[PATCH v2 1/4] staging: sm750fb: wrong type for print

[PATCH v2 1/4] staging: sm750fb: wrong type for print

[Sudip Mukherjee]

mention correct format specifier while printing.
fixes all the build warnings about incorrect argument type while
printing.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
---

V2: Giedrius commented resource_size_t can be either u64 or u32
depending on if CONFIG_PHYS_ADDR_T_64BIT. based on his comments i
should have kept the datatype as resource_size_t and  used %pa as the
format specifier. But since this is a framebuffer device and it
should follow what the framebuffer layer is suggesting in
struct fb_fix_screeninfo at smem_start and mmio_start.
adding Tomi for his comments.

this patch will give checkpatch warnings about use of printk.
this patch was mainly to fix the build warnings. printk will be
converted to pr_* and dev_* in a later patch.

 drivers/staging/sm750fb/sm750.c    | 24 ++++++++++++------------
 drivers/staging/sm750fb/sm750.h    |  8 ++++----
 drivers/staging/sm750fb/sm750_hw.c |  4 ++--
 3 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/drivers/staging/sm750fb/sm750.c b/drivers/staging/sm750fb/sm750.c
index 520c69e..753869e 100644
--- a/drivers/staging/sm750fb/sm750.c
+++ b/drivers/staging/sm750fb/sm750.c
@@ -530,20 +530,20 @@ static int lynxfb_ops_mmap(struct fb_info * info, struct vm_area_struct * vma)
 	if (vma->vm_pgoff > (~0UL >> PAGE_SHIFT))
 		return -EINVAL;
 	off = vma->vm_pgoff << PAGE_SHIFT;
-	printk("lynxfb mmap pgoff: %x\n", vma->vm_pgoff);
-	printk("lynxfb mmap off 1: %x\n", off);
+	printk("lynxfb mmap pgoff: %lx\n", vma->vm_pgoff);
+	printk("lynxfb mmap off 1: %lx\n", off);
 	
 	/* frame buffer memory */
 	start = info->fix.smem_start;
 	len = PAGE_ALIGN((start & ~PAGE_MASK) + info->fix.smem_len);
 	
-	printk("lynxfb mmap start 1: %x\n", start);
+	printk("lynxfb mmap start 1: %lx\n", start);
 	printk("lynxfb mmap len 1: %x\n", len);
 	
 	if (off >= len) {
 		/* memory mapped io */
 		off -= len;
-		printk("lynxfb mmap off 2: %x\n", off);
+		printk("lynxfb mmap off 2: %lx\n", off);
 		if (info->var.accel_flags) {
 			printk("lynxfb mmap accel flags true");
 			return -EINVAL;
@@ -551,28 +551,28 @@ static int lynxfb_ops_mmap(struct fb_info * info, struct vm_area_struct * vma)
 		start = info->fix.mmio_start;
 		len = PAGE_ALIGN((start & ~PAGE_MASK) + info->fix.mmio_len);
 		
-		printk("lynxfb mmap start 2: %x\n", start);
+		printk("lynxfb mmap start 2: %lx\n", start);
 		printk("lynxfb mmap len 2: %x\n", len);
 	}
 	start &= PAGE_MASK;
-	printk("lynxfb mmap start 3: %x\n", start);
-	printk("lynxfb mmap vm start: %x\n", vma->vm_start);
-	printk("lynxfb mmap vm end: %x\n", vma->vm_end);
+	printk("lynxfb mmap start 3: %lx\n", start);
+	printk("lynxfb mmap vm start: %lx\n", vma->vm_start);
+	printk("lynxfb mmap vm end: %lx\n", vma->vm_end);
 	printk("lynxfb mmap len: %x\n", len);
-	printk("lynxfb mmap off: %x\n", off);
+	printk("lynxfb mmap off: %lx\n", off);
 	if ((vma->vm_end - vma->vm_start + off) > len)
 	{
 		return -EINVAL;
 	}
 	off += start;
-	printk("lynxfb mmap off 3: %x\n", off);
+	printk("lynxfb mmap off 3: %lx\n", off);
 	vma->vm_pgoff = off >> PAGE_SHIFT;
 	/* This is an IO map - tell maydump to skip this VMA */
 	vma->vm_flags |= VM_IO | VM_DONTEXPAND | VM_DONTDUMP;
 	vma->vm_page_prot = vm_get_page_prot(vma->vm_flags);
 	fb_pgprotect(file, vma, off);
-	printk("lynxfb mmap off 4: %x\n", off);
-	printk("lynxfb mmap pgprot: %x\n", vma->vm_page_prot);
+	printk("lynxfb mmap off 4: %lx\n", off);
+	printk("lynxfb mmap pgprot: %lx\n", (unsigned long) pgprot_val(vma->vm_page_prot));
 	if (io_remap_pfn_range(vma, vma->vm_start, off >> PAGE_SHIFT,
 			     vma->vm_end - vma->vm_start, vma->vm_page_prot))
 		return -EAGAIN;
diff --git a/drivers/staging/sm750fb/sm750.h b/drivers/staging/sm750fb/sm750.h
index 711676c..d39968c 100644
--- a/drivers/staging/sm750fb/sm750.h
+++ b/drivers/staging/sm750fb/sm750.h
@@ -59,10 +59,10 @@ struct lynx_share{
 		}mtrr;
 #endif
 	/* all smi graphic adaptor got below attributes */
-	resource_size_t vidmem_start;
-	resource_size_t vidreg_start;
-	resource_size_t vidmem_size;
-	resource_size_t vidreg_size;
+	unsigned long vidmem_start;
+	unsigned long vidreg_start;
+	__u32 vidmem_size;
+	__u32 vidreg_size;
 	volatile unsigned char __iomem * pvReg;
 	unsigned char __iomem * pvMem;
 	/* locks*/
diff --git a/drivers/staging/sm750fb/sm750_hw.c b/drivers/staging/sm750fb/sm750_hw.c
index cd971bd..ec2d499 100644
--- a/drivers/staging/sm750fb/sm750_hw.c
+++ b/drivers/staging/sm750fb/sm750_hw.c
@@ -36,7 +36,7 @@ int hw_sm750_map(struct lynx_share* share,struct pci_dev* pdev)
 	share->vidreg_start  = pci_resource_start(pdev,1);
 	share->vidreg_size = MB(2);
 
-	pr_info("mmio phyAddr = %x\n",share->vidreg_start);
+	pr_info("mmio phyAddr = %lx\n", share->vidreg_start);
 
 	/* reserve the vidreg space of smi adaptor
 	 * if you do this, u need to add release region code
@@ -73,7 +73,7 @@ int hw_sm750_map(struct lynx_share* share,struct pci_dev* pdev)
 	 * @hw_sm750_getVMSize function can be safe.
 	 * */
 	share->vidmem_size = hw_sm750_getVMSize(share);
-	pr_info("video memory phyAddr = %x, size = %d bytes\n",
+	pr_info("video memory phyAddr = %lx, size = %u bytes\n",
 	share->vidmem_start,share->vidmem_size);
 
 	/* reserve the vidmem space of smi adaptor */
-- 
1.8.1.2

[PATCH v2 2/4] staging: sm750fb: remove pragma optimize

[Sudip Mukherjee]

remove use of #pragma optimize which will usually be ignored by the
compiler.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
---

v2: same as v1

 drivers/staging/sm750fb/ddk750_swi2c.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/staging/sm750fb/ddk750_swi2c.c b/drivers/staging/sm750fb/ddk750_swi2c.c
index b53407b..cae6b9b 100644
--- a/drivers/staging/sm750fb/ddk750_swi2c.c
+++ b/drivers/staging/sm750fb/ddk750_swi2c.c
@@ -217,8 +217,6 @@ static unsigned char swI2CReadSDA(void)
         return 0;
 }
 
-#pragma optimize( "", off )
-
 /*
  *  This function sends ACK signal
  */
@@ -356,7 +354,6 @@ unsigned char swI2CReadByte(unsigned char ack)
 
     return data;
 }
-#pragma optimize( "", on )
 
 /*
  * This function initializes GPIO port for SW I2C communication.
-- 
1.8.1.2

[PATCH v2 3/4] staging: sm750fb: correctly define SM750LE_REVISION_ID

[Sudip Mukherjee]

check if it is already defined before defining SM750LE_REVISION_ID
again and at the same time mention correct data type.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
---

v2: removed the redundant cast in sm750_hw.c

 drivers/staging/sm750fb/ddk750_chip.h | 4 +++-
 drivers/staging/sm750fb/sm750_hw.c    | 2 +-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/staging/sm750fb/ddk750_chip.h b/drivers/staging/sm750fb/ddk750_chip.h
index 1c78875..d761b72 100644
--- a/drivers/staging/sm750fb/ddk750_chip.h
+++ b/drivers/staging/sm750fb/ddk750_chip.h
@@ -1,7 +1,9 @@
 #ifndef DDK750_CHIP_H__
 #define DDK750_CHIP_H__
 #define DEFAULT_INPUT_CLOCK 14318181 /* Default reference clock */
-#define SM750LE_REVISION_ID (char)0xfe
+#ifndef SM750LE_REVISION_ID
+#define SM750LE_REVISION_ID ((unsigned char)0xfe)
+#endif
 
 /* This is all the chips recognized by this library */
 typedef enum _logical_chip_type_t
diff --git a/drivers/staging/sm750fb/sm750_hw.c b/drivers/staging/sm750fb/sm750_hw.c
index ec2d499..a2b7fe2 100644
--- a/drivers/staging/sm750fb/sm750_hw.c
+++ b/drivers/staging/sm750fb/sm750_hw.c
@@ -277,7 +277,7 @@ int hw_sm750_crtc_checkMode(struct lynxfb_crtc* crtc,struct fb_var_screeninfo* v
 		case 16:
 			break;
 		case 32:
-			if(share->revid == (unsigned char)SM750LE_REVISION_ID){
+			if (share->revid == SM750LE_REVISION_ID) {
 				pr_debug("750le do not support 32bpp\n");
 				return -EINVAL;
 			}
-- 
1.8.1.2

[PATCH v2 4/4] staging: sm750fb: fix undeclared function

[Sudip Mukherjee]

kbuild test robot reported that for microblaze-allyesconfig
chan_to_field() and lynxfb_ops_set_par() were not defined. These two
functions were defined under CONFIG_PM, so for any archtecture if
CONFIG_PM is not defined we will have this error.

while moving the lynxfb_suspend() function some very obvious
checkpatch errors, like space after comma, space after if, space
before opening brace, were taken care of.

Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
---

v2: some extra spaces in the comments were removed.

 drivers/staging/sm750fb/sm750.c | 110 +++++++++++++++++++---------------------
 1 file changed, 53 insertions(+), 57 deletions(-)

diff --git a/drivers/staging/sm750fb/sm750.c b/drivers/staging/sm750fb/sm750.c
index 753869e..476dc5c 100644
--- a/drivers/staging/sm750fb/sm750.c
+++ b/drivers/staging/sm750fb/sm750.c
@@ -303,62 +303,6 @@ static int lynxfb_ops_pan_display(struct fb_var_screeninfo *var,
     return ret;
 }
 
-
-
-
-#ifdef CONFIG_PM
-static int lynxfb_suspend(struct pci_dev * pdev,pm_message_t mesg)
-{
-	struct fb_info * info;
-	struct lynx_share * share;
-	int ret;
-	
-
-	if(mesg.event == pdev->dev.power.power_state.event)
-		return 0;
-
-	ret = 0;
-	share = pci_get_drvdata(pdev);
-	switch (mesg.event) {
-	case PM_EVENT_FREEZE:
-	case PM_EVENT_PRETHAW:
-		pdev->dev.power.power_state = mesg;
-		return 0;
-	}
-
-	console_lock();
-	if (mesg.event & PM_EVENT_SLEEP) {
-		info = share->fbinfo[0];
-		if(info)
-			fb_set_suspend(info, 1);/* 1 means do suspend*/
-
-		info = share->fbinfo[1];
-		if(info)
-			fb_set_suspend(info, 1);/* 1 means do suspend*/
-
-		ret = pci_save_state(pdev);
-		if(ret){
-			pr_err("error:%d occured in pci_save_state\n",ret);
-			return ret;
-		}
-
-		/* set chip to sleep mode	*/
-		if(share->suspend)
-			(*share->suspend)(share);
-
-		pci_disable_device(pdev);
-		ret = pci_set_power_state(pdev,pci_choose_state(pdev,mesg));
-		if(ret){
-			pr_err("error:%d occured in pci_set_power_state\n",ret);
-			return ret;
-		}
-	}
-
-	pdev->dev.power.power_state = mesg;
-	console_unlock();
-	return ret;
-}
-
 static int lynxfb_ops_set_par(struct fb_info * info)
 {
 	struct lynxfb_par * par;
@@ -369,7 +313,6 @@ static int lynxfb_ops_set_par(struct fb_info * info)
 	struct fb_fix_screeninfo * fix;
 	int ret;
 	unsigned int line_length;
-	
 
 	if(!info)
 		return -EINVAL;
@@ -441,6 +384,7 @@ static int lynxfb_ops_set_par(struct fb_info * info)
 		ret = output->proc_setMode(output,var,fix);
 	return ret;
 }
+
 static inline unsigned int chan_to_field(unsigned int chan,struct fb_bitfield * bf)
 {
 	chan &= 0xffff;
@@ -448,6 +392,58 @@ static inline unsigned int chan_to_field(unsigned int chan,struct fb_bitfield *
 	return chan << bf->offset;
 }
 
+#ifdef CONFIG_PM
+static int lynxfb_suspend(struct pci_dev *pdev, pm_message_t mesg)
+{
+	struct fb_info *info;
+	struct lynx_share *share;
+	int ret;
+
+	if (mesg.event == pdev->dev.power.power_state.event)
+		return 0;
+
+	ret = 0;
+	share = pci_get_drvdata(pdev);
+	switch (mesg.event) {
+	case PM_EVENT_FREEZE:
+	case PM_EVENT_PRETHAW:
+		pdev->dev.power.power_state = mesg;
+		return 0;
+	}
+
+	console_lock();
+	if (mesg.event & PM_EVENT_SLEEP) {
+		info = share->fbinfo[0];
+		if (info)
+			/* 1 means do suspend*/
+			fb_set_suspend(info, 1);
+		info = share->fbinfo[1];
+		if (info)
+			/* 1 means do suspend*/
+			fb_set_suspend(info, 1);
+
+		ret = pci_save_state(pdev);
+		if (ret) {
+			pr_err("error:%d occurred in pci_save_state\n", ret);
+			return ret;
+		}
+
+		/* set chip to sleep mode*/
+		if (share->suspend)
+			(*share->suspend)(share);
+
+		pci_disable_device(pdev);
+		ret = pci_set_power_state(pdev, pci_choose_state(pdev, mesg));
+		if (ret) {
+			pr_err("error:%d occurred in pci_set_power_state\n", ret);
+			return ret;
+		}
+	}
+
+	pdev->dev.power.power_state = mesg;
+	console_unlock();
+	return ret;
+}
 
 static int lynxfb_resume(struct pci_dev* pdev)
 {
-- 
1.8.1.2

Re: [PATCH v2 1/4] staging: sm750fb: wrong type for print

[Lad, Prabhakar]

On Mon, Mar 9, 2015 at 7:35 AM, Sudip Mukherjee
<sudipm.mukherjee@gmail.com> wrote:
> mention correct format specifier while printing.
> fixes all the build warnings about incorrect argument type while
> printing.
>
> Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
> ---
>
> V2: Giedrius commented resource_size_t can be either u64 or u32
> depending on if CONFIG_PHYS_ADDR_T_64BIT. based on his comments i
> should have kept the datatype as resource_size_t and  used %pa as the
> format specifier. But since this is a framebuffer device and it
> should follow what the framebuffer layer is suggesting in
> struct fb_fix_screeninfo at smem_start and mmio_start.
> adding Tomi for his comments.
>
> this patch will give checkpatch warnings about use of printk.
> this patch was mainly to fix the build warnings. printk will be
> converted to pr_* and dev_* in a later patch.
>
>  drivers/staging/sm750fb/sm750.c    | 24 ++++++++++++------------
>  drivers/staging/sm750fb/sm750.h    |  8 ++++----
>  drivers/staging/sm750fb/sm750_hw.c |  4 ++--
>  3 files changed, 18 insertions(+), 18 deletions(-)
>
> diff --git a/drivers/staging/sm750fb/sm750.c b/drivers/staging/sm750fb/sm750.c
> index 520c69e..753869e 100644
> --- a/drivers/staging/sm750fb/sm750.c
> +++ b/drivers/staging/sm750fb/sm750.c
> @@ -530,20 +530,20 @@ static int lynxfb_ops_mmap(struct fb_info * info, struct vm_area_struct * vma)
>         if (vma->vm_pgoff > (~0UL >> PAGE_SHIFT))
>                 return -EINVAL;
>         off = vma->vm_pgoff << PAGE_SHIFT;
> -       printk("lynxfb mmap pgoff: %x\n", vma->vm_pgoff);
> -       printk("lynxfb mmap off 1: %x\n", off);
> +       printk("lynxfb mmap pgoff: %lx\n", vma->vm_pgoff);
> +       printk("lynxfb mmap off 1: %lx\n", off);
>
wouldn't it makes sense to change printk to some pr_*() ?

Cheers,
--Prabhakar Lad

Re: [PATCH v2 3/4] staging: sm750fb: correctly define SM750LE_REVISION_ID

[Lad, Prabhakar]

On Mon, Mar 9, 2015 at 7:35 AM, Sudip Mukherjee
<sudipm.mukherjee@gmail.com> wrote:
> check if it is already defined before defining SM750LE_REVISION_ID
> again and at the same time mention correct data type.
>
> Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
> ---
>
> v2: removed the redundant cast in sm750_hw.c
>
>  drivers/staging/sm750fb/ddk750_chip.h | 4 +++-
>  drivers/staging/sm750fb/sm750_hw.c    | 2 +-
>  2 files changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/staging/sm750fb/ddk750_chip.h b/drivers/staging/sm750fb/ddk750_chip.h
> index 1c78875..d761b72 100644
> --- a/drivers/staging/sm750fb/ddk750_chip.h
> +++ b/drivers/staging/sm750fb/ddk750_chip.h
> @@ -1,7 +1,9 @@
>  #ifndef DDK750_CHIP_H__
>  #define DDK750_CHIP_H__
>  #define DEFAULT_INPUT_CLOCK 14318181 /* Default reference clock */
> -#define SM750LE_REVISION_ID (char)0xfe
> +#ifndef SM750LE_REVISION_ID
> +#define SM750LE_REVISION_ID ((unsigned char)0xfe)

BTW who is defining this somewhere else ?

Cheers,
--Prabhakar Lad

Re: [PATCH v2 1/4] staging: sm750fb: wrong type for print

[Sudip Mukherjee]

On Mon, Mar 09, 2015 at 07:42:53AM +0000, Lad, Prabhakar wrote:
> On Mon, Mar 9, 2015 at 7:35 AM, Sudip Mukherjee
> <sudipm.mukherjee@gmail.com> wrote:
> >
> >
> > this patch will give checkpatch warnings about use of printk.
> > this patch was mainly to fix the build warnings. printk will be
> > converted to pr_* and dev_* in a later patch.
> >
> > +       printk("lynxfb mmap pgoff: %lx\n", vma->vm_pgoff);
> > +       printk("lynxfb mmap off 1: %lx\n", off);
> >
> wouldn't it makes sense to change printk to some pr_*() ?
yes, i could have. But usually Greg does not accept patches which is doing more than one type of change. That is why I mentioned in the comments that this will be converted to pr_* and dev_* in a later patch.

regards
sudip

> 
> Cheers,
> --Prabhakar Lad

Re: [PATCH v2 3/4] staging: sm750fb: correctly define SM750LE_REVISION_ID

[Sudip Mukherjee]

On Mon, Mar 09, 2015 at 07:47:40AM +0000, Lad, Prabhakar wrote:
> On Mon, Mar 9, 2015 at 7:35 AM, Sudip Mukherjee
> <sudipm.mukherjee@gmail.com> wrote:
> > check if it is already defined before defining SM750LE_REVISION_ID
> > again and at the same time mention correct data type.
> >
> > Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
> > ---
> >
> > v2: removed the redundant cast in sm750_hw.c
> >
> >  drivers/staging/sm750fb/ddk750_chip.h | 4 +++-
> >  drivers/staging/sm750fb/sm750_hw.c    | 2 +-
> >  2 files changed, 4 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/staging/sm750fb/ddk750_chip.h b/drivers/staging/sm750fb/ddk750_chip.h
> > index 1c78875..d761b72 100644
> > --- a/drivers/staging/sm750fb/ddk750_chip.h
> > +++ b/drivers/staging/sm750fb/ddk750_chip.h
> > @@ -1,7 +1,9 @@
> >  #ifndef DDK750_CHIP_H__
> >  #define DDK750_CHIP_H__
> >  #define DEFAULT_INPUT_CLOCK 14318181 /* Default reference clock */
> > -#define SM750LE_REVISION_ID (char)0xfe
> > +#ifndef SM750LE_REVISION_ID
> > +#define SM750LE_REVISION_ID ((unsigned char)0xfe)
> 
> BTW who is defining this somewhere else ?
it is defined in ddk750_chip.h which is again being used by ddk750_chip.c .


regards
sudip

> 
> Cheers,
> --Prabhakar Lad

Re: [PATCH v2 1/4] staging: sm750fb: wrong type for print

[Dan Carpenter]

On Mon, Mar 09, 2015 at 01:05:03PM +0530, Sudip Mukherjee wrote:
> mention correct format specifier while printing.
> fixes all the build warnings about incorrect argument type while
> printing.
> 
> Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
> ---
> 
> V2: Giedrius commented resource_size_t can be either u64 or u32
> depending on if CONFIG_PHYS_ADDR_T_64BIT. based on his comments i
> should have kept the datatype as resource_size_t and  used %pa as the
> format specifier. But since this is a framebuffer device and it
> should follow what the framebuffer layer is suggesting in
> struct fb_fix_screeninfo at smem_start and mmio_start.
> adding Tomi for his comments.


This should all have gone into the changelog because it changes how the
code works.

regards,
dan carpenter

Re: [PATCH v2 1/4] staging: sm750fb: wrong type for print

[Sudip Mukherjee]

On Mon, Mar 09, 2015 at 02:53:37PM +0300, Dan Carpenter wrote:
> On Mon, Mar 09, 2015 at 01:05:03PM +0530, Sudip Mukherjee wrote:
> > 
> > V2: Giedrius commented resource_size_t can be either u64 or u32
> > depending on if CONFIG_PHYS_ADDR_T_64BIT. based on his comments i
> > should have kept the datatype as resource_size_t and  used %pa as the
> > format specifier. But since this is a framebuffer device and it
> > should follow what the framebuffer layer is suggesting in
> > struct fb_fix_screeninfo at smem_start and mmio_start.
> > adding Tomi for his comments.
> 
> 
> This should all have gone into the changelog because it changes how the
> code works.

ok, then should i send a v3 of only this patch or the whole series?

regards
sudip

> 
> regards,
> dan carpenter
> 

Re: [PATCH v2 4/4] staging: sm750fb: fix undeclared function

[Dan Carpenter]

On Mon, Mar 09, 2015 at 01:05:06PM +0530, Sudip Mukherjee wrote:
> kbuild test robot reported that for microblaze-allyesconfig
> chan_to_field() and lynxfb_ops_set_par() were not defined. These two
> functions were defined under CONFIG_PM, so for any archtecture if
> CONFIG_PM is not defined we will have this error.
> 
> while moving the lynxfb_suspend() function some very obvious
> checkpatch errors, like space after comma, space after if, space
> before opening brace, were taken care of.

I have a script to review patches moving functions around but these
white space changes break my script so I have to review it by hand.
Sucks.

>  static int lynxfb_ops_set_par(struct fb_info * info)
>  {
>  	struct lynxfb_par * par;
> @@ -369,7 +313,6 @@ static int lynxfb_ops_set_par(struct fb_info * info)
>  	struct fb_fix_screeninfo * fix;
>  	int ret;
>  	unsigned int line_length;
> -	
>  
>  	if(!info)
>  		return -EINVAL;
> @@ -441,6 +384,7 @@ static int lynxfb_ops_set_par(struct fb_info * info)
>  		ret = output->proc_setMode(output,var,fix);
>  	return ret;
>  }
> +
>  static inline unsigned int chan_to_field(unsigned int chan,struct fb_bitfield * bf)
>  {
>  	chan &= 0xffff;

These white space changes are not related.

regards,
dan carpenter

Re: [PATCH v2 4/4] staging: sm750fb: fix undeclared function

[Sudip Mukherjee]

On Mon, Mar 09, 2015 at 03:42:22PM +0300, Dan Carpenter wrote:
> On Mon, Mar 09, 2015 at 01:05:06PM +0530, Sudip Mukherjee wrote:
> > kbuild test robot reported that for microblaze-allyesconfig
> > chan_to_field() and lynxfb_ops_set_par() were not defined. These two
> > functions were defined under CONFIG_PM, so for any archtecture if
> > CONFIG_PM is not defined we will have this error.
> > 
> > while moving the lynxfb_suspend() function some very obvious
> > checkpatch errors, like space after comma, space after if, space
> > before opening brace, were taken care of.
> 
> I have a script to review patches moving functions around but these
> white space changes break my script so I have to review it by hand.
> Sucks.

oops . sorry ..
> 
> >  static int lynxfb_ops_set_par(struct fb_info * info)
<snip>
> >  static inline unsigned int chan_to_field(unsigned int chan,struct fb_bitfield * bf)
> >  {
> >  	chan &= 0xffff;
> 
> These white space changes are not related.
if you want i can break it into multiple patches, so that reviewing can be easy and your script will not break :) .
Actually I thought, since this is a vendor crude driver there will be many such changes, so if i can combine some changes together then atleast the number of patches can be kept low and also i thought of clubbing these changes together as Joe Perches once told me "Don't get carried away with patch type separation" (reference: https://lkml.org/lkml/2015/1/1/2).

regards
sudip

> 
> regards,
> dan carpenter
> 

Re: [PATCH v2 4/4] staging: sm750fb: fix undeclared function

[Dan Carpenter]

On Mon, Mar 09, 2015 at 06:25:03PM +0530, Sudip Mukherjee wrote:
> Actually I thought, since this is a vendor crude driver there will be many such changes, so if i can combine some changes together then atleast the number of patches can be kept low and also i thought of clubbing these changes together as Joe Perches once told me "Don't get carried away with patch type separation" (reference: https://lkml.org/lkml/2015/1/1/2).

How to break up patches is more art than science.  I wouldn't have
complained about the minor extra white space changes except that I was
already going to email you about breaking my scripts.

regards,
dan carpenter