OSDN Git Service
Dan Gohman [Thu, 7 Feb 2008 16:28:05 +0000 (16:28 +0000)]
Add SourceValue information for outgoing argument stores on x86.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46854
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Thu, 7 Feb 2008 16:18:00 +0000 (16:18 +0000)]
Don't abort if a MemOperand is missing a SourceValue; just print it
as <unknown>. And make some minor adjustments to the MemOperand
dump format.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46853
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Thu, 7 Feb 2008 08:53:49 +0000 (08:53 +0000)]
Fix a x86-64 codegen deficiency. Allow gv + offset when using rip addressing mode.
Before:
_main:
subq $8, %rsp
leaq _X(%rip), %rax
movsd 8(%rax), %xmm1
movss _X(%rip), %xmm0
call _t
xorl %ecx, %ecx
movl %ecx, %eax
addq $8, %rsp
ret
Now:
_main:
subq $8, %rsp
movsd _X+8(%rip), %xmm1
movss _X(%rip), %xmm0
call _t
xorl %ecx, %ecx
movl %ecx, %eax
addq $8, %rsp
ret
Notice there is another idiotic codegen issue that needs to be fixed asap:
xorl %ecx, %ecx
movl %ecx, %eax
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46850
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Thu, 7 Feb 2008 08:29:53 +0000 (08:29 +0000)]
In some cases, e.g. ADD32ri, no transformation is made. Guide against it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46849
91177308-0d34-0410-b5e6-
96231b3b80d8
Nick Lewycky [Thu, 7 Feb 2008 06:36:26 +0000 (06:36 +0000)]
Don't make up new directives. (".set_foobar")
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46848
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Thu, 7 Feb 2008 02:30:40 +0000 (02:30 +0000)]
Fix a typo in a comment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46836
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Wed, 6 Feb 2008 23:09:15 +0000 (23:09 +0000)]
Add support to FoldingSet for hashing APInt objects.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46833
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Wed, 6 Feb 2008 22:27:42 +0000 (22:27 +0000)]
Re-apply the memory operand changes, with a fix for the static
initializer problem, a minor tweak to the way the
DAGISelEmitter finds load/store nodes, and a renaming of the
new PseudoSourceValue objects.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46827
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Wed, 6 Feb 2008 22:07:17 +0000 (22:07 +0000)]
It's PR1925, not PR1609.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46825
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Wed, 6 Feb 2008 20:03:07 +0000 (20:03 +0000)]
Temporarily reverting:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-
20080128/057882.html
This is causing a miscompilation on PPC G5 and just now seeing it on iMac x86-64.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46822
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Wed, 6 Feb 2008 19:16:53 +0000 (19:16 +0000)]
Fix a number of local register allocator issues: PR1609.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46821
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 6 Feb 2008 18:32:06 +0000 (18:32 +0000)]
some more notes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46818
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 6 Feb 2008 18:00:48 +0000 (18:00 +0000)]
make it clear :)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46817
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 6 Feb 2008 18:00:06 +0000 (18:00 +0000)]
add a bunch of random notes and fix some issues noticed by Gabor.
This is not ready for review or hacking, this is a work in progress :)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46816
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Wed, 6 Feb 2008 08:00:32 +0000 (08:00 +0000)]
RegAllocaLocal still *requires* LiveVariables since it runs PHIElimination, followed by TwoAddress which requires LiveVariables. We cannot run LiveVariables on non-SSA code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46813
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 6 Feb 2008 06:30:34 +0000 (06:30 +0000)]
a starter shell for 2.2 release notes
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46810
91177308-0d34-0410-b5e6-
96231b3b80d8
Tanya Lattner [Wed, 6 Feb 2008 00:54:55 +0000 (00:54 +0000)]
Throttle the non-local dependence analysis for basic blocks with more than 50 predecessors. Added command line option to play with this threshold.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46790
91177308-0d34-0410-b5e6-
96231b3b80d8
Anton Korobeynikov [Tue, 5 Feb 2008 23:34:40 +0000 (23:34 +0000)]
Don't dereference an invalid pointer if string is empty.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46781
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Tue, 5 Feb 2008 22:50:29 +0000 (22:50 +0000)]
Fix PR1975: dag isel emitter produces patterns that isel wrong flag result.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46776
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Tue, 5 Feb 2008 22:44:06 +0000 (22:44 +0000)]
Move to getCALLSEQ_END to ensure CALLSEQ_END node produces a flag. This is consistent with the definition in td file.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46775
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 5 Feb 2008 21:26:23 +0000 (21:26 +0000)]
Make RenamePass faster by making the 'is this a new phi node'
check more intelligent. This speeds up mem2reg from 5.29s to
0.79s on a synthetic testcase with tons of predecessors and
phi nodes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46767
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Tue, 5 Feb 2008 20:46:33 +0000 (20:46 +0000)]
Implement sseregparm.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46764
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Tue, 5 Feb 2008 20:04:18 +0000 (20:04 +0000)]
If a vr is already marked alive in a bb, then it has PHI uses that are visited earlier, then it is not killed in the def block (i.e. not dead).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46763
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 5 Feb 2008 19:43:53 +0000 (19:43 +0000)]
regenerate
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46761
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 5 Feb 2008 19:43:40 +0000 (19:43 +0000)]
Make the check for GCC version more robust, fix shared library
dependencies in makefile, and fix llvm_cv_no_link_all_option
on darwin.
Patch by Shantonu Sen, more info here:
http://lists.cs.uiuc.edu/pipermail/llvmdev/2008-February/012410.html
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46760
91177308-0d34-0410-b5e6-
96231b3b80d8
Ted Kremenek [Tue, 5 Feb 2008 18:50:25 +0000 (18:50 +0000)]
Added FoldingSet profiling support to ImmutableSet.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46757
91177308-0d34-0410-b5e6-
96231b3b80d8
Ted Kremenek [Tue, 5 Feb 2008 17:30:43 +0000 (17:30 +0000)]
Changed profiling method for ImmutableMap to once again just use its
unique ImutAVLTree* for profiling.
Modified ImutAVLTree:
(1) changed ComputeHash() to ComputeDigest() and
(2) changed Profile() to use the computed digest and
(3) modified insertion of IMutAVLTree into the FoldingSet owned by
the ImutAVLTreeFactory object to use profiling instead of computing
a direct hash. This fixes a bug where our abuse of the FoldingSet would
not work when the FoldingSet was resized.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46753
91177308-0d34-0410-b5e6-
96231b3b80d8
Nate Begeman [Tue, 5 Feb 2008 08:49:09 +0000 (08:49 +0000)]
Ident mnemonics appropriately
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46746
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 5 Feb 2008 06:18:42 +0000 (06:18 +0000)]
dump the module *before* we delete it, not after.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46741
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 5 Feb 2008 04:45:32 +0000 (04:45 +0000)]
Fix a bug compiling PR1978 (perhaps not the only one though) which
was incorrectly simplifying "x == (gep x, 1, i)" into false, even
though i could be negative. As it turns out, all the code to
handle this already existed, we just need to disable the incorrect
optimization case and let the general case handle it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46739
91177308-0d34-0410-b5e6-
96231b3b80d8
Owen Anderson [Tue, 5 Feb 2008 04:34:03 +0000 (04:34 +0000)]
Fix an obscure read-after-free bug that Duncan found.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46738
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Tue, 5 Feb 2008 01:12:10 +0000 (01:12 +0000)]
Missed one.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46733
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Tue, 5 Feb 2008 00:25:31 +0000 (00:25 +0000)]
This should also work on x86 now.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46730
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Tue, 5 Feb 2008 00:25:13 +0000 (00:25 +0000)]
Unbreak teh build.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46729
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Mon, 4 Feb 2008 23:27:29 +0000 (23:27 +0000)]
Do not unconditionally redefine vec_ext_v16qi and
vec_ext_v4si builtins. This is a hack; they should
be defined here, then resolved in the X86 BE.
However there is enough other stuff missing in the
X86 BE for SSE41 that this will do for now.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46727
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Mon, 4 Feb 2008 23:10:38 +0000 (23:10 +0000)]
Typo.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46725
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Mon, 4 Feb 2008 23:06:48 +0000 (23:06 +0000)]
Dwarf requires variable entries to be in the source order. Right now, since we are recording variable information at isel time this means parameters would appear in the reverse order. The short term fix is to issue recordVariable() at asm printing time instead.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46724
91177308-0d34-0410-b5e6-
96231b3b80d8
Nate Begeman [Mon, 4 Feb 2008 23:04:24 +0000 (23:04 +0000)]
This method should be virtual
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46723
91177308-0d34-0410-b5e6-
96231b3b80d8
Nate Begeman [Mon, 4 Feb 2008 21:44:06 +0000 (21:44 +0000)]
Eliminate some redundant code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46720
91177308-0d34-0410-b5e6-
96231b3b80d8
Ted Kremenek [Mon, 4 Feb 2008 21:17:02 +0000 (21:17 +0000)]
Modified 'Profile' method of ImmutableMap to use the 'Profile' method
of the contained ImutAVLTree root.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46719
91177308-0d34-0410-b5e6-
96231b3b80d8
Devang Patel [Mon, 4 Feb 2008 21:16:10 +0000 (21:16 +0000)]
Do not set time out, lets assume assembler will do its job in reasonable amount of time.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46718
91177308-0d34-0410-b5e6-
96231b3b80d8
Ted Kremenek [Mon, 4 Feb 2008 21:15:24 +0000 (21:15 +0000)]
Modified node creation of ImutAVLTree to do a hash lookup for an existing
node in the FoldingSet of nodes held by the Factory object. If we
we find a node with a matching hash, we do a full structural comparison.
Nodes are also now inserted into the FoldingSet only when we mark them
Immutable, as their children can change during intermediate-rebalancing.
The 'Profile' method for ImutAVLTree is no longer used when looking up
existing ImutAVLTrees with a given set of contents; instead the Profile method
is used by other clients that wish to insert such a tree into a folding set.
This means that we are not using FoldingSet in ImutAVLTreeFactory in the way
it was intended, but instead are using it as an opaque hashtable.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46717
91177308-0d34-0410-b5e6-
96231b3b80d8
Ted Kremenek [Mon, 4 Feb 2008 21:11:17 +0000 (21:11 +0000)]
Added "bucket_iterators" to FoldingSet. Bucket iterators allow iteration
over all the nodes in a particular bucket.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46716
91177308-0d34-0410-b5e6-
96231b3b80d8
Devang Patel [Mon, 4 Feb 2008 20:57:54 +0000 (20:57 +0000)]
Set error message.
Patch by Shantonu Sen.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46715
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 4 Feb 2008 18:07:02 +0000 (18:07 +0000)]
Crashes LegalizeTypes with "Do not know how to
expand the result of this operator!" (node: ctlz).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46713
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 4 Feb 2008 18:05:42 +0000 (18:05 +0000)]
Crashes LegalizeTypes with "Do not know how to split
this operator's operand" (node: extract_subvector).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46712
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 4 Feb 2008 18:02:37 +0000 (18:02 +0000)]
remove target triple to make this test more "generic"
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46711
91177308-0d34-0410-b5e6-
96231b3b80d8
Ted Kremenek [Mon, 4 Feb 2008 17:14:20 +0000 (17:14 +0000)]
Fixed 80 col. violation.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46709
91177308-0d34-0410-b5e6-
96231b3b80d8
Ted Kremenek [Mon, 4 Feb 2008 16:57:26 +0000 (16:57 +0000)]
constified operator~().
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46707
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 4 Feb 2008 09:40:27 +0000 (09:40 +0000)]
Crashed the new type legalizer. Not likely to catch
any bugs in the future since to get the crash you also
need hacked in fake libcall support (which creates odd
but legal trees), but since adding it doesn't hurt...
Thanks to Chris for this ultimately reduced version.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46706
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 4 Feb 2008 09:29:17 +0000 (09:29 +0000)]
I don't see how NodeUpdated can be called with a
ReadyToProcess node - add an assertion to check
this. Add an assertion to NodeDeleted that checks
that processed/ready nodes are indeed not deleted.
It is because they are never deleted that none of
the maps can have a deleted node as the source of
a mapping. It does however seem to be possible in
theory to have a deleted value as the target of a
mapping, however this has not yet been spotted in
the wild. Still mulling on what to do about this.
[The theoretical situation is this: a node A is
expanded/promoted/whatever to a newly created node
B. Thus A->B is added to a map. When the subtree
rooted at B is legalized it is conceivable that B
is deleted due to RAUW on a node somewhere above
it].
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46705
91177308-0d34-0410-b5e6-
96231b3b80d8
Nate Begeman [Mon, 4 Feb 2008 06:00:24 +0000 (06:00 +0000)]
The rest of the SSE4.1 intrinsic patterns that are obvious to me. Getting
Evan's help with the rest.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46697
91177308-0d34-0410-b5e6-
96231b3b80d8
Nate Begeman [Mon, 4 Feb 2008 05:34:34 +0000 (05:34 +0000)]
Some more SSE 4.1 intrinsic patterns.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46696
91177308-0d34-0410-b5e6-
96231b3b80d8
Owen Anderson [Mon, 4 Feb 2008 04:55:24 +0000 (04:55 +0000)]
Make this test more aggressive, to cover recent improvements.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46695
91177308-0d34-0410-b5e6-
96231b3b80d8
Owen Anderson [Mon, 4 Feb 2008 04:53:00 +0000 (04:53 +0000)]
Be more precise when eliminating pointers bue to memcpy's. This allows more
stores to be deleted in some cases.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46694
91177308-0d34-0410-b5e6-
96231b3b80d8
Owen Anderson [Mon, 4 Feb 2008 02:59:58 +0000 (02:59 +0000)]
Allow GVN to hack on memcpy's, making them open to further optimization.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46693
91177308-0d34-0410-b5e6-
96231b3b80d8
Nick Lewycky [Sun, 3 Feb 2008 16:35:19 +0000 (16:35 +0000)]
Tag this test with the PR reference.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46688
91177308-0d34-0410-b5e6-
96231b3b80d8
Nick Lewycky [Sun, 3 Feb 2008 16:33:09 +0000 (16:33 +0000)]
There are some cases where icmp(add) can be folded into a new icmp. Handle them.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46687
91177308-0d34-0410-b5e6-
96231b3b80d8
Nick Lewycky [Sun, 3 Feb 2008 08:19:11 +0000 (08:19 +0000)]
Hack on vectors too.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46684
91177308-0d34-0410-b5e6-
96231b3b80d8
Nick Lewycky [Sun, 3 Feb 2008 07:42:09 +0000 (07:42 +0000)]
Fold away one multiply in instcombine. This would normally be caught in
reassociate anyways, but they could be generated during instcombine's run.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46683
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 3 Feb 2008 07:30:27 +0000 (07:30 +0000)]
Fix typo
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46682
91177308-0d34-0410-b5e6-
96231b3b80d8
Nate Begeman [Sun, 3 Feb 2008 07:18:54 +0000 (07:18 +0000)]
SSE 4.1 Intrinsics and detection
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46681
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 3 Feb 2008 07:13:32 +0000 (07:13 +0000)]
handle the case where a node can become ready to process
multiple times due to a RAUW.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46680
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 3 Feb 2008 07:08:51 +0000 (07:08 +0000)]
Use the new infrastructure for listening to node updates to
keep the LegalizeTypes node flags up to date when doing a RAUW.
This fixes a nasty bug that Duncan ran into and makes the
previous (nonbuggy case) more efficent.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46679
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 3 Feb 2008 07:01:05 +0000 (07:01 +0000)]
the world doesn't need my debugging code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46678
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 3 Feb 2008 06:49:24 +0000 (06:49 +0000)]
Change the 'global modification' APIs in SelectionDAG to take a new
DAGUpdateListener object pointer instead of just returning a vector
of deleted nodes. This makes the interfaces more efficient (no more
allocating a vector [at least a malloc], filling it in, then walking
it) and more clean. This also allows the client to be notified of
nodes that are *changed* but not deleted.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46677
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 3 Feb 2008 05:43:57 +0000 (05:43 +0000)]
explicitly include Compiler.h instead of getting it from tblgen in the middle of a class.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46676
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 3 Feb 2008 03:35:22 +0000 (03:35 +0000)]
Generalize the SDOperand->SDOperand form of
SelectionDAG::ReplaceAllUsesWith to handle replacement of
an SDOperand with *any* sdoperand, not just one for a node with
a single result. Note that this has a horrible FIXME'd hack in it
to work around PR1975. This should be removed when PR1975 is fixed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46674
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 3 Feb 2008 03:20:59 +0000 (03:20 +0000)]
don't do ReplaceUses on a result that doesn't exist.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46673
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 3 Feb 2008 02:05:04 +0000 (02:05 +0000)]
add a -view-legalize-types-dags option, for viewing the dags going into legalize types.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46672
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 2 Feb 2008 08:39:46 +0000 (08:39 +0000)]
Get rid of the annoying blank lines before labels.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46667
91177308-0d34-0410-b5e6-
96231b3b80d8
Nick Lewycky [Sat, 2 Feb 2008 08:29:58 +0000 (08:29 +0000)]
Don't use uninitialized values. Fixes vec_align.ll on X86 Linux.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46666
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 2 Feb 2008 05:06:29 +0000 (05:06 +0000)]
Unbreak ppc debug support.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46665
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 2 Feb 2008 04:07:54 +0000 (04:07 +0000)]
SDIsel processes llvm.dbg.declare by recording the variable debug information descriptor and its corresponding stack frame index in MachineModuleInfo. This only works if the local variable is "homed" in the stack frame. It does not work for byval parameter, etc.
Added ISD::DECLARE node type to represent llvm.dbg.declare intrinsic. Now the intrinsic calls are lowered into a SDNode and lives on through out the codegen passes.
For now, since all the debugging information recording is done at isel time, when a ISD::DECLARE node is selected, it has the side effect of also recording the variable. This is a short term solution that should be fixed in time.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46659
91177308-0d34-0410-b5e6-
96231b3b80d8
Devang Patel [Sat, 2 Feb 2008 01:43:30 +0000 (01:43 +0000)]
Add comment explaining what is lower level analysis pass.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46658
91177308-0d34-0410-b5e6-
96231b3b80d8
Devang Patel [Sat, 2 Feb 2008 01:32:38 +0000 (01:32 +0000)]
Do not strip symbols when EH is ON.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46657
91177308-0d34-0410-b5e6-
96231b3b80d8
Gordon Henriksen [Sat, 2 Feb 2008 01:07:50 +0000 (01:07 +0000)]
Fixing a bug creating floating point constants of type other
than double through the C bindings. Thanks to Tomas Lindquist
Olsen for reporting it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46656
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 2 Feb 2008 00:17:00 +0000 (00:17 +0000)]
Frame index can be negative.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46655
91177308-0d34-0410-b5e6-
96231b3b80d8
Lauro Ramos Venancio [Fri, 1 Feb 2008 21:25:59 +0000 (21:25 +0000)]
CBackend: Implement unaligned load/store.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46646
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Fri, 1 Feb 2008 20:37:16 +0000 (20:37 +0000)]
Don't drop function/call return attributes like 'nounwind'.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46645
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 1 Feb 2008 09:12:11 +0000 (09:12 +0000)]
Update llvm.dbg.func.start.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46636
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 1 Feb 2008 09:10:45 +0000 (09:10 +0000)]
Remove the nasty LABEL hack with a much less evil one. Now llvm.dbg.func.start implies a stoppoint is set. SelectionDAGISel records a new source line but does not create a ISD::LABEL node for this special stoppoint. Asm printer will magically print this label. This ensures nothing is emitted before.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46635
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 1 Feb 2008 02:05:57 +0000 (02:05 +0000)]
Rename RecordLabel to RecordSourceLine because that's what it is doing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46628
91177308-0d34-0410-b5e6-
96231b3b80d8
Anton Korobeynikov [Thu, 31 Jan 2008 23:02:33 +0000 (23:02 +0000)]
Add convenient std::string helpers to StringMap. Patch by Mikhail Glushenkov!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46625
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Thu, 31 Jan 2008 21:00:00 +0000 (21:00 +0000)]
Revert 46556 and 46585. Dan please fix the PseudoSourceValue problem and re-commit.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46623
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Thu, 31 Jan 2008 18:34:01 +0000 (18:34 +0000)]
Chris' change to print an approximation to long doubles
exposed a bug in APFloat's long double->double conversion of
NaNs. Broke several things in the ieee part of gcc testsuite.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46617
91177308-0d34-0410-b5e6-
96231b3b80d8
Anton Korobeynikov [Thu, 31 Jan 2008 17:43:25 +0000 (17:43 +0000)]
Reapply the patch with fix.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46615
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 31 Jan 2008 17:24:51 +0000 (17:24 +0000)]
revert anton's recent stringmap patch, which breaks clang.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46614
91177308-0d34-0410-b5e6-
96231b3b80d8
Anton Korobeynikov [Thu, 31 Jan 2008 12:30:40 +0000 (12:30 +0000)]
Make StringMap to be more STL'ish. Patch by Mikhail Glushenkov!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46612
91177308-0d34-0410-b5e6-
96231b3b80d8
Anton Korobeynikov [Thu, 31 Jan 2008 12:10:41 +0000 (12:10 +0000)]
Whitespace cleanup
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46611
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Thu, 31 Jan 2008 10:05:13 +0000 (10:05 +0000)]
Add a comment for a nasty short term hack.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46610
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Thu, 31 Jan 2008 09:59:15 +0000 (09:59 +0000)]
Add an extra operand to LABEL nodes which distinguishes between debug, EH, or misc labels. This fixes the EH breakage. However I am not convinced this is *the* solution.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46609
91177308-0d34-0410-b5e6-
96231b3b80d8
Christopher Lamb [Thu, 31 Jan 2008 07:27:46 +0000 (07:27 +0000)]
Allow ComplexExpressions in InstrInfo.td files to be slightly more... complex! ComplexExpressions can now have attributes which affect how TableGen interprets
the pattern when generating matchin code.
The first (and currently, only) attribute causes the immediate parent node of the ComplexPattern operand to be passed into the matching code rather than the node at the root of the entire DAG containing the pattern.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46606
91177308-0d34-0410-b5e6-
96231b3b80d8
Christopher Lamb [Thu, 31 Jan 2008 07:09:08 +0000 (07:09 +0000)]
Add more thorough error checking for NULL register classes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46605
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Thu, 31 Jan 2008 04:06:00 +0000 (04:06 +0000)]
Add x86 specific getFrameIndexOffset(). This fixes local variable debugging info.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46598
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Thu, 31 Jan 2008 03:37:28 +0000 (03:37 +0000)]
MRegisterInfo::getLocation() is a really bad idea. Its function is to calculate the offset from frame pointer to a stack slot and then storing the delta in a MachineLocation object. The name is bad (it implies a getter), and MRegisterInfo doesn't need to know about MachineLocation.
Replace getLocation() with getFrameIndexOffset() which returns the delta from frame pointer to stack slot. Dwarf writer can then use the information for whatever it wants.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46597
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Thu, 31 Jan 2008 03:33:38 +0000 (03:33 +0000)]
Makes the same change in ppc backend: avoid inserting prologue before debug labels.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46596
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Thu, 31 Jan 2008 01:05:10 +0000 (01:05 +0000)]
Avoid unnecessarily casting away const, fixing a FIXME.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46591
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Thu, 31 Jan 2008 01:01:48 +0000 (01:01 +0000)]
Avoid unnecessarily casting away const.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46590
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Thu, 31 Jan 2008 00:44:33 +0000 (00:44 +0000)]
Fix a typo in a comment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46588
91177308-0d34-0410-b5e6-
96231b3b80d8