search for: gperftools

Hey guys, I am currently working on a project that uses JIT compilation to compile incoming user requests to native code. Are there some best practises related to profiling the generated code? My project uses gperftools pprof for profiling etc. Is there a way to hook the two up? Are there any other profiling method that works? This page describes how to debug JIT code with GDB. I wonder if something similar could be done for gperftools/pprof? Regards, -- Priyendra -------------- next part -------------- An HTML a...

...03, 2013 3:11 AM To: llvmdev at cs.uiuc.edu Subject: [LLVMdev] Profiling LLVM JIT code Hey guys, I am currently working on a project that uses JIT compilation to compile incoming user requests to native code. Are there some best practises related to profiling the generated code? My project uses gperftools pprof for profiling etc. Is there a way to hook the two up? Are there any other profiling method that works? This page describes how to debug JIT code with GDB. I wonder if something similar could be done for gperftools/pprof? Regards, -- Priyendra -------------- next part -------------- An HTML...

...> ** ** > > Hey guys,**** > > ** ** > > I am currently working on a project that uses JIT compilation to compile > incoming user requests to native code. Are there some best practises > related to profiling the generated code?**** > > ** ** > > My project uses gperftools pprof for profiling etc. Is there a way to hook > the two up? Are there any other profiling method that works? This page > describes how to debug JIT code with GDB. I wonder if something similar > could be done for gperftools/pprof?**** > > ** ** > > Regards,**** > > -- P...

...2020 at 11:12:39PM +0000, Eric Wong wrote: > Olly Betts <olly at survex.com> wrote: > > If most (or at least a significant amount) is CPU time then it would > > be useful to profile to see if there are any low-hanging fruit. I've > > been mostly using the profiler in gperftools lately if you want to > > try this and don't know what to use. > > Haven't tried gperftools, but I'm somewhat familiar with Linux `perf`. > Just installed the -dbgsym packages from Debian, attached is > "perf record ... && perf report" output. The r...

...#39;re much less, then it's spending a lot of time waiting for I/O, which in this case means loading files from disk. If most (or at least a significant amount) is CPU time then it would be useful to profile to see if there are any low-hanging fruit. I've been mostly using the profiler in gperftools lately if you want to try this and don't know what to use. It would also be interesting to compare with xapian git master (if you're not using that already - you don't seem to mention a version). The handling of shards has changed in some possibly significant ways. > Perhaps creati...

...ocid assignment and per-shard gaps has anything to do with the slowdowns. Works great with 2..8 shards... > If most (or at least a significant amount) is CPU time then it would > be useful to profile to see if there are any low-hanging fruit. I've > been mostly using the profiler in gperftools lately if you want to > try this and don't know what to use. Haven't tried gperftools, but I'm somewhat familiar with Linux `perf`. Just installed the -dbgsym packages from Debian, attached is "perf record ... && perf report" output. > It would also be interest...

Going back to the "prioritizing aggregated DBs" thread from February 2020, I've got 390 Xapian shards for 130 public inboxes I want to search against(*). There's more on the horizon (we're expecting tens of thousands of public inboxes). After bumping RLIMIT_NOFILE and running ->add_database a bunch, the actual queries seem to be taking ~30s (not good :x). Now I'm

[ https://www.centos.org/search.php?query=atomicorp&mid=30&action=showall&andor=AND forum search] returns a 404. Can the forum search https://www.centos.org/forums/search.php? be used with parameters that will provide the supporting material for the warning "Many CentOS users have had problems after enabling this repo"? -------------- next part -------------- An HTML

Thanks for the suggestion James, it reduces the commit by about ~900 MB (14,9 GB -> 14 GB). Unfortunately it does not solve the performance problem. The heap is global to the application and thread-safe, so every malloc/free locks it, which evidently doesn’t scale. We could manually create thread-local heaps, but I didn’t want to go there. Ultimately allocated blocks need to share ownership

See this bug for background information: https://bugzilla.redhat.com/show_bug.cgi?id=1350465 Thanks: Tomáš Golembiovský --- v2v/virt-v2v.pod | 38 ++++++++++++++++++++++++++------------ 1 file changed, 26 insertions(+), 12 deletions(-) diff --git a/v2v/virt-v2v.pod b/v2v/virt-v2v.pod index e68d75cf8..0943bf305 100644 --- a/v2v/virt-v2v.pod +++ b/v2v/virt-v2v.pod @@ -1909,18 +1909,32 @@ that

See this bug for background information: https://bugzilla.redhat.com/show_bug.cgi?id=1350465 --- v2v/virt-v2v.pod | 39 +++++++++++++++++++++++++++------------ 1 file changed, 27 insertions(+), 12 deletions(-) diff --git a/v2v/virt-v2v.pod b/v2v/virt-v2v.pod index e68d75cf8..93d1a9ecd 100644 --- a/v2v/virt-v2v.pod +++ b/v2v/virt-v2v.pod @@ -1909,18 +1909,33 @@ that guest through the RHV-M UI,

If the connection is not local, paths of disks will refer to the remote host, which were mistakenly handled as local paths (in the best case failing to open a non-existing disk, and in the worst case opening a different disk!). In case the disks are remote resources like ssh or ceph, nothing guarantees that the hostname can be reached from the local machine, or even that it is actually the same on

Hopefully this time ...