Age | Commit message (Collapse) | Author |
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19931 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19930 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
the same.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19929 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19928 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19926 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Based on the ilist changes avoid allocating an entire Use object for the
end of the Use chain. This saves 8 bytes of memory for each Value allocated
in the program. For 176.gcc, this reduces us from 69.5M -> 66.0M, a 5.3%
memory savings.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19925 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19924 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19923 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19922 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Add a new method, destroySentinal, that is used to delete it (instead of
requiring use of delete.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19921 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19920 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19919 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
cases it represented them as 'unsigned's, which are not enough for 64-bit
hosts. In other cases, it represented them as uint64_t's, which are
inefficient for 32-bit hosts.
This patch unifies all of the sizes to use size_t instead.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19918 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
This file was schizophrenic when it came to representing sizes. In some
cases it represented them as 'unsigneds', which are not enough for 64-bit
hosts. In other cases, it represented them as uint64_t's, which are
inefficient for 32-bit hosts.
This patch unifies all of the sizes to use size_t instead.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19917 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
really need. This reduces 4M of memory consumption reading 176.gcc.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19916 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19915 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
things that were getting stuck in the nightly tester.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19914 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
used here.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19913 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19912 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
BB as the load are not included in the Cand* sets at all.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19911 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
not invalidated on entry and on exit of the block. This fixes some N^2
behavior in common cases, and speeds up gcc another 5% to 22.35s.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19910 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19909 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19908 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
This speeds up 176.gcc from 25.73s to 23.48s, which is 9.5%
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19907 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
will cause us to miss cases where the input pointer to a load could be value
numbered to another load. Something like this:
%X = load int* %P1
%Y = load int* %P2
Those are obviously the same if P1/P2 are the same. The code this patch
removes attempts to handle that. However, since GCSE iterates, this doesn't
actually buy us anything: GCSE will first replace P1 or P2 with the other
one, then the load can be value numbered as equal.
Removing this code speeds up gcse a lot. On 176.gcc in debug mode, this
speeds up gcse from 29.08s -> 25.73s, a 13% savings.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19906 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
%A = alloca int
%V = load int* %A
value number %V to undef, not 0.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19905 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
memuse at the end, thus it is signed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19904 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19903 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19902 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19901 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19900 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19899 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19898 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19897 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Adjust TmpInstruction to work with the new User model.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19896 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19895 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19894 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19893 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19892 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
Adjust to changes in the User class, operand handling is very different.
PHI node and switch statements must handle explicit resizing of operand
lists.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19891 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
manage their operands.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19890 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19889 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19888 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
class.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19887 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19886 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
directly embed their operands.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19885 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19884 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
and num operands in the User class. this allows us to embed the operands
directly in the subclasses if possible. For example, for binary operators
we store the two operands in the derived class.
The has several effects:
1. it improves locality because the operands and instruction are together
2. it makes accesses to operands faster (one less load) if you access them
through the derived class pointer. For example this:
Value *GetBinaryOperatorOp(BinaryOperator *I, int i) {
return I->getOperand(i);
}
Was compiled to:
_Z19GetBinaryOperatorOpPN4llvm14BinaryOperatorEi:
movl 4(%esp), %edx
movl 8(%esp), %eax
sall $4, %eax
movl 24(%edx), %ecx
addl %ecx, %eax
movl (%eax), %eax
ret
and is now compiled to:
_Z19GetBinaryOperatorOpPN4llvm14BinaryOperatorEi:
movl 8(%esp), %eax
movl 4(%esp), %edx
sall $4, %eax
addl %edx, %eax
movl 44(%eax), %eax
ret
Accesses through "Instruction*" are unmodified.
3. This reduces memory consumption (by about 3%) by eliminating 1 word of
vector overhead and a malloc header on a seperate object.
4. This speeds up gccas about 10% (both debug and release builds) on
large things (such as 176.gcc). For example, it takes a debug build
from 172.9 -> 155.6s and a release gccas from 67.7 -> 61.8s
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19883 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19882 91177308-0d34-0410-b5e6-96231b3b80d8
|
|
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19881 91177308-0d34-0410-b5e6-96231b3b80d8
|