- 17 Mar, 2019 1 commit
-
-
Federico G. Schwindt authored
-
- 16 Mar, 2019 2 commits
-
-
Federico G. Schwindt authored
-
Federico G. Schwindt authored
-
- 15 Mar, 2019 12 commits
-
-
Pål Hermunn Johansen authored
-
Pål Hermunn Johansen authored
Happy new year!
-
Pål Hermunn Johansen authored
-
Pål Hermunn Johansen authored
-
Nils Goroll authored
This reverts commit de3e581e. tests/saintmode/test04.vtc from varnish-modules revealed that there may be use cases where it is implied that vcl_backend_error will leave a ttl=0 object and the sale object will remain in place. This can be resolved by: - return(abandon) for these cases or - making the HSH_Kill conditional on bo->fetch_objcore->ttl > 0 but I already has concerns regarding the impact of this change and have them even more now. We need time to ponder about the consequences and a documented inefficiency is better than unexpected behaviour in a release.
-
Nils Goroll authored
Our stack space canary in v00004.vtc fell off the twig on fedora rawhide i686 Thank you to @ingvarha for testing, reporting and all your support overall!
-
Poul-Henning Kamp authored
-
Nils Goroll authored
Fixes #2946 in the sense that we want to treat backend synth like any other object and kill the stale object it replaces. The two HSH_Kill calls could also be subsumed in vbf_fetch_thread(), but after an irc discussion, phk decided that he wanted to aim for even more consolidation later. The test checks that we remove the stale_oc, but not when abandoning the bereq.
-
Poul-Henning Kamp authored
-
Poul-Henning Kamp authored
-
Geoff Simmons authored
-
Federico G. Schwindt authored
-
- 14 Mar, 2019 16 commits
-
-
Federico G. Schwindt authored
-
Federico G. Schwindt authored
-
Federico G. Schwindt authored
-
Federico G. Schwindt authored
-
Federico G. Schwindt authored
-
Federico G. Schwindt authored
-
Geoff Simmons authored
-
Dridi Boukelmoune authored
I'm keeping that information in both the changes and upgrading documents to increase the chances that people will notice it. The imprecise "changes" statement predated the decision to drop Python 2 entirely and was spotted by Geoff.
-
Federico G. Schwindt authored
-
Federico G. Schwindt authored
-
Federico G. Schwindt authored
Reported by coverity.
-
Federico G. Schwindt authored
-
Geoff Simmons authored
Unless we think of something else to add, this version of the docs is suitable for the 6.2 release.
-
Geoff Simmons authored
-
Nils Goroll authored
-
Nils Goroll authored
Fixes #2943
-
- 13 Mar, 2019 3 commits
-
-
Federico G. Schwindt authored
Patch by @phk, test by yours truly.
-
Dridi Boukelmoune authored
-
Dridi Boukelmoune authored
-
- 12 Mar, 2019 6 commits
-
-
Dridi Boukelmoune authored
See dd28fc73 for the whole story.
-
Dridi Boukelmoune authored
Dealing with code style violations around the return keyword, I figured I could automate that fix, so I tried with my favourite refactoring tool and following those steps: $ cat >return.cocci <<EOF @@ expression rv; @@ - return rv; + return (rv); EOF $ spatch --dir . --in-place --sp-file return.cocci Unfortunately that leads to proper return statements having their return value enclosed twice, like `return ((rv))`. So a bit more steps to go: $ cat >rereturn.cocci <<EOF @@ expression rv; @@ - return ((rv)); + return (rv); EOF $ spatch --dir . --in-place --sp-file rreeturn.cocci At this point you don't want to touch bundled code, right? So let's take care of libvgz and shout out to teken that saw no changes at all! $ git checkout -- lib/libvgz/ The next step is to manually review all the changes, because Coccinelle is not aware of how we break lines and unbreaks them. That's quite tedious but I was waiting for someone and they are still late as I'm describing the whole process. Care was taken not to touch varnishtest to make it easier to sync with vtest. A dedicated patch will follow.
-
Dridi Boukelmoune authored
-
Poul-Henning Kamp authored
-
Federico G. Schwindt authored
-
Geoff Simmons authored
-