RTC_RD_TIME

RTC_RD_TIME

[Jorge Almeida]

A function to read the hardware clock (the battery powered thingy in
the motherboard):

(man 4 rtc)

void read_hw(struct rtc_time* rtctime){
   int rtcfd=open("/dev/rtc0" , O_RDONLY);
   ioctl(rtcfd, RTC_RD_TIME, rtctime);
   close(rtcfd);
}

(error checking omitted)

The problem:
warning: overflow in implicit constant conversion [-Woverflow]

(an arrow pointing to RTC_RD_TIME)

The same program compiles without warnings with glibc and dietlibc.

Headers:

#define _DEFAULT_SOURCE
#include <unistd.h>
#include <stdlib.h>
#include <stdint.h>
#include <string.h>
#include <sys/types.h>
#include <sys/wait.h>
#include <sys/time.h>
#include <time.h>
#include <linux/unistd.h>
#include <math.h>
#include <sys/ioctl.h>
#include <sys/stat.h>
#include <linux/vt.h>
#include <linux/rtc.h>
#include <fcntl.h>
#include <errno.h>
#include <stdio.h>

TIA

Jorge Almeida

Re: RTC_RD_TIME

[Rich Felker]

On Tue, Aug 29, 2017 at 10:28:14PM +0100, Jorge Almeida wrote:
> A function to read the hardware clock (the battery powered thingy in
> the motherboard):
> 
> (man 4 rtc)
> 
> void read_hw(struct rtc_time* rtctime){
>    int rtcfd=open("/dev/rtc0" , O_RDONLY);
>    ioctl(rtcfd, RTC_RD_TIME, rtctime);
>    close(rtcfd);
> }
> 
> (error checking omitted)
> 
> The problem:
> warning: overflow in implicit constant conversion [-Woverflow]
> 
> (an arrow pointing to RTC_RD_TIME)
> 
> The same program compiles without warnings with glibc and dietlibc.

I don't know any good fix. The problem is that the glibc (and others')
ioctl function has a signature mismatching the standard one, with an
unsigned argument instead of signed, and the macro values are outside
the range of signed int. Fortunately the warning is harmless but
obviously it can impact builds with -Werror or policy about warnings..

Rich

Re: RTC_RD_TIME

[Jorge Almeida]

On Wed, Aug 30, 2017 at 2:43 AM, Rich Felker <dalias@libc.org> wrote:
> On Tue, Aug 29, 2017 at 10:28:14PM +0100, Jorge Almeida wrote:

>>    ioctl(rtcfd, RTC_RD_TIME, rtctime);

>> The problem:
>> warning: overflow in implicit constant conversion [-Woverflow]
>>

>>
>> The same program compiles without warnings with glibc and dietlibc.
>

> the range of signed int. Fortunately the warning is harmless but

Good enough. Thanks.

Jorge

Re: RTC_RD_TIME

[Alexander Monakov]

On Tue, 29 Aug 2017, Rich Felker wrote:
> I don't know any good fix. The problem is that the glibc (and others')
> ioctl function has a signature mismatching the standard one, with an
> unsigned argument instead of signed, and the macro values are outside
> the range of signed int. Fortunately the warning is harmless but
> obviously it can impact builds with -Werror or policy about warnings..

Type confusion causes musl to wrongly pass sign-extended value to the
kernel (instead of zero-extended). That's a real issue, and it's a bit
disappointing you have not mentioned it.

(BSD flavours as well as Linux have 'unsigned long' there, so I don't
understand why it was ok for POSIX to document request type as 'int';
wouldn't it be appropriate for musl to just follow existing practice
here rather than the letter of the standard?)

Without changing the prototype, I think a possible fix is

#ifdef __GNUC__
#define ioctl(fd, req, ...) ioctl(fd, (int)(req), ##__VA_ARGS__)
#endif

(plus a fix in ioctl.c for the sign extension issue)

although I don't understand how the quoted warning appears; I can
produce another one with -Wsign-conversion, but not this.

Alexander

Re: RTC_RD_TIME

[Jorge Almeida]

On Wed, Aug 30, 2017 at 1:55 PM, Alexander Monakov <amonakov@ispras.ru> wrote:


> although I don't understand how the quoted warning appears; I can
> produce another one with -Wsign-conversion, but not this.
>

CC=/opt/bin/musl-gcc -static
FLAGS=-Os -march=native -fomit-frame-pointer -pipe -Wall -pedantic
-falign-functions=1 -falign-jumps=1 -falign-loops=1 -fno-unwind-tables
-fdata-sections -ffunction-sections -Wl,--gc-sections
-fno-asynchronous-unwind-tables -std=c99

(CC) $(FLAGS) -o foo foo.c

Jorge

Re: RTC_RD_TIME

[Alexander Monakov]

Thanks, that clears it up. It needs -pedantic and a 64-bit target.

void f(int);
void g()
{
  f(1u<<31|sizeof 0);
}

So casting sizes to 'int' or such in implementation of _IO[RW]* macros
would avoid this particular warning too (but not -Wsign-conversion).

Alexander