[PATCH RFC] scsi: target: tcmu: optimize use of flush_dcache_page

[PATCH RFC] scsi: target: tcmu: optimize use of flush_dcache_page

[Bodo Stroesser]

(scatter|gather)_data_area() need to flush dcache after
writing data to or before reading data from a page in
uio data area.
The two routines are able to handle data transfer to/from
such a page in fragments and flush the cache after each
fragment is copied by calling the wrapper
tcmu_flush_dcache_range().

That means:
1) flush_dcache_page() can be called multiple times for
   the same page.
2) Calling flush_dcache_page() indirectly using the
   wrapper does not make sense, because each call of the
   wrapper is for one single page only and the calling
   routine already has the correct page pointer.

Therefore I changed (scatter|gather)_data_area() such,
that instead of calling tcmu_flush_dcache_range()
before/after each memcpy, it now calls flush_dcache_page()
before unmapping a page (when writing is complete for
that page) or after mapping a page (when starting to read
the page).

Signed-off-by: Bodo Stroesser <bstroesser@ts.fujitsu.com>
---

Since I have x86 machines only, where flush_dcache_page()
is empty, this patch indeed is compile tested only.
Therefore it needs thorough review or ideally a test on
an architecture that has a non empty flush_dcache_page().


 drivers/target/target_core_user.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/target/target_core_user.c b/drivers/target/target_core_user.c
index cdb4848d23c6..d922a7087c31 100644
--- a/drivers/target/target_core_user.c
+++ b/drivers/target/target_core_user.c
@@ -676,8 +676,10 @@ static void scatter_data_area(struct tcmu_dev *udev,
 		from = kmap_atomic(sg_page(sg)) + sg->offset;
 		while (sg_remaining > 0) {
 			if (block_remaining = 0) {
-				if (to)
+				if (to) {
+					flush_dcache_page(page);
 					kunmap_atomic(to);
+				}
 
 				block_remaining = DATA_BLOCK_SIZE;
 				dbi = tcmu_cmd_get_dbi(tcmu_cmd);
@@ -722,7 +724,6 @@ static void scatter_data_area(struct tcmu_dev *udev,
 				memcpy(to + offset,
 				       from + sg->length - sg_remaining,
 				       copy_bytes);
-				tcmu_flush_dcache_range(to, copy_bytes);
 			}
 
 			sg_remaining -= copy_bytes;
@@ -731,8 +732,10 @@ static void scatter_data_area(struct tcmu_dev *udev,
 		kunmap_atomic(from - sg->offset);
 	}
 
-	if (to)
+	if (to) {
+		flush_dcache_page(page);
 		kunmap_atomic(to);
+	}
 }
 
 static void gather_data_area(struct tcmu_dev *udev, struct tcmu_cmd *cmd,
@@ -778,13 +781,13 @@ static void gather_data_area(struct tcmu_dev *udev, struct tcmu_cmd *cmd,
 				dbi = tcmu_cmd_get_dbi(cmd);
 				page = tcmu_get_block_page(udev, dbi);
 				from = kmap_atomic(page);
+				flush_dcache_page(page);
 			}
 			copy_bytes = min_t(size_t, sg_remaining,
 					block_remaining);
 			if (read_len < copy_bytes)
 				copy_bytes = read_len;
 			offset = DATA_BLOCK_SIZE - block_remaining;
-			tcmu_flush_dcache_range(from, copy_bytes);
 			memcpy(to + sg->length - sg_remaining, from + offset,
 					copy_bytes);
 
-- 
2.12.3

[PATCH RFC] scsi: target: tcmu: optimize use of flush_dcache_page

[Bodo Stroesser]

(scatter|gather)_data_area() need to flush dcache after
writing data to or before reading data from a page in
uio data area.
The two routines are able to handle data transfer to/from
such a page in fragments and flush the cache after each
fragment is copied by calling the wrapper
tcmu_flush_dcache_range().

That means:
1) flush_dcache_page() can be called multiple times for
   the same page.
2) Calling flush_dcache_page() indirectly using the
   wrapper does not make sense, because each call of the
   wrapper is for one single page only and the calling
   routine already has the correct page pointer.

Therefore I changed (scatter|gather)_data_area() such,
that instead of calling tcmu_flush_dcache_range()
before/after each memcpy, it now calls flush_dcache_page()
before unmapping a page (when writing is complete for
that page) or after mapping a page (when starting to read
the page).

Signed-off-by: Bodo Stroesser <bstroesser@ts.fujitsu.com>
---

Since I have x86 machines only, where flush_dcache_page()
is empty, this patch indeed is compile tested only.
Therefore it needs thorough review or ideally a test on
an architecture that has a non empty flush_dcache_page().


 drivers/target/target_core_user.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/target/target_core_user.c b/drivers/target/target_core_user.c
index cdb4848d23c6..d922a7087c31 100644
--- a/drivers/target/target_core_user.c
+++ b/drivers/target/target_core_user.c
@@ -676,8 +676,10 @@ static void scatter_data_area(struct tcmu_dev *udev,
 		from = kmap_atomic(sg_page(sg)) + sg->offset;
 		while (sg_remaining > 0) {
 			if (block_remaining == 0) {
-				if (to)
+				if (to) {
+					flush_dcache_page(page);
 					kunmap_atomic(to);
+				}
 
 				block_remaining = DATA_BLOCK_SIZE;
 				dbi = tcmu_cmd_get_dbi(tcmu_cmd);
@@ -722,7 +724,6 @@ static void scatter_data_area(struct tcmu_dev *udev,
 				memcpy(to + offset,
 				       from + sg->length - sg_remaining,
 				       copy_bytes);
-				tcmu_flush_dcache_range(to, copy_bytes);
 			}
 
 			sg_remaining -= copy_bytes;
@@ -731,8 +732,10 @@ static void scatter_data_area(struct tcmu_dev *udev,
 		kunmap_atomic(from - sg->offset);
 	}
 
-	if (to)
+	if (to) {
+		flush_dcache_page(page);
 		kunmap_atomic(to);
+	}
 }
 
 static void gather_data_area(struct tcmu_dev *udev, struct tcmu_cmd *cmd,
@@ -778,13 +781,13 @@ static void gather_data_area(struct tcmu_dev *udev, struct tcmu_cmd *cmd,
 				dbi = tcmu_cmd_get_dbi(cmd);
 				page = tcmu_get_block_page(udev, dbi);
 				from = kmap_atomic(page);
+				flush_dcache_page(page);
 			}
 			copy_bytes = min_t(size_t, sg_remaining,
 					block_remaining);
 			if (read_len < copy_bytes)
 				copy_bytes = read_len;
 			offset = DATA_BLOCK_SIZE - block_remaining;
-			tcmu_flush_dcache_range(from, copy_bytes);
 			memcpy(to + sg->length - sg_remaining, from + offset,
 					copy_bytes);
 
-- 
2.12.3