Commit graph

6716 commits

Author SHA1 Message Date
Eelco Dolstra ef09da58f2
nix path-info: Hopefully fix macOS build failure
https://hydra.nixos.org/build/80480356
2018-08-31 17:30:45 +02:00
Eelco Dolstra e3731a1a1f
Merge pull request #2387 from graham-at-target/add-ids-to-important-sections
docs: Add IDs to important sections
2018-08-31 17:24:38 +02:00
Graham Christensen 149d10c308 docs: Add IDs to important sections 2018-08-31 10:06:33 -04:00
Graham Christensen 2df21b78b9 docs: Add some examples to fetchGit 2018-08-31 10:00:32 -04:00
Eelco Dolstra c0c31b58a4
Add localhost alias for ::1 to the sandbox 2018-08-31 11:12:51 +02:00
Eelco Dolstra adab8b916a
Merge pull request #2383 from cleverca22/fix-error-9
fix `error: unknown serve command 9`
2018-08-31 11:07:16 +02:00
Michael Bishop 0767e402f1 fix error: unknown serve command 9 2018-08-30 20:10:08 -03:00
Eelco Dolstra 39f1722f36
nix upgrade-nix: Handle .nix-profile being a link to .../profiles/per-user/...
Also some cosmetic improvements.
2018-08-30 21:18:56 +02:00
Eelco Dolstra f08b14c9d0
nix upgrade-nix: Improve error message if 'nix' is not in a profile
E.g.

  $ nix upgrade-nix
  error: directory '/home/eelco/Dev/nix/inst/bin' does not appear to be part of a Nix profile

instead of

  $ nix upgrade-nix
  error: '/home/eelco/Dev/nix/inst' is not a symlink
2018-08-30 17:14:17 +02:00
Eelco Dolstra ddeda0b62e
Merge branch 'nix-upgrade-profile' of https://github.com/LnL7/nix 2018-08-30 17:10:59 +02:00
Eelco Dolstra 264e66f696
Add simple test for nix upgrade-nix 2018-08-30 17:10:28 +02:00
Eelco Dolstra 308689f94b
Merge pull request #2365 from erikarvstedt/fix-docs
Docs: Fix install prefix
2018-08-30 16:24:52 +02:00
Eelco Dolstra 99828245f8
printSize() fixes
Fix a 32-bit overflow that resulted in negative numbers being printed;
use fmt() instead of boost::format(); change -H to -h for consistency
with 'ls' and 'du'; make the columns narrower (since they can't be
bigger than 1024.0).
2018-08-30 16:22:04 +02:00
Benjamin Hipple c908df881f
Avoid overflow and use boost::format
If the user has an object greater than 1024 yottabytes, it'll just display it as
N yottabytes instead of overflowing.

Swaps to use boost::format strings instead of std::setw and std::setprecision.
2018-08-30 16:22:04 +02:00
Benjamin Hipple 3407a5d936
Add human readable closure sizes to nix path-info
Unfortunately, -h is already taken as a short option by --help, so we have to
use a different letter or the capitalized version.

Resolves #2363
2018-08-30 16:22:04 +02:00
Eelco Dolstra 5e83b0227f
nix: Remove the -h flag 2018-08-30 16:14:19 +02:00
Eelco Dolstra 145db703e5
Remove boost from the closure
This reduces the size of the closure by 45 MiB.
2018-08-30 16:03:06 +02:00
Eelco Dolstra 64d7d1a884
Update release notes 2018-08-30 15:28:18 +02:00
Eelco Dolstra 245d01701d
upload-release.pl: Handle prerelease versions 2018-08-30 15:21:58 +02:00
Eelco Dolstra 1286d86b95
Merge pull request #2336 from ivan/repl-trailing-spaces
repl: don't add trailing spaces to history lines
2018-08-29 18:13:29 +02:00
Domen Kožar d16ff76c69
nix.conf: clarify -j0 doesn't affect preferLocalBuild 2018-08-29 14:33:14 +01:00
Eelco Dolstra 20d74a3257
Merge pull request #2378 from aszlig/int64
libexpr: Use int64_t for NixInt
2018-08-29 13:30:50 +02:00
Domen Kožar 54df4bb0b5
nix.conf: mention -j0 is useful 2018-08-29 11:51:51 +01:00
aszlig 0ad643ed5c
libexpr: Use int64_t for NixInt
Using a 64bit integer on 32bit systems will come with a bit of a
performance overhead, but given that Nix doesn't use a lot of integers
compared to other types, I think the overhead is negligible also
considering that 32bit systems are in decline.

The biggest advantage however is that when we use a consistent integer
size across all platforms it's less likely that we miss things that we
break due to that. One example would be:

https://github.com/NixOS/nixpkgs/pull/44233

On Hydra it will evaluate, because the evaluator runs on a 64bit
machine, but when evaluating the same on a 32bit machine it will fail,
so using 64bit integers should make that consistent.

While the change of the type in value.hh is rather easy to do, we have a
few more options available for doing the conversion in the lexer:

  * Via an #ifdef on the architecture and using strtol() or strtoll()
    accordingly depending on which architecture we are. For the #ifdef
    we would need another AX_COMPILE_CHECK_SIZEOF in configure.ac.
  * Using istringstream, which would involve copying the value.
  * As we're already using boost, lexical_cast might be a good idea.

Spoiler: I went for the latter, first of all because lexical_cast does
have an overload for const char* and second of all, because it doesn't
involve copying around the input string. Also, because istringstream
seems to come with a bigger overhead than boost::lexical_cast:

https://www.boost.org/doc/libs/release/doc/html/boost_lexical_cast/performance.html

The first method (still using strtol/strtoll) also wasn't something I
pursued further, because it is also locale-aware which I doubt is what
we want, given that the regex for int is [0-9]+.

Signed-off-by: aszlig <aszlig@nix.build>
Fixes: #2339
2018-08-29 01:05:52 +02:00
Daiderd Jordan d85bb4814f
upgrade-nix: resolve profile symlinks
The profile present in PATH is not necessarily the actual profile
location. User profiles are generally added as $HOME/.nix-profile
in which case the indirect profile link needs to be resolved first.

/home/user/.nix-profile -> /nix/var/nix/profiles/per-user/user/profile
/nix/var/nix/profiles/per-user/user/profile -> profile-15-link
/nix/var/nix/profiles/per-user/user/profile-14-link -> /nix/store/hyi4kkjh3bwi2z3wfljrkfymz9904h62-user-environment
/nix/var/nix/profiles/per-user/user/profile-15-link -> /nix/store/6njpl3qvihz46vj911pwx7hfcvwhifl9-user-environment

To upgrade nix here we want /nix/var/nix/profiles/per-user/user/profile-16-link
instead of /home/user/.nix-profile-1-link. The latter is not a gcroot
and would be garbage collected, resulting in a broken profile.

Fixes #2175
2018-08-25 20:50:44 +02:00
Daiderd Jordan 414397759a
upgrade-nix: add --dry-run 2018-08-25 20:25:43 +02:00
Eelco Dolstra c651b7bdc9
Revert "Fix parser/lexer generation with parallel make"
This reverts commit d277442df5.

Make sucks.
2018-08-23 00:23:35 +02:00
Eelco Dolstra 1b01954a3d
Merge pull request #2368 from dezgeg/drop-dead-code
Drop dead code
2018-08-22 21:32:49 +02:00
Erik Arvstedt 8ad2defdf0 Docs: Fix install prefix
shell.nix defines the install prefix as $(pwd)/inst
2018-08-21 21:46:43 +02:00
Eelco Dolstra 954d1f4d0a
Disable the Ubuntu 17.10 build
Ubuntu 17.10 doesn't have libbrotli.

https://hydra.nixos.org/build/79867741
2018-08-21 15:33:29 +02:00
Eelco Dolstra ebe3d2d370 Improve 'coroutine has finished' error message 2018-08-21 15:22:04 +02:00
Eelco Dolstra 6317c65937 Fix warnings in compression.cc 2018-08-21 15:20:23 +02:00
Eelco Dolstra 17a92dfb7d Fix another 'coroutine has finished' during decompression
https://hydra.nixos.org/build/79867739
2018-08-21 15:19:20 +02:00
Tuomas Tynkkynen 2894197de7 Drop all references to NIX_INDENT_MAKE
Dead code since 6669a3b477
2018-08-20 02:01:59 +03:00
Tuomas Tynkkynen 458282be59 Drop all references to --disable-init-state
It's all dead code since 2014 (commit 0c6d62cf27).
2018-08-20 01:51:23 +03:00
Eelco Dolstra cc7b4386b1
nix run: Restore CPU affinity
Fixes #2359.
2018-08-19 12:05:08 +02:00
Eelco Dolstra 9b1bdf2db8
FIx floating point evaluation
Fixes #2361.
2018-08-19 11:59:49 +02:00
Graham Christensen 02098d2073 fetchGit: use a better caching scheme
The current usage technically works by putting multiple different
repos in to the same git directory. However, it is very slow as
Git tries very hard to find common commits between the two
repositories. If the two repositories are large (like Nixpkgs and
another long-running project,) it is maddeningly slow.

This change busts the cache for existing deployments, but users
will be promptly repaid in per-repository performance.
2018-08-17 11:27:34 -04:00
Eelco Dolstra d277442df5
Fix parser/lexer generation with parallel make
Fun fact: rules with multiple targets don't work properly with 'make
-j'. For example, a rule like

  a b: c
    touch a b

is equivalent to

  a: c
    touch a b

  b: c
    touch a b

so with 'make -j', the 'touch' command will be run twice. See
e.g. https://stackoverflow.com/questions/2973445/gnu-makefile-rule-generating-a-few-targets-from-a-single-source-file.
2018-08-17 12:59:23 +02:00
Eelco Dolstra 19265ed26c
Merge pull request #2351 from volth/update-config
update config/config.{sub,guess}
2018-08-13 22:24:22 +02:00
Eelco Dolstra 746cf2d27e
Merge pull request #2350 from symphorien/xattr-cifs
ignore when listxattr fails with ENODATA
2018-08-13 22:20:04 +02:00
volth bad27dc475 update config/config.{sub,guess}
Just
curl 'http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.sub;hb=HEAD'   > config/config.sub
curl 'http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.guess;hb=HEAD' > config/config.guess

Those files are 5 years old and failed to guess new archs ("ppc64-linux")
2018-08-13 20:00:17 +00:00
Symphorien Gibol 5b19a6663b ignore when listxattr fails with ENODATA
This happens on CIFS and means the remote filesystem has no extended
attributes.
2018-08-13 21:11:36 +02:00
Linus Heckemann d7402c9cd5 dirOf: allow use on non-absolute paths 2018-08-13 11:27:35 +02:00
Eelco Dolstra f72c907ad8
Slightly questionable workaround for #2342 2018-08-10 11:34:41 +02:00
Eelco Dolstra ada4e90267
S3BinaryCacheStore: Share TransferManager
TransferManager allocates a lot of memory (50 MiB by default), and it
might leak but I'm not sure about that. In any case it was causing
OOMs in hydra-queue-runner. So allocate only one TransferManager per
S3BinaryCacheStore.

Hopefully fixes https://github.com/NixOS/hydra/issues/586.
2018-08-09 20:48:45 +02:00
Eelco Dolstra c87f4b9324
nix run: Respect propagated-user-env-packages
Also, add $path/bin to $PATH even if it doesn't exist. This makes
'man' work properly (since it looks for ../share/man relative to $PATH
entries).
2018-08-09 13:01:03 +02:00
Eelco Dolstra a0b971dd9c
S3BinaryCacheStore: Don't use the transfer status callback
This callback is executed on a different thread, so exceptions thrown
from the callback are not caught:

  Aug 08 16:25:48 chef hydra-queue-runner[11967]: terminate called after throwing an instance of 'nix::Error'
  Aug 08 16:25:48 chef hydra-queue-runner[11967]:   what():  AWS error: failed to upload 's3://nix-cache/19dbddlfb0vp68g68y19p9fswrgl0bg7.ls'

Therefore, just check the transfer status after it completes. Also
include the S3 error message in the exception.
2018-08-08 21:39:11 +02:00
Ivan Kozik ec49ea28dc repl: don't add trailing spaces to history lines 2018-08-07 03:59:34 +00:00
Eelco Dolstra 103ad1598c
Merge pull request #2334 from dtzWill/fix/revert-progress-bar-refresh-for-now
Revert "progress-bar: re-draw last update if nothing new for 1sec."
2018-08-06 19:01:08 +02:00