similar to: Publication

I've found what appears to be a bug in instcombine. Specifically, the transformation of -(X/C) to X/(-C) is invalid if C == INT_MIN. Specifically, if I have > define i32 @foo(i32 %x) #0 { > entry: > %div = sdiv i32 %x, -2147483648 > %sub = sub nsw i32 0, %div > ret i32 %sub > } then opt -instcombine will produce > define i32 @foo(i32 %x) #0 { > entry: > %sub

There is in fact a transform out there somewhere that reverses yours. define i64 @foo(i64 %a) { %b = shl i64 %a, 5 %c = add i64 %b, %a ret i64 %c } becomes define i64 @foo(i64 %a) { %c = mul i64 %a, 33 ret i64 %c } ~Craig On Wed, Sep 13, 2017 at 10:11 AM, Craig Topper <craig.topper at gmail.com> wrote: > Your code seems fine. InstCombine can infinite loop if some other

Hello, I am in need of moderators to help moderate each session. If you are interested, please respond to this mail and indicate which slot you can moderate. Its a long shot as you would miss talks, but if anyone wants a break to do work or check email while also staffing the registration desk, please let me know if you can do it during any of the sessions below. Thanks! Options: Moderator #1

I'm still in need for moderators for the following sessions: > Moderator #3 (11:15-12:45, Regency 1) > Verified LLVM: Formalizing the semantics of the LLVM Intermediate Representation for Verified Program Transformations- Santosh Nagarakatte, University of Pennsylvania/Rutgers University > Integrated Security, using LLVM for Dynamic and Static Security Tasks- Jared Carlson,

LLVM developers, Can you add the following paper to the publication list? "Formal Verification of SSA Optimizations for LLVM" Jianzhou Zhao, Santosh Nagarakatte, Milo M K Martin and Steve Zdancewic Proceedings of the 34th ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI) 2013 link to ACM DL:

On Mon, Aug 22, 2011 at 2:49 PM, Santosh Nagarakatte <santosh.nagarakatte at gmail.com> wrote: > Hi all, > > I have been trying to understand the use of undef in both sequential > and concurrent programs. > > >From the LLVM Language Reference Manual, I see the following > definition of undef. > "Undef can be used anywhere a constant is expected, and indicates

On Mon, Aug 22, 2011 at 8:46 PM, Jianzhou Zhao <jianzhou at seas.upenn.edu> wrote: > On Mon, Aug 22, 2011 at 6:08 PM, Eli Friedman <eli.friedman at gmail.com> wrote: >> On Mon, Aug 22, 2011 at 2:49 PM, Santosh Nagarakatte >> <santosh.nagarakatte at gmail.com> wrote: >>> Hi all, >>> >>> I have been trying to understand the use of undef in

On Mon, Aug 22, 2011 at 6:08 PM, Eli Friedman <eli.friedman at gmail.com> wrote: > On Mon, Aug 22, 2011 at 2:49 PM, Santosh Nagarakatte > <santosh.nagarakatte at gmail.com> wrote: >> Hi all, >> >> I have been trying to understand the use of undef in both sequential >> and concurrent programs. >> >> >From the LLVM Language Reference Manual, I

Hi all, I have been trying to understand the use of undef in both sequential and concurrent programs. >From the LLVM Language Reference Manual, I see the following definition of undef. "Undef can be used anywhere a constant is expected, and indicates that the user of the value may receive an unspecified bit-pattern". LLVM Language Reference manual also demonstrates how optimizers

On Mon, Aug 22, 2011 at 3:40 PM, Jianzhou Zhao <jianzhou at seas.upenn.edu> wrote: > On Mon, Aug 22, 2011 at 6:08 PM, Eli Friedman <eli.friedman at gmail.com> wrote: >> On Mon, Aug 22, 2011 at 2:49 PM, Santosh Nagarakatte >> <santosh.nagarakatte at gmail.com> wrote: >>> Hi all, >>> >>> I have been trying to understand the use of undef in

On Tue, Aug 23, 2011 at 12:08 AM, Eli Friedman <eli.friedman at gmail.com> wrote: > On Mon, Aug 22, 2011 at 8:46 PM, Jianzhou Zhao <jianzhou at seas.upenn.edu> wrote: >> On Mon, Aug 22, 2011 at 6:08 PM, Eli Friedman <eli.friedman at gmail.com> wrote: >>> On Mon, Aug 22, 2011 at 2:49 PM, Santosh Nagarakatte >>> <santosh.nagarakatte at gmail.com>

On Mon, Aug 22, 2011 at 3:49 PM, Eli Friedman <eli.friedman at gmail.com> wrote: > On Mon, Aug 22, 2011 at 3:40 PM, Jianzhou Zhao <jianzhou at seas.upenn.edu> wrote: >> On Mon, Aug 22, 2011 at 6:08 PM, Eli Friedman <eli.friedman at gmail.com> wrote: >>> On Mon, Aug 22, 2011 at 2:49 PM, Santosh Nagarakatte >>> <santosh.nagarakatte at gmail.com>

On Mon, Aug 22, 2011 at 6:49 PM, Eli Friedman <eli.friedman at gmail.com> wrote: > On Mon, Aug 22, 2011 at 3:40 PM, Jianzhou Zhao <jianzhou at seas.upenn.edu> wrote: >> On Mon, Aug 22, 2011 at 6:08 PM, Eli Friedman <eli.friedman at gmail.com> wrote: >>> On Mon, Aug 22, 2011 at 2:49 PM, Santosh Nagarakatte >>> <santosh.nagarakatte at gmail.com>

Hi all, I am testing link time optimizations with clang on linux x86_64 machine running Ubuntu Maverick. I am using clang with llvm release-3.0 I followed the instructions on http://llvm.org/docs/LinkTimeOptimization.html However, when I try to compile with one file being bitcode and other being native code exactly as shown in the documentation, I get the following errors clang a.o main.o -o

On Mon, Aug 22, 2011 at 7:27 PM, Jeffrey Yasskin <jyasskin at google.com> wrote: > On Mon, Aug 22, 2011 at 3:49 PM, Eli Friedman <eli.friedman at gmail.com> wrote: >> On Mon, Aug 22, 2011 at 3:40 PM, Jianzhou Zhao <jianzhou at seas.upenn.edu> wrote: >>> On Mon, Aug 22, 2011 at 6:08 PM, Eli Friedman <eli.friedman at gmail.com> wrote: >>>> On

On Mon, Aug 22, 2011 at 5:29 PM, Jianzhou Zhao <jianzhou at seas.upenn.edu> wrote: > On Mon, Aug 22, 2011 at 7:27 PM, Jeffrey Yasskin <jyasskin at google.com> wrote: >> On Mon, Aug 22, 2011 at 3:49 PM, Eli Friedman <eli.friedman at gmail.com> wrote: >>> On Mon, Aug 22, 2011 at 3:40 PM, Jianzhou Zhao <jianzhou at seas.upenn.edu> wrote: >>>> On

LLVM developers, I was wondering if the program would still be safe if I strip the byval attributes from the parameters in the entire bitcode. LLVM language reference manual states that "The attribute implies that a hidden copy of the pointee is made between the caller and the callee, so the callee is unable to modify the value in the callee. This attribute is only valid on LLVM pointer

On Mon, Aug 22, 2011 at 4:20 PM, Jianzhou Zhao <jianzhou at seas.upenn.edu> wrote: > On Mon, Aug 22, 2011 at 6:49 PM, Eli Friedman <eli.friedman at gmail.com> wrote: >> On Mon, Aug 22, 2011 at 3:40 PM, Jianzhou Zhao <jianzhou at seas.upenn.edu> wrote: >>> On Mon, Aug 22, 2011 at 6:08 PM, Eli Friedman <eli.friedman at gmail.com> wrote: >>>> On

On Mon, Aug 22, 2011 at 7:35 PM, Jeffrey Yasskin <jyasskin at google.com> wrote: > On Mon, Aug 22, 2011 at 4:20 PM, Jianzhou Zhao <jianzhou at seas.upenn.edu> wrote: >> On Mon, Aug 22, 2011 at 6:49 PM, Eli Friedman <eli.friedman at gmail.com> wrote: >>> On Mon, Aug 22, 2011 at 3:40 PM, Jianzhou Zhao <jianzhou at seas.upenn.edu> wrote: >>>> On

Folks, I just added two publications to www-pubs (r280004), as it used to be, but it seems the results are not up yet: http://llvm.org/pubs/ Also, my SVN www-pubs seems to only have papers up to 2012, while the page has it up to 2015. I'm confused, what's the right way to update pubs nowadays? cheers, --renato