about summary refs log tree commit diff
path: root/src
AgeCommit message (Collapse)AuthorFilesLines
2009-03-22 * NAR archives: handle files larger than 2^32 bytes. Previously itEelco Dolstra6-17/+47
would just silently store only (fileSize % 2^32) bytes. * Use posix_fallocate if available when unpacking archives. * Provide a better error message when trying to unpack something that isn't a NAR archive.
2009-03-18 * Unify exportReferencesGraph and exportBuildReferencesGraph, and makeEelco Dolstra2-60/+36
sure that it works as expected when you pass it a derivation. That is, we have to make sure that all build-time dependencies are built, and that they are all in the input closure (otherwise remote builds might fail, for example). This is ensured at instantiation time by adding all derivations and their sources to inputDrvs and inputSrcs.
2009-03-18 * Acquire the locks on the output paths before trying to run the buildEelco Dolstra1-145/+71
hook. This fixes a problem with log files being partially or completely filled with 0's because another nix-store process truncates the log file. It should also be more efficient.
2009-03-17 * Cleanup.Eelco Dolstra1-22/+9
2009-03-06 (no commit message)Eelco Dolstra1-1/+1
2009-03-05 * Install some headers in the right location.Eelco Dolstra2-2/+6
2009-02-27 * Allow options from the Nix config file to be overriden from theEelco Dolstra3-0/+21
command line (e.g. "--option build-use-chroot true").
2009-02-16 * Release output locks as soon as possible, not when the destructor ofEelco Dolstra3-0/+14
the DerivationGoal runs. Otherwise, if a goal is a top-level goal, then the lock won't be released until nix-store finishes. With --keep-going and lots of top-level goals, it's possible to run out of file descriptors (this happened sometimes in the build farm for Nixpkgs). Also, for failed derivation, it won't be possible to build it again until the lock is released. * Idem for locks on build users: these weren't released in a timely manner for failed top-level derivation goals. So if there were more than (say) 10 such failed builds, you would get an error about having run out of build users.
2009-02-05 added primop functions __isBool, __isString, __isIntMarc Weber1-0/+25
2009-02-02 * Build hooks: use nix-store --import. This prevents a redundantEelco Dolstra2-9/+27
scan for runtime dependencies (i.e. the local machine shouldn't do a scan that the remote machine has already done). Also pipe directly into `nix-store --import': don't use a temporary file.
2009-01-27 Add the "addErrorContext" builtin to add more information in the stack trace.Nicolas Pierron1-0/+14
2009-01-27 * Make it compile on Debian 4.0 (which doesn't define PER_LINUX32_3GBEelco Dolstra1-1/+1
in sys/personality.h).
2009-01-13 * When using a build hook, distinguish between transient failuresEelco Dolstra1-4/+20
(e.g. an SSH connection problem) and permanent failures (i.e. the builder failed). This matters to Hydra (it wants to know whether it makes sense to retry a build).
2009-01-12 * Support i686-linux builds directly on x86_64-linux NixEelco Dolstra1-1/+18
installations. This is implemented using the personality() syscall, which causes uname to return "i686" in child processes.
2009-01-12 * Make Nix build with Bison 2.4.Eelco Dolstra1-0/+1
2008-12-16 * nix-store --verify: repair bad hash fields in the metadata file.Eelco Dolstra2-5/+16
2008-12-16 * Pass --use-atime / --max-atime to the daemon.Eelco Dolstra3-1/+9
2008-12-15 * Delete the chroot if it already exists (e.g. left over from anEelco Dolstra1-0/+1
interrupted build).
2008-12-12 * Put chroots in the Nix store. This ensures that we can create hardEelco Dolstra2-2/+14
links to the inputs.
2008-12-12 * Simplify deleting .lock files in /nix/store: just don't delete themEelco Dolstra2-31/+19
if they belong a path that's currently being built. This gets rid of some Cygwin-specific code.
2008-12-12 * Detect whether unshare() is available.Eelco Dolstra1-4/+1
2008-12-12 * We can't use string objects in signal handlers because they mightEelco Dolstra1-4/+4
allocate memory, which is verboten in signal handlers. This caused random failures in the test suite on Mac OS X (triggered by the spurious SIGPOLL signals on Mac OS X, which should also be fixed).
2008-12-12 * Use a PathSet for the chroot directories so that we don'tEelco Dolstra1-6/+7
accidentally bind-mount a directory twice.
2008-12-12 * Define _GNU_SOURCE. Hopefully this fixes the build on Debian 4.0Eelco Dolstra1-0/+3
(http://hydra.nixos.org/nixlog/384/1).
2008-12-11 * Don't provide the whole Nix store in the chroot, but only theEelco Dolstra1-10/+44
closure of the inputs. This really enforces that there can't be any undeclared dependencies on paths in the store. This is done by creating a fake Nix store and creating bind-mounts or hard-links in the fake store for all paths in the closure. After the build, the build output is moved from the fake store to the real store. TODO: the chroot has to be on the same filesystem as the Nix store for this to work, but this isn't enforced yet. (I.e. it only works currently if /tmp is on the same FS as /nix/store.)
2008-12-11 * Revert r13150: now that we use private namespaces for the chroot, weEelco Dolstra4-9/+1
don't have to put the chroot in /nix/var/nix/chroots anymore. They're back in /tmp now.
2008-12-11 * Provide a minimal /etc/passwd in the chroot to keep some buildersEelco Dolstra1-1/+12
happy.
2008-12-11 * Do chroot builds in a private namespace. This means that all theEelco Dolstra1-134/+45
bind-mounts we do are only visible to the builder process and its children. So accidentally doing "rm -rf" on the chroot directory won't wipe out /nix/store and other bind-mounted directories anymore. Also, the bind-mounts in the private namespace disappear automatically when the builder exits.
2008-12-11 * Open the connection to the daemon lazily (on demand) so thatEelco Dolstra3-6/+38
read-only operations (like nix-env -qa) work properly when the daemon isn't running.
2008-12-04 * Build on Debian 4.0 (GCC 4.1.1).Eelco Dolstra1-1/+1
2008-12-04 * Propagate --max-silent-time to remote machines.Eelco Dolstra1-0/+1
2008-12-04 * When using build hooks, for any nix-store -r build operation, it isEelco Dolstra1-1/+12
necessary that at least one build hook doesn't return "postpone", otherwise nix-store will barf ("waiting for a build slot, yet there are no running children"). So inform the build hook when this is the case, so that it can start a build even when that would exceed the maximum load on a machine.
2008-12-04 * Don't ignore errors from writing to stderr. That way, whenEelco Dolstra1-5/+1
nix-store -r (or some other operation) is started via ssh, it will at least have a chance of terminating quickly when the connection is killed. Right now it just runs to completion, because it never notices that stderr is no longer connected to anything. Of course it would be better if sshd would just send a SIGHUP, but it doesn't (https://bugzilla.mindrot.org/show_bug.cgi?id=396).
2008-12-04 (no commit message)Eelco Dolstra1-3/+4
2008-12-04 * Dirty hack to make nix-push work properly on derivations: theEelco Dolstra1-8/+23
derivation should be a source rather than a derivation dependency of the call to the NAR derivation. Otherwise the derivation (and all its dependencies) will be built as a side-effect, which may not even succeed.
2008-12-03 * addToStore() in nix-worker: don't write the NAR dump received fromEelco Dolstra3-25/+85
the client to a temporary directory, as that is highly inefficient.
2008-12-03 * A simple API for parsing NAR archives.Eelco Dolstra3-26/+80
2008-12-03 * Backwards compatibility.Eelco Dolstra1-4/+9
2008-12-03 * Pass HashType values instead of strings.Eelco Dolstra11-23/+36
2008-12-03 * Made addToStore() a lot more efficient: it no longer reads the pathEelco Dolstra2-35/+29
being copied 3 times in the worst case. It doesn't run in constant space, but it didn't do that anyway.
2008-12-03 * Unify the treatment of sources copied to the store, and recursiveEelco Dolstra9-72/+140
SHA-256 outputs of fixed-output derivations. I.e. they now produce the same store path: $ nix-store --add x /nix/store/j2fq9qxvvxgqymvpszhs773ncci45xsj-x $ nix-store --add-fixed --recursive sha256 x /nix/store/j2fq9qxvvxgqymvpszhs773ncci45xsj-x the latter being the same as the path that a derivation derivation { name = "x"; outputHashAlgo = "sha256"; outputHashMode = "recursive"; outputHash = "..."; ... }; produces. This does change the output path for such fixed-output derivations. Fortunately they are quite rare. The most common use is fetchsvn calls with SHA-256 hashes. (There are a handful of those is Nixpkgs, mostly unstable development packages.) * Documented the computation of store paths (in store-api.cc).
2008-11-25 * Ignore carriage returns.Eelco Dolstra1-0/+1
2008-11-25 * Handle prematurely ended logfiles, i.e. make sure we emit enoughEelco Dolstra1-5/+17
close tags.
2008-11-20 * Nix daemon: reload the configuration file after forking (NIX-100).Eelco Dolstra3-1/+14
2008-11-19 * Primop builtins.storePath for declaring a store path as aEelco Dolstra1-0/+22
dependency. `storePath /nix/store/bla' gives exactly the same result as `toPath /nix/store/bla', except that the former includes /nix/store/bla in the dependency context of the string. Useful in some generated Nix expressions like nix-push, which now finally does the right thing wrt distributed builds. (Previously the path to be packed wasn't an explicit dependency, so it wouldn't be copied to the remote machine.)
2008-11-19 * Files in the info directory starting with "." are temporary filesEelco Dolstra1-2/+5
and don't indicate path validity.
2008-11-19 * nix-store --dump-db: be more streamy.Eelco Dolstra1-3/+3
2008-11-14 * Prevent zombies. Previous the SIGCHLD handler only reaped oneEelco Dolstra1-1/+1
zombie at a time, so if multiple children died before the handler got to run, some of them would not be cleaned up.
2008-11-14 * Restore SIGPIPE to SIG_DFL when running the builder. This preventsEelco Dolstra1-0/+13
subtle and often hard-to-reproduce bugs where programs in pipes either barf with a "Broken pipe" message or not, depending on the exact timing conditions. This particularly happened in GNU M4 (and Bison, which uses M4).
2008-11-12 * Some somewhat ad hoc mechanism to allow the build farm to monitorEelco Dolstra6-7/+71
build progress.