similar to: [LLVMdev] Landing my new development on the trunk ...

On Wed, Oct 27, 2010 at 1:29 PM, Brian West <bnwest at rice.edu> wrote: > Here is the patch for the new Operator Strength Reduction optimization > pass that I have written.  The bulk of the code is in > > lib/Transforms/Scalar/OperatorStrengthReduce.cpp > > The algorithm finds reduction opportunities in both array accesses and > explicit multiplications within loops.

On 10/27/10 8:34 PM, Eli Friedman wrote: > On Wed, Oct 27, 2010 at 1:29 PM, Brian West<bnwest at rice.edu> wrote: >> Here is the patch for the new Operator Strength Reduction optimization >> pass that I have written. The bulk of the code is in >> >> lib/Transforms/Scalar/OperatorStrengthReduce.cpp >> >> The algorithm finds reduction opportunities in

On Thu, Oct 28, 2010 at 9:38 AM, Brian West <bnwest at rice.edu> wrote: >> 3. LLVM already has a significant amount of infrastructure for loop >> passes; why does this pass have its own code for finding loops? > > I saw the loop infrastructure for CFG loops. This algorithm finds loops in > the data flow (more precisely: strongly-connected components in the >

Eli Friedman <eli.friedman <at> gmail.com> writes: > > Empirically the OSR optimization is compile-time faster than LSR. I have > > also noticed that OSR has more "analysis" requirements: Induction Variable > > User, Natural Loop Information, Canonicalize natural loops, and Scalar > > Evolution Analysis. Both OSR and LSR require the Dominator Tree

Eli Friedman <eli.friedman <at> gmail.com> writes: > >> > I did not mention in the original email (and should have) that OSR needs > >> > -instcombine to be run after it for cleanup. Also -licm, -reassociate, -gvn > >> > and -sccp can be enabling optimizations for OSR. > >> > >> Hmm... perhaps that could be partially fixed

> > A big downside of the current LSR algorithm is it's slow. I had initially > hoped that some of the heuristics would protect it better, but the problem > is > more complex than I had expected. I haven't done any measurements, > but it's likely that OSR is faster, which may interest some people > regardless > of how the output compares. > A few years

On 11/13/10 11:05 PM, Daniel Berlin wrote: > > A big downside of the current LSR algorithm is it's slow. I had > initially > hoped that some of the heuristics would protect it better, but the > problem is > more complex than I had expected. I haven't done any measurements, > but it's likely that OSR is faster, which may interest some

On Fri, Oct 29, 2010 at 11:18 AM, Brian West <bnwest at rice.edu> wrote: > Eli Friedman <eli.friedman <at> gmail.com> writes: >> >> > I did not mention in the original email (and should have) that OSR >> >> > needs >> >> > -instcombine to be run after it for cleanup. Also -licm, >> >> > -reassociate, -gvn >>

On 10/29/10 1:26 PM, Eli Friedman wrote: > Sure, but you know which induction variables you created; you can just > zap the unused ones at the end of the pass, no? This is feasible. We would have to collect more information during OSR proper pass and add logic to cleanup at the end. >> FWIW I noticed that other optimizations (as seen in StandardPasses.h) are >> followed by

On Oct 29, 2010, at 12:20 PM, Brian West wrote: > On 10/29/10 1:26 PM, Eli Friedman wrote: >> Sure, but you know which induction variables you created; you can just >> zap the unused ones at the end of the pass, no? > This is feasible. We would have to collect more information during OSR > proper pass and add logic to cleanup at the end. > >>> FWIW I noticed

On Mon, Nov 15, 2010 at 10:38 AM, Brian West <bnwest at rice.edu> wrote: > On 11/13/10 11:05 PM, Daniel Berlin wrote: > > A big downside of the current LSR algorithm is it's slow. I had >> initially >> hoped that some of the heuristics would protect it better, but the problem >> is >> more complex than I had expected. I haven't done any

Evan Cheng <evan.cheng <at> apple.com> writes: > Eli is right. We do need to see some benchmark numbers and understand how the pass will fit in the target > independent optimizer. While we encourage contribution, we typically don't commit new passes unless it > introduce new functionalities that have active clients. It would also help if you provide us with compile

Allow the user to define the type of partition and its number from the command line. The following example creates partition numer 4 with type 1c (hidden FAT32 LBA): isohybrid -fstype=28 -pentry=4 hybrid.iso --- utils/isohybrid.in | 11 ++++++++--- 1 files changed, 8 insertions(+), 3 deletions(-) diff --git a/utils/isohybrid.in b/utils/isohybrid.in index 83f9dc0..d67a2b5 100644 ---

Hi, don't know if this is the right list. Please post a better place, otherwise. I'm currently writing a LLVM ModulePass and ran into strange segfaults or endless loops within LLVM. My main question is, if this is a programming error or API misuse from me or a LLVM bug? Here is some minimal code, that triggers the bug: ---------------- class DebugPass : public ModulePass { public:

This moves the xen_modified_memory call from cpu_physical_memory_map to cpu_physical_memory_unmap because the memory could be migrated before the device model have written to it. But because we need to know the guest address and to avoid rewriting a new function, the call is moved to qemu_invalidate_entry. So this later has to new parameters, the length of the mapping and if it was a write.

When compiling a release version on Linux with GCC 3.4.4, I get the following warnings repeated multiple times when linking LLVMCore.o: /usr/bin/ld: `.gnu.linkonce.t._ZNK4llvm11Instruction13getOpcodeNameEv' referenced in section `.gnu.linkonce.r._ZNK4llvm11Instruction13getOpcodeNameEv' of /proj/llvm/build/lib/VMCore/Release/Instruction.o: defined in discarded section

Hi, openssh_cvs as of today, SCO Open Server 3.0, socklen_t this typedef doesn't exist on SCO OSR 3, and "configure" properly detects this, leading to /* #undef HAVE_SOCKLEN_T */ in config.h. Problem: I can't find any place where this is actually being used? I'd expect something like #ifndef HAVE_SOCKLEN_T typdef int socklen_t; #endif ("int" is what the

utils/mkdiskimage.in: New option -S to sync zeroizing after each cylinder and to issue pacifier messages. --- utils/mkdiskimage_patched_03.in 2014-01-30 20:54:27.000000000 +0100 +++ utils/mkdiskimage.in 2014-01-30 20:59:19.000000000 +0100 @@ -139,6 +139,7 @@ if ( $file eq '' || $c < 1 || $h < 1 || print STDERR " -4 use partition entry 4 (standard for

I'm iterating over MCInsts and I'd like to examine particular instructions. For example, I'd like to look at all x86 CALL64m instructions. I may be missing something, but it seems like my only option is to use MCInstPrinter::getOpcodeName and compare strings. (Of course, I could iterate through the opcodes and build up a table of the ones I'm interested in to avoid string

Hello Everyone, I need some help in retrieving the numeric value of an instruction operand from LLVM IR. this is what the IR looks like for a simple add function that adds two i32 integers define i32 @summ(i32 %a, i32 %b) #0 { entry: %add = add nsw i32 %b, %a ret i32 %add } i would like to know the integer value of %a and %b. I've tried -i->getOpcodeName() which gives me the