Thread (5 messages) 5 messages, 4 authors, 2007-03-07

Re: [PATCH] atyfb: Replace busy loop with mdelay()

From: Andrew Morton <akpm@linux-foundation.org>
Date: 2007-03-05 23:54:58
Subsystem: framebuffer layer, the rest · Maintainers: Helge Deller, Linus Torvalds

On Thu, 22 Feb 2007 02:20:33 +0200
Ville Syrj__l__ [off-list ref] wrote:
quoted hunk ↗ jump to hunk
Replace a busy loop with mdelay() in atyfb clock setup code. This fixes
boot hangs several people have experienced.

Signed-off-by: Ville Syrjala <syrjala@sci.fi>
---

This patch was already forwareded by James to some of you, but here it 
is with the proper changelog, signed-off-by line, and a language fix in 
the comment part. Also Cc:d stable@kernel.org as this seems like -stable 
material to me.

 drivers/video/aty/mach64_ct.c |    9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

Index: linux-2.6.20.1/drivers/video/aty/mach64_ct.c
===================================================================
--- linux-2.6.20.1.orig/drivers/video/aty/mach64_ct.c
+++ linux-2.6.20.1/drivers/video/aty/mach64_ct.c
@@ -598,7 +598,6 @@ static void aty_resume_pll_ct(const stru
 	struct atyfb_par *par = info->par;
 
 	if (par->mclk_per != par->xclk_per) {
-		int i;
 		/*
 		* This disables the sclk, crashes the computer as reported:
 		* aty_st_pll_ct(SPLL_CNTL2, 3, info);
@@ -609,12 +608,10 @@ static void aty_resume_pll_ct(const stru
 		aty_st_pll_ct(SCLK_FB_DIV, pll->ct.sclk_fb_div, par);
 		aty_st_pll_ct(SPLL_CNTL2, pll->ct.spll_cntl2, par);
 		/*
-		 * The sclk has been started. However, I believe the first clock
-		 * ticks it generates are not very stable. Hope this primitive loop
-		 * helps for Rage Mobilities that sometimes crash when
-		 * we switch to sclk. (Daniel Mantione, 13-05-2003)
+		 * SCLK has been started. Wait for the PLL to lock. 5 ms
+		 * should be enough according to mach64 programmer's guide.
 		 */
-		for (i=0;i<=0x1ffff;i++);
+		mdelay(5);
 	}
 
 	aty_st_pll_ct(PLL_REF_DIV, pll->ct.pll_ref_div, par);
A similar but different patch just got merged into mainline:


commit 8690ba446defe2e2b81803756c099d2943dfd5fd
Author: David Miller [off-list ref]
Date:   Mon Mar 5 01:35:36 2007 -0800

    [PATCH] video/aty/mach64_ct.c: fix bogus delay loop
    
    CT based mach64 cards were reported to hang on sparc64 boxes when
    compiled with gcc-4.1.x and later.
    
    Looking at this piece of code, it's no surprise.  A critical
    delay was implemented as an empty for() loop, and gcc 4.0.x
    and previous did not optimize it away, so we did get a delay.
    
    But gcc-4.1.x and later can optimize it away, and we get crashes.
    
    Use a real udelay() to fix this.  Fix verified on SunBlade100.
    
    Signed-off-by: David S. Miller [off-list ref]
    Signed-off-by: Linus Torvalds [off-list ref]
diff --git a/drivers/video/aty/mach64_ct.c b/drivers/video/aty/mach64_ct.c
index f3b487b..1fdcfdb 100644
--- a/drivers/video/aty/mach64_ct.c
+++ b/drivers/video/aty/mach64_ct.c
@@ -598,7 +598,6 @@ static void aty_resume_pll_ct(const stru
        struct atyfb_par *par = info->par;
 
        if (par->mclk_per != par->xclk_per) {
-               int i;
                /*
                * This disables the sclk, crashes the computer as reported:
                * aty_st_pll_ct(SPLL_CNTL2, 3, info);
@@ -614,7 +613,7 @@ static void aty_resume_pll_ct(const stru
                 * helps for Rage Mobilities that sometimes crash when
                 * we switch to sclk. (Daniel Mantione, 13-05-2003)
                 */
-               for (i=0;i<=0x1ffff;i++);
+               udelay(500);
        }
 
        aty_st_pll_ct(PLL_REF_DIV, pll->ct.pll_ref_div, par);

hopefully that works OK for you.


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help