Insert a header into the new testcase containing a sample RUN line a FIXME and
an XFAIL. Then insert the formated C code and finally the LLVM-IR without
attributes, the module ID or the target triple.
llvm-svn: 211612
We use llvm.codegen intrinsic to generate code for embedded LLVM-IR
strings. The reason we introduce such a intrinsic is that previous
clang/opt tools was NOT linked with various LLVM targets and their
AsmParsers and AsmPrinters. Since clang/opt been linked with all the
needed libraries, we no longer need the llvm.codegen intrinsic.
llvm-svn: 211573
+ Collect reduction dependences
+ Introduced TYPE_RED in Dependences.h which can be used to obtain the
reduction dependences
+ Used TYPE_RED to prevent parallelization while we do not have a privatizing
code generation
+ Relax the dependences for non-parallel code generation
+ Add privatization dependences to ensure correctness
+ 12 Test cases to check for reduction and privatization dependences
llvm-svn: 211369
We do this currently only for test cases where we have integer offsets that
clearly access array dimensions out-of-bound.
-; for (long i = 0; i < n; i++)
-; for (long j = 0; j < m; j++)
-; for (long k = 0; k < o; k++)
+; for (long i = 0; i < n - 3; i++)
+; for (long j = 4; j < m; j++)
+; for (long k = 0; k < o - 7; k++)
; A[i+3][j-4][k+7] = 1.0;
This will be helpful if we later want to simplify the access functions under the
assumption that they do not access memory out of bounds.
llvm-svn: 210179
Without this patch, the testcase would fail on the delinearization of the second
array:
; void foo(long n, long m, long o, double A[n][m][o]) {
; for (long i = 0; i < n; i++)
; for (long j = 0; j < m; j++)
; for (long k = 0; k < o; k++) {
; A[i+3][j-4][k+7] = 1.0;
; A[i][0][k] = 2.0;
; }
; }
; CHECK: [n, m, o] -> { Stmt_for_body6[i0, i1, i2] -> MemRef_A[3 + i0, -4 + i1, 7 + i2] };
; CHECK: [n, m, o] -> { Stmt_for_body6[i0, i1, i2] -> MemRef_A[i0, 0, i2] };
Here is the output of FileCheck on the testcase without this patch:
; CHECK: [n, m, o] -> { Stmt_for_body6[i0, i1, i2] -> MemRef_A[i0, 0, i2] };
^
<stdin>:26:2: note: possible intended match here
[n, m, o] -> { Stmt_for_body6[i0, i1, i2] -> MemRef_A[o0] };
^
It is possible to find a good delinearization for A[i][0][k] only in the context
of the delinearization of both array accesses.
There are two ways to delinearize together all array subscripts touching the
same base address: either duplicate the code from scop detection to first gather
all array references and then run the delinearization; or as implemented in this
patch, use the same delinearization info that we computed during scop detection.
llvm-svn: 210117
+ CL-option --polly-tile-sizes=<int,...,int>
The i'th value is used as a tile size for dimension i, if
there is no i'th value, the value of --polly-default-tile-size is
used
+ CL-option --polly-default-tile-size=int
Used if no tile size is given for a dimension i
+ 3 Simple testcases
llvm-svn: 209753
Tag the GPGPU codegen test cases as unsupported if the nvptx target is not
included in the current llvm build.
Contributed-by: Yabin Hu <yabin.hwu@gmail.com>
llvm-svn: 208779
Commit r206510 falsely advertised to fix the load cases, even though it only
fixed the store case. This commit adds the same fix for the load case including
the missing test coverage.
llvm-svn: 206577
Even tough we may want to generate a vector load, the address from which to load
still is a scalar. Make sure even if previous address computations may have been
vectorized, that the addresses are also available as scalars.
This fixes http://llvm.org/PR19469
Reported-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
llvm-svn: 206510
The following example shows a non-parallel loop
void f(int a[]) {
int i;
for (i = 0; i < 10; ++i)
A[i] = A[i+5];
}
which, in case we import a schedule that limits the iteration domain
to 0 <= i < 5, becomes parallel. Previously we crashed in such cases, now we
just recognize it as parallel.
This fixes http://llvm.org/PR19435
Reported-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
llvm-svn: 206318
During code preperation trivial PHI nodes (mainly introduced by lcssa) are
deleted to decrease the number of introduced allocas (==> dependences). However
simply replacing them by their only incoming value would cause the independent
block pass to introduce new allocas. To prevent this we try to share stack slots
during code preperarion, hence to reuse a already created alloca 'to demote' the
trivial PHI node. This works if we know that the value stored in this alloca
will be the incoming value of the trivial PHI at the end of the predecessor
block of this trivial PHI.
Contributed-by: Johannes Doerfert <doerfert@cs.uni-saarland.de>
llvm-svn: 205320
We explicitly specifying all filenames instead of assuming some naming
convention used by clang and opt.
Contributed-by: Johannes Doerfert <doerfert@cs.uni-saarland.de>
llvm-svn: 204726
For complex examples it may happen that we do not compute dependences. In this
case we do not want to crash, but just not detect parallel loops.
llvm-svn: 204470
This patch enables vectorization of loops containing backward array
traversal (array stride is -1).
Contributed-by: Chris Jenneisch <chrisj@codeaurora.org>
llvm-svn: 204257
In case we are at the innermost band, we try to prepare for vectorization. This
means, we look for the innermost parallel loop and strip mine this loop to the
innermost level using a strip-mine factor corresponding to the number of vector
iterations.
For whatever reason, the code that implemented this feature was broken. We now
added a comment, a test case and obviously also the right code.
llvm-svn: 203544
This is necessary to avoid test failures in the CLooG test suite due to the
recent isl update.
We also need to update two polly test cases which rely on a certain order in the
textual description that isl chooses for its sets and maps. Changes here are not
often, but we should probably switch to a check that verifies such maps are
semantically equivalent instead of represented identically.
llvm-svn: 203476
For now we only mark innermost loops for the loop vectorizer. We could later
also mark not-innermost loops to enable the introduction of openmp parallelism.
llvm-svn: 202854
In 'obsequi' we have a scop in which the current dead code elimination works,
but the generated code is way too complex. To avoid this trouble (and to not
disable the DCE entirely) we add an additional approximative step before
the actual dead code elimination. This should fix one of the two current
nightly-test issues.
Polly could be improved to handle 'obsequi' by teaching it to introduce only a
single parameter for (%1 and zext %1) which halves the number of parameters and
allows polly to derive a simpler representation for the set of live iterations.
However, this needs some time to investigate.
I will commit a test case as soon as we have a reduced one.
llvm-svn: 202010
In case we do not have valid dependences, we do not run dead code elimination or
the schedule optimizer. This fixes an infinite loop in the dead code
elimination (PR12110).
llvm-svn: 201982
Instead of giving a choice between a precise (but possibly very complex)
analysis and an approximative analysis we now use a hybrid approach which uses N
precise steps followed by one approximating step. The precision of the analysis
can be changed by increasing N. With a default of 'N' = 2, we get fully precise
results for our current test cases and should not run into performance problems
for more complex test cases. We can adjust this value when we got more
experience with this dead code elimination.
llvm-svn: 201888
We now skip the debug intrinsics which is a lot better than crashing due to
uncopied metadata references. We should step by step investigate which debug
intrinsics we can copy without trouble.
We still keep the debug location metadata.
llvm-svn: 201860
This pass eliminates loop iterations that compute results that are not used
later on. This can help e.g. in D, where the default zero-initialization is
often unnecessary if right after new values are assigned to an array.
Contributed-by: Peter Conn <conn.peter@gmail.com>
llvm-svn: 201817
We do not have a use for this information at the moment. If we need this at some
point, the "instruction -> access" mapping needs to be enhanced as a single
instruction could then possibly perform multiple accesses.
This patch allows us to build the polyhedral information for scops with scalar
dependences.
llvm-svn: 201815
In rare cases the modification of one scop can effect the validity of other
scops, as code generation of an earlier scop may make the scalar evolution
functions derived for later scops less precise. The example that triggered this
patch was a scop that contained an 'or' expression as follows:
%add13710 = or i32 %j.19, 1
--> {(1 + (4 * %l)),+,2}<nsw><%for.body81>
Scev could only analyze the 'or' as it knew %j.19 is a multiple of 2. This
information was not available after the first scop was code generated (or
independent-blocks was run on it) and SCEV could not derive a precise SCEV
expression any more. This means we could not any more code generate this SCoP.
My current understanding is that there is always the risk that an earlier code
generation change invalidates later scops. As the example we have seen here is
difficult to avoid, we use this occasion to guard us against all such
invalidations.
This patch "solves" this issue by verifying right before we start working on
a detected scop, if this scop is in fact still valid. This adds a certain
overhead. However the verification we run is anyways very fast and secondly
it is only run on detected scops. So the overhead should not be very large. As
a later optimization we could detect scops only on demand, such that we need
to run scop-detections always only a single time.
This should fix the single last failure in the LLVM test-suite for the new
scev-based code generation.
llvm-svn: 201593
There does not seem to be a reason that we can not support PHI nodes outside of
the scop that reference values within the SCoP. Or at least, the attached test
case seems to do the right thing. We remove the assert for now.
llvm-svn: 200427
In rare cases, a region R which is itself not valid has an indirect child region
that is valid. When R becomes part of a valid region by expansion of another
region, then all children of R have to be erased from the set of valid regions.
This patch ensures that indirect children are erased in addition to direct
children.
Contributed-by: Armin Groesslinger <armin.groesslinger@uni-passau.de>
Tobias: I added a reduced test case and adjusted the logic of the patch to
only recurse until the first child is found.
llvm-svn: 200411
Array base addresses need to be invariant in the region considered. The base
address has to be computed outside the region, or, when it is computed inside,
the value must not change with the iterations of the loops. For example, when a
two-dimensional array is represented as a pointer to pointers the base address
A[i] in an access A[i][j] changes with i; therefore, such regions have to be
rejected.
Contributed by: Armin Größlinger <armin.groesslinger@uni-passau.de>
llvm-svn: 200314
This is not only not necessary, but in case -03 changes this can actually
cause arbitrarily failing test cases such as, e.g., a recent change by Chandler
that caused -O3 to unroll the loop body, which made the loop we wanted to
detect disappear and consequently this test case fail.
llvm-svn: 200204
Count the number of computational steps that have been used to solve the
dependence problem and abort in case we reach the "compute-out". This ensures we
do not hang forever in cases the dependence problem is too difficult to solve.
There is just a single case in the LLVM test-suite that runs into the
compute-out. Even in this case, we can probably coalesce some of the parameters
(i32 b, i32 b zext i64, ...) to simplify the problem enough to not hit the
compute out. However, for now we set the compute out in place to address the
general issue. The compute out was choosen such that it stops on a recent laptop
after about 8 seconds.
llvm-svn: 200156
We now report the following:
$ polly-clang -O3 -mllvm -polly -mllvm -polly-report test.c -c \
-gline-tables-only
note: Polly detected an optimizable loop region (scop) in function 'foo'
test.c:2: Start of scop
test.c:3: End of scop
note: Polly detected an optimizable loop region (scop) in function 'bar'
test.c:9: Start of scop
test.c:13: End of scop
llvm-svn: 197558
When constructing a scop sometimes the exact representation of a statement or
condition would be very complex, but there is a common case which is a lot
simpler, but which is only valid under certain assumptions. The assumed context
records the assumptions taken during the construction of this scop and that need
to be code generated as a run-time test.
At the moment, we do not yet model any assumptions, but only added the
AssumedContext as well as the isl-ast generation support. As a next step,
this needs to be hooked up with the isl code generation.
if (1) /* run-time condition */
{ /* optimized code */ }
else
{ /* original code */ }
llvm-svn: 193652
SCoP invariant parameters with the different start value would deter parameter
sharing. For example, when compiling the following C code:
void foo(float *input) {
for (long j = 0; j < 8; j++) {
// SCoP begin
for (long i = 0; i < 8; i++) {
float x = input[j * 64 + i + 1];
input[j * 64 + i] = x * x;
}
}
}
Polly would creat two parameters for these memory accesses:
p_0: {0,+,256}
p_2: {4,+,256}
[j * 64 + i + 1] => MemRef_input[o0] : 4o0 = p_1 + 4i0
[j * 64 + i] => MemRef_input[o0] : 4o0 = p_0 + 4i0
These parameters only differ from start value. To enable parameter sharing,
we split the start value from SCEVAddRecExpr, so they would share a single
parameter that always has zero start value:
p0: {0,+,256}<%for.cond1.preheader>
[j * 64 + i + 1] => MemRef_input[o0] : 4o0 = 4 + p_1 + 4i0
[j * 64 + i] => MemRef_input[o0] : 4o0 = p_0 + 4i0
Such translation can make the polly-dependence much faster.
Contributed-by: Star Tan <tanmx_star@yeah.net>
llvm-svn: 187728
In case we detect that the schedule the user wants to import is invalid we
refuse it _and_ free the isl_maps containing it.
Another bug found thanks to Rafael.
llvm-svn: 187339
We now use __isl_take to annotate the uses of the isl_set where we got the
memory management wrong.
Thanks to Rafael! His pipefail work hardened our test environment and exposed
this bug nicely.
llvm-svn: 187338
Ensure that the scalar write access corresponds to the result of a load
instruction appears after the generic read access corresponds to the load
instruction.
llvm-svn: 186419
Previously this happend to work for integers up to i64, but we got it wrong
for larger numbers. Fix this and add test cases to verify this keeps working.
Reported by: Sven Verdoolaege <skimo at kotnet dot org>
llvm-svn: 183986
When a region header is part of a loop, then all entering edges of this region
should not come from the loop but outside the region. Otherwise, the loop may be
only partially part of the region, which would cause troubles in handling
induction variables.
Currently, we can only model induction variables that are either fully part of
the scop (loop induction variable) or induction variables that are scop-
invariant (parameter). A loop that is only partially part of the
scop causes troubles, as there is no good way to handle the induction
variable in the independent blocks pass.
Contributed-by: Star Tan <tanmx_star@yeah.net>
llvm-svn: 183800
The original test case showed a problem with the independet blocks pass and
we decided to XFAIL it for now. Unfortunately the failure is not detected if
we build without asserts and the verification of the independent block pass
is not run. This change tests now for the actual reason of the failure and
should trigger even in a non asserts build. We did not yet solve the underlying
bug, but this should at least make the test suite behavior consistent.
llvm-svn: 183025
When the Polly code generation was written we did not correctly update the
LoopInfo data, but still claimed that the loop information is correct. This
does not only lead to missed optimizations, but it can also cause
miscompilations in case passes such as LoopSimplify are run after Polly.
Reported-by: Sergei Larin <slarin@codeaurora.org>
llvm-svn: 181987
BeforeBB
|
v
GuardBB
/ \
__ PreHeaderBB \
/ \ / |
latch HeaderBB |
\ / \ /
< \ /
\ /
ExitBB
This does not only remove the need for an explicit loop rotate pass, but it also
gives us the possibility to skip the construction of the guard condition in case
the loop is known to be executed at least once. We do not yet exploit this, but
by implementing this analysis in the isl code generator we should be able to
remove more guards than the generic loop rotate pass can. Another point is that
loop rotation can introduce additional PHI nodes, which may hide that a loop can
be executed in parallel. This change avoids this complication and will make it
easier to move the openmp code generation into a separate pass.
llvm-svn: 181986
Use the new cl::OptionCategory support to move the Polly options into a separate
option category. The aim is to hide most options and show by default only the
options a user needs to influence '-O3 -polly'. The available options probably
need some care, but here is the current status:
Polly Options:
Configure the polly loop optimizer
-enable-polly-openmp - Generate OpenMP parallel code
-polly - Enable the polly optimizer (only at -O3)
-polly-no-tiling - Disable tiling in the scheduler
-polly-only-func=<function-name> - Only run on a single function
-polly-report - Print information about the activities
of Polly
-polly-vectorizer - Select the vectorization strategy
=none - No Vectorization
=polly - Polly internal vectorizer
=unroll-only - Only grouped unroll the vectorize
candidate loops
=bb - The Basic Block vectorizer driven by
Polly
llvm-svn: 181295
In the classical (non -polly-codegen-scev) mode, we assume that we can always
recreate PHI nodes during code generation. This is not true. We can only
reconstruct them from the polyhedral information, in case the entire loop of the
PHI node is part of the SCoP and consequently the PHI node was translated in
the polyhedral description.
llvm-svn: 179674
We now support regions with multiple entries and multiple exits natively.
Regions are not needed to be simplified to single entry and single exit.
We need to XFAIL two test cases as this change increases the scop coverage
and uncoveres two failures in the independent blocks pass. The first failure
will be fixed in a subsequent commit, the second one is in the non-default
-polly-codegen-scev mode and still needs to be fixed.
Contributed-by: Star Tan <tanmx_star@yeah.net>
llvm-svn: 179673
Regions that have multiple entry edges are very common. A simple if condition
yields e.g. such a region:
if
/ \
then else
\ /
for_region
This for_region contains two entry edges 'then' -> 'for_region' and 'else' -> 'for_region'.
Previously we scheduled the RegionSimplify pass to translate such regions into
simple regions. With this patch, we now support them natively when the region is
in -loop-simplify form, which means the entry block should not be a loop header.
Contributed by: Star Tan <tanmx_star@yeah.net>
llvm-svn: 179586
We do not only need to understand that 'k * p' is a parameter expression, but
also need to store this expression in the set of parameters. Before this patch
we wrongly stored the two individual parameters %k and %p.
Reported by: Sebastian Pop <spop@codeaurora.org>
llvm-svn: 179485
Statements with an empty iteration domain may not have a schedule assigned by
the isl schedule optimizer. As Polly expects each statement to have a schedule,
we keep the old schedule for such statements.
This fixes http://llvm.org/PR15645`
Reported-by: Johannes Doerfert <johannesdoerfert@gmx.de>
llvm-svn: 179233
Regions that have multiple exit edges are very common. A simple if condition
yields e.g. such a region:
if
/ \
then else
\ /
after
Region: if -> after
This regions contains the bbs 'if', 'then', 'else', but not 'after'. It has
two exit edges 'then' -> 'after' and 'else' -> 'after'.
Previously we scheduled the RegionSimplify pass to translate such regions into
simple regions. With this patch, we now support them natively.
Contributed-by: Star Tan <tanmx_star@yeah.net>
llvm-svn: 179159
Fix inspired from c2d4a0627e95c34a819b9d4ffb4db62daa78dade.
Given the following code
for (i = 0; i < 10; i++) {
;
}
S: A[i] = 0
When translate the data reference A[i] in statement S using scev, we need to
retrieve the scev of 'i' at the location of 'S'. If we do not do this the
scev that we obtain will be expressed as {0,+,1}_for and will reference loop
iterators that do not surround 'S'. What we really want is the scev to be
instantiated to the value of 'i' after the loop. This value is {10}.
This used to crash in:
int loopDimension = getLoopDepth(Expr->getLoop());
isl_aff *LAff = isl_aff_set_coefficient_si(
isl_aff_zero_on_domain(LocalSpace), isl_dim_in, loopDimension, 1);
(gdb) p Expr->dump()
{8,+,8}<nw><%do.body>
(gdb) p getLoopDepth(Expr->getLoop())
$5 = 0
isl_space *Space = isl_space_set_alloc(Ctx, 0, NbLoopSpaces);
isl_local_space *LocalSpace = isl_local_space_from_space(Space);
As we are trying to create a memory access in a stmt that is outside all loops,
LocalSpace has 0 dimensions:
(gdb) p NbLoopSpaces
$12 = 0
(gdb) p Statement.BB->dump()
if.then: ; preds = %do.end
%0 = load float* %add.ptr, align 4
store float %0, float* %q.1.reg2mem, align 4
br label %if.end.single_exit
and so the scev for %add.ptr should be taken at the place where it is used,
i.e., it should be the value on the last iteration of the do.body loop, and not
"{8,+,8}<nw><%do.body>".
llvm-svn: 179148
After this commit, polly is clang-format clean. This can be tested with
'ninja polly-check-format'. Updates to clang-format may change this, but the
differences will hopefully be both small and general improvements to the
formatting.
We currently have some not very nice formatting for a couple of items, DEBUG()
stmts for example. I believe the benefit of being clang-format clean outweights
the not perfect layout of this code.
llvm-svn: 177796
Given the following code
for (i = 0; i < 10; i++) {
;
}
S: A[i] = 0
When code generating S using scev based code generation, we need to retrieve
the scev of 'i' at the location of 'S'. If we do not do this the scev that
we obtain will be expressed as {0,+,1}_for and will reference loop iterators
that do not surround 'S' and that we consequently do not know how to code
generate. What we really want is the scev to be instantiated to the value of 'i'
after the loop. This value is {10} and it can be code generated without
troubles.
llvm-svn: 177777
We now detect scops without a canonical induction variable and can generate a
polyhedral representation for them. There was no modification necessary to
code generate these scops.
llvm-svn: 177643
When using the scev based code generation, we now do not rely on the presence
of a canonical induction variable any more. This commit prepares the path to
(conditionally) disable the induction variable canonicalization pass.
llvm-svn: 177548
When doing SCEV based code generation, we ignore instructions calculating values
that are fully defined by a SCEV expression. The values that are calculated by
this instructions are recalculated on demand.
This commit improves the check to verify if certain instructions can be ignored
and recalculated on demand.
llvm-svn: 177313
In my previous commits I failed to realise that my new requires lines fully
disabled these tests. We now properly check if we are in an asserts build and
only disable the tests if assertions are not available.
Reported-by: Sean Silva <silvas@purdue.edu>
llvm-svn: 176900
This fixes issues caused by the following commit:
r176733 | jvoung | 2013-03-08 17:56:31 -0500
Disable statistics on Release builds and move tests that depend on -stats.
Reported by: Jack Howarth <howarth@bromo.med.uc.edu>
llvm-svn: 176856
We need to remove one dimension. Any is correct as long as it exists. We have
choosen for whatever reason the dimension #dims - 2. This is incorrect if
there is just one dimension. For CLooG this case did never happen. For isl
however, the case can happen and causes undefined behavior including crashes.
We choose now always the last dimension #dims - 1. We could have choosen
dimension '0' but the last dimension is what we remove conceptionally in the
algorithm, so it seems better to actually program it that way.
While at it remove another piece of undefined behavior.
llvm-svn: 174894
Similar to LLVM we now follow the policy of only having LLVM-IR level tests in
the Polly test suite. Testing for miscompilation of larger programs should be
done with the llvm test suite.
llvm-svn: 167255
Instead of calculating exact value (flow) dependences, it is also possible to
calculate memory based dependences. Sometimes memory based dependences are a lot
easier to calculate. To evaluate the benefits, we add an option to calculate
memory based dependences (use -polly-value-dependences=false).
llvm-svn: 167251
The detection of values that need to be copied in to the generated OpenMP
subfunction also detects the array base addresses needed in the SCoP. Hence, it
is not necessary to unconditionally copy all the base addresses to the generated
function.
Test cases are modified to reflect this change. Arrays which are global
variables do not occur in the struct passed to the subfunction anymore. A test
case for base address copy-in is added in copy_in_array.{c,ll}.
Committed with slight modifications
Contributed by: Armin Groesslinger <armin.groesslinger@uni-passau.de>
llvm-svn: 167215
In addition to the arrays and clast variables a SCoP statement may also refer to
values defined before the SCoP or to function arguments. Detect these values and
add them to the set of values passed to the function generated for OpenMP
parallel execution of a clast.
Committed with additional test cases and some refactoring.
Contributed by: Armin Groesslinger <armin.groesslinger@uni-passau.de>
llvm-svn: 167214
When generating OpenMP or GPGPU code the original ValueMap and ClastVars must be
kept. We already recovered the original ClastVars by reverting the changes, but
we did not keep the content of the ValueMap. This patch keeps now an explicit
copy of both maps and restores them after generating OpenMP or GPGPU code.
This is an adapted version of a patch contributed by:
Armin Groesslinger <armin.groesslinger@uni-passau.de>
llvm-svn: 167213
On Linux there is no difference between shared modules and shared libaries, both
are '.so' files. However, on darwin only shared modules are '.so' files. Shared
libraries have the '.dynlib' suffix.
Fix test cases on darwin by expecting a shared module suffix for Polly instead
of a shared library suffix.
This fixes PR14135
Reported by: Jack Howarth <howarth@bromo.med.uc.edu>
llvm-svn: 166402
The bug was within isl. To fix it, we simply update the isl version that
is used by Polly. We still have some changes within Polly to be able to
write a proper test case.
Reported-by: Sameer Sahasrabuddhe <Sameer.Sahasrabuddhe@amd.com>
llvm-svn: 166021
Previously isl always generated '<=' or '>='. However, in many cases '<' or '>'
leads to simpler code. This commit updates isl and adds the relevant code
generation support to Polly.
llvm-svn: 166020
At the moment we can handle such arrays only by conservatively assuming that
each access to such an array may touch any element in the array. It would be
great if we could improve Polly/LLVM at some point, such that we can
recover the multi-dimensionality of the accesses.
llvm-svn: 163619
This ensures that the isl sets/maps we operate on have the same parameter
dimensions. Operations on objects with different parameter dimensions are not
allow and trigger assertions.
llvm-svn: 163618
Translate the selected parallel loop body into a ptx string and run it with the
cuda driver API. We limit this preliminary implementation to target the
following special test cases:
- Support only 2-dimensional parallel loops with or without only one innermost
non-parallel loop.
- Support write memory access to only one array in a SCoP.
The patch was committed with smaller changes to the build system:
There is now a flag to enable gpu code generation explictly. This was required
as we need the llvm.codegen() patch applied on the llvm sources, to compile this
feature correctly. Also, enabling gpu code generation does not require cuda.
This requirement was removed to allow 'make polly-test' runs, even without an
installed cuda runtime.
Contributed by: Yabin Hu <yabin.hwu@gmail.com>
llvm-svn: 161239
I did not take into account, that this patch fails to compile without the
llvm.codegen patch applied. This breaks buildbots.
I revert this until we found a solution to commit this without buildbots
complaining.
This reverts commit cb43ab80e94434e780a66be3b9a6ad466822fe33.
llvm-svn: 160165
Translate the selected parallel loop body into a ptx string and run it
with cuda driver API. We limit this preliminary implementation to
target the following special test cases:
- Support only 2-dimensional parallel loops with or without only one
innermost non-parallel loop.
- Support write memory access to only one array in a SCoP.
Contributed by: Yabin Hu <yabin.hwu@gmail.com>
llvm-svn: 160164
Derive the maximal and minimal values of a parameter from the type it has. Add
this information to the scop context. This information is needed, to derive
optimal types during code generation.
llvm-svn: 157245
This is an incomplete implementation of the SCEV based code generation.
When finished it will remove the need for -indvars -enable-iv-rewrite.
For the moment it is still disabled. Even though it passes 'make polly-test',
there are still loose ends especially in respect of OpenMP code generation.
llvm-svn: 155717
This fixes two crashes that appeared in case of:
- A load of a non vectorizable type (e.g. float**)
- An instruction that is not vectorizable (e.g. call)
llvm-svn: 154586
Grouped unrolling means that we unroll a loop such that the different instances
of a certain statement are scheduled right after each other, but we do
not generate any vector code. The idea here is that we can schedule the
bb vectorizer right afterwards and use it heuristics to decide when
vectorization should be performed.
llvm-svn: 154251
To avoid overflows we still use a larger type (i64) while calculating the value
of the old ivs. However, we truncate the result to the type of the old iv when
providing it to the new code.
A corresponding test case is added to the polly test suite. Also, a failing test
case is fixed.
This fixes PR12311.
Contributed by: Tsingray Liu <tsingrayliu@gmail.com>
llvm-svn: 153952
When deriving new values for the statements of a SCoP, we assumed that parameter
values are constant within the SCoP and consquently do not need to be rewritten.
For OpenMP code generation this assumption is wrong, as such values are not
available in the OpenMP subfunction and consequently also may need to be
rewritten.
Committed with some changes.
Contributed-By: Johannes Doerfert <s9jodoer@stud.uni-saarland.de>
llvm-svn: 153838
This also adds support for modifiable write accesses (until now only read
accesses where supported). We currently do not derive an exact type for the
expression, but assume that i64 is good enough. This will be improved in future
patches.
Contributed by: Yabin Hu <yabin.hwu@gmail.com>
llvm-svn: 153319
We currently do not support pointer types in affine expressions. Hence, we
disallow in the SCoP detection. Later we may decide to add support for them.
This fixes PR12277
Reported-By: Sebastian Pop <sebpop@gmail.com>
llvm-svn: 152928
This also fixes UMax where we did not correctly keep track of the parameters.
Fixes PR12275.
Reported-By: Sebastian Pop <sebpop@gmail.com>
llvm-svn: 152913
The FinalRead statement represented a virtual read that is executed after the
SCoP. It was used when we verified the correctness of a schedule by checking if
it yields the same FLOW dependences as the original code. This is only works, if
we have a final read that reads all memory at the end of the SCoP.
We now switched to just checking if a schedule does not introduce negative
dependences and also consider WAW WAR dependences. This restricts the schedules
a little bit more, but we do not have any optimizer that would calculate a more
complex schedule. Hence, for now final reads are obsolete.
llvm-svn: 152319
In case we can not analyze an access function, we do not discard the SCoP, but
assume conservatively that all memory accesses that can be derived from our base
pointer may be accessed.
Patch provided by: Marcello Maggioni <hayarms@gmail.com>
llvm-svn: 146972
Parameters can be complex SCEV expressions, but they can also be single scalar
values. If a parameters is such a simple scalar value and the value is named,
use this name to name the isl parameter dimensions.
llvm-svn: 144641
Instead of using TempScop to find parameters, we detect them directly
on the SCEV. This allows us to remove the TempScop parameter detection
in a subsequent commit.
This fixes a bug reported by Marcello Maggioni <hayarms@gmail.com>
llvm-svn: 144087
It may happen that we generate the code of a basic block from the original
scop is code generated several times. The new naming scheme reduces confusing
that earlier appeared as the version numbers of the new basic blocks could
have been interpreted as part of the name of the original basic block.
llvm-svn: 139092
Due to the recent introduction of isl_id, parameters need now always to be
aligned. This was not yet taken care of in the code path of vectorization and
dependence analysis.
llvm-svn: 138555
I am planning to eliminate the TempScopInfo pass. To simplify this I remove
some features that may later be added to the ScopInfo pass.
The interchange pass is currently strongly tested and furthermore ment to be
replaced by the general scheduling optimizer. Reductions itself can later
be added easily.
llvm-svn: 138219
Code is generated for a simple access function change imported
from JSCOP file. An access of A[i] is changed to A[0]. The code
for A[0] is generated directly without refering to isl function calls.
llvm-svn: 136789
Instead of deleting the old code, keep it on the side in an if-branch. It will
either be deleted by the dead code elimination or we can use it as fallback.
llvm-svn: 131352
Johannes Doerfert