[PATCH] video: fbdev: savage: Replace mdelay with usleep_range in savage_init_hw

[PATCH] video: fbdev: savage: Replace mdelay with usleep_range in savage_init_hw

[Jia-Ju Bai]

savage_init_hw() is never called in atomic context.

The call chains ending up at savage_init_hw() are:
[1] savage_init_hw() <- savagefb_probe()
[2] savage_init_hw() <- savagefb_resume()

savagefb_probe() is only set as ".probe" in struct pci_driver.
savagefb_resume) is only set as ".resume" in struct pci_driver.
These functions are not called in atomic context.

Despite never getting called from atomic context, savage_init_hw()
calls mdelay() to busily wait.
This is not necessary and can be replaced with usleep_range to
avoid busy waiting.

This is found by a static analysis tool named DCNS written by myself.
And I also manually check it.

Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
---
 drivers/video/fbdev/savage/savagefb_driver.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/video/fbdev/savage/savagefb_driver.c b/drivers/video/fbdev/savage/savagefb_driver.c
index c204683..c09d742 100644
--- a/drivers/video/fbdev/savage/savagefb_driver.c
+++ b/drivers/video/fbdev/savage/savagefb_driver.c
@@ -1892,11 +1892,11 @@ static int savage_init_hw(struct savagefb_par *par)
 	vga_out8(0x3d4, 0x66, par);
 	cr66 = vga_in8(0x3d5, par);
 	vga_out8(0x3d5, cr66 | 0x02, par);
-	mdelay(10);
+	usleep_range(10000, 11000);
 
 	vga_out8(0x3d4, 0x66, par);
 	vga_out8(0x3d5, cr66 & ~0x02, par);	/* clear reset flag */
-	mdelay(10);
+	usleep_range(10000, 11000);
 
 
 	/*
@@ -1906,11 +1906,11 @@ static int savage_init_hw(struct savagefb_par *par)
 	vga_out8(0x3d4, 0x3f, par);
 	cr3f = vga_in8(0x3d5, par);
 	vga_out8(0x3d5, cr3f | 0x08, par);
-	mdelay(10);
+	usleep_range(10000, 11000);
 
 	vga_out8(0x3d4, 0x3f, par);
 	vga_out8(0x3d5, cr3f & ~0x08, par);	/* clear reset flags */
-	mdelay(10);
+	usleep_range(10000, 11000);
 
 	/* Savage ramdac speeds */
 	par->numClocks = 4;
-- 
1.9.1

Re: [PATCH] video: fbdev: savage: Replace mdelay with usleep_range in savage_init_hw

[Bartlomiej Zolnierkiewicz]

On Tuesday, April 10, 2018 09:05:59 AM Jia-Ju Bai wrote:
> savage_init_hw() is never called in atomic context.
> 
> The call chains ending up at savage_init_hw() are:
> [1] savage_init_hw() <- savagefb_probe()
> [2] savage_init_hw() <- savagefb_resume()
> 
> savagefb_probe() is only set as ".probe" in struct pci_driver.
> savagefb_resume) is only set as ".resume" in struct pci_driver.
> These functions are not called in atomic context.
> 
> Despite never getting called from atomic context, savage_init_hw()
> calls mdelay() to busily wait.
> This is not necessary and can be replaced with usleep_range to
> avoid busy waiting.
> 
> This is found by a static analysis tool named DCNS written by myself.
> And I also manually check it.
> 
> Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>

Patch queued for 4.18, thanks.

Best regards,
--
Bartlomiej Zolnierkiewicz
Samsung R&D Institute Poland
Samsung Electronics