From mboxrd@z Thu Jan 1 00:00:00 1970 From: Yang Xu Date: Fri, 12 Jul 2019 14:00:11 +0800 Subject: [LTP] [PATCH V2] getrlimit03: adjust a bit of code to compatiable with mips32 In-Reply-To: References: <1562826026-158569-1-git-send-email-hongzhi.song@windriver.com> <408194681.32599099.1562831328308.JavaMail.zimbra@redhat.com> Message-ID: <5D2821EB.6070807@cn.fujitsu.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: ltp@lists.linux.it on 2019/07/12 13:46, Hongzhi, Song wrote: > > On 7/11/19 3:48 PM, Jan Stancek wrote: >> >> ----- Original Message ----- >>> Error info: >>> getrlimit03.c:104: FAIL: __NR_prlimit64(0) had rlim_cur = >>> ffffffffffffffff but __NR_getrlimit(0) had rlim_cur = 7fffffff >>> >>> According to kernel code: [arch/mips/include/uapi/asm/resource.h] >>> RLIM_INFINITY is set to 0x7fffffffUL instead of ULONG_MAX on mips32. >>> >>> /* >>> * SuS says limits have to be unsigned. >>> * Which makes a ton more sense anyway, >>> * but we keep the old value on MIPS32, >>> * for compatibility: >>> */ >>> #ifndef __mips64 >>> # define RLIM_INFINITY 0x7fffffffUL >>> #endif >>> >>> Adding conditional statement about mips to fix this. >>> >>> Signed-off-by: Hongzhi.Song >>> --- >>> testcases/kernel/syscalls/getrlimit/getrlimit03.c | 10 ++++++++-- >>> 1 file changed, 8 insertions(+), 2 deletions(-) >>> >>> diff --git a/testcases/kernel/syscalls/getrlimit/getrlimit03.c >>> b/testcases/kernel/syscalls/getrlimit/getrlimit03.c >>> index e4d56c4..1903558 100644 >>> --- a/testcases/kernel/syscalls/getrlimit/getrlimit03.c >>> +++ b/testcases/kernel/syscalls/getrlimit/getrlimit03.c >>> @@ -61,7 +61,13 @@ struct rlimit_ulong { >>> unsigned long rlim_cur; >>> unsigned long rlim_max; >>> }; >>> +#ifdef __mips >>> +#ifndef __mips64 >>> +const unsigned long RLIM_INFINITY_UL = 0x7fffffffUL; >>> +#else >>> const unsigned long RLIM_INFINITY_UL = ULONG_MAX; >>> +#endif >>> +#endif >> Hi, >> >> This will break every other arch, because it's now undefined >> everywhere except mips. > > > Is there a good way to filter mips32? > > I tried > > 65 #ifdef __mips > 66 #ifndef __mips64 > 67 const unsigned long RLIM_INFINITY_UL = 0x7fffffffUL; > 68 #endif > 69 const unsigned long RLIM_INFINITY_UL = ULONG_MAX; > 70 #else > 71 const unsigned long RLIM_INFINITY_UL = ULONG_MAX; > 72 #endif > > but mips32 thinks the RLIM_INFINITY_UL micro is redefined. Hi hongzhi You redefine RLIM_INFINITY_UL macro indeed when on mips32 because line 67 and 69. It should be as below: +#ifdef __mips +#ifndef __mips64 +const unsigned long RLIM_INFINITY_UL = 0x7fffffffUL; #endif +#else +const unsigned long RLIM_INFINITY_UL = ULONG_MAX; +#endif > > >> >>> static int getrlimit_ulong(int resource, struct rlimit_ulong *rlim) >>> { >>> @@ -101,8 +107,8 @@ static int compare_u64_ulong(int resource, uint64_t >>> val_u64, >>> { >>> if ((val_u64 > RLIM_INFINITY_UL && val_ul != RLIM_INFINITY_UL) || >>> (val_u64 <= RLIM_INFINITY_UL && val_ul != val_u64)) { >>> - tst_res(TFAIL, "__NR_prlimit64(%d) had %s = %" PRIx64 " but " >>> __NR_getrlimit_ulong_str "(%d) had %s = %lx", >>> - resource, kind, val_u64, >>> + tst_res(TFAIL, "SIGNED_GETRLIMIT = %d __WORDSIZE %d >>> ULONG_MAX = %lu >>> RLIM_INFINITY_UL = %lu __NR_prlimit64(%d) had %s = %" PRIx64 " but " >>> __NR_getrlimit_ulong_str "(%d) had %s = %lx", >>> + SIGNED_GETRLIMIT, __WORDSIZE, ULONG_MAX, >>> RLIM_INFINITY_UL, resource, >> I suggest to split it into another tst_res() message, line above is >> nearly 200 characters. > > > Sorry, I shouldn't append debug code to patch. > > --Hongzhi > > >> >> Regards, >> Jan >> >>> kind, val_u64, >>> resource, kind, val_ul); >>> return -1; >>> } >>> -- >>> 2.8.1 >>> >>> >>> -- >>> Mailing list info: https://lists.linux.it/listinfo/ltp >>> >