From: Hillf Danton <hdanton@sina.com> To: Marco Elver <elver@google.com> Cc: Hillf Danton <hdanton@sina.com>, Andrew Morton <akpm@linux-foundation.org>, Alexander Potapenko <glider@google.com>, Andrey Konovalov <andreyknvl@google.com>, Linux ARM <linux-arm-kernel@lists.infradead.org>, Linux Memory Management List <linux-mm@kvack.org> Subject: Re: [PATCH v2 01/10] mm: add Kernel Electric-Fence infrastructure Date: Fri, 18 Sep 2020 07:45:39 +0800 [thread overview] Message-ID: <20200917234539.13276-1-hdanton@sina.com> (raw) In-Reply-To: <CANpmjNPPs=7gZRiXER9umH0cROMwJ6FEPsqeBMo7RMfmxhvCOw@mail.gmail.com> On Thu, 17 Sep 2020 15:36:38 +0200 Marco Elver wrote: > On Thu, 17 Sep 2020 at 15:26, Hillf Danton wrote: > > Tue, 15 Sep 2020 15:20:37 +0200 > > > From: Alexander Potapenko <glider@google.com> > > > > > > This adds the Kernel Electric-Fence (KFENCE) infrastructure. KFENCE is a > > > low-overhead sampling-based memory safety error detector of heap > > > use-after-free, invalid-free, and out-of-bounds access errors. > > > > > > KFENCE is designed to be enabled in production kernels, and has near > > > zero performance overhead. Compared to KASAN, KFENCE trades performance > > > for precision. The main motivation behind KFENCE's design, is that with > > > enough total uptime KFENCE will detect bugs in code paths not typically > > > exercised by non-production test workloads. One way to quickly achieve a > > > large enough total uptime is when the tool is deployed across a large > > > fleet of machines. > > > > > > KFENCE objects each reside on a dedicated page, at either the left or > > > right page boundaries. The pages to the left and right of the object > > > page are "guard pages", whose attributes are changed to a protected > > > state, and cause page faults on any attempted access to them. Such page > > > faults are then intercepted by KFENCE, which handles the fault > > > gracefully by reporting a memory access error. > > > > To help understand the magic of KFENCE, a simple diagram looks needed to > > illustrate the relations between obj and guard pages, something like the > > below asiic chart. > > > > |-----------------|-----------------------------------|------------------| > > | left guard page | the page containing KFENCE object | right guard page | > > |-----------------|-----------------------------------|------------------| > > > > Would the one we have in Documentation be what you're after? Yes, that's great. > > https://lore.kernel.org/linux-arm-kernel/20200915132046.3332537-10-elver@google.com/ > (at "The following figure illustrates the page layout::") > > Let us know if you'd like that copied into the commit message. Then the copy can help more readers. Thanks Hillf
WARNING: multiple messages have this Message-ID (diff)
From: Hillf Danton <hdanton@sina.com> To: Marco Elver <elver@google.com> Cc: Hillf Danton <hdanton@sina.com>, Andrey Konovalov <andreyknvl@google.com>, Linux Memory Management List <linux-mm@kvack.org>, Alexander Potapenko <glider@google.com>, Andrew Morton <akpm@linux-foundation.org>, Linux ARM <linux-arm-kernel@lists.infradead.org> Subject: Re: [PATCH v2 01/10] mm: add Kernel Electric-Fence infrastructure Date: Fri, 18 Sep 2020 07:45:39 +0800 [thread overview] Message-ID: <20200917234539.13276-1-hdanton@sina.com> (raw) In-Reply-To: <CANpmjNPPs=7gZRiXER9umH0cROMwJ6FEPsqeBMo7RMfmxhvCOw@mail.gmail.com> On Thu, 17 Sep 2020 15:36:38 +0200 Marco Elver wrote: > On Thu, 17 Sep 2020 at 15:26, Hillf Danton wrote: > > Tue, 15 Sep 2020 15:20:37 +0200 > > > From: Alexander Potapenko <glider@google.com> > > > > > > This adds the Kernel Electric-Fence (KFENCE) infrastructure. KFENCE is a > > > low-overhead sampling-based memory safety error detector of heap > > > use-after-free, invalid-free, and out-of-bounds access errors. > > > > > > KFENCE is designed to be enabled in production kernels, and has near > > > zero performance overhead. Compared to KASAN, KFENCE trades performance > > > for precision. The main motivation behind KFENCE's design, is that with > > > enough total uptime KFENCE will detect bugs in code paths not typically > > > exercised by non-production test workloads. One way to quickly achieve a > > > large enough total uptime is when the tool is deployed across a large > > > fleet of machines. > > > > > > KFENCE objects each reside on a dedicated page, at either the left or > > > right page boundaries. The pages to the left and right of the object > > > page are "guard pages", whose attributes are changed to a protected > > > state, and cause page faults on any attempted access to them. Such page > > > faults are then intercepted by KFENCE, which handles the fault > > > gracefully by reporting a memory access error. > > > > To help understand the magic of KFENCE, a simple diagram looks needed to > > illustrate the relations between obj and guard pages, something like the > > below asiic chart. > > > > |-----------------|-----------------------------------|------------------| > > | left guard page | the page containing KFENCE object | right guard page | > > |-----------------|-----------------------------------|------------------| > > > > Would the one we have in Documentation be what you're after? Yes, that's great. > > https://lore.kernel.org/linux-arm-kernel/20200915132046.3332537-10-elver@google.com/ > (at "The following figure illustrates the page layout::") > > Let us know if you'd like that copied into the commit message. Then the copy can help more readers. Thanks Hillf _______________________________________________ linux-arm-kernel mailing list linux-arm-kernel@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
next prev parent reply other threads:[~2020-09-17 23:45 UTC|newest] Thread overview: 61+ messages / expand[flat|nested] mbox.gz Atom feed top 2020-09-15 13:20 [PATCH v2 00/10] KFENCE: A low-overhead sampling-based memory safety error detector Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` [PATCH v2 01/10] mm: add Kernel Electric-Fence infrastructure Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-17 13:26 ` Hillf Danton 2020-09-17 13:26 ` Hillf Danton 2020-09-17 13:36 ` Marco Elver 2020-09-17 13:36 ` Marco Elver 2020-09-17 23:45 ` Hillf Danton [this message] 2020-09-17 23:45 ` Hillf Danton 2020-09-15 13:20 ` [PATCH v2 02/10] x86, kfence: enable KFENCE for x86 Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` [PATCH v2 03/10] arm64, kfence: enable KFENCE for ARM64 Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` [PATCH v2 04/10] mm, kfence: insert KFENCE hooks for SLAB Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-17 9:37 ` Christopher Lameter 2020-09-17 9:37 ` Christopher Lameter 2020-09-17 9:37 ` Christopher Lameter 2020-09-17 9:47 ` Alexander Potapenko 2020-09-17 9:47 ` Alexander Potapenko 2020-09-17 9:47 ` Alexander Potapenko 2020-09-15 13:20 ` [PATCH v2 05/10] mm, kfence: insert KFENCE hooks for SLUB Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-17 9:40 ` Christopher Lameter 2020-09-17 9:40 ` Christopher Lameter 2020-09-17 9:40 ` Christopher Lameter 2020-09-17 9:51 ` Alexander Potapenko 2020-09-17 9:51 ` Alexander Potapenko 2020-09-17 9:51 ` Alexander Potapenko 2020-09-15 13:20 ` [PATCH v2 06/10] kfence, kasan: make KFENCE compatible with KASAN Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` [PATCH v2 07/10] kfence, kmemleak: make KFENCE compatible with KMEMLEAK Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` [PATCH v2 08/10] kfence, lockdep: make KFENCE compatible with lockdep Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` [PATCH v2 09/10] kfence, Documentation: add KFENCE documentation Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` [PATCH v2 10/10] kfence: add test suite Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-15 13:20 ` Marco Elver 2020-09-16 5:13 ` kernel test robot 2020-09-15 13:49 ` [PATCH v2 00/10] KFENCE: A low-overhead sampling-based memory safety error detector Dmitry Vyukov 2020-09-15 13:49 ` Dmitry Vyukov 2020-09-15 13:49 ` Dmitry Vyukov 2020-09-18 11:17 ` Qian Cai 2020-09-18 11:17 ` Qian Cai 2020-09-18 11:17 ` Qian Cai 2020-09-18 11:59 ` Marco Elver 2020-09-18 11:59 ` Marco Elver 2020-09-18 11:59 ` Marco Elver
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20200917234539.13276-1-hdanton@sina.com \ --to=hdanton@sina.com \ --cc=akpm@linux-foundation.org \ --cc=andreyknvl@google.com \ --cc=elver@google.com \ --cc=glider@google.com \ --cc=linux-arm-kernel@lists.infradead.org \ --cc=linux-mm@kvack.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.