This is the original patch in my GNUInstallDirs series, now last to merge as the final piece!
It arose as a new draft of D28234. I initially did the unorthodox thing of pushing to that when I wasn't the original author, but since I ended up
- Using `GNUInstallDirs`, rather than mimicking it, as the original author was hesitant to do but others requested.
- Converting all the packages, not just LLVM, effecting many more projects than LLVM itself.
I figured it was time to make a new revision.
I have used this patch series (and many back-ports) as the basis of https://github.com/NixOS/nixpkgs/pull/111487 for my distro (NixOS), which was merged last spring (2021). It looked like people were generally on board in D28234, but I make note of this here in case extra motivation is useful.
---
As pointed out in the original issue, a central tension is that LLVM already has some partial support for these sorts of things. Variables like `COMPILER_RT_INSTALL_PATH` have already been dealt with. Variables like `LLVM_LIBDIR_SUFFIX` however, will require further work, so that we may use `CMAKE_INSTALL_LIBDIR`.
These remaining items will be addressed in further patches. What is here is now rote and so we should get it out of the way before dealing more intricately with the remainder.
Reviewed By: #libunwind, #libc, #libc_abi, compnerd
Differential Revision: https://reviews.llvm.org/D99484
It's NFC because shadow of pointer is clean so origins will not be
propagated anyway.
Depends on D117275
Reviewed By: kda, eugenis
Differential Revision: https://reviews.llvm.org/D117276
Similar to the ld64 command-line options. These use the same underlying
mechanisms as -l and -hidden-l, but allow specifying an absolute path to
the archive. This is often more convenient for a one-off, or when adding
a new search path could change how existing -l options are resolved.
Differential Revision: https://reviews.llvm.org/D117360
Rather than hardcoding all constants, we now use the input tensor to drive the
code setup. Of course, we still need to hardcode dim-2 of A and the final
verification in CHECK is input dependent, but overall this sets a slightly
better example of tensor setup in general.
Reviewed By: bixia
Differential Revision: https://reviews.llvm.org/D117349
Character substrings weren't being folded correctly;
add tests and rework the implementation so that substrings
of literals and named constant character scalars & arrays
are properly folded for use in constant expressions.
Differential Revision: https://reviews.llvm.org/D117343
Currently, when connecting to a remote iOS device from the command line
on Apple Silicon, we end up using the host platform (PlatfromMacOSX)
instead of remote-ios (PlatformRemoteiOS). This happens because
PlatfromMacOSX includes arm64-apple-ios and arm64e-apple-ios as
compatible architectures, presumably to support debugging iOS Apps on
Apple Silicon [1].
This is a problem for debugging remote ios devices, because the host
platform doesn't look for an expanded shared cache on disk and as a
result we end up reading everything from memory, incurring a significant
performance hit.
The crux of this patch is to make PlatfromMacOSX *not* compatible with
arm64(e)-apple-ios. This also means that we now use remote-ios
(PlatformRemoteiOS) as the platform for debugging iOS apps on Apple
Silicon. This has the (unintended) side effect that unlike we do for the
host platform, we no longer check our local shared cache, and incur a
performance hit on debugging these apps.
To avoid that, PlatformRemoteiOS now also check the local cache to
support this use case, which is cheap enough to do unconditionally for
PlatformRemoteiOS.
[1] https://support.apple.com/guide/app-store/iphone-ipad-apps-mac-apple-silicon-fird2c7092da/mac
Differential revision: https://reviews.llvm.org/D117340
Summary:
Address @smeenai feedback https://reviews.llvm.org/D117061#inline-1122106:
>CMake has if(IN_LIST) now, which you can use instead of the string(FIND)
IN_LIST is available since CMake 3.3 released in 2015.
Reviewed By: smeenai
FBD33590959
This is a fix for a crash in the HexagonOptAddrMode pass that was looking
for the third operand (offset) in the following instruction that does not,
in fact, have a third operand:
$r1 = L2_loadw_locked $r1
Additionally, this patch also adds an addrMode value to vgather pseudos
in the Hexagon backend.
Differential Revision: https://reviews.llvm.org/D117133
Since 2959e082e1, we conservatively
assume all inputs are enabled by default. This isn't the best
interface for controlling these anyway, since it's not granular and
only allows trimming the last fields.
ENTRY statement names in module subprograms were not acceptable for
use as a "module procedure" in a generic interface, but should be.
ENTRY statements need to have symbols with place-holding
SubprogramNameDetails created for them in order to be visible in
generic interfaces. Those symbols are created from the "program
tree" data structure. This patch adds ENTRY statement names to the
program tree data structure and uses them to generate SubprogramNameDetails
symbols.
Differential Revision: https://reviews.llvm.org/D117345
This commit adds small tests for the combination of:
{exact, no_exact} x { EQ, NE, UGT, UGE, ULT, ULE, SGT, SGE, SLT, SLE}
This is related to the changes in D117338.
Internal writes to character arrays should not blank-fill
records (elements) past the last one that was written to.
Differential Revision: https://reviews.llvm.org/D117342
EHTerminateScope is used to implement C++ noexcept semantics. Per C++
[except.terminate], it is implemented-defined whether no, some, or all
cleanups are run prior to terminatation.
Therefore, the code to run cleanups on the way towards termination is
unnecessary, and may be omitted.
After this change, we will still run some cleanups: any cleanups in a
function called from the noexcept function will continue to run, while
those in the noexcept function itself will not.
Differential Revision: https://reviews.llvm.org/D113620
In the process of rewriting `alloca`s and `phi`s that use them, the SROA
pass can try to insert a non-PHI instruction by calling
`getFirstInsertionPt()`, which is not possible in a catchswitch BB. This
CL makes we bail out on these cases.
Reviewed By: dschuff
Differential Revision: https://reviews.llvm.org/D117168
In C++20 compound assignment to volatile (here `LocalData[I]++`) is
deprecated, so `mutex_test.cpp` fails to compile.
Simply changing it to `LocalData[I] = LocalData[I] + 1` fixes it.
Differential Revision: https://reviews.llvm.org/D117359
After an ENDFILE statement, a WRITE is an error without
a prior BACKSPACE. Also fix the return value for the case
of formatted integer input with no input digits to be false
(exposed by new test).
Differential Revision: https://reviews.llvm.org/D117346
Very old (pre-'77 standard) codes would use arrays initialized
with Hollerith literals, typically in DATA, as modifiable
formats.
Differential Revision: https://reviews.llvm.org/D117344
The DW_FORM_addr form of highPC address is written in absolute addres,
the data form is written in offset-from-low pc format.
Due to the large test binary the test is prepared separately in
https://github.com/rafaelauler/bolt-tests/pull/8
Vladislav Khmelevsky,
Advanced Software Technology Lab, Huawei
Reviewed By: ayermolo
Differential Revision: https://reviews.llvm.org/D117217
This is consistent with the behavior of Doxygen, and allows users to
write strings with C escapes or document input/output formats containing
special characters (@ or \) without escaping them, which might be
confusing. For example, if a function wants to document its expected
input format as "user@host" it doesn't have to write user\@host instead,
which would look right in the documentation but confusing in the code.
Now users can just use double quotes (which they might do anyway).
This fixes a lot of false positives of -Wdocumentation-unknown-command,
but it could also fix issues with -Wdocumentation if the text triggers
an actual command.
Reviewed By: gribozavr2
Differential Revision: https://reviews.llvm.org/D116190
Inspection of the first character can just be handled by the loop as
well, it does exactly the same thing. Dereferencing the pointer a second
time shouldn't be an issue: the middle end can eliminate that second
read as it's separated from the first only by a pure function call.
Reviewed By: gribozavr2
Differential Revision: https://reviews.llvm.org/D116186
strdup needs either scudo enabled or fullbuild disabled, this properly
adds the second condition
Reviewed By: sivachandra, lntue
Differential Revision: https://reviews.llvm.org/D117138
Currently loop interchange only supports loops with one inner loop
induction variable. This patch adds support for transformation with
more than one inner loop induction variables. The induction PHIs and
induction increment instructions are moved/duplicated properly to the
new outer header and the new outer latch, respectively.
Reviewed By: bmahjour
Differential Revision: https://reviews.llvm.org/D114917
`getNumRegionInvocations` was originally added for the async reference counting, but turned out to be not useful, and currently is not used anywhere (couldn't find any uses in public github repos). Removing dead code.
Reviewed By: Mogball, mehdi_amini
Differential Revision: https://reviews.llvm.org/D117347
Tail merge is slow and of low value. With regular string deduplication, we can
just use the return value of StringTableBuilder::add.
There is no noticeable performance increase because without deduplication
`__cstring` is quite small (7.6MiB for chromium_framework).
Reviewed By: #lld-macho, Jez Ng
Differential Revision: https://reviews.llvm.org/D117273
This commit optimizes the code sequence:
icmp-XXX (ashr-exact (X, C_1), C_2).
Instcombine already implements this optimization for sgt, and this
patch adds support to additional predicates. The transformation is legal
for all predicates if the 'exact' flag is set, and to SGE, UGE, SLT, ULT
when the exact flag is not present.
This pattern is found in the std::vector bounds checks code of the at()
method.
Alive2 proof:
https://alive2.llvm.org/ce/z/JT_WL8
Differential Revision: https://reviews.llvm.org/D117252
Fixes https://github.com/llvm/llvm-project/issues/24784.
With config:
```
AllowShortFunctionsOnASingleLine: Inline
NamespaceIndentation: All
```
The code:
```
namespace Test
{
void f()
{
return;
}
}
```
was incorrectly formatted to:
```
namespace Test
{
void f() { return; }
}
```
since the function `f` was considered being inside a class/struct/record.
That's because the check was simplistic and only checked for a non-zero indentation level of the line starting `f`.
Reviewed By: MyDeveloperDay, HazardyKnusperkeks
Differential Revision: https://reviews.llvm.org/D117142
This seems to be a leftover from a long time ago when there was
an ISD::VBIT_CONVERT and a MVT::Vector. It looks like in those days
the vector type was carried in a VTSDNode.
As far as I know, these days ComputeValueTypes would have already
assigned "Result" the same type we're getting from TLI.getValueType
here. Thus the BITCAST is always a NOP. Verified by adding an assert
and running check-llvm.
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D117335
In clang-format 12, `CompactNamespaces` misformatted the code when `AllowShortLambdasOnASingleLine` is set to false and `BraceWrapping.BeforeLambdaBody` is true.
Input:
```
namespace out {
namespace in {
}
} // namespace out::in
```
Expected output:
```
namespace out { namespace in {
}} // namespace out::in
```
Output from v12:
```
namespace out {
namespace in {
}
} // namespace out::in
```
Config triggering the issue:
```
---
AllowShortLambdasOnASingleLine: None
BraceWrapping:
BeforeLambdaBody : true
BreakBeforeBraces: Custom
CompactNamespaces: true
...
```
Seems there's a corner case when `AllowShortLambdasOnASingleLine` is false, and `BraceWrapping.BeforeLambdaBody` is true, that causes CompactNamespaces to stop working.
The cause was a misannotation of `{` opening brace after `namespace` as a lambda opening brace.
The regression was probably introduced with [this commit](fa0118e6e5).
Originally contributed by Ahmed Mahdy (@aybassiouny). Thank you!
Reviewed By: Wawha, HazardyKnusperkeks
Differential Revision: https://reviews.llvm.org/D99031
Allocation functions should be marked with onlyAccessesInaccessibleMemory (when that is correct for the given function) which is checked elsewhere so this check is no longer needed.
Differential Revision: https://reviews.llvm.org/D117180
John Ericson