| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <Z4TBLhxvV6llCcWb@xsang-OptiPlex-9020>
Date: Mon, 13 Jan 2025 15:30:54 +0800
From: Oliver Sang <oliver.sang@...el.com>
To: Barry Song <21cnbao@...il.com>
CC: <lkp@...el.com>, <akpm@...ux-foundation.org>, <anshuman.khandual@....com>,
<baolin.wang@...ux.alibaba.com>, <bp@...en8.de>, <catalin.marinas@....com>,
<chrisl@...nel.org>, <dave.hansen@...ux.intel.com>, <david@...hat.com>,
<gshan@...hat.com>, <hpa@...or.com>, <ioworker0@...il.com>,
<kasong@...cent.com>, <kirill.shutemov@...ux.intel.com>,
<linux-arm-kernel@...ts.infradead.org>, <linux-kernel@...r.kernel.org>,
<linux-mm@...ck.org>, <mark.rutland@....com>, <mingo@...hat.com>,
<oe-kbuild-all@...ts.linux.dev>, <ryan.roberts@....com>,
<shahuang@...hat.com>, <tglx@...utronix.de>, <v-songbaohua@...o.com>,
<wangkefeng.wang@...wei.com>, <will@...nel.org>, <x86@...nel.org>,
<ying.huang@...el.com>, <yosryahmed@...gle.com>, <zhengtangquan@...o.com>,
<oliver.sang@...el.com>
Subject: Re: [PATCH 2/3] mm: Support tlbbatch flush for a range of PTEs
hi, Barry,
On Mon, Jan 13, 2025 at 01:56:26PM +1300, Barry Song wrote:
> On Mon, Jan 6, 2025 at 11:08 PM kernel test robot <lkp@...el.com> wrote:
> >
> > Hi Barry,
> >
> > kernel test robot noticed the following build errors:
> >
> > [auto build test ERROR on akpm-mm/mm-everything]
> >
> > url: https://github.com/intel-lab-lkp/linux/commits/Barry-Song/mm-set-folio-swapbacked-iff-folios-are-dirty-in-try_to_unmap_one/20250106-112638
> > base: https://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git mm-everything
> > patch link: https://lore.kernel.org/r/20250106031711.82855-3-21cnbao%40gmail.com
> > patch subject: [PATCH 2/3] mm: Support tlbbatch flush for a range of PTEs
> > config: riscv-randconfig-001-20250106 (https://download.01.org/0day-ci/archive/20250106/202501061736.FoHcInHJ-lkp@intel.com/config)
> > compiler: riscv64-linux-gcc (GCC) 14.2.0
> > reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20250106/202501061736.FoHcInHJ-lkp@intel.com/reproduce)
> >
> > If you fix the issue in a separate patch/commit (i.e. not just a new version of
> > the same patch/commit), kindly add following tags
> > | Reported-by: kernel test robot <lkp@...el.com>
> > | Closes: https://lore.kernel.org/oe-kbuild-all/202501061736.FoHcInHJ-lkp@intel.com/
> >
>
> Sorry. My bad, does the below diff fix the build?
yes, below diff fixes the build. thanks
Tested-by: kernel test robot <oliver.sang@...el.com>
>
> diff --git a/arch/riscv/include/asm/tlbflush.h b/arch/riscv/include/asm/tlbflush.h
> index 72e559934952..7f3ea687ce33 100644
> --- a/arch/riscv/include/asm/tlbflush.h
> +++ b/arch/riscv/include/asm/tlbflush.h
> @@ -61,7 +61,8 @@ void flush_pmd_tlb_range(struct vm_area_struct *vma, unsigned long start,
> bool arch_tlbbatch_should_defer(struct mm_struct *mm);
> void arch_tlbbatch_add_pending(struct arch_tlbflush_unmap_batch *batch,
> struct mm_struct *mm,
> - unsigned long uaddr);
> + unsigned long uaddr,
> + unsigned long size);
> void arch_flush_tlb_batched_pending(struct mm_struct *mm);
> void arch_tlbbatch_flush(struct arch_tlbflush_unmap_batch *batch);
>
> diff --git a/arch/riscv/mm/tlbflush.c b/arch/riscv/mm/tlbflush.c
> index 9b6e86ce3867..aeda64a36d50 100644
> --- a/arch/riscv/mm/tlbflush.c
> +++ b/arch/riscv/mm/tlbflush.c
> @@ -187,7 +187,8 @@ bool arch_tlbbatch_should_defer(struct mm_struct *mm)
>
> void arch_tlbbatch_add_pending(struct arch_tlbflush_unmap_batch *batch,
> struct mm_struct *mm,
> - unsigned long uaddr)
> + unsigned long uaddr,
> + unsigned long size)
> {
> cpumask_or(&batch->cpumask, &batch->cpumask, mm_cpumask(mm));
> }
>
> > All errors (new ones prefixed by >>):
> >
> > mm/rmap.c: In function 'set_tlb_ubc_flush_pending':
> > >> mm/rmap.c:685:9: error: too many arguments to function 'arch_tlbbatch_add_pending'
> > 685 | arch_tlbbatch_add_pending(&tlb_ubc->arch, mm, uaddr, size);
> > | ^~~~~~~~~~~~~~~~~~~~~~~~~
> > In file included from arch/riscv/include/asm/pgtable.h:113,
> > from include/linux/pgtable.h:6,
> > from include/linux/mm.h:30,
> > from mm/rmap.c:55:
> > arch/riscv/include/asm/tlbflush.h:62:6: note: declared here
> > 62 | void arch_tlbbatch_add_pending(struct arch_tlbflush_unmap_batch *batch,
> > | ^~~~~~~~~~~~~~~~~~~~~~~~~
> >
> >
> > vim +/arch_tlbbatch_add_pending +685 mm/rmap.c
> >
> > 663
> > 664 /*
> > 665 * Bits 0-14 of mm->tlb_flush_batched record pending generations.
> > 666 * Bits 16-30 of mm->tlb_flush_batched bit record flushed generations.
> > 667 */
> > 668 #define TLB_FLUSH_BATCH_FLUSHED_SHIFT 16
> > 669 #define TLB_FLUSH_BATCH_PENDING_MASK \
> > 670 ((1 << (TLB_FLUSH_BATCH_FLUSHED_SHIFT - 1)) - 1)
> > 671 #define TLB_FLUSH_BATCH_PENDING_LARGE \
> > 672 (TLB_FLUSH_BATCH_PENDING_MASK / 2)
> > 673
> > 674 static void set_tlb_ubc_flush_pending(struct mm_struct *mm, pte_t pteval,
> > 675 unsigned long uaddr,
> > 676 unsigned long size)
> > 677 {
> > 678 struct tlbflush_unmap_batch *tlb_ubc = ¤t->tlb_ubc;
> > 679 int batch;
> > 680 bool writable = pte_dirty(pteval);
> > 681
> > 682 if (!pte_accessible(mm, pteval))
> > 683 return;
> > 684
> > > 685 arch_tlbbatch_add_pending(&tlb_ubc->arch, mm, uaddr, size);
> > 686 tlb_ubc->flush_required = true;
> > 687
> > 688 /*
> > 689 * Ensure compiler does not re-order the setting of tlb_flush_batched
> > 690 * before the PTE is cleared.
> > 691 */
> > 692 barrier();
> > 693 batch = atomic_read(&mm->tlb_flush_batched);
> > 694 retry:
> > 695 if ((batch & TLB_FLUSH_BATCH_PENDING_MASK) > TLB_FLUSH_BATCH_PENDING_LARGE) {
> > 696 /*
> > 697 * Prevent `pending' from catching up with `flushed' because of
> > 698 * overflow. Reset `pending' and `flushed' to be 1 and 0 if
> > 699 * `pending' becomes large.
> > 700 */
> > 701 if (!atomic_try_cmpxchg(&mm->tlb_flush_batched, &batch, 1))
> > 702 goto retry;
> > 703 } else {
> > 704 atomic_inc(&mm->tlb_flush_batched);
> > 705 }
> > 706
> > 707 /*
> > 708 * If the PTE was dirty then it's best to assume it's writable. The
> > 709 * caller must use try_to_unmap_flush_dirty() or try_to_unmap_flush()
> > 710 * before the page is queued for IO.
> > 711 */
> > 712 if (writable)
> > 713 tlb_ubc->writable = true;
> > 714 }
> > 715
> >
> > --
> > 0-DAY CI Kernel Test Service
> > https://github.com/intel/lkp-tests/wiki
>
Powered by blists - more mailing lists