search for: sunk

I'm looking at fixing some layering violations in LLVM & came across a few in the CodeView handling, specifically: lib/MC/MCCodeView includes several llvm/DebugInfo/CodeView headers I guess MC could be made dependent on DebugInfoCodeView? But probably these things should be sunk into BinaryFormat as is the case for DWARF features used by MC? include/llvm/Object/COFF.h includes include/llvm/DebugInfo/CodeView/CVDebugRecord.h Also seems like this could/should/needs to be sunk into BinaryFormat? I'm open to ideas & happy to do the work, or help in any way that mig...

Yes, some of the headers and stuff that are just raw structure definitions and enums could probably be sunk into BinaryFormat.. How'd you find this? Curious why it hasn't been breaking in modules builds for a long time. On Wed, Mar 21, 2018 at 11:31 AM David Blaikie <dblaikie at gmail.com> wrote: > I'm looking at fixing some layering violations in LLVM & came across a few &gt...

...e modules build wouldn't fail - it knows nothing about .cpp modularity/grouping, just headers)) On Wed, Mar 21, 2018 at 11:42 AM Zachary Turner <zturner at google.com> wrote: > Yes, some of the headers and stuff that are just raw structure definitions > and enums could probably be sunk into BinaryFormat.. > > How'd you find this? Curious why it hasn't been breaking in modules > builds for a long time. > > On Wed, Mar 21, 2018 at 11:31 AM David Blaikie <dblaikie at gmail.com> wrote: > >> I'm looking at fixing some layering violations in L...

...f the identified problems regarding loop unrolling, loop unswitching, etc. Credit to John McCall for talking this over with me and seeding the core ideas. Today, convergent operations may only be moved into control-equivalent locations, or, in layman’s terms, a convergent operation may neither be sunk into nor hoisted out of, a condition. This causes problems for full loop unrolling, as the control dependence on the loop counter is eliminated, but our intuition indicates that this dependence was somehow trivial. More concretely, all know uses of convergent are OK with full unrolling, making th...

...9;t fail - it knows nothing about .cpp modularity/grouping, just headers)) On Wed, Mar 21, 2018 at 11:42 AM Zachary Turner <zturner at google.com<mailto:zturner at google.com>> wrote: Yes, some of the headers and stuff that are just raw structure definitions and enums could probably be sunk into BinaryFormat.. How'd you find this? Curious why it hasn't been breaking in modules builds for a long time. On Wed, Mar 21, 2018 at 11:31 AM David Blaikie <dblaikie at gmail.com<mailto:dblaikie at gmail.com>> wrote: I'm looking at fixing some layering violations in LL...

...- it knows nothing > about .cpp modularity/grouping, just headers)) > > > > On Wed, Mar 21, 2018 at 11:42 AM Zachary Turner <zturner at google.com> > wrote: > > Yes, some of the headers and stuff that are just raw structure definitions > and enums could probably be sunk into BinaryFormat.. > > > > How'd you find this? Curious why it hasn't been breaking in modules > builds for a long time. > > > > On Wed, Mar 21, 2018 at 11:31 AM David Blaikie <dblaikie at gmail.com> wrote: > > I'm looking at fixing some layering...

...layering violations in LLVM & came across a >> few in the CodeView handling, specifically: >> >> lib/MC/MCCodeView includes several llvm/DebugInfo/CodeView headers >> I guess MC could be made dependent on DebugInfoCodeView? But probably >> these things should be sunk into BinaryFormat as is the case for DWARF >> features used by MC? >> > > I'd be OK with that. We could very easily introduce true link dependencies > on that library in the near future. > OK - added the LLVMBuild.txt change in r328595 > include/llvm/Object/COFF.h...

...oking at fixing some layering violations in LLVM & came across a few > in the CodeView handling, specifically: > > lib/MC/MCCodeView includes several llvm/DebugInfo/CodeView headers > I guess MC could be made dependent on DebugInfoCodeView? But probably > these things should be sunk into BinaryFormat as is the case for DWARF > features used by MC? > I'd be OK with that. We could very easily introduce true link dependencies on that library in the near future. > include/llvm/Object/COFF.h includes > include/llvm/DebugInfo/CodeView/CVDebugRecord.h > Also s...

Full_Name: Tadashi Kadowaki Version: 2.4.0 OS: Redhat Linux 9 Submission from: (NULL) (58.12.166.67) Doesn't nls function support subset? It seems not to work. And, there are no information in the online help. Has it sunk into oblivion?

...tions deleted because all callers found 496 inline - Number of functions inlined 12 instcombine - Number of constant folds 512 instcombine - Number of dead inst eliminated 3 instcombine - Number of dead stores eliminated 70 instcombine - Number of instructions sunk 876 instcombine - Number of insts combined 2 licm - Number of call insts hoisted or sunk 539 licm - Number of instructions hoisted out of loop 37 licm - Number of instructions sunk out of loop 7 licm - Number of load insts hoisted or su...

...ieve and so not available to the general public :)) Some discussion can be found here: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20100531/102160.html. Unfortunately, no real testcase is provided, just abstract scenario description. Basically a EFLAGS-clobbering instruction is not sunk if EFLAGS might be live out of the current block and since a conditional branch doesn't mark its EFLAGS use as a kill, that effectively means it is never sunk. But how can that happen? I think the only way for it is when EFLAGS def and use are located in different blocks but that is impossible...

...across a >>>> few in the CodeView handling, specifically: >>>> >>>> lib/MC/MCCodeView includes several llvm/DebugInfo/CodeView headers >>>> I guess MC could be made dependent on DebugInfoCodeView? But probably >>>> these things should be sunk into BinaryFormat as is the case for DWARF >>>> features used by MC? >>>> >>> >>> I'd be OK with that. We could very easily introduce true link >>> dependencies on that library in the near future. >>> >> >> OK - added the L...

In plotmath expressions, R's degree symbol, e.g. shown by plot(1, main = parse(text = "1*degree*C")) has sunk to halfway the text line, instead of touching its top. In older R versions this looked much better. -- Edzer Pebesma Institute for Geoinformatics Heisenbergstrasse 2, 48151 Muenster, Germany Phone: +49 251 8333081

...in LLVM & came across a >>> few in the CodeView handling, specifically: >>> >>> lib/MC/MCCodeView includes several llvm/DebugInfo/CodeView headers >>> I guess MC could be made dependent on DebugInfoCodeView? But probably >>> these things should be sunk into BinaryFormat as is the case for DWARF >>> features used by MC? >>> >> >> I'd be OK with that. We could very easily introduce true link >> dependencies on that library in the near future. >> > > OK - added the LLVMBuild.txt change in r328595 &...

...2)   I'd like to return a vector that contains only the values that are shared across ALL THREE vectors. So the statement would return a vector that looked like this: 4,5,6,7,8,9,10,11,12,52   For some reason I initially thought that a cbind and a unique() would handle it, but then common sense sunk in.  I think the sleep deprivation is starting to take it's toll.  I've got to believe that there is a simple solution to this dilema.   Thanks in adance for any help! C [[alternative HTML version deleted]]

...available to the general public J) > Some discussion can be found here: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20100531/102160.html. Unfortunately, no real testcase is provided, just abstract scenario description. > > Basically a EFLAGS-clobbering instruction is not sunk if EFLAGS might be live out of the current block and since a conditional branch doesn’t mark its EFLAGS use as a kill, that effectively means it is never sunk. But how can that happen? I think the only way for it is when EFLAGS def and use are located in different blocks but that is impossible with...

...;>> few in the CodeView handling, specifically: >>>>> >>>>> lib/MC/MCCodeView includes several llvm/DebugInfo/CodeView headers >>>>> I guess MC could be made dependent on DebugInfoCodeView? But >>>>> probably these things should be sunk into BinaryFormat as is the case for >>>>> DWARF features used by MC? >>>>> >>>> >>>> I'd be OK with that. We could very easily introduce true link >>>> dependencies on that library in the near future. >>>> >>&g...

...ll sites analyzed 34,766 5 instcombine # constant folds 3,979,078 2,004 instcombine # dead inst eliminated 6,323 2 instcombine # dead stores eliminated 1,522 4 instcombine # factorizations 254,146 66 instcombine # instructions sunk 10,427,131 1,749 instcombine # insts combined 57,943 -205 instcombine # reassociations 1,072 1 instsimplify # expansions 135,129 1 instsimplify # reassociations 121,777 246 instsimplify # redundant instructions removed 27,612...

...onstant expression. For example you can output the equivalent of if (x == 0) llvm.trap() // or maybe some special trap routine y = whatever/x ... use y ... rather than the constant expression "whatever/x". If the optimizers can prove that x is never zero then this will be sunk back into a constant expression (if "whatever" is a constant expression). Ciao, Duncan. > > Thanks, > Cameron > > P.S. Is it your intention to hide the simple constant expression from the > constant folder? I.e. (x == 0) ? 0 : x.

...tified problems regarding loop unrolling, loop unswitching, etc. Credit to John McCall for talking this over with me and seeding the core ideas. > > Today, convergent operations may only be moved into control-equivalent locations, or, in layman’s terms, a convergent operation may neither be sunk into nor hoisted out of, a condition. This causes problems for full loop unrolling, as the control dependence on the loop counter is eliminated, but our intuition indicates that this dependence was somehow trivial. More concretely, all know uses of convergent are OK with full unrolling, making th...