All of lore.kernel.org
 help / color / mirror / Atom feed
From: Stephen Boyd <sboyd@codeaurora.org>
To: Mark Brown <broonie@kernel.org>
Cc: Samuel Ortiz <sameo@linux.intel.com>,
	Lee Jones <lee.jones@linaro.org>,
	Srinivas Ramana <sramana@codeaurora.org>,
	linux-kernel@vger.kernel.org, linux-arm-msm@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH 3/8] regmap: Add support for using regmap over ssbi
Date: Fri, 13 Dec 2013 13:37:07 -0800	[thread overview]
Message-ID: <20131213213707.GO21983@codeaurora.org> (raw)
In-Reply-To: <20131213104128.GT11044@sirena.org.uk>

On 12/13, Mark Brown wrote:
> On Thu, Dec 12, 2013 at 03:13:01PM -0800, Stephen Boyd wrote:
> 
> > > bulk_read() should decay to individual reads if there isn't a block
> > > operaton and it's not like the hardware actually supports bulk reads
> > > anyway.
> 
> > So regmap_bulk_read() should work if I don't have a map->bus? To make it
> > work with reg_read/write I had to do this. I'm not sure how to make
> > bulk_write work.

I came up with this (possibly ugly) patch. It works for my
specific case but I'm not sure if unpacking the val bits into an
unsigned int and passing that to _regmap_write() is sane. What do
you think?

---8<----
diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c
index 1ccd61b..333c059 100644
--- a/drivers/base/regmap/regmap.c
+++ b/drivers/base/regmap/regmap.c
@@ -1514,11 +1514,9 @@ int regmap_bulk_write(struct regmap *map, unsigned int reg, const void *val,
 {
 	int ret = 0, i;
 	size_t val_bytes = map->format.val_bytes;
-	void *wval;
+	void *wval = NULL;
 
-	if (!map->bus)
-		return -EINVAL;
-	if (!map->format.parse_inplace)
+	if (map->bus && !map->format.parse_inplace)
 		return -EINVAL;
 	if (reg % map->reg_stride)
 		return -EINVAL;
@@ -1528,7 +1526,7 @@ int regmap_bulk_write(struct regmap *map, unsigned int reg, const void *val,
 	/* No formatting is require if val_byte is 1 */
 	if (val_bytes == 1) {
 		wval = (void *)val;
-	} else {
+	} else if (map->bus) {
 		wval = kmemdup(val, val_count * val_bytes, GFP_KERNEL);
 		if (!wval) {
 			ret = -ENOMEM;
@@ -1538,24 +1536,37 @@ int regmap_bulk_write(struct regmap *map, unsigned int reg, const void *val,
 		for (i = 0; i < val_count * val_bytes; i += val_bytes)
 			map->format.parse_inplace(wval + i);
 	}
-	/*
-	 * Some devices does not support bulk write, for
-	 * them we have a series of single write operations.
-	 */
-	if (map->use_single_rw) {
-		for (i = 0; i < val_count; i++) {
-			ret = _regmap_raw_write(map,
+
+	if (map->bus) {
+		/*
+		 * Some devices does not support bulk write, for
+		 * them we have a series of single write operations.
+		 */
+		if (map->use_single_rw) {
+			for (i = 0; i < val_count; i++) {
+				ret = _regmap_raw_write(map,
 						reg + (i * map->reg_stride),
 						val + (i * val_bytes),
 						val_bytes);
-			if (ret != 0)
-				return ret;
+				if (ret != 0)
+					return ret;
+			}
+		} else {
+			ret = _regmap_raw_write(map, reg, wval,
+						val_bytes * val_count);
 		}
 	} else {
-		ret = _regmap_raw_write(map, reg, wval, val_bytes * val_count);
+		for (i = 0; i < val_count; i++) {
+			unsigned int ival;
+			ival = *(unsigned int *)(val + (i * val_bytes));
+			ret = _regmap_write(map, reg + (i * map->reg_stride),
+					ival);
+			if (ret != 0)
+				goto out;
+		}
 	}
 
-	if (val_bytes != 1)
+	if (val_bytes != 1 && map->bus)
 		kfree(wval);
 
 out:

-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
hosted by The Linux Foundation

WARNING: multiple messages have this Message-ID (diff)
From: sboyd@codeaurora.org (Stephen Boyd)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 3/8] regmap: Add support for using regmap over ssbi
Date: Fri, 13 Dec 2013 13:37:07 -0800	[thread overview]
Message-ID: <20131213213707.GO21983@codeaurora.org> (raw)
In-Reply-To: <20131213104128.GT11044@sirena.org.uk>

On 12/13, Mark Brown wrote:
> On Thu, Dec 12, 2013 at 03:13:01PM -0800, Stephen Boyd wrote:
> 
> > > bulk_read() should decay to individual reads if there isn't a block
> > > operaton and it's not like the hardware actually supports bulk reads
> > > anyway.
> 
> > So regmap_bulk_read() should work if I don't have a map->bus? To make it
> > work with reg_read/write I had to do this. I'm not sure how to make
> > bulk_write work.

I came up with this (possibly ugly) patch. It works for my
specific case but I'm not sure if unpacking the val bits into an
unsigned int and passing that to _regmap_write() is sane. What do
you think?

---8<----
diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c
index 1ccd61b..333c059 100644
--- a/drivers/base/regmap/regmap.c
+++ b/drivers/base/regmap/regmap.c
@@ -1514,11 +1514,9 @@ int regmap_bulk_write(struct regmap *map, unsigned int reg, const void *val,
 {
 	int ret = 0, i;
 	size_t val_bytes = map->format.val_bytes;
-	void *wval;
+	void *wval = NULL;
 
-	if (!map->bus)
-		return -EINVAL;
-	if (!map->format.parse_inplace)
+	if (map->bus && !map->format.parse_inplace)
 		return -EINVAL;
 	if (reg % map->reg_stride)
 		return -EINVAL;
@@ -1528,7 +1526,7 @@ int regmap_bulk_write(struct regmap *map, unsigned int reg, const void *val,
 	/* No formatting is require if val_byte is 1 */
 	if (val_bytes == 1) {
 		wval = (void *)val;
-	} else {
+	} else if (map->bus) {
 		wval = kmemdup(val, val_count * val_bytes, GFP_KERNEL);
 		if (!wval) {
 			ret = -ENOMEM;
@@ -1538,24 +1536,37 @@ int regmap_bulk_write(struct regmap *map, unsigned int reg, const void *val,
 		for (i = 0; i < val_count * val_bytes; i += val_bytes)
 			map->format.parse_inplace(wval + i);
 	}
-	/*
-	 * Some devices does not support bulk write, for
-	 * them we have a series of single write operations.
-	 */
-	if (map->use_single_rw) {
-		for (i = 0; i < val_count; i++) {
-			ret = _regmap_raw_write(map,
+
+	if (map->bus) {
+		/*
+		 * Some devices does not support bulk write, for
+		 * them we have a series of single write operations.
+		 */
+		if (map->use_single_rw) {
+			for (i = 0; i < val_count; i++) {
+				ret = _regmap_raw_write(map,
 						reg + (i * map->reg_stride),
 						val + (i * val_bytes),
 						val_bytes);
-			if (ret != 0)
-				return ret;
+				if (ret != 0)
+					return ret;
+			}
+		} else {
+			ret = _regmap_raw_write(map, reg, wval,
+						val_bytes * val_count);
 		}
 	} else {
-		ret = _regmap_raw_write(map, reg, wval, val_bytes * val_count);
+		for (i = 0; i < val_count; i++) {
+			unsigned int ival;
+			ival = *(unsigned int *)(val + (i * val_bytes));
+			ret = _regmap_write(map, reg + (i * map->reg_stride),
+					ival);
+			if (ret != 0)
+				goto out;
+		}
 	}
 
-	if (val_bytes != 1)
+	if (val_bytes != 1 && map->bus)
 		kfree(wval);
 
 out:

-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
hosted by The Linux Foundation

  parent reply	other threads:[~2013-12-13 21:37 UTC|newest]

Thread overview: 77+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-12-10 23:35 [PATCH 0/8] Modernize pm8921 with irqdomains + regmap Stephen Boyd
2013-12-10 23:35 ` Stephen Boyd
2013-12-10 23:35 ` [PATCH 1/8] mfd: ssbi: Remove platform data structs and hide ssbi type enum Stephen Boyd
2013-12-10 23:35   ` Stephen Boyd
2013-12-11  9:27   ` Lee Jones
2013-12-11  9:27     ` Lee Jones
2013-12-10 23:35 ` [PATCH 2/8] mfd: ssbi: Constify buffer in ssbi_write Stephen Boyd
2013-12-10 23:35   ` Stephen Boyd
2013-12-11  9:28   ` Lee Jones
2013-12-11  9:28     ` Lee Jones
2013-12-10 23:35 ` [PATCH 3/8] regmap: Add support for using regmap over ssbi Stephen Boyd
2013-12-10 23:35   ` Stephen Boyd
2013-12-10 23:50   ` Mark Brown
2013-12-10 23:50     ` Mark Brown
2013-12-11  0:13     ` Stephen Boyd
2013-12-11  0:13       ` Stephen Boyd
2013-12-11  0:51       ` Mark Brown
2013-12-11  0:51         ` Mark Brown
2013-12-11  1:32         ` Stephen Boyd
2013-12-11  1:32           ` Stephen Boyd
2013-12-11 13:27           ` Mark Brown
2013-12-11 13:27             ` Mark Brown
2013-12-12 23:13             ` Stephen Boyd
2013-12-12 23:13               ` Stephen Boyd
2013-12-13 10:41               ` Mark Brown
2013-12-13 10:41                 ` Mark Brown
2013-12-13 17:14                 ` [PATCH] regmap: Allow regmap_bulk_read() to work for "no-bus" regmaps Stephen Boyd
2013-12-13 17:14                   ` Stephen Boyd
2013-12-13 17:14                   ` Stephen Boyd
2013-12-16 20:57                   ` Mark Brown
2013-12-16 20:57                     ` Mark Brown
2013-12-13 21:37                 ` Stephen Boyd [this message]
2013-12-13 21:37                   ` [PATCH 3/8] regmap: Add support for using regmap over ssbi Stephen Boyd
2013-12-16 21:01                   ` Mark Brown
2013-12-16 21:01                     ` Mark Brown
2013-12-17  2:30                     ` [PATCH] regmap: Allow regmap_bulk_write() to work for "no-bus" regmaps Stephen Boyd
2013-12-17  2:30                       ` Stephen Boyd
2013-12-18 18:45                       ` Mark Brown
2013-12-18 18:45                         ` Mark Brown
2013-12-23 20:05                         ` Stephen Boyd
2013-12-23 20:05                           ` Stephen Boyd
2013-12-24 12:53                           ` Mark Brown
2013-12-24 12:53                             ` Mark Brown
2013-12-26 19:34                             ` Stephen Boyd
2013-12-26 19:34                               ` Stephen Boyd
2013-12-26 21:52                               ` [PATCH v2] " Stephen Boyd
2013-12-26 21:52                                 ` Stephen Boyd
2013-12-30 12:42                                 ` Mark Brown
2013-12-30 12:42                                   ` Mark Brown
2013-12-10 23:35 ` [PATCH 4/8] mfd: ssbi: Mark match table const Stephen Boyd
2013-12-10 23:35   ` Stephen Boyd
2013-12-11  9:29   ` Lee Jones
2013-12-11  9:29     ` Lee Jones
2013-12-10 23:35 ` [PATCH 5/8] mfd: Move pm8xxx-irq.c contents into only driver that uses it Stephen Boyd
2013-12-10 23:35   ` Stephen Boyd
2013-12-11  9:35   ` Lee Jones
2013-12-11  9:35     ` Lee Jones
2013-12-12 19:06     ` Stephen Boyd
2013-12-12 19:06       ` Stephen Boyd
2013-12-10 23:35 ` [PATCH 6/8] mfd: pm8921: Update for genirq changes Stephen Boyd
2013-12-10 23:35   ` Stephen Boyd
2013-12-11  9:48   ` Lee Jones
2013-12-11  9:48     ` Lee Jones
2013-12-10 23:35 ` [PATCH 7/8] mfd: pm8921: Migrate to irqdomains Stephen Boyd
2013-12-10 23:35   ` Stephen Boyd
2013-12-11  9:53   ` Lee Jones
2013-12-11  9:53     ` Lee Jones
2013-12-11 21:30   ` Courtney Cavin
2013-12-11 21:30     ` Courtney Cavin
2013-12-11 21:30     ` Courtney Cavin
2013-12-12 19:05     ` Stephen Boyd
2013-12-12 19:05       ` Stephen Boyd
2013-12-12 19:05       ` Stephen Boyd
2013-12-10 23:35 ` [PATCH 8/8] mfd: pm8921: Use ssbi regmap Stephen Boyd
2013-12-10 23:35   ` Stephen Boyd
2013-12-11  9:55   ` Lee Jones
2013-12-11  9:55     ` Lee Jones

Reply instructions:

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

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

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

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

  git send-email \
    --in-reply-to=20131213213707.GO21983@codeaurora.org \
    --to=sboyd@codeaurora.org \
    --cc=broonie@kernel.org \
    --cc=lee.jones@linaro.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sameo@linux.intel.com \
    --cc=sramana@codeaurora.org \
    /path/to/YOUR_REPLY

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

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.