Commit f4ff2b49 authored by Ralf Jung's avatar Ralf Jung
Browse files

Merge branch 'ralf/no-coq-8.10' into 'master'

drop support for Coq 8.10

Closes #388

See merge request !603
parents 32bde4f0 bd2926f5
......@@ -32,6 +32,11 @@ build-coq.dev:
variables:
OPAM_PINS: "coq version dev"
build-coq.8.13.dev:
<<: *template
variables:
OPAM_PINS: "coq version 8.13.dev"
build-coq.8.12.1:
<<: *template
variables:
......@@ -48,11 +53,6 @@ build-coq.8.11.2:
variables:
OPAM_PINS: "coq version 8.11.2"
build-coq.8.10.2:
<<: *template
variables:
OPAM_PINS: "coq version 8.10.2"
# Nightly job with a known-to-work Coq version
# (so failures must be caused by std++)
build-stdpp.dev-coq.8.12.1:
......@@ -65,7 +65,7 @@ build-stdpp.dev-coq.8.12.1:
- schedules
- api
# Nightly job with latest Coq branch
# Nightly job with latest Coq beta branch
build-stdpp.dev-coq.8.13.dev:
<<: *template
variables:
......
......@@ -5,7 +5,7 @@ lemma.
## Iris master
With this release, we dropped support for Coq 8.9.
With this release, we dropped support for Coq 8.9 and Coq 8.10.
We also split Iris into multiple opam packages: `coq-iris` no longer contains
HeapLang, which is now in a separate package `coq-iris-heap-lang`.
......
......@@ -30,7 +30,7 @@ Importing Iris has some side effects as the library sets some global options.
This version is known to compile with:
- Coq 8.10.2 / 8.11.2 / 8.12.1
- Coq 8.11.2 / 8.12.1
- A development version of [std++](https://gitlab.mpi-sws.org/iris/stdpp)
If you need to work with Coq 8.7 or Coq 8.8, please check out the
......
......@@ -13,7 +13,7 @@ iris.prelude, iris.algebra, iris.si_logic, iris.bi, iris.proofmode, iris.base_lo
"""
depends: [
"coq" { (>= "8.10.2" & < "8.14~") | (= "dev") }
"coq" { (>= "8.11" & < "8.14~") | (= "dev") }
"coq-stdpp" { (= "dev.2020-11-11.2.6385b547") | (= "dev") }
]
......
......@@ -154,21 +154,21 @@ Tactic failure: iInv: invariant "H2" not found.
PROP : bi
============================
forall (P : PROP) (_ : True), bi_entails P P
forall P (_ : True), bi_entails P P
"p2"
: string
1 subgoal
PROP : bi
============================
forall P : PROP, and True (bi_entails P P)
forall P, and True (bi_entails P P)
"p3"
: string
1 subgoal
PROP : bi
============================
ex (fun P : PROP => bi_entails P P)
ex (fun P => bi_entails P P)
"p4"
: string
1 subgoal
......
......@@ -5,7 +5,6 @@ From iris.base_logic.lib Require Import invariants cancelable_invariants na_inva
From iris.bi Require Import ascii.
Set Default Proof Using "Type".
Unset Printing Use Implicit Types. (* FIXME: remove once we drop support for Coq <=8.11. *)
Section base_logic_tests.
Context {M : ucmraT}.
......
......@@ -29,17 +29,17 @@
Objective0 : Objective Q
𝓟, 𝓠 : PROP
============================
"H2" : ∀ i : I, Q i
"H2" : ∀ i, Q i
"H3" : 𝓟
"H4" : 𝓠
--------------------------------------∗
∀ i : I, 𝓟 ∗ 𝓠 ∗ Q i
∀ i, 𝓟 ∗ 𝓠 ∗ Q i
1 subgoal
I : biIndex
PROP : bi
FU0 : BiFUpd PROP * ()
FU : BiFUpd PROP
P, Q : monPred
i : I
============================
......@@ -50,7 +50,7 @@
I : biIndex
PROP : bi
FU0 : BiFUpd PROP * ()
FU : BiFUpd PROP
P : monPred
i : I
============================
......
From iris.proofmode Require Import tactics monpred.
From iris.base_logic.lib Require Import invariants.
Unset Printing Use Implicit Types. (* FIXME: remove once we drop support for Coq <=8.11. *)
Section tests.
Context {I : biIndex} {PROP : bi}.
......@@ -102,11 +101,7 @@ Section tests.
iAssumption.
Qed.
(* This is a hack to avoid avoid coq bug #5735: sections variables ignore hint
modes. So we assume the instances in a way that cannot be used by type
class resolution, and then separately declare the instance as such. *)
Context (FU0 : BiFUpd PROP * unit).
Instance FU : BiFUpd PROP := fst FU0.
Context (FU : BiFUpd PROP).
Lemma test_apply_fupd_intro_mask E1 E2 P :
E2 E1 P - |={E1,E2}=> |={E2,E1}=> P.
......
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment