subject:"\[valgrind\] \[Bug 433439\] FreeBSD support, part 3"

[valgrind] [Bug 433439] FreeBSD support, part 3

2021-10-06 Thread Paul Floyd

https://bugs.kde.org/show_bug.cgi?id=433439 Paul Floyd changed: What|Removed |Added Resolution|--- |FIXED Status|REPORTED

[valgrind] [Bug 433439] FreeBSD support, part 3

2021-10-05 Thread Paul Floyd

https://bugs.kde.org/show_bug.cgi?id=433439 Paul Floyd changed: What|Removed |Added Attachment #141834|0 |1 is obsolete|

[valgrind] [Bug 433439] FreeBSD support, part 3

2021-09-27 Thread Paul Floyd

https://bugs.kde.org/show_bug.cgi?id=433439 --- Comment #6 from Paul Floyd --- For sysctl the problem is that I created separate patches for include and coregrind (patch 12) The definition is Int VG_(sysctl)(Int *name, UInt namelen, void *oldp, SizeT *oldlenp, const void *newp, SizeT newlen)

[valgrind] [Bug 433439] FreeBSD support, part 3

2021-09-26 Thread Mark Wielaard

https://bugs.kde.org/show_bug.cgi?id=433439 --- Comment #5 from Mark Wielaard --- (In reply to Paul Floyd from comment #4) > Why do you think making sysctl newp const might be unsafe? > > The signature on FreeBSD is > > int > sysctl(const int *name, u_int namelen, void *oldp, size_t

[valgrind] [Bug 433439] FreeBSD support, part 3

2021-09-26 Thread Paul Floyd

https://bugs.kde.org/show_bug.cgi?id=433439 --- Comment #4 from Paul Floyd --- Why do you think making sysctl newp const might be unsafe? The signature on FreeBSD is int sysctl(const int *name, u_int namelen, void *oldp, size_t *oldlenp, const void *newp, size_t newlen); On

[valgrind] [Bug 433439] FreeBSD support, part 3

2021-09-26 Thread Mark Wielaard

https://bugs.kde.org/show_bug.cgi?id=433439 --- Comment #3 from Mark Wielaard --- Created attachment 141932 --> https://bugs.kde.org/attachment.cgi?id=141932=edit freebsd3 patch This variant of the patch, with the changes suggested in my previous comment, works for me on fedora x86_64. --