Drop the gid 60 compatibility path
gid 60 was never a decision -- it was the ENROLL token's timeout field being read as a gid, then made self-consistent. Keeping it readable alongside gid = uid would be a second code path, a second group to reason about, and a migration to eventually retire, all to avoid re-enrolling one finger on one dev phone. Re-enrol it instead.
This commit is contained in:
parent
1a54b5480c
commit
f2a696bbbe
2 changed files with 6 additions and 8 deletions
|
|
@ -71,13 +71,12 @@ export namespace fingerprintd::store {
|
|||
// We choose the caller's Linux uid, which makes the mapping total and
|
||||
// needs no allocation table.
|
||||
//
|
||||
// The dev phone's existing template sits under gid 60, which was never a
|
||||
// decision: it came from the ENROLL token's timeout field being read as
|
||||
// the gid, and was then made self-consistent. It is kept readable as a
|
||||
// legacy group so an enrolled finger is not stranded, but nothing new is
|
||||
// written there.
|
||||
// The research harness used gid 60, which was never a decision -- it was
|
||||
// the ENROLL token's timeout field being read as the gid, then made
|
||||
// self-consistent. There is no compatibility path for it: the finger it
|
||||
// enrolled gets re-enrolled under the uid. Nothing is worth carrying a
|
||||
// second group id for.
|
||||
using Gid = std::uint32_t;
|
||||
inline constexpr Gid LegacyGid = 60;
|
||||
inline Gid GidForUid(std::uint32_t uid) { return uid; }
|
||||
|
||||
// common.max_enrolling_fingers, from the trustlet config we ship.
|
||||
|
|
|
|||
Loading…
Reference in a new issue