Hi Rob,
On 6 August 2017 at 05:58, Rob Clark robdclark@gmail.com wrote:
On Sun, Aug 6, 2017 at 1:16 AM, Simon Glass sjg@chromium.org wrote:
On 4 August 2017 at 07:16, Bin Meng bmeng.cn@gmail.com wrote:
Hi Rob,
On Fri, Aug 4, 2017 at 8:51 PM, Rob Clark robdclark@gmail.com wrote:
stdin might not be set, which would cause iomux_doenv() to fail therefore causing probe_usb_keyboard() to fail. Furthermore if we do have iomux enabled, the sensible thing (in terms of user experience) would be to simply add ourselves to the list of stdin devices.
This fixes an issue with usbkbd on dragonboard410c with distro- bootcmd, where stdin is not set (so stdinname is null).
Signed-off-by: Rob Clark robdclark@gmail.com
v2: address Bin's review comments v3: fix fail with free()ing if usbkbd is already in stdin env variable pointed out by Simon
(the real v3 this time)
As I mentioned, it's the email title, not the commit title with version number. If you prefer format-patch, then use --subject-prefix="PATCH v3"
common/usb_kbd.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+)
Reviewed-by: Simon Glass sjg@chromium.org
Question below
diff --git a/common/usb_kbd.c b/common/usb_kbd.c index d2d29cc98f..d71eae61ec 100644 --- a/common/usb_kbd.c +++ b/common/usb_kbd.c @@ -517,7 +517,22 @@ static int probe_usb_keyboard(struct usb_device *dev)
stdinname = getenv("stdin");#if CONFIG_IS_ENABLED(CONSOLE_MUX)
Would this work:
if (CONFIG_IS_ENABLED(CONSOLE_MUX) { .. }
The #ifdef adds a code path that is not tested, so if possible we should try to use the compiler.
I gave this a quick try.. it would require either adding an unconditional #include <iomux.h> in usb_kbd.c or dropping the #ifdef CONFIG_CONSOLE_MUX guarding the #include <iomux.h> in console.h.. not sure which is preferred?
I don't think we should put #ifdefs around header files #includes so the first option seems best. There are still some header files in U-Boot which 'do stuff' like ensure that an option is set, etc. We should over time fix those up.
The #include in console.h seems wrong as well. It would be good to move that to console.c . .>
I suspect it would cause problems with -O0 but when I tried KCFLAGS=-O0 there were enough other unrelated compile errors, that I guess this isn't a legit use-case.
Yes we have had that for a while. I don't think people use it anymore.
If you want I can send a v4 that uses "if (CONFIG_IS_ENABLED(...))" (or a separate patch on top.. in which case, do you mind removing the "(v3)" in $subject when you apply, or do you prefer I spam list with yet another resend?) And in either case let me know what you prefer about iomux.h
I think a v4 patch is best. But you should not have the version in the subject there since it will end up in the commit. If you use patman it will produce the patch correctly.
BR, -R
char *devname = DEVNAME;char *newstdin = NULL;/** stdin might not be set yet.. either way, with console-mux the* sensible thing to do is add ourselves to the list of stdio* devices:*/if (stdinname && !strstr(stdinname, DEVNAME)) {newstdin = malloc(strlen(stdinname) + strlen(","DEVNAME) + 1);sprintf(newstdin, "%s,"DEVNAME, stdinname);stdinname = newstdin;} else if (!stdinname) {stdinname = devname;} error = iomux_doenv(stdin, stdinname);free(newstdin); if (error) return error;#else
Regards, Simon
Regards, Simon