Paul wrote (but it got bounced): On Fri, 16 Aug 2019, Andrew Cagney wrote:
> Given the code is isolated and (hopeful) treated read-only - itt > should still build on old systems. However, I struggle to see a > reason for building it for Fedora 30, and would again like to disable > it. I'm okay with that. > (this shouldn't be confused with the klips userland code which can > continue to build). For now that's good. Once we release 3.30, provided it has KLIPS and XFRMi support (KLIPS for linux < 4.x) then in 3.31 we will completely remove KLIPS. That gives us one release that supports both, that people can use for testing/migration. Paul Perhaps we should create a klips git repo (cloned from libreswan then pared down to just the klips kernel module), perhaps. I've pushed the below commit 2bad6cc6ee877a5cc8d4691a3bb1985f6ba1f42e (HEAD -> master) Author: Andrew Cagney <[email protected]> Date: Tue Sep 3 13:58:05 2019 -0400 testing: only build/install the KLIPS kernel module when KVM_KLIPS_MODULE=true (default is false) Not to be confused with {KVM_,}USE_KLIPS which enables support for KLIPS in pluto and still defaults to true. On Fri, 16 Aug 2019 at 18:15, Andrew Cagney <[email protected]> wrote: > If you look through a recent test result on > https://testing.libreswan.org/ you'll notice KERNEL errors - these are > from the KLIPS module panicking. This has problem seemingly been > known for some time but forgotten - KLIPS will panic on >=4.x kernels > (at least those shipped by fedora). > > Given the code is isolated and (hopeful) treated read-only - itt > should still build on old systems. However, I struggle to see a > reason for building it for Fedora 30, and would again like to disable > it. > > (this shouldn't be confused with the klips userland code which can > continue to build). > > Andrew >
_______________________________________________ Swan-dev mailing list [email protected] https://lists.libreswan.org/mailman/listinfo/swan-dev
