| Summary: | net-misc/curl-7.67.0[esni] fails to build against dev-libs/openssl-1.1.1d-r2 | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Michał Bartoszkiewicz <mbartoszkiewicz> |
| Component: | Current packages | Assignee: | Anthony Basile <blueness> |
| Status: | RESOLVED FIXED | ||
| Severity: | normal | CC: | 1i5t5.duncan, arfrever.fta, conikost, jstein, ua_gentoo_bugzilla |
| Priority: | Normal | ||
| Version: | unspecified | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Attachments: | Build log | ||
|
Description
Michał Bartoszkiewicz
2019-11-09 08:49:15 UTC
Created attachment 595518 [details]
Build log
In reply to Michał Bartoszkiewicz from comment #0) > From reading the configure script, it also appears that esni is only > available when building against openssl, so there probably should be some > REQUIRED_USE constraints to enforce that in the ebuild. Yes. Same "No ESNI support found" error here, with CURL_SSL=gnutls. The RELEASE-NOTES file footnotes this link for ESNI: https://github.com/curl/curl/pull/4011 ... which says under limits: * Check for available resources, although extensible, refers only to specific work in progress (described here) to implement ESNI for OpenSSL, as this is the immediate motivation for the proposed change.  That "described here" link in turn is to: https://github.com/sftcd/openssl/tree/master/esnistuff ... which is an esni-specific openssl fork that notes (with a late Oct, 2019 date) that it's not PR-ready for mainline openssl yet. So unless the maintainer wants to play with it enough to hassle use-mask, I'd say just kill the USE=esni flag entirely and hard-disable it until support has matured substantially. Meanwhile, for users like the reporter and I who tried it, simply disabling USE=esni again will suffice. After all, however nice it sounds it's not like we'll be disabling something we can't do without, since we've been doing without until now. =:^) (In reply to Duncan from comment #2) > > So unless the maintainer wants to play with it enough to hassle use-mask, > I'd say just kill the USE=esni flag entirely and hard-disable it until > support has matured substantially. > I'm going to hard disable it for now. |
