February 2014 - Wine-Devel - list.winehq.org

Re: Rename "Version" field to "Earliest Affected Version" in Bugzilla?

by Vincent Povirk

> The original version is useful to test the original bug, it may be the > difference between resolving it as FIXED (was able to reproduce and no > longer happens) or WORKSFORME (could not reproduce in the reported > version and nowadays). Why is the original version better for this purpose than a later version that was tested and known to be affected by the bug?

Feb. 10, 2014

Re: Rename "Version" field to "Earliest Affected Version" in Bugzilla?

by Bruno Jesus

On Mon, Feb 10, 2014 at 10:14 PM, Vincent Povirk <madewokherd(a)gmail.com> wrote: >> That's a bad idea IMO. Version field as "version where the bug was >> found" provides useful information as an "upper bound", i.e. for v_N, >> the bug appeared in V_i...V_N-1, V_N >> If you change the meaning of the field, you'd have to examine the bug >> history to find that info. > > When would that information be useful apart from regressions? If it's > only useful for regressions, aren't we optimizing for an uncommon > case? The original version is useful to test the original bug, it may be the difference between resolving it as FIXED (was able to reproduce and no longer happens) or WORKSFORME (could not reproduce in the reported version and nowadays). Bruno

Feb. 10, 2014

Re: Rename "Version" field to "Earliest Affected Version" in Bugzilla?

by Vincent Povirk

> That's a bad idea IMO. Version field as "version where the bug was > found" provides useful information as an "upper bound", i.e. for v_N, > the bug appeared in V_i...V_N-1, V_N > If you change the meaning of the field, you'd have to examine the bug > history to find that info. When would that information be useful apart from regressions? If it's only useful for regressions, aren't we optimizing for an uncommon case? Generally, regressions will have that information in a comment anyway, and indirectly in the Regression SHA1 field.

Feb. 10, 2014

Re: Rename "Version" field to "Earliest Affected Version" in Bugzilla?

by Frédéric Delanoy

On Mon, Feb 10, 2014 at 6:32 PM, Vincent Povirk <madewokherd(a)gmail.com> wrote: > In fact, I would like to propose that we change the field's definition > to be the LATEST WINE VERSION in which the bug is known to be present. > This way, people retesting bugs won't have to flood them with comments > of "still present in version x.y.x", and one can quickly search to > find e.g. download bugs that weren't tested recently. That's a bad idea IMO. Version field as "version where the bug was found" provides useful information as an "upper bound", i.e. for v_N, the bug appeared in V_i...V_N-1, V_N If you change the meaning of the field, you'd have to examine the bug history to find that info. -- Frédéric Delanoy

Feb. 10, 2014

Re: Rename "Version" field to "Earliest Affected Version" in Bugzilla?

by Rosanne DiMesio

> On Mon, Feb 10, 2014 at 5:32 PM, Vincent Povirk <madewokherd(a)gmail.com> wrote: > > I've not seen any explanation in this > > thread of when any affected version is not good enough, and you really > > need to know the version the reporter was using. The only situation I can think of where this would be important is when the bug reporter claims it is a regression, but does not do a regression test. Someone who comes along later and is willing to run the regression test will want to know the first bad version. However, presumably that info would still be available in the bug's history field, so changing it in the version field shouldn't be a problem. -- Rosanne DiMesio <dimesio(a)earthlink.net>

Feb. 10, 2014

Re: [PATCH 1/5] wined3d: Introduce a sub resource structure.

by Stefan Dösinger

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am 2014-02-10 17:59, schrieb Henri Verbeet: > On 10 February 2014 13:28, Stefan Dösinger <stefan(a)codeweavers.com> > wrote: >> This structure will be a base structure for buffers, surfaces >> and volumes. It will take care of location management, opengl >> buffers, mapping and (at least for now) memory allocation. Sub >> resources will also store the pointer to the container resource. >> > I'm not convinced this is really the correct approach to this. Do you have any specific concerns or alternative ideas? -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQIcBAEBAgAGBQJS+R2bAAoJEN0/YqbEcdMwERQP/1eDnXTbI6Hzs5HK28TXGOlc wo3fCvLsp0ZCYOaWLLIU7qT6DJpxjfC6YqKgCRoPpjI9bhHL0hK/vdC89paN0CN6 Qtuq5vktDCaZ9RtuEaAeaQo005d9SxfDh5fLYFEIo+VAjRvb7cQLusiHaiei86Ut WetV6fM9NfHb1E+TOCIAnhkDb2mwhak4iGjIALcpNcRq8OEKOrdDNejk+jHB9bsK aUxU9j3om2EIc9FLHrncosH1GQE4rA4uBV82R3/vNaAGh2deFAuMDFNMJc0M9GrS 8zOPISJxhJzXOVT8fCEfKLcTt7+Aebj+7NIw9BjPP/eg8SBd4SQJ9iM3B2eeVM39 TnzbapXmuUDTbgf6wsebAx9M5S3w+m3yyLJQFC4UJ8lFO8pNqOVvF7gv0NScS7VN kBB/npqr9tLK+Wqa5LOenNYZOdqmUcvuPjDuzI/rrpeiPE3/QdDeFm4uDJ7Ea+9Y ahUk7OkXMTropCrn0j552OTkDasiSfDAgI95shZ0mKtlF+bB28Ef+/GL7cNTZ1bW 1aDb1GsWqsEuEoJI0I+s6Ui6QIo3DSJol2qPOgqo0U/E4qXzhg7z4tBZ8wlSxYdZ MmbXHnkikQPyiud82HgafUXqFzHYfBsJG/UK6OkwgARnLc8d5J1yWdtBzW8hDG79 NZA6/TaxAyI2OJhW/QUH =OQkX -----END PGP SIGNATURE-----

Feb. 10, 2014

Re: Rename "Version" field to "Earliest Affected Version" in Bugzilla?

by Jerome Leclanche

On Mon, Feb 10, 2014 at 5:32 PM, Vincent Povirk <madewokherd(a)gmail.com> wrote: >> IMO that field is important for testing bugs in the correct or at >> least a close wine version in order to reproduce the issue. > > But it should be equally useful for this as long as it's a version > where the bug is present. It doesn't matter whether it was the > originally reported version. I've not seen any explanation in this > thread of when any affected version is not good enough, and you really > need to know the version the reporter was using. I believe there is no > situation where that is important, and people changing it back are > just adding noise to the bug for no good reason. > > In fact, I would like to propose that we change the field's definition > to be the LATEST WINE VERSION in which the bug is known to be present. > This way, people retesting bugs won't have to flood them with comments > of "still present in version x.y.x", and one can quickly search to > find e.g. download bugs that weren't tested recently. > > As a bugzilla triager, I am extremely highly in favour of that. It makes a lot more sense for users, is a lot more easily explained and won't cause so much meta-drama. But then again, it's whatever is most useful to the devs that really matters. J. Leclanche

Feb. 10, 2014

Re: Rename "Version" field to "Earliest Affected Version" in Bugzilla?

by Vincent Povirk

> IMO that field is important for testing bugs in the correct or at > least a close wine version in order to reproduce the issue. But it should be equally useful for this as long as it's a version where the bug is present. It doesn't matter whether it was the originally reported version. I've not seen any explanation in this thread of when any affected version is not good enough, and you really need to know the version the reporter was using. I believe there is no situation where that is important, and people changing it back are just adding noise to the bug for no good reason. In fact, I would like to propose that we change the field's definition to be the LATEST WINE VERSION in which the bug is known to be present. This way, people retesting bugs won't have to flood them with comments of "still present in version x.y.x", and one can quickly search to find e.g. download bugs that weren't tested recently.

Feb. 10, 2014

Re: [PATCH 1/5] wined3d: Introduce a sub resource structure.

by Henri Verbeet

On 10 February 2014 13:28, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > This structure will be a base structure for buffers, surfaces and > volumes. It will take care of location management, opengl buffers, > mapping and (at least for now) memory allocation. Sub resources will > also store the pointer to the container resource. > I'm not convinced this is really the correct approach to this.

Feb. 10, 2014

Re: Rename "Version" field to "Earliest Affected Version" in Bugzilla?

by Bruno Jesus

On Mon, Feb 10, 2014 at 11:18 AM, Andrew Eikum <aeikum(a)codeweavers.com> wrote: > On Mon, Feb 10, 2014 at 04:28:48PM +0900, Dmitry Timoshkov wrote: >> Jerome Leclanche <adys.wh(a)gmail.com> wrote: >> > What is the reasoning behind that, really? What is the point of having >> > a version field that tells the developers absolutely nothing? >> >> One need to be a developer to comment on this I'd guess, but telling an >> exact used Wine version in the bug report is already a very good start. >> > > I don't know. As a developer, I've never used that field for anything, > and I find the "user changed version, admin changed it back" dance to > be a distraction in my Bugs mailbox. IMO that field is important for testing bugs in the correct or at least a close wine version in order to reproduce the issue. I vote for setting the field as admin only, and maybe a "set once" for users when they report the bug, it then can be reset by admins later. Regards, Bruno

Feb. 10, 2014