[<prev] [next>] [day] [month] [year] [list]
Message-ID: <202109142222.1Ztk3E6a-lkp@intel.com>
Date: Tue, 14 Sep 2021 22:08:27 +0800
From: kernel test robot <lkp@...el.com>
To: Jakub Jelinek <jakub@...hat.com>
Cc: kbuild-all@...ts.01.org, linux-kernel@...r.kernel.org,
"Peter Zijlstra (Intel)" <peterz@...radead.org>,
Andrew Morton <akpm@...ux-foundation.org>,
Linux Memory Management List <linux-mm@...ck.org>
Subject: arch/mips/mm/tlbex.c:2599 check_pabits() warn: always true condition
'(fillbits >= ((__builtin_constant_p(0)) ?(((0) < 2) ?0:63 -
__builtin_clzll(0)):((4 <= 4)) ?__ilog2_u32(0):__ilog2_u64(0))) => (0-u32max
>= 0)'
tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
head: d0ee23f9d78be5531c4b055ea424ed0b489dfe9b
commit: 2f78788b55baa3410b1ec91a576286abe1ad4d6a ilog2: improve ilog2 for constant arguments
date: 9 months ago
config: mips-randconfig-m031-20210914 (attached as .config)
compiler: mips-linux-gcc (GCC) 11.2.0
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@...el.com>
New smatch warnings:
arch/mips/mm/tlbex.c:2599 check_pabits() warn: always true condition '(fillbits >= ((__builtin_constant_p(0)) ?(((0) < 2) ?0:63 - __builtin_clzll(0)):((4 <= 4)) ?__ilog2_u32(0):__ilog2_u64(0))) => (0-u32max >= 0)'
Old smatch warnings:
arch/mips/mm/tlbex.c:1741 iPTE_SW() warn: inconsistent indenting
vim +2599 arch/mips/mm/tlbex.c
c5b367835cfc7a Steven J. Hill 2015-02-26 2569
00bf1c691d082c Paul Burton 2015-09-22 2570 static void check_pabits(void)
00bf1c691d082c Paul Burton 2015-09-22 2571 {
00bf1c691d082c Paul Burton 2015-09-22 2572 unsigned long entry;
00bf1c691d082c Paul Burton 2015-09-22 2573 unsigned pabits, fillbits;
00bf1c691d082c Paul Burton 2015-09-22 2574
00bf1c691d082c Paul Burton 2015-09-22 2575 if (!cpu_has_rixi || !_PAGE_NO_EXEC) {
00bf1c691d082c Paul Burton 2015-09-22 2576 /*
00bf1c691d082c Paul Burton 2015-09-22 2577 * We'll only be making use of the fact that we can rotate bits
00bf1c691d082c Paul Burton 2015-09-22 2578 * into the fill if the CPU supports RIXI, so don't bother
00bf1c691d082c Paul Burton 2015-09-22 2579 * probing this for CPUs which don't.
00bf1c691d082c Paul Burton 2015-09-22 2580 */
00bf1c691d082c Paul Burton 2015-09-22 2581 return;
00bf1c691d082c Paul Burton 2015-09-22 2582 }
00bf1c691d082c Paul Burton 2015-09-22 2583
00bf1c691d082c Paul Burton 2015-09-22 2584 write_c0_entrylo0(~0ul);
00bf1c691d082c Paul Burton 2015-09-22 2585 back_to_back_c0_hazard();
00bf1c691d082c Paul Burton 2015-09-22 2586 entry = read_c0_entrylo0();
00bf1c691d082c Paul Burton 2015-09-22 2587
00bf1c691d082c Paul Burton 2015-09-22 2588 /* clear all non-PFN bits */
00bf1c691d082c Paul Burton 2015-09-22 2589 entry &= ~((1 << MIPS_ENTRYLO_PFN_SHIFT) - 1);
00bf1c691d082c Paul Burton 2015-09-22 2590 entry &= ~(MIPS_ENTRYLO_RI | MIPS_ENTRYLO_XI);
00bf1c691d082c Paul Burton 2015-09-22 2591
00bf1c691d082c Paul Burton 2015-09-22 2592 /* find a lower bound on PABITS, and upper bound on fill bits */
00bf1c691d082c Paul Burton 2015-09-22 2593 pabits = fls_long(entry) + 6;
00bf1c691d082c Paul Burton 2015-09-22 2594 fillbits = max_t(int, (int)BITS_PER_LONG - pabits, 0);
00bf1c691d082c Paul Burton 2015-09-22 2595
00bf1c691d082c Paul Burton 2015-09-22 2596 /* minus the RI & XI bits */
00bf1c691d082c Paul Burton 2015-09-22 2597 fillbits -= min_t(unsigned, fillbits, 2);
00bf1c691d082c Paul Burton 2015-09-22 2598
00bf1c691d082c Paul Burton 2015-09-22 @2599 if (fillbits >= ilog2(_PAGE_NO_EXEC))
00bf1c691d082c Paul Burton 2015-09-22 2600 fill_includes_sw_bits = true;
00bf1c691d082c Paul Burton 2015-09-22 2601
00bf1c691d082c Paul Burton 2015-09-22 2602 pr_debug("Entry* registers contain %u fill bits\n", fillbits);
00bf1c691d082c Paul Burton 2015-09-22 2603 }
00bf1c691d082c Paul Burton 2015-09-22 2604
:::::: The code at line 2599 was first introduced by commit
:::::: 00bf1c691d082c1945fdba032c03a9a82e9e7e61 MIPS: tlbex: Avoid placing software PTE bits in Entry* PFN fields
:::::: TO: Paul Burton <paul.burton@...tec.com>
:::::: CC: Ralf Baechle <ralf@...ux-mips.org>
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
Download attachment ".config.gz" of type "application/gzip" (38505 bytes)
Powered by blists - more mailing lists