Eliminate the "store" global variable
Also, move a few free-standing functions into StoreAPI and Derivation.
Also, introduce a non-nullable smart pointer, ref<T>, which is just a
wrapper around std::shared_ptr ensuring that the pointer is never
null. (For reference-counted values, this is better than passing a
"T&", because the latter doesn't maintain the refcount. Usually, the
caller will have a shared_ptr keeping the value alive, but that's not
always the case, e.g., when passing a reference to a std::thread via
std::bind.)
2016-02-04 13:28:26 +00:00
|
|
|
#include "derivations.hh"
|
2019-09-03 14:02:12 +00:00
|
|
|
#include "parsed-derivations.hh"
|
2012-04-30 23:15:34 +00:00
|
|
|
#include "globals.hh"
|
Eliminate the "store" global variable
Also, move a few free-standing functions into StoreAPI and Derivation.
Also, introduce a non-nullable smart pointer, ref<T>, which is just a
wrapper around std::shared_ptr ensuring that the pointer is never
null. (For reference-counted values, this is better than passing a
"T&", because the latter doesn't maintain the refcount. Usually, the
caller will have a shared_ptr keeping the value alive, but that's not
always the case, e.g., when passing a reference to a std::thread via
std::bind.)
2016-02-04 13:28:26 +00:00
|
|
|
#include "local-store.hh"
|
|
|
|
#include "store-api.hh"
|
2016-07-21 12:13:35 +00:00
|
|
|
#include "thread-pool.hh"
|
2020-07-27 20:45:34 +00:00
|
|
|
#include "topo-sort.hh"
|
2020-09-21 16:40:11 +00:00
|
|
|
#include "callback.hh"
|
2021-05-18 12:30:32 +00:00
|
|
|
#include "closure.hh"
|
2021-12-13 20:08:42 +00:00
|
|
|
#include "filetransfer.hh"
|
2006-09-04 21:06:23 +00:00
|
|
|
|
|
|
|
namespace nix {
|
2004-06-18 18:09:32 +00:00
|
|
|
|
2019-12-05 18:11:09 +00:00
|
|
|
void Store::computeFSClosure(const StorePathSet & startPaths,
|
|
|
|
StorePathSet & paths_, bool flipDirection, bool includeOutputs, bool includeDerivers)
|
2005-01-19 11:16:11 +00:00
|
|
|
{
|
2021-05-18 12:30:32 +00:00
|
|
|
std::function<std::set<StorePath>(const StorePath & path, std::future<ref<const ValidPathInfo>> &)> queryDeps;
|
|
|
|
if (flipDirection)
|
|
|
|
queryDeps = [&](const StorePath& path,
|
|
|
|
std::future<ref<const ValidPathInfo>> & fut) {
|
|
|
|
StorePathSet res;
|
|
|
|
StorePathSet referrers;
|
|
|
|
queryReferrers(path, referrers);
|
2023-02-01 15:07:54 +00:00
|
|
|
for (auto& ref : referrers)
|
2021-05-18 12:30:32 +00:00
|
|
|
if (ref != path)
|
|
|
|
res.insert(ref);
|
|
|
|
|
|
|
|
if (includeOutputs)
|
2023-02-01 15:07:54 +00:00
|
|
|
for (auto& i : queryValidDerivers(path))
|
2021-05-18 12:30:32 +00:00
|
|
|
res.insert(i);
|
|
|
|
|
|
|
|
if (includeDerivers && path.isDerivation())
|
2023-02-01 15:07:54 +00:00
|
|
|
for (auto& [_, maybeOutPath] : queryPartialDerivationOutputMap(path))
|
2021-06-12 10:24:53 +00:00
|
|
|
if (maybeOutPath && isValidPath(*maybeOutPath))
|
2021-06-11 07:24:24 +00:00
|
|
|
res.insert(*maybeOutPath);
|
2021-05-18 12:30:32 +00:00
|
|
|
return res;
|
|
|
|
};
|
|
|
|
else
|
|
|
|
queryDeps = [&](const StorePath& path,
|
|
|
|
std::future<ref<const ValidPathInfo>> & fut) {
|
|
|
|
StorePathSet res;
|
|
|
|
auto info = fut.get();
|
2023-02-01 15:07:54 +00:00
|
|
|
for (auto& ref : info->references)
|
2023-01-14 21:38:43 +00:00
|
|
|
if (ref != path)
|
|
|
|
res.insert(ref);
|
2021-05-18 12:30:32 +00:00
|
|
|
|
|
|
|
if (includeOutputs && path.isDerivation())
|
2023-02-01 15:07:54 +00:00
|
|
|
for (auto& [_, maybeOutPath] : queryPartialDerivationOutputMap(path))
|
2021-06-11 07:24:24 +00:00
|
|
|
if (maybeOutPath && isValidPath(*maybeOutPath))
|
|
|
|
res.insert(*maybeOutPath);
|
2021-05-18 12:30:32 +00:00
|
|
|
|
|
|
|
if (includeDerivers && info->deriver && isValidPath(*info->deriver))
|
|
|
|
res.insert(*info->deriver);
|
|
|
|
return res;
|
|
|
|
};
|
|
|
|
|
|
|
|
computeClosure<StorePath>(
|
|
|
|
startPaths, paths_,
|
|
|
|
[&](const StorePath& path,
|
|
|
|
std::function<void(std::promise<std::set<StorePath>>&)>
|
|
|
|
processEdges) {
|
|
|
|
std::promise<std::set<StorePath>> promise;
|
|
|
|
std::function<void(std::future<ref<const ValidPathInfo>>)>
|
|
|
|
getDependencies =
|
|
|
|
[&](std::future<ref<const ValidPathInfo>> fut) {
|
|
|
|
try {
|
|
|
|
promise.set_value(queryDeps(path, fut));
|
|
|
|
} catch (...) {
|
|
|
|
promise.set_exception(std::current_exception());
|
|
|
|
}
|
|
|
|
};
|
|
|
|
queryPathInfo(path, getDependencies);
|
|
|
|
processEdges(promise);
|
|
|
|
});
|
2005-02-14 17:35:10 +00:00
|
|
|
}
|
2006-03-06 11:21:15 +00:00
|
|
|
|
2019-12-05 18:11:09 +00:00
|
|
|
void Store::computeFSClosure(const StorePath & startPath,
|
|
|
|
StorePathSet & paths_, bool flipDirection, bool includeOutputs, bool includeDerivers)
|
2016-11-10 16:45:04 +00:00
|
|
|
{
|
2019-12-05 18:11:09 +00:00
|
|
|
StorePathSet paths;
|
2020-06-16 20:20:18 +00:00
|
|
|
paths.insert(startPath);
|
2019-12-05 18:11:09 +00:00
|
|
|
computeFSClosure(paths, paths_, flipDirection, includeOutputs, includeDerivers);
|
2016-11-10 16:45:04 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2020-06-22 17:08:11 +00:00
|
|
|
std::optional<ContentAddress> getDerivationCA(const BasicDerivation & drv)
|
2020-06-12 14:49:09 +00:00
|
|
|
{
|
2020-06-22 17:08:11 +00:00
|
|
|
auto out = drv.outputs.find("out");
|
2020-10-12 23:51:23 +00:00
|
|
|
if (out == drv.outputs.end())
|
|
|
|
return std::nullopt;
|
2022-03-25 19:40:52 +00:00
|
|
|
if (auto dof = std::get_if<DerivationOutput::CAFixed>(&out->second)) {
|
2020-10-12 23:51:23 +00:00
|
|
|
return std::visit(overloaded {
|
2021-10-01 17:25:22 +00:00
|
|
|
[&](const TextInfo & ti) -> std::optional<ContentAddress> {
|
2020-10-13 16:22:30 +00:00
|
|
|
if (!ti.references.empty())
|
|
|
|
return std::nullopt;
|
2023-02-28 17:46:00 +00:00
|
|
|
return ti.hash;
|
2020-10-13 16:22:30 +00:00
|
|
|
},
|
2021-10-01 17:25:22 +00:00
|
|
|
[&](const FixedOutputInfo & fi) -> std::optional<ContentAddress> {
|
2023-01-06 20:37:31 +00:00
|
|
|
if (!fi.references.empty())
|
2020-10-13 16:22:30 +00:00
|
|
|
return std::nullopt;
|
2023-02-28 17:46:00 +00:00
|
|
|
return fi.hash;
|
2020-10-13 16:22:30 +00:00
|
|
|
},
|
2023-04-01 20:40:32 +00:00
|
|
|
}, dof->ca.raw);
|
2020-07-30 17:38:24 +00:00
|
|
|
}
|
2020-06-12 14:49:09 +00:00
|
|
|
return std::nullopt;
|
|
|
|
}
|
|
|
|
|
2021-04-05 13:48:18 +00:00
|
|
|
void Store::queryMissing(const std::vector<DerivedPath> & targets,
|
2019-12-05 18:11:09 +00:00
|
|
|
StorePathSet & willBuild_, StorePathSet & willSubstitute_, StorePathSet & unknown_,
|
2020-07-30 11:10:49 +00:00
|
|
|
uint64_t & downloadSize_, uint64_t & narSize_)
|
2006-03-06 11:21:15 +00:00
|
|
|
{
|
2017-08-31 14:02:36 +00:00
|
|
|
Activity act(*logger, lvlDebug, actUnknown, "querying info about missing paths");
|
|
|
|
|
2016-07-21 15:40:40 +00:00
|
|
|
downloadSize_ = narSize_ = 0;
|
2006-03-06 11:21:15 +00:00
|
|
|
|
2021-12-13 20:08:42 +00:00
|
|
|
// FIXME: make async.
|
|
|
|
ThreadPool pool(fileTransferSettings.httpConnections);
|
download-from-binary-cache: parallelise fetching of NAR info files
Getting substitute information using the binary cache substituter has
non-trivial latency overhead. A package or NixOS system configuration
can have hundreds of dependencies, and in the worst case (when the
local info cache is empty) we have to do a separate HTTP request for
each of these. If the ping time to the server is t, getting N info
files will take tN seconds; e.g., with a ping time of 0.1s to
nixos.org, sequentially downloading 1000 info files (a typical NixOS
config) will take at least 100 seconds.
To fix this problem, the binary cache substituter can now perform
requests in parallel. This required changing the substituter
interface to support a function querySubstitutablePathInfos() that
queries multiple paths at the same time, and rewriting queryMissing()
to take advantage of parallelism. (Due to local caching,
parallelising queryMissing() is sufficient for most use cases, since
it's almost always called before building a derivation and thus fills
the local info cache.)
For example, parallelism speeds up querying all 1056 paths in a
particular NixOS system configuration from 116s to 2.6s. It works so
well because the eccentricity of the top-level derivation in the
dependency graph is only 9. So we only need 10 round-trips (when
using an unlimited number of parallel connections) to get everything.
Currently we do a maximum of 150 parallel connections to the server.
Thus it's important that the binary cache server (e.g. nixos.org) has
a high connection limit. Alternatively we could use HTTP pipelining,
but WWW::Curl doesn't support it and libcurl has a hard-coded limit of
5 requests per pipeline.
2012-07-06 23:08:20 +00:00
|
|
|
|
2016-07-21 15:40:40 +00:00
|
|
|
struct State
|
|
|
|
{
|
2019-12-05 18:11:09 +00:00
|
|
|
std::unordered_set<std::string> done;
|
|
|
|
StorePathSet & unknown, & willSubstitute, & willBuild;
|
2020-07-30 11:10:49 +00:00
|
|
|
uint64_t & downloadSize;
|
|
|
|
uint64_t & narSize;
|
2016-07-21 15:40:40 +00:00
|
|
|
};
|
download-from-binary-cache: parallelise fetching of NAR info files
Getting substitute information using the binary cache substituter has
non-trivial latency overhead. A package or NixOS system configuration
can have hundreds of dependencies, and in the worst case (when the
local info cache is empty) we have to do a separate HTTP request for
each of these. If the ping time to the server is t, getting N info
files will take tN seconds; e.g., with a ping time of 0.1s to
nixos.org, sequentially downloading 1000 info files (a typical NixOS
config) will take at least 100 seconds.
To fix this problem, the binary cache substituter can now perform
requests in parallel. This required changing the substituter
interface to support a function querySubstitutablePathInfos() that
queries multiple paths at the same time, and rewriting queryMissing()
to take advantage of parallelism. (Due to local caching,
parallelising queryMissing() is sufficient for most use cases, since
it's almost always called before building a derivation and thus fills
the local info cache.)
For example, parallelism speeds up querying all 1056 paths in a
particular NixOS system configuration from 116s to 2.6s. It works so
well because the eccentricity of the top-level derivation in the
dependency graph is only 9. So we only need 10 round-trips (when
using an unlimited number of parallel connections) to get everything.
Currently we do a maximum of 150 parallel connections to the server.
Thus it's important that the binary cache server (e.g. nixos.org) has
a high connection limit. Alternatively we could use HTTP pipelining,
but WWW::Curl doesn't support it and libcurl has a hard-coded limit of
5 requests per pipeline.
2012-07-06 23:08:20 +00:00
|
|
|
|
2016-07-21 15:40:40 +00:00
|
|
|
struct DrvState
|
|
|
|
{
|
|
|
|
size_t left;
|
|
|
|
bool done = false;
|
2019-12-05 18:11:09 +00:00
|
|
|
StorePathSet outPaths;
|
2016-07-21 15:40:40 +00:00
|
|
|
DrvState(size_t left) : left(left) { }
|
|
|
|
};
|
download-from-binary-cache: parallelise fetching of NAR info files
Getting substitute information using the binary cache substituter has
non-trivial latency overhead. A package or NixOS system configuration
can have hundreds of dependencies, and in the worst case (when the
local info cache is empty) we have to do a separate HTTP request for
each of these. If the ping time to the server is t, getting N info
files will take tN seconds; e.g., with a ping time of 0.1s to
nixos.org, sequentially downloading 1000 info files (a typical NixOS
config) will take at least 100 seconds.
To fix this problem, the binary cache substituter can now perform
requests in parallel. This required changing the substituter
interface to support a function querySubstitutablePathInfos() that
queries multiple paths at the same time, and rewriting queryMissing()
to take advantage of parallelism. (Due to local caching,
parallelising queryMissing() is sufficient for most use cases, since
it's almost always called before building a derivation and thus fills
the local info cache.)
For example, parallelism speeds up querying all 1056 paths in a
particular NixOS system configuration from 116s to 2.6s. It works so
well because the eccentricity of the top-level derivation in the
dependency graph is only 9. So we only need 10 round-trips (when
using an unlimited number of parallel connections) to get everything.
Currently we do a maximum of 150 parallel connections to the server.
Thus it's important that the binary cache server (e.g. nixos.org) has
a high connection limit. Alternatively we could use HTTP pipelining,
but WWW::Curl doesn't support it and libcurl has a hard-coded limit of
5 requests per pipeline.
2012-07-06 23:08:20 +00:00
|
|
|
|
2019-12-05 18:11:09 +00:00
|
|
|
Sync<State> state_(State{{}, unknown_, willSubstitute_, willBuild_, downloadSize_, narSize_});
|
download-from-binary-cache: parallelise fetching of NAR info files
Getting substitute information using the binary cache substituter has
non-trivial latency overhead. A package or NixOS system configuration
can have hundreds of dependencies, and in the worst case (when the
local info cache is empty) we have to do a separate HTTP request for
each of these. If the ping time to the server is t, getting N info
files will take tN seconds; e.g., with a ping time of 0.1s to
nixos.org, sequentially downloading 1000 info files (a typical NixOS
config) will take at least 100 seconds.
To fix this problem, the binary cache substituter can now perform
requests in parallel. This required changing the substituter
interface to support a function querySubstitutablePathInfos() that
queries multiple paths at the same time, and rewriting queryMissing()
to take advantage of parallelism. (Due to local caching,
parallelising queryMissing() is sufficient for most use cases, since
it's almost always called before building a derivation and thus fills
the local info cache.)
For example, parallelism speeds up querying all 1056 paths in a
particular NixOS system configuration from 116s to 2.6s. It works so
well because the eccentricity of the top-level derivation in the
dependency graph is only 9. So we only need 10 round-trips (when
using an unlimited number of parallel connections) to get everything.
Currently we do a maximum of 150 parallel connections to the server.
Thus it's important that the binary cache server (e.g. nixos.org) has
a high connection limit. Alternatively we could use HTTP pipelining,
but WWW::Curl doesn't support it and libcurl has a hard-coded limit of
5 requests per pipeline.
2012-07-06 23:08:20 +00:00
|
|
|
|
2021-04-05 13:48:18 +00:00
|
|
|
std::function<void(DerivedPath)> doPath;
|
download-from-binary-cache: parallelise fetching of NAR info files
Getting substitute information using the binary cache substituter has
non-trivial latency overhead. A package or NixOS system configuration
can have hundreds of dependencies, and in the worst case (when the
local info cache is empty) we have to do a separate HTTP request for
each of these. If the ping time to the server is t, getting N info
files will take tN seconds; e.g., with a ping time of 0.1s to
nixos.org, sequentially downloading 1000 info files (a typical NixOS
config) will take at least 100 seconds.
To fix this problem, the binary cache substituter can now perform
requests in parallel. This required changing the substituter
interface to support a function querySubstitutablePathInfos() that
queries multiple paths at the same time, and rewriting queryMissing()
to take advantage of parallelism. (Due to local caching,
parallelising queryMissing() is sufficient for most use cases, since
it's almost always called before building a derivation and thus fills
the local info cache.)
For example, parallelism speeds up querying all 1056 paths in a
particular NixOS system configuration from 116s to 2.6s. It works so
well because the eccentricity of the top-level derivation in the
dependency graph is only 9. So we only need 10 round-trips (when
using an unlimited number of parallel connections) to get everything.
Currently we do a maximum of 150 parallel connections to the server.
Thus it's important that the binary cache server (e.g. nixos.org) has
a high connection limit. Alternatively we could use HTTP pipelining,
but WWW::Curl doesn't support it and libcurl has a hard-coded limit of
5 requests per pipeline.
2012-07-06 23:08:20 +00:00
|
|
|
|
2019-12-05 18:11:09 +00:00
|
|
|
auto mustBuildDrv = [&](const StorePath & drvPath, const Derivation & drv) {
|
2016-07-21 15:40:40 +00:00
|
|
|
{
|
|
|
|
auto state(state_.lock());
|
2020-06-16 20:20:18 +00:00
|
|
|
state->willBuild.insert(drvPath);
|
2016-07-21 15:40:40 +00:00
|
|
|
}
|
2012-07-30 23:55:41 +00:00
|
|
|
|
2016-07-21 15:40:40 +00:00
|
|
|
for (auto & i : drv.inputDrvs)
|
2021-04-05 13:48:18 +00:00
|
|
|
pool.enqueue(std::bind(doPath, DerivedPath::Built { i.first, i.second }));
|
2016-07-21 15:40:40 +00:00
|
|
|
};
|
download-from-binary-cache: parallelise fetching of NAR info files
Getting substitute information using the binary cache substituter has
non-trivial latency overhead. A package or NixOS system configuration
can have hundreds of dependencies, and in the worst case (when the
local info cache is empty) we have to do a separate HTTP request for
each of these. If the ping time to the server is t, getting N info
files will take tN seconds; e.g., with a ping time of 0.1s to
nixos.org, sequentially downloading 1000 info files (a typical NixOS
config) will take at least 100 seconds.
To fix this problem, the binary cache substituter can now perform
requests in parallel. This required changing the substituter
interface to support a function querySubstitutablePathInfos() that
queries multiple paths at the same time, and rewriting queryMissing()
to take advantage of parallelism. (Due to local caching,
parallelising queryMissing() is sufficient for most use cases, since
it's almost always called before building a derivation and thus fills
the local info cache.)
For example, parallelism speeds up querying all 1056 paths in a
particular NixOS system configuration from 116s to 2.6s. It works so
well because the eccentricity of the top-level derivation in the
dependency graph is only 9. So we only need 10 round-trips (when
using an unlimited number of parallel connections) to get everything.
Currently we do a maximum of 150 parallel connections to the server.
Thus it's important that the binary cache server (e.g. nixos.org) has
a high connection limit. Alternatively we could use HTTP pipelining,
but WWW::Curl doesn't support it and libcurl has a hard-coded limit of
5 requests per pipeline.
2012-07-06 23:08:20 +00:00
|
|
|
|
2016-07-21 15:40:40 +00:00
|
|
|
auto checkOutput = [&](
|
2021-03-05 00:49:46 +00:00
|
|
|
const StorePath & drvPath, ref<Derivation> drv, const StorePath & outPath, ref<Sync<DrvState>> drvState_)
|
2016-07-21 15:40:40 +00:00
|
|
|
{
|
|
|
|
if (drvState_->lock()->done) return;
|
download-from-binary-cache: parallelise fetching of NAR info files
Getting substitute information using the binary cache substituter has
non-trivial latency overhead. A package or NixOS system configuration
can have hundreds of dependencies, and in the worst case (when the
local info cache is empty) we have to do a separate HTTP request for
each of these. If the ping time to the server is t, getting N info
files will take tN seconds; e.g., with a ping time of 0.1s to
nixos.org, sequentially downloading 1000 info files (a typical NixOS
config) will take at least 100 seconds.
To fix this problem, the binary cache substituter can now perform
requests in parallel. This required changing the substituter
interface to support a function querySubstitutablePathInfos() that
queries multiple paths at the same time, and rewriting queryMissing()
to take advantage of parallelism. (Due to local caching,
parallelising queryMissing() is sufficient for most use cases, since
it's almost always called before building a derivation and thus fills
the local info cache.)
For example, parallelism speeds up querying all 1056 paths in a
particular NixOS system configuration from 116s to 2.6s. It works so
well because the eccentricity of the top-level derivation in the
dependency graph is only 9. So we only need 10 round-trips (when
using an unlimited number of parallel connections) to get everything.
Currently we do a maximum of 150 parallel connections to the server.
Thus it's important that the binary cache server (e.g. nixos.org) has
a high connection limit. Alternatively we could use HTTP pipelining,
but WWW::Curl doesn't support it and libcurl has a hard-coded limit of
5 requests per pipeline.
2012-07-06 23:08:20 +00:00
|
|
|
|
2016-07-21 15:40:40 +00:00
|
|
|
SubstitutablePathInfos infos;
|
2020-06-17 19:03:05 +00:00
|
|
|
querySubstitutablePathInfos({{outPath, getDerivationCA(*drv)}}, infos);
|
2012-11-26 14:39:10 +00:00
|
|
|
|
2016-07-21 15:40:40 +00:00
|
|
|
if (infos.empty()) {
|
|
|
|
drvState_->lock()->done = true;
|
|
|
|
mustBuildDrv(drvPath, *drv);
|
|
|
|
} else {
|
|
|
|
{
|
|
|
|
auto drvState(drvState_->lock());
|
|
|
|
if (drvState->done) return;
|
|
|
|
assert(drvState->left);
|
|
|
|
drvState->left--;
|
2020-06-16 20:20:18 +00:00
|
|
|
drvState->outPaths.insert(outPath);
|
2016-07-21 15:40:40 +00:00
|
|
|
if (!drvState->left) {
|
|
|
|
for (auto & path : drvState->outPaths)
|
2021-04-05 13:48:18 +00:00
|
|
|
pool.enqueue(std::bind(doPath, DerivedPath::Opaque { path } ));
|
download-from-binary-cache: parallelise fetching of NAR info files
Getting substitute information using the binary cache substituter has
non-trivial latency overhead. A package or NixOS system configuration
can have hundreds of dependencies, and in the worst case (when the
local info cache is empty) we have to do a separate HTTP request for
each of these. If the ping time to the server is t, getting N info
files will take tN seconds; e.g., with a ping time of 0.1s to
nixos.org, sequentially downloading 1000 info files (a typical NixOS
config) will take at least 100 seconds.
To fix this problem, the binary cache substituter can now perform
requests in parallel. This required changing the substituter
interface to support a function querySubstitutablePathInfos() that
queries multiple paths at the same time, and rewriting queryMissing()
to take advantage of parallelism. (Due to local caching,
parallelising queryMissing() is sufficient for most use cases, since
it's almost always called before building a derivation and thus fills
the local info cache.)
For example, parallelism speeds up querying all 1056 paths in a
particular NixOS system configuration from 116s to 2.6s. It works so
well because the eccentricity of the top-level derivation in the
dependency graph is only 9. So we only need 10 round-trips (when
using an unlimited number of parallel connections) to get everything.
Currently we do a maximum of 150 parallel connections to the server.
Thus it's important that the binary cache server (e.g. nixos.org) has
a high connection limit. Alternatively we could use HTTP pipelining,
but WWW::Curl doesn't support it and libcurl has a hard-coded limit of
5 requests per pipeline.
2012-07-06 23:08:20 +00:00
|
|
|
}
|
2008-08-04 11:44:50 +00:00
|
|
|
}
|
download-from-binary-cache: parallelise fetching of NAR info files
Getting substitute information using the binary cache substituter has
non-trivial latency overhead. A package or NixOS system configuration
can have hundreds of dependencies, and in the worst case (when the
local info cache is empty) we have to do a separate HTTP request for
each of these. If the ping time to the server is t, getting N info
files will take tN seconds; e.g., with a ping time of 0.1s to
nixos.org, sequentially downloading 1000 info files (a typical NixOS
config) will take at least 100 seconds.
To fix this problem, the binary cache substituter can now perform
requests in parallel. This required changing the substituter
interface to support a function querySubstitutablePathInfos() that
queries multiple paths at the same time, and rewriting queryMissing()
to take advantage of parallelism. (Due to local caching,
parallelising queryMissing() is sufficient for most use cases, since
it's almost always called before building a derivation and thus fills
the local info cache.)
For example, parallelism speeds up querying all 1056 paths in a
particular NixOS system configuration from 116s to 2.6s. It works so
well because the eccentricity of the top-level derivation in the
dependency graph is only 9. So we only need 10 round-trips (when
using an unlimited number of parallel connections) to get everything.
Currently we do a maximum of 150 parallel connections to the server.
Thus it's important that the binary cache server (e.g. nixos.org) has
a high connection limit. Alternatively we could use HTTP pipelining,
but WWW::Curl doesn't support it and libcurl has a hard-coded limit of
5 requests per pipeline.
2012-07-06 23:08:20 +00:00
|
|
|
}
|
2016-07-21 15:40:40 +00:00
|
|
|
};
|
download-from-binary-cache: parallelise fetching of NAR info files
Getting substitute information using the binary cache substituter has
non-trivial latency overhead. A package or NixOS system configuration
can have hundreds of dependencies, and in the worst case (when the
local info cache is empty) we have to do a separate HTTP request for
each of these. If the ping time to the server is t, getting N info
files will take tN seconds; e.g., with a ping time of 0.1s to
nixos.org, sequentially downloading 1000 info files (a typical NixOS
config) will take at least 100 seconds.
To fix this problem, the binary cache substituter can now perform
requests in parallel. This required changing the substituter
interface to support a function querySubstitutablePathInfos() that
queries multiple paths at the same time, and rewriting queryMissing()
to take advantage of parallelism. (Due to local caching,
parallelising queryMissing() is sufficient for most use cases, since
it's almost always called before building a derivation and thus fills
the local info cache.)
For example, parallelism speeds up querying all 1056 paths in a
particular NixOS system configuration from 116s to 2.6s. It works so
well because the eccentricity of the top-level derivation in the
dependency graph is only 9. So we only need 10 round-trips (when
using an unlimited number of parallel connections) to get everything.
Currently we do a maximum of 150 parallel connections to the server.
Thus it's important that the binary cache server (e.g. nixos.org) has
a high connection limit. Alternatively we could use HTTP pipelining,
but WWW::Curl doesn't support it and libcurl has a hard-coded limit of
5 requests per pipeline.
2012-07-06 23:08:20 +00:00
|
|
|
|
2021-04-05 13:48:18 +00:00
|
|
|
doPath = [&](const DerivedPath & req) {
|
2012-07-30 23:55:41 +00:00
|
|
|
|
2016-07-21 15:40:40 +00:00
|
|
|
{
|
|
|
|
auto state(state_.lock());
|
2021-04-05 13:24:42 +00:00
|
|
|
if (!state->done.insert(req.to_string(*this)).second) return;
|
2016-07-21 15:40:40 +00:00
|
|
|
}
|
download-from-binary-cache: parallelise fetching of NAR info files
Getting substitute information using the binary cache substituter has
non-trivial latency overhead. A package or NixOS system configuration
can have hundreds of dependencies, and in the worst case (when the
local info cache is empty) we have to do a separate HTTP request for
each of these. If the ping time to the server is t, getting N info
files will take tN seconds; e.g., with a ping time of 0.1s to
nixos.org, sequentially downloading 1000 info files (a typical NixOS
config) will take at least 100 seconds.
To fix this problem, the binary cache substituter can now perform
requests in parallel. This required changing the substituter
interface to support a function querySubstitutablePathInfos() that
queries multiple paths at the same time, and rewriting queryMissing()
to take advantage of parallelism. (Due to local caching,
parallelising queryMissing() is sufficient for most use cases, since
it's almost always called before building a derivation and thus fills
the local info cache.)
For example, parallelism speeds up querying all 1056 paths in a
particular NixOS system configuration from 116s to 2.6s. It works so
well because the eccentricity of the top-level derivation in the
dependency graph is only 9. So we only need 10 round-trips (when
using an unlimited number of parallel connections) to get everything.
Currently we do a maximum of 150 parallel connections to the server.
Thus it's important that the binary cache server (e.g. nixos.org) has
a high connection limit. Alternatively we could use HTTP pipelining,
but WWW::Curl doesn't support it and libcurl has a hard-coded limit of
5 requests per pipeline.
2012-07-06 23:08:20 +00:00
|
|
|
|
2021-03-02 03:50:41 +00:00
|
|
|
std::visit(overloaded {
|
2021-09-30 21:31:21 +00:00
|
|
|
[&](const DerivedPath::Built & bfd) {
|
2021-03-02 03:50:41 +00:00
|
|
|
if (!isValidPath(bfd.drvPath)) {
|
2016-07-21 15:40:40 +00:00
|
|
|
// FIXME: we could try to substitute the derivation.
|
|
|
|
auto state(state_.lock());
|
2021-03-02 03:50:41 +00:00
|
|
|
state->unknown.insert(bfd.drvPath);
|
2016-07-21 15:40:40 +00:00
|
|
|
return;
|
|
|
|
}
|
2012-11-26 16:15:09 +00:00
|
|
|
|
2021-03-05 00:49:46 +00:00
|
|
|
StorePathSet invalid;
|
2020-08-07 19:09:26 +00:00
|
|
|
/* true for regular derivations, and CA derivations for which we
|
|
|
|
have a trust mapping for all wanted outputs. */
|
|
|
|
auto knownOutputPaths = true;
|
2021-03-02 03:50:41 +00:00
|
|
|
for (auto & [outputName, pathOpt] : queryPartialDerivationOutputMap(bfd.drvPath)) {
|
2020-08-07 19:09:26 +00:00
|
|
|
if (!pathOpt) {
|
|
|
|
knownOutputPaths = false;
|
|
|
|
break;
|
|
|
|
}
|
2023-01-11 23:57:18 +00:00
|
|
|
if (bfd.outputs.contains(outputName) && !isValidPath(*pathOpt))
|
2021-03-05 00:49:46 +00:00
|
|
|
invalid.insert(*pathOpt);
|
2020-08-07 19:09:26 +00:00
|
|
|
}
|
|
|
|
if (knownOutputPaths && invalid.empty()) return;
|
|
|
|
|
2021-03-02 03:50:41 +00:00
|
|
|
auto drv = make_ref<Derivation>(derivationFromPath(bfd.drvPath));
|
|
|
|
ParsedDerivation parsedDrv(StorePath(bfd.drvPath), *drv);
|
2006-03-06 11:21:15 +00:00
|
|
|
|
2020-08-07 19:09:26 +00:00
|
|
|
if (knownOutputPaths && settings.useSubstitutes && parsedDrv.substitutesAllowed()) {
|
2016-07-21 15:40:40 +00:00
|
|
|
auto drvState = make_ref<Sync<DrvState>>(DrvState(invalid.size()));
|
|
|
|
for (auto & output : invalid)
|
2021-03-02 03:50:41 +00:00
|
|
|
pool.enqueue(std::bind(checkOutput, bfd.drvPath, drv, output, drvState));
|
download-from-binary-cache: parallelise fetching of NAR info files
Getting substitute information using the binary cache substituter has
non-trivial latency overhead. A package or NixOS system configuration
can have hundreds of dependencies, and in the worst case (when the
local info cache is empty) we have to do a separate HTTP request for
each of these. If the ping time to the server is t, getting N info
files will take tN seconds; e.g., with a ping time of 0.1s to
nixos.org, sequentially downloading 1000 info files (a typical NixOS
config) will take at least 100 seconds.
To fix this problem, the binary cache substituter can now perform
requests in parallel. This required changing the substituter
interface to support a function querySubstitutablePathInfos() that
queries multiple paths at the same time, and rewriting queryMissing()
to take advantage of parallelism. (Due to local caching,
parallelising queryMissing() is sufficient for most use cases, since
it's almost always called before building a derivation and thus fills
the local info cache.)
For example, parallelism speeds up querying all 1056 paths in a
particular NixOS system configuration from 116s to 2.6s. It works so
well because the eccentricity of the top-level derivation in the
dependency graph is only 9. So we only need 10 round-trips (when
using an unlimited number of parallel connections) to get everything.
Currently we do a maximum of 150 parallel connections to the server.
Thus it's important that the binary cache server (e.g. nixos.org) has
a high connection limit. Alternatively we could use HTTP pipelining,
but WWW::Curl doesn't support it and libcurl has a hard-coded limit of
5 requests per pipeline.
2012-07-06 23:08:20 +00:00
|
|
|
} else
|
2021-03-02 03:50:41 +00:00
|
|
|
mustBuildDrv(bfd.drvPath, *drv);
|
2006-03-06 11:21:15 +00:00
|
|
|
|
2021-03-02 03:50:41 +00:00
|
|
|
},
|
2021-09-30 21:31:21 +00:00
|
|
|
[&](const DerivedPath::Opaque & bo) {
|
2012-07-30 23:55:41 +00:00
|
|
|
|
2021-03-02 03:50:41 +00:00
|
|
|
if (isValidPath(bo.path)) return;
|
2016-07-21 15:40:40 +00:00
|
|
|
|
|
|
|
SubstitutablePathInfos infos;
|
2021-03-02 03:50:41 +00:00
|
|
|
querySubstitutablePathInfos({{bo.path, std::nullopt}}, infos);
|
2016-07-21 15:40:40 +00:00
|
|
|
|
|
|
|
if (infos.empty()) {
|
|
|
|
auto state(state_.lock());
|
2021-03-02 03:50:41 +00:00
|
|
|
state->unknown.insert(bo.path);
|
2016-07-21 15:40:40 +00:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2021-03-02 03:50:41 +00:00
|
|
|
auto info = infos.find(bo.path);
|
2016-07-21 15:40:40 +00:00
|
|
|
assert(info != infos.end());
|
|
|
|
|
|
|
|
{
|
|
|
|
auto state(state_.lock());
|
2021-03-02 03:50:41 +00:00
|
|
|
state->willSubstitute.insert(bo.path);
|
2016-07-21 15:40:40 +00:00
|
|
|
state->downloadSize += info->second.downloadSize;
|
|
|
|
state->narSize += info->second.narSize;
|
|
|
|
}
|
|
|
|
|
2023-01-14 21:38:43 +00:00
|
|
|
for (auto & ref : info->second.references)
|
2021-04-05 13:48:18 +00:00
|
|
|
pool.enqueue(std::bind(doPath, DerivedPath::Opaque { ref }));
|
2021-03-02 03:50:41 +00:00
|
|
|
},
|
2021-04-05 13:24:42 +00:00
|
|
|
}, req.raw());
|
2016-07-21 15:40:40 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
for (auto & path : targets)
|
|
|
|
pool.enqueue(std::bind(doPath, path));
|
|
|
|
|
|
|
|
pool.process();
|
2006-03-06 11:21:15 +00:00
|
|
|
}
|
2006-09-04 21:06:23 +00:00
|
|
|
|
2012-07-30 23:55:41 +00:00
|
|
|
|
2019-12-05 18:11:09 +00:00
|
|
|
StorePaths Store::topoSortPaths(const StorePathSet & paths)
|
2011-12-30 15:02:50 +00:00
|
|
|
{
|
2020-07-27 20:45:34 +00:00
|
|
|
return topoSort(paths,
|
|
|
|
{[&](const StorePath & path) {
|
|
|
|
try {
|
2023-01-14 21:38:43 +00:00
|
|
|
return queryPathInfo(path)->references;
|
2020-07-27 20:45:34 +00:00
|
|
|
} catch (InvalidPath &) {
|
2021-10-14 08:04:13 +00:00
|
|
|
return StorePathSet();
|
2020-07-27 20:45:34 +00:00
|
|
|
}
|
|
|
|
}},
|
|
|
|
{[&](const StorePath & path, const StorePath & parent) {
|
|
|
|
return BuildError(
|
|
|
|
"cycle detected in the references of '%s' from '%s'",
|
|
|
|
printStorePath(path),
|
|
|
|
printStorePath(parent));
|
|
|
|
}});
|
2011-12-30 15:02:50 +00:00
|
|
|
}
|
|
|
|
|
2021-05-26 14:09:02 +00:00
|
|
|
std::map<DrvOutput, StorePath> drvOutputReferences(
|
2021-06-21 14:37:45 +00:00
|
|
|
const std::set<Realisation> & inputRealisations,
|
|
|
|
const StorePathSet & pathReferences)
|
2021-05-19 08:35:31 +00:00
|
|
|
{
|
2021-05-26 14:09:02 +00:00
|
|
|
std::map<DrvOutput, StorePath> res;
|
2011-12-30 15:02:50 +00:00
|
|
|
|
2021-05-26 14:09:02 +00:00
|
|
|
for (const auto & input : inputRealisations) {
|
|
|
|
if (pathReferences.count(input.outPath)) {
|
|
|
|
res.insert({input.id, input.outPath});
|
|
|
|
}
|
2021-05-19 08:35:31 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
return res;
|
|
|
|
}
|
2011-12-30 15:02:50 +00:00
|
|
|
|
2021-05-26 14:09:02 +00:00
|
|
|
std::map<DrvOutput, StorePath> drvOutputReferences(
|
2021-05-19 08:35:31 +00:00
|
|
|
Store & store,
|
|
|
|
const Derivation & drv,
|
|
|
|
const StorePath & outputPath)
|
|
|
|
{
|
|
|
|
std::set<Realisation> inputRealisations;
|
|
|
|
|
2022-03-11 12:23:23 +00:00
|
|
|
for (const auto & [inputDrv, outputNames] : drv.inputDrvs) {
|
2022-05-04 05:44:32 +00:00
|
|
|
const auto outputHashes =
|
2021-05-19 08:35:31 +00:00
|
|
|
staticOutputHashes(store, store.readDerivation(inputDrv));
|
2022-03-11 12:23:23 +00:00
|
|
|
for (const auto & outputName : outputNames) {
|
2022-05-04 05:44:32 +00:00
|
|
|
auto outputHash = get(outputHashes, outputName);
|
|
|
|
if (!outputHash)
|
|
|
|
throw Error(
|
|
|
|
"output '%s' of derivation '%s' isn't realised", outputName,
|
|
|
|
store.printStorePath(inputDrv));
|
2021-05-19 08:35:31 +00:00
|
|
|
auto thisRealisation = store.queryRealisation(
|
2022-05-04 05:44:32 +00:00
|
|
|
DrvOutput{*outputHash, outputName});
|
2021-05-19 08:35:31 +00:00
|
|
|
if (!thisRealisation)
|
|
|
|
throw Error(
|
2022-03-11 12:23:23 +00:00
|
|
|
"output '%s' of derivation '%s' isn't built", outputName,
|
2021-05-19 08:35:31 +00:00
|
|
|
store.printStorePath(inputDrv));
|
|
|
|
inputRealisations.insert(*thisRealisation);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
auto info = store.queryPathInfo(outputPath);
|
|
|
|
|
2023-01-14 21:38:43 +00:00
|
|
|
return drvOutputReferences(Realisation::closure(store, inputRealisations), info->references);
|
2021-05-19 08:35:31 +00:00
|
|
|
}
|
2022-03-11 12:23:23 +00:00
|
|
|
|
2023-01-11 23:57:18 +00:00
|
|
|
OutputPathMap resolveDerivedPath(Store & store, const DerivedPath::Built & bfd, Store * evalStore_)
|
|
|
|
{
|
|
|
|
auto & evalStore = evalStore_ ? *evalStore_ : store;
|
|
|
|
|
|
|
|
OutputPathMap outputs;
|
|
|
|
auto drv = evalStore.readDerivation(bfd.drvPath);
|
|
|
|
auto outputHashes = staticOutputHashes(store, drv);
|
|
|
|
auto drvOutputs = drv.outputsAndOptPaths(store);
|
|
|
|
auto outputNames = std::visit(overloaded {
|
|
|
|
[&](const OutputsSpec::All &) {
|
|
|
|
StringSet names;
|
|
|
|
for (auto & [outputName, _] : drv.outputs)
|
|
|
|
names.insert(outputName);
|
|
|
|
return names;
|
|
|
|
},
|
|
|
|
[&](const OutputsSpec::Names & names) {
|
2023-01-11 16:54:43 +00:00
|
|
|
return static_cast<std::set<std::string>>(names);
|
2023-01-11 23:57:18 +00:00
|
|
|
},
|
2023-01-15 20:16:14 +00:00
|
|
|
}, bfd.outputs.raw());
|
2023-01-11 23:57:18 +00:00
|
|
|
for (auto & output : outputNames) {
|
|
|
|
auto outputHash = get(outputHashes, output);
|
|
|
|
if (!outputHash)
|
|
|
|
throw Error(
|
|
|
|
"the derivation '%s' doesn't have an output named '%s'",
|
|
|
|
store.printStorePath(bfd.drvPath), output);
|
2023-03-17 14:33:48 +00:00
|
|
|
if (experimentalFeatureSettings.isEnabled(Xp::CaDerivations)) {
|
2023-01-11 23:57:18 +00:00
|
|
|
DrvOutput outputId { *outputHash, output };
|
|
|
|
auto realisation = store.queryRealisation(outputId);
|
|
|
|
if (!realisation)
|
|
|
|
throw MissingRealisation(outputId);
|
|
|
|
outputs.insert_or_assign(output, realisation->outPath);
|
|
|
|
} else {
|
|
|
|
// If ca-derivations isn't enabled, assume that
|
|
|
|
// the output path is statically known.
|
|
|
|
auto drvOutput = get(drvOutputs, output);
|
|
|
|
assert(drvOutput);
|
|
|
|
assert(drvOutput->second);
|
|
|
|
outputs.insert_or_assign(output, *drvOutput->second);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return outputs;
|
2021-05-19 08:35:31 +00:00
|
|
|
}
|
2022-03-11 12:23:23 +00:00
|
|
|
|
2006-09-04 21:06:23 +00:00
|
|
|
}
|