12 Jun
2023
12 Jun
'23
7:49 a.m.
Le 12/06/2023 à 09:17, Yann Sionneau a écrit :
Le 08/06/2023 à 10:39, Yann a écrit :
OK I'll just send a fix for that that applies on top of a revert of
08d46f1ce21e4ec51b2b1626beeaea6cbe7fdc6b
Regards,
--
Yann
June 8, 2023 2:05 AM, "Damien Le Moal"
<Damien.LeMoal(a)wdc.com> wrote:
At most I can see those warnings but it still builds fine:
librt/clock_nanosleep.c: In function 'clock_nanosleep':
librt/clock_nanosleep.c:43:22: warning: implicit declaration of
function 'LIBC_CANCEL_ASYNC'; did you mean 'LIBC_CANCEL_HANDLED'?
[-Wimplicit-function-declaration]
43 | int oldstate = LIBC_CANCEL_ASYNC ();
| ^~~~~~~~~~~~~~~~~
| LIBC_CANCEL_HANDLED
librt/clock_nanosleep.c:48:7: warning: implicit declaration of
function 'LIBC_CANCEL_RESET'; did you mean 'LIBC_CANCEL_HANDLED'?
[-Wimplicit-function-declaration]
48 | LIBC_CANCEL_RESET (oldstate);
| ^~~~~~~~~~~~~~~~~
| LIBC_CANCEL_HANDLED
Also, I tried to build with and without your patch and then I compared
the 2 resulting librt/clock_nanosleep.os object files: they show up
exactly the same.
I am guessing that now it works because of a newer compiler that does
a better job at optimizing because if I build clock_nanosleep.os with
-save-temps I can see the output of pre-processor being:
if (1)
r = ({ long _sys_result; { register long int _a7 __asm__ ("a7");
register long _a3 __asm__ ("a3"); long _a3tmp; register long _a2
__asm__ ("a2"); long _a2tmp; register long _a1 __asm__ ("a1"); long
_a1tmp; long _a0tmp; register long _a0 __asm__ ("a0"); _a0tmp = (long)
(clock_id); _a0 = _a0tmp; _a1tmp = (long) (flags); _a1 = _a1tmp;
_a2tmp = (long) (req); _a2 = _a2tmp; _a3tmp = (long) (rem); _a3 =
_a3tmp; _a7 = (115); __asm__ volatile ( "scall\n\t" : "=r" (_a0) :
"r"(_a7) , "r" (_a0), "r" (_a1), "r" (_a2),
"r" (_a3) : "memory");
_sys_result = _a0; } _sys_result; });
else
{
int oldstate = LIBC_CANCEL_ASYNC ();
r = ({ long _sys_result; { register long int _a7 __asm__ ("a7");
register long _a3 __asm__ ("a3"); long _a3tmp; register long _a2
__asm__ ("a2"); long _a2tmp; register long _a1 __asm__ ("a1"); long
_a1tmp; long _a0tmp; register long _a0 __asm__ ("a0"); _a0tmp = (long)
(clock_id); _a0 = _a0tmp; _a1tmp = (long) (flags); _a1 = _a1tmp;
_a2tmp = (long) (req); _a2 = _a2tmp; _a3tmp = (long) (rem); _a3 =
_a3tmp; _a7 = (115); __asm__ volatile ( "scall\n\t" : "=r" (_a0) :
"r"(_a7) , "r" (_a0), "r" (_a1), "r" (_a2),
"r" (_a3) : "memory");
_sys_result = _a0; } _sys_result; })
;
LIBC_CANCEL_RESET (oldstate);
}
So basically if (1) { } else { call to undefined function }.
I guess the `else` part is being optimized out and it silents the issue.
Regards,
Note that to test patches/regressions for risc-v
NOMMU builds, you
can simply
use buildroot and run:
make sipeed_maix_bit_defconfig
make
To check that it builds.
Hello Damien,
With buildroot master branch, I can build sipeed_maix_bit_defconfig
config with the reverted patch with no build error.
How can I reproduce the issue?