JDK-4318701 : -Xss request larger than shell-imposed limit causes SEGV/assert
Type:Bug
Component:hotspot
Sub-Component:runtime
Affected Version:1.3.0
Priority:P4
Status:Closed
Resolution:Fixed
OS:generic
CPU:generic
Submitted:2000-03-03
Updated:2012-10-08
Resolved:2000-03-06
The Version table provides details related to the release that this issue/RFE will be addressed.
Unresolved : Release in which this issue/RFE will be addressed. Resolved: Release in which this issue/RFE has been resolved. Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.
Need code to check that the main thread doesn't inadvertently try using more
stack that available via shell-imposed limit.
Comments
CONVERTED DATA
BugTraq+ Release Management Values
COMMIT TO FIX:
kestrel-solaris
FIXED IN:
kestrel-solaris
INTEGRATED IN:
kestrel-solaris
14-06-2004
SUGGESTED FIX
As in description; a 2-line change in os_solaris.cpp::(os::initialize_thread).
Date: Fri, 3 Mar 2000 19:54:04 -0800 (PST)
From: "Y. S. Ramakrishna" <###@###.###>
Subject: Re: another very simple review
To: ###@###.###
MIME-Version: 1.0
Content-MD5: crL0ovjFNCdQJwsw550vEQ==
Oops.. here's the webrev pointer:
http://analemma.eng/net/jde1.eng/export/jtech1/ysr/HSPOT/webrev/index.html
I also included one other change: turning on UseStrictCompilerSafepoints
as default, as a "fix" for
4317961 [sparc] Compiler safepoints do not work
I tested with Volano, SPEC, and Java2D demo, as well as the
test "Srdjan" supplied by Srdjan, all with both -Xcomp and in
mixed mode, using compiler2. I believe it is safe to
turn on this compiler flag. However, i am not removing
it completely yet, just in case we discover a regression,
and need to experiment some more in the future.
As far as the performance implication of the above change
is concerned, no change was observed in either SPEC or
in Volano scores.
> Date: Fri, 3 Mar 2000 19:23:12 -0800 (PST)
> From: "Y. S. Ramakrishna" <###@###.###>
> Subject: another very simple review
> To: ###@###.###
> MIME-Version: 1.0
> Content-MD5: UKZvQ4HoBLywN+cpGiXYOA==
>
>
>
> 4318701 -Xss request larger than shell-imposed limit causes SEGV/assert
>
> We now check (in the main thread's os::thread_initialize, which is
> Solaris-specific) that the requested stack size doesn't exceed the
> stack size limit for this thread as imposed by the shell. If so,
> we exit immediately with an appropriate error message. A less
> draconian possibility would have been to (effectively) ignore
> the -Xss request and use the shell-imposed limit in such a case,
> but it can be argued that users might not want to be thus misled.
>
> Tests: tested various large settings of stack while
> adjusting the shell-imposed limit;
> ran Volano.
>
Date: Sun, 5 Mar 2000 09:40:37 -0800 (PST)
From: "Y. S. Ramakrishna" <###@###.###>
To: ###@###.###
Subject: Code Manager notification
Mime-Version: 1.0
Event: putback-to
Parent workspace: /net/jano/export/disk05/hotspot/ws/main/baseline
(jano:/export/disk05/hotspot/ws/main/baseline)
Child workspace: /net/jde1/export/jtech1/ysr/HSPOT
(jde1:/export/jtech1/ysr/HSPOT)
User: ysr
Comment:
4317961 [sparc] Compiler safepoints do not work
4318701 -Xss request larger than shell-imposed limit causes SEGV/assert
Details:
4317961: Turn on UseStrictCompilerSafepoints as default. [globals.hpp]
4318701: In os_solaris.cpp [os::initialize_thread] add appropriate checks.
Tests: Volano, SPEC, Srdjan (test supplied by Srdjan)
Performance: no change observed in Volano (-Xcomp & mixed), using compiler2
Reviewed by: Srdjan
Files:
update: src/os/solaris/vm/os_solaris.cpp
update: src/share/vm/runtime/globals.hpp
11-06-2004
EVALUATION
y.s.ramakrishna@eng 2000-03-03: fixed as in description. Fix will be put back
after some testing.