From: David Hildenbrand <david@redhat.com> To: "Liang, Liang (Leo)" <Liang.Liang@amd.com>, Mike Rapoport <rppt@linux.ibm.com> Cc: "Deucher, Alexander" <Alexander.Deucher@amd.com>, "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>, amd-gfx list <amd-gfx@lists.freedesktop.org>, Andrew Morton <akpm@linux-foundation.org>, "Huang, Ray" <Ray.Huang@amd.com>, "Koenig, Christian" <Christian.Koenig@amd.com>, "Rafael J. Wysocki" <rafael@kernel.org>, George Kennedy <george.kennedy@oracle.com> Subject: Re: slow boot with 7fef431be9c9 ("mm/page_alloc: place pages to tail in __free_pages_core()") Date: Tue, 16 Mar 2021 10:08:10 +0100 [thread overview] Message-ID: <9e869214-7a3b-e86d-4832-7117f7c6090f@redhat.com> (raw) In-Reply-To: <MW3PR12MB4537B6D5911092E456220818F36B9@MW3PR12MB4537.namprd12.prod.outlook.com> On 16.03.21 09:58, Liang, Liang (Leo) wrote: > [AMD Public Use] > > Hi David, > > root@scbu-Chachani:~# cat /proc/mtrr > reg00: base=0x000000000 ( 0MB), size= 2048MB, count=1: write-back > reg01: base=0x0ffe00000 ( 4094MB), size= 2MB, count=1: write-protect > reg02: base=0x100000000 ( 4096MB), size= 16MB, count=1: write-protect ^ there it is https://wiki.osdev.org/MTRR "Reads allocate cache lines on a cache miss. All writes update main memory. Cache lines are not allocated on a write miss. Write hits invalidate the cache line and update main memory. " AFAIU, writes completely bypass caches and store directly to main mamory. If there are cache lines from a previous read, they are invalidated. So I think especially slow will be read(addr), write(addr), read(addr), ... which is what we have in the kstream benchmark. The question is: who sets this up without owning the memory? Is the memory actually special/slow or is that setting wrong? Buggy firmware/BIOS? Buggy device driver? > reg03: base=0x0ffde0000 ( 4093MB), size= 128KB, count=1: write-protect > reg04: base=0x0ff000000 ( 4080MB), size= 512KB, count=1: write-protect -- Thanks, David / dhildenb
WARNING: multiple messages have this Message-ID (diff)
From: David Hildenbrand <david@redhat.com> To: "Liang, Liang (Leo)" <Liang.Liang@amd.com>, Mike Rapoport <rppt@linux.ibm.com> Cc: "Rafael J. Wysocki" <rafael@kernel.org>, "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>, amd-gfx list <amd-gfx@lists.freedesktop.org>, George Kennedy <george.kennedy@oracle.com>, "Huang, Ray" <Ray.Huang@amd.com>, "Deucher, Alexander" <Alexander.Deucher@amd.com>, Andrew Morton <akpm@linux-foundation.org>, "Koenig, Christian" <Christian.Koenig@amd.com> Subject: Re: slow boot with 7fef431be9c9 ("mm/page_alloc: place pages to tail in __free_pages_core()") Date: Tue, 16 Mar 2021 10:08:10 +0100 [thread overview] Message-ID: <9e869214-7a3b-e86d-4832-7117f7c6090f@redhat.com> (raw) In-Reply-To: <MW3PR12MB4537B6D5911092E456220818F36B9@MW3PR12MB4537.namprd12.prod.outlook.com> On 16.03.21 09:58, Liang, Liang (Leo) wrote: > [AMD Public Use] > > Hi David, > > root@scbu-Chachani:~# cat /proc/mtrr > reg00: base=0x000000000 ( 0MB), size= 2048MB, count=1: write-back > reg01: base=0x0ffe00000 ( 4094MB), size= 2MB, count=1: write-protect > reg02: base=0x100000000 ( 4096MB), size= 16MB, count=1: write-protect ^ there it is https://wiki.osdev.org/MTRR "Reads allocate cache lines on a cache miss. All writes update main memory. Cache lines are not allocated on a write miss. Write hits invalidate the cache line and update main memory. " AFAIU, writes completely bypass caches and store directly to main mamory. If there are cache lines from a previous read, they are invalidated. So I think especially slow will be read(addr), write(addr), read(addr), ... which is what we have in the kstream benchmark. The question is: who sets this up without owning the memory? Is the memory actually special/slow or is that setting wrong? Buggy firmware/BIOS? Buggy device driver? > reg03: base=0x0ffde0000 ( 4093MB), size= 128KB, count=1: write-protect > reg04: base=0x0ff000000 ( 4080MB), size= 512KB, count=1: write-protect -- Thanks, David / dhildenb _______________________________________________ amd-gfx mailing list amd-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/amd-gfx
next prev parent reply other threads:[~2021-03-16 9:10 UTC|newest] Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-03-11 14:41 slow boot with 7fef431be9c9 ("mm/page_alloc: place pages to tail in __free_pages_core()") Deucher, Alexander 2021-03-11 14:41 ` Deucher, Alexander 2021-03-11 15:02 ` David Hildenbrand 2021-03-11 15:02 ` David Hildenbrand 2021-03-12 14:06 ` Deucher, Alexander 2021-03-12 14:06 ` Deucher, Alexander 2021-03-12 14:12 ` David Hildenbrand 2021-03-12 14:12 ` David Hildenbrand 2021-03-12 15:39 ` Deucher, Alexander 2021-03-12 15:39 ` Deucher, Alexander 2021-03-12 15:48 ` David Hildenbrand 2021-03-12 15:48 ` David Hildenbrand 2021-03-12 16:15 ` Deucher, Alexander 2021-03-12 16:15 ` Deucher, Alexander 2021-03-12 16:19 ` Liang, Liang (Leo) 2021-03-12 16:19 ` Liang, Liang (Leo) 2021-03-12 16:46 ` David Hildenbrand 2021-03-12 16:46 ` David Hildenbrand 2021-03-13 4:04 ` Liang, Liang (Leo) 2021-03-13 4:04 ` Liang, Liang (Leo) 2021-03-13 9:05 ` David Hildenbrand 2021-03-13 9:05 ` David Hildenbrand 2021-03-13 13:48 ` Mike Rapoport 2021-03-13 13:48 ` Mike Rapoport 2021-03-15 13:04 ` David Hildenbrand 2021-03-15 13:04 ` David Hildenbrand 2021-03-16 0:36 ` Liang, Liang (Leo) 2021-03-16 0:36 ` Liang, Liang (Leo) 2021-03-16 6:50 ` Mike Rapoport 2021-03-16 6:50 ` Mike Rapoport 2021-03-16 8:00 ` Liang, Liang (Leo) 2021-03-16 8:00 ` Liang, Liang (Leo) 2021-03-16 8:37 ` David Hildenbrand 2021-03-16 8:37 ` David Hildenbrand 2021-03-16 8:43 ` Liang, Liang (Leo) 2021-03-16 8:43 ` Liang, Liang (Leo) 2021-03-16 8:54 ` David Hildenbrand 2021-03-16 8:54 ` David Hildenbrand 2021-03-16 8:58 ` Liang, Liang (Leo) 2021-03-16 8:58 ` Liang, Liang (Leo) 2021-03-16 9:08 ` David Hildenbrand [this message] 2021-03-16 9:08 ` David Hildenbrand 2021-03-16 10:30 ` Mike Rapoport 2021-03-16 10:30 ` Mike Rapoport 2021-03-16 11:02 ` Liang, Liang (Leo) 2021-03-16 11:02 ` Liang, Liang (Leo) 2021-03-16 11:10 ` David Hildenbrand 2021-03-16 11:10 ` David Hildenbrand
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=9e869214-7a3b-e86d-4832-7117f7c6090f@redhat.com \ --to=david@redhat.com \ --cc=Alexander.Deucher@amd.com \ --cc=Christian.Koenig@amd.com \ --cc=Liang.Liang@amd.com \ --cc=Ray.Huang@amd.com \ --cc=akpm@linux-foundation.org \ --cc=amd-gfx@lists.freedesktop.org \ --cc=george.kennedy@oracle.com \ --cc=linux-kernel@vger.kernel.org \ --cc=rafael@kernel.org \ --cc=rppt@linux.ibm.com \ /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.