Why the GTK+ wayland backend can't be enabled in linux distros, at all

[Pierre-Loup A. Griffais]

On 03/27/2012 06:35 PM, Thiago Macieira wrote:

[...]

> 1) prelink refused to work on ELF modules with text relocations,
> which excluded the NVidia libs outright

I'm using the --reloc-only option, which works fine even on our non-PIC
x86 binaries.

> Prelinking helps with load time processing, but not with reducing
> memory footprint.

You'll find that this is wrong, it is necessary to achieve the results I 
described below.

>> However, the linux ELF loader currently has a bug that causes the
>> second type of relocation to be overwritten even if prelink had
>> already initialized them to the correct value. This causes the
>> memory in the relocation sections to get dirtied without anything
>> being changed, which contradicts our goal.
>
> That might explain my results too.
>
>> Attached is a proof-of-concept patch which fixes this issue for the
>> x86_64 Linux ELF loader. By both applying this patch to the loader
>> and prelinking libnvidia-glcore.so, distros can reduce the memory
>> overhead of linking an application against the NVIDIA libGL to just
>> a few kB:
>
> Interesting. Those are very good results. Can you provide the same
> results when two applications are running? Our hope is that those
> 6544 kB of Private_Clean (which are sharable) become actually shared
> and move to Shared_Clean.

Yes, they do:

------

7fc000000000-7fc00192a000 r-xp 00000000 08:01 1591924
  /usr/lib/libnvidia-glcore.so.295.33
Size:              25768 kB
Rss:                9108 kB
Pss:                3037 kB
Shared_Clean:       9108 kB
Shared_Dirty:          0 kB
Private_Clean:         0 kB
Private_Dirty:         0 kB
Referenced:         9108 kB
Anonymous:             0 kB
AnonHugePages:         0 kB
Swap:                  0 kB
KernelPageSize:        4 kB
MMUPageSize:           4 kB
Locked:                0 kB
7fc00192a000-7fc001b2a000 ---p 0192a000 08:01 1591924
  /usr/lib/libnvidia-glcore.so.295.33
Size:               2048 kB
Rss:                   0 kB
Pss:                   0 kB
Shared_Clean:          0 kB
Shared_Dirty:          0 kB
Private_Clean:         0 kB
Private_Dirty:         0 kB
Referenced:            0 kB
Anonymous:             0 kB
AnonHugePages:         0 kB
Swap:                  0 kB
KernelPageSize:        4 kB
MMUPageSize:           4 kB
Locked:                0 kB
7fc001b2a000-7fc002309000 rwxp 0192a000 08:01 1591924
  /usr/lib/libnvidia-glcore.so.295.33
Size:               8060 kB
Rss:                6592 kB
Pss:                3320 kB
Shared_Clean:       6544 kB
Shared_Dirty:          0 kB
Private_Clean:         0 kB
Private_Dirty:        48 kB
Referenced:         6592 kB
Anonymous:            48 kB
AnonHugePages:         0 kB
Swap:                  0 kB
KernelPageSize:        4 kB
MMUPageSize:           4 kB
Locked:                0 kB

------

As you can see, everything functions as expected and the NVIDIA driver
isn't the problem here; it's very unfortunate that everyone already
relayed the opposite information.

> Has this patch been submitted to the glibc folks? They have just had
> a change of governance.

No, it's just a proof-of-concept at this stage. I'll submit it for
review later after cleaning it up and adding support for i386 relocations.

Thanks,
  - Pierre-Loup