In my continuing quest to get the HotSpot Java virtual machine to work under valgrind, I found that it wants to execute this opcode:
0x66 0xDD 0x4 This is (as far as I have been able to determine) a floating point load (fldl) with a 0x66 size prefix. Now the size prefix is useless with this instruction. The AMD decoder in the svn head of valgrind explicitly says that putting 0x66 with any of the fpu-related opcodes is bad. However, I needed to put in special dispensation for fnsave and frstor, since 0x66 has meaning for them (and HotSpot appears to use that sequence). The AMD manual explicitly says that 0x66 is *ignored* if presented with fpu opcodes on which it has no effect. So ... I am running tests with the decoder revised to allow 0x66 on the fpu instructions. Does anyone know of a reason why doing this would be bad/wrong? Regards -- Eliot ------------------------------------------------------------------------------ Keep Your Developer Skills Current with LearnDevNow! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-d2d _______________________________________________ Valgrind-users mailing list Valgrind-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/valgrind-users
