From: Michal Hocko <mhocko@kernel.org> To: John Hubbard <jhubbard@nvidia.com> Cc: Andrew Morton <akpm@linux-foundation.org>, Vlastimil Babka <vbabka@suse.cz>, David Rientjes <rientjes@google.com>, Mel Gorman <mgorman@suse.de>, Johannes Weiner <hannes@cmpxchg.org>, Al Viro <viro@zeniv.linux.org.uk>, linux-mm@kvack.org, LKML <linux-kernel@vger.kernel.org>, Anatoly Stepanov <astepanov@cloudlinux.com>, Paolo Bonzini <pbonzini@redhat.com>, Mike Snitzer <snitzer@redhat.com>, "Michael S. Tsirkin" <mst@redhat.com>, "Theodore Ts'o" <tytso@mit.edu> Subject: Re: [PATCH 1/6] mm: introduce kv[mz]alloc helpers Date: Thu, 19 Jan 2017 09:45:11 +0100 [thread overview] Message-ID: <20170119084510.GF30786@dhcp22.suse.cz> (raw) In-Reply-To: <37232cc6-af8b-52e2-3265-9ef0c0d26e5f@nvidia.com> On Thu 19-01-17 00:37:08, John Hubbard wrote: > > > On 01/18/2017 12:21 AM, Michal Hocko wrote: > > On Tue 17-01-17 21:59:13, John Hubbard wrote: [...] > > > * Reclaim modifiers - __GFP_NORETRY and __GFP_NOFAIL should not be passed in. > > > * Passing in __GFP_REPEAT is supported, but note that it is ignored for small > > > * (<=64KB) allocations, during the kmalloc attempt. > > > > > __GFP_REPEAT is fully > > > * honored for all allocation sizes during the second part: the vmalloc attempt. > > > > this is not true to be really precise because vmalloc doesn't respect > > the given gfp mask all the way down (look at the pte initialization). > > > > I'm having some difficulty in locating that pte initialization part, am I on > the wrong code path? Here's what I checked, before making the claim about > __GFP_REPEAT being honored: > > kvmalloc_node > __vmalloc_node_flags > __vmalloc_node > __vmalloc_node_range > __vmalloc_area_node map_vm_area vmap_page_range vmap_page_range_noflush vmap_pud_range pud_alloc __pud_alloc pud_alloc_one pud will be allocated but the same pattern repeats on the pmd and pte levels. This is btw. one of the reasons why vmalloc with gfp flags is tricky! moreover > alloc_pages_node this is order-0 request so... > __alloc_pages_node > __alloc_pages > __alloc_pages_nodemask > __alloc_pages_slowpath > > > ...and __alloc_pages_slowpath does the __GFP_REPEAT handling: > > /* > * Do not retry costly high order allocations unless they are > * __GFP_REPEAT > */ > if (order > PAGE_ALLOC_COSTLY_ORDER && !(gfp_mask & __GFP_REPEAT)) > goto nopage; ... this doesn't apply -- Michal Hocko SUSE Labs
WARNING: multiple messages have this Message-ID (diff)
From: Michal Hocko <mhocko@kernel.org> To: John Hubbard <jhubbard@nvidia.com> Cc: Andrew Morton <akpm@linux-foundation.org>, Vlastimil Babka <vbabka@suse.cz>, David Rientjes <rientjes@google.com>, Mel Gorman <mgorman@suse.de>, Johannes Weiner <hannes@cmpxchg.org>, Al Viro <viro@zeniv.linux.org.uk>, linux-mm@kvack.org, LKML <linux-kernel@vger.kernel.org>, Anatoly Stepanov <astepanov@cloudlinux.com>, Paolo Bonzini <pbonzini@redhat.com>, Mike Snitzer <snitzer@redhat.com>, "Michael S. Tsirkin" <mst@redhat.com>, Theodore Ts'o <tytso@mit.edu> Subject: Re: [PATCH 1/6] mm: introduce kv[mz]alloc helpers Date: Thu, 19 Jan 2017 09:45:11 +0100 [thread overview] Message-ID: <20170119084510.GF30786@dhcp22.suse.cz> (raw) In-Reply-To: <37232cc6-af8b-52e2-3265-9ef0c0d26e5f@nvidia.com> On Thu 19-01-17 00:37:08, John Hubbard wrote: > > > On 01/18/2017 12:21 AM, Michal Hocko wrote: > > On Tue 17-01-17 21:59:13, John Hubbard wrote: [...] > > > * Reclaim modifiers - __GFP_NORETRY and __GFP_NOFAIL should not be passed in. > > > * Passing in __GFP_REPEAT is supported, but note that it is ignored for small > > > * (<=64KB) allocations, during the kmalloc attempt. > > > > > __GFP_REPEAT is fully > > > * honored for all allocation sizes during the second part: the vmalloc attempt. > > > > this is not true to be really precise because vmalloc doesn't respect > > the given gfp mask all the way down (look at the pte initialization). > > > > I'm having some difficulty in locating that pte initialization part, am I on > the wrong code path? Here's what I checked, before making the claim about > __GFP_REPEAT being honored: > > kvmalloc_node > __vmalloc_node_flags > __vmalloc_node > __vmalloc_node_range > __vmalloc_area_node map_vm_area vmap_page_range vmap_page_range_noflush vmap_pud_range pud_alloc __pud_alloc pud_alloc_one pud will be allocated but the same pattern repeats on the pmd and pte levels. This is btw. one of the reasons why vmalloc with gfp flags is tricky! moreover > alloc_pages_node this is order-0 request so... > __alloc_pages_node > __alloc_pages > __alloc_pages_nodemask > __alloc_pages_slowpath > > > ...and __alloc_pages_slowpath does the __GFP_REPEAT handling: > > /* > * Do not retry costly high order allocations unless they are > * __GFP_REPEAT > */ > if (order > PAGE_ALLOC_COSTLY_ORDER && !(gfp_mask & __GFP_REPEAT)) > goto nopage; ... this doesn't apply -- Michal Hocko SUSE Labs -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2017-01-19 9:04 UTC|newest] Thread overview: 129+ messages / expand[flat|nested] mbox.gz Atom feed top 2017-01-12 15:37 [PATCH 0/6 v3] kvmalloc Michal Hocko 2017-01-12 15:37 ` Michal Hocko 2017-01-12 15:37 ` [PATCH 1/6] mm: introduce kv[mz]alloc helpers Michal Hocko 2017-01-12 15:37 ` Michal Hocko 2017-01-16 4:34 ` John Hubbard 2017-01-16 4:34 ` John Hubbard 2017-01-16 8:47 ` Michal Hocko 2017-01-16 8:47 ` Michal Hocko 2017-01-16 19:09 ` John Hubbard 2017-01-16 19:09 ` John Hubbard 2017-01-16 19:40 ` Michal Hocko 2017-01-16 19:40 ` Michal Hocko 2017-01-16 21:15 ` John Hubbard 2017-01-16 21:15 ` John Hubbard 2017-01-16 21:48 ` Michal Hocko 2017-01-16 21:48 ` Michal Hocko 2017-01-16 21:57 ` John Hubbard 2017-01-16 21:57 ` John Hubbard 2017-01-17 7:51 ` Michal Hocko 2017-01-17 7:51 ` Michal Hocko 2017-01-18 5:59 ` John Hubbard 2017-01-18 5:59 ` John Hubbard 2017-01-18 8:21 ` Michal Hocko 2017-01-18 8:21 ` Michal Hocko 2017-01-19 8:37 ` John Hubbard 2017-01-19 8:37 ` John Hubbard 2017-01-19 8:45 ` Michal Hocko [this message] 2017-01-19 8:45 ` Michal Hocko 2017-01-19 9:09 ` John Hubbard 2017-01-19 9:09 ` John Hubbard 2017-01-19 9:56 ` Michal Hocko 2017-01-19 9:56 ` Michal Hocko 2017-01-19 21:28 ` John Hubbard 2017-01-19 21:28 ` John Hubbard 2017-01-26 12:09 ` Michal Hocko 2017-01-26 12:09 ` Michal Hocko 2017-01-30 8:42 ` Vlastimil Babka 2017-01-30 8:42 ` Vlastimil Babka 2017-01-12 15:37 ` [PATCH 2/6] mm: support __GFP_REPEAT in kvmalloc_node for >=64kB Michal Hocko 2017-01-12 15:37 ` Michal Hocko 2017-01-12 16:12 ` Michael S. Tsirkin 2017-01-12 16:12 ` Michael S. Tsirkin 2017-01-14 2:42 ` Tetsuo Handa 2017-01-14 2:42 ` Tetsuo Handa 2017-01-14 8:45 ` Michal Hocko 2017-01-14 8:45 ` Michal Hocko 2017-01-24 15:40 ` Michael S. Tsirkin 2017-01-24 15:40 ` Michael S. Tsirkin 2017-01-12 15:37 ` [PATCH 3/6] rhashtable: simplify a strange allocation pattern Michal Hocko 2017-01-12 15:37 ` Michal Hocko 2017-01-12 15:37 ` [PATCH 4/6] ila: " Michal Hocko 2017-01-12 15:37 ` Michal Hocko 2017-01-12 15:37 ` [PATCH 5/6] treewide: use kv[mz]alloc* rather than opencoded variants Michal Hocko 2017-01-12 15:37 ` Michal Hocko 2017-01-12 15:37 ` Michal Hocko 2017-01-12 15:57 ` David Sterba 2017-01-12 15:57 ` David Sterba 2017-01-12 15:57 ` David Sterba 2017-01-12 16:05 ` Christian Borntraeger 2017-01-12 16:05 ` Christian Borntraeger 2017-01-12 16:05 ` Christian Borntraeger 2017-01-12 16:54 ` Ilya Dryomov 2017-01-12 16:54 ` Ilya Dryomov 2017-01-12 16:54 ` Ilya Dryomov 2017-01-12 17:18 ` Michal Hocko 2017-01-12 17:18 ` Michal Hocko 2017-01-12 17:18 ` Michal Hocko 2017-01-12 17:00 ` Dan Williams 2017-01-12 17:00 ` Dan Williams 2017-01-12 17:00 ` Dan Williams 2017-01-12 17:26 ` Kees Cook 2017-01-12 17:26 ` Kees Cook 2017-01-12 17:26 ` Kees Cook 2017-01-12 17:37 ` Michal Hocko 2017-01-12 17:37 ` Michal Hocko 2017-01-12 17:37 ` Michal Hocko 2017-01-20 13:41 ` Vlastimil Babka 2017-01-20 13:41 ` Vlastimil Babka 2017-01-20 13:41 ` Vlastimil Babka 2017-01-24 15:00 ` Michal Hocko 2017-01-24 15:00 ` Michal Hocko 2017-01-24 15:00 ` Michal Hocko 2017-01-25 11:15 ` Vlastimil Babka 2017-01-25 11:15 ` Vlastimil Babka 2017-01-25 11:15 ` Vlastimil Babka 2017-01-25 13:09 ` Michal Hocko 2017-01-25 13:09 ` Michal Hocko 2017-01-25 13:09 ` Michal Hocko 2017-01-25 13:40 ` Ilya Dryomov 2017-01-25 13:40 ` Ilya Dryomov 2017-01-25 13:40 ` Ilya Dryomov 2017-01-12 17:29 ` Michal Hocko 2017-01-12 17:29 ` Michal Hocko 2017-01-12 17:29 ` Michal Hocko 2017-01-14 3:01 ` Tetsuo Handa 2017-01-14 3:01 ` Tetsuo Handa 2017-01-14 8:49 ` Michal Hocko 2017-01-14 8:49 ` Michal Hocko 2017-01-12 20:14 ` Boris Ostrovsky 2017-01-12 20:14 ` Boris Ostrovsky 2017-01-12 20:14 ` Boris Ostrovsky 2017-01-13 1:11 ` Dilger, Andreas 2017-01-13 1:11 ` Dilger, Andreas 2017-01-13 1:11 ` Dilger, Andreas 2017-01-14 10:56 ` Leon Romanovsky 2017-01-14 10:56 ` Leon Romanovsky 2017-01-16 7:33 ` Michal Hocko 2017-01-16 7:33 ` Michal Hocko 2017-01-16 7:33 ` Michal Hocko 2017-01-16 8:28 ` Leon Romanovsky 2017-01-16 8:28 ` Leon Romanovsky 2017-01-16 8:18 ` Tariq Toukan 2017-01-16 8:18 ` Tariq Toukan 2017-01-16 8:18 ` Tariq Toukan 2017-01-12 15:37 ` [RFC PATCH 6/6] net: use kvmalloc with __GFP_REPEAT rather than open coded variant Michal Hocko 2017-01-12 15:37 ` Michal Hocko 2017-01-12 15:37 ` Michal Hocko 2017-01-24 15:17 ` [PATCH 0/6 v3] kvmalloc Michal Hocko 2017-01-24 15:17 ` Michal Hocko 2017-01-24 16:00 ` Eric Dumazet 2017-01-24 16:00 ` Eric Dumazet 2017-01-25 13:10 ` Michal Hocko 2017-01-25 13:10 ` Michal Hocko 2017-01-24 19:17 ` Alexei Starovoitov 2017-01-24 19:17 ` Alexei Starovoitov 2017-01-25 13:10 ` Michal Hocko 2017-01-25 13:10 ` Michal Hocko 2017-01-25 13:21 ` Michal Hocko 2017-01-25 13:21 ` Michal Hocko
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=20170119084510.GF30786@dhcp22.suse.cz \ --to=mhocko@kernel.org \ --cc=akpm@linux-foundation.org \ --cc=astepanov@cloudlinux.com \ --cc=hannes@cmpxchg.org \ --cc=jhubbard@nvidia.com \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=mgorman@suse.de \ --cc=mst@redhat.com \ --cc=pbonzini@redhat.com \ --cc=rientjes@google.com \ --cc=snitzer@redhat.com \ --cc=tytso@mit.edu \ --cc=vbabka@suse.cz \ --cc=viro@zeniv.linux.org.uk \ /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.