|
Patch 2739 : RealityEngine rollup #4 for IRIX 6.2 : [IRIX 6.2]
INDEX
RELATIONS
RELEASE NOTES
1. Patch SG0002739 Release Note
This release note describes patch SG0002739 to IRIX 6.2.
The primary purpose of patch SG0002739 is to consolidate a
number of RealityEngine bug fixes that were applied after
the release of IRIX 6.2. It replaces patch SG0001442, patch
SG0001702, and patch SG0002038 which previously served that
purpose.
In addition to bug fixes, patch SG0002739 also provides
significant performance improvements for the IrisGL and
OpenGL groups of commands that are detailed below.
1.1 Supported Hardware Platforms
This patch is for Onyx/PowerOnyx RealityEngine2 and VTX, and
Crimson RealityEngine. The software cannot be installed on
other configurations.
1.2 Supported Software Platforms
Patch SG0002739 contains bug fixes for the Execution
Environment (eoe) of a system running IRIX 6.2. The
software cannot be installed on other configurations.
1.3 Bugs Fixed by Patch SG0002739
This patch contains fixes for the following bugs in IRIX
6.2. Bug numbers from the Silicon Graphics bug tracking
system are included for reference.
o Graphics hang on REII IP19 when loading texture memory.
Bug number 472831.
o DoubleBuffer application goes SingleBuffered in stereo.
Bug number 481989.
o Pixmap rendering is broken on RealityEngine. Bug
number 454622
o Incorrect primitive lighting when using
GL_AMBIENT_AND_DIFFUSE parameter to glColorMaterial().
Bug number 341355.
o A pipe hang caused by X server drawing large bitmap
images. Bug number 419755.
o Incorrect X server drawing of bitmaps under certain
circumstances. Bug number 372836.
o Corruption of dynamically allocated graphics subsystem
memory (heap memory) that occurred when multiple
graphics contexts allocate memory (i.e., for histogram,
pixel maps, OpenGL's attribute stack etc.) such that
the total size of the graphics subsystem heap exceeds
the available heap memory. Under certain circumstances
such corruption would cause a pipe hang. Bug number
358341.
o Incorrect restoration of many parts of OpenGL state
after glPopAttrib(). Examples are the stencil
writemask, user clip planes, TexGen environments, etc.
Bug numbers 406391, 407193.
o Incorrect handling of glCopyPixels() with convolution
enabled. In addition to possible pipe hangs the
determination of the ability to use optimized path
based on the values of the color matrix and color
buffer write mask was incorrect. The failure to select
an optimized paths had a very significant performance
implication. Bug numbers 365068 and 412633.
o Incorrect handling of GL_FUNC_REVERSE_SUBTRACT_EXT
blend function. Bug number 370840.
o Incorrect handling of OpenGL's 1D and 2D evaluators'
enable state. Bug number 357517.
o Incorrect error detection and handling for a variety of
OpenGL commands both for illegal command parameter
values and for issuing commands that are not allowed
inside glBegin(...)/glEnd() while inside a
glBegin(...)/glEnd(). One of the problems that has
been fixed is described by bug number 372086.
o glGetTexGen() of the eye or object plane equations
always returned the default values rather than the
actual equations set by the user (although these
equations were used when generating texture
coordinates). Bug number 407195.
o glGetClipPlane() would return incorrect results when
more than 16 graphics contexts were running
simultaneously (the actual equations used by the system
were however correct). Bug number 421407.
o Incorrect handling of linear fog; bug number 374451.
Note that there were two issues in this case. First,
there was a bug that caused certain fog parameters to
be corrupted; this has been fixed. Second, there was
no support for the IB2 chip, which has significantly
improved fog capability compared to the older IB1. In
order to preserve compatibility with previous releases,
IB2 capability is not enabled by default. One must set
the environment variable RE_FOG_MODE to the value 2 in
order to enable it. Note that IB2 capability must be
used if compatibility between RealityEngine and
InfiniteReality is desired.
o There was a memory leak in glColorTableSGI() when
loading texture color lookup tables. Bug number
437745.
o Iconifying and de-iconifying a running application
could cause a segmentation violation. The conditions
needed to cause this are exceedingly rare, except in
real-time applications running with non-degrading
priority. Bug number 452106.
o Cursor colors with component values greater than 127
were handled incorrectly. (No associated bug number.)
o Suspension and resumption of selection and feedback
modes were handled incorrectly. Bug number 458806.
o A problem (introduced in patch 1442) with accidental
enabling of user clip planes was fixed. This bug was
manifest as incorrect feedback results after user clip
planes had been set. (No associated bug number.)
o When using pixie with N32 OpenGL programs, the pixied
program dumps core. Bug number 372990.
o Use of a glDrawPixels() command with an image of a
particular size could cause display-list corruption,
leading to a segmentation fault. Bug number 433594.
o The change introduced in patch SG0001702 to fix bug
374451 accidentally created new problems with
exponential and EXP2 fog, for both OpenGL and IrisGL.
Bug number 467103.
1.4 Performance improvements and functional enhancements inPatch SG00027391.4.1 Performance improvements
o The performance of IrisGL and OpenGL mode changes
(i.e., setting of the color mask, depth test, blending
and fog functions, enables, material binding, etc.),
and OpenGL matrix queries have been improved by more
than 30%.
o OpenGL switching between different contexts
(glMakeCurrent/Read) performance has been improved by
about 25%.
o The performance of OpenGL pushing and popping of the
current state (glPush/PopAttrib(GL_CURRENT_BIT)) has
been improved by a factor of more than 15.
o The performance of OpenGL and IrisGL evaluators has
been improved by more than 40%.
o glDrawPixels of small images (< 50K bytes) has been
tuned.
1.4.2 Functional enhancements
o The size of the available graphics subsystem dynamic
memory (heap memory) has been doubled. Thus, for
example, OpenGL applications can use a larger number of
big pixel lookup tables (PixelMaps) simultaneously with
a large histogram.
o The maximum width of an OpenGL viewport has been
increased from 2048 to 2560. Bug number 448477.
o IB2-style fog (more accurate than the default, provided
your system has revision 2 IB chips) is now enabled by
setting the RE_FOG_MODE environment variable to 2. See
notes under ``Bugs Fixed'' above.
1.5 Subsystems Included in Patch SG0002739
This patch release includes these subsystems:
o patchSG0002739.eoe_sw.gfx
o patchSG0002739.eoe_sw.unix
o patchSG0002739.nfs_sw.dskless_client
1.6 Installation Instructions
Because you want to install only the patches for problems
you have encountered, patch software is not installed by
default. After reading the descriptions of the bugs fixed
in this patch (see Section 1.3), determine the patches that
meet your specific needs.
If, after reading Sections 1.1 and 1.2 of these release
notes, you are unsure whether your hardware and software
meet the requirements for installing a particular patch, run
inst. The inst program does not allow you to install
patches that are incompatible with your hardware or
software.
Patch software is installed like any other Silicon Graphics
software product. Follow the instructions in your Software
Installation Administrator's Guide to bring up the miniroot
form of the software installation tools.
Follow these steps to select a patch for installation:
1. At the Inst> prompt, type
install patchSGxxxxxxx
where xxxxxxx is the patch number.
2. Initiate the installation sequence. Type
Inst> go
3. You may find that two patches have been marked as
incompatible. (The installation tools reject an
installation request if an incompatibility is
detected.) If this occurs, you must deselect one of
the patches.
Inst> keep patchSGxxxxxxx
where xxxxxxx is the patch number.
4. After completing the installation process, exit the
inst program by typing
Inst> quit
1.7 Patch Removal Instructions
To remove a patch, use the versions remove command as you
would for any other software subsystem. The removal process
reinstates the original version of software unless you have
specifically removed the patch history from your system.
versions remove patchSGxxxxxxx
where xxxxxxx is the patch number.
To keep a patch but increase your disk space, use the
versions removehist command to remove the patch history.
versions removehist patchSGxxxxxxx
where xxxxxxx is the patch number.
INST SUBSYSTEM REQUIREMENTS No Requirements Information Available. INST SUBSYSTEM CHECKSUMS These checksums help to provide a 'signature' for the patch inst image which can be used to authenticate other inst images. You can obtain this kind of output by running sum -r on the image (from the command line): 39545 6 patchSG0002739 06864 17059 patchSG0002739.eoe_sw 57365 10877 patchSG0002739.eoe_sw32 55153 8447 patchSG0002739.eoe_sw64 13955 74 patchSG0002739.idb 11444 16 patchSG0002739.nfs_sw INST SUBSYSTEM FILE LISTINGS The following lists the files which get installed from each subsystem in the patch:
DOWNLOAD PATCH
|
||||||||||||||||||||||||||||||||||||
Document Id: 20021117070727-IRIXPatch-1164
|
||||||||||||||||||||||||||||||||||||