Re: [PATCH] mm: page_alloc: ignore init_on_free=1 for page alloc

2021-03-30 Thread Andrey Konovalov
On Tue, Mar 30, 2021 at 12:07 AM Vlastimil Babka wrote: > > On 3/30/21 12:00 AM, Andrey Konovalov wrote: > > On Mon, Mar 29, 2021 at 2:10 PM Vlastimil Babka wrote: > >> > >> > commit 855a9c4018f3219db8be7e4b9a65ab22aebfde82 > >> > Author: Andrey Konovalov > >> > Date: Thu Mar 18 17:01:40 2021

Re: [PATCH] mm: page_alloc: ignore init_on_free=1 for page alloc

2021-03-29 Thread Vlastimil Babka
On 3/30/21 12:00 AM, Andrey Konovalov wrote: > On Mon, Mar 29, 2021 at 2:10 PM Vlastimil Babka wrote: >> >> > commit 855a9c4018f3219db8be7e4b9a65ab22aebfde82 >> > Author: Andrey Konovalov >> > Date: Thu Mar 18 17:01:40 2021 +1100 >> > >> > kasan, mm: integrate page_alloc init with HW_TAGS

Re: [PATCH] mm: page_alloc: ignore init_on_free=1 for page alloc

2021-03-29 Thread Andrey Konovalov
On Mon, Mar 29, 2021 at 2:10 PM Vlastimil Babka wrote: > > > commit 855a9c4018f3219db8be7e4b9a65ab22aebfde82 > > Author: Andrey Konovalov > > Date: Thu Mar 18 17:01:40 2021 +1100 > > > > kasan, mm: integrate page_alloc init with HW_TAGS > > But the mmotm patch/-next commit also changes

Re: [PATCH] mm: page_alloc: ignore init_on_free=1 for page alloc

2021-03-29 Thread Vlastimil Babka
On 3/26/21 2:48 PM, David Hildenbrand wrote: > On 26.03.21 12:26, Sergei Trofimovich wrote: >> init_on_free=1 does not guarantee that free pages contain only zero bytes. >> >> Some examples: >> 1. page_poison=on takes presedence over init_on_alloc=1 / ini_on_free=1 > > s/ini_on_free/init_on_free/

Re: [PATCH] mm: page_alloc: ignore init_on_free=1 for page alloc

2021-03-27 Thread Sergei Trofimovich
On Fri, 26 Mar 2021 17:25:22 + Sergei Trofimovich wrote: > On Fri, 26 Mar 2021 15:17:00 +0100 > Vlastimil Babka wrote: > > > On 3/26/21 12:26 PM, Sergei Trofimovich wrote: > > > init_on_free=1 does not guarantee that free pages contain only zero bytes. > > > > > > Some examples: > > >

Re: [PATCH] mm: page_alloc: ignore init_on_free=1 for page alloc

2021-03-26 Thread Sergei Trofimovich
On Fri, 26 Mar 2021 15:17:00 +0100 Vlastimil Babka wrote: > On 3/26/21 12:26 PM, Sergei Trofimovich wrote: > > init_on_free=1 does not guarantee that free pages contain only zero bytes. > > > > Some examples: > > 1. page_poison=on takes presedence over init_on_alloc=1 / ini_on_free=1 > >

Re: [PATCH] mm: page_alloc: ignore init_on_free=1 for page alloc

2021-03-26 Thread Sergei Trofimovich
On Fri, 26 Mar 2021 16:00:34 +0100 Andrey Konovalov wrote: > On Fri, Mar 26, 2021 at 2:49 PM David Hildenbrand wrote: > > > > > I observed use of poisoned pages as the crash on ia64 booted with > > > init_on_free=1 init_on_alloc=1 (CONFIG_PAGE_POISONING=y config). > > > There pmd page

Re: [PATCH] mm: page_alloc: ignore init_on_free=1 for page alloc

2021-03-26 Thread Andrey Konovalov
On Fri, Mar 26, 2021 at 2:49 PM David Hildenbrand wrote: > > > I observed use of poisoned pages as the crash on ia64 booted with > > init_on_free=1 init_on_alloc=1 (CONFIG_PAGE_POISONING=y config). > > There pmd page contained 0x poison pages and led to early crash. > > > > The change

Re: [PATCH] mm: page_alloc: ignore init_on_free=1 for page alloc

2021-03-26 Thread Vlastimil Babka
On 3/26/21 12:26 PM, Sergei Trofimovich wrote: > init_on_free=1 does not guarantee that free pages contain only zero bytes. > > Some examples: > 1. page_poison=on takes presedence over init_on_alloc=1 / ini_on_free=1 Yes, and it spits out a message that you enabled both and poisoning takes

Re: [PATCH] mm: page_alloc: ignore init_on_free=1 for page alloc

2021-03-26 Thread David Hildenbrand
On 26.03.21 12:26, Sergei Trofimovich wrote: init_on_free=1 does not guarantee that free pages contain only zero bytes. Some examples: 1. page_poison=on takes presedence over init_on_alloc=1 / ini_on_free=1 s/ini_on_free/init_on_free/ 2. free_pages_prepare() always poisons pages:

[PATCH] mm: page_alloc: ignore init_on_free=1 for page alloc

2021-03-26 Thread Sergei Trofimovich
init_on_free=1 does not guarantee that free pages contain only zero bytes. Some examples: 1. page_poison=on takes presedence over init_on_alloc=1 / ini_on_free=1 2. free_pages_prepare() always poisons pages: if (want_init_on_free()) kernel_init_free_pages(page, 1 << order);