lix -

Age	Commit message (Collapse)	Author
2010-12-08	* Ignore the result of sqlite3_reset().	Eelco Dolstra

2010-12-06	* `nix-store --verify --check-contents': don't hold the global GC lock	Eelco Dolstra
	while checking the contents, since this operation can take a very long time to finish. Also, fill in missing narSize fields in the DB while doing this.
2010-12-05	* Retry a transaction if SQLite returns SQLITE_BUSY. This can happen	Eelco Dolstra
	even with a very long busy timeout, because SQLITE_BUSY is also returned to resolve deadlocks. This should get rid of random "database is locked" errors. This is kind of hard to test though. * Fix a horrible bug in deleteFromStore(): deletePathWrapped() should be called after committing the transaction, not before, because the commit might not succeed.
2010-11-29	* Quick hack around one of the pathological cases in bsdiff	Eelco Dolstra
	(namely, when there is a long region of 0s). On one 86 MiB case, this cut patch generation time from 44m to 2m.
2010-11-25	* Sync with the trunk.	Eelco Dolstra

2010-11-25	* DrvInfo contains pointers to expressions, so DrvInfos should be	Eelco Dolstra
	traced by the garbage collector. Otherwise "nix-env -u" can crash randomly.
2010-11-17	* Before a build, show the disk space that the downloaded store paths	Eelco Dolstra
	will approximately require.
2010-11-17	* Add an operation `nix-store -q --size'.	Eelco Dolstra

2010-11-17	* Implement RemoteStore::queryPathInfo().	Eelco Dolstra

2010-11-16	* Store the size of a store path in the database (to be precise, the	Eelco Dolstra
	size of the NAR serialisation of the path, i.e., `nix-store --dump PATH'). This is useful for Hydra.
2010-11-16	* Sync with the trunk.	Eelco Dolstra

2010-10-29	* Document Boehm GC support.	Eelco Dolstra

2010-10-29	* Set libgc's initial heap size to 384 MiB to prevent garbage	Eelco Dolstra
	collection in most cases (and therefore its performance overhead).
2010-10-28	* new(UseGC) is inexplicably slower than GC_MALLOC, so prefer the	Eelco Dolstra
	latter.
2010-10-28	* Handle out of memory condition.	Eelco Dolstra

2010-10-24	* When allocating an attribute set, reserve enough space for all	Eelco Dolstra
	elements. This prevents the vector from having to resize itself.
2010-10-24	* Keep attribute sets in sorted order to speed up attribute lookups.	Eelco Dolstra
	* Simplify the representation of attributes in the AST. * Change the behaviour of listToAttrs() in case of duplicate names.
2010-10-24	* Don't create thunks for variable lookups (if possible). This	Eelco Dolstra
	significantly reduces the number of values allocated (e.g. from 8.7m to 4.9m for the Bittorrent test).
2010-10-24	* Store attribute sets as a vector instead of a map (i.e. a red-black	Eelco Dolstra
	tree). This saves a lot of memory. The vector should be sorted so that names can be looked up using binary search, but this is not the case yet. (Surprisingly, looking up attributes using linear search doesn't have a big impact on performance.) Memory consumption for $ nix-instantiate /etc/nixos/nixos/tests -A bittorrent.test --readonly-mode on x86_64-linux with GC enabled is now 185 MiB (compared to 946 MiB on the trunk).
2010-10-23	* Fix compiling without Boehm.	Eelco Dolstra
	* Fix the stats.
2010-10-23	* Optimise string constants by putting them in the symbol table.	Eelco Dolstra

2010-10-23	* Make Value smaller by not storing redundant PrimOp info.	Eelco Dolstra
	* Clear pointers in Values after overwriting them to make sure that no objects are kept alive unnecessarily.
2010-10-23	* Remove allocValues().	Eelco Dolstra

2010-10-22	* In environments, store pointers to values rather than values. This	Eelco Dolstra
	improves GC effectiveness a bit more (because a live value doesn't keep other values in the environment plus the parent environments alive), and removes the need for copy nodes.
2010-10-22	* Store Value nodes outside of attribute sets. I.e., Attr now stores	Eelco Dolstra
	a pointer to a Value, rather than the Value directly. This improves the effectiveness of garbage collection a lot: if the Value is stored inside the set directly, then any live pointer to the Value causes all other attributes in the set to be live as well.
2010-10-22	* Make building against the Boehm GC a configure option.	Eelco Dolstra

2010-10-20	* Keep some more stats.	Eelco Dolstra

2010-10-20	* Use the Boehm garbage collector to reclaim unused memory in the Nix	Eelco Dolstra
	expression evaluator.
2010-10-14	* Wrap deleteFromStore() in a transaction. Otherwise there might be a	Eelco Dolstra
	race with other processes that add new referrers to a path, resulting in the garbage collector crashing with "foreign key constraint failed". (Nix/4) * Make --gc --print-dead etc. interruptible.
2010-10-04	* Make sure that config.h is included before the system headers,	Eelco Dolstra
	because it defines _FILE_OFFSET_BITS. Without this, on OpenSolaris the system headers define it to be 32, and then the 32-bit stat() ends up being called with a 64-bit "struct stat", or vice versa. This also ensures that we get 64-bit file sizes everywhere. * Remove the redundant call to stat() in parseExprFromFile(). The file cannot be a symlink because that's the exit condition of the loop before.
2010-10-04	* If std::tr1::unordered_set is unavailable, use std::set.	Eelco Dolstra

2010-10-04	* Sync with the trunk.	Eelco Dolstra

2010-10-04	* In printMsg(), ignore failing writes to stderr if we're in an	Eelco Dolstra
	exception handler, otherwise throw an exception. We need to ignore write errors in exception handlers to ensure that cleanup code runs to completion if the other side of stderr has been closed unexpectedly.
2010-10-04	* Hack needed for GCC 4.3.2 on OpenSolaris.	Eelco Dolstra

2010-09-14	* Doh. Remove debug message.	Eelco Dolstra

2010-09-01	* Only do "pragma journal_mode = ..." if the current journal mode	Eelco Dolstra
	differs from the desired mode. There is an open SQLite ticket `Executing "PRAGMA journal_mode" may delete journal file while it is in use.'
2010-08-31	* Always print hook output on stderr, even if --no-build-output is	Eelco Dolstra
	set. * In the build hook, print a trace message to allow Hydra to pick up the name of the remote machine used for the build.
2010-08-31	`nix-store --verify' improvements:	Eelco Dolstra
	* If a path has disappeared, check its referrers first, and don't try to invalidate paths that have valid referrers. Otherwise we get a foreign key constraint violation. * Read the whole Nix store directory instead of statting each valid path, which is slower. * Acquire the global GC lock.
2010-08-30	* When using the build hook, distinguish between the stderr of the	Eelco Dolstra
	hook script proper, and the stdout/stderr of the builder. Only the latter should be saved in /nix/var/log/nix/drvs. * Allow the verbosity to be set through an option. * Added a flag --quiet to lower the verbosity level.
2010-08-30	* Disable the GC reachability check for now (when creating new roots),	Eelco Dolstra
	as it's hopelessly inefficient.
2010-08-27	* Experimental feature: allow a derivation to tell the build hook that	Eelco Dolstra
	it requires a certain feature on the build machine, e.g. requiredSystemFeatures = [ "kvm" ]; We need this in Hydra to make sure that builds that require KVM support are forwarded to machines that have KVM support. Probably this should also be enforced for local builds.
2010-08-27	* Oops - "null" was displayed as "true".	Eelco Dolstra

2010-08-25	* Made the build hook mechanism more efficient. Rather than starting	Eelco Dolstra
	the hook every time we want to ask whether we can run a remote build (which can be very often), we now reuse a hook process for answering those queries until it accepts a build. So if there are N derivations to be built, at most N hooks will be started.
2010-08-24	* Don't complain if the stored hash of a path is zero (unknown).	Eelco Dolstra

2010-08-24	* Handle the unlikely case where a derivation has no dependencies at	Eelco Dolstra
	all.
2010-08-24	* Sync with the trunk.	Eelco Dolstra

2010-08-12	* Make --cores work when building through the Nix daemon.	Eelco Dolstra

2010-08-04	* Sync with the trunk.	Eelco Dolstra

2010-08-04	* Use SQLite 3.7.0's write-ahead logging (WAL mode). This is a lot	Eelco Dolstra
	faster than the old mode when fsyncs are enabled, because it only performs an fsync() when doing a checkpoint, rather than at every commit. Some timings for doing a "nix-instantiate /etc/nixos/nixos -A system" after modifying the stdenv setup script: 42.5s - SQLite 3.6.23 with truncate mode and fsync 3.4s - SQLite 3.6.23 with truncate mode and no fsync 32.1s - SQLite 3.7.0 with truncate mode and fsync 16.8s - SQLite 3.7.0 with WAL mode and fsync, auto-checkpoint every 1000 pages 8.3s - SQLite 3.7.0 with WAL mode and fsync, auto-checkpoint every 8192 pages 1.7s - SQLite 3.7.0 with WAL mode and no fsync The default is now to use WAL mode with fsyncs. Because WAL doesn't work on remote filesystems such as NFS (as it uses shared memory), truncate mode can be re-enabled by setting the "use-sqlite-wal" option to false.
2010-08-04	* Remove "auto" and "guess" as synonyms for 0 in the handling of	Eelco Dolstra
	build-cores and --cores. They're superfluous and just complicate the parsing.