Merge branch 'multi-output-hashDerivationModulo' of github.com:Ericson2314/nix into misc-ca

src/libexpr/primops.cc

@@ -800,7 +800,9 @@ static void prim_derivationStrict(EvalState & state, const Pos & pos, Value * *
                 });
         }
 
-        Hash h = hashDerivationModulo(*state.store, Derivation(drv), true);
+        // Regular, non-CA derivation should always return a single hash and not
+        // hash per output.
+        Hash h = std::get<0>(hashDerivationModulo(*state.store, Derivation(drv), true));
 
         for (auto & i : outputs) {
             auto outPath = state.store->makeOutputPath(i, h, drvName);

src/libstore/derivations.cc

@@ -369,50 +369,83 @@ bool BasicDerivation::isFixedOutput() const
 
 DrvHashes drvHashes;
 
+/* pathDerivationModulo and hashDerivationModulo are mutually recursive
+ */
 
-/* Returns the hash of a derivation modulo fixed-output
-   subderivations.  A fixed-output derivation is a derivation with one
-   output (`out') for which an expected hash and hash algorithm are
-   specified (using the `outputHash' and `outputHashAlgo'
-   attributes).  We don't want changes to such derivations to
-   propagate upwards through the dependency graph, changing output
-   paths everywhere.
+/* Look up the derivation by value and memoize the
+   `hashDerivationModulo` call.
+ */
+static const DrvHashModulo & pathDerivationModulo(Store & store, const StorePath & drvPath)
+{
+    auto h = drvHashes.find(drvPath);
+    if (h == drvHashes.end()) {
+        assert(store.isValidPath(drvPath));
+        // Cache it
+        h = drvHashes.insert_or_assign(
+            drvPath,
+            hashDerivationModulo(
+                store,
+                store.readDerivation(drvPath),
+                false)).first;
+    }
+    return h->second;
+}
 
-   For instance, if we change the url in a call to the `fetchurl'
-   function, we do not want to rebuild everything depending on it
-   (after all, (the hash of) the file being downloaded is unchanged).
-   So the *output paths* should not change.  On the other hand, the
-   *derivation paths* should change to reflect the new dependency
-   graph.
+/* See the header for interface details. These are the implementation details.
 
-   That's what this function does: it returns a hash which is just the
-   hash of the derivation ATerm, except that any input derivation
-   paths have been replaced by the result of a recursive call to this
-   function, and that for fixed-output derivations we return a hash of
-   its output path. */
-Hash hashDerivationModulo(Store & store, const Derivation & drv, bool maskOutputs)
+   For fixed-output derivations, each hash in the map is not the
+   corresponding output's content hash, but a hash of that hash along
+   with other constant data. The key point is that the value is a pure
+   function of the output's contents, and there are no preimage attacks
+   either spoofing an output's contents for a derivation, or
+   spoofing a derivation for an output's contents.
+
+   For regular derivations, it looks up each subderivation from its hash
+   and recurs. If the subderivation is also regular, it simply
+   substitutes the derivation path with its hash. If the subderivation
+   is fixed-output, however, it takes each output hash and pretends it
+   is a derivation hash producing a single "out" output. This is so we
+   don't leak the provenance of fixed outputs, reducing pointless cache
+   misses as the build itself won't know this.
+ */
+DrvHashModulo hashDerivationModulo(Store & store, const Derivation & drv, bool maskOutputs)
 {
     /* Return a fixed hash for fixed-output derivations. */
     if (drv.isFixedOutput()) {
-        DerivationOutputs::const_iterator i = drv.outputs.begin();
-        auto hash = std::get<DerivationOutputFixed>(i->second.output);
-        return hashString(htSHA256, "fixed:out:"
-            + hash.hash.printMethodAlgo() + ":"
-            + hash.hash.hash.to_string(Base16, false) + ":"
-            + store.printStorePath(i->second.path(store, drv.name)));
+        std::map<std::string, Hash> outputHashes;
+        for (const auto & i : drv.outputs) {
+            auto & dof = std::get<DerivationOutputFixed>(i.second.output);
+            auto hash = hashString(htSHA256, "fixed:out:"
+                + dof.hash.printMethodAlgo() + ":"
+                + dof.hash.hash.to_string(Base16, false) + ":"
+                + store.printStorePath(i.second.path(store, drv.name)));
+            outputHashes.insert_or_assign(i.first, std::move(hash));
+        }
+        return outputHashes;
     }
 
     /* For other derivations, replace the inputs paths with recursive
-       calls to this function.*/
+       calls to this function. */
     std::map<std::string, StringSet> inputs2;
     for (auto & i : drv.inputDrvs) {
-        auto h = drvHashes.find(i.first);
-        if (h == drvHashes.end()) {
-            assert(store.isValidPath(i.first));
-            h = drvHashes.insert_or_assign(i.first, hashDerivationModulo(store,
-                store.readDerivation(i.first), false)).first;
-        }
-        inputs2.insert_or_assign(h->second.to_string(Base16, false), i.second);
+        const auto & res = pathDerivationModulo(store, i.first);
+        std::visit(overloaded {
+            // Regular non-CA derivation, replace derivation
+            [&](Hash drvHash) {
+                inputs2.insert_or_assign(drvHash.to_string(Base16, false), i.second);
+            },
+            // CA derivation's output hashes
+            [&](CaOutputHashes outputHashes) {
+                std::set<std::string> justOut = { "out" };
+                for (auto & output : i.second) {
+                    /* Put each one in with a single "out" output.. */
+                    const auto h = outputHashes.at(output);
+                    inputs2.insert_or_assign(
+                        h.to_string(Base16, false),
+                        justOut);
+                }
+            },
+        }, res);
     }
 
     return hashString(htSHA256, drv.unparse(store, maskOutputs, &inputs2));

src/libstore/derivations.hh

@@ -6,6 +6,7 @@
 #include "content-address.hh"
 
 #include <map>
+#include <variant>
 
 
 namespace nix {
@@ -114,10 +115,42 @@ Derivation readDerivation(const Store & store, const Path & drvPath, std::string
 // FIXME: remove
 bool isDerivation(const string & fileName);
 
-Hash hashDerivationModulo(Store & store, const Derivation & drv, bool maskOutputs);
+// known CA drv's output hashes, current just for fixed-output derivations
+// whose output hashes are always known since they are fixed up-front.
+typedef std::map<std::string, Hash> CaOutputHashes;
+
+typedef std::variant<
+    Hash, // regular DRV normalized hash
+    CaOutputHashes
+> DrvHashModulo;
+
+/* Returns hashes with the details of fixed-output subderivations
+   expunged.
+
+   A fixed-output derivation is a derivation whose outputs have a
+   specified content hash and hash algorithm. (Currently they must have
+   exactly one output (`out'), which is specified using the `outputHash'
+   and `outputHashAlgo' attributes, but the algorithm doesn't assume
+   this.) We don't want changes to such derivations to propagate upwards
+   through the dependency graph, changing output paths everywhere.
+
+   For instance, if we change the url in a call to the `fetchurl'
+   function, we do not want to rebuild everything depending on it---after
+   all, (the hash of) the file being downloaded is unchanged.  So the
+   *output paths* should not change. On the other hand, the *derivation
+   paths* should change to reflect the new dependency graph.
+
+   For fixed-output derivations, this returns a map from the name of
+   each output to its hash, unique up to the output's contents.
+
+   For regular derivations, it returns a single hash of the derivation
+   ATerm, after subderivations have been likewise expunged from that
+   derivation.
+ */
+DrvHashModulo hashDerivationModulo(Store & store, const Derivation & drv, bool maskOutputs);
 
 /* Memoisation of hashDerivationModulo(). */
-typedef std::map<StorePath, Hash> DrvHashes;
+typedef std::map<StorePath, DrvHashModulo> DrvHashes;
 
 extern DrvHashes drvHashes; // FIXME: global, not thread-safe
 

src/libstore/local-store.cc

@@ -563,7 +563,9 @@ void LocalStore::checkDerivationOutputs(const StorePath & drvPath, const Derivat
     }
 
     else {
-        Hash h = hashDerivationModulo(*this, drv, true);
+        // Regular, non-CA derivation should always return a single hash and not
+        // hash per output.
+        Hash h = std::get<0>(hashDerivationModulo(*this, drv, true));
         for (auto & i : drv.outputs)
             check(makeOutputPath(i.first, h, drvName), i.second.path(*this, drv.name), i.first);
     }

src/nix/develop.cc

@@ -135,7 +135,7 @@ StorePath getDerivationEnvironment(ref<Store> store, const StorePath & drvPath)
     drv.env["_outputs_saved"] = drv.env["outputs"];
     drv.env["outputs"] = "out";
     drv.inputSrcs.insert(std::move(getEnvShPath));
-    Hash h = hashDerivationModulo(*store, drv, true);
+    Hash h = std::get<0>(hashDerivationModulo(*store, drv, true));
     auto shellOutPath = store->makeOutputPath("out", h, drvName);
     drv.outputs.insert_or_assign("out", DerivationOutput { .output = DerivationOutputInputAddressed {
                 .path = shellOutPath