Probably need to address these to both raman and sudu Please have a look at the file(s) below to see if we can improve reentrantcy across the code base by making modifications to: caps/include/nsCCapsManagerFactory.h: static nsIFactory *m_pNSIFactory; caps/include/nsPrivilege.h: static PRBool theInited; caps/include/nsPrivilegeManager.h: static PRBool theSecurityInited; caps/include/nsPrivilegeManager.h: static char * SignedAppletDBName; caps/include/nsPrivilegeManager.h: static PRBool theInited; caps/include/nsTarget.h: static PRBool theInited;

RX tasks aren't going to make M3. close the reentrantcy tracking tasks if this specific area has been looked at and doesn't seem to be a problem. Otherwise each of these areas still need some looking at.

moving to m5. raman can you have a look at these files and see if they can be cleaned up to comply with the reentrancy goals... A reentrant function f is a nested function (not a leaf; it calls other functions) that may be reentered (indirect recursion) from under any other function g() that it calls. Therefore f must avoid sampling a static or global variable and holding it across such a call to g(), if the recursive f called from g could modify the static and vary conditions that f believes are invariant. We believe much of our non-static/global data is partitioned by window, not reachable by other windows except through statics/globals. Therefore we're starting the hunt with statics and globals. Please have a look at the file(s) below to see if we can improve reentrantcy across the code base by making modifications to:

I apolosize for not taking a look at any of the CAPS related bugs (I have been swamped with pushing the netcenter products out. I am close to shipping my 2nd product out and one more product to go. I am on the hook here until the last guy is out. I really apolosize for not being able to look at CAPS stuff). I will try to do a pull and build of mozilla this week end. If someone already has a tree, I could come over and knock these problems off within couple of hours. I would really appreciate that. thanks very much, raman

raman, any update on this and other CAPPs work? I think we are going to need CAPPs support in one of the upcoming milestones

raman, any update on this and other CAPPs work? I think we are going to need CAPPs support in one of the upcoming milestones

moving to norris, and taking off the m6 list

Accepting bug and changing target milestone based upon current schedule rollup.

OK, i've been reading over the comments and i'm not quite sure i understand the overall impact of reentrant functions though i think i get the idea. in terms of the state of the statics listed below here is a short status update. caps/include/nsCCapsManagerFactory.h: static nsIFactory *m_pNSIFactory; --- this file doesn't exist anymore so neither does this static var caps/include/nsPrivilege.h: static PRBool theInited; --- this boolean has been removed, i'm not sure if it ever served a purpose. caps/include/nsPrivilegeManager.h: static PRBool theSecurityInited; --- this static var no longer exists, with the new factory scheme, there exists a static method var -- -in GetPrivilegeManager(), maybe you should check and see if that's kosher. i think it's alright. caps/include/nsPrivilegeManager.h: static char * SignedAppletDBName; ---this one still exists, i'm not sure what it does so idon't what to get rid of it just yet, i'll take a look at ---this one soon caps/include/nsPrivilegeManager.h: static PRBool theInited; ---gone caps/include/nsTarget.h: static PRBool theInited; ---still there, haven't really gotten around to fixing up nsTarget API yet, ---improvements should be coming soon

Atthis point the only static variable in caps is static nsScriptSecurityManager *ssecMan = NULL; which is used to implement the singleton status of nsScriptSecurityManager. jband said this was an acceptable use until the new glorious factory code is done.