[PATCH v5 0/3] MR3105: Add a few additional nullchecks (Coverity)

[Fabian Maurer (＠DarkShadow44)]

--
  v5: mmdevapi: Add nullcheck for parameter that can be null (Coverity)
      winealsa: Remove superflous check
      wineoss: Remove superflous check
 
https://gitlab.winehq.org/wine/wine/-/merge_requests/3105

[PATCH v5 1/3] wineoss: Remove superflous check

[Fabian Maurer]

From: Fabian Maurer <dark.shadow4(a)web.de>

---
 dlls/wineoss.drv/oss.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/dlls/wineoss.drv/oss.c b/dlls/wineoss.drv/oss.c
index 9a7a0fbb5d9..89e626f39d6 100644
--- a/dlls/wineoss.drv/oss.c
+++ b/dlls/wineoss.drv/oss.c
@@ -551,7 +551,7 @@ static HRESULT setup_oss_device(AUDCLNT_SHAREMODE share, int fd,
     if(ret == S_FALSE && !out)
         ret = AUDCLNT_E_UNSUPPORTED_FORMAT;
 
-    if(ret == S_FALSE && out){
+    if(ret == S_FALSE){
         closest->Format.nBlockAlign =
             closest->Format.nChannels * closest->Format.wBitsPerSample / 8;
         closest->Format.nAvgBytesPerSec =
-- 
GitLab


https://gitlab.winehq.org/wine/wine/-/merge_requests/3105

[PATCH v5 2/3] winealsa: Remove superflous check

[Fabian Maurer]

From: Fabian Maurer <dark.shadow4(a)web.de>

---
 dlls/winealsa.drv/alsa.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/dlls/winealsa.drv/alsa.c b/dlls/winealsa.drv/alsa.c
index 94fc3c6fa4b..aa1ad4abe6c 100644
--- a/dlls/winealsa.drv/alsa.c
+++ b/dlls/winealsa.drv/alsa.c
@@ -2014,7 +2014,7 @@ exit:
     if(params->result == S_FALSE && !params->fmt_out)
         params->result = AUDCLNT_E_UNSUPPORTED_FORMAT;
 
-    if(params->result == S_FALSE && params->fmt_out) {
+    if(params->result == S_FALSE) {
         closest->Format.nBlockAlign = closest->Format.nChannels * closest->Format.wBitsPerSample / 8;
         closest->Format.nAvgBytesPerSec = closest->Format.nBlockAlign * closest->Format.nSamplesPerSec;
         if(closest->Format.wFormatTag == WAVE_FORMAT_EXTENSIBLE)
-- 
GitLab


https://gitlab.winehq.org/wine/wine/-/merge_requests/3105

[PATCH v5 3/3] mmdevapi: Add nullcheck for parameter that can be null (Coverity)

[Fabian Maurer]

From: Fabian Maurer <dark.shadow4(a)web.de>

"out" got checked for null earlier, so we should do it here as well
Also added a test that crashes otherwise

Signed-off-by: Fabian Maurer <dark.shadow4(a)web.de>
---
 dlls/mmdevapi/client.c       |  3 +++
 dlls/mmdevapi/tests/render.c | 16 ++++++++++++++++
 2 files changed, 19 insertions(+)

diff --git a/dlls/mmdevapi/client.c b/dlls/mmdevapi/client.c
index 8146f8a53b4..c2107196be0 100644
--- a/dlls/mmdevapi/client.c
+++ b/dlls/mmdevapi/client.c
@@ -649,6 +649,9 @@ static HRESULT WINAPI client_IsFormatSupported(IAudioClient3 *iface, AUDCLNT_SHA
 
     wine_unix_call(is_format_supported, ¶ms);
 
+    if (params.result == S_FALSE && !out)
+        return AUDCLNT_E_UNSUPPORTED_FORMAT;
+
     if (params.result == S_FALSE)
         *out = ¶ms.fmt_out->Format;
     else
diff --git a/dlls/mmdevapi/tests/render.c b/dlls/mmdevapi/tests/render.c
index 8e000f03acb..3e24ac45944 100644
--- a/dlls/mmdevapi/tests/render.c
+++ b/dlls/mmdevapi/tests/render.c
@@ -139,11 +139,23 @@ static void test_audioclient(void)
     HRESULT hr;
     ULONG ref;
     WAVEFORMATEX *pwfx, *pwfx2;
+    WAVEFORMATEXTENSIBLE format_float_error;
     REFERENCE_TIME t1, t2;
     HANDLE handle;
     BOOL offload_capable;
     AudioClientProperties client_props;
 
+    format_float_error.Format.cbSize = sizeof(WAVEFORMATEXTENSIBLE);
+    format_float_error.Format.wFormatTag = WAVE_FORMAT_EXTENSIBLE;
+    format_float_error.Format.nAvgBytesPerSec = 384000;
+    format_float_error.Format.nBlockAlign = 8;
+    format_float_error.Format.nChannels = 2;
+    format_float_error.Format.nSamplesPerSec = 48000;
+    format_float_error.Format.wBitsPerSample = 32;
+    format_float_error.Samples.wValidBitsPerSample = 4;
+    format_float_error.dwChannelMask = 3;
+    format_float_error.SubFormat = KSDATAFORMAT_SUBTYPE_IEEE_FLOAT;
+
     hr = IMMDevice_Activate(dev, &IID_IAudioClient3, CLSCTX_INPROC_SERVER,
             NULL, (void**)&ac3);
     ok(hr == S_OK ||
@@ -258,6 +270,10 @@ static void test_audioclient(void)
            "IsFormatSupported(Exclusive) call returns %08lx\n", hr);
         hexcl = hr;
 
+        hr = IAudioClient_IsFormatSupported(ac, AUDCLNT_SHAREMODE_EXCLUSIVE, &format_float_error.Format, NULL);
+        ok(hr == S_OK || hr == AUDCLNT_E_UNSUPPORTED_FORMAT || hr == AUDCLNT_E_EXCLUSIVE_MODE_NOT_ALLOWED,
+                  "IsFormatSupported(Exclusive) call returns %08lx\n", hr);
+
         pwfx2 = (WAVEFORMATEX*)0xDEADF00D;
         hr = IAudioClient_IsFormatSupported(ac, AUDCLNT_SHAREMODE_EXCLUSIVE, pwfx, &pwfx2);
         ok(hr == hexcl, "IsFormatSupported(Exclusive) call returns %08lx\n", hr);
-- 
GitLab

https://gitlab.winehq.org/wine/wine/-/merge_requests/3105

[PATCH v5 3/3] mmdevapi: Add nullcheck for parameter that can be null (Coverity)

[Marvin]

Hi,

It looks like your patch introduced the new failures shown below.
Please investigate and fix them before resubmitting your patch.
If they are not new, fixing them anyway would help a lot. Otherwise
please ask for the known failures list to be updated.

The tests also ran into some preexisting test failures. If you know how
to fix them that would be helpful. See the TestBot job for the details:

The full results can be found at:
https://testbot.winehq.org/JobDetails.pl?Key=139565

Your paranoid android.


=== w8adm (32 bit report) ===

mmdevapi:
render.c:1358: Test failed: GetBuffer large (20671) at iteration 7

=== w1064v1507 (64 bit report) ===

mmdevapi:
render.c:1358: Test failed: GetBuffer large (20671) at iteration 3

=== w1064_adm (64 bit report) ===

mmdevapi:
render.c:1358: Test failed: GetBuffer large (22500) at iteration 4

[Fabian Maurer (＠DarkShadow44)]

On Wed Jul 26 17:29:17 2023 +0000, Fabian Maurer wrote:
> changed this line in [version 4 of the diff](/wine/wine/-/merge_requests/3105/diffs?diff_id=59476&start_sha=c4d570f6b39ce4ddfd7bd958461ffafefa3626cd#99913733944fc35298a259705e992a42b21541f5_653_653)
> Any news regarding this?

-- 
https://gitlab.winehq.org/wine/wine/-/merge_requests/3105#note_50920

[Nikolay Sivov (＠nsivov)]

On Sat Nov  4 14:58:36 2023 +0000, Fabian Maurer wrote:
> Any news regarding this?
> This now will produce different return value for out == NULL vs out != NULL, this does not look right.

-- 
https://gitlab.winehq.org/wine/wine/-/merge_requests/3105#note_50922

[Davide Beatrici (＠davidebeatrici)]

On Sat Nov  4 15:15:09 2023 +0000, Nikolay Sivov wrote:
> This now will produce different return value for out == NULL vs out !=
> NULL, this does not look right.
> Also see !3575.

-- 
https://gitlab.winehq.org/wine/wine/-/merge_requests/3105#note_50928

[Davide Beatrici (＠davidebeatrici)]

On Wed Jul 26 17:34:28 2023 +0000, Fabian Maurer wrote:
> Adjusted the code. It is a bit confusing, since the testbot and my
> machine behave differently... I had changes that failed only locally,
> and changes that seem to have only failed on the testbot. I hope this
> iteration one works everywhere.
> EDIT: Or do you want me to move the check into pulse? Because it's the
> only driver that doesn't have that
> ```
>     if(ret == S_FALSE && !out)
>         ret = AUDCLNT_E_UNSUPPORTED_FORMAT;
> ```
> logic.
> See !3575.

-- 
https://gitlab.winehq.org/wine/wine/-/merge_requests/3105#note_50930