Message ID | 20231016071332.597654-1-fe@dev.tdt.de |
---|---|
Headers | show |
Series | ledtrig-tty: add additional tty state evaluation | expand |
Florian Eckert wrote on 2023-10-16 09:13: > Has complained about the following construct: Who is "Has" or who/what has complained? > drivers/leds/trigger/ledtrig-tty.c:362:3: error: a label can only be > part of a statement and a declaration is not a statement > > Hence move the variable definition to the beginning of the function. > > Reported-by: kernel test robot <lkp@intel.com> > Closes: > https://lore.kernel.org/oe-kbuild-all/202309270440.IJB24Xap-lkp@intel.com/ > Signed-off-by: Florian Eckert <fe@dev.tdt.de> > --- > @@ -124,8 +125,6 @@ static void ledtrig_tty_work(struct work_struct > *work) > > if (icount.rx != trigger_data->rx || > icount.tx != trigger_data->tx) { > - unsigned long interval = LEDTRIG_TTY_INTERVAL; > - Is this kernel test robot broken? I see no label definition here. And this variable declaration is at the start of a new block which does not even require C99 support. > led_blink_set_oneshot(trigger_data->led_cdev, &interval, > &interval, 0); Maarten
On 16. 10. 23, 11:12, Florian Eckert wrote: > > > On 2023-10-16 10:46, m.brock@vanmierlo.com wrote: >> Florian Eckert wrote on 2023-10-16 09:13: >>> Has complained about the following construct: >> >> Who is "Has" or who/what has complained? > > The test robot who does not agree with my change in the v1 patchset. Well, you should have put subject to that sentence, so that we can understand. And not to parse "Has" as a tool/person name ;). >>> drivers/leds/trigger/ledtrig-tty.c:362:3: error: a label can only be >>> part of a statement and a declaration is not a statement >>> >>> Hence move the variable definition to the beginning of the function. >>> >>> Reported-by: kernel test robot <lkp@intel.com> >>> Closes: >>> https://lore.kernel.org/oe-kbuild-all/202309270440.IJB24Xap-lkp@intel.com/ >>> Signed-off-by: Florian Eckert <fe@dev.tdt.de> >>> --- >>> @@ -124,8 +125,6 @@ static void ledtrig_tty_work(struct work_struct >>> *work) >>> >>> if (icount.rx != trigger_data->rx || >>> icount.tx != trigger_data->tx) { >>> - unsigned long interval = LEDTRIG_TTY_INTERVAL; >>> - >> >> Is this kernel test robot broken? > > The test robot does nothing wrong. > >> I see no label definition here. case is a label. >> And this variable declaration is at the start of a new block which >> does not >> even require C99 support. Nah. The block begins after the switch. So """ switch (X) { type var; case X: } would work. Moving the def after case is no longer at the block beginning. So just wrap put the case code in a block like we are used to: """ case X: { type var; } """. regards,
Florian Eckert wrote on 2023-10-16 11:12: > On 2023-10-16 10:46, m.brock@vanmierlo.com wrote: >> Florian Eckert wrote on 2023-10-16 09:13: >>> Has complained about the following construct: >> >> Who is "Has" or who/what has complained? > > The test robot who does not agree with my change in the v1 patchset. You don't have to explain to me, just fix the comment. >>> drivers/leds/trigger/ledtrig-tty.c:362:3: error: a label can only be >>> part of a statement and a declaration is not a statement >>> >>> Hence move the variable definition to the beginning of the function. >>> >>> Reported-by: kernel test robot <lkp@intel.com> >>> Closes: >>> https://lore.kernel.org/oe-kbuild-all/202309270440.IJB24Xap-lkp@intel.com/ >>> Signed-off-by: Florian Eckert <fe@dev.tdt.de> >>> --- >>> @@ -124,8 +125,6 @@ static void ledtrig_tty_work(struct work_struct >>> *work) >>> >>> if (icount.rx != trigger_data->rx || >>> icount.tx != trigger_data->tx) { >>> - unsigned long interval = LEDTRIG_TTY_INTERVAL; >>> - >> >> Is this kernel test robot broken? > > The test robot does nothing wrong. > >> I see no label definition here. >> And this variable declaration is at the start of a new block which >> does >> not even require C99 support. > > I made change in patch set v1, that moves the definition of the > variable > `interval` into the switch case statement. > https://lore.kernel.org/linux-leds/20230926093607.59536-3-fe@dev.tdt.de/ > The robot complained about this. > > So I decided to move the definition of the variable 'interval' to > function > head to make the test robot happy in the commit. So this commit > prepares > the code for my change. > > If it is more common, I can merge this patch [1] into the next patch > [2] > of this set. Yes, please. You're fixing a problem that does not exist yet (and never will), because the patch that introduces it is not yet applied. So fix the proposed patch instead of patching the patch. > [1] > https://lore.kernel.org/linux-leds/20231016071332.597654-4-fe@dev.tdt.de/ > [2] > https://lore.kernel.org/linux-leds/20231016071332.597654-5-fe@dev.tdt.de/ > > > Florian Maarten