Really the problem here is the return values of fit_load_[kernel|fdt]() are a
little all over the place. However we don't want to somehow get
through not having set kernel_end and having it just be random unused
data.

The compiler complained on an --enable-gcov build:

  In file included from ../../hw/core/loader-fit.c:20:
  /home/alex/lsrc/qemu.git/include/qemu/osdep.h: In function ‘load_fit’:
  /home/alex/lsrc/qemu.git/include/qemu/osdep.h:486:45: error: ‘kernel_end’ may 
be used uninitialized [-Werror=maybe-uninitialized]
    486 | #define ROUND_UP(n, d) ROUND_DOWN((n) + (d) - 1, (d))
        |                                             ^
  ../../hw/core/loader-fit.c:270:12: note: ‘kernel_end’ was declared here
    270 |     hwaddr kernel_end;
        |            ^~~~~~~~~~

Reviewed-by: Manos Pitsidianakis <manos.pitsidiana...@linaro.org>
Signed-off-by: Alex Bennée <alex.ben...@linaro.org>
Reviewed-by: Aleksandar Rikalo <arik...@gmail.com>
Message-Id: <20240705084047.857176-5-alex.ben...@linaro.org>

diff --git a/hw/core/loader-fit.c b/hw/core/loader-fit.c
index 9f20007dbb..7ccc9d5fbc 100644
--- a/hw/core/loader-fit.c
+++ b/hw/core/loader-fit.c
@@ -267,7 +267,7 @@ int load_fit(const struct fit_loader *ldr, const char 
*filename, void *opaque)
     const char *def_cfg_name;
     char path[FIT_LOADER_MAX_PATH];
     int itb_size, configs, cfg_off, off;
-    hwaddr kernel_end;
+    hwaddr kernel_end = 0;
     int ret;
 
     itb = load_device_tree(filename, &itb_size);
-- 
2.39.2


Reply via email to