harmony-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mikhail Fursov (JIRA)" <j...@apache.org>
Subject [jira] Updated: (HARMONY-5119) [drlvm][jit][opt] Bug in assigning opcode group to the instruction with operands having different sizes
Date Wed, 14 Nov 2007 11:38:43 GMT

     [ https://issues.apache.org/jira/browse/HARMONY-5119?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Mikhail Fursov updated HARMONY-5119:
------------------------------------

    Estimated Complexity: Advanced
                 Summary: [drlvm][jit][opt] Bug in assigning opcode group to the instruction
with operands having different sizes  (was: [drlvm][jitrino] bug in assigning opcode group
to the instruction with operands having different sizes)

> [drlvm][jit][opt] Bug in assigning opcode group to the instruction with operands having
different sizes
> -------------------------------------------------------------------------------------------------------
>
>                 Key: HARMONY-5119
>                 URL: https://issues.apache.org/jira/browse/HARMONY-5119
>             Project: Harmony
>          Issue Type: Bug
>          Components: DRLVM
>         Environment: IA32
>            Reporter: Dmitry Pronichkin
>            Assignee: Mikhail Fursov
>         Attachments: opcode_group.patch
>
>
> Currently IA32 Encoder erroneously assigns opcode group to the instruction that has different
opcodes, if operand sizes are different. The problem is in method bool Encoder::matches(const
OpcodeGroup* og, const FindInfo& fi, bool any), that maps operand indices from extended
to native form. This sometimes leads to ignoring the last operand and assigning the wrong
opcode group.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message