linux-kernel - Re: [PATCH 07/25] mm/csky: Use mm_fault

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

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 18 Jun 2020 10:15:50 -0700
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     Peter Xu <peterx@...hat.com>
Cc:     Guo Ren <guoren@...nel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Gerald Schaefer <gerald.schaefer@...ibm.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Andrea Arcangeli <aarcange@...hat.com>,
        linux-csky@...r.kernel.org
Subject: Re: [PATCH 07/25] mm/csky: Use mm_fault_accounting()

On Thu, Jun 18, 2020 at 7:38 AM Peter Xu <peterx@...hat.com> wrote:
>
> GUP needs the per-task accounting, but not the perf events.  We can do that by
> slightly changing the new approach into:
>
>         bool major = (ret & VM_FAULT_MAJOR) || (flags & FAULT_FLAG_TRIED);
>
>         if (major)
>                 current->maj_flt++;
>         else
>                 current->min_flt++;
>
>         if (!regs)
>                 return ret;
>
>         if (major)
>                 perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1, regs, address);
>         else
>                 perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1, regs, address);

Ack, I think this is the right thing to do.

No normal situation will ever notice the difference, with remote
accesses being as rare and specialized as they are. But being able to
remote the otherwise unused 'tsk' parameter sounds like the right
thing to do too.

It might be worth adding a comment about why.

Also, honestly, how about we remove the 'major' variable entirely, and
instead make the code be something like

        unsigned long *flt;
        int event_type;
        ...

        /* Major fault */
        if ((ret & VM_FAULT_MAJOR) || (flags & FAULT_FLAG_TRIED)) {
                flt = &current->maj_flt;
                event_type = PERF_COUNT_SW_PAGE_FAULTS_MAJ;
        } else {
                flt = &current->min_flt;
                event_type = PERF_COUNT_SW_PAGE_FAULTS_MIN;
        }
        *flt++;
        if (regs)
                perf_sw_event(event_type, 1, regs, address);

instead. Less source code duplication, and I bet it improves code
generation too.

             Linus