May 2021 - Wine-devel - list.winehq.org

[PATCH] po: Update Lithuanian translation
by Aurimas Fišeras 31 May '21

31 May '21

2 1

[PATCH 1/5] ws2_32/tests: Call getsockname() on the right socket.
by Zebediah Figura 31 May '21

31 May '21

Also fixes spurious test failures wherein the broken call would return ENOTSOCK. Signed-off-by: Zebediah Figura <z.figura12(a)gmail.com> --- dlls/ws2_32/tests/afd.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/dlls/ws2_32/tests/afd.c b/dlls/ws2_32/tests/afd.c index 97b35e1f3ea..3f03c14a94f 100644 --- a/dlls/ws2_32/tests/afd.c +++ b/dlls/ws2_32/tests/afd.c @@ -565,7 +565,7 @@ static void test_poll(void) ret = bind(client, (const struct sockaddr *)&bind_addr, sizeof(bind_addr)); ok(!ret, "got error %u\n", WSAGetLastError()); len = sizeof(addr); - ret = getsockname(listener, (struct sockaddr *)&addr, &len); + ret = getsockname(client, (struct sockaddr *)&addr, &len); ok(!ret, "got error %u\n", WSAGetLastError()); check_poll(client, event, AFD_POLL_WRITE); @@ -1046,7 +1046,7 @@ static void test_recv(void) ret = bind(client, (const struct sockaddr *)&bind_addr, sizeof(bind_addr)); ok(!ret, "got error %u\n", WSAGetLastError()); len = sizeof(addr); - ret = getsockname(listener, (struct sockaddr *)&addr, &len); + ret = getsockname(client, (struct sockaddr *)&addr, &len); ok(!ret, "got error %u\n", WSAGetLastError()); memset(buffer, 0xcc, sizeof(buffer)); -- 2.30.2

3 11

[PATCH resend 4/4] msxml3/tests: Test for methods that break lock with writer domdoc dest.
by Jefferson Carpenter 31 May '21

31 May '21

I actually don't like 4/4 as much since the 2 helper methods I added `begin_writing_to_domdoc` and `check_domdoc_lock_broken` pretty much arbitrarily factor out code. Nevertheless, I would be fine maintaining this test, so please let me know what you think. thanks, Jefferson

2 1

[PATCH resend 3/4] msxml3/tests: Test for disabled document methods with writer domdoc dest.
by Jefferson Carpenter 31 May '21

31 May '21

2 1

[PATCH 1/5] server: Don't change the status of an already terminated async.
by Zebediah Figura 31 May '21

31 May '21

Signed-off-by: Zebediah Figura <z.figura12(a)gmail.com> --- server/async.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) diff --git a/server/async.c b/server/async.c index 1ac5117edb9..a5fa61dc6e5 100644 --- a/server/async.c +++ b/server/async.c @@ -154,12 +154,7 @@ void async_terminate( struct async *async, unsigned int status ) { assert( status != STATUS_PENDING ); - if (async->status != STATUS_PENDING) - { - /* already terminated, just update status */ - async->status = status; - return; - } + if (async->status != STATUS_PENDING) return; /* already terminated */ async->status = status; if (async->iosb && async->iosb->status == STATUS_PENDING) async->iosb->status = status; @@ -457,7 +452,7 @@ static int cancel_async( struct process *process, struct object *obj, struct thr restart: LIST_FOR_EACH_ENTRY( async, &process->asyncs, struct async, process_entry ) { - if (async->status == STATUS_CANCELLED) continue; + if (async->status != STATUS_PENDING) continue; if ((!obj || (async->fd && get_fd_user( async->fd ) == obj)) && (!thread || async->thread == thread) && (!iosb || async->data.iosb == iosb)) -- 2.30.2

4 9

[PATCH resend 2/4] msxml3/tests: Test for disabled element methods with writer domdoc dest.
by Jefferson Carpenter 31 May '21

31 May '21

2 1

[PATCH v4 1/2] msxml3: Null check newChild in DOMDocument insertBefore.
by Jefferson Carpenter 31 May '21

31 May '21

Thanks for bearing with me. Regarding commit messages I was thinking DOMDocument is the name of a clsid, so that's the name of the object that should be mentioned. Element doesn't seem to be associated with a clsid, so I just said 'domelem'. Let me know if you would have a different opinion on that.

2 2

[PATCH v4 2/2] msxml3: Null check newNode in domelem insertBefore.
by Jefferson Carpenter 31 May '21

31 May '21

2 1

[PATCH resend 1/4] msxml3/tests: Test for disabled node methods with writer domdoc dest.
by Jefferson Carpenter 31 May '21

31 May '21

Feel free to comment only on 1/4, and I can extrapolate to 2/4 and 3/4. I would be fine with dropping this patch set, if you don't think it's worthwhile to test these methods. thanks, Jefferson

2 1

[PATCH] d3dcompiler: Use return value in read_dword.
by Jacek Caban 31 May '21

31 May '21

Fixes warnings like on clang with MSVC target, where enums may be signed: dlls/d3dcompiler_43/reflection.c:1160:22: warning: passing 'D3D_PRIMITIVE *' (aka 'enum D3D_PRIMITIVE *') to parameter of type 'DWORD *' (aka 'unsigned int *') converts between pointers to integer types with different sign [-Wpointer-sign] Signed-off-by: Jacek Caban <jacek(a)codeweavers.com> --- dlls/d3dcompiler_43/d3dcompiler_private.h | 8 +- dlls/d3dcompiler_43/reflection.c | 144 +++++++++++----------- dlls/d3dcompiler_43/utils.c | 14 +-- 3 files changed, 84 insertions(+), 82 deletions(-)

2 1