[haiku-bugs] Re: [Haiku] #8007: [app_server] fully unresponsive when resizing window of big textfile
- From: "jua" <trac@xxxxxxxxxxxx>
- Date: Sat, 06 Oct 2012 14:04:14 -0000
#8007: [app_server] fully unresponsive when resizing window of big textfile
-----------------------------+----------------------------
Reporter: ttcoder | Owner: axeld
Type: bug | Status: new
Priority: normal | Milestone: R1
Component: System/Kernel | Version: R1/Development
Resolution: | Keywords:
Blocked By: | Blocking: 7882, 8136
Has a Patch: 1 | Platform: All
-----------------------------+----------------------------
Comment (by jua):
Hey bonefish! Yeah you are right that the SPLASH-2 suite(*) is not meant
for testing locking primitives themselves, it's for testing concurrent
calculation performance. They however do make extensive use of various
locking primitives, although in the version I'm using it's pthreads, not
the kernel locks. (If the kernel locks were public, it would actually be
possible to use them directly: the modified splash-2 uses generalized
macros for synchronization primitives which are replaced by the actual
synchronization API calls by an M4 script before compiling. They supply an
M4 script for pthreads, but it would be possible to create a new one to
make them use the kernel locks directly...)
The reason why I thought SPLASH-2 benchmarks would be still relevant is
that many of them cause intense freezing and thus I figured there is some
connection to the kernel lock contention too, just like when building a
Haiku image. I'm now preparing two more images (patched/unpatched) to test
compiling Haiku with both.. but that will take some hours. I will report
back when I have results.
(*) I'm using the modified version found here:
http://www.capsl.udel.edu/splash/
--
Ticket URL: <http://dev.haiku-os.org/ticket/8007#comment:65>
Haiku <http://dev.haiku-os.org>
Haiku - the operating system.
Other related posts:
