December 2016 - Wine-Devel - list.winehq.org

Re: webservices: allocate enough memory for WS_XML_TEXT_TYPE_DOUBLE (Valgrind) (try 2)

by Alexandre Julliard

Hans Leidekker <hans(a)codeweavers.com> writes: > On Wed, 2016-12-21 at 09:38 +0100, Alexandre Julliard wrote: >> Austin English <austinenglish(a)gmail.com> writes: >> >> > On Wed, Dec 14, 2016 at 4:45 PM, Marvin <testbot(a)winehq.org> wrote: >> >> Thank you for your contribution to Wine! >> >> >> >> This is an automated notification to let you know that your patch has >> >> been reviewed and its status set to "Assigned". >> >> >> >> This means that the patch has been assigned to a reviewer. If the >> >> reviewer signs off on the patch, it will be committed, otherwise >> >> you'll receive advice on how to improve it. If you don't hear anything >> >> after a couple of weeks, please resend the patch. >> > >> > This shows in the tracker as assigned to Hans, but Hans signed off a week ago. >> >> Henri objected, and I agree, that we need to understand what's going on, >> and in particular why 48 would be the correct size. > > Valgrind ignores the rounding mode: > > Valgrind, setting RC_DOWN: > writer.c:364: Test failed: 2097: got 39 expected 22 > writer.c:366: Test failed: 2097: got <t>100000000000000.0156249999999999</t> expected <t>100000000000000</t> > > no Valgrind, not setting RC_DOWN: > writer.c:364: Test failed: 2097: got 39 expected 22 > writer.c:366: Test failed: 2097: got <t>100000000000000.0312499999999999</t> expected <t>100000000000000</t> > > 32 characters will be needed at most (excluding XML tags) because the > formatting code switches to exponential notation beyond 1^14 and precision > is capped at 0.0000000000000001, or 16 digits. Thanks for looking into this. Could you please resend a patch with the correct size and an appropriate comment? -- Alexandre Julliard julliard(a)winehq.org

Dec. 21, 2016

Re: webservices: allocate enough memory for WS_XML_TEXT_TYPE_DOUBLE (Valgrind) (try 2)

by Hans Leidekker

On Wed, 2016-12-21 at 09:38 +0100, Alexandre Julliard wrote: > Austin English <austinenglish(a)gmail.com> writes: > > > On Wed, Dec 14, 2016 at 4:45 PM, Marvin <testbot(a)winehq.org> wrote: > >> Thank you for your contribution to Wine! > >> > >> This is an automated notification to let you know that your patch has > >> been reviewed and its status set to "Assigned". > >> > >> This means that the patch has been assigned to a reviewer. If the > >> reviewer signs off on the patch, it will be committed, otherwise > >> you'll receive advice on how to improve it. If you don't hear anything > >> after a couple of weeks, please resend the patch. > > > > This shows in the tracker as assigned to Hans, but Hans signed off a week ago. > > Henri objected, and I agree, that we need to understand what's going on, > and in particular why 48 would be the correct size. Valgrind ignores the rounding mode: Valgrind, setting RC_DOWN: writer.c:364: Test failed: 2097: got 39 expected 22 writer.c:366: Test failed: 2097: got <t>100000000000000.0156249999999999</t> expected <t>100000000000000</t> no Valgrind, not setting RC_DOWN: writer.c:364: Test failed: 2097: got 39 expected 22 writer.c:366: Test failed: 2097: got <t>100000000000000.0312499999999999</t> expected <t>100000000000000</t> 32 characters will be needed at most (excluding XML tags) because the formatting code switches to exponential notation beyond 1^14 and precision is capped at 0.0000000000000001, or 16 digits.

Dec. 21, 2016

Re: webservices: allocate enough memory for WS_XML_TEXT_TYPE_DOUBLE (Valgrind) (try 2)

by Alexandre Julliard

Austin English <austinenglish(a)gmail.com> writes: > On Wed, Dec 14, 2016 at 4:45 PM, Marvin <testbot(a)winehq.org> wrote: >> Thank you for your contribution to Wine! >> >> This is an automated notification to let you know that your patch has >> been reviewed and its status set to "Assigned". >> >> This means that the patch has been assigned to a reviewer. If the >> reviewer signs off on the patch, it will be committed, otherwise >> you'll receive advice on how to improve it. If you don't hear anything >> after a couple of weeks, please resend the patch. > > This shows in the tracker as assigned to Hans, but Hans signed off a week ago. Henri objected, and I agree, that we need to understand what's going on, and in particular why 48 would be the correct size. -- Alexandre Julliard julliard(a)winehq.org

Dec. 21, 2016

Re: webservices: allocate enough memory for WS_XML_TEXT_TYPE_DOUBLE (Valgrind) (try 2)

by Austin English

On Wed, Dec 14, 2016 at 4:45 PM, Marvin <testbot(a)winehq.org> wrote: > Thank you for your contribution to Wine! > > This is an automated notification to let you know that your patch has > been reviewed and its status set to "Assigned". > > This means that the patch has been assigned to a reviewer. If the > reviewer signs off on the patch, it will be committed, otherwise > you'll receive advice on how to improve it. If you don't hear anything > after a couple of weeks, please resend the patch. This shows in the tracker as assigned to Hans, but Hans signed off a week ago. -- -Austin GPG: 14FB D7EA A041 937B

Dec. 20, 2016

Re: Can't build crosstests

by Fabian Maurer

On Tuesday, December 20, 2016 8:28:12 AM CET Andrew Eikum wrote: > I think your "-O0 -rdynamic" options are coming from someplace outside > of configure. Oh yes, you're right... I had a wrong CFLAGS environment variable set. Sorry for that, but thanks for the help!

Dec. 20, 2016

Re: Can't build crosstests

by Andrew Eikum

On Tue, Dec 20, 2016 at 02:37:13PM +0100, Fabian Maurer wrote: > On Tuesday, December 20, 2016 9:00:13 AM CET Jactry Zeng wrote: > > You can have more information about this failure in config.log. > Thanks, I somehow completely missed that... > > The error is: > >configure:8781: i686-w64-mingw32-gcc -c -g -O0 -rdynamic conftest.c >&5 > >i686-w64-mingw32-gcc: error: unrecognized command line option '-rdynamic' > I'm on Arch Linux with mingw-w64-gcc 6.2.1 > Seems like the script passes wrong options to gcc. > I think your "-O0 -rdynamic" options are coming from someplace outside of configure. I'm also on Arch Linux and I see: configure:8924: result: ccache i686-w64-mingw32-gcc configure:8940: checking whether ccache i686-w64-mingw32-gcc works configure:8953: ccache i686-w64-mingw32-gcc -c -g -O2 conftest.c >&5 I also noticed the configure line numbers are different, which seems strange as configure is tracked by git. The string 'rdynamic' doesn't appear in Wine's configure script. Andrew

Dec. 20, 2016

Re: Can't build crosstests

by Fabian Maurer

On Tuesday, December 20, 2016 9:00:13 AM CET Jactry Zeng wrote: > You can have more information about this failure in config.log. Thanks, I somehow completely missed that... The error is: >configure:8781: i686-w64-mingw32-gcc -c -g -O0 -rdynamic conftest.c >&5 >i686-w64-mingw32-gcc: error: unrecognized command line option '-rdynamic' I'm on Arch Linux with mingw-w64-gcc 6.2.1 Seems like the script passes wrong options to gcc.

Dec. 20, 2016

Re: user32/tests: Disable automatic shell restart during SetShellWindow tests.

by Hans Leidekker

On Tue, 2016-12-20 at 05:31 -0600, Marvin wrote: > === w864 (32 bit win) === > win.c:2305: Test failed: expected !100 > win.c:2305: Test failed: expected !100 > > === w864 (64 bit win) === > win.c:2305: Test failed: expected !100 > win.c:2305: Test failed: expected !100 This failure is also seen with a whitespace patch: https://testbot.winehq.org/JobDetails.pl?Key=27439

Dec. 20, 2016

Re: user32/tests: Disable automatic shell restart during SetShellWindow tests.

by Marvin

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=27438 Your paranoid android. === w864 (32 bit win) === win.c:2305: Test failed: expected !100 win.c:2305: Test failed: expected !100 === w864 (64 bit win) === win.c:2305: Test failed: expected !100 win.c:2305: Test failed: expected !100

Dec. 20, 2016

Re: Can't build crosstests

by Jactry Zeng

Hi Fabian, You can have more information about this failure in config.log. 2016-12-20 8:56 GMT+08:00 Fabian Maurer <dark.shadow4(a)web.de>: > Hello, > > I'm trying to build the tests for windows, but I can't get it working > anymore. > I have mingw installed and it does work for cross-compiling with > code-blocks, > but wine says it's broken. > The configure script fails with > >checking for i686-w64-mingw32-gcc... i686-w64-mingw32-gcc > >checking whether i686-w64-mingw32-gcc works... no > > An idea how to find out what's going wrong? > > > -- Regards, Jactry Zeng

Dec. 20, 2016