search for: bondhugula

...ers. > > Cool. I’d really recommend you connect with some of the loop optimization > people working on MLIR to learn more about what they are doing, because it > is directly related to this and I’d love for there to be more communication. > > I’ve cc'd Nicolas Vasilache, Uday Bondhugula, and Albert Cohen as examples > that would be great to connect with. > You may have already seen my discussion with Uday on the mailing list. I would like to discuss approaches with all 3 of them, at latest at EuroLLVM (or contact be before that, e.g. on this mailing-list thread). Michael -...

Am Mo., 27. Jan. 2020 um 22:06 Uhr schrieb Uday Kumar Reddy Bondhugula < uday at polymagelabs.com>: > Hi Michael, > > Although the approach to use a higher order in-memory abstraction like the > loop tree will make it easier than what you have today, if you used MLIR > for this representation, you already get a round trippable textual format >...

...really recommend you connect with some of the loop >> optimization people working on MLIR to learn more about what they are >> doing, because it is directly related to this and I’d love for there to be >> more communication. >> >> I’ve cc'd Nicolas Vasilache, Uday Bondhugula, and Albert Cohen as >> examples that would be great to connect with. >> > > You may have already seen my discussion with Uday on the mailing list. I > would like to discuss approaches with all 3 of them, at latest at EuroLLVM > (or contact be before that, e.g. on this maili...

...State of the art: - POPL 2016 http://conf.researchr.org/event/POPL-2016/popl-2016-papers-smo-an-integrated-approach-to-intra-array-and-inter-array-storage-optimization http://dl.acm.org/citation.cfm?id=2837636 There is a tool available but it has received little publicity yet: https://github.com/bondhugula/smo Yet the input is in for of polyhedral sets/maps, not code. - CC 2016, Darte, Isoard, Yuki https://www.conference-publishing.com/list.php?Event=CC16 Better heuristic. Currently limited to single-array SMO. Again, much potential for research and implementation. I'll be glad to discuss tran...

2016-06-27 15:52 GMT+05:00 4lbert C0hen <4lbert.h.c0hen at gmail.com>: > Dear Roman and all, > > Such features would be extremely useful to implement array expansion (scalar > and array renaming, privatization with new subscript expressions of higher > dimension) and storage mapping optimization (generalizing array > contraction). It would be interesting to have these

Am Mi., 5. Feb. 2020 um 18:13 Uhr schrieb Chris Lattner <clattner at nondot.org>: > If I understand your claims, you are claiming both that red/green trees are essential for a loop optimizer, and that this essential nature “justifies” the cost of reinventing an entire compiler infrastructure is lower than the benefit of using (e.g.) MLIR to do this. I haven’t seen evidence of either

Vinay Bondhugula (2): Compatibility with older versions of X Don't call conversion_proc for older Xservers git tag: xf86-input-vmmouse-12.5.1 http://xorg.freedesktop.org/archive/individual/driver/xf86-input-vmmouse-12.5.1.tar.bz2 MD5: 1e2feafb8e4f1b670aeb39bcac9fbb18 xf86-input-vmmouse-12.5.1.tar...

Vinay Bondhugula (3): More xorg and xfree86 backwards compatibility Fix VMWARE_CTRL and Xinerama Initialization Prepare for 10.16.2 git tag: xf86-video-vmware-10.16.2 http://xorg.freedesktop.org/archive/individual/driver/xf86-video-vmware-10.16.2.tar.bz2 MD5: d79b660e1effb86628930e5c01989216 xf...

Adam Jackson (2): Fix distcheck. vmware 10.16.0 Bankim Bhavsar (2): Xv extension for VMware's video driver Video Overlay: Source video parameters Matthieu Herrb (1): Makefile.am: nuke RCS Id Vinay (1): Trivial libpciaccess related fix Vinay Bondhugula (1): Port to libpciaccess. git tag: xf86-video-vmware-10.16.0 http://xorg.freedesktop.org/archive/individual/driver/xf86-video-vmware-10.16.0.tar.bz2 MD5: 2768dd5c5c3867ac6fc390ae15dc85bd xf86-video-vmware-10.16.0.tar.bz2 SHA1: 57d87ce92d481d85506b13707139a5d5e18f61d3 xf86-video-vmware-10...

Adam Jackson (1): vmmouse 12.5.0 Matthieu Herrb (1): Makefile.am: nuke RCS Id Vinay Bondhugula (2): Fix an old sign bug for the relative mode Fix device to screen space coordinate conversion git tag: xf86-input-vmmouse-12.5.0 http://xorg.freedesktop.org/archive/individual/driver/xf86-input-vmmouse-12.5.0.tar.bz2 MD5: 367357d01f33a03888559e2e5cc6e08e xf86-input-vmmouse-12.5.0.t...

Am Do., 30. Jan. 2020 um 04:40 Uhr schrieb Uday Kumar Reddy Bondhugula <uday at polymagelabs.com>: > There are multiple ways regions in MLIR can be viewed, but the more relevant point here is you do have a loop tree structure native in the IR with MLIR. Regions in MLIR didn't evolve from modeling inlined calls - the affine.for/affine.if were originally th...

In the 2018 LLVM DevMtg [1], I presented some shortcomings of how LLVM optimizes loops. In summary, the biggest issues are (a) the complexity of writing a new loop optimization pass (including needing to deal with a variety of low-level issues, a significant amount of required boilerplate, the difficulty of analysis preservation, etc.), (b) independent optimization heuristics and a fixed pass