modemmanager-next and cryptohome failures are flakes on unit tests that use timeouts. Usually these pass on retry, but I guess that's not always the case.

I can look at the cryptohome test and make it less flaky but the modemmanager one comes from upstream and so I wouldn't worry about that one too much.

The flashrom build failure looks really odd -- it's hard to tell because the build output is interleaved between the makefile checking for libpci and the actual build, but it looks like it did successfully compile something with libz. Don't know why there would be an issue there, but it seems transient somehow.

Additionally, sys-libs/zlib is in the depgrah for flashrom at depth 2.

$ equery-wolf depgraph --depth=2 flashrom
[...]
 `--  sys-apps/flashrom-9999  [missing keyword]
   `--  sys-apps/pciutils-3.4.1  (sys-apps/pciutils) unknown
[...]
    `--  sys-libs/zlib-1.2.11  (>=sys-libs/zlib-1.2.8-r1) unknown  [static-libs(+) abi_x86_32(-)? abi_x86_64(-)? abi_x86_x32(-)? abi_mips_n32(-)? abi_mips_n64(-)? abi_mips_o32(-)? abi_ppc_32(-)? abi_ppc_64(-)? abi_s390_32(-)? abi_s390_64(-)?]

Really not sure why this would happen but it seems like nothing is actually broken to me, and dependencies are sorted out.

Assigning own the flake.  Dropping priority because I'm not aware that this is a serious issue currently.