[vbox-dev] Fixes for Kernel 5.3 wanted

Cyrax evvke at hotmail.com
Mon Aug 5 22:13:59 UTC 2019


On 05/08/2019 19:30, Valdis Klētnieks wrote:
> On Mon, 05 Aug 2019 10:50:47 -0500, Larry Finger said:
> 
>>> /tmp/vbox.0/linux/VBoxNetFlt-linux.c:928:66: error: ‘skb_frag_t’ {aka ‘struct
>>> bio_vec’} has no member named ‘page_offset’; did you mean ‘bv_offset’?
> 
>> I do not see these errors when building the kernel modules with kernel
>> 5.3.0-rc3. The patch I posted is sufficient for me. Note I am using an
>> unmodified kernel from mainline. A kernel obtained from a disto might have other
>> modifications.
> 
> This isn't a 5.3.0-rc3 issue.
> 
> The kernel patches causing the issue are a set of 7 a in the linux-next tree
> and targeting the 5.4 kernel .  As such, Cyrax's patch is incorrect, for production
> it should be comparing >= 5.4.0.   It will however DTRT on a linux-next kernel
> that calls itself 5.3.something, but will *not* work on a mainline Linus tree.
> 
> (Hey Cyrax, thanks for the patch anyhow, from another linux-next user :)
> 
> The patch series is here:
> 
> https://www.spinics.net/lists/netdev/msg585545.html
> 
> commit d5c3a62d0bb9b763e9378fe8f4cd79502e16cce8
> Merge: 7e24b4ed5ac4 8842d285bafa
> Author: David S. Miller <davem at davemloft.net>
> Date:   Mon Jul 22 20:47:56 2019 -0700
> 
>      Merge branch 'Convert-skb_frag_t-to-bio_vec'
>      
>      Matthew Wilcox says:
>      
>      ====================
>      Convert skb_frag_t to bio_vec
>      
>      The skb_frag_t and bio_vec are fundamentally the same (page, offset,
>      length) tuple.  This patch series unifies the two, leaving the
>      skb_frag_t typedef in place.  This has the immediate advantage that
>      we already have iov_iter support for bvecs and don't need to add
>      support for iterating skbuffs.  It enables a long-term plan to use
>      bvecs more broadly within the kernel and should make network-storage
>      drivers able to do less work converting between skbuffs and biovecs.
>      
>      It will consume more memory on 32-bit kernels.  If that proves
>      problematic, we can look at ways of addressing it.
>      
>      v3: Rebase on latest Linus with net-next merged.
>        - Reorder the uncontroversial 'Use skb accessors' patches first so you
>          can apply just those two if you want to hold off on the full
>          conversion.
>        - Convert all the users of 'struct skb_frag_struct' to skb_frag_t.
>      ====================
>      
>      Signed-off-by: David S. Miller <davem at davemloft.net>
> 
> 
> 

Oopsie! :$ :-D
I got confused about what kernels I have installed on and on what I have 
been dabbling on.
Oh, well.



More information about the vbox-dev mailing list