Bug 1131 - target $var writing (target lvalue contexts)
Summary: target $var writing (target lvalue contexts)
Status: RESOLVED FIXED
Alias: None
Product: systemtap
Classification: Unclassified
Component: translator (show other bugs)
Version: unspecified
: P2 normal
Target Milestone: days
Assignee: Graydon Hoare
URL:
Keywords:
Depends on: 905
Blocks:
  Show dependency tree / graph
 
Reported: 2005-07-28 16:48 UTC by Graydon Hoare
Modified: 2005-12-13 19:00 UTC (History)
0 users

See Also:
Host:
Target:
Build:
Last reconfirmed:

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Graydon Hoare 2005-07-28 16:48:45 UTC 
currently $var is only legal in an rvalue context; add some support for writing
to target vars
Comment 1 Graydon Hoare 2005-10-01 00:12:58 UTC 
It's actually somewhat hard to test this feature, but as far as I know it's now
working. By "working" I mean: writing using operator =, to base types. No
support for += or other op= forms at the moment, and no support for pointer
types or strings.
Comment 2 Frank Ch. Eigler 2005-10-03 18:05:40 UTC 
Do the not-working parts behave like clean failures?  Testsuite?  Is guru mode
enforced for $var lvalue operations?
Comment 3 Graydon Hoare 2005-10-05 00:47:16 UTC 
The not-working parts fail as exceptions, yes. I've now predicated it on guru
mode, and added testsuite/buildok/twenty.stp (previously I was hunting around
for a test which would be safe to *run*, which is hard to find. most writing to
the kernel is sort of dangerous!)
Comment 4 Frank Ch. Eigler 2005-12-13 19:00:24 UTC 
belated