summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* fdio: Allow using AT_FDCWD with GlnxTmpfilewip/alexl/fix-tmpfilesAlexander Larsson2017-05-192-4/+2
| | | | | | | | We use fd >= 0 to check for validity instead of src_dfd >= 0 because everything works just fine with src_dfd == AT_FDCWD, except that is negative so the checks break. This fixes flatpak which uses AT_FDCWD.
* fdio: Add missing return in tmpfile error caseColin Walters2017-05-171-0/+1
| | | | Just noticed this while reading the code.
* fdio: Redo tmpfile API with GLnxTmpfile structColin Walters2017-05-152-44/+68
| | | | | | | | | | | | | | | | | The core problem with the previous tmpfile code is we don't have an autocleanup that calls `unlinkat` in the non-`O_TMPFILE` case. And even if we did, it'd be awkward still since the `glnx_link_tmpfile_at()` call *consumes* the tmpfile. Fix this by introducing a struct with a cleanup macro. This simplifies a number of the callers in libostree - a notable case is where we had two arrays, one of fds, one of paths. It makes other places in libostree a bit more complex, but that's because some of the commit code paths want to deal with temporary *symlinks* too. Most callers are better though - in libglnx itself, `glnx_file_copy_at()` now correctly unlinks on failure for example.
* glnx-errors.h: add a glnx_throw_prefix() variantJonathan Lebon2017-05-113-9/+59
| | | | | | For completeness. It just looks much cleaner than doing the `, FALSE` trick. It also takes care of appending the ': ' for you like its errno version.
* fdio: Expose glnx_regfile_copy_bytes(), rewrite: GNU style, POSIX errnov2017.1Colin Walters2017-04-282-91/+114
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | NOTE: This changes the error handling API of `glnx_loop_write()` to be "old school POSIX" instead of "systemd". In ostree in a few places we use `g_output_stream_splice()`. I thought this would use `splice()`, but actually it doesn't today. They also, if a cancellable is provided, end up dropping into `poll()` for every read and write. (In addition to copying data to/from userspace). My opinion on this is - for *local files* that's dumb. In the big picture, you really only need cancellation when copying gigabytes. Down the line, we could perhaps add a `glnx_copy_bytes_cancellable()` that only did that check e.g. every gigabyte of copied data. And when we do that we should use `g_cancellable_set_error_if_cancelled()` rather than a `poll()` with the regular file FD, since regular files are *always* readable and writable. For my use case with rpm-ostree though, we don't have gigabyte sized files, and seeing all of the `poll()` calls in strace is annoying. So let's have the non-cancellable file copying API that's modern and uses both reflink and `sendfile()` if available, in that order. My plan at some point once this is tested more is to migrate this code into GLib. Note that in order to keep our APIs consistent, I switched the systemd-imported code to "old school POSIX" error conventions. Otherwise we'd have *3* (POSIX, systemd, and GError) and particularly given the first two are easily confused, it'd be a recipe for bugs.
* fdio: Mostly port to new code styleColin Walters2017-04-251-149/+59
| | | | | | | | | | There's one function that did `unlinkat()` in the cleanup section, not doing that yet. Note I uncovered a few bugs in a few places where we didn't preserve errno before doing an `unlinkat()` in error paths in a few cases. I also tried to prefix a few more error cases with the system call name.
* Port most code (except fdio) to new styleColin Walters2017-04-253-80/+31
| | | | There's a lot more fdio code, starting with some of the easier ones.
* tests/xattrs: Skip on filesystems with no user xattr supportColin Walters2017-04-211-0/+15
| | | | | | Like tmpfs. See: https://github.com/flatpak/flatpak/issues/686
* glnx-fdio: Add wrappers around fstat() and fstatat() to handle errorsPhilip Withnall2017-04-211-0/+51
| | | | | | | | Add two inline wrappers around fstat() and fstatat() which handle retrying on EINTR and return other errors using GError, to be consistent with other glnx functions. Signed-off-by: Philip Withnall <withnall@endlessm.com>
* glnx-dirfd: Add variants of glnx_mkdtempat() which open the directoryPhilip Withnall2017-04-212-0/+78
| | | | | | | | At the moment, it’s not possible for them to do this race-free (since openat(O_DIRECTORY | O_CREAT | O_EXCL) doesn’t work), but in future this could be possible. In any case, it’s a useful thing to want to do. Signed-off-by: Philip Withnall <withnall@endlessm.com>
* glnx-shutil: Add glnx_shutil_mkdir_p_at_open()Philip Withnall2017-04-212-0/+45
| | | | | | | | | This is a variant of glnx_shutil_mkdir_p_at() which opens the given directory and returns a dirfd to it. Currently, the implementation cannot be race-free (due to a kernel bug), but it could eventually be made race-free. Signed-off-by: Philip Withnall <withnall@endlessm.com>
* glnx-local-alloc: Make check for invalid FDs more generalPhilip Withnall2017-04-211-2/+2
| | | | | | | | In general, all FDs < 0 are invalid (and should not have close() called on them), so check that. This could have caused problems if a function returned an error value < -1. Signed-off-by: Philip Withnall <withnall@endlessm.com>
* errors: Fix legacy set_prefix_error_from_errno()Colin Walters2017-03-242-4/+8
| | | | | | | We were missing the previous automatic `: ` addition; noticed in a failing ostree test. Fix this by just calling the new API as the non-prefix case does too.
* glnx-errors.h: add glnx_null_throw[_*] variantsJonathan Lebon2017-03-233-8/+62
| | | | | | | | | | | | | | | | | | | | | | | These are equivalent to the non-null throw, except that the returned value is a NULL pointer. They can be used in functions where one wants to return a pointer. E.g.: GKeyFile *foo(GError **error) { return glnx_null_throw (error, "foobar"); } The function call redirections are wrapped around a compound statement expression[1] so that they represent a single top-level expression. This allows us to avoid -Wunused-value warnings vs using a comma operator if the return value isn't used. I made the 'args...' absorb the fmt argument as well so that callers can still use it without always having to specify at least one additional variadic argument. I had to check to be sure that the expansion is all done by the preprocessor, so we don't need to worry about stack intricacies. [1] https://gcc.gnu.org/onlinedocs/gcc/Statement-Exprs.html
* errors: Add glnx_throw() and testsColin Walters2017-03-223-1/+109
| | | | | | | | Following up to the previous commit, also shorten our use of `g_set_error (..., G_IO_ERROR_FAILED, ...)`. There's a lot of this in libostree at least. See also https://bugzilla.gnome.org/show_bug.cgi?id=774061
* errors: Add new glnx_throw_errno{,_prefix}() APIsColin Walters2017-03-223-40/+80
| | | | | | | | | | | | | | | | | | | | We have a *lot* of code of the form: ``` if (unlinkat (fd, pathname) < 0) { glnx_set_error_from_errno (error); goto out; } ``` After conversion to `return FALSE style` which is in progress, it's way shorter, and clearer like this: ``` if (unlinkat (fd, pathname) < 0) return glnx_throw_errno (error); ```
* fdio: Expose wrappers for renameat2() EXCHANGE and NOREPLACEColin Walters2017-03-025-22/+236
| | | | | | | | | | | I want the `RENAME_EXCHANGE` version for rpm-ostree, to atomically swap `/usr/share/rpm` (a directory) with a new verison. While we're here we might as well expose `RENAME_NOREPLACE` in case something else wants it. These both have fallbacks to the non-atomic version. Closes: https://github.com/GNOME/libglnx/pull/36
* fix bug found by -Wmaybe-uninitializedJonathan Lebon2017-03-021-1/+1
|
* tests/xattrs: Fix possible NULL allocationColin Walters2017-02-211-1/+1
| | | | | | | | This showed up in the ostree runs with `-fsanitize=undefined` - if we happened to get `0` then `g_malloc` would return `NULL`. However, what's interesting is it seemed to happen *consistently*. I think what's going on is GCC proved that the value *could* be zero, and hence it *could* return NULL, and hence it was undefined behavior. Hooray for `-fsanitize=undefined`.
* COPYING: Bump to LGPL 2.1 due to systemd importColin Walters2017-02-111-91/+112
| | | | | | | | | | | We originally inherited LGPL 2.0 from glib I think. But I didn't notice when importing systemd code it's LGPL 2.1. While individual file licenses still apply; I'm not going to bother bumping all of them to 2.1, the complete module should be viewed as under 2.1. Bump the master COPYING file accordingly.
* xattrs: Add a test case for previous commitsColin Walters2017-01-292-0/+289
| | | | | This is actually the first test case in libglnx 🙌; hopefully the consumers are prepared for us injecting into `TESTS`.
* xattrs: Handle ERANGEColin Walters2017-01-291-0/+6
| | | | | This is symmetric with an earlier commit which handled a transition from `size != 0` -> `size = 0`. Now if xattrs are added we retry.
* xattrs: Dedup fd reading codeColin Walters2017-01-291-57/+18
| | | | | | | By taking both fd and path into one copy of the reader func, exactly like we do in `read_xattr_name_array`, we can abstract over the difference. Preparatory cleanup for more work here.
* xattrs: Handle xattrs changing size concurrentlyColin Walters2017-01-291-11/+18
| | | | | | We should be robust in the face of this and return a snapshot of the current value we saw, not transiently fail. This is the semantics we expect with ostree upgrades for `/etc` for example.
* listxattr: Don't assume that first call to listxattr gives correct sizeWilliam Manley2017-01-291-12/+20
| | | | | | | | | | | | | | | To get the right sized buffer to pass to `flistattr` and `llistattr` we first call them with a zero byte buffer. They then return the number of bytes they'll actually need to operate. We would `malloc` and then call again assuming that the size we got originally was correct. On my computer at least this isn't always the case. I've seen instances where the first call returns 23B, but then on the second one returns no data at all. Getting these non-existant xattrs would then cause ostree to fail. I'm not sure why it's behaving this way on my machine. I suspect its some interaction with overlayfs but I haven't proven this.
* dirfd: Use better and faster random algorithm for gen_temp_name()Colin Walters2017-01-261-26/+4
| | | | | | | | | | | | | | | | I was looking at ostree performance, and a surprising amount of time was spent in `glnx_gen_temp_name()`. We end up calling it from the main loop, and the iteration here shows up in my perf profiles. The glibc algorithm here that we adopted is *very* dated; let's switch to use `GRand`, which gives us a better algorithm. It'd be even better of course to use `getrandom()`, but we should do that in glib at some point. While I had the patient open, I extended the charset with lowercase, to better avoid collisions.
* dirfd: Set initialized flag for itersColin Walters2016-10-271-0/+4
| | | | | | | | | And use it when deinitializing, to avoid calling `closedir(NULL)`. In practice, this doesn't matter, because `closedir` *does* handle `NULL` in glibc. However, I'm playing with the GCC `-fsanitize=undefined`, and it aborts because `closedir` is tagged as requiring a non-`NULL` pointer.
* fdio: Make GLnxFileCopyFlags actually flagsColin Walters2016-10-251-3/+3
| | | | | | | | | I wanted to add a new one, and realized it was wrong. Luckily, I think we were safe until now, since the set of bits for `(0, 1, 2)` is actually distinct. Although, hm, callers specifying `GLNX_FILE_COPY_OVERWRITE` may have not actually been getting that.
* build: Add --enable-wrpseudo-compatColin Walters2016-10-052-2/+11
| | | | | | | | | See https://mail.gnome.org/archives/ostree-list/2016-October/msg00003.html Basically https://github.com/wrpseudo/pseudo doesn't implement newer APIs like renameat2() and O_TMPFILE, so on the host side (as potentially opposed to the target system) we want to be able to disable them.
* Distribute libglnx.m4Dan Nicholson2016-08-311-1/+5
| | | | This is needed by ostree when creating a tarball with make dist.
* Remove libcontainerColin Walters2016-08-305-357/+0
| | | | No longer used by anything; see https://github.com/projectatomic/rpm-ostree/pull/429
* libcontainer: Add a fd-relative APIColin Walters2016-08-072-11/+31
| | | | | | I'm porting rpm-ostree and need this. Of course all this libcontainer stuff will be nuked in favor of bubblewrap when everything comes together.
* Add missing files to libglnx distributionSimon McVittie2016-08-051-0/+2
| | | | Signed-off-by: Simon McVittie <smcv@debian.org>
* fdio: Only invoke fallocate() for sizes > 0Colin Walters2016-08-041-5/+8
| | | | | | | In some cases we want to replace with zero size, and `posix_fallocate()` is documented to return `EINVAL` in this case. Making this change since I noticed it elsewhere.
* Add --disable-otmpfileColin Walters2016-08-032-1/+10
| | | | | | | | Some systems have bugs with it, so let's allow downstreams to easily disable it. https://bugzilla.gnome.org/show_bug.cgi?id=769453 https://github.com/ostreedev/ostree/issues/421
* shutil: Use new API to iterate ensuring d_typeColin Walters2016-07-291-40/+1
| | | | This drops a lot of duplicate code.
* text_percent_internal: compare uints before printingJonathan Lebon2016-07-221-2/+4
| | | | | | | A wild sordid tale of substractions and unsigned integers leads this team of variables down a loonng path... Reported-by: Gatis Paeglis <gatis.paeglis@qt.io>
* fdio: Add unlinkat() in error paths for tmpfilesColin Walters2016-07-081-1/+9
| | | | | | | | | | This is kind of an ABI change but it's for the better I think; on error we consistently clean up the temp file. This is obviously necessary without `O_TMPFILE`. With it, we still need an error cleanup in the case where we're trying to replace an existing file. I noticed this in ostree's `tests/test-refs.sh` which intentionally tries to rename a file over a directory path.
* fdio: Use correct dfd with O_TMPFILE in rename caseColin Walters2016-07-081-1/+1
| | | | | | | | While auditing this code to figure out why ostree's `tests/test-refs.sh` was failing, while the bug turned out to be different, I noticed that in the case where `dfd != target_dfd`, we failed to do the right `renameat()`. (No code I'm aware of does this now).
* fdio: Add open_tmpfile_linkable() and link_tmpfile_at()Colin Walters2016-07-015-31/+406
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We had a bug previously where we failed to clean up a temporary file in an error path. This is a classic case where the new `O_TMPFILE` API in Linux is nicer. To implement this, as usual we start with some original bits from systemd. But in this case I ended up having to heavily modify it because systemd doesn't support "link into place and overwrite". They don't actually use their tempfile code much at all in fact - as far as I can tell, just in the coredump code. Whereas in many apps, ostree included, a very common use case is atomically updating an existing file, which is `glnx_file_replace_contents_at()`, including subtleties like doing an `fdatasync()` if the file already existed. Implementing this then is slightly weird since we need to link() the file into place, then rename() after. It's still better though because if we e.g. hit `ENOSPC` halfway through, we'll clean up the file automatically. We still do keep the mode where we error out if the file exists. Finally, the ostree core though does have a more unusual case where we want to ignore EEXIST (allow concurrent object writers), so add support for that now. Note: One really confusing bug I had here was that `O_TMPFILE` ignores the provided mode, and this caused ostree to write refs that weren't world readable. Rework things so we always call `fchmod()`, but as a consequence we're no longer honoring umask in the default case. I doubt anyone will care, and if they do we should probably fix ostree to consistently use a mode inherited from the repo or something.
* glnx_release_lock_file - Don't close fd -1 (i.e. if we never locked)Alexander Larsson2016-06-281-1/+2
| | | | | This happens a lot if you use autocleanup for lock files, and the function returns early without the lock being taken.
* fdio: Delete .tmp file on failureYu Qi Zhang2016-06-161-0/+2
| | | | | | | | We noticed the temp files being left over in ostree when (mistakenly) trying to replace the contents of a subpath that wasn't a directory. In the future we should look at the systemd code using `O_TMPFILE` here.
* dirfd: Fix inverted precondition in previous tmpname commitColin Walters2016-05-311-1/+1
| | | | I swear I tested it...
* Introduce glnx_gen_temp_name()Colin Walters2016-05-302-21/+39
| | | | | | | We have multiple copies growing again of this code. glibc has this API internally and uses it in multiple places, let's do the same. Closes: #14
* text_percent_internal: only pad right in the text caseJonathan Lebon2016-05-101-7/+5
| | | | | | | Padding in the percentage case was useless (and actually didn't work properly) since all the real estate is taken up by the text and the bar. We only need padding in the text case, in case the new string is shorter.
* fdio: Add glnx_stream_fstatColin Walters2016-05-032-0/+34
| | | | | Migrated from libgsystem's `gs_stream_fstat()`. It's a small function but I end up using it in OSTree a fair bit.
* libcontainer: Always set PATH when running in new rootColin Walters2016-05-021-3/+8
| | | | | | | For rpm-ostree's use we always run in a new root, so we don't want to inherit the host system's PATH. For example, NixOS uses PATH for its software namespacing, but one could be using rpm-ostree to build CentOS commits.
* console: Fix glnx_console_textColin Walters2016-05-011-24/+29
| | | | Not sure if it ever worked. We need to not print the bars, etc.
* Don't touch errno in glnx_fd_closeAlexander Larsson2016-03-101-2/+7
| | | | | | We're ignoring the result from the close, but it can still affect errno, which is bad if you use this in functions that sets errno, because errno can unexpectedly change after you've set it.
* console: Add an API to just emit textColin Walters2016-03-102-18/+33
| | | | We had this internally, just need to expose it.
View Lorry Controller Status