[Bug 28932] New: new avivo PLL calculation in Radeon driver fails with certain modelines

[Bug 28932] New: new avivo PLL calculation in Radeon driver fails with certain modelines

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=28932

           Summary: new avivo PLL calculation in Radeon driver fails with
                    certain modelines
           Product: Drivers
           Version: 2.5
    Kernel Version: 2.6.38-rc4
          Platform: All
        OS/Version: Linux
              Tree: Mainline
            Status: NEW
          Severity: normal
          Priority: P1
         Component: Video(DRI - non Intel)
        AssignedTo: drivers_video-dri@kernel-bugs.osdl.org
        ReportedBy: bitbytebit@gmail.com
        Regression: Yes


This patch https://bugs.freedesktop.org/attachment.cgi?id=41943 fails to
calculate the PLL values for certain lower dotclock modelines.  I can change
the code to use the legacy PLL calculation and things work again.  

Here's an example modeline that fails now, used to work flawless...


# mario 256x224@59.19 15.2699Khz
    ModeLine          "256x224x59.19" 5.130681 256 272 296 336 224 233 236 258
-HSync -VSync

This modeline now with the newer avivo PLL calculation ends up being 133Hz
instead of 59.19 and the Horizontal frequency is 39khz or so.  It's basically
completely wrong, way off of what it should be.

Things work fine for anything that seems to have a dotclock above 12 Mhz or so,
from what I suspect the newer method isn't taking into account possible lower
dotclock and Horizontal frequency modelines.  

Thanks,
Chris

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb
--

[Bug 28932] new avivo PLL calculation in Radeon driver fails with certain modelines

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=28932





--- Comment #1 from Alex Deucher <alexdeucher@gmail.com>  2011-02-12 08:51:26 ---
Created an attachment (id=47492)
 --> (https://bugzilla.kernel.org/attachment.cgi?id=47492)
possible fix

This patch should fix it.  The post divider was overflowing.

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb
--

[Bug 28932] new avivo PLL calculation in Radeon driver fails with certain modelines

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=28932


Chris Kennedy <bitbytebit@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |CODE_FIX




--- Comment #2 from Chris Kennedy <bitbytebit@gmail.com>  2011-02-12 16:53:50 ---
This fixes the issue, now these lower dotclock modelines work again.

Thanks,
Chris

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb
--
_______________________________________________
Dri-devel mailing list
Dri-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dri-devel

[Bug 28932] new avivo PLL calculation in Radeon driver fails with certain modelines

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=28932





--- Comment #3 from Alex Deucher <alexdeucher@gmail.com>  2011-02-13 23:26:28 ---
Created an attachment (id=47672)
 --> (https://bugzilla.kernel.org/attachment.cgi?id=47672)
alternate patch

Please try this patch instead of the previous one and let me know which works
better or if they both work ok.

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb
--
_______________________________________________
Dri-devel mailing list
Dri-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dri-devel

[Bug 28932] new avivo PLL calculation in Radeon driver fails with certain modelines

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=28932





--- Comment #4 from Chris Kennedy <bitbytebit@gmail.com>  2011-02-14 07:31:36 ---
This works works good too, basically same as the last one, can't tell any
difference.

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb
--
_______________________________________________
Dri-devel mailing list
Dri-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dri-devel

[Bug 28932] new avivo PLL calculation in Radeon driver fails with certain modelines

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=28932


Rafael J. Wysocki <rjw@sisk.pl> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |rjw@sisk.pl
             Blocks|                            |27352




--- Comment #5 from Rafael J. Wysocki <rjw@sisk.pl>  2011-02-14 23:45:22 ---
Handled-By : Alex Deucher <alexdeucher@gmail.com>
Patch : https://bugzilla.kernel.org/attachment.cgi?id=47672

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb
--
_______________________________________________
Dri-devel mailing list
Dri-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dri-devel

[Bug 28932] new avivo PLL calculation in Radeon driver fails with certain modelines

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=28932


Florian Mickler <florian@mickler.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |florian@mickler.org




--- Comment #6 from Chris Kennedy <bitbytebit@gmail.com>  2011-02-18 13:36:01 ---
I think there's an issue with some legacy chips plus low dotclocks and that
same patch, oddly the only little part I can see which could do this is...

@@ -849,7 +951,7 @@ void radeon_compute_pll(struct radeon_pll *pll,
         max_fractional_feed_div = pll->max_frac_feedback_div;
     }

-    for (post_div = max_post_div; post_div >= min_post_div; --post_div) {
+    for (post_div = min_post_div; post_div <= max_post_div; ++post_div) {
         uint32_t ref_div;

         if ((pll->flags & RADEON_PLL_NO_ODD_POST_DIV) && (post_div & 1))


Although it's basically looking like that must be the issue.  A person has a
ATI Radeon 9200SE 5964 (AGP)and with 2.6.38-rc4-git4+ it works while with
2.6.38-rc5 there are certain (not all) modelines with low dotclocks that end up
being totally out of sync.

Here's an example of a non-working modeline for him with the newer kernel, vs.
the older one step back before the patch with the above change...


# toki 256x224@59.61 15.6774Khz
    ModeLine          "256x224x59.61" 5.518455 256 272 304 352 224 235 238 263
-HSync -VSync


There's a lot of other ones thought that can work with the same lower
dotclocks, so I'm not sure why that is, but I can have him test with that
change in the legacy pll computation reversed.   Yet from what I can tell, that
is the only code his Radeon should be hitting that changed in the patch.

Thanks,
Chris

--- Comment #7 from Florian Mickler <florian@mickler.org>  2011-02-20 00:03:57 ---
merged in .38-rc5: 
commit a4b40d5d97f5c9ad0b7f4bf2818291ca184bb433
Author: Alex Deucher <alexdeucher@gmail.com>
Date:   Mon Feb 14 11:43:10 2011 -0500

    drm/radeon/kms: add bounds checking to avivo pll algo

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb
--
_______________________________________________
Dri-devel mailing list
Dri-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dri-devel

[Bug 28932] new avivo PLL calculation in Radeon driver fails with certain modelines

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=28932


Alex Deucher <alexdeucher@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |alexdeucher@gmail.com




--- Comment #8 from Alex Deucher <alexdeucher@gmail.com>  2011-02-21 04:51:56 ---
Comment 6 is not relevant to this bug and has been opened as a separate issue:
bug 29502.

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb
--
_______________________________________________
Dri-devel mailing list
Dri-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dri-devel

[Bug 28932] new avivo PLL calculation in Radeon driver fails with certain modelines

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=28932


Florian Mickler <florian@mickler.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |CLOSED




--- Comment #9 from Florian Mickler <florian@mickler.org>  2011-02-21 08:42:00 ---
Thx, closing.

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb
--
_______________________________________________
Dri-devel mailing list
Dri-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dri-devel