Commit Graph

1810 Commits

Author SHA1 Message Date
Chris Lattner
9982e211ce Move sparc-specific livevar code into lib/Target/Sparc
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10733 91177308-0d34-0410-b5e6-96231b3b80d8
2004-01-09 18:15:24 +00:00
Chris Lattner
1d415a967b Fix more incestuous #includage.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10732 91177308-0d34-0410-b5e6-96231b3b80d8
2004-01-09 16:17:09 +00:00
Chris Lattner
46de01e0c8 Move InstrSelection into lib/Target/Sparc, as it's sparc specific. This
makes the incestuous #include'ing of sparc internal headers much less
disturbing.  :)


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10729 91177308-0d34-0410-b5e6-96231b3b80d8
2004-01-09 06:22:34 +00:00
Chris Lattner
75e260990d Move lib/Codegen/RegAlloc into lib/Target/Sparc, as it is sparc specific
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10728 91177308-0d34-0410-b5e6-96231b3b80d8
2004-01-09 06:17:12 +00:00
Chris Lattner
f70e0c216c Clean up a lot of the code I added yesterday by exposing the IntrinsicLowering
implementation from the TargetMachine directly.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10636 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-28 21:23:38 +00:00
Alkis Evlogimenos
5e30002af7 Add TargetInstrInfo::isMoveInstr() to support coalescing in register
allocation.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10633 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-28 17:35:08 +00:00
Chris Lattner
aed386e6b0 Whoops, don't try to lower non intrinsic calls
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10632 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-28 09:53:23 +00:00
Chris Lattner
4482715f3d implement support for the intrinsic lowering functionality
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10629 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-28 09:47:19 +00:00
Chris Lattner
37b1826aab Eliminate some code that is not needed now that we have the intrinsic lowering pass
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10628 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-28 09:46:33 +00:00
Chris Lattner
f8742b3e03 Use the intrinsic lowering functionality
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10626 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-28 09:43:35 +00:00
Chris Lattner
f045328dd1 finegrainify namespacification
add new getIntPtrType() method


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10579 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-22 05:01:15 +00:00
Misha Brukman
3615845b3e Really release memory used by functions. Patch by Chris.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10572 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-22 03:47:58 +00:00
Alkis Evlogimenos
0ef76cadbb Move FP_REG_KILL closer to the return instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10567 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-21 16:47:43 +00:00
Alkis Evlogimenos
9abc81790e Move FP_REG_KILL closer to the actual branch instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10563 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-20 17:28:15 +00:00
Alkis Evlogimenos
e0bb3e766d Remove floating point killer pass. This is now implemented in the
instruction selector by adding a new pseudo-instruction
FP_REG_KILL. This instruction implicitly defines all x86 fp registers
and is a terminator so that passes which add machine code at the end
of basic blocks (like phi elimination) do not add instructions between
it and the branch or return instruction.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10562 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-20 16:22:59 +00:00
Chris Lattner
655239cc6b Finally, _actually delete the machine code_ for a function, after it has
been emitted.  Also, since the FPK pass is causing memory access violations,
disable it.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10559 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-20 10:20:19 +00:00
Chris Lattner
9068574a29 Fix memory leak in the stackifier, due to the machinebasicblocks not holding
instructions on an ilist


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10556 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-20 10:12:17 +00:00
Chris Lattner
f2e49d4c1a Finegrainify namespacification
Minor cleanups to killer pass


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10555 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-20 09:58:55 +00:00
Chris Lattner
583b9d8455 Hoist some sparc specific code into the sparc target
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10554 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-20 09:17:40 +00:00
Chris Lattner
61838824c2 Remove a horrible API
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10551 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-20 09:15:01 +00:00
Chris Lattner
2828970819 A minor cleanup for better encapsulation
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10545 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-20 02:03:14 +00:00
Chris Lattner
1e60a9165d Rip JIT specific stuff out of TargetMachine, as per PR176
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10542 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-20 01:22:19 +00:00
Misha Brukman
97f9cf2de2 * Converted C-style comments to C++
* Doxygenified comments
* Reordered #includes


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10503 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-17 22:08:20 +00:00
Misha Brukman
3d0ad411a5 Reordered #includes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10502 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-17 22:06:28 +00:00
Misha Brukman
c3402cd735 Doxygenified some comments, reduced extraneous space.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10501 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-17 22:06:08 +00:00
Misha Brukman
d71295a684 Reorganized the Sparc backend to be more modular -- each different
implementation of a Target{RegInfo, InstrInfo, Machine, etc} now has a separate
header and a separate implementation file.

This means that instead of a massive SparcInternals.h that forces a
recompilation of the whole target whenever a minor detail is changed, you should
only recompile a few files.

Note that SparcInternals.h is still around; its contents should be minimized.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10500 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-17 22:04:00 +00:00
Alkis Evlogimenos
4d7af65903 Change interface of MachineOperand as follows:
a) remove opIsUse(), opIsDefOnly(), opIsDefAndUse()
    b) add isUse(), isDef()
    c) rename opHiBits32() to isHiBits32(),
              opLoBits32() to isLoBits32(),
              opHiBits64() to isHiBits64(),
              opLoBits64() to isLoBits64().

This results to much more readable code, for example compare
"op.opIsDef() || op.opIsDefAndUse()" to "op.isDef()" a pattern used
very often in the code.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10461 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-14 13:24:17 +00:00
Alkis Evlogimenos
97323a47d8 Change preserve all claim to just preserve live variables and phielimination.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10460 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-14 10:14:23 +00:00
Alkis Evlogimenos
359b65f782 Add a floating point killer pass. This pass runs before register
allocaton on the X86 to add information to the machine code denoting
that our floating point stackifier cannot handle virtual point
register that are alive across basic blocks. This pass adds an
implicit def of all virtual floating point register at the end of each
basic block.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10446 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-13 05:36:22 +00:00
John Criswell
52b62de100 Modified cast so that it converts the int to a long before casting to
a pointer.  This evades a warning emitted by GCC when we cast from
unsigned int (32 bit) to void * (64 bit) on SparcV9.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10435 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-12 19:57:48 +00:00
Chris Lattner
65b05ce022 Implement the TargetMachine::getJITStubForFunction method for X86, finegrainify
namespacification.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10430 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-12 07:11:18 +00:00
Brian Gaeke
27f7a71a25 Turn off "attribute weak" to pacify Mac OS X's system compiler, which prints a
warning whenever it sees it.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10391 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-11 00:24:36 +00:00
John Criswell
f5ba89df1f This appears to fix Bug 172 and does not break any other feature tests or
regression tests.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10388 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-10 22:51:41 +00:00
John Criswell
0284a42fd3 Upon Chris's suggestion, moved the #ifdef's to the generated C code.
This makes LLVM and the generated C code more portable.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10377 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-10 18:17:57 +00:00
John Criswell
01b2d61311 Fixed the CBE on Solaris/Sparc. We need to define the return value of
the write() system call because it returns 64 bits on Solaris 64 bit,
and an implicit return value of int says it returns 32 bits.
Admittedly, this is a bit of a hack.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10375 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-10 18:06:34 +00:00
Chris Lattner
9b527704f7 Add an option to enable the SSA based peephole optimizer.
Eventually this pass will provide substantially better code in the interim between when we
have a crappy isel and nice isel.  Unfortunately doing so requires fixing the backend to
actually SUPPORT all of the fancy addressing modes that we now generate, and writing a DCE
pass for machine code.  Each of these is a fairly substantial job, so this will remain disabled
for the immediate future.  :(


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10276 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-01 05:18:30 +00:00
Chris Lattner
4537076ee8 Add a new SSA-based peephole optimizer which includes copy propagation and
folding of instructions into addressing modes.  This creates lots of dead
instructions, which are currently not deleted.  It also creates a lot of
instructions that the X86 backend currently cannot handle.  :(


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10275 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-01 05:15:28 +00:00
Chris Lattner
f2d2925452 generalize the instruction types permitted a bit
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10274 91177308-0d34-0410-b5e6-96231b3b80d8
2003-12-01 05:13:56 +00:00
Chris Lattner
e1cc79f1cf Finegrainify namespacification
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10270 91177308-0d34-0410-b5e6-96231b3b80d8
2003-11-30 06:13:25 +00:00
Chris Lattner
4f7e173cbd output foo(void) as appropriate
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10232 91177308-0d34-0410-b5e6-96231b3b80d8
2003-11-26 00:09:17 +00:00
Chris Lattner
bc771c5608 Fix bug in previous checkin
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10231 91177308-0d34-0410-b5e6-96231b3b80d8
2003-11-25 23:44:40 +00:00
Chris Lattner
23e907082a Do not depend on the gep index types to determine what flavor of index it is
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10225 91177308-0d34-0410-b5e6-96231b3b80d8
2003-11-25 20:49:55 +00:00
Brian Gaeke
f9c86cf5cf Support constant casting constant pointers to ints/uints, and/or
the other way around, instead of failing a large, tumor-like assertion.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10171 91177308-0d34-0410-b5e6-96231b3b80d8
2003-11-22 07:18:25 +00:00
Brian Gaeke
dfcc9cfaa8 Constant shift expressions, meet InstSelectSimple. Yow!!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10170 91177308-0d34-0410-b5e6-96231b3b80d8
2003-11-22 06:49:41 +00:00
Brian Gaeke
2dd3e1b435 Stub for constant shift expr support.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10168 91177308-0d34-0410-b5e6-96231b3b80d8
2003-11-22 05:18:35 +00:00
Brian Gaeke
0415b6c57a Fix PR149 - support constant shift expressions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10167 91177308-0d34-0410-b5e6-96231b3b80d8
2003-11-22 05:02:56 +00:00
Misha Brukman
a59f41fc6c * Add code to flush the ICache, which any self-respecting SMC must do
* Restore registers *after* everything else to avoid any possible side effects

This fixes McCat-imp.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10147 91177308-0d34-0410-b5e6-96231b3b80d8
2003-11-21 23:48:54 +00:00
Chris Lattner
62b767b8d7 Fix PR123
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10069 91177308-0d34-0410-b5e6-96231b3b80d8
2003-11-18 17:47:05 +00:00
Chris Lattner
c59c118e44 ISO 9899 7.13.2.1 (3) says that "[...] the values of objects of automatic
storage duration that are local to the function containing the invocation of the
[...] setjmp macro that do not have volatile-qualified type and have been
changed between the setjmp invocation and longjmp call are indeterminate."

As such, we have to mark all variables in a function that uses 'invoke' as
volatile.

This fixes PR77


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@10035 91177308-0d34-0410-b5e6-96231b3b80d8
2003-11-16 22:06:14 +00:00
Misha Brukman
c6b1584de6 Include the file before we close the llvm namespace.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@9954 91177308-0d34-0410-b5e6-96231b3b80d8
2003-11-13 00:23:05 +00:00