Re: [PATCH 01/15] backports: handle RHEL 7.6 kernel

On Sun, 2020-02-23 at 23:24 +0100, Hauke Mehrtens wrote:

> > +++ b/backport/backport-include/linux/time.h
> > @@ -1,5 +1,11 @@
> >  #ifndef __BACKPORT_LINUX_TIME_H
> >  #define __BACKPORT_LINUX_TIME_H
> > +
> > +#if RHEL_RELEASE_CODE >= RHEL_RELEASE_VERSION(7,6)
> > +#include_next <linux/time64.h>
> > +#include <uapi/linux/time.h>
> > +#endif
> > +
> 
> Something is strange with the time header file handling.

Too many changes in this area, I guess ;-)

(I'd appreciate if you trimmed your quoting a bit, it's hard to find
your responses in the pages-long text)

> > +EXPORT_SYMBOL_GPL(get_user_pages_unlocked);
> > +#endif
> > +
> 
> Please check if this is needed at all.

I guess we can

> > -static struct scatterlist *scatterwalk_ffwd(struct scatterlist dst[2],
> > +#if RHEL_RELEASE_CODE < RHEL_RELEASE_VERSION(7,6)
> > + static struct scatterlist *scatterwalk_ffwd(struct scatterlist dst[2],
> 
> Why  did you add a space?

Heh.

> > @@ -441,3 +450,4 @@ int backport_genlmsg_multicast_allns(const struct genl_family *family,
> >  	return genlmsg_mcast(skb, portid, group, flags);
> >  }
> >  EXPORT_SYMBOL_GPL(backport_genlmsg_multicast_allns);
> > +#endif /* RHEL_RELEASE_CODE < RHEL_RELEASE_VERSION(7,6) */
> 
> Please check again if all these ifdefs for redhat are needed in this
> function. It looks like there are some bigger ifdes around most of the
> code in here.

This stuff is very tricky, and they tend to backport pieces of it ...

johannes

--
To unsubscribe from this list: send the line "unsubscribe backports" in