[Qemu-devel] [RFC, PATCH 2/2] bt-sdp: Fix if statement with empty body, spotted by clang

[Qemu-devel] [RFC, PATCH 2/2] bt-sdp: Fix if statement with empty body, spotted by clang

[Blue Swirl]

Fix clang error:
  CC    bt-sdp.o
/src/qemu/hw/bt-sdp.c:174:17: error: if statement has empty body [-Wempty-body]
    if (len > 1);

However, fixing this means that some code that was previously
ignored by the compiler now gets compiled, resulting in this error:
  CC    bt-sdp.o
cc1: warnings being treated as errors
/src/qemu/hw/bt-sdp.c: In function 'sdp_svc_search':
/src/qemu/hw/bt-sdp.c:184: error: 'max' may be used uninitialized in
this function

I could not figure out how to calculate max, so I just invented some
figure to suppress the error.

Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
---
 hw/bt-sdp.c |    4 +++-
 1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/hw/bt-sdp.c b/hw/bt-sdp.c
index b8732d0..c0bfd4d 100644
--- a/hw/bt-sdp.c
+++ b/hw/bt-sdp.c
@@ -171,12 +171,14 @@ static ssize_t sdp_svc_search(struct
bt_l2cap_sdp_state_s *sdp,
     } else
         start = 0;

-    if (len > 1);
+    if (len > 1) {
         return -SDP_INVALID_SYNTAX;
+    }

     /* Output the results */
     len = 4;
     count = 0;
+    max = INT_MAX; /* XXX: probably horribly incorrect */
     end = start;
     for (i = 0; i < sdp->services; i ++)
         if (sdp->service_list[i].match) {
-- 
1.6.2.4

[Qemu-devel] Re: [RFC, PATCH 2/2] bt-sdp: Fix if statement with empty body, spotted by clang

[andrzej zaborowski]

On 18 April 2010 10:52, Blue Swirl <blauwirbel@gmail.com> wrote:
> Fix clang error:
>  CC    bt-sdp.o
> /src/qemu/hw/bt-sdp.c:174:17: error: if statement has empty body [-Wempty-body]
>    if (len > 1);
>
> However, fixing this means that some code that was previously
> ignored by the compiler now gets compiled, resulting in this error:
>  CC    bt-sdp.o
> cc1: warnings being treated as errors
> /src/qemu/hw/bt-sdp.c: In function 'sdp_svc_search':
> /src/qemu/hw/bt-sdp.c:184: error: 'max' may be used uninitialized in
> this function
>
> I could not figure out how to calculate max, so I just invented some
> figure to suppress the error.

Thanks, good catches, I applied fixes to both problems.

Turns out we were assigning the value of the "Max" parameter of the
request to the wrong variable, which was later overwritten anyway.  I
don't know what I was thinking...

Cheers