similar to: *********Recuperate return addresses of virtual and non virtual C++ functions with Clang and LLVM

I am working with this setup: LLVM+Clang Compiler on an Ubuntu 64bit OS. Questions: Is it possible to get the return addresses of virtual function call and non virtual C++ methods inside Clang and LLVM. Is thre a way to get the candidate addresses where the return address points back after each function return was performed. I have a few questions and hopefully you guys can help me out. The

I am working with this setup: LLVM+Clang Compiler on an Ubuntu 64bit OS. Questions: Is it possible to get the return addresses of virtual function call and non virtual C++ methods inside Clang and LLVM. Is thre a way to get the candidate addresses where the return address points back after each function return was performed. I have a few questions and hopefully you guys can help me out. The

Dear list, I apologies first for my English, hope you will understand well my question. I am working on 1/2 hour piezometric data, time unit is second. They present daily oscillation when using the spectrum() function. What I am really interested in, is to find the amplitude corresponding to this oscillation. I work with a college using Matlab, and although we apply the same methodology, our

On Tue, Oct 3, 2017 at 1:46 PM, Teresa Johnson via llvm-dev < llvm-dev at lists.llvm.org> wrote: > > > On Tue, Oct 3, 2017 at 1:38 PM, Graham Yiu <gyiu at ca.ibm.com> wrote: > >> Hi Teresa, >> >> Actually, enabling the new pass manager manually seems to have solved >> this issue, so this problem is only valid for the old pass manager. >> >

Thanks Easwaran. This is what we've observed as well, where the old PM inliner was only looking hot/cold callee information, which have signficantly smaller boosts/penalties compared to callsite information. Teresa, do you know if there is some documentation/video/presentation on how PGO information is represented in LLVM and what information is passed via the IR? I'm finding some

Hi Teresa, Actually, enabling the new pass manager manually seems to have solved this issue, so this problem is only valid for the old pass manager. Thanks, Graham Yiu LLVM Compiler Development IBM Toronto Software Lab Office: (905) 413-4077 C2-707/8200/Markham Email: gyiu at ca.ibm.com From: Teresa Johnson <tejohnson at google.com> To: Graham Yiu <gyiu at ca.ibm.com> Cc:

On Fri, Dec 15, 2017 at 12:22 AM, Sean Silva <chisophugis at gmail.com> wrote: > IIUC, this proposal is just saying that we should infer a static profile > for entry counts just like we do for branch probabilities. In the case of > entry counts, we do not hide that information behind an analysis like BPI, > so currently just annotating synthetic PGO entry counts is a simple >

Hello, My team and I noticed that callsite hotness information is not preserved from compile to link step with LTO/thinLTO enabled. As a result, the link step inlining pass remains conservative when inlining callsites known to be hot (ie. without the 'HotCallSiteThreshold' which is set at 3000 by default). There are likely many cross-module inlining opportunities lost this way, and

Hi Paul, It is the 4th time in two weeks that you repost the exact same two emails, that’s a bit high repost rate for the mailing-list, especially considering that you already got answers that you ignored previously: - http://lists.llvm.org/pipermail/llvm-dev/2016-December/108060.html - http://lists.llvm.org/pipermail/llvm-dev/2016-December/108051.html Adding an alternate email in case you

> On Dec 21, 2016, at 8:38 AM, Paul <paul at sec.in.tum.de> wrote: > > Hi Mehdi, > > thanks a lot for your email. Actually I was disappointed by the fact that I was not able to get a useful answer to my two questions. > As I observed that the questions move down along the list with time passing I decided to re-post them in the hope that the questions > will have a

On Fri, Dec 15, 2017 at 11:56 AM, Sean Silva <chisophugis at gmail.com> wrote: > > > On Fri, Dec 15, 2017 at 11:27 AM, Xinliang David Li <davidxl at google.com> > wrote: > >> >> >> On Fri, Dec 15, 2017 at 11:13 AM, Sean Silva <chisophugis at gmail.com> >> wrote: >> >>> >>> >>> On Fri, Dec 15, 2017 at 10:22

Hi Mehdi, thanks a lot for your email. Actually I was disappointed by the fact that I was not able to get a useful answer to my two questions. As I observed that the questions move down along the list with time passing I decided to re-post them in the hope that the questions will have a better visibility. Could you help me with the two questions? Finding an expert in this area would be of great

On Fri, Dec 15, 2017 at 11:13 AM, Sean Silva <chisophugis at gmail.com> wrote: > > > On Fri, Dec 15, 2017 at 10:22 AM, Easwaran Raman <eraman at google.com> > wrote: > >> >> >> On Fri, Dec 15, 2017 at 12:22 AM, Sean Silva <chisophugis at gmail.com> >> wrote: >> >>> IIUC, this proposal is just saying that we should infer a

I want to detect bad casts in C++ code by using the Clang compiler. The approach is similar to what Caver and TypeSan do but without using the compiler-rt. Caver and TypeSan: https://www.usenix.org/system/files/conference/usenixsecurity15/sec15-paper-lee.pdf https://nebelwelt.net/publications/files/16CCS2.pdf For example if I have the following C++ code snippet where I want to cast object b into

I want to detect bad casts in C++ code by using the Clang compiler. The approach is similar to what Caver and TypeSan do but without using the compiler-rt. Caver and TypeSan: https://www.usenix.org/system/files/conference/usenixsecurity15/sec15-paper-lee.pdf https://nebelwelt.net/publications/files/16CCS2.pdf For example if I have the following C++ code snippet where I want to cast object b into

Functions in LLVM IR have a function_entry_count metadata that is attached in PGO compilation. By using the entry count together with the block frequency info, the compiler computes the profile count of call instructions based on which the hotness/coldness of callsites can be determined. Experiments have shown that using a higher threshold for hot callsites results in improved runtime performance

The profile summary is saved in the global metadata ASAIK. If we want to calculate if something is hot/cold while choosing functions for importing, we would either need to read whole Module (which we clearly don't want to do) or duplicate this information in the summary, so we could get it without reading Module. 2016-09-02 15:49 GMT-07:00 Mehdi Amini <mehdi.amini at apple.com>: >

2016-09-02 15:04 GMT-07:00 Xinliang David Li <davidxl at google.com>: > On Fri, Sep 2, 2016 at 2:58 PM, Piotr Padlewski > <piotr.padlewski at gmail.com> wrote: > > Hi, > > I am working right now on importing based on PGO/FDO data. There is one > > issue that I found - when we calculate the list of imports, we can't get > the > > ProfileSummaryInfo,

On Fri, Sep 2, 2016 at 3:30 PM, Xinliang David Li <davidxl at google.com> wrote: > On Fri, Sep 2, 2016 at 3:16 PM, Piotr Padlewski > <piotr.padlewski at gmail.com> wrote: > > > > > > 2016-09-02 15:04 GMT-07:00 Xinliang David Li <davidxl at google.com>: > >> > >> On Fri, Sep 2, 2016 at 2:58 PM, Piotr Padlewski > >>

Hi, I am working right now on importing based on PGO/FDO data. There is one issue that I found - when we calculate the list of imports, we can't get the ProfileSummaryInfo, which is the best and I think only valid way of checking if callsite/callee is hot (isHotCount()). There are 2 solutions that I come up with Teresa and Easwaran: 1. Add PGO data to summary 2. Replace