Re: [Qemu-devel] [PATCH v3 4/5] milkymist-softusb: Don't map RAM memory regions in the device itself

[Peter Maydell]

On 28 March 2013 18:31, Michael Walle <address@hidden> wrote:
> Am Donnerstag 28 März 2013, 18:55:59 schrieb Anthony Liguori:cc>
>>
>> Breaks the build:
>>
>> address@hidden qemu]$ make
>>   CC    lm32-softmmu/hw/lm32/../milkymist-softusb.o
>> /home/aliguori/git/qemu/hw/lm32/../milkymist-softusb.c: In function
>> ‘softusb_mouse_hid_datain’:
>> /home/aliguori/git/qemu/hw/lm32/../milkymist-softusb.c:183:38: error: ‘m’
>> may be used uninitialized in this function [-Werror=maybe-uninitialized]
>> /home/aliguori/git/qemu/hw/lm32/../milkymist-softusb.c:179:13: note: ‘m’
>> was declared here /home/aliguori/git/qemu/hw/lm32/../milkymist-softusb.c:
>> In function ‘softusb_kbd_hid_datain’:
>> /home/aliguori/git/qemu/hw/lm32/../milkymist-softusb.c:197:38: error: ‘m’
>> may be used uninitialized in this function [-Werror=maybe-uninitialized]
>> /home/aliguori/git/qemu/hw/lm32/../milkymist-softusb.c:193:13: note: ‘m’
>> was declared here cc1: all warnings being treated as errors
>> make[1]: *** [hw/lm32/../milkymist-softusb.o] Error 1
>
> are you sure, this patch breaks the build, or was is broken before?
>
> i'll send a patch soon.

My compiler doesn't complain. I suspect you may be right and
the problem was already there. Anyway I think that adding
a 'memset(buf, 0, len);' after the error_report() calls
in softusb_read_pmem()/softusb_read_dmem() will fix it.

-- PMM