On Mon, Oct 15, 2012 at 7:58 PM, Pete Batard wrote:
> On 15 October 2012 12:15, Sean McBride wrote:
> >>3) Look at how other software that uses -Wcast-align handles the same
> issue.
> >
> > I'm not exactly sure what you're asking...
>
> I was suggesting that, before we do anything, we look at w
On 15 October 2012 14:04, Sean McBride wrote:
> I see logging the 'long' as an advantage,
Could we please avoid using a "what-frigging-size-is-this?" long and
use an int##_t instead?
Regards,
/Pete
--
Don't let slow si
On 15 October 2012 12:15, Sean McBride wrote:
>>3) Look at how other software that uses -Wcast-align handles the same issue.
>
> I'm not exactly sure what you're asking...
I was suggesting that, before we do anything, we look at what projects that:
1. Chose to use -Wcast-align
2. Ran into the war
Hi Greg,
Sorry the previous email goes to the wrong linux kernel list.
Now usbutils git almost builds successfully out of the box under
Mac OS Xand Cygwin (using libusbx). Just wondering if you can
accept the minor fix for Mac OS X and suggest a way to fix
cygwin build.
For Cygwin, there is a co
Hi Greg,
Now usbutils git almost builds successfully out of the box under Mac OS X
and Cygwin (using libusbx). Just wondering if you can accept the minor
fix for Mac OS X and suggest a way to fix cygwin build.
For Cygwin, there is a conflict with Cygwin's w32api package.
DATADIR conflicts with M
On Sun, Oct 14, 2012 at 2:35 AM, Ludovic Rousseau
wrote:
>> I just tried your Xcode project file under Mac OS X Lion and it works
>> fine. There is a warning though.
>>
>> /Users/xiaofanc/work/libusbx/ludovic/libusbx/libusb/io.c:1877:35:
>> Implicit conversion loses integer precision: 'long' to 'i
john smith wrote:
>
> But I got another "performance issue" I need your help again.
>
> My test app is doing a loop, it seem that every time it loops, it
> takes SIGNIFICANT SLOWER than last time, if you look at debug trace,
> on the Throughput (Please see attached)
>
> 1) Throughput : 1.373530823
On Mon, 15 Oct 2012 18:52:18 +0200, Ludovic Rousseau said:
>I don't know why your code would be better. Maybe it is but you should
>explain.
>
>One problem with your code is that the value used for the debug may
>not be the value really used if an integer overflow occurs.
>You log "timeout_ms" but
On Mon, 15 Oct 2012 09:56:29 -0700, Orin Eman said:
>> 1) making libusb_control_transfer_get_setup not inline, so it's iffy cast
>> is hidden in a .c file
>>
>
>That should wait for the next API break as it would change the binary
>interface.
By adding a new exported symbol you mean? Of course
On Mon, Oct 15, 2012 at 9:15 AM, Sean McBride wrote:
> On Fri, 12 Oct 2012 23:25:14 +0100, Pete Batard said:
>
> >On 2012.10.12 22:06, Sean McBride wrote:
> >> 3) ???
> >
> >3) Look at how other software that uses -Wcast-align handles the same
> issue.
> >
> >There's probably hardly any piece of
2012/10/15 Sean McBride :
> On Sat, 13 Oct 2012 20:35:24 +0200, Ludovic Rousseau said:
>
>>> /Users/xiaofanc/work/libusbx/ludovic/libusbx/libusb/io.c:1877:35:
>>> Implicit conversion loses integer precision: 'long' to 'int'
>>> timeout_ms = (tv->tv_sec * 1000) + (tv->tv_usec / 1000);
>>
>>It
On Fri, 12 Oct 2012 23:25:14 +0100, Pete Batard said:
>On 2012.10.12 22:06, Sean McBride wrote:
>> 3) ???
>
>3) Look at how other software that uses -Wcast-align handles the same issue.
>
>There's probably hardly any piece of significant C code out there that's
>not going to cast a char* to str
On Sat, 13 Oct 2012 20:35:24 +0200, Ludovic Rousseau said:
>> /Users/xiaofanc/work/libusbx/ludovic/libusbx/libusb/io.c:1877:35:
>> Implicit conversion loses integer precision: 'long' to 'int'
>> timeout_ms = (tv->tv_sec * 1000) + (tv->tv_usec / 1000);
>
>It is fixed in https://github.com/Lud
On Wed, Oct 10, 2012 at 11:54 AM, Xiaofan Chen wrote:
> I am testing libftdi.git and somehow the example
> find_all_pp has segmentation fault problems under
> Windows, Mac OS X and Linux. I am not so sure
> if this is a libftdi problem or libusbx problem.
>
> The following is the log under Ubuntu
Hello Xiaofan
On 15/10/2012 10:15, Xiaofan Chen wrote:
>>
>> Ah - now that sounds very promising. What I still do not get is how the
>> WinUSB driver co-exists with the in-box mass storage stack? Can you load the
>> WinUSB driver for the same VID/PID but use the 2nd interface while mass
>> storage
On Mon, Oct 15, 2012 at 5:05 PM, g...@novadsp.com wrote:
>> It may not be as difficult as you think to make your firmware a
>> USB composite device. In you case, you need to add a second
>> interface (say Interface 1) without any endpoint associated
>> (basically just using the control endpoint).
Hello Xiaofan
On 13/10/2012 11:33, Xiaofan Chen wrote:
> It may not be as difficult as you think to make your firmware a
> USB composite device. In you case, you need to add a second
> interface (say Interface 1) without any endpoint associated
> (basically just using the control endpoint).
>
> In
17 matches
Mail list logo
