ON(!res) */
+	if (resource_size(res) > 0x100)
+		sd_ctrl_write16(host, CTL_RESET_SDIO, 0x0000);
+	msleep(10);
+	sd_ctrl_write16(host, CTL_RESET_SD, 0x0001);
+	if (resource_size(res) > 0x100)
+		sd_ctrl_write16(host, CTL_RESET_SDIO, 0x0001);
+	msleep(10);
+}
+
+static void tmio_mmc_reset_work(struct work_struct *work)
+{
+	struct tmio_mmc_host *host = container_of(work, struct tmio_mmc_host,
+						  delayed_reset_work.work);
+	struct mmc_request *mrq;
+	unsigned long flags;
+
+	spin_lock_irqsave(&host->lock, flags);
+	mrq = host->mrq;
+
+	/* request already finished */
+	if (!mrq
+	    || time_is_after_jiffies(host->last_req_ts +
+		msecs_to_jiffies(2000))) {
+		spin_unlock_irqrestore(&host->lock, flags);
+		return;
+	}
+
+	dev_warn(&host->pdev->dev,
+		"timeout waiting for hardware interrupt (CMD%u)\n",
+		mrq->cmd->opcode);
+
+	if (host->data)
+		host->data->error = -ETIMEDOUT;
+	else if (host->cmd)
+		host->cmd->error = -ETIMEDOUT;
+	else
+		mrq->cmd->error = -ETIMEDOUT;
+
+	host->cmd = NULL;
+	host->data = NULL;
+	host->mrq = NULL;
+	host->force_pio = false;
+
+	spin_unlock_irqrestore(&host->lock, flags);
+
+	tmio_mmc_reset(host);
+
+	mmc_request_done(host->mmc, mrq);
+}
+
+static void tmio_mmc_finish_request(struct tmio_mmc_host *host)
+{
+	struct mmc_request *mrq = host->mrq;
+
+	if (!mrq)
+		return;
+
+	host->mrq = NULL;
+	host->cmd = NULL;
+	host->data = NULL;
+	host->force_pio = false;
+
+	cancel_delayed_work(&host->delayed_reset_work);
+
+	mmc_request_done(host->mmc, mrq);
+}
+
+/* These are the bitmasks the tmio chip requires to implement the MMC response
+ * types. Note that R1 and R6 are the same in this scheme. */
+#define APP_CMD        0x0040
+#define RESP_NONE      0x0300
+#define RESP_R1        0x0400
+#define RESP_R1B       0x0500
+#define RESP_R2        0x0600
+#define RESP_R3        0x0700
+#define DATA_PRESENT   0x0800
+#define TRANSFER_READ  0x1000
+#define TRANSFER_MULTI 0x2000
+#define SECURITY_CMD   0x4000
+
+static int tmio_mmc_start_command(struct tmio_mmc_host *host, struct mmc_command *cmd)
+{
+	struct mmc_data *data = host->data;
+	int c = cmd->opcode;
+
+	/* Command 12 is handled by hardware */
+	if (cmd->opcode == 12 && !cmd->arg) {
+		sd_ctrl_write16(host, CTL_STOP_INTERNAL_ACTION, 0x001);
+		return 0;
+	}
+
+	switch (mmc_resp_type(cmd)) {
+	case MMC_RSP_NONE: c |= RESP_NONE; break;
+	case MMC_RSP_R1:   c |= RESP_R1;   break;
+	case MMC_RSP_R1B:  c |= RESP_R1B;  break;
+	case MMC_RSP_R2:   c |= RESP_R2;   break;
+	case MMC_RSP_R3:   c |= RESP_R3;   break;
+	default:
+		pr_debug("Unknown response type %d\n", mmc_resp_type(cmd));
+		return -EINVAL;
+	}
+
+	host->cmd = cmd;
+
+/* FIXME - this seems to be ok commented out but the spec suggest this bit
+ *         should be set when issuing app commands.
+ *	if(cmd->flags & MMC_FLAG_ACMD)
+ *		c |= APP_CMD;
+ */
+	if (data) {
+		c |= DATA_PRESENT;
+		if (data->blocks > 1) {
+			sd_ctrl_write16(host, CTL_STOP_INTERNAL_ACTION, 0x100);
+			c |= TRANSFER_MULTI;
+		}
+		if (data->flags & MMC_DATA_READ)
+			c |= TRANSFER_READ;
+	}
+
+	tmio_mmc_enable_mmc_irqs(host, TMIO_MASK_CMD);
+
+	/* Fire off the command */
+	sd_ctrl_write32(host, CTL_ARG_REG, cmd->arg);
+	sd_ctrl_write16(host, CTL_SD_CMD, c);
+
+	return 0;
+}
+
+/*
+ * This chip always returns (at least?) as much data as you ask for.
+ * I'm unsure what happens if you ask for less than a block. This should be
+ * looked into to ensure that a funny length read doesn't hose the controller.
+ */
+static void tmio_mmc_pio_irq(struct tmio_mmc_host *host)
+{
+	struct mmc_data *data = host->data;
+	void *sg_virt;
+	unsigned short *buf;
+	unsigned int count;
+	unsigned long flags;
+
+	if ((host->chan_tx || host->chan_rx) && !host->force_pio) {
+		pr_err("PIO IRQ in DMA mode!\n");
+		return;
+	} else if (!data) {
+		pr_debug("Spurious PIO IRQ\n");
+		return;
+	}
+
+	sg_virt = tmio_mmc_kmap_atomic(host->sg_ptr, &flags);
+	buf = (unsigned short *)(sg_virt + host->sg_off);
+
+	count = host->sg_ptr->length - host->sg_off;
+	if (count > data->blksz)
+		count = data->blksz;
+
+	pr_debug("count: %08x offset: %08x flags %08x\n",
+		 count, host->sg_off, data->flags);
+
+	/* Transfer the data */
+	if (data->flags & MMC_DATA_READ)
+		sd_ctrl_read16_rep(host, CTL_SD_DATA_PORT, buf, count >> 1);
+	else
+		sd_ctrl_write16_rep(host, CTL_SD_DATA_PORT, buf, count >> 1);
+
+	host->sg_off += count;
+
+	tmio_mmc_kunmap_atomic(host->sg_ptr, &flags, sg_virt);
+
+	if (host->sg_off == host->sg_ptr->length)
+		tmio_mmc_next_sg(host);
+
+	return;
+}
+
+static void tmio_mmc_check_bounce_buffer(struct tmio_mmc_host *host)
+{
+	if (host->sg_ptr == &host->bounce_sg) {
+		unsigned long flags;
+		void *sg_vaddr = tmio_mmc_kmap_atomic(host->sg_orig, &flags);
+		memcpy(sg_vaddr, host->bounce_buf, host->bounce_sg.length);
+		tmio_mmc_kunmap_atomic(host->sg_orig, &flags, sg_vaddr);
+	}
+}
+
+/* needs to be called with host->lock held */
+void tmio_mmc_do_data_irq(struct tmio_mmc_host *host)
+{
+	struct mmc_data *data = host->data;
+	struct mmc_command *stop;
+
+	host->data = NULL;
+
+	if (!data) {
+		dev_warn(&host->pdev->dev, "Spurious data end IRQ\n");
+		return;
+	}
+	stop = data->stop;
+
+	/* FIXME - return correct transfer count on errors */
+	if (!data->error)
+		data->bytes_xfered = data->blocks * data->blksz;
+	else
+		data->bytes_xfered = 0;
+
+	pr_debug("Completed data request\n");
+
+	/*
+	 * FIXME: other drivers allow an optional stop command of any given type
+	 *        which we dont do, as the chip can auto generate them.
+	 *        Perhaps we can be smarter about when to use auto CMD12 and
+	 *        only issue the auto request when we know this is the desired
+	 *        stop command, allowing fallback to the stop command the
+	 *        upper layers expect. For now, we do what works.
+	 */
+
+	if (data->flags & MMC_DATA_READ) {
+		if (host->chan_rx && !host->force_pio)
+			tmio_mmc_check_bounce_buffer(host);
+		dev_dbg(&host->pdev->dev, "Complete Rx request %p\n",
+			host->mrq);
+	} else {
+		dev_dbg(&host->pdev->dev, "Complete Tx request %p\n",
+			host->mrq);
+	}
+
+	if (stop) {
+		if (stop->opcode == 12 && !stop->arg)
+			sd_ctrl_write16(host, CTL_STOP_INTERNAL_ACTION, 0x000);
+		else
+			BUG();
+	}
+
+	tmio_mmc_finish_request(host);
+}
+
+static void tmio_mmc_data_irq(struct tmio_mmc_host *host)
+{
+	struct mmc_data *data;
+	spin_lock(&host->lock);
+	data = host->data;
+
+	if (!data)
+		goto out;
+
+	if (host->chan_tx && (data->flags & MMC_DATA_WRITE) && !host->force_pio) {
+		/*
+		 * Has all data been written out yet? Testing on SuperH showed,
+		 * that in most cases the first interrupt comes already with the
+		 * BUSY status bit clear, but on some operations, like mount or
+		 * in the beginning of a write / sync / umount, there is one
+		 * DATAEND interrupt with the BUSY bit set, in this cases
+		 * waiting for one more interrupt fixes the problem.
+		 */
+		if (!(sd_ctrl_read32(host, CTL_STATUS) & TMIO_STAT_CMD_BUSY)) {
+			tmio_mmc_disable_mmc_irqs(host, TMIO_STAT_DATAEND);
+			tasklet_schedule(&host->dma_complete);
+		}
+	} else if (host->chan_rx && (data->flags & MMC_DATA_READ) && !host->force_pio) {
+		tmio_mmc_disable_mmc_irqs(host, TMIO_STAT_DATAEND);
+		tasklet_schedule(&host->dma_complete);
+	} else {
+		tmio_mmc_do_data_irq(host);
+		tmio_mmc_disable_mmc_irqs(host, TMIO_MASK_READOP | TMIO_MASK_WRITEOP);
+	}
+out:
+	spin_unlock(&host->lock);
+}
+
+static void tmio_mmc_cmd_irq(struct tmio_mmc_host *host,
+	unsigned int stat)
+{
+	struct mmc_command *cmd = host->cmd;
+	int i, addr;
+
+	spin_lock(&host->lock);
+
+	if (!host->cmd) {
+		pr_debug("Spurious CMD irq\n");
+		goto out;
+	}
+
+	host->cmd = NULL;
+
+	/* This controller is sicker than the PXA one. Not only do we need to
+	 * drop the top 8 bits of the first response word, we also need to
+	 * modify the order of the response for short response command types.
+	 */
+
+	for (i = 3, addr = CTL_RESPONSE ; i >= 0 ; i--, addr += 4)
+		cmd->resp[i] = sd_ctrl_read32(host, addr);
+
+	if (cmd->flags &  MMC_RSP_136) {
+		cmd->resp[0] = (cmd->resp[0] << 8) | (cmd->resp[1] >> 24);
+		cmd->resp[1] = (cmd->resp[1] << 8) | (cmd->resp[2] >> 24);
+		cmd->resp[2] = (cmd->resp[2] << 8) | (cmd->resp[3] >> 24);
+		cmd->resp[3] <<= 8;
+	} else if (cmd->flags & MMC_RSP_R3) {
+		cmd->resp[0] = cmd->resp[3];
+	}
+
+	if (stat & TMIO_STAT_CMDTIMEOUT)
+		cmd->error = -ETIMEDOUT;
+	else if (stat & TMIO_STAT_CRCFAIL && cmd->flags & MMC_RSP_CRC)
+		cmd->error = -EILSEQ;
+
+	/* If there is data to handle we enable data IRQs here, and
+	 * we will ultimatley finish the request in the data_end handler.
+	 * If theres no data or we encountered an error, finish now.
+	 */
+	if (host->data && !cmd->error) {
+		if (host->data->flags & MMC_DATA_READ) {
+			if (host->force_pio || !host->chan_rx)
+				tmio_mmc_enable_mmc_irqs(host, TMIO_MASK_READOP);
+			else
+				tasklet_schedule(&host->dma_issue);
+		} else {
+			if (host->force_pio || !host->chan_tx)
+				tmio_mmc_enable_mmc_irqs(host, TMIO_MASK_WRITEOP);
+			else
+				tasklet_schedule(&host->dma_issue);
+		}
+	} else {
+		tmio_mmc_finish_request(host);
+	}
+
+out:
+	spin_unlock(&host->lock);
+}
+
+static irqreturn_t tmio_mmc_irq(int irq, void *devid)
+{
+	struct tmio_mmc_host *host = devid;
+	struct tmio_mmc_data *pdata = host->pdata;
+	unsigned int ireg, irq_mask, status;
+	unsigned int sdio_ireg, sdio_irq_mask, sdio_status;
+
+	pr_debug("MMC IRQ begin\n");
+
+	status = sd_ctrl_read32(host, CTL_STATUS);
+	irq_mask = sd_ctrl_read32(host, CTL_IRQ_MASK);
+	ireg = status & TMIO_MASK_IRQ & ~irq_mask;
+
+	sdio_ireg = 0;
+	if (!ireg && pdata->flags & TMIO_MMC_SDIO_IRQ) {
+		sdio_status = sd_ctrl_read16(host, CTL_SDIO_STATUS);
+		sdio_irq_mask = sd_ctrl_read16(host, CTL_SDIO_IRQ_MASK);
+		sdio_ireg = sdio_status & TMIO_SDIO_MASK_ALL & ~sdio_irq_mask;
+
+		sd_ctrl_write16(host, CTL_SDIO_STATUS, sdio_status & ~TMIO_SDIO_MASK_ALL);
+
+		if (sdio_ireg && !host->sdio_irq_enabled) {
+			pr_warning("tmio_mmc: Spurious SDIO IRQ, disabling! 0x%04x 0x%04x 0x%04x\n",
+				   sdio_status, sdio_irq_mask, sdio_ireg);
+			tmio_mmc_enable_sdio_irq(host->mmc, 0);
+			goto out;
+		}
+
+		if (host->mmc->caps & MMC_CAP_SDIO_IRQ &&
+			sdio_ireg & TMIO_SDIO_STAT_IOIRQ)
+			mmc_signal_sdio_irq(host->mmc);
+
+		if (sdio_ireg)
+			goto out;
+	}
+
+	pr_debug_status(status);
+	pr_debug_status(ireg);
+
+	if (!ireg) {
+		tmio_mmc_disable_mmc_irqs(host, status & ~irq_mask);
+
+		pr_warning("tmio_mmc: Spurious irq, disabling! "
+			"0x%08x 0x%08x 0x%08x\n", status, irq_mask, ireg);
+		pr_debug_status(status);
+
+		goto out;
+	}
+
+	while (ireg) {
+		/* Card insert / remove attempts */
+		if (ireg & (TMIO_STAT_CARD_INSERT | TMIO_STAT_CARD_REMOVE)) {
+			tmio_mmc_ack_mmc_irqs(host, TMIO_STAT_CARD_INSERT |
+				TMIO_STAT_CARD_REMOVE);
+			mmc_detect_change(host->mmc, msecs_to_jiffies(100));
+		}
+
+		/* CRC and other errors */
+/*		if (ireg & TMIO_STAT_ERR_IRQ)
+ *			handled |= tmio_error_irq(host, irq, stat);
+ */
+
+		/* Command completion */
+		if (ireg & (TMIO_STAT_CMDRESPEND | TMIO_STAT_CMDTIMEOUT)) {
+			tmio_mmc_ack_mmc_irqs(host,
+				     TMIO_STAT_CMDRESPEND |
+				     TMIO_STAT_CMDTIMEOUT);
+			tmio_mmc_cmd_irq(host, status);
+		}
+
+		/* Data transfer */
+		if (ireg & (TMIO_STAT_RXRDY | TMIO_STAT_TXRQ)) {
+			tmio_mmc_ack_mmc_irqs(host, TMIO_STAT_RXRDY | TMIO_STAT_TXRQ);
+			tmio_mmc_pio_irq(host);
+		}
+
+		/* Data transfer completion */
+		if (ireg & TMIO_STAT_DATAEND) {
+			tmio_mmc_ack_mmc_irqs(host, TMIO_STAT_DATAEND);
+			tmio_mmc_data_irq(host);
+		}
+
+		/* Check status - keep going until we've handled it all */
+		status = sd_ctrl_read32(host, CTL_STATUS);
+		irq_mask = sd_ctrl_read32(host, CTL_IRQ_MASK);
+		ireg = status & TMIO_MASK_IRQ & ~irq_mask;
+
+		pr_debug("Status at end of loop: %08x\n", status);
+		pr_debug_status(status);
+	}
+	pr_debug("MMC IRQ end\n");
+
+out:
+	return IRQ_HANDLED;
+}
+
+static int tmio_mmc_start_data(struct tmio_mmc_host *host,
+	struct mmc_data *data)
+{
+	struct tmio_mmc_data *pdata = host->pdata;
+
+	pr_debug("setup data transfer: blocksize %08x  nr_blocks %d\n",
+		 data->blksz, data->blocks);
+
+	/* Some hardware cannot perform 2 byte requests in 4 bit mode */
+	if (host->mmc->ios.bus_width == MMC_BUS_WIDTH_4) {
+		int blksz_2bytes = pdata->flags & TMIO_MMC_BLKSZ_2BYTES;
+
+		if (data->blksz < 2 || (data->blksz < 4 && !blksz_2bytes)) {
+			pr_err("%s: %d byte block unsupported in 4 bit mode\n",
+			       mmc_hostname(host->mmc), data->blksz);
+			return -EINVAL;
+		}
+	}
+
+	tmio_mmc_init_sg(host, data);
+	host->data = data;
+
+	/* Set transfer length / blocksize */
+	sd_ctrl_write16(host, CTL_SD_XFER_LEN, data->blksz);
+	sd_ctrl_write16(host, CTL_XFER_BLK_COUNT, data->blocks);
+
+	tmio_mmc_start_dma(host, data);
+
+	return 0;
+}
+
+/* Process requests from the MMC layer */
+static void tmio_mmc_request(struct mmc_host *mmc, struct mmc_request *mrq)
+{
+	struct tmio_mmc_host *host = mmc_priv(mmc);
+	int ret;
+
+	if (host->mrq)
+		pr_debug("request not null\n");
+
+	host->last_req_ts = jiffies;
+	wmb();
+	host->mrq = mrq;
+
+	if (mrq->data) {
+		ret = tmio_mmc_start_data(host, mrq->data);
+		if (ret)
+			goto fail;
+	}
+
+	ret = tmio_mmc_start_command(host, mrq->cmd);
+	if (!ret) {
+		schedule_delayed_work(&host->delayed_reset_work,
+				      msecs_to_jiffies(2000));
+		return;
+	}
+
+fail:
+	host->mrq = NULL;
+	host->force_pio = false;
+	mrq->cmd->error = ret;
+	mmc_request_done(mmc, mrq);
+}
+
+/* Set MMC clock / power.
+ * Note: This controller uses a simple divider scheme therefore it cannot
+ * run a MMC card at full speed (20MHz). The max clock is 24MHz on SD, but as
+ * MMC wont run that fast, it has to be clocked at 12MHz which is the next
+ * slowest setting.
+ */
+static void tmio_mmc_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
+{
+	struct tmio_mmc_host *host = mmc_priv(mmc);
+
+	if (ios->clock)
+		tmio_mmc_set_clock(host, ios->clock);
+
+	/* Power sequence - OFF -> UP -> ON */
+	if (ios->power_mode == MMC_POWER_OFF || !ios->clock) {
+		/* power down SD bus */
+		if (ios->power_mode == MMC_POWER_OFF && host->set_pwr)
+			host->set_pwr(host->pdev, 0);
+		tmio_mmc_clk_stop(host);
+	} else if (ios->power_mode == MMC_POWER_UP) {
+		/* power up SD bus */
+		if (host->set_pwr)
+			host->set_pwr(host->pdev, 1);
+	} else {
+		/* start bus clock */
+		tmio_mmc_clk_start(host);
+	}
+
+	switch (ios->bus_width) {
+	case MMC_BUS_WIDTH_1:
+		sd_ctrl_write16(host, CTL_SD_MEM_CARD_OPT, 0x80e0);
+	break;
+	case MMC_BUS_WIDTH_4:
+		sd_ctrl_write16(host, CTL_SD_MEM_CARD_OPT, 0x00e0);
+	break;
+	}
+
+	/* Let things settle. delay taken from winCE driver */
+	udelay(140);
+}
+
+static int tmio_mmc_get_ro(struct mmc_host *mmc)
+{
+	struct tmio_mmc_host *host = mmc_priv(mmc);
+	struct tmio_mmc_data *pdata = host->pdata;
+
+	return ((pdata->flags & TMIO_MMC_WRPROTECT_DISABLE) ||
+		!(sd_ctrl_read32(host, CTL_STATUS) & TMIO_STAT_WRPROTECT));
+}
+
+static int tmio_mmc_get_cd(struct mmc_host *mmc)
+{
+	struct tmio_mmc_host *host = mmc_priv(mmc);
+	struct tmio_mmc_data *pdata = host->pdata;
+
+	if (!pdata->get_cd)
+		return -ENOSYS;
+	else
+		return pdata->get_cd(host->pdev);
+}
+
+static const struct mmc_host_ops tmio_mmc_ops = {
+	.request	= tmio_mmc_request,
+	.set_ios	= tmio_mmc_set_ios,
+	.get_ro         = tmio_mmc_get_ro,
+	.get_cd		= tmio_mmc_get_cd,
+	.enable_sdio_irq = tmio_mmc_enable_sdio_irq,
+};
+
+int __devinit tmio_mmc_host_probe(struct tmio_mmc_host **host,
+				  struct platform_device *pdev,
+				  struct tmio_mmc_data *pdata)
+{
+	struct tmio_mmc_host *_host;
+	struct mmc_host *mmc;
+	struct resource *res_ctl;
+	int ret;
+	u32 irq_mask = TMIO_MASK_CMD;
+
+	res_ctl = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res_ctl)
+		return -EINVAL;
+
+	mmc = mmc_alloc_host(sizeof(struct tmio_mmc_host), &pdev->dev);
+	if (!mmc)
+		return -ENOMEM;
+
+	_host = mmc_priv(mmc);
+	_host->pdata = pdata;
+	_host->mmc = mmc;
+	_host->pdev = pdev;
+	platform_set_drvdata(pdev, mmc);
+
+	_host->set_pwr = pdata->set_pwr;
+	_host->set_clk_div = pdata->set_clk_div;
+
+	/* SD control register space size is 0x200, 0x400 for bus_shift=1 */
+	_host->bus_shift = resource_size(res_ctl) >> 10;
+
+	_host->ctl = ioremap(res_ctl->start, resource_size(res_ctl));
+	if (!_host->ctl) {
+		ret = -ENOMEM;
+		goto host_free;
+	}
+
+	mmc->ops = &tmio_mmc_ops;
+	mmc->caps = MMC_CAP_4_BIT_DATA | pdata->capabilities;
+	mmc->f_max = pdata->hclk;
+	mmc->f_min = mmc->f_max / 512;
+	mmc->max_segs = 32;
+	mmc->max_blk_size = 512;
+	mmc->max_blk_count = (PAGE_CACHE_SIZE / mmc->max_blk_size) *
+		mmc->max_segs;
+	mmc->max_req_size = mmc->max_blk_size * mmc->max_blk_count;
+	mmc->max_seg_size = mmc->max_req_size;
+	if (pdata->ocr_mask)
+		mmc->ocr_avail = pdata->ocr_mask;
+	else
+		mmc->ocr_avail = MMC_VDD_32_33 | MMC_VDD_33_34;
+
+	tmio_mmc_clk_stop(_host);
+	tmio_mmc_reset(_host);
+
+	ret = platform_get_irq(pdev, 0);
+	if (ret < 0)
+		goto unmap_ctl;
+
+	_host->irq = ret;
+
+	tmio_mmc_disable_mmc_irqs(_host, TMIO_MASK_ALL);
+	if (pdata->flags & TMIO_MMC_SDIO_IRQ)
+		tmio_mmc_enable_sdio_irq(mmc, 0);
+
+	ret = request_irq(_host->irq, tmio_mmc_irq, IRQF_DISABLED |
+		IRQF_TRIGGER_FALLING, dev_name(&pdev->dev), _host);
+	if (ret)
+		goto unmap_ctl;
+
+	spin_lock_init(&_host->lock);
+
+	/* Init delayed work for request timeouts */
+	INIT_DELAYED_WORK(&_host->delayed_reset_work, tmio_mmc_reset_work);
+
+	/* See if we also get DMA */
+	tmio_mmc_request_dma(_host, pdata);
+
+	mmc_add_host(mmc);
+
+	/* Unmask the IRQs we want to know about */
+	if (!_host->chan_rx)
+		irq_mask |= TMIO_MASK_READOP;
+	if (!_host->chan_tx)
+		irq_mask |= TMIO_MASK_WRITEOP;
+
+	tmio_mmc_enable_mmc_irqs(_host, irq_mask);
+
+	*host = _host;
+
+	return 0;
+
+unmap_ctl:
+	iounmap(_host->ctl);
+host_free:
+	mmc_free_host(mmc);
+
+	return ret;
+}
+EXPORT_SYMBOL(tmio_mmc_host_probe);
+
+void tmio_mmc_host_remove(struct tmio_mmc_host *host)
+{
+	mmc_remove_host(host->mmc);
+	cancel_delayed_work_sync(&host->delayed_reset_work);
+	tmio_mmc_release_dma(host);
+	free_irq(host->irq, host);
+	iounmap(host->ctl);
+	mmc_free_host(host->mmc);
+}
+EXPORT_SYMBOL(tmio_mmc_host_remove);
+
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/mmc/host/via-sdmmc.c b/drivers/mmc/host/via-sdmmc.c
index 9ed84ddb478..4dfe2c02ea9 100644
--- a/drivers/mmc/host/via-sdmmc.c
+++ b/drivers/mmc/host/via-sdmmc.c
@@ -802,12 +802,9 @@ static const struct mmc_host_ops via_sdc_ops = {
 
 static void via_reset_pcictrl(struct via_crdr_mmc_host *host)
 {
-	void __iomem *addrbase;
 	unsigned long flags;
 	u8 gatt;
 
-	addrbase = host->pcictrl_mmiobase;
-
 	spin_lock_irqsave(&host->lock, flags);
 
 	via_save_pcictrlreg(host);
@@ -1090,14 +1087,13 @@ static int __devinit via_sd_probe(struct pci_dev *pcidev,
 	struct mmc_host *mmc;
 	struct via_crdr_mmc_host *sdhost;
 	u32 base, len;
-	u8 rev, gatt;
+	u8  gatt;
 	int ret;
 
-	pci_read_config_byte(pcidev, PCI_CLASS_REVISION, &rev);
 	pr_info(DRV_NAME
 		": VIA SDMMC controller found at %s [%04x:%04x] (rev %x)\n",
 		pci_name(pcidev), (int)pcidev->vendor, (int)pcidev->device,
-		(int)rev);
+		(int)pcidev->revision);
 
 	ret = pci_enable_device(pcidev);
 	if (ret)
diff --git a/drivers/mmc/host/wbsd.c b/drivers/mmc/host/wbsd.c
index 7fca0a386ba..62e5a4d171e 100644
--- a/drivers/mmc/host/wbsd.c
+++ b/drivers/mmc/host/wbsd.c
@@ -484,7 +484,7 @@ static void wbsd_fill_fifo(struct wbsd_host *host)
 
 	/*
 	 * Check that we aren't being called after the
-	 * entire buffer has been transfered.
+	 * entire buffer has been transferred.
 	 */
 	if (host->num_sg == 0)
 		return;
@@ -828,7 +828,7 @@ static void wbsd_request(struct mmc_host *mmc, struct mmc_request *mrq)
 	/*
 	 * If this is a data transfer the request
 	 * will be finished after the data has
-	 * transfered.
+	 * transferred.
 	 */
 	if (cmd->data && !cmd->error) {
 		/*
@@ -904,7 +904,7 @@ static void wbsd_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 			setup &= ~WBSD_DAT3_H;
 
 			/*
-			 * We cannot resume card detection immediatly
+			 * We cannot resume card detection immediately
 			 * because of capacitance and delays in the chip.
 			 */
 			mod_timer(&host->ignore_timer, jiffies + HZ / 100);