geom_uzip to update zlib version 1.0.4 to 1.2.11.

[delphij]

Summary

Update geom_uzip to use newer zlib version.

This change depends on D20193 and D19706.
Based on a new proposal, this also depends on D21156.

Cross reference :
https://reviews.freebsd.org/D19706?id=55444#inline-120590

Test Plan

% kldload geom_uzip
% mdconfig -a -t vnode -f <uzip-file>
% mount -r /dev/md#.uzip /mnt/<path>

[sobomax]

Looks good, as long as it compiles.

[ota_j.email.ne.jp]

Thank you for taking a look.

This change-set requires 2 other change-sets, crc32 relocation and dual zlib compilation before ready to land.

I actually didn't expect this is marked as "Buildable 24253"....

[delphij]

sys/geom/uzip/g_uzip_zlib.c

114

I think inflateInit() still expects zfree be set?

[ota_j.email.ne.jp]

sys/geom/uzip/g_uzip_zlib.c

110

compressBounce is taken out by -DZ_SOLO.
We will need to adjust zlib side to expose with -DZ_SOLO.

114

I will check. Using -DZ_SOLO changes implementation and don't recall.

[ota_j.email.ne.jp]

Given we added -DZ_SOLO to zlib, pass zfree function explicitly.
Expose compressBound() function even with -DZ_SOLO.
compress2() doesn't work with -DZ_SOLO and not worth exposing as is.
Alternative is improve MY_ZCALLOC conditions along with Z_SOLO for compress2 to work as well.

[ota_j.email.ne.jp]

Enhancce MY_ZCALLOC to expose functions in compress.c.
This also allows to use default zfree.

[ota_j.email.ne.jp]

Expose functions from compress.c with modified MY_ZCALLC handling.

[delphij]

Rebase over D21156.

[ota_j.email.ne.jp]

sys/geom/uzip/g_uzip_zlib.c

113

Given geom_uzip uses nowait kmalloc,

	zp->zs.zalloc = zcalloc_nowait;

is the equivalent choice.

I am not sure which one provides better system handling and user experiences, though.

It will be nice if we can get file-system and kernel memory experts inputs.

135

geom_uzip uses M_NOWAIT option.

[delphij]

sys/geom/uzip/g_uzip_zlib.c

113

My version of zcalloc was the nowait variant :)

[delphij]

Properly initialize zstream structure.

[ota_j.email.ne.jp]

sys/geom/uzip/g_uzip_zlib.c

113

Yes, that was the case.
I somehow got confused and mixed up.