Bug 18420 - [915gv GEM] readpix fail with tiling on
[915gv GEM] readpix fail with tiling on
Status: CLOSED FIXED
Product: Mesa
Classification: Unclassified
Component: Drivers/DRI/i915
unspecified
Other Linux (All)
: low major
Assigned To: Eric Anholt
:
Depends on:
Blocks: 18841
  Show dependency treegraph
 
Reported: 2008-11-06 23:47 UTC by liuhaien
Modified: 2009-11-03 01:19 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
xorg.0.log (24.41 KB, text/plain)
2008-11-06 23:47 UTC, liuhaien
Details
xorg conf file (3.26 KB, text/plain)
2008-11-06 23:47 UTC, liuhaien
Details
xorg.0.log with tiling on (18.71 KB, text/plain)
2009-03-15 18:56 UTC, liuhaien
Details
screenshot (125.31 KB, image/png)
2009-03-15 18:57 UTC, liuhaien
Details
xorg.0.log with Modedebug on (44.90 KB, text/plain)
2009-03-16 22:21 UTC, liuhaien
Details

Note You need to log in before you can comment on or make changes to this bug.
Description liuhaien 2008-11-06 23:47:17 UTC
Created attachment 20123 [details]
xorg.0.log

System Environment:
--------------------------

--Platform:915gv

--Architecture(32-bit,64-bit,compatiblity):32bit 

GEM_kernel (drm-intel-next)  2db46b5668dfed59acacd75dfc49ecf947ea5aeb
Libdrm:         (master)87e90c73620b88005fcca5fd40aaaad0b08932e1
Mesa:           (master)df94fd17647937975df031dcaa1ac24b2d79ce1b
Xserver:                (master)b4ca9dc2393ace2415163094b916f0d17ecba9ee
Xf86_video_intel:              
(master)226efa7c74c0ed8a7b974a3727c5d7322c8ef815


Bug detailed description:
-------------------------
  startx ,run glean cases.on 915gv, a lot of glean cases run failed,for
example,blendFunc,fragProg1,logicOp,orthoPosRandTris,orthoPosRandRects ...and
so on.we just get 13 pass in 42 cases.
Comment 1 liuhaien 2008-11-06 23:47:35 UTC
Created attachment 20124 [details]
xorg conf file
Comment 2 Eric Anholt 2008-11-25 02:08:05 UTC
(insert usual question here about readpix demo)
Comment 3 Gordon Jin 2008-12-05 22:43:26 UTC
Haien, please try readpix, though I know it's troublesome as the machine is in AutoDTI lab.
Comment 4 liuhaien 2008-12-09 00:20:34 UTC
(In reply to comment #3)
> Haien, please try readpix, though I know it's troublesome as the machine is in
> AutoDTI lab.
> 

readpix and gearbox also failed.
Comment 5 liuhaien 2008-12-09 01:17:00 UTC
(In reply to comment #4)
> (In reply to comment #3)
> > Haien, please try readpix, though I know it's troublesome as the machine is in
> > AutoDTI lab.
> > 
> 
> readpix and gearbox also failed.
> 

btw,readipx and gearbox can work well if set tiling off.
Comment 6 Eric Anholt 2009-03-13 16:24:16 UTC
Please attach the screenshot of readpix, as well as dmidecode output.  From the registers, it looks like you've got a single channel of memory and no extra swizzling occurring, and the code at that time should have handled it correctly.
Comment 7 liuhaien 2009-03-15 18:50:39 UTC
dmidecode |grep -A16 "Memory Device$" shows
Memory Device
        Array Handle: 0x1000
        Error Information Handle: Not Provided
        Total Width: 64 bits
        Data Width: 64 bits
        Size: 256 MB
        Form Factor: DIMM
        Set: None
        Locator: DIMM_1
        Bank Locator: Not Specified
        Type: DDR
        Type Detail: Synchronous
        Speed: 533 MHz (1.9 ns)
        Manufacturer: FFFFFFFFFFFFFFFF
        Serial Number: FFFFFFFF
        Asset Tag: Not Specified
        Part Number:
--
Memory Device
        Array Handle: 0x1000
        Error Information Handle: Not Provided
        Total Width: 64 bits
        Data Width: 64 bits
        Size: 256 MB
        Form Factor: DIMM
        Set: None
        Locator: DIMM_2
        Bank Locator: Not Specified
        Type: DDR
        Type Detail: Synchronous
        Speed: 533 MHz (1.9 ns)
        Manufacturer: CE00000000000000
        Serial Number: 73111D28
        Asset Tag: Not Specified
        Part Number: M3 78T3354CZ3-CD5
Comment 8 liuhaien 2009-03-15 18:56:16 UTC
Created attachment 23891 [details]
xorg.0.log with tiling on
Comment 9 liuhaien 2009-03-15 18:57:19 UTC
Created attachment 23892 [details]
screenshot
Comment 10 liuhaien 2009-03-15 19:22:40 UTC
sorry,this is 915gv, not 915gm. The previous log was wrong
Comment 11 Eric Anholt 2009-03-16 10:44:51 UTC
Still requires ModeDebug.
Comment 12 liuhaien 2009-03-16 22:21:12 UTC
Created attachment 23952 [details]
xorg.0.log with Modedebug on
Comment 13 liuhaien 2009-03-16 22:21:57 UTC
(In reply to comment #11)
> Still requires ModeDebug.
> 

ok, Xorg.0.log with Modedebug on is re-attached
Comment 14 Eric Anholt 2009-03-23 18:13:19 UTC
Likely fixed by:

commit a243a32388e4828dc5fe8e9dd129a3042bf5b6b3
Author: Eric Anholt <eric@anholt.net>
Date:   Thu Mar 12 16:27:11 2009 -0700

    drm/i915: Change DCC tiling detection case to cover only mobile parts.
    
    Later spec investigation has revealed that every 9xx mobile part has
    had this register in this format.  Also, no non-mobile parts have been shown
    to have this register.  So make all mobile use the same code, and all
    non-mobile use the hack 965 detection.
    
    Signed-off-by: Eric Anholt <eric@anholt.net>
Comment 15 liuhaien 2009-03-25 00:10:08 UTC
still fails against this commit.so reopen it.
Comment 16 Gordon Jin 2009-03-30 18:26:22 UTC
remove from Q1 release blocker
Comment 17 Eric Anholt 2009-10-30 17:36:46 UTC
commit 7c8bed62e0165a0be3363f7abf81bf9e30341e00
Author: Eric Anholt <eric@anholt.net>
Date:   Fri Oct 30 15:33:11 2009 -0700

    intel: Use GTT mapping when available for swrast.
    
    This improves piglit quick.tests runtime from 19:33 minutes to 6:06 on
    my GM45.  It should also hide most of the A17 swizzling issues, though
    they'll still exist when swapping occurs (which is the kernel's problem
    either way).
Comment 18 Gordon Jin 2009-11-03 01:19:05 UTC
this machine is not in our control now, closing.