On Thu, 2018-12-13 at 19:27 +0000, Nadav Amit wrote: > > On Dec 13, 2018, at 11:02 AM, Edgecombe, Rick P <rick.p.edgecombe@xxxxxxxxx> > > wrote: > > > > On Wed, 2018-12-12 at 23:40 +0000, Nadav Amit wrote: > > > > On Dec 11, 2018, at 4:03 PM, Rick Edgecombe <rick.p.edgecombe@xxxxxxxxx> > > > > wrote: > > > > > > > > Add new flags for handling freeing of special permissioned memory in > > > > vmalloc, > > > > and remove places where the handling was done in module.c. > > > > > > > > This will enable this flag for all architectures. > > > > > > > > Signed-off-by: Rick Edgecombe <rick.p.edgecombe@xxxxxxxxx> > > > > --- > > > > kernel/module.c | 43 ++++++++++++------------------------------- > > > > 1 file changed, 12 insertions(+), 31 deletions(-) > > > > > > I count on you for merging your patch-set with mine, since clearly they > > > conflict. > > > > Yes, I can rebase on top of yours if you omit the changes around > > module_memfree > > for your next version. It should fit together pretty cleanly for BPF and > > modules > > I think. Not sure what you are planning for kprobes and ftrace. > > Are you asking after looking at the latest version of my patch-set? > > Kprobes is done and ack'd. ftrace needs to be broken into two separate > changes (setting x after writing, and using text_poke interfaces), unless > Steven ack’s them. The changes introduce some overhead (3x), but I think it > is a reasonable slowdown for a debug feature. > > Can you have a look at the series I’ve sent and let me know which patches > to drop? It would be best (for me) if the two series are fully merged. Looking at v7, the only extra thing could be the tweaks to the existing lines in kprobes free_insn_page be omitted, since those lines would just be removed in my later changes.
