- Wine-gitlab - list.winehq.org

[PATCH v2 0/1] MR3598: ntdll: Add error and warning logs in more cases of memory exhaustion.
by Tim Clem (＠tclem) 17 Aug '23

17 Aug '23

In my opinion we're far too quiet about mmap and allocation failures that will likely be fatal to the application. It generally takes some close inspection of +virtual or even +relay to detect running out of address space, e.g.. In my testing the added logs here shouldn't be too noisy - I've avoided them in the `try` functions and in cases where an error code is normal. I made the cross-process allocation failure a warning, because the target process itself will presumably log an error if it goes wrong. I also explicitly logged errno in a few cases where we assume its value. That can help catch some exotic error cases. -- v2: ntdll: Add error and warning logs in more cases of memory exhaustion. https://gitlab.winehq.org/wine/wine/-/merge_requests/3598

2 1

[PATCH 0/1] MR308: vkd3d-shader/hlsl: Initialize all constant values in hlsl_fold_constant_swizzles().
by Francisco Casas (＠fcasas) 17 Aug '23

17 Aug '23

Even though the uninitialized values shouldn't be used in the output binary program, they still show up in it, and affect the checksum, so we better make them zeroes. This error was reported by valgrind: ``` libtool --mode=execute valgrind --track-origins=yes vkd3d-build/tests/shader_runner vkd3d/tests/hlsl/static-initializer.shader_test ``` ``` ==46180== Conditional jump or move depends on uninitialised value(s) ==46180== at 0x48D98C7: parse_dxbc.isra.0 (dxbc.c:182) ==46180== by 0x48DA044: vkd3d_shader_parse_dxbc (dxbc.c:308) ==46180== by 0x488D1B6: vkd3d_shader_parse_dxbc_source_type (vkd3d_shader_utils.h:37) ==46180== by 0x488D1B6: create_shader_stage.isra.0 (state.c:1988) ==46180== by 0x48926B6: d3d12_pipeline_state_init_graphics (state.c:3084) ==46180== by 0x4893A96: d3d12_pipeline_state_create_graphics (state.c:3280) ==46180== by 0x4878498: d3d12_device_CreateGraphicsPipelineState (device.c:2619) ==46180== by 0x1FFEFFECC7: ??? ==46180== by 0xE8: ??? ==46180== by 0x47: ??? ==46180== by 0x61: ??? ==46180== by 0x660066000000023: ??? ==46180== by 0x661066100000044: ??? ==46180== Uninitialised value was created by a stack allocation ==46180== at 0x48F3FF0: hlsl_fold_constant_swizzles (hlsl_constant_ops.c:1010) ``` Thank you valgrind! :smile: -- https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/308

2 1

[PATCH 0/1] MR3598: ntdll: Add error and warning logs in more cases of memory exhaustion.
by Tim Clem (＠tclem) 17 Aug '23

17 Aug '23

In my opinion we're far too quiet about mmap and allocation failures that will likely be fatal to the application. It generally takes some close inspection of +virtual or even +relay to detect running out of address space, e.g.. In my testing the added logs here shouldn't be too noisy - I've avoided them in the `try` functions and in cases where an error code is normal. I made the cross-process allocation failure a warning, because the target process itself will presumably log an error if it goes wrong. I also explicitly logged errno in a few cases where we assume its value. That can help catch some exotic error cases. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/3598

2 1

[PATCH v6 0/1] MR3581: gdiplus: add support for GdipPathIterNextPathType
by Bartosz Kosiorek (＠gang65) 17 Aug '23

17 Aug '23

I have covered implementation with additional test cases. -- v6: gdiplus: add support for GdipPathIterNextPathType https://gitlab.winehq.org/wine/wine/-/merge_requests/3581

4 3

Re: [PATCH v3 0/5] MR3572: mf: Make session_get_node_object() more robust.
by Nikolay Sivov (＠nsivov) 17 Aug '23

17 Aug '23

I think it's better to have media engine patches in a separate MR, after making sure they work on Windows first. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/3572#note_42634

1 0

Re: [PATCH v3 0/5] MR3572: mf: Make session_get_node_object() more robust.
by Nikolay Sivov (＠nsivov) 17 Aug '23

17 Aug '23

Nikolay Sivov (@nsivov) commented about dlls/mf/session.c: > > /* fallthrough */ > case SESSION_STATE_PAUSED: > + case SESSION_STATE_STARTED: > + if (session->state == SESSION_STATE_STARTED && !(session->caps & MFSESSIONCAP_SEEK)) > + { > + WARN("Seeking is not supported for this session.\n"); > + session_command_complete(session); > + return; > + } Why does it check for STARTED state? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/3572#note_42632

1 0

Re: [PATCH v3 0/5] MR3572: mf: Make session_get_node_object() more robust.
by Nikolay Sivov (＠nsivov) 17 Aug '23

17 Aug '23

Nikolay Sivov (@nsivov) commented about dlls/mf/session.c: > } > } > > + if (session->state == SESSION_STATE_STARTED) > + { > + session_flush_transform_output_nodes(session); > + session_set_source_output_nodes_seeking(session); > + } Same here, isn't it possible to seek from paused state too? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/3572#note_42633

1 0

Re: [PATCH v3 0/5] MR3572: mf: Make session_get_node_object() more robust.
by Nikolay Sivov (＠nsivov) 17 Aug '23

17 Aug '23

Nikolay Sivov (@nsivov) commented about dlls/mf/session.c: > return hr; > } > > +static void session_flush_transform_output_nodes(struct media_session *session) This could be call session_flush_nodes(), or flush_topology/flush_pipeline, no need to make the name that specific. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/3572#note_42631

1 0

[PATCH 0/1] MR3587: mshtml: Make skip_prefix static in OmNavigator_get_appVersion.
by Alex Henrie (＠alexhenrie) 17 Aug '23

17 Aug '23

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/3587

4 6

[PATCH v6 0/3] MR2168: win32u: Match second name when enum fonts.
by Keming Liang (＠Keming) 17 Aug '23

17 Aug '23

the patch can resolve the issue: https://bugs.winehq.org/show_bug.cgi?id=54483 and can also solve many crashed issue related to font(MS Shell Dlg etc.) -- v6: add a skip log when there is no simsun font in test env. https://gitlab.winehq.org/wine/wine/-/merge_requests/2168

3 5