similar to: [LLVMdev] which Register allocator to use with llc -O0

I will be removing RegAllocLinearScan and VirtRegRewriter from trunk shortly after we cut the 3.0 release branch. LLVM 3.0 will still ship with the linear scan register allocator, but the default will be the new greedy allocator. Linear scan can be enabled by passing '-regalloc=linearscan -join-physregs' to llc. RegAllocLinearScan and VirtRegRewriter need to go away soon because they

What do you mean by "current top of tree"?   Pankaj ________________________________ From: Eric Christopher <echristo at apple.com> To: Pankaj Gode <godepankaj at yahoo.com> Cc: "llvmdev at cs.uiuc.edu" <llvmdev at cs.uiuc.edu> Sent: Monday, October 3, 2011 10:32 PM Subject: Re: [LLVMdev] collect end line number for scope On Oct 3, 2011, at 4:36 AM,

Hi all, I've come across a problem with register allocation which I have been unable to track down the root cause of. 6728B %vreg304<def> = COPY %vreg278; VRF128:%vreg304,%vreg278 6736B %vreg302<def> = COPY %vreg278; VRF128:%vreg302,%vreg278 6752B %vreg278<def,tied1> = foo %vreg278<tied0>, %vreg277, 14, pred:1, pred:%noreg, 5; VRF128:%vreg278 VRF64_l:%vreg277 * bar

Hi, He is referring to current SVN head. Regards, Alex On Tue, Oct 4, 2011 at 12:08 PM, Pankaj Gode <godepankaj at yahoo.com> wrote: > What do you mean by "current top of tree"? > > Pankaj > From: Eric Christopher <echristo at apple.com> > To: Pankaj Gode <godepankaj at yahoo.com> > Cc: "llvmdev at cs.uiuc.edu" <llvmdev at

Hi Eric,   Thanks for patch information. I have checked the patch with my code and I see that it works. But when I extract the variable and check the metadata information to retrieve the scope information, I do not get exact scope, after optimization. As you said that this is a work in progress, so can we expect full version on this in llvm3.3 release ?   Regards, Pankaj  

Hi All,   Due to 'LowerDbgDeclare' call ( as part of 'instruction combining' optimization), a local variable gets declared and initialized inside the basic blocks it is used in. Is there anyway I can avoid this ?   This is with reference with my previous question. http://lists.cs.uiuc.edu/pipermail/llvmdev/2013-April/061644.html     Regards, Pankaj -------------- next part

On 9/26/2017 6:47 PM, Matthias Braun wrote: > >> On Sep 26, 2017, at 3:33 PM, Geoff Berry <gberry at codeaurora.org >> <mailto:gberry at codeaurora.org>> wrote: >> >> >> >> On 9/26/2017 6:11 PM, Matthias Braun wrote: >>>> On Sep 26, 2017, at 2:39 PM, Geoff Berry via llvm-dev >>>> <llvm-dev at lists.llvm.org

> On Sep 26, 2017, at 8:24 PM, Geoff Berry via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > On 9/26/2017 6:47 PM, Matthias Braun wrote: >>> On Sep 26, 2017, at 3:33 PM, Geoff Berry <gberry at codeaurora.org <mailto:gberry at codeaurora.org> <mailto:gberry at codeaurora.org <mailto:gberry at codeaurora.org>>> wrote: >>> >>>

There was a patch committed the other day that should, at least, work around some of the behavior you describe. Optimized debug info is an area that's being worked on in ToT and previous releases are particularly bad. I'd use that. -eric On Mon, Apr 29, 2013 at 7:02 AM, Pankaj Gode <godepankaj at yahoo.com> wrote: > Hi All, > > Due to 'LowerDbgDeclare' call ( as

Hi,   The link and the information shared was helpful.   I will make my problem definition more clear. While I am "asm printing" target code, I also want to emit scope related information. Scope related information includes, - for each scope, start line, end line, start column, end column and - scope heirarchy. As scope is delimited by "{" and "}" (for an input

Hi Quentin, Jonas, Splitting fits in with PBQP reasonably well, at least conceptually. The PBQP graph is designed to be mutable, so there is no problem with updating it when splitting. As I see it, there are two logical places to integrate splitting into PBQP: 1) Split during spilling -- If a PBQP solution selects the spill option for a node, rather than spill immediately, split the interval

Hi, Can anyone help me with the stack slot coloring optimization? This corresponding file is /lib/codegen/stackslotcoloring.cpp. It is said this optimization was for stack slot overlay for frame size reduction, after register allocation phase. And this transformation pass relies on the LiveStack analysis pass. How, when checking the source code, it seems the LiveStack analysis has not been

Both approaches are not exclusive. I would even think it makes sense to have a pre-split pass to prepare the graph, with a global view, and later on use use trySplit (or an equivalent) to handle the local coloring issues. From: Quentin Colombet [mailto:qcolombet at apple.com] Sent: 09 March 2015 23:08 To: Lang Hames Cc: Jonas Paulsson; llvmdev at cs.uiuc.edu; Arnaud De Grandmaison Subject:

Hi All,   int global; int func( int t) {    //scope 1   {      ....   } <-----   return x; }   For the above code, i want to collect endline (indicated by <---) for the scope. Can we get this information from the Dwarf Information in llvm 2.9 ?     Thanks & Regards, Pankaj -------------- next part -------------- An HTML attachment was scrubbed... URL:

Pankaj, If you want to decorate MachineInstrs then for the end of scope you're not looking at "}" but instead you're looking at _last_ machine instruction in that scope. Now, if you want to find out start and end MachineInstrs for a lexical scope (and corresponding line numbers) then see CodeGen/LexicalScopes pass. It collects lexical scope information and maps MIs to respective

Hi All, I was checking for GIT or SVN versions corresponding to llvm3.1 release by checking svn and git logs, but failed to find the exact match. I checked with r156747, which is the last version it shows in      http://llvm.org/svn/llvm-project/llvm/tags/RELEASE_31/ but when I 'diff' this with the 'downloaded llvm3.1 release'(http://llvm.org/releases/download.html#3.1), then I

Hal's advice helps me a lot to understand the implementation much better. Thanks so much! So, now I am able to state my problem more clearly: 1) There are two kinds of locals, i.e., the local variables originated from the source code (like C/C++), and the compilation generated temporaries. After instruction selection phase, the former is seen as frame indexes, while the latter is seen as

Hi Kristof, Thanks for the measurements. > On May 24, 2017, at 6:00 AM, Kristof Beyls <kristof.beyls at arm.com> wrote: > >> >> On 23 May 2017, at 21:48, Quentin Colombet <qcolombet at apple.com <mailto:qcolombet at apple.com>> wrote: >> >> Great! >> I thought I had to look at our pipeline at O0 to make sure optimized regalloc was

Hi Kristof, Thanks for going back so fast! > On May 24, 2017, at 12:57 PM, Kristof Beyls <kristof.beyls at arm.com> wrote: > >> >> On 24 May 2017, at 19:31, Quentin Colombet <qcolombet at apple.com <mailto:qcolombet at apple.com>> wrote: >> >> Hi Kristof, >> >> Thanks for the measurements. >> >>> On May 24, 2017, at

Hi Kristof, > On May 25, 2017, at 2:09 AM, Kristof Beyls <kristof.beyls at arm.com> wrote: > >> >> On 24 May 2017, at 22:01, Quentin Colombet <qcolombet at apple.com <mailto:qcolombet at apple.com>> wrote: >> >> Hi Kristof, >> >> Thanks for going back so fast! >> >>> On May 24, 2017, at 12:57 PM, Kristof Beyls