[Bug gprofng/32193] [display html] Files missing for a hardware event counter experiment
https://sourceware.org/bugzilla/show_bug.cgi?id=32193 --- Comment #1 from Ruud van der Pas --- I received two test cases, verified the bug and will investigate this. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/32193] [display html] Files missing for a hardware event counter experiment
https://sourceware.org/bugzilla/show_bug.cgi?id=32193 Ruud van der Pas changed: What|Removed |Added Status|NEW |ASSIGNED -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/32193] [display html] Files missing for a hardware event counter experiment
https://sourceware.org/bugzilla/show_bug.cgi?id=32193 Ruud van der Pas changed: What|Removed |Added Assignee|vladimir.mezentsev at oracle dot c |ruud.vanderpas at oracle dot com |om | CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/32193] New: [display html] Files missing for a hardware event counter experiment
https://sourceware.org/bugzilla/show_bug.cgi?id=32193 Bug ID: 32193 Summary: [display html] Files missing for a hardware event counter experiment Product: binutils Version: 2.43.1 Status: NEW Severity: normal Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- This bug has been reported by the Danish Technical University (DTU) in Denmark. In case a hardware event counter experiment is used as input, the source and disassembly files are missing. They are not produced at all. Everything else works just fine and the tool does not crash, but the information is missing. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/32085] Source file not recognized for gcc 11.4.0-compiled code
https://sourceware.org/bugzilla/show_bug.cgi?id=32085 Ruud van der Pas changed: What|Removed |Added CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/31459] need a way to suppress the stderr message
https://sourceware.org/bugzilla/show_bug.cgi?id=31459 --- Comment #3 from Ruud van der Pas --- Great to hear this works for you, James! It is interesting that the outfile option on the collect command was never documented, but is clearly useful. The unfortunate thing is that the outfile option as supported on gp-display-text has a different behaviour: $ gprofng display text -outfile LOG -limit 5 -functions graph.2.thr.er Print limit set to 5 $ cat LOG Functions sorted by metric: Exclusive Total CPU Time Excl. TotalIncl. Total Name CPUCPU sec. %sec. % 15.905 100.00 15.905 100.00 6.464 40.64 6.467 40.66 verify_bfs_tree._omp_fn.0 4.700 29.55 4.716 29.65 make_bfs_tree._omp_fn.0 0.716 4.50 0.716 4.50 mrg_get_uint_orig 0.695 4.37 0.695 4.37 __aarch64_ldadd8_sync $ As you can see, outfile stores the output from the command, but not the diagnostics. This is why it might be cleaner and clearer to introduce a new option on the gp-collect-app command. Something like -gp-output. It would be the same as --outfile, but possibly avoid confusion. Your feedback is much appreciated! -- You are receiving this mail because: You are on the CC list for the bug.
bug-binutils@gnu.org
https://sourceware.org/bugzilla/show_bug.cgi?id=31390 Ruud van der Pas changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED --- Comment #7 from Ruud van der Pas --- A patch has been tested and submitted. -- You are receiving this mail because: You are on the CC list for the bug.
bug-binutils@gnu.org
https://sourceware.org/bugzilla/show_bug.cgi?id=31390 --- Comment #5 from Ruud van der Pas --- Thanks for the update! Great to hear that this works for you now. I plan to work on a patch to eliminate the reliance on bignum and use bigint instead. -- You are receiving this mail because: You are on the CC list for the bug.
bug-binutils@gnu.org
https://sourceware.org/bugzilla/show_bug.cgi?id=31390 --- Comment #3 from Ruud van der Pas --- I found that after installing module bignum, the error that you reported manifested itself. I will dig some deeper, but as a workaround, the following works for me. At line 25 in file gp-display-html (e.g. $ sudo vi `which gp-display-html`) you could change "use bignum" by "use bigint". The program runs to completion then. -- You are receiving this mail because: You are on the CC list for the bug.
bug-binutils@gnu.org
https://sourceware.org/bugzilla/show_bug.cgi?id=31390 --- Comment #2 from Ruud van der Pas --- I have tried to reproduce the issue, but on Oracle Linux (OL), the perl-bignum package is installed as part of the Perl package. I am unfortunately not familiar with ArchLinux, but hopefully a "yum install perl", or "yum install perl-bignum" works. Otherwise, you may want to try using cpan and cpanm to install the bignum Perl module. This is well documented in this page: http://www.cpan.org/modules/INSTALL.html I tried this myself and the following commands worked for me: $ cpan App::cpanminus I relied on the defaults, but did set it up such that I could use sudo for subsequent installations: What approach do you want? (Choose 'local::lib', 'sudo' or 'manual') [local::lib] sudo Next, I used cpanm to install bignum: $ cpanm --sudo bignum This also pulled in some other modules. This is what I saw on my machine: $ cpanm --sudo bignum --> Working on bignum Fetching http://www.cpan.org/authors/id/P/PJ/PJACKLAM/bignum-0.67.tar.gz ... OK Configuring bignum-0.67 ... OK ==> Found dependencies: Math::BigRat, Math::BigInt --> Working on Math::BigRat Fetching http://www.cpan.org/authors/id/P/PJ/PJACKLAM/Math-BigInt-2.003002.tar.gz ... OK Configuring Math-BigInt-2.003002 ... OK Building and testing Math-BigInt-2.003002 ... OK Successfully installed Math-BigInt-2.003002 (upgraded from 0.2614) Building and testing bignum-0.67 ... OK Successfully installed bignum-0.67 (upgraded from 0.49) 2 distributions installed $ Please give this a try and let me know what your experiences are. If you still have an issue, I'll be happy to help to try to resolve it. -- You are receiving this mail because: You are on the CC list for the bug.
bug-binutils@gnu.org
https://sourceware.org/bugzilla/show_bug.cgi?id=31390 --- Comment #1 from Ruud van der Pas --- Thank you for the feedback. I'm sorry you ran into this. Technically, this is not a bug and this is why I have lowered the priority to P3. I have also labeled this as an enhancement, because it is a documentation issue. The Perl code that implements the "gprofng display html" functionality, needs the bigint module. Unfortunately, this is not part of the standard module set and needs to be installed separately before running this tool. This is not an ideal situation, but at the moment, the only thing I can do is to document this and give a recipe how to obtain and install this missing module. This is the reason I made it an enhancement. I will work on recipe for this and update this bugzilla once I have it. -- You are receiving this mail because: You are on the CC list for the bug.
bug-binutils@gnu.org
https://sourceware.org/bugzilla/show_bug.cgi?id=31390 Ruud van der Pas changed: What|Removed |Added Last reconfirmed||2024-02-21 Status|UNCONFIRMED |ASSIGNED Ever confirmed|0 |1 -- You are receiving this mail because: You are on the CC list for the bug.
bug-binutils@gnu.org
https://sourceware.org/bugzilla/show_bug.cgi?id=31390 Ruud van der Pas changed: What|Removed |Added Severity|normal |enhancement Priority|P2 |P3 CC||ruud.vanderpas at oracle dot com Assignee|vladimir.mezentsev at oracle dot c |ruud.vanderpas at oracle dot com |om | -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/31346] [gp-display-text] For functions and methods show the level in the call stack in the relevant views
https://sourceware.org/bugzilla/show_bug.cgi?id=31346 Ruud van der Pas changed: What|Removed |Added See Also||https://sourceware.org/bugz ||illa/show_bug.cgi?id=31347 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/31347] [gp-display-text] Support a filter to select call stacks for a target function at a specified level
https://sourceware.org/bugzilla/show_bug.cgi?id=31347 Ruud van der Pas changed: What|Removed |Added See Also||https://sourceware.org/bugz ||illa/show_bug.cgi?id=31346 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/31347] [gp-display-text] Support a filter to select call stacks for a target function at a specified level
https://sourceware.org/bugzilla/show_bug.cgi?id=31347 Ruud van der Pas changed: What|Removed |Added CC||ruud.vanderpas at oracle dot com Priority|P2 |P3 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/31347] New: [gp-display-text] Support a filter to select call stacks for a target function at a specified level
https://sourceware.org/bugzilla/show_bug.cgi?id=31347 Bug ID: 31347 Summary: [gp-display-text] Support a filter to select call stacks for a target function at a specified level Product: binutils Version: 2.42 Status: NEW Severity: enhancement Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- This is related to bugzilla #31346. Once the functionality requested in this RFE is available, it will be useful to have a filter that allows the user to show the call stacks for a target function at a specific level in the call stack. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/31346] [gp-display-text] For functions and methods show the level in the call stack in the relevant views
https://sourceware.org/bugzilla/show_bug.cgi?id=31346 --- Comment #1 from Ruud van der Pas --- This definitely sounds like a useful addition we will consider. We need to decide how to make this information available. This should also be in such a way that the gprofng tools that make use of gp-display-text, like the GUI, can easily access this information and make it available to the user. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/31346] [gp-display-text] For functions and methods show the level in the call stack in the relevant views
https://sourceware.org/bugzilla/show_bug.cgi?id=31346 Ruud van der Pas changed: What|Removed |Added CC||ruud.vanderpas at oracle dot com Priority|P2 |P3 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/31346] New: [gp-display-text] For functions and methods show the level in the call stack in the relevant views
https://sourceware.org/bugzilla/show_bug.cgi?id=31346 Bug ID: 31346 Summary: [gp-display-text] For functions and methods show the level in the call stack in the relevant views Product: binutils Version: 2.42 Status: NEW Severity: enhancement Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- This is a request from a Java development team: "In the Functions, Callers - Callees and CallTree view, when you click on a method name, or node to also show at what level or number in the depth of the stack trace that method or node is. The idea is to show the user is this stack 300 deep, and is the method / node I just clicked on number 153 of 300? Or more generally display in some means that value the numeric stack depth of the currently selected method or node, and maybe the total depth of that stack trace." -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30943] [display html] Do not generate disassembly by default
https://sourceware.org/bugzilla/show_bug.cgi?id=30943 Ruud van der Pas changed: What|Removed |Added Status|NEW |ASSIGNED --- Comment #2 from Ruud van der Pas --- I will investigate this further, but upfront I'm not sure this is a good approach to improve the performance. It will also complicate the code, since there won't be any links to the html files with the disassembly. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30942] [display html] Improve the performance
https://sourceware.org/bugzilla/show_bug.cgi?id=30942 Ruud van der Pas changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED --- Comment #4 from Ruud van der Pas --- I identified an opportunity to improve the performance in the part that generates the caller-callee information. This has a noticeable positive impact. Especially on applications with a larger number of functions. The fix is in binutils 2.42. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30439] [display html] Assertion error in caller-callee structure creation
https://sourceware.org/bugzilla/show_bug.cgi?id=30439 Ruud van der Pas changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED --- Comment #2 from Ruud van der Pas --- The fix is in binutils 2.42. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30438] [display html] Assertion error in function structure creation
https://sourceware.org/bugzilla/show_bug.cgi?id=30438 Ruud van der Pas changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED --- Comment #2 from Ruud van der Pas --- The fixe is in binutils 2.42. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/31169] [display text] Source code locations can not be found in a C++ application
https://sourceware.org/bugzilla/show_bug.cgi?id=31169 Ruud van der Pas changed: What|Removed |Added CC||ruud.vanderpas at oracle dot com Priority|P2 |P3 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/31169] New: [display text] Source code locations can not be found in a C++ application
https://sourceware.org/bugzilla/show_bug.cgi?id=31169 Bug ID: 31169 Summary: [display text] Source code locations can not be found in a C++ application Product: binutils Version: 2.41 Status: NEW Severity: normal Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- This is a C++ application developed at the University of Bristol. The developers run into the problem that "gprofng display text" cannot locate the source file(s), while the application and library objects have been compiled with -g. This problem seems to be related to interpreting the DWARF output. If -gdwarf-2 is added on the g++ compile lines, the source files can be located and source line level information is printed. If however, -gdwarf-3 is used, the locations of the source files cannot be found and consequently, no source line level information is printed. A test case exhibiting this problem has been made available. The program needs to be compiled with g++ 8.5 and that has been used in the tests. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30679] gprofng man pages improvements
https://sourceware.org/bugzilla/show_bug.cgi?id=30679 Ruud van der Pas changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED --- Comment #4 from Ruud van der Pas --- The updates are upstream now. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30895] gprofng is not relocatable
https://sourceware.org/bugzilla/show_bug.cgi?id=30895 --- Comment #2 from Ruud van der Pas --- The upcoming docs update has coverage of GPROFNG_SYSCONFDIR -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30895] gprofng is not relocatable
https://sourceware.org/bugzilla/show_bug.cgi?id=30895 Ruud van der Pas changed: What|Removed |Added Assignee|ruud.vanderpas at oracle dot com |vladimir.mezentsev at oracle dot c ||om -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30896] gprofng display text: please provide a way (or document how) to include the object name next to the called functions
https://sourceware.org/bugzilla/show_bug.cgi?id=30896 Ruud van der Pas changed: What|Removed |Added Priority|P2 |P3 CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30896] gprofng display text: please provide a way (or document how) to include the object name next to the called functions
https://sourceware.org/bugzilla/show_bug.cgi?id=30896 --- Comment #4 from Ruud van der Pas --- This is indeed confusing. The keyword "name" in a metrics definition is different than the "-name" option. As was pointed out in Bugzilla 30898, this is the correct way to include the object name: $ gprofng display text -name short:soname -metrics e.%totalcpu:name For example: $ gprofng display text -name short:soname -metrics e.%totalcpu:name -functions experiment.2.thr.er Current metrics: e.%totalcpu:name Current Sort Metric: Exclusive Total CPU Time ( e.%totalcpu ) Functions sorted by metric: Exclusive Total CPU Time Excl. TotalName CPU sec. % 1.861 100.00[] 1.771 95.16 mxv_core [] 0.050 2.69 drand48 [] 0.020 1.08 erand48_r [] 0.020 1.08 init_data [] 0. 0. @0x1de20 [] 0. 0. __libc_start_call_main [] 0. 0. __libc_start_main_impl [] 0. 0. main [] 0. 0. start_thread [] $ I will add a clarification and example to the documentation. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30898] consider multi-threading for gprofng display text
https://sourceware.org/bugzilla/show_bug.cgi?id=30898 Ruud van der Pas changed: What|Removed |Added CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30898] consider multi-threading for gprofng display text
https://sourceware.org/bugzilla/show_bug.cgi?id=30898 Ruud van der Pas changed: What|Removed |Added Priority|P2 |P3 Assignee|ruud.vanderpas at oracle dot com |vladimir.mezentsev at oracle dot c ||om --- Comment #4 from Ruud van der Pas --- I think there are several issues brought up in this RFE. It will be cleaner and easier to submit patches for, if we split this RFE into several others. This is why I have created two RFEs for "display html" and cross referenced them: 30942 and 30943. We can then keep this one for the "display text" related performance issues mentioned. I have also changed this RFE from P2 to P3, and re-assigned it to Vladimir, since he is the lead maintainer for "display text". -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30898] consider multi-threading for gprofng display text
https://sourceware.org/bugzilla/show_bug.cgi?id=30898 Ruud van der Pas changed: What|Removed |Added See Also||https://sourceware.org/bugz ||illa/show_bug.cgi?id=30943 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30943] [display html] Do not generate disassembly by default
https://sourceware.org/bugzilla/show_bug.cgi?id=30943 Ruud van der Pas changed: What|Removed |Added Assignee|vladimir.mezentsev at oracle dot c |ruud.vanderpas at oracle dot com |om | Priority|P2 |P3 CC||ruud.vanderpas at oracle dot com See Also||https://sourceware.org/bugz ||illa/show_bug.cgi?id=30898 --- Comment #1 from Ruud van der Pas --- I'm not sure that this is the right approach. Most performance analysts will want to look at the disassembly at some point. Instead, it will be better to explore opportunities to improve the performance. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30943] New: [display html] Do not generate disassembly by default
https://sourceware.org/bugzilla/show_bug.cgi?id=30943 Bug ID: 30943 Summary: [display html] Do not generate disassembly by default Product: binutils Version: 2.41 Status: NEW Severity: enhancement Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- In Bugzilla 30898, John Doe reports the following: "Maybe the default for "gprofng display html" should be to NOT include disassembly (this may be a separate issue though)?" -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30942] [display html] Improve the performance
https://sourceware.org/bugzilla/show_bug.cgi?id=30942 Ruud van der Pas changed: What|Removed |Added Status|NEW |ASSIGNED --- Comment #2 from Ruud van der Pas --- Apologies. The reference in the first comment should be to #30898. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30898] consider multi-threading for gprofng display text
https://sourceware.org/bugzilla/show_bug.cgi?id=30898 Ruud van der Pas changed: What|Removed |Added See Also||https://sourceware.org/bugz ||illa/show_bug.cgi?id=30942 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30942] [display html] Improve the performance
https://sourceware.org/bugzilla/show_bug.cgi?id=30942 Ruud van der Pas changed: What|Removed |Added See Also||https://sourceware.org/bugz ||illa/show_bug.cgi?id=30898 Assignee|vladimir.mezentsev at oracle dot c |ruud.vanderpas at oracle dot com |om | Priority|P2 |P3 CC||ruud.vanderpas at oracle dot com --- Comment #1 from Ruud van der Pas --- Since "display html" is written in Perl, the code may not be as efficient as "display text", which has been written in C/C++ and heavily optimized by the compiler. This does not mean there may not be room for source level performance improvement(s). This case needs to be investigated and where possible, improved. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30942] New: [display html] Improve the performance
https://sourceware.org/bugzilla/show_bug.cgi?id=30942 Bug ID: 30942 Summary: [display html] Improve the performance Product: binutils Version: 2.41 Status: NEW Severity: enhancement Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- In Bugzilla 30989, John Doe reports the following: "recording a sample program that uses lots of shared object files (most loaded with dlopen) using "gprofng collect app" is quite fast; doing the "gprofng display html" afterwards was... unexpected. The first process took several minutes and wrote hundreds of MBs (mostly disassembly) to disk /tmp/install/bin/gp-display-text -limit 100 -viewmode machine -compare off -script display.2.html/gp-script /tmp/binutils-2.41/build/test.2.er The second process ran even much longer (actually I've killt it after taking a single complete core for over 2 hours) perl /tmp/install/bin/gp-display-html --whoami=gprofng display html test.2.er Both processes ran on a single core, so the main question here is: is it possible to use multiple cores for those tasks?" -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29759] [display html] Support for aarch64 is needed
https://sourceware.org/bugzilla/show_bug.cgi?id=29759 Ruud van der Pas changed: What|Removed |Added Resolution|FIXED |--- Status|RESOLVED|REOPENED --- Comment #2 from Ruud van der Pas --- This one was accidentally closed and has to be reopened. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30679] gprofng man pages improvements
https://sourceware.org/bugzilla/show_bug.cgi?id=30679 --- Comment #2 from Ruud van der Pas --- Hi Matthias, Thanks very much for the feedback. Regarding the warnings, I have fixed this in my workspace for the docs. We're working on some other doc changes that will be too late for 2.41, but we plan to release a doc patch in the foreseeable future. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30679] gprofng man pages improvements
https://sourceware.org/bugzilla/show_bug.cgi?id=30679 Ruud van der Pas changed: What|Removed |Added CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30439] [display html] Assertion error in caller-callee structure creation
https://sourceware.org/bugzilla/show_bug.cgi?id=30439 Ruud van der Pas changed: What|Removed |Added Priority|P2 |P3 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30438] [display html] Assertion error in function structure creation
https://sourceware.org/bugzilla/show_bug.cgi?id=30438 Ruud van der Pas changed: What|Removed |Added Priority|P2 |P3 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29763] [docs] The user guide needs to be expanded
https://sourceware.org/bugzilla/show_bug.cgi?id=29763 Ruud van der Pas changed: What|Removed |Added Resolution|--- |FIXED Status|ASSIGNED|RESOLVED --- Comment #2 from Ruud van der Pas --- There has been a major overhaul in April. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30439] [display html] Assertion error in caller-callee structure creation
https://sourceware.org/bugzilla/show_bug.cgi?id=30439 Ruud van der Pas changed: What|Removed |Added Assignee|vladimir.mezentsev at oracle dot c |ruud.vanderpas at oracle dot com |om | Status|NEW |ASSIGNED CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30439] New: [display html] Assertion error in caller-callee structure creation
https://sourceware.org/bugzilla/show_bug.cgi?id=30439 Bug ID: 30439 Summary: [display html] Assertion error in caller-callee structure creation Product: binutils Version: 2.40 Status: NEW Severity: normal Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- Execution terminates with the following error: [Assertion error] generate_caller_callee - 6 elements in name exceeds limit What happens is that a line with the information for a specific function has additional fields that are not expected. This is the line triggering the problem: 5:0x0002bc0d *@0x2bc0d () -- no functions found 0. The string "-- no functions found" is not expected and this results in the assertion. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30438] [display html] Assertion error in function structure creation
https://sourceware.org/bugzilla/show_bug.cgi?id=30438 Ruud van der Pas changed: What|Removed |Added Status|NEW |ASSIGNED -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30438] [display html] Assertion error in function structure creation
https://sourceware.org/bugzilla/show_bug.cgi?id=30438 Ruud van der Pas changed: What|Removed |Added CC||ruud.vanderpas at oracle dot com Assignee|vladimir.mezentsev at oracle dot c |ruud.vanderpas at oracle dot com |om | -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30438] New: [display html] Assertion error in function structure creation
https://sourceware.org/bugzilla/show_bug.cgi?id=30438 Bug ID: 30438 Summary: [display html] Assertion error in function structure creation Product: binutils Version: 2.40 Status: NEW Severity: normal Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- For a specific experiment directory, the following error is reported by "gprofng display html": [Assertion error] get_function_info - unexpected: number of fields = 10 This error indicates that the input data is not as expected, resulting in an assertion error. As a result, the execution terminates prematurely. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29763] [docs] The user guide needs to be expanded
https://sourceware.org/bugzilla/show_bug.cgi?id=29763 --- Comment #1 from Ruud van der Pas --- The recent changes committed upstream are a big step forward. Many more options have been documented now. The index lists most of the commands and options, and the text of all man pages is included in the user guide. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29521] [docs] man pages are not in the release tarball
https://sourceware.org/bugzilla/show_bug.cgi?id=29521 Ruud van der Pas changed: What|Removed |Added Resolution|--- |FIXED Status|ASSIGNED|RESOLVED --- Comment #10 from Ruud van der Pas --- The dependence on help2man has been eliminated. All man pages are generated using Texinfo. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29759] [display html] Support for aarch64 is needed
https://sourceware.org/bugzilla/show_bug.cgi?id=29759 Ruud van der Pas changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED --- Comment #1 from Ruud van der Pas --- The dependence on help2man has been eliminated. All man pages are now generated using Texinfo. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29521] [docs] man pages are not in the release tarball
https://sourceware.org/bugzilla/show_bug.cgi?id=29521 --- Comment #9 from Ruud van der Pas --- I think this bug has been addressed and resolved. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30283] [display text] The metrics command should honor the order of the visibility specifications
https://sourceware.org/bugzilla/show_bug.cgi?id=30283 Ruud van der Pas changed: What|Removed |Added CC||ruud.vanderpas at oracle dot com Severity|normal |enhancement Priority|P2 |P3 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30283] New: [display text] The metrics command should honor the order of the visibility specifications
https://sourceware.org/bugzilla/show_bug.cgi?id=30283 Bug ID: 30283 Summary: [display text] The metrics command should honor the order of the visibility specifications Product: binutils Version: 2.40 Status: NEW Severity: normal Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- The metrics command is used to specify the metrics to be shown. The visibility part is used to select whether this should be the absolute number, and/or the percentage, etc. I would expect the order in the visibility specifications to be honored. For example, ".%" shows the percentage in the second column, but "%." should show the percentage first. As the example below demonstrates, this order is currently ignored. This should be changed. $ gprofng display text -limit 5 -metrics e.%totalcpu -functions exp.er Print limit set to 5 Current metrics: e.%totalcpu:name Current Sort Metric: Exclusive Total CPU Time ( e.%totalcpu ) Functions sorted by metric: Exclusive Total CPU Time Excl. TotalName CPU sec. % 2.422 100.00 2.302 95.04 mxv_core 0.070 2.89 init_data 0.020 0.83 erand48_r 0.010 0.41 __drand48_iterate $ gprofng display text -limit 5 -metrics e%.totalcpu -functions exp.er Print limit set to 5 Current metrics: e.%totalcpu:name Current Sort Metric: Exclusive Total CPU Time ( e.%totalcpu ) Functions sorted by metric: Exclusive Total CPU Time Excl. TotalName CPU sec. % 2.422 100.00 2.302 95.04 mxv_core 0.070 2.89 init_data 0.020 0.83 erand48_r 0.010 0.41 __drand48_iterate $ -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30225] New: [display text] The overview command displays info for the first experiment only
https://sourceware.org/bugzilla/show_bug.cgi?id=30225 Bug ID: 30225 Summary: [display text] The overview command displays info for the first experiment only Product: binutils Version: 2.40 Status: NEW Severity: normal Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- The "overview" command supported by "gprofng display text" prints a summary of the statistics for the experiment(s) specified. If multiple experiments are specified, only the results for the first experiment are shown though. The results for the other experiment(s) are missing. This is an example: $ gprofng display text -overview mxv.1.thr.er mxv.2.thr.er Experiment(s): Experiment :mxv.1.thr.er Target: '../bindir/mxv-pthreads -m 8000 -n 4000 -v -t 1' Host : ruud-arm-system (aarch64, Linux 5.15.0-5.76.5.1.el9uek.aarch64) Start Time: Thu Mar 9 06:33:23 2023 Duration : 7.124 Seconds Metrics: Experiment Duration (Seconds): [7.124] Clock Profiling [X]Total CPU Time - totalcpu (Seconds): [*0.183] Notes: '*' indicates hot metrics, '[X]' indicates currently enabled metrics. The metrics command can be used to change selections. The metric_list command lists all available metrics. $ As can be seen above, only the results for "mxv.1.thr.er" are shown. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30225] [display text] The overview command displays info for the first experiment only
https://sourceware.org/bugzilla/show_bug.cgi?id=30225 Ruud van der Pas changed: What|Removed |Added Priority|P2 |P3 CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29521] [docs] man pages are not in the release tarball
https://sourceware.org/bugzilla/show_bug.cgi?id=29521 --- Comment #7 from Ruud van der Pas --- I'm working on the fix for this and will use this opportunity to enhance the gprofng user guide with the contents of the man pages. In this way, users have access to a description of all the options without the need to install gprofng. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30195] [display text] Source code location can not be found
https://sourceware.org/bugzilla/show_bug.cgi?id=30195 Ruud van der Pas changed: What|Removed |Added CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30195] New: [display text] Source code location can not be found
https://sourceware.org/bugzilla/show_bug.cgi?id=30195 Bug ID: 30195 Summary: [display text] Source code location can not be found Product: binutils Version: 2.40 Status: NEW Severity: normal Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- As the log of my session shown below shows, gprofng cannot find the location of a source file, but the code has been compiled with -g. This error has been observed using gcc 11 on Oracle Linux 9. $ uname -r 5.15.0-5.76.5.1.el9uek.aarch64 $ $ gcc --version gcc (GCC) 11.3.1 20220421 (Red Hat 11.3.1-2.1.0.1) Copyright (C) 2021 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. $ make gcc -o ../objects/main.o -c -g -O -Wall -Werror=undef -Wstrict-prototypes main.c gcc -o ../objects/manage_data.o -c -g -O -Wall -Werror=undef -Wstrict-prototypes manage_data.c gcc -o ../objects/workload.o -c -g -O -Wall -Werror=undef -Wstrict-prototypes workload.c gcc -o ../objects/mxv.o -c -g -O -Wall -Werror=undef -Wstrict-prototypes mxv.c gcc -o ../bindir/mxv-pthreads ../objects/main.o ../objects/manage_data.o ../objects/workload.o ../objects/mxv.o -lm -lpthread ldd ../bindir/mxv-pthreads linux-vdso.so.1 (0x8a081000) libm.so.6 => /lib64/libm.so.6 (0x89fa3000) libc.so.6 => /lib64/libc.so.6 (0x89df5000) /lib/ld-linux-aarch64.so.1 (0x8a044000) $ cd ../bindir $ gprofng collect app ./mxv-pthreads -m 8000 -n 4000 CPU clock frequency: 2500 Creating experiment directory test.1.er (Process ID: 2101962) ... mxv: error check passed - rows = 8000 columns = 4000 threads = 1 $ gprofng display text -source mxv_core test.1.er Error: Source location not recorded in experiment $ -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30194] [collect app] Investigate occurrences of @PC from a shared library
https://sourceware.org/bugzilla/show_bug.cgi?id=30194 Ruud van der Pas changed: What|Removed |Added CC||ruud.vanderpas at oracle dot com Priority|P2 |P3 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30194] New: [collect app] Investigate occurrences of @PC from a shared library
https://sourceware.org/bugzilla/show_bug.cgi?id=30194 Bug ID: 30194 Summary: [collect app] Investigate occurrences of @PC from a shared library Product: binutils Version: 2.40 Status: NEW Severity: normal Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- While gprofng lists @PC for several functions in shared libraries, perf seems to be able to resolve the function name. This was reported by a user who has been using gprofng on Ubuntu. Below are the function views as shown by gprofng and perf. For example, __strncmp_sse42 shows up in the perf output, but gprofng reports this as @0x1711f1 (). It will be good to investigate where this difference comes from and whether gprofng can be improved to resolve the function name(s). This is what gprofng display text shows: Excl. Total Incl. TotalName CPU CPU sec. % sec. % 0.690 100.00 0.690 100.00 0.140 20.29 0.490 71.01 ada::parser::parse_url(std::basic_string_view >, ada::url const*, ada::encoding_type) 0.130 18.84 0.130 18.84 @0x1711f1 () 0.050 7.25 0.510 73.91 ada::parse(std::basic_string_view >, ada::url const*, ada::encoding_type) 0.050 7.25 0.650 94.20 void BasicBench_AdaURL(benchmark::State&) 0.040 5.80 0.040 5.80 ada::unicode::to_lower_ascii(char*, unsigned long) [clone .isra.0] 0.040 5.80 0.040 5.80 malloc 0.030 4.35 0.030 4.35 ioctl 0.030 4.35 0.070 10.14 operator new(unsigned long) 0.030 4.35 0.090 13.04 std::__cxx11::basic_string, std::allocator >::_M_append(char const*, unsigned long) 0.030 4.35 0.030 4.35 std::basic_string_view >::substr(unsigned long, unsigned long) const 0.020 2.90 0.020 2.90 @0x9dcd3 () 0.020 2.90 0.040 5.80 free 0.010 1.45 0.010 1.45 @0x7a30 () 0.010 1.45 0.010 1.45 @0x9e160 () 0.010 1.45 0.010 1.45 ada::checkers::has_hex_prefix(std::basic_string_view >) 0.010 1.45 0.010 1.45 ada::url::~url() 0.010 1.45 0.010 1.45 bool std::all_of(char const*, char const*, bool (*)(char) noexcept) ... This is what perf reports (perf record/perf report): Samples: 28K of event 'cycles:u', Event count (approx.): 23444784147 Overhead CommandShared Object Symbol 29.08% bbc_bench bbc_bench [.] ada::parser::parse_url 10.71% bbc_bench libc.so.6 [.] __strncmp_sse42 9.16% bbc_bench bbc_bench [.] BasicBench_AdaURL 5.69% bbc_bench libc.so.6 [.] wcstold_l_internal 5.55% bbc_bench libc.so.6 [.] __strncasecmp_l_sse42 3.66% bbc_bench bbc_bench [.] ada::parse 3.62% bbc_bench libstdc++.so.6.0.30 [.] std::__cxx11::basic_string, std::allocator >::push_back 3.31% bbc_bench bbc_bench [.] std::basic_string_view >::substr 3.00% bbc_bench libstdc++.so.6.0.30 [.] std::__cxx11::basic_string, std::allocator >::_M_mutate 2.83% bbc_bench bbc_bench [.] ada::unicode::to_lower_ascii 2.79% bbc_bench libc.so.6 [.] wcstod_l_internal 2.64% bbc_bench libstdc++.so.6.0.30 [.] std::__cxx11::basic_string, std::allocator >::_M_append 2.59% bbc_bench bbc_bench [.] std::basic_string_view >::find 2.31% bbc_bench bbc_bench [.] ada::url::~url 2.26% bbc_bench libstdc++.so.6.0.30 [.] std::__cxx11::basic_string, std::allocator >::_M_create 1.05% bbc_bench libstdc++.so.6.0.30 [.] operator new 0.91% bbc_bench bbc_bench [.] standard_url::~standard_url -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30089] New: [display text] Incorrect thread count shown on aarch64
https://sourceware.org/bugzilla/show_bug.cgi?id=30089 Bug ID: 30089 Summary: [display text] Incorrect thread count shown on aarch64 Product: binutils Version: 2.40 Status: NEW Severity: normal Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- As the sample output below shows, the number of threads shown with "gprofng display text" is incorrect on aarch64. This tested was conducted on an Oracle Linux 9 ("OL9") aarch64 system with multiple cores. The sample output below shows that the performance improves when increasing the number of threads from 1 to 2. Then the gprofng experiment is conducted using 2 threads. The printed elapsed time of about 0.6 seconds confirms that both threads are used in the computation. But the output of "gprofng display text" shows the information for one thread only. LOG of the session = $ cat run.res + exe=../bindir/mxv-pthreads + ../bindir/mxv-pthreads -t 1 -v Verbose mode enabled Allocated data structures Timings for the initialization phase elapsed time init_data = 0.066946 cpu time init_data = 0.066852 system time init_data = 0.008289 Initialized matrix and vectors Defined workload distribution Assigned work to threads Thread 0 has been created Timings for the computational phase elapsed time mxv = 1.253865 cpu time mxv = 1.251356 system time mxv = 0.00 Matrix vector multiplication has completed Verify correctness of result Error check passed mxv: error check passed - rows = 2000 columns = 3000 threads = 1 + ../bindir/mxv-pthreads -t 2 -v Verbose mode enabled Allocated data structures Timings for the initialization phase elapsed time init_data = 0.067124 cpu time init_data = 0.067025 system time init_data = 0.004770 Initialized matrix and vectors Defined workload distribution Assigned work to threads Thread 0 has been created Thread 1 has been created Timings for the computational phase elapsed time mxv = 0.613867 cpu time mxv = 1.222008 system time mxv = 0.003657 Matrix vector multiplication has completed Verify correctness of result Error check passed mxv: error check passed - rows = 2000 columns = 3000 threads = 2 + gprofng --version GNU gprofng binutils version 2.40.50 Copyright (C) 2023 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <https://gnu.org/licenses/gpl.html>. This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. + gprofng collect app -O mxv.2.thr.er ../bindir/mxv-pthreads -t 2 -v CPU clock frequency: 2500 Creating experiment directory mxv.2.thr.er (Process ID: 1147696) ... Verbose mode enabled Allocated data structures Timings for the initialization phase elapsed time init_data = 0.067561 cpu time init_data = 0.067397 system time init_data = 0.014307 Initialized matrix and vectors Defined workload distribution Assigned work to threads Thread 0 has been created Thread 1 has been created Timings for the computational phase elapsed time mxv = 0.618479 cpu time mxv = 1.230675 system time mxv = 0.00 Matrix vector multiplication has completed Verify correctness of result Error check passed mxv: error check passed - rows = 2000 columns = 3000 threads = 2 + gprofng display text -threads ./mxv.2.thr.er Objects sorted by metric: Exclusive Total CPU Time Excl. TotalName CPU sec. % 0.060 100.00 0.060 100.00 Process 1, Thread 1 $ -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30089] [display text] Incorrect thread count shown on aarch64
https://sourceware.org/bugzilla/show_bug.cgi?id=30089 Ruud van der Pas changed: What|Removed |Added CC||ruud.vanderpas at oracle dot com Priority|P2 |P3 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30088] [collect app] CPU frequency is inadvertently printed in the help output on aarch64
https://sourceware.org/bugzilla/show_bug.cgi?id=30088 Ruud van der Pas changed: What|Removed |Added Priority|P2 |P3 Severity|minor |normal CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/30088] New: [collect app] CPU frequency is inadvertently printed in the help output on aarch64
https://sourceware.org/bugzilla/show_bug.cgi?id=30088 Bug ID: 30088 Summary: [collect app] CPU frequency is inadvertently printed in the help output on aarch64 Product: binutils Version: 2.40 Status: NEW Severity: minor Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- On an aarch64 system, the output from --help includes the CPU frequency: $ gprofng collect app --help CPU clock frequency: 2500 Usage: gprofng collect app [OPTION(S)] TARGET [TARGET_ARGUMENTS] Collect performance data on the target program. In addition to Program Counter PC) sampling, hardware event counters and various tracing options are supported. This first line should not be there. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29521] [docs] man pages are not in the release tarball
https://sourceware.org/bugzilla/show_bug.cgi?id=29521 --- Comment #5 from Ruud van der Pas --- To further clarify this patch, we now use special controls in the gprofng.texi user guide file to generate the gprofng.1 man page. This eliminates the dependence on help2man to create this page. The other four man pages still rely on help2man, but in case help2man has not been installed, this step will be skipped. So no more creation of empty files. In the next step to resolve this issue, we plan to provide separate Texinfo files for the 4 tools and remove the usage of help2man from the gprofng build and installation process. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29521] [docs] man pages are not in the release tarball
https://sourceware.org/bugzilla/show_bug.cgi?id=29521 --- Comment #3 from Ruud van der Pas --- Unfortunately we did not get to address this before 2.40 was released, but we are actively looking into a solution. We plan to either embed the man page texts in .texi files (or perhaps a single file), or we see if help2man can be endorsed by binutils. The latter will not be in time for 2.40 though and perhaps an interim approach is needed for this situation. Either way, we want to resolve this asap now. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29759] [display html] Support for aarch64 is needed
https://sourceware.org/bugzilla/show_bug.cgi?id=29759 Ruud van der Pas changed: What|Removed |Added Status|NEW |ASSIGNED -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29763] [docs] The user guide needs to be expanded
https://sourceware.org/bugzilla/show_bug.cgi?id=29763 Ruud van der Pas changed: What|Removed |Added Status|NEW |ASSIGNED -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29763] [docs] The user guide needs to be expanded
https://sourceware.org/bugzilla/show_bug.cgi?id=29763 Ruud van der Pas changed: What|Removed |Added CC||ruud.vanderpas at oracle dot com Assignee|vladimir.mezentsev at oracle dot c |ruud.vanderpas at oracle dot com |om | Priority|P2 |P3 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29763] New: [docs] The user guide needs to be expanded
https://sourceware.org/bugzilla/show_bug.cgi?id=29763 Bug ID: 29763 Summary: [docs] The user guide needs to be expanded Product: binutils Version: 2.39 Status: NEW Severity: enhancement Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- The current version of the user guide gets people going, but lacks the details needed. All the gprofng tools need to be documented. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29760] [collect app] Support profiling an already running application
https://sourceware.org/bugzilla/show_bug.cgi?id=29760 Ruud van der Pas changed: What|Removed |Added CC||ruud.vanderpas at oracle dot com Priority|P2 |P3 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29760] New: [collect app] Support profiling an already running application
https://sourceware.org/bugzilla/show_bug.cgi?id=29760 Bug ID: 29760 Summary: [collect app] Support profiling an already running application Product: binutils Version: 2.39 Status: NEW Severity: enhancement Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- Currently, gprofng collect app only provides meaningful results if the application is started under its control. An important range of applications is always running though. Support for these is important to have. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29759] [display html] Support for aarch64 is needed
https://sourceware.org/bugzilla/show_bug.cgi?id=29759 Ruud van der Pas changed: What|Removed |Added Assignee|vladimir.mezentsev at oracle dot c |ruud.vanderpas at oracle dot com |om | CC||ruud.vanderpas at oracle dot com Priority|P2 |P3 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29759] New: [display html] Support for aarch64 is needed
https://sourceware.org/bugzilla/show_bug.cgi?id=29759 Bug ID: 29759 Summary: [display html] Support for aarch64 is needed Product: binutils Version: 2.39 Status: NEW Severity: enhancement Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- Currently, gprofng display html only supports x86_64. There should be at least some basic support for aarch64 as well. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29758] [collect app] Support hardware event counters on recent processors
https://sourceware.org/bugzilla/show_bug.cgi?id=29758 Ruud van der Pas changed: What|Removed |Added Priority|P2 |P3 CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29758] New: [collect app] Support hardware event counters on recent processors
https://sourceware.org/bugzilla/show_bug.cgi?id=29758 Bug ID: 29758 Summary: [collect app] Support hardware event counters on recent processors Product: binutils Version: 2.39 Status: NEW Severity: enhancement Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- The support for hardware event counters has fallen behind and needs to be updated to handle more recent processors from Intel, AMD, and Arm. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29700] [display collect] Issue a warning if callstack frames are no longer captured
https://sourceware.org/bugzilla/show_bug.cgi?id=29700 Ruud van der Pas changed: What|Removed |Added Priority|P2 |P3 CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29700] [collect app] Issue a warning if callstack frames are no longer captured
https://sourceware.org/bugzilla/show_bug.cgi?id=29700 Ruud van der Pas changed: What|Removed |Added Summary|collect app] Issue a|[collect app] Issue a |warning if callstack frames |warning if callstack frames |are no longer captured |are no longer captured -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29701] [gprofng] Streamline the gprofng supported environment variables
https://sourceware.org/bugzilla/show_bug.cgi?id=29701 Ruud van der Pas changed: What|Removed |Added Priority|P2 |P3 Summary|Streamline the gprofng |[gprofng] Streamline the |supported environment |gprofng supported |variables |environment variables CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29700] collect app] Issue a warning if callstack frames are no longer captured
https://sourceware.org/bugzilla/show_bug.cgi?id=29700 Ruud van der Pas changed: What|Removed |Added Summary|[display collect] Issue a |collect app] Issue a |warning if callstack frames |warning if callstack frames |are no longer captured |are no longer captured -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29701] New: Streamline the gprofng supported environment variables
https://sourceware.org/bugzilla/show_bug.cgi?id=29701 Bug ID: 29701 Summary: Streamline the gprofng supported environment variables Product: binutils Version: 2.39 Status: NEW Severity: enhancement Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- Currently, various components of gprofng use environment variables. The names are inconsistent though and also not always sufficiently descriptive. The proposal is to have all these environment variables start with GP_ and reconsider the names. These variables should also be documented. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29700] New: [display collect] Issue a warning if callstack frames are no longer captured
https://sourceware.org/bugzilla/show_bug.cgi?id=29700 Bug ID: 29700 Summary: [display collect] Issue a warning if callstack frames are no longer captured Product: binutils Version: 2.39 Status: NEW Severity: enhancement Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- The data collection engine uses a buffer to store the call stack data. Once the buffer is full, callstack frames are no longer captured. The default limit is 256 both for C/C++ and Java. If frames have been dropped, this is shown in the output from "gprofng display text" (), but it will be helpful to also print a diagnostic run-time message in case this has occurred. This message should also explain how to increase the buffer size by changing the value through environment variables GPROFNG_MAX_CALL_STACK_DEPTH (C/C++) or GPROFNG_JAVA_MAX_CALL_STACK_DEPTH (Java). The complication is that this warning should not be sent to stdout/stderr per-se, or the user needs to have a way to disable these, because perhaps this behaviour is known and desired. This is why it might be better to include such a warning in a log file. An added benefit is that other tools (e.g. display html) may pick this up. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29633] [display text] Support derived metrics
https://sourceware.org/bugzilla/show_bug.cgi?id=29633 Ruud van der Pas changed: What|Removed |Added Priority|P2 |P3 CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29631] [display text] Incorrect percentages shown in the source and disassembly output
https://sourceware.org/bugzilla/show_bug.cgi?id=29631 Ruud van der Pas changed: What|Removed |Added CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29632] [display text] Support units for the metrics
https://sourceware.org/bugzilla/show_bug.cgi?id=29632 Ruud van der Pas changed: What|Removed |Added CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29632] [display text] Support units for the metrics
https://sourceware.org/bugzilla/show_bug.cgi?id=29632 Ruud van der Pas changed: What|Removed |Added Priority|P2 |P3 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29633] New: [display text] Support derived metrics
https://sourceware.org/bugzilla/show_bug.cgi?id=29633 Bug ID: 29633 Summary: [display text] Support derived metrics Product: binutils Version: 2.39 Status: NEW Severity: enhancement Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- Very often, derived metrics provide more insight. For example, to normalize certain events against the number of instructions executed. Similar to what is already done for the IPC and CPI derived metrics, the user should have the possibility to include the division of two columns in the output. Multiple of such derived metrics should be supported. This brings up a UI issue how the user can pass this on. Possibly by expanding the metric definition with the two metrics to be divided. For example e.cyclesDIVinsts. By using an operator, like DIV here, we can also support other basic arithmetic operators. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29632] New: [display text] Support units for the metrics
https://sourceware.org/bugzilla/show_bug.cgi?id=29632 Bug ID: 29632 Summary: [display text] Support units for the metrics Product: binutils Version: 2.39 Status: NEW Severity: enhancement Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- The numbers shown with display text can be quite large. For example when measuring hardware events like cache misses, or instructions executed. To make the analysis easier, it will be helpful to use units, like K, M, G, and T. For example "123 K instructions executed". Ideally, the user can specify this through an option. By default, the current formatting is used. This is to avoid that layered tools may otherwise break. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29631] New: [display text] Incorrect percentages shown in the source and disassembly output
https://sourceware.org/bugzilla/show_bug.cgi?id=29631 Bug ID: 29631 Summary: [display text] Incorrect percentages shown in the source and disassembly output Product: binutils Version: 2.39 Status: NEW Severity: normal Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- Created attachment 14366 --> https://sourceware.org/bugzilla/attachment.cgi?id=14366&action=edit The experiment directory used to produce the output given The source and disassembly output gives a value of 0. for the percentages of all the metrics. Attached is an experiment directory that was used to verify this issue. I also include the output below. The following command was used to generate this output: $ gprofng display text -source mxv_core -dis mxv_core test.1.er = Source file: /home/opc/00.dev/experiments/mxv/src/mxv.c Object file: mxv-pthreads.exe (found as test.1.er/archives/mxv-pthreads.exe__wAocSNXHc8) Load Object: mxv-pthreads.exe (found as test.1.er/archives/mxv-pthreads.exe__wAocSNXHc8) Excl. Total Incl. Total CPU CPU sec. %sec. % 1. #include "mydefs.h" 2. 0. 0. 0. 0. 3. void *driver_mxv(void *thread_arguments) 0. 0. 0. 0. 4. { 0. 0. 0. 0. 5.thread_data *local_data; 6. 0. 0. 0. 0. 7.local_data = (thread_data *) thread_arguments; 8. 0. 0. 0. 0. 9.bool do_work = local_data->do_work; 0. 0. 0. 0.10.int64_t repeat_count= local_data->repeat_count; 0. 0. 0. 0.11.uint64_t row_index_start = local_data->row_index_start; 0. 0. 0. 0.12.uint64_t row_index_end = local_data->row_index_end; 0. 0. 0. 0.13.uint64_t m = local_data->m; 0. 0. 0. 0.14.uint64_t n = local_data->n; 0. 0. 0. 0.15.double *b = local_data->b; 0. 0. 0. 0.16.double *c = local_data->c; 0. 0. 0. 0.17.double **A = local_data->A; 18. 0. 0. 0. 0.19.if (do_work) { 0. 0. 0. 0.20. for (int64_t r=0; r 0. 0. 0. 0.32. void __attribute__ ((noinline)) mxv_core(uint64_t row_index_start, uint64_t row_index_end, uint64_t m, uint64_t n, double **restrict A, double *restrict b, double *restrict c) 0. 0. 0. 0.33. { 0. 0. 0. 0.34.for (uint64_t i=row_index_start; i<=row_index_end; i++) { 0. 0. 0. 0.35.double row_sum = 0.0; ## 5.904 0. 5.904 0.36.for (int64_t j=0; jdo_work; 10.int64_t repeat_count= local_data->repeat_count; 11.uint64_t row_index_start = local_data->row_index_start; 12.uint64_t row_index_end = local_data->row_index_end; 13.uint64_t m = local_data->m; 14.uint64_t n = local_data->n; 15.double *b = local_data->b; 16.double *c = local_data->c; 17.double **A = local_data->A; 18. 19.if (do_work) { 20. for (int64_t r=0; r 0. 0.0. 0. [33] 4021ae: mov0x8(%rsp),%r10 34.for (uint64_t i=row_index_start; i<=row_index_end; i++) { 0. 0.0. 0. [34] 4021b3: cmp%rsi,%rdi 0. 0.0. 0. [34] 4021b6: jbe0x37 0. 0.0. 0. [34] 4021b8: ret 35.double row_sum = 0.0; 36.for (int64_t j=0; j
[Bug gprofng/29521] man pages are not in the release tarball
https://sourceware.org/bugzilla/show_bug.cgi?id=29521 --- Comment #1 from Ruud van der Pas --- Thanks for the feedback. This was definitely not intentional and will be addressed as soon as possible. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29521] [docs] man pages are not in the release tarball
https://sourceware.org/bugzilla/show_bug.cgi?id=29521 Ruud van der Pas changed: What|Removed |Added Summary|man pages are not in the|[docs] man pages are not in |release tarball |the release tarball -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29465] [docs] File version.texi is created in the binutils source directory
https://sourceware.org/bugzilla/show_bug.cgi?id=29465 --- Comment #1 from Ruud van der Pas --- After exploring several alternatives, it was decided to no longer dynamically generate file version.texi. Instead, the information in version.texi will be updated as part of a patch for the documentation. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29465] [docs] File version.texi is created in the binutils source directory
https://sourceware.org/bugzilla/show_bug.cgi?id=29465 Ruud van der Pas changed: What|Removed |Added Summary|[gprofng doc] File |[docs] File version.texi is |version.texi is created in |created in the binutils |the binutils source |source directory |directory | CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29470] [test suite] The test suite should be made more flexible
https://sourceware.org/bugzilla/show_bug.cgi?id=29470 Ruud van der Pas changed: What|Removed |Added CC||ruud.vanderpas at oracle dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29470] New: [test suite] The test suite should be made more flexible
https://sourceware.org/bugzilla/show_bug.cgi?id=29470 Bug ID: 29470 Summary: [test suite] The test suite should be made more flexible Product: binutils Version: 2.40 (HEAD) Status: NEW Severity: normal Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: ruud.vanderpas at oracle dot com Target Milestone: --- Currently, the gprofng test suite runs for too long. This is not in line with other tests in binutils. The set up, and perhaps also the information shown, should be made more flexible. One proposal to make the duration and/or number of tests, user selectable. Or to introduce two different commands. The default and something like "check-expensive", or "check-extensive" for a more thorough series of tests. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29465] [gprofng doc] File version.texi is created in the binutils source directory
https://sourceware.org/bugzilla/show_bug.cgi?id=29465 Ruud van der Pas changed: What|Removed |Added Summary|[display html] File |[gprofng doc] File |version.texi is created in |version.texi is created in |the binutils source |the binutils source |directory |directory -- You are receiving this mail because: You are on the CC list for the bug.