Sooooo many of these had incorrect or strange main module includes.
I have manually inspected all of these, and fixed the main module
include to be the nearest plausible thing I could find. If you own or
care about any of these source files, I encourage you to take some time
and check that these edits were sensible. I can't have broken anything
(I strictly added headers, and reordered them, never removed), but they
may not be the headers you'd really like to identify as containing the
API being implemented.
Many forward declarations and missing includes were added to a header
files to allow them to parse cleanly when included first. The main
module rule does in fact have its merits. =]
llvm-svn: 169131
new hash_value infrastructure, and replace their implementations using
hash_combine. This removes a complete copy of Jenkin's lookup3 hash
function (which is both significantly slower and lower quality than the
one implemented in hash_combine) along with a somewhat scary xor-only
hash function.
Now that APInt and APFloat can be passed directly to hash_combine,
simplify the rest of the LLVMContextImpl hashing to use the new
infrastructure.
llvm-svn: 152004
toString() now takes an optional bool argument that,
depending on the radix, adds the appropriate prefix
to the integer's string representation that makes it into a
meaningful C literal, e.g.:
hexademical: '-f' becomes '-0xf'
octal: '77' becomes '077'
binary: '110' becomes '0b110'
Patch by nobled@dreamwidth.org!
llvm-svn: 133032
of a constant had a minor typo introduced when copying it from the book, which
caused it to favor negative approximations over positive approximations in many
cases. Positive approximations require fewer operations beyond the multiplication.
In the case of division by 3, we still generate code that is a single instruction
larger than GCC's code.
llvm-svn: 126097
zextOrTrunc(), and APSInt methods extend(), extOrTrunc() and new method
trunc(), to be const and to return a new value instead of modifying the
object in place.
llvm-svn: 121120
logic to use the new APInt methods. Among other things this
implements rdar://8501501 - llvm.smul.with.overflow.i32 should constant fold
which comes from "clang -ftrapv", originally brought to my attention from PR8221.
llvm-svn: 116457
payloads. APFloat's internal folding routines always make QNaNs now,
instead of sometimes making QNaNs and sometimes SNaNs depending on the
type.
llvm-svn: 97364
This adds location info for all llvm_unreachable calls (which is a macro now) in
!NDEBUG builds.
In NDEBUG builds location info and the message is off (it only prints
"UREACHABLE executed").
llvm-svn: 75640
Make llvm_unreachable take an optional string, thus moving the cerr<< out of
line.
LLVM_UNREACHABLE is now a simple wrapper that makes the message go away for
NDEBUG builds.
llvm-svn: 75379
large for the testsuite) took over six minutes to compile on my Mac.
The patched LLVM-GCC compiles that testcase in three seconds (GCC
takes less than one second). This hash function is more complex
(about 35 instructions on x86) than what Chris wanted, but I expect it
will be well-behaved with arbitrary inputs.
Thank you to everyone who responded to my previous request for advice.
llvm-svn: 66962
causing assertion failures in getSExtValue().
Fix it by making highWordBits actually contain what its name says,
and add some more unit-tests for APInt.
This fixes PR3419.
llvm-svn: 63107
use raw_ostream instead of std::ostream. Among other goodness,
this speeds up llvm-dis of kc++ with a release build from 0.85s
to 0.49s (88% faster).
Other interesting changes:
1) This makes Value::print be non-virtual.
2) AP[S]Int and ConstantRange can no longer print to ostream directly,
use raw_ostream instead.
3) This fixes a bug in raw_os_ostream where it didn't flush itself
when destroyed.
4) This adds a new SDNode::print method, instead of only allowing "dump".
A lot of APIs have both std::ostream and raw_ostream versions, it would
be useful to go through and systematically anihilate the std::ostream
versions.
This passes dejagnu, but there may be minor fallout, plz let me know if
so and I'll fix it.
llvm-svn: 55263
returning an std::string by value, it fills in a SmallString/SmallVector
passed in. This significantly reduces string thrashing in some cases.
More specifically, this:
- Adds an operator<< and a print method for APInt that allows you to
directly send them to an ostream.
- Reimplements APInt::toString to be much simpler and more efficient
algorithmically in addition to not thrashing strings quite as much.
This speeds up llvm-dis on kc++ by 7%, and may also slightly speed up the
asmprinter. This also fixes a bug I introduced into the asmwriter in a
previous patch w.r.t. alias printing.
llvm-svn: 54873
multiplicative inverse of a given number. Modify udivrem to allow input and
output pairs of arguments to overlap. Patch is based on the work by Chandler
Carruth.
llvm-svn: 52638
over-shift-right should return -1. So here it should be signed-extended,
when bitwidth larger than 64.
test case: llvm/test/ExecutionEngine/2008-06-05-APInt-OverAShr.ll
llvm-svn: 51999
arbitrary range of bits embedded in the middle of another bignum.
This kind of operation is desirable in many cases of software
floating point, e.g. converting bignum integers to floating point
numbers of fixed precision (you want to extract the precision most
significant bits).
Elsewhere, add an assertion, and exit the shift functions early if
the shift count is zero.
llvm-svn: 42745
part widths. Also, return the number of parts actually required to
hold the result's value.
Remove an over-cautious condition from rounding of float->hex conversion.
llvm-svn: 42669
bit width instead of number of words allocated, which
makes it actually work for int->APF conversions.
Adjust callers. Add const to one of the APInt constructors
to prevent surprising match when called with const
argument.
llvm-svn: 42210
double from some of the many places in the optimizers
it appears, and do something reasonable with x86
long double.
Make APInt::dump() public, remove newline, use it to
dump ConstantSDNode's.
Allow APFloats in FoldingSet.
Expand X86 backend handling of long doubles (conversions
to/from int, mostly).
llvm-svn: 41967
two's complement bignum arithmetic. They could be used to
implement much of APInt, but the idea is they are enough to
implement APFloat as well, which the current APInt interface
is not suited for.
Patch by Neil Booth!
llvm-svn: 41124
not being generated correctly because the shl operator does not mutate its
object but returns a new value. Also, make the distinction between radix
16 and the others more clear.
llvm-svn: 37111
types:
1. Functions to compute div/rem at the same time.
2. Further assurance that an APInt with 0 bitwidth cannot be constructed.
3. Left and right rotate operations.
4. An exactLogBase2 function which requires an exact power of two or it
returns -1.
llvm-svn: 37025
this to a NOTE: because pari/gp results start to get rounded incorrectly
after 192 bits of precision. APInt and pari/gp never differ by more than
1, but APInt is more accurate because it does not lose precision after 192
bits as does pari/gp.
llvm-svn: 34834
the bit width of negative numbers by computing the minimum bit width for a
negative value. E.g. 0x1800000000000000 could be just 0x8000000000000000
llvm-svn: 34695
1. Add unsigned and signed versions of methods so a "bool" argument doesn't
need to be passed in.
2. Make the various getMin/getMax functions all be inline since they are
so simple.
3. Simplify sdiv and srem code.
llvm-svn: 34680
2. Rewrite operator=(const APInt& RHS) to allow the RHS to be a different
bit width than the LHS. This makes it possible to use APInt as the key
of a DenseMap, as needed for the IntConstants map in Constants.cpp
3. Fix operator=(uint64_t) to clear unused bits in case the client assigns
a value that has more bits than the APInt allows.
4. Assert that bit widths are equal in operator==
5. Revise getHashValue() to put the bit width in the low order six bits.
This should help to make i1 0, i2 0, ... i64 0 all distinct in the
IntConstants DenseMap.
llvm-svn: 34646
2. Fix countTrailingZeros to use a faster algorithm.
3. Simplify sext() slightly by using isNegative().
4. Implement ashr using word-at-a-time logic instead of bit-at-a-time
5. Rename locals named isNegative so they don't clash with method name.
6. Fix fromString to compute negated value correctly.
llvm-svn: 34629
2. Move comments for methods to .h file, delete them in .cpp file.
3. All places that were doing manual clear of high order bits now call the
clearUnusedBits() method in order to not depend on undefined behavior
of the >> operator when the number of bits shifted equals the word size.
4. Reduced # of loc by using the new result of clearUnusedBits() method.
5. Simplified logic (decreased indentation) in a few places.
6. Added code comments to larger functions that needed them.
7. Added FIXME notes about weak implementations of things (e.g. bit-by-bit
shift right is sub-optimal).
llvm-svn: 34603
2. Clean up comments, style, coding standards, etc.
3. Simplify a constructor.
Extended testing revealed some additional bugs in shifting. I'll fix these
tomorrow.
llvm-svn: 34559
to be safe.
2. Make internal functions that return a carry/borrow return bool instead
of uint64_t because the carry/borrow can only be in range [0,1].
3. Assert that the pointers to KnuthDiv are all different so that the
result and operands can't overlap.
4. Add debug output to KnuthDiv function.
5. Fix a problem with KnuthDiv by separating the b's complement operation
from the subtraction borrow operation. This fixes a wide range of
division problems, but alas, not all of them.
llvm-svn: 34554
memory (as done in fromString).
2. Implement Knuth divide more closely to what is recommended in his book.
3. Fix computation of the remainder for Knuth Divide (bad shifting).
4. Remove some cruft from the file
llvm-svn: 34518
word size. This fixes all reads of uninitialized data (buffer over read)
and makes APInt.cpp memory clean, per valgrind. The only remaining
problem is division in a few cases.
llvm-svn: 34483
2. Change 0 initialization of union to larger component so all is zeroed.
3. Fix the borrow logic in subtraction so it works for > 128 bits.
4. Rewrite fromString to use a simpler but correct algorithm and also to
not set the bit width contrary to the user's request.
5. Optimize toString a bit by making it only do one Knuth divide per
iteration instead of two.
With these changes, all arithmetic passes (verified by pari/GP) up to
1024 bits except for certain division cases.
llvm-svn: 34463
1. Ensure pVal is set to 0 in each constructor.
2. Fix roundToDouble to make correct calculations and not read beyond the
end of allocated memory.
3. Implement Knuth's "classical algorithm" for division from scratch and
eliminate buffer overflows and uninitialized mememory reads. Document
it properly too.
4. Implement a wrapper function for KnuthDiv which handles the 64-bit to
32-bit conversion and back. It also implement short division for the
n == 1 case that Knuth's algorithm can't handle.
5. Simplify the logic of udiv and urem a little, make them exit early, and
have them use the "divide" wrapper function to perform the division
or remainder operation.
6. Move the toString function to the end of the file, closer to where
the division functions are located.
Note: division is still broken for some > 64 bit values, but at least it
doesn't crash any more.
llvm-svn: 34449
2. Implement toString for power-of-2 radix without using divide and always
printing full words. This allows hex/binary to look at the bit
respresentation of the APInt as well as avoid bugs in divide.
llvm-svn: 34396
2. Consolidate memory allocation into just two inline functions.
3. Convert "unsigned" to uint32_t to gaurantee its size.
4. Eliminate magic constants and replace with symbolic equivalent.
5. Improve code documentation slightly.
6. Simplify the logical operator code because bitwidths must be the same.
7. Fix indentation per coding standards.
8. Use exit-early style to reduce indentation in several functions.
llvm-svn: 34389
1. Function style changes.
2. 80-col violations.
3. Better names for things.
4. Arrange constructors so they all take bit width first.
5. Add named signed and unsigned comparison functions and remove the
corresponding operators.
6. Remove operator&& and operator|| but provide a getBoolValue function which
converts to bool as comparison against 0. This allows the normal && and
|| operators to be used as if (X.getBoolValue() && Y.getBoolValue())
Note: this still doesn't function 100% yet. I'm working on the bugs now.
llvm-svn: 34353
1. Make getMinValue() returns the right value.
2. Fix the ByteSwap() crash problem.
3. Make Postfix increment work correctly.
4. Fix some bugs in LogBase2, Hi/LoBits and UDiv.
llvm-svn: 34304
1. Make all the operators use uppercase
2. Rename APIntRoundToDouble method just RoundToDouble, the APInt is
redundant.
3. Turn the class on for compilation.
llvm-svn: 34253
Craig Topper