Merge branch 'jk/http-backend-deadlock' into maint
Communication between the HTTP server and http_backend process can
lead to a dead-lock when relaying a large ref negotiation request.
Diagnose the situation better, and mitigate it by reading such a
request first into core (to a reasonable limit).
* jk/http-backend-deadlock:
http-backend: spool ref negotiation requests to buffer
t5551: factor out tag creation
http-backend: fix die recursion with custom handler
Merge branch 'jh/filter-empty-contents' into maint
The clean/smudge interface did not work well when filtering an
empty contents (failed and then passed the empty input through).
It can be argued that a filter that produces anything but empty for
an empty input is nonsense, but if the user wants to do strange
things, then why not?
* jh/filter-empty-contents:
sha1_file: pass empty buffer to index empty file
l10n: de.po: change error message from "sagen" to "Meinten Sie"
We should not use "sagen" if someone has written something wrong.
Although it's "say" in English, we should not use it in German
and instead use our normal error message.
Merge branch 'rs/plug-leak-in-pack-bitmaps' into maint
The code to read pack-bitmap wanted to allocate a few hundred
pointers to a structure, but by mistake allocated and leaked memory
enough to hold that many actual structures. Correct the allocation
size and also have it on stack, as it is small enough.
Various documentation mark-up fixes to make the output more
consistent in general and also make AsciiDoctor (an alternative
formatter) happier.
* jk/asciidoc-markup-fix:
doc: convert AsciiDoc {?foo} to ifdef::foo[]
doc: put example URLs and emails inside literal backticks
doc: drop backslash quoting of some curly braces
doc: convert \--option to --option
doc/add: reformat `--edit` option
doc: fix length of underlined section-title
doc: fix hanging "+"-continuation
doc: fix unquoted use of "{type}"
doc: fix misrendering due to `single quote'
Merge branch 'mh/write-refs-sooner-2.4' into maint
Multi-ref transaction support we merged a few releases ago
unnecessarily kept many file descriptors open, risking to fail with
resource exhaustion. This is for 2.4.x track.
* mh/write-refs-sooner-2.4:
ref_transaction_commit(): fix atomicity and avoid fd exhaustion
ref_transaction_commit(): remove the local flags variable
ref_transaction_commit(): inline call to write_ref_sha1()
rename_ref(): inline calls to write_ref_sha1() from this function
commit_ref_update(): new function, extracted from write_ref_sha1()
write_ref_to_lockfile(): new function, extracted from write_ref_sha1()
t7004: rename ULIMIT test prerequisite to ULIMIT_STACK_SIZE
update-ref: test handling large transactions properly
ref_transaction_commit(): fix atomicity and avoid fd exhaustion
ref_transaction_commit(): remove the local flags variable
ref_transaction_commit(): inline call to write_ref_sha1()
rename_ref(): inline calls to write_ref_sha1() from this function
commit_ref_update(): new function, extracted from write_ref_sha1()
write_ref_to_lockfile(): new function, extracted from write_ref_sha1()
t7004: rename ULIMIT test prerequisite to ULIMIT_STACK_SIZE
update-ref: test handling large transactions properly
The ref API did not handle cases where 'refs/heads/xyzzy/frotz' is
removed at the same time as 'refs/heads/xyzzy' is added (or vice
versa) very well.
* mh/ref-directory-file:
reflog_expire(): integrate lock_ref_sha1_basic() errors into ours
ref_transaction_commit(): delete extra "the" from error message
ref_transaction_commit(): provide better error messages
rename_ref(): integrate lock_ref_sha1_basic() errors into ours
lock_ref_sha1_basic(): improve diagnostics for ref D/F conflicts
lock_ref_sha1_basic(): report errors via a "struct strbuf *err"
verify_refname_available(): report errors via a "struct strbuf *err"
verify_refname_available(): rename function
refs: check for D/F conflicts among refs created in a transaction
ref_transaction_commit(): use a string_list for detecting duplicates
is_refname_available(): use dirname in first loop
struct nonmatching_ref_data: store a refname instead of a ref_entry
report_refname_conflict(): inline function
entry_matches(): inline function
is_refname_available(): convert local variable "dirname" to strbuf
is_refname_available(): avoid shadowing "dir" variable
is_refname_available(): revamp the comments
t1404: new tests of ref D/F conflicts within transactions
The "log --decorate" enhancement in Git 2.4 that shows the commit
at the tip of the current branch e.g. "HEAD -> master", did not
work with --decorate=full.
* mg/log-decorate-HEAD:
log: do not shorten decoration names too early
log: decorate HEAD with branch name under --decorate=full, too
There was a commented-out (instead of being marked to expect
failure) test that documented a breakage that was fixed since the
test was written; turn it into a proper test.
* sb/t1020-cleanup:
subdirectory tests: code cleanup, uncomment test
core.excludesfile (defaulting to $XDG_HOME/git/ignore) is supposed
to be overridden by repository-specific .git/info/exclude file, but
the order was swapped from the beginning. This belatedly fixes it.
* jc/gitignore-precedence:
ignore: info/exclude should trump core.excludesfile
The connection initiation code for "ssh" transport tried to absorb
differences between the stock "ssh" and Putty-supplied "plink" and
its derivatives, but the logic to tell that we are using "plink"
variants were too loose and falsely triggered when "plink" appeared
anywhere in the path (e.g. "/home/me/bin/uplink/ssh").
* bc/connect-plink:
connect: improve check for plink to reduce false positives
t5601: fix quotation error leading to skipped tests
connect: simplify SSH connection code path
Merge branch 'tb/blame-resurrect-convert-to-git' into maint
Some time ago, "git blame" (incorrectly) lost the convert_to_git()
call when synthesizing a fake "tip" commit that represents the
state in the working tree, which broke folks who record the history
with LF line ending to make their project portabile across
platforms while terminating lines in their working tree files with
CRLF for their platform.
* tb/blame-resurrect-convert-to-git:
blame: CRLF in the working tree and LF in the repo
* pt/xdg-config-path:
path.c: remove home_config_paths()
git-config: replace use of home_config_paths()
git-commit: replace use of home_config_paths()
credential-store.c: replace home_config_paths() with xdg_config_home()
dir.c: replace home_config_paths() with xdg_config_home()
attr.c: replace home_config_paths() with xdg_config_home()
path.c: implement xdg_config_home()
t0302: "unreadable" test needs POSIXPERM
t0302: test credential-store support for XDG_CONFIG_HOME
git-credential-store: support XDG_CONFIG_HOME
git-credential-store: support multiple credential files
"git rev-list --objects $old --not --all" to see if everything that
is reachable from $old is already connected to the existing refs
was very inefficient.
* jk/still-interesting:
limit_list: avoid quadratic behavior from still_interesting
"hash-object --literally" introduced in v2.2 was not prepared to
take a really long object type name.
* jc/hash-object:
write_sha1_file(): do not use a separate sha1[] array
t1007: add hash-object --literally tests
hash-object --literally: fix buffer overrun with extra-long object type
git-hash-object.txt: document --literally option
Merge branch 'jk/filter-branch-use-of-sed-on-incomplete-line' into maint
"filter-branch" corrupted commit log message that ends with an
incomplete line on platforms with some "sed" implementations that
munge such a line. Work it around by avoiding to use "sed".
* jk/filter-branch-use-of-sed-on-incomplete-line:
filter-branch: avoid passing commit message through sed
Merge branch 'jk/stash-require-clean-index' into maint
"git stash pop/apply" forgot to make sure that not just the working
tree is clean but also the index is clean. The latter is important
as a stash application can conflict and the index will be used for
conflict resolution.
* jk/stash-require-clean-index:
stash: require a clean index to apply
t3903: avoid applying onto dirty index
t3903: stop hard-coding commit sha1s
Merge branch 'jk/git-no-more-argv0-path-munging' into maint
We have prepended $GIT_EXEC_PATH and the path "git" is installed in
(typically "/usr/bin") to $PATH when invoking subprograms and hooks
for almost eternity, but the original use case the latter tried to
support was semi-bogus (i.e. install git to /opt/foo/git and run it
without having /opt/foo on $PATH), and more importantly it has
become less and less relevant as Git grew more mainstream (i.e. the
users would _want_ to have it on their $PATH). Stop prepending the
path in which "git" is installed to users' $PATH, as that would
interfere the command search order people depend on (e.g. they may
not like versions of programs that are unrelated to Git in /usr/bin
and want to override them by having different ones in /usr/local/bin
and have the latter directory earlier in their $PATH).
* jk/git-no-more-argv0-path-munging:
stop putting argv[0] dirname at front of PATH
Merge branch 'jk/http-backend-deadlock-2.3' into jk/http-backend-deadlock
* jk/http-backend-deadlock-2.3:
http-backend: spool ref negotiation requests to buffer
t5551: factor out tag creation
http-backend: fix die recursion with custom handler
Merge branch 'jk/http-backend-deadlock-2.2' into jk/http-backend-deadlock-2.3
* jk/http-backend-deadlock-2.2:
http-backend: spool ref negotiation requests to buffer
t5551: factor out tag creation
http-backend: fix die recursion with custom handler
http-backend: spool ref negotiation requests to buffer
When http-backend spawns "upload-pack" to do ref
negotiation, it streams the http request body to
upload-pack, who then streams the http response back to the
client as it reads. In theory, git can go full-duplex; the
client can consume our response while it is still sending
the request. In practice, however, HTTP is a half-duplex
protocol. Even if our client is ready to read and write
simultaneously, we may have other HTTP infrastructure in the
way, including the webserver that spawns our CGI, or any
intermediate proxies.
In at least one documented case[1], this leads to deadlock
when trying a fetch over http. What happens is basically:
1. Apache proxies the request to the CGI, http-backend.
2. http-backend gzip-inflates the data and sends
the result to upload-pack.
3. upload-pack acts on the data and generates output over
the pipe back to Apache. Apache isn't reading because
it's busy writing (step 1).
This works fine most of the time, because the upload-pack
output ends up in a system pipe buffer, and Apache reads
it as soon as it finishes writing. But if both the request
and the response exceed the system pipe buffer size, then we
deadlock (Apache blocks writing to http-backend,
http-backend blocks writing to upload-pack, and upload-pack
blocks writing to Apache).
We need to break the deadlock by spooling either the input
or the output. In this case, it's ideal to spool the input,
because Apache does not start reading either stdout _or_
stderr until we have consumed all of the input. So until we
do so, we cannot even get an error message out to the
client.
The solution is fairly straight-forward: we read the request
body into an in-memory buffer in http-backend, freeing up
Apache, and then feed the data ourselves to upload-pack. But
there are a few important things to note:
1. We limit the in-memory buffer to prevent an obvious
denial-of-service attack. This is a new hard limit on
requests, but it's unlikely to come into play. The
default value is 10MB, which covers even the ridiculous
100,000-ref negotation in the included test (that
actually caps out just over 5MB). But it's configurable
on the off chance that you don't mind spending some
extra memory to make even ridiculous requests work.
2. We must take care only to buffer when we have to. For
pushes, the incoming packfile may be of arbitrary
size, and we should connect the input directly to
receive-pack. There's no deadlock problem here, though,
because we do not produce any output until the whole
packfile has been read.
For upload-pack's initial ref advertisement, we
similarly do not need to buffer. Even though we may
generate a lot of output, there is no request body at
all (i.e., it is a GET, not a POST).
Test-adapted-from: Dennis Kaarsemaker <dennis@kaarsemaker.net> Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Fix remaining instances where "pack-file" is used instead of
"packfile". Some places remain where we still use "pack-file",
This is the case when we explicitly refer to a file with a
".pack" extension as opposed to a data source providing a pack
data stream.
Signed-off-by: Patrick Steinhardt <ps@pks.im> Signed-off-by: Junio C Hamano <gitster@pobox.com>
If you run "git stash --help", you get the help for stash
(this magic is done by the git wrapper itself). But if you
run "git stash drop --help", you get an error. We
cannot show help specific to "stash drop", of course, but we
can at least give the user the normal stash manpage.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
The option parser for git-stash stuffs unknown flags into
the $FLAGS variable, where they can be accessed by the
individual commands. However, most commands do not even look
at these extra flags, leading to unexpected results like
this:
We should notice the extra flags and bail. Rather than
annotate each command to reject a non-empty $FLAGS variable,
we can notice that "stash show" is the only command that
actually _wants_ arbitrary flags. So we switch the default
mode to reject unknown flags, and let stash_show() opt into
the feature.
Reported-by: Vincent Legoll <vincent.legoll@gmail.com> Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
One of our tests in t5551 creates a large number of tags,
and jumps through some hoops to do it efficiently. Let's
factor that out into a function so we can make other similar
tests.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
pack-bitmaps: plug memory leak, fix allocation size for recent_bitmaps
Use an automatic variable for recent_bitmaps, an array of pointers.
This way we don't allocate too much and don't have to free the memory
at the end. The old code over-allocated because it reserved enough
memory to store all of the structs it is only pointing to and never
freed it. 160 64-bit pointers take up 1280 bytes, which is not too
much to be placed on the stack.
MAX_XOR_OFFSET is turned into a preprocessor constant to make it
constant enough for use in an non-variable array declaration.
Noticed-by: Stefan Beller <stefanbeller@gmail.com> Suggested-by: Jeff King <peff@peff.net> Signed-off-by: Rene Scharfe <l.s.r@web.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
clone: call transport_set_verbosity before anything else on the newly created transport
Commit 2879bc3 made the progress and verbosity options sent to remote helper
earlier than they previously were. But nothing else after that would send
updates if the value is changed later on with transport_set_verbosity.
While for fetch and push, transport_set_verbosity is the first thing that
is done after creating the transport, it was not the case for clone. So
commit 2879bc3 broke changing progress and verbosity for clone, for urls
requiring a remote helper only (so, not git:// urls, for instance).
Moving transport_set_verbosity to just after the transport is created
works around the issue.
Signed-off-by: Mike Hommey <mh@glandium.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Back when these tests were written, we wanted to make sure that Git
notices it is in a bare repository and "git show -s HEAD" would
refrain from complaining that HEAD might mean a file it sees in its
current working directory (because it does not). But the version of
Git back then didn't behave well, without (doubly) being told that
it is inside a bare repository by exporting "GIT_DIR=.". The form
of the test we originally wanted to have was left commented out as
a reminder.
Nowadays the test as originally intended works, so add it to the
test suite. We'll keep the old test that explicitly sets GIT_DIR=.
to make sure that use case will not regress.
Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Even though "git clean" takes pathspec to limit the part of the
working tree to be cleaned, it checked the paths it encounters
during its directory traversal with lstat(2), before checking if
the path is within the pathspec.
Ignore paths outside pathspec and proceed without checking with
lstat(2). Even if such a path is unreadable due to e.g. EPERM,
"git clean" should not care.
Signed-off-by: David Turner <dturner@twopensource.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
There are several "raw formats", and describing --raw as "Generate the
raw format" in the documentation for git-log seems to imply that it
generates the raw *log* format.
Clarify the wording by saying "raw diff format" explicitly, and make a
special-case for "git log": "git log --raw" does not just change the
format, it shows something which is not shown by default.
Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since b814da8 (pull: add pull.ff configuration, 2014-01-15) git-pull
supported setting --(no-)ff via the pull.ff configuration value.
However, as it only matches the string values of "true" and "false", it
does not support other boolean aliases such as "on", "off", "1", "0".
This is inconsistent with the merge.ff setting, which supports these
aliases.
Fix this by using the bool_or_string_config function to retrieve the
value of pull.ff.
Signed-off-by: Paul Tan <pyokagan@gmail.com> Reviewed-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since b814da8 (pull: add pull.ff configuration, 2014-01-15), running
git-pull with the configuration pull.ff=false or pull.ff=only is
equivalent to passing --no-ff and --ff-only to git-merge. However, if
pull.ff=true, no switch is passed to git-merge. This leads to the
confusing behavior where pull.ff=false or pull.ff=only is able to
override merge.ff, while pull.ff=true is unable to.
Fix this by adding the --ff switch if pull.ff=true, and add a test to
catch future regressions.
Furthermore, clarify in the documentation that pull.ff overrides
merge.ff.
Signed-off-by: Paul Tan <pyokagan@gmail.com> Reviewed-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since efb779f (merge, pull: add '--(no-)log' command line option,
2008-04-06) git-pull supported the (--no-)log switch and would pass it
to git-merge.
96e9420 (merge: Make '--log' an integer option for number of shortlog
entries, 2010-09-08) implemented support for the --log=<n> switch, which
would explicitly set the number of shortlog entries. However, git-pull
does not recognize this option, and will instead pass it to git-fetch,
leading to "unknown option" errors.
Fix this by matching --log=* in addition to --log and --no-log.
Implement a test for this use case.
Signed-off-by: Paul Tan <pyokagan@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
`git add` of an empty file with a filter pops complaints from
`copy_fd` about a bad file descriptor.
This traces back to these lines in sha1_file.c:index_core:
if (!size) {
ret = index_mem(sha1, NULL, size, type, path, flags);
The problem here is that content to be added to the index can be
supplied from an fd, or from a memory buffer, or from a pathname. This
call is supplying a NULL buffer pointer and a zero size.
Downstream logic takes the complete absence of a buffer to mean the
data is to be found elsewhere -- for instance, these, from convert.c:
~If there's a buffer, write from that, otherwise the data must be coming
from an open fd.~
Perfectly reasonable logic in a routine that's going to write from
either a buffer or an fd.
So change `index_core` to supply an empty buffer when indexing an empty
file.
There's a patch out there that instead changes the logic quoted above to
take a `-1` fd to mean "use the buffer", but it seems to me that the
distinction between a missing buffer and an empty one carries intrinsic
semantics, where the logic change is adapting the code to handle
incorrect arguments.
Signed-off-by: Jim Hill <gjthill@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
http-backend: fix die recursion with custom handler
When we die() in http-backend, we call a custom handler that
writes an HTTP 500 response to stdout, then reports the
error to stderr. Our routines for writing out the HTTP
response may themselves die, leading to us entering die()
again.
When it was originally written, that was OK; our custom
handler keeps a variable to notice this and does not
recurse. However, since cd163d4 (usage.c: detect recursion
in die routines and bail out immediately, 2012-11-14), the
main die() implementation detects recursion before we even
get to our custom handler, and bails without printing
anything useful.
We can handle this case by doing two things:
1. Installing a custom die_is_recursing handler that
allows us to enter up to one level of recursion. Only
the first call to our custom handler will try to write
out the error response. So if we die again, that is OK.
If we end up dying more than that, it is a sign that we
are in an infinite recursion.
2. Reporting the error to stderr before trying to write
out the HTTP response. In the current code, if we do
die() trying to write out the response, we'll exit
immediately from this second die(), and never get a
chance to output the original error (which is almost
certainly the more interesting one; the second die is
just going to be along the lines of "I tried to write
to stdout but it was closed").
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
The problem is that setup_rerere() will not actually take
the lock if the rerere system is disabled. We should notice
this and return early. We can return with a success code
here, because we know there is nothing to forget.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
The former seems to just be syntactic sugar for the latter.
And as it's sugar that AsciiDoctor doesn't understand, it
would be nice to avoid it. Since there are only two spots,
and the resulting source is not significantly harder to
read, it's worth doing.
Note that this does slightly affect the generated HTML (it
has an extra newline), but the rendered result for both HTML
and docbook should be the same (since the newline is not
syntactically significant there).
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach the codepaths that read .gitignore and .gitattributes files
that these files encoded in UTF-8 may have UTF-8 BOM marker at the
beginning; this makes it in line with what we do for configuration
files already.
* cn/bom-in-gitignore:
attr: skip UTF8 BOM at the beginning of the input file
config: use utf8_bom[] from utf.[ch] in git_parse_source()
utf8-bom: introduce skip_utf8_bom() helper
add_excludes_from_file: clarify the bom skipping logic
dir: allow a BOM at the beginning of exclude files
Access to objects in repositories that borrow from another one on a
slow NFS server unnecessarily got more expensive due to recent code
becoming more cautious in a naive way not to lose objects to pruning.
* jk/prune-mtime:
sha1_file: only freshen packs once per run
sha1_file: freshen pack objects before loose
reachable: only mark local objects as recent
Merge branch 'jk/init-core-worktree-at-root' into maint
We avoid setting core.worktree when the repository location is the
".git" directory directly at the top level of the working tree, but
the code misdetected the case in which the working tree is at the
root level of the filesystem (which arguably is a silly thing to
do, but still valid).
* jk/init-core-worktree-at-root:
init: don't set core.worktree when initializing /.git
The DECORATE_SHORT_REFS option given to load_ref_decorations()
affects the way a copy of the refname is stored for each decorated
commit, and this forces later steps like current_pointed_by_HEAD()
to adjust their behaviour based on this initial settings.
Instead, we can always store the full refname and then shorten them
when producing the output.
log: decorate HEAD with branch name under --decorate=full, too
The previous step to teach "log --decorate" to show "HEAD -> master"
instead of "HEAD, master" when showing the commit at the tip of the
'master' branch, when the 'master' branch is checked out, did not
work for "log --decorate=full".
The commands in the "log" family prepare commit decorations for all
refs upfront, and the actual string used in a decoration depends on
how load_ref_decorations() is called very early in the process. By
default, "git log --decorate" stores names with common prefixes such
as "refs/heads" stripped; "git log --decorate=full" stores the full
refnames.
When the current_pointed_by_HEAD() function has to decide if "HEAD"
points at the branch a decoration describes, however, what was
passed to load_ref_decorations() to decide to strip (or keep) such a
common prefix is long lost. This makes it impossible to reliably
tell if a decoration that stores "refs/heads/master", for example,
is the 'master' branch (under "--decorate" with prefix omitted) or
'refs/heads/master' branch (under "--decorate=full").
Keep what was passed to load_ref_decorations() in a global next to
the global variable name_decoration, and use that to decide how to
match what was read from "HEAD" and what is in a decoration.
Text like "{foo}" triggers an AsciiDoc attribute; we have to
write "\{foo}" to suppress this. But when the "foo" is not a
syntactically valid attribute, we can skip the quoting. This
makes the source nicer to read, and looks better under
Asciidoctor. With AsciiDoc itself, this patch produces no
changes.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Older versions of AsciiDoc would convert the "--" in
"--option" into an emdash. According to 565e135
(Documentation: quote double-dash for AsciiDoc, 2011-06-29),
this is fixed in AsciiDoc 8.3.0. According to bf17126, we
don't support anything older than 8.4.1 anyway, so we no
longer need to worry about quoting.
Even though this does not change the output at all, there
are a few good reasons to drop the quoting:
1. It makes the source prettier to read.
2. We don't quote consistently, which may be confusing when
reading the source.
3. Asciidoctor does not like the quoting, and renders a
literal backslash.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
but AsciiDoctor is more strict. Let's match the underline to
the title (which also makes the source prettier to read).
The output from AsciiDoc is the same either way.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
In list content that wants to continue to a second
paragraph, the "+" continuation and subsequent paragraph
need to be left-aligned. Otherwise AsciiDoc seems to insert
only a linebreak.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Curly braces open an "attribute" in AsciiDoc; if there's no
such attribute, strange things may happen. In this case, the
unquoted "{type}" causes AsciiDoc to omit an entire line of
text from the output. We can fix it by putting the whole
phrase inside literal backticks (which also lets us get rid
of ugly backslash escaping).
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
AsciiDoc misparses some text that contains a `literal`
word followed by a fancy `single quote' word, and treats
everything from the start of the literal to the end of the
quote as a single-quoted phrase.
We can work around this by switching the latter to be a
literal, as well. In the first case, this is perhaps what
was intended anyway, as it makes us consistent with the the
earlier literals in the same paragraph. In the second, the
output is arguably better, as we will format our commit
references as <code> blocks.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Merge branch 'mh/write-refs-sooner-2.3' into mh/write-refs-sooner-2.4
* mh/write-refs-sooner-2.3:
ref_transaction_commit(): fix atomicity and avoid fd exhaustion
ref_transaction_commit(): remove the local flags variable
ref_transaction_commit(): inline call to write_ref_sha1()
rename_ref(): inline calls to write_ref_sha1() from this function
commit_ref_update(): new function, extracted from write_ref_sha1()
write_ref_to_lockfile(): new function, extracted from write_ref_sha1()
t7004: rename ULIMIT test prerequisite to ULIMIT_STACK_SIZE
update-ref: test handling large transactions properly
ref_transaction_commit(): fix atomicity and avoid fd exhaustion
The old code was roughly
for update in updates:
acquire locks and check old_sha
for update in updates:
if changing value:
write_ref_to_lockfile()
commit_ref_update()
for update in updates:
if deleting value:
unlink()
rewrite packed-refs file
for update in updates:
if reference still locked:
unlock_ref()
This has two problems.
Non-atomic updates
==================
The atomicity of the reference transaction depends on all pre-checks
being done in the first loop, before any changes have started being
committed in the second loop. The problem is that
write_ref_to_lockfile() (previously part of write_ref_sha1()), which
is called from the second loop, contains two more checks:
* It verifies that new_sha1 is a valid object
* If the reference being updated is a branch, it verifies that
new_sha1 points at a commit object (as opposed to a tag, tree, or
blob).
If either of these checks fails, the "transaction" is aborted during
the second loop. But this might happen after some reference updates
have already been permanently committed. In other words, the
all-or-nothing promise of "git update-ref --stdin" could be violated.
So these checks have to be moved to the first loop.
The old code locked all of the references in the first loop, leaving
all of the lockfiles open until later loops. Since we might be
updating a lot of references, this could result in file descriptor
exhaustion.
The solution
============
After this patch, the code looks like
for update in updates:
acquire locks and check old_sha
if changing value:
write_ref_to_lockfile()
else:
close_ref()
for update in updates:
if changing value:
commit_ref_update()
for update in updates:
if deleting value:
unlink()
rewrite packed-refs file
for update in updates:
if reference still locked:
unlock_ref()
This fixes both problems:
1. The pre-checks in write_ref_to_lockfile() are now done in the first
loop, before any changes have been committed. If any of the checks
fails, the whole transaction can now be rolled back correctly.
2. All lockfiles are closed in the first loop immediately after they
are created (either by write_ref_to_lockfile() or by close_ref()).
This means that there is never more than one open lockfile at a
time, preventing file descriptor exhaustion.
To simplify the bookkeeping across loops, add a new REF_NEEDS_COMMIT
bit to update->flags, which keeps track of whether the corresponding
lockfile needs to be committed, as opposed to just unlocked. (Since
"struct ref_update" is internal to the refs module, this change is not
visible to external callers.)
This change fixes two tests in t1400.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu> Signed-off-by: Junio C Hamano <gitster@pobox.com>
ref_transaction_commit(): remove the local flags variable
Instead, work directly with update->flags. This has the advantage that
the REF_DELETING bit, set in the first loop, can be read in the second
loop instead of having to be recomputed. Plus, it was potentially
confusing having both update->flags and flags, which sometimes had
different values.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu> Signed-off-by: Junio C Hamano <gitster@pobox.com>
t7004: rename ULIMIT test prerequisite to ULIMIT_STACK_SIZE
During creation of the patch series our discussion we could have a
more descriptive name for the prerequisite for the test so it stays
unique when other limits of ulimit are introduced.
Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu> Signed-off-by: Junio C Hamano <gitster@pobox.com>
update-ref: test handling large transactions properly
Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Merge branch 'mh/write-refs-sooner-2.2' into mh/write-refs-sooner-2.3
* mh/write-refs-sooner-2.2:
ref_transaction_commit(): fix atomicity and avoid fd exhaustion
ref_transaction_commit(): remove the local flags variable
ref_transaction_commit(): inline call to write_ref_sha1()
rename_ref(): inline calls to write_ref_sha1() from this function
commit_ref_update(): new function, extracted from write_ref_sha1()
write_ref_to_lockfile(): new function, extracted from write_ref_sha1()
t7004: rename ULIMIT test prerequisite to ULIMIT_STACK_SIZE
update-ref: test handling large transactions properly
ref_transaction_commit(): fix atomicity and avoid fd exhaustion
The old code was roughly
for update in updates:
acquire locks and check old_sha
for update in updates:
if changing value:
write_ref_to_lockfile()
commit_ref_update()
for update in updates:
if deleting value:
unlink()
rewrite packed-refs file
for update in updates:
if reference still locked:
unlock_ref()
This has two problems.
Non-atomic updates
==================
The atomicity of the reference transaction depends on all pre-checks
being done in the first loop, before any changes have started being
committed in the second loop. The problem is that
write_ref_to_lockfile() (previously part of write_ref_sha1()), which
is called from the second loop, contains two more checks:
* It verifies that new_sha1 is a valid object
* If the reference being updated is a branch, it verifies that
new_sha1 points at a commit object (as opposed to a tag, tree, or
blob).
If either of these checks fails, the "transaction" is aborted during
the second loop. But this might happen after some reference updates
have already been permanently committed. In other words, the
all-or-nothing promise of "git update-ref --stdin" could be violated.
So these checks have to be moved to the first loop.
The old code locked all of the references in the first loop, leaving
all of the lockfiles open until later loops. Since we might be
updating a lot of references, this could result in file descriptor
exhaustion.
The solution
============
After this patch, the code looks like
for update in updates:
acquire locks and check old_sha
if changing value:
write_ref_to_lockfile()
else:
close_ref()
for update in updates:
if changing value:
commit_ref_update()
for update in updates:
if deleting value:
unlink()
rewrite packed-refs file
for update in updates:
if reference still locked:
unlock_ref()
This fixes both problems:
1. The pre-checks in write_ref_to_lockfile() are now done in the first
loop, before any changes have been committed. If any of the checks
fails, the whole transaction can now be rolled back correctly.
2. All lockfiles are closed in the first loop immediately after they
are created (either by write_ref_to_lockfile() or by close_ref()).
This means that there is never more than one open lockfile at a
time, preventing file descriptor exhaustion.
To simplify the bookkeeping across loops, add a new REF_NEEDS_COMMIT
bit to update->flags, which keeps track of whether the corresponding
lockfile needs to be committed, as opposed to just unlocked. (Since
"struct ref_update" is internal to the refs module, this change is not
visible to external callers.)
This change fixes two tests in t1400.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu> Signed-off-by: Junio C Hamano <gitster@pobox.com>