git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3610 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3609 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3608 d4fa192b-c00b-0410-8231-f00ffab90ce4

Merge cache_vrt_acl.c into cache_vrt.c, one one-line function is silly
in a source-file.

(All the work is done by vcc generated code now)



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3607 d4fa192b-c00b-0410-8231-f00ffab90ce4

Rework the binary heap, we use for expiry processing, to deal more
gracefully with large number of objects.

Previously we kept all objects in a single array which resultined
in increasingly infrequent but increasingly demanding calls to
calloc(3) with the consequent massive memory copies.  We also did
not release memory again if unused.

Now we stripe the array into rows of 64k objects each.

This number is a compromise between space wastage, max 1MB on a
64bit machine, and the desire to not add and delete rows all the
time.  With 64k objects in a row, even on a very busy server would
only add a new row every 5...10 seconds during ramp up.

Delete unused rows, but keep a hysteresis of an entire empty row
to avoid silly add-delete-add-delete-add-delete behaviour at row
boundaries.

Streamline some of the functions a bit.

Fixes #210



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3606 d4fa192b-c00b-0410-8231-f00ffab90ce4

Have I mentioned that I think POSIX is a bunch of amateurs ?

There is no NULL value for pthread_t and you have to call a function
to compare them.

Remember to set lock "held" in CondWait



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3605 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3604 d4fa192b-c00b-0410-8231-f00ffab90ce4

Take the full step and wrap all our mutex operations in proper C functions
instead of increasingly unwieldy macros.

Amongst other things, this will make it much easier to do lock
profiling, contest statistics, asserts etc.

This commit is largely mechanically generated and should not result in
any changed functionality.

Locks retain the "mtx" monicker, as a reminder that they are mutexes.

No performance impact expected.



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3603 d4fa192b-c00b-0410-8231-f00ffab90ce4

Fixes #368

r3389 | phk | 2008-11-11 21:40:37 +0100 (ti., 11 nov. 2008) | 4 lines

Fix this test-case to not rely on being able to compile /dev/null
now that this is no longer possible.




git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3602 d4fa192b-c00b-0410-8231-f00ffab90ce4

Change the .tcl script corresponding to r3364, fixes build failure



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3601 d4fa192b-c00b-0410-8231-f00ffab90ce4

Unfortunately the thread support in FlexeLint is not quite up to our
standards of creative locking.



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3600 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3599 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3598 d4fa192b-c00b-0410-8231-f00ffab90ce4

Move the || true bit of invoking tclsh to the Makefile, since
missing would otherwise not pass the file name to tclsh.

Thanks to des for spotting this.



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3597 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3596 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3595 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3594 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3593 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3592 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3591 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3590 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3589 d4fa192b-c00b-0410-8231-f00ffab90ce4

edits to bases from generating huge diffs.



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3588 d4fa192b-c00b-0410-8231-f00ffab90ce4

Clean up some of the fallout from the Solaris patch - mostly configure
script breakage.  In particular, disable the sendfile() check, as Solaris
doesn't have a working sendfile() any more than Linux does.  Actually, the
Solaris sendfile code was never compiled or used, because the #ifdefs in
cache_pool.c were all wrong.



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3587 d4fa192b-c00b-0410-8231-f00ffab90ce4

Add "config.h" where it's missing.



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3586 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3585 d4fa192b-c00b-0410-8231-f00ffab90ce4

Strictly speaking, we should test $target, not $host - although in practice,
varnish is not cross-buildable anyway due to the SO_RCVTIMEO / SO_SNDTIMEO
tests.



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3584 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3583 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3582 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3581 d4fa192b-c00b-0410-8231-f00ffab90ce4

I need a printout for a code-walkthrough and this commit wraps lines,
removes trailing space, space before TAB and other iregularities.

Some generated files have been compacted a bit.

A few changes to comments along the way.



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3580 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3579 d4fa192b-c00b-0410-8231-f00ffab90ce4

Fix up the spec file so it actually includes the workaround for make check on ppc64.

Original commit message:

* Thu Oct 16 2008 Ingvar Hagelund <ingvar@linpro.no> - 2.0-2
- Readded the debugflag patch. It's so practical
- Added a strange workaround for make check on ppc64



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3578 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3577 d4fa192b-c00b-0410-8231-f00ffab90ce4

Be more stringent about our timeout: fail even if we did get a
socket, but took to.



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3563 d4fa192b-c00b-0410-8231-f00ffab90ce4

Assert fcntl(2) have not failed.



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3562 d4fa192b-c00b-0410-8231-f00ffab90ce4

Add a missing error check.

Fixes #409



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3561 d4fa192b-c00b-0410-8231-f00ffab90ce4

Changed rpm summary string as requested by the Fedora project



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3560 d4fa192b-c00b-0410-8231-f00ffab90ce4

Fix typo (s/timeout/interval/) in default parameters for backend health

Thanks to Jonny @ globo for noticing.



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3559 d4fa192b-c00b-0410-8231-f00ffab90ce4

Update license to remove the advertising clause, reflecting similar change
in the FreeBSD original.

Approved by:    des



git-svn-id: http://www.varnish-cache.org/svn/branches/2.0@3558 d4fa192b-c00b-0410-8231-f00ffab90ce4