| Age | Commit message (Collapse) | Author |
|---|
|
Cache generation needs a way of quickly iterating over the unique potion
of the dependencies to be able to share them. By linking them together
we can reduce the speed penality (~ 80%) with only a small reduction in
saved size (~ 20%).
Git-Dch: Ignore
|
|
Having dependency data separated from the link between version/package
and the dependency allows use to work on sharing the depdency data a bit
as it turns out that many dependencies are in fact duplicates. How many
are duplicates various heavily with the sources configured, but for a
single Debian release the ballpark is 2 duplicates for each dependency
already (e.g. libc6 counts 18410 dependencies, but only 45 unique). Add
more releases and the duplicates count only rises to get ~6 for 3
releases. For each architecture a user has configured which given the
shear number of dependencies amounts to MBs of duplication.
We can cut down on this number, but pay a heavy price for it: In my
many releases(3) + architectures(3) test we have a 10% (~ 0.5 sec)
increase in cache creationtime, but also 10% less cachesize (~ 10 MB).
Further work is needed to rip the whole benefits from this through, so
this is just the start.
Git-Dch: Ignore
|
|
With a bit of trickery and the Curiously recurring template pattern we
can free us from our use of virtual in the iterators were it is unneeded
bloat as we never deal with pointers to iterators and similar such.
Git-Dch: Ignore
|
|
C++11 slightly changes the API again to const_iterator, but we are find
with iterators in the C++03 style for now as long as they look and
behave equally to the methods of the standard containers.
Git-Dch: Ignore
|
|
Some of them modify the ABI, but given that we prepare a big one
already, these few hardly count for much.
Git-Dch: Ignore
|
|
We used to read the Release file for each Packages file and store the
data in the PackageFile struct even through potentially many Packages
(and Translation-*) files could use the same data. The point of the
exercise isn't the duplicated data through. Having the Release files as
first-class citizens in the Cache allows us to properly track their
state as well as allows us to use the information also for files which
aren't in the cache, but where we know to which Release file they
belong (Sources are an example for this).
This modifies the pkgCache structs, especially the PackagesFile struct
which depending on how libapt users access the data in these structs can
mean huge breakage or no visible change. As a single data point:
aptitude seems to be fine with this. Even if there is breakage it is
trivial to fix in a backportable way while avoiding breakage for
everyone would be a huge pain for us.
Note that not all PackageFile structs have a corresponding ReleaseFile.
In particular the dpkg/status file as well as *.deb files have not. As
these have only a Archive property need, the Component property takes
over this duty and the ReleaseFile remains zero. This is also the reason
why it isn't needed nor particularily recommended to change from
PackagesFile to ReleaseFile blindly. Sticking with the earlier is
usually the better option.
|
|
|
|
A version belongs to a section and has hence a section member of its
own. A package on the other hand can have multiple versions from
different sections. This was "solved" by using the section which was
parsed first as order of sources.list defines, but that is obviously a
horribly unpredictable thing.
Users are way better of with the Section() as returned by the version
they are dealing with. It is likely the same for all versions of a
package, but in the few cases it isn't, it is important (like packages
moving from main/* to contrib/* or into oldlibs …).
Backport of 7a66977 which actually instantly removes the member.
|
|
Git-Dch: Ignore
|
|
Accessing the package records to acquire this information is pretty
costly, so that information wasn't used so far in many places. The most
noticeable user by far is EDSP at the moment, but there are ideas to
change that which this commit tries to enable.
|
|
They both store the same information, so this field just takes up space
in the Package struct for no good reason. We mark it "just" as deprecated
instead of instantly removing it though as it isn't misleading like
Section was and is potentially used in the wild more often.
|
|
A version belongs to a section and has hence a section member of its
own. A package on the other hand can have multiple versions from
different sections. This was "solved" by using the section which was
parsed first as order of sources.list defines, but that is obviously a
horribly unpredictable thing.
We therefore directly remove this struct member to free some space and
mark the access method as deprecated, which is told to return the
section of the 'newest' known version, which is at least predictable,
but possible not what it returned before – but nobody knows.
Users are way better of with the Section() as returned by the version
they are dealing with. It is likely the same for all versions of a
package, but in the few cases it isn't, it is important (like packages
moving from main/* to contrib/* or into oldlibs …).
|
|
Git-Dch: Ignore
Reported-By: gcc -Wsuggest-attribute={pure,const,noreturn}
|
|
Beside being a bit cleaner it hopefully also resolves oddball problems
I have with high levels of parallel jobs.
Git-Dch: Ignore
Reported-By: iwyu (include-what-you-use)
|
|
Git-Dch: Ignore
Reported-By: gcc -Wcast-qual
|
|
Git-Dch: Ignore
Reported-By: gcc -Wpedantic
|
|
|
|
Adds on top of Version 2 to all displayed version numbers the
architecture as well as the MultiArch flag for consumption by the hooks.
Most of the time the architecture will be the same for both versions
displayed, but packages might change from "all" to "any" (or back)
between versions so we can't display the architecture for packages.
Pseudo-Format for Version 3:
<name> <version> <arch> <m-a-flag> <compare> <version> <arch> <m-a-flag>
Examples:
stuff - - none < 1 amd64 none **CONFIGURE**
libsame 1 i386 same < 2 i386 same **CONFIGURE**
stuff 2 i386 none > 1 i386 none **CONFIGURE**
libsame 2 i386 same > - - none **REMOVE**
toolkit 1 all foreign > - - none **REMOVE**
Closes: #712116
|
|
- provide DepIterator::IsSatisfied as a nicer shorthand for DepCheck
|
|
- add an IsMultiArchImplicit() method for Dep- and PrvIterator
|
|
- return the correct version arch for all+foreign, too
The flag is interpreted at a few other places in different styles so
this commit ensures that the flag check is consistent everywhere
(checking for Same in flag style is a bit too much as it isn't used
in combination with others anyway, but who knows and just for
consistency)
|
|
|
|
|
|
|
|
pseudo-package related and/or private
|
|
- clean up mess with the "all" handling in MultiArch to
fix LP: #733741 cleanly for everyone now
|
|
|
|
- return "all" instead of native architecture without breaking the abi
(too much) by extending enum instead of using bitflags (LP: #733741)
With the next abi break that enum should be a char bitflag instead
|
|
Multi-Arch; instead, Arch: all packages only satisfy dependencies for
the native arch, except where the Arch: all package is declared
Multi-Arch: foreign. (Closes: #613584)
This has the sideeffect that arch:all packages internally show up as
coming from the native arch - so packages with the architecture "all"
doesn't exist any longer in the pkgcache
|
|
- do not segfault if cache is not build (Closes: #254770)
|
|
- change the debug outputs to display also arch of the
package and version dependencies information
|
|
- prefer non-virtual packages in FindPreferredPkg
|
|
|
|
- make the used MMap moveable (and therefore dynamic resizeable) by
applying (some) mad pointer magic (Closes: #195018)
|
|
- add another round of const& madness as the previous round accidently
NOT override the virtual GetCandidateVer() method (Closes: #587725)
|
|
|
|
- add simple wrapper around std::set for cache structures
|
|
- get the best matching arch package from a group with FindPreferredPkg
|
|
|
|
- let pkgCache::Iterator inherent std::iterator
|
|
|
|
(0 instead of the HeaderP) This breaks the Cache Validation functionality
as the end() doesn't test for NULL.
(The fault was introduced with the rewriting of the CacheIterators)
|
|
responseable for displaying a package name and the architecture in a
uniform way. Pretty option can be used to not append the architecture if
it is the native architecture or all - and use it all over the place in
the commandline tools.
|
|
Arch all packages are now represented by arch depending packages which
all depend on a package with the same name and the special arch "all".
This packages has NO dependencies, but beside this the same information.
It is the only package which has a size, the arch depending ones all
have a zero size. While the arch depending pseudo packages are used
for dependency resolution the arch "all" package is used for downloading
and ordering of the package.
|
|
with previous usecases. You now need to requested with Arch(true) the
return of the architecture this version (and pseudo package) was created for.
|
|
the information from the parent package now
|
|
for creating the dependencies needed for our groups:
For now for all groups only one package can be installed at the same
time which conflicts with each other packages in the group.
The exceptions are architecture all package.
Also, the Multi-Arch field is now parsed, but not used for now.
|
|
infrastructor for packages.
APT is now aware of the fact that a package A in architecture X can't
satisfy a dependency on package A in architecture Y - to handle these
packages are now identified by name and architecture, so different
architectures of the same package are handled internally as completly
different packages. This is great for pinning, dependency checking and
in many other situations, but sometimes we need to know which archs are
available for a given package: Here Groups come to our rescue!
|
|
This should not change the public interface, but it removes the
friend connection between the iterators and pkgcache as it is unused.
|
|
* add hook for MarkInstall and MarkDelete (closes: #470035)
* add the various foldmarkers in apt-pkg & cmdline (no code change)
* versions with a pin of -1 shouldn't be a candidate (Closes: #355237)
* prefer mmap as memory allocator in MMap instead of a static char
array which can (at least in theory) grow dynamic
* eliminate (hopefully all) segfaults in pkgcachegen.cc and mmap.cc
which can arise if cache doesn't fit into the mmap (Closes: #535218)
* display warnings instead of errors if the parts dirs doesn't exist
* honor the dpkg hold state in new Marker hooks (closes: #64141)
|
