[Mono-devel-list] fixed statement

[Eric Durand Tremblay]

Do you or anybody know how to do this with Reflection.Emit (or something 
else).   For the compiled regex, we use pointer.  I don't want our class 
to break when the gc will change.

Eric Durand-Tremblay

Bernie Solomon a écrit :

>This should be a property of the local variable - it should be
>flagged as pinned but mcs doesn't appear to do this.
>
>However since mono doesn't currently using a compacting
>garbage collector it doesn't matter if compiling and running
>with mono alone as data won't move anyway.
>
>Of course different runtimes do compact and mcs compiled
>code could then fail so mcs does need to be fixed ;-) too.
>
>Bernie Solomon
>
>----- Original Message ----- 
>From: "Eric Durand Tremblay" <eric.durand-tremblay.1 at ulaval.ca>
>To: "Mono Development" <mono-devel-list at lists.ximian.com>
>Sent: Wednesday, March 17, 2004 1:42 PM
>Subject: [Mono-devel-list] fixed statement
>
>
>  
>
>>Can somebody tell me how mono handle fixed statement.
>>
>>For something like this :
>>
>>string bar = "something";
>>
>>fixed ( char * foo = bar) {
>>    foo[0] = 'c';
>>}
>>
>>MCS output :
>>
>>        IL_0000:  ldstr "bar"
>>        IL_0005:  stloc.1
>>        IL_0006:  ldloc.1
>>        IL_0007:  stloc.2
>>        IL_0008:  ldloc.2
>>        IL_0009:  conv.i
>>        IL_000a:  call int32 class
>>
>>    
>>
>[mscorlib]'System.Runtime.CompilerServices.RuntimeHelpers'::'get_OffsetToStr
>ingData'()
>  
>
>>        IL_000f:  add
>>        IL_0010:  stloc.0
>>        IL_0011:  ldloc.0
>>        IL_0012:  conv.i
>>        IL_0013:  ldc.i4.0
>>        IL_0014:  ldc.i4.2
>>        IL_0015:  mul
>>        IL_0016:  conv.i
>>        IL_0017:  add
>>        IL_0018:  ldc.i4.s 0x63
>>        IL_001a:  stind.i2
>>        IL_001b:  ldnull
>>        IL_001c:  stloc.2
>>        IL_001d:  ret
>>
>>There is nothing in this that "lock" the garbage collector
>>(msdn define lock as: Prevents relocation of a variable by the garbage
>>collector.)
>>
>>Thanks
>>
>>Eric Durand-Tremblay
>>
>>_______________________________________________
>>Mono-devel-list mailing list
>>Mono-devel-list at lists.ximian.com
>>http://lists.ximian.com/mailman/listinfo/mono-devel-list
>>
>>    
>>
>
>
>  
>