Re: [PATCH 1/2] d3dx9_36/tests: Add D3DXSaveSurfaceToFileInMemory D3DXIFF_DDS tests (try 4)
2015-12-11 10:59 GMT+01:00 Alistair Leslie-Hughes leslie_alistair@hotmail.com:
The DDS_CAPS_TEXTURE | DDSCAPS_ALPHA caps check was to address testbot failures http://testbot.winehq.org/JobDetails.pl?Key=18888
Yes, I understand that. What I meant is that I don't get the header->caps == DDS_CAPS_TEXTURE case ever, except on Wine, so the check should be simply:
todo_wine ok(header->caps == (DDS_CAPS_TEXTURE | DDSCAPS_ALPHA), "Got unexpected caps %x.\n", header->caps);
Unless you can actually get that one to fail on Windows. If that's the case I'd like to know where does it fail for you.
2015-12-11 16:02 GMT+01:00 Matteo Bruni matteo.mystral@gmail.com:
2015-12-11 10:59 GMT+01:00 Alistair Leslie-Hughes < leslie_alistair@hotmail.com>:
The DDS_CAPS_TEXTURE | DDSCAPS_ALPHA caps check was to address testbot
failures
Yes, I understand that. What I meant is that I don't get the header->caps == DDS_CAPS_TEXTURE case ever, except on Wine,
Apparently the job shows that header->caps == DDS_CAPS_TEXTURE in 5 tests corresponding to winxp, win2003, vista, win2008 and win7 and DDS_CAPS_TEXTURE | DDSCAPS_ALPHA for 3 tests corresponding to win8 and win10. So if you never had header->caps == DDS_CAPS_TEXTURE, I'm wondering if you ever launched tests on windows below win8.
so the check should be simply:
todo_wine ok(header->caps == (DDS_CAPS_TEXTURE | DDSCAPS_ALPHA), "Got unexpected caps %x.\n", header->caps);
Unless you can actually get that one to fail on Windows. If that's the case I'd like to know where does it fail for you.
2015-12-11 18:10 GMT+01:00 Christian Costa titan.costa@gmail.com:
2015-12-11 16:02 GMT+01:00 Matteo Bruni matteo.mystral@gmail.com:
2015-12-11 10:59 GMT+01:00 Alistair Leslie-Hughes leslie_alistair@hotmail.com:
The DDS_CAPS_TEXTURE | DDSCAPS_ALPHA caps check was to address testbot failures http://testbot.winehq.org/JobDetails.pl?Key=18888
Yes, I understand that. What I meant is that I don't get the header->caps == DDS_CAPS_TEXTURE case ever, except on Wine,
Apparently the job shows that header->caps == DDS_CAPS_TEXTURE in 5 tests corresponding to winxp, win2003, vista, win2008 and win7 and DDS_CAPS_TEXTURE | DDSCAPS_ALPHA for 3 tests corresponding to win8 and win10. So if you never had header->caps == DDS_CAPS_TEXTURE, I'm wondering if you ever launched tests on windows below win8.
The test isn't run at all on the testbot on versions below Windows 8 because there is no d3dx9_36.dll installed or CreateDevice fails, as the logs show.
As I already mentioned in my reply to the previous version of this patch, I personally only ran the test on Windows 7 and XP (real hardware) which gave the same results as Win8 and Win10 from the testbot.
I don't have the previous mails at hand. Did you run Alistair's tests in your winxp and win7?
2015-12-11 18:23 GMT+01:00 Matteo Bruni matteo.mystral@gmail.com:
2015-12-11 18:10 GMT+01:00 Christian Costa titan.costa@gmail.com:
2015-12-11 16:02 GMT+01:00 Matteo Bruni matteo.mystral@gmail.com:
2015-12-11 10:59 GMT+01:00 Alistair Leslie-Hughes leslie_alistair@hotmail.com:
The DDS_CAPS_TEXTURE | DDSCAPS_ALPHA caps check was to address testbot failures http://testbot.winehq.org/JobDetails.pl?Key=18888
Yes, I understand that. What I meant is that I don't get the header->caps == DDS_CAPS_TEXTURE case ever, except on Wine,
Apparently the job shows that header->caps == DDS_CAPS_TEXTURE in 5 tests corresponding to winxp, win2003, vista, win2008 and win7 and DDS_CAPS_TEXTURE | DDSCAPS_ALPHA for 3 tests corresponding to win8 and win10. So if you never had header->caps == DDS_CAPS_TEXTURE, I'm wondering if
you
ever launched tests on windows below win8.
The test isn't run at all on the testbot on versions below Windows 8 because there is no d3dx9_36.dll installed or CreateDevice fails, as the logs show.
As I already mentioned in my reply to the previous version of this patch, I personally only ran the test on Windows 7 and XP (real hardware) which gave the same results as Win8 and Win10 from the testbot.
2015-12-11 18:43 GMT+01:00 Christian Costa titan.costa@gmail.com:
I don't have the previous mails at hand. Did you run Alistair's tests in your winxp and win7?
Yes, modified to print the actual value of the caps field.
FWIW: https://www.winehq.org/pipermail/wine-devel/2015-December/110767.html
P.S.: please bottom-post on the mailing list.
2015-12-11 18:59 GMT+01:00 Matteo Bruni matteo.mystral@gmail.com:
2015-12-11 18:43 GMT+01:00 Christian Costa titan.costa@gmail.com:
I don't have the previous mails at hand. Did you run Alistair's tests in your winxp and win7?
Yes, modified to print the actual value of the caps field.
FWIW: https://www.winehq.org/pipermail/wine-devel/2015-December/110767.html
Thanks. You said "I don't get the header->caps == DDS_CAPS_TEXTURE case ever, except on Wine". Do you mean that if you use native version of d3dx9_36 from both your winxp and win7, you get only DDS_CAPS_TEXTURE with Alistair's tests?
P.S.: please bottom-post on the mailing list.
Thanks. You said "I don't get the header->caps == DDS_CAPS_TEXTURE case ever, except on Wine". Do you mean that if you use native version of d3dx9_36 from both your winxp and win7, you get only DDS_CAPS_TEXTURE with Alistair's tests?
I haven't tried the native DLL on Wine but I expect it to give the same result as on Windows. It isn't an interesting case anyway. What I meant is that at this point only Wine (builtin) gives a different result, as far as I am aware, so that check should only allow the caps == (DDSCAPS_TEXTURE | DDSCAPS_ALPHA) case and be marked with todo_wine. Unless there is some Windows version that gives a different result but I don't see any evidence of that as of now.
Le 11/12/2015 19:27, Matteo Bruni a écrit :
Thanks. You said "I don't get the header->caps == DDS_CAPS_TEXTURE case ever, except on Wine". Do you mean that if you use native version of d3dx9_36 from both your winxp and win7, you get only DDS_CAPS_TEXTURE with Alistair's tests?
I haven't tried the native DLL on Wine but I expect it to give the same result as on Windows. It isn't an interesting case anyway. What I meant is that at this point only Wine (builtin) gives a different result, as far as I am aware, so that check should only allow the caps == (DDSCAPS_TEXTURE | DDSCAPS_ALPHA) case and be marked with todo_wine. Unless there is some Windows version that gives a different result but I don't see any evidence of that as of now.
I agree to mark the test as todo_wine with caps == (DDSCAPS_TEXTURE | DDSCAPS_ALPHA) for now. DDSCAPS_ALPHA is probably set depending on the D3D format of the surface so another patch can be written to set this value correctly with a additional test without alpha channel. The test used an A8R8G8B8 that's why DDSCAPS_ALPHA is set. My patch was originally written to fix Dragon Age screenshot for saved games. I didn't see the DDSCAPS_ALPHA set probably because the surface didn't have an alpha channel.
-
Christian Costa -
Matteo Bruni