Skip to contentSkip to conversation

ufs: Adds glue drivers to ufshcd

PATCH Superseded Vinayak Holikatti 3 participants 6 messages Jul 13, 2012 – Jul 19, 2012

Subsystems (2)

v2 of 6 Superseded Latest is v6 →

lore
Share & tools
lore
#0 lore raw

v2 cover ufs: Adds glue drivers to ufshcd

This patch set adds following features
- Seprates PCI specific code from ufshcd.c to make it as core
- Adds PCI glue driver ufshcd-pci.c
- Adds Platform glue driver ufshcd-pltfrm.c
- Update correct transfer size in Command UPIU

Vinayak Holikatti (4):
[SCSI] drivers/scsi/ufs: Remove PCI specific code [SCSI] drivers/scsi/ufs: Add PCI glue driver [SCSI] ufs: Add Platform glue driver for ufshcd [SCSI] ufs: Correct the expected data transfer size

 drivers/scsi/ufs/Kconfig         |   37 +++-
 drivers/scsi/ufs/Makefile        |    2 +
 drivers/scsi/ufs/ufshcd-pci.c    |  227 ++++++++++++++++++++
 drivers/scsi/ufs/ufshcd-pltfrm.c |  215 +++++++++++++++++++
 drivers/scsi/ufs/ufshcd.c        |  422 ++++++--------------------------------
 drivers/scsi/ufs/ufshcd.h        |  206 +++++++++++++++++++
 drivers/scsi/ufs/ufshcd_common.h |   53 +++++
 7 files changed, 805 insertions(+), 357 deletions(-)
 create mode 100644 drivers/scsi/ufs/ufshcd-pci.c
 create mode 100644 drivers/scsi/ufs/ufshcd-pltfrm.c
 create mode 100644 drivers/scsi/ufs/ufshcd.h
 create mode 100644 drivers/scsi/ufs/ufshcd_common.h
Signature · 1 line
1.7.9.5
#1 lore raw

PATCH v2 1/4 drivers/scsi/ufs: Remove PCI specific code

This patch removes the PCI specific code to make it into core ufs driver. This separation allows room for other bus specific glue drivers to be added in future

---
 drivers/scsi/ufs/Kconfig  |   15 ++-
 drivers/scsi/ufs/ufshcd.c |  270 ++++++++++++---------------------------------
 2 files changed, 80 insertions(+), 205 deletions(-)
drivers/scsi/ufs/Kconfig +12 -3
index 8f27f9d..8c0cd22 100644
@@ -43,7 +43,16 @@
 # USA.
 
 config SCSI_UFSHCD
-	tristate "Universal Flash Storage host controller driver"
-	depends on PCI && SCSI
+	tristate "Universal Flash Storage Controller Driver Core"
+	depends on SCSI
 	---help---
-	This is a generic driver which supports PCIe UFS Host controllers.
+	This selects the support for UFS devices in Linux, say Y and make
+	  sure that you know the name of your UFS host adapter (the card
+	  inside your computer that "speaks" the UFS protocol, also
+	  called UFS Host Controller), because you will be asked for it.
+	  The module will be called ufshcd.
+
+	  To compile this driver as a module, choose M here and read
+	  <file:Documentation/scsi/ufs.txt>.
+	  However, do not compile this as a module if your root file system
+	  (the one containing the directory /) is located on a UFS device.
drivers/scsi/ufs/ufshcd.c +68 -202

502 lines not on this page · raw message

Signature · 1 line
1.7.9.5
maintainer
#3 lore raw
Quoted 8 lines · Arnd Bergmann
Quoted 5 lines · Vinayak Holikatti

This patch set adds following features
- Seprates PCI specific code from ufshcd.c to make it as core
- Adds PCI glue driver ufshcd-pci.c
- Adds Platform glue driver ufshcd-pltfrm.c
- Update correct transfer size in Command UPIU

I need a maintainer ack for this to go upstream.

Also, looking at this, I think this patch series isn't bisectable:
Patch 1 removes PCI support and patch 2 adds it back in a different form. However, any PCI based UFS system would stop working if the bisect landed at patch 1. I think you can fix this just by combining patches 1 and 2.

James

#4 lore raw
Quoted 10 lines · James Bottomley
Quoted 8 lines · Arnd Bergmann
Quoted 5 lines · Vinayak Holikatti

This patch set adds following features
- Seprates PCI specific code from ufshcd.c to make it as core
- Adds PCI glue driver ufshcd-pci.c
- Adds Platform glue driver ufshcd-pltfrm.c
- Update correct transfer size in Command UPIU

Quoted 7 lines

I need a maintainer ack for this to go upstream.

Also, looking at this, I think this patch series isn't bisectable:
Patch 1 removes PCI support and patch 2 adds it back in a different form. However, any PCI based UFS system would stop working if the bisect landed at patch 1. I think you can fix this just by combining patches 1 and 2.

I suggested to split the patch in two in a private review that we did on the linaro mailing list. I gave my Ack because the split was done, but I failed to notice that it was done differently from what I suggested in

Quoted 4 lines · Arnd Bergmann

I would recommend that you split this patch into two separate changesets, where you do all the changes to existing code in the first patch, and only move but don't change code in the second one that creates the new file.

I agree that breaking bisectibility by ripping out the PCI code first is not good. The version 2 is not actually easier to review at all than the first version, it just splits the changes by file, which is pointless.

Arnd

#5 lore raw
Quoted 30 lines · Arnd Bergmann
Quoted 10 lines · James Bottomley
Quoted 8 lines · Arnd Bergmann
Quoted 5 lines · Vinayak Holikatti

This patch set adds following features
- Seprates PCI specific code from ufshcd.c to make it as core
- Adds PCI glue driver ufshcd-pci.c
- Adds Platform glue driver ufshcd-pltfrm.c
- Update correct transfer size in Command UPIU

Quoted 7 lines

I need a maintainer ack for this to go upstream.

Also, looking at this, I think this patch series isn't bisectable:
Patch 1 removes PCI support and patch 2 adds it back in a different form. However, any PCI based UFS system would stop working if the bisect landed at patch 1. I think you can fix this just by combining patches 1 and 2.

I suggested to split the patch in two in a private review that we did on the linaro mailing list. I gave my Ack because the split was done, but I failed to notice that it was done differently from what I suggested in

Quoted 4 lines · Arnd Bergmann

I would recommend that you split this patch into two separate changesets, where you do all the changes to existing code in the first patch, and only move but don't change code in the second one that creates the new file.

Quoted 4 lines

I agree that breaking bisectibility by ripping out the PCI code first is not good. The version 2 is not actually easier to review at all than the first version, it just splits the changes by file, which is pointless.

I will roll out another patch series which will address the current issue. Firstly will modify existing code with PCI code and then in the second move the code to new file ufshcd-pci.c

Quoted 3 lines

Arnd