similar to: LLC crash while handling DEBUG info

Hi David If you look at line https://github.com/llvm/llvm-project/blob/master/llvm/lib/IR/Verifier.cpp#L1160 there is IR verification which asserts that only in case of `spFlags = DISPFlagDefinition`, the compilation unit (`unit` field) should be present. Otherwise, it should *not* be present. In the crash case, `spFlags = DISPFlagOptimized`. So, I guess, `unit` field should *not* be present,

I am bit confused - `unit` must be present for definitions, and `optimized ` is also a `definition`, so, `unit` must be present for `optimized ` too. Am I right? Mahesha On Sun, May 31, 2020 at 10:14 PM David Blaikie <dblaikie at gmail.com> wrote: > definition and optimized are orthogonal (a function could be both, or > neither) - one says this DISubprogram describes a function

Let's forget about my malformed IR if it is adding additional confusion here. I mentioned it here to ease the conversation, but if it is causing confusion rather than making the discussion flow easier, then we better ignore it. The whole triggering point for this email initiative is - one of the applications is crashing with the stack trace that I mentioned earlier. The crash is during the

Hi All , We have a scenario in our debugger to handle the file index in the debug_ine info like $llvm-dwarfdump -debug-line test.o file_names[ 1]: name: "test.cpp" dir_index: 0 mod_time: 0x00000000 length: 0x00000000 file_names[ 2]: name: "test.cpp" dir_index: 1 mod_time: 0x00000000 length: 0x00000000

> Could you please describe what is the benefit of that? Currently there are two ways to provide DISPFlagDefinition, via bool and SPFlag, I would like to make it only via SPFlags, it will be NFC and it will make the changes in parser simpler for moving five flags from from DIFlags to DISPFlags. Currently parser checks the presence of SPFlags to see if the definition is present in bool or spflag

Try using $PWD/test.cpp on the clang command line. I am seeing the duplicate DIFile entries, but not yet able to reproduce a .debug_line section with multiple directory entries. --paulr From: llvm-dev <llvm-dev-bounces at lists.llvm.org> On Behalf Of Tomar, Sourabh Singh via llvm-dev Sent: Tuesday, September 1, 2020 1:07 PM To: Umesh Kalappa <umesh.kalappa0 at gmail.com>; cfe-dev at

Yes, removing the support for isLocal, isDefinition fields completely from ll files, currently the LLParser still parses it. I want to remove it and update the all the ll files which still uses it. Also the metadata read will support old format, no changes in that. so if ll file has isLocal and isDefinition it will result in parser error. But the bitcode read will work as usual. - Chirag.

Hello, In regard to the review request https://reviews.llvm.org/D74470, I am trying to move five of the DIFlags to DISPFlag for the moment namely DIFlagExplicit, DIFlagPrototyped, DIFlagNoReturn, DIFlagThunk, DIFlagAllCallsDescribed. The llvm ir format for DISubprogram currently has backword compatibility where the isLocal, isDefinition, virtuality, isOptimized and SPFlags are mutually exclusive.

Hi David, Sorry for taking a few days to reply. It's not easy for you to compile a Haskell file to see the problem as the debug information is still WIP. Below I prove the IR for a simple hello world program which you can feed into llc. https://gist.github.com/05296933e37e87533a51d493b46aa48d The `out.ir` file can be passed straight to `llc`. Can you see anything obviously wrong? Matt

There certainly is support; after all AMD supports both OpenCL and HIP (a dialect of C++ very close to cuda). AMD device libraries (in bitcode form) are installed when ROCm ( https://rocm.github.io/ ) is installed. AMD device libraries are mostly written in (OpenCL) C and open source at https://github.com/RadeonOpenCompute/ROCm-Device-Libs . They are configured by linking in a number tiny

Hi dev list, I am currently exploring the integration of AMDGPU/ROCm into the PACXX project and observing some strange behavior of the AMDGPU backend. The following IR is generated for a simple address space test that copies from global to shared memory and back to global after a barrier synchronization. Here is the IR is attached as as1.ll The output is as follows: 0 0 0 0 0 0 0 0 0 0 0 0 0

Finally I kind of modified llvm to generate assembly that can run on AMDGPU pro drivers. One problem is the performance of the code generated by llvm is about 10% slower than amdgpu's online compiler. Anything I can tune the performance up the performance of llvm?\ Thanks! On Tue, Sep 4, 2018 at 9:23 AM 董昌道 <dongchangdao at gmail.com> wrote: > I am writing a miner of crypto

Am Mi., 18. Nov. 2020 um 12:17 Uhr schrieb David Chisnall <David.Chisnall at cl.cam.ac.uk>: > On 18/11/2020 18:03, Michael Kruse wrote: > > This is missing the probably largest group: Users of clang who compile > > clang themselves (e.g. because their OS does not come with a package > > for clang, or is too old) > > I think our target for this group should be that

Hi All, I am first time trying build CLANG+LLVM using cmake+ninja build system. I updated all my CLANG+LLVM sources to current trunk, and I successfully built it using classic *make* build system. But, trying to build the same with cmake+ninja build system resulting in following build failures for compiler-rt sources. Am I missing something basics here? ==================== cmake command used:

I'm seeing this too. CC'ing the author ubsan stuff. Richard, I know you were OK with only supporting Clang-bootstraps, but I don't think that's terribly viable here. We should be able to build ubsan's runtime with standards conforming code unless there is some fairly extreme reason not to... On Fri, Oct 12, 2012 at 7:19 AM, Mahesha HS <mahesha.llvm at gmail.com> wrote:

On Sat, Oct 13, 2012 at 1:09 AM, David Blaikie <dblaikie at gmail.com> wrote: > On Fri, Oct 12, 2012 at 6:14 PM, Chandler Carruth <chandlerc at google.com> > wrote: > > I'm seeing this too. CC'ing the author ubsan stuff. > > > > Richard, I know you were OK with only supporting Clang-bootstraps, but I > > don't think that's terribly viable

I use latest cmake+ninja which are built from latest sources. ================================= > cmake --version cmake version 2.8.9.20121011-g2876 ================================= -- mahesha On Fri, Oct 12, 2012 at 7:35 PM, Mahesha HS <mahesha.llvm at gmail.com> wrote: > Hi All, > > I am first time trying build CLANG+LLVM using cmake+ninja build > system. I updated all

On Sat, Oct 13, 2012 at 2:33 AM, Mahesha HS <mahesha.llvm at gmail.com> wrote: > On Sat, Oct 13, 2012 at 1:31 PM, Tobias Grosser <tobias at grosser.es> wrote: >> On 10/13/2012 04:38 AM, Mahesha HS wrote: >>> >>> On Sat, Oct 13, 2012 at 5:14 AM, Eli Friedman <eli.friedman at gmail.com> >>> wrote: >>>> >>>> On Wed, Oct 10,

On Fri, Oct 12, 2012 at 6:14 PM, Chandler Carruth <chandlerc at google.com> wrote: > I'm seeing this too. CC'ing the author ubsan stuff. > > Richard, I know you were OK with only supporting Clang-bootstraps, but I > don't think that's terribly viable here. Just out of curiosity - why isn't that viable? I'd sort of hope to treat optional sanitizer runtimes

Hi Eli, Attached zipped file, named, "fopenmp_option_support.tar.gz" contains the first patch, along with relevant *test case*. This patch is to support the option "-fopenmp" option in Clang. Following files are changed in this patch. Please start going through this patch, and let me know comments. Meanwhile, I will prepare next patch.