Re: [PATCH v4 1/4] mm,page_alloc: Bail out earlier on -ENOMEM in alloc_contig_migrate_range

2021-03-15 Thread David Hildenbrand
On 10.03.21 16:08, Oscar Salvador wrote: Currently, __alloc_contig_migrate_range can generate -EINTR, -ENOMEM or -EBUSY, and report them down the chain. The problem is that when migrate_pages() reports -ENOMEM, we keep going till we exhaust all the try-attempts (5 at the moment) instead of

Re: [PATCH v4 1/4] mm,page_alloc: Bail out earlier on -ENOMEM in alloc_contig_migrate_range

2021-03-15 Thread Vlastimil Babka
On 3/10/21 4:08 PM, Oscar Salvador wrote: > Currently, __alloc_contig_migrate_range can generate -EINTR, -ENOMEM or > -EBUSY, > and report them down the chain. > The problem is that when migrate_pages() reports -ENOMEM, we keep going till > we > exhaust all the try-attempts (5 at the moment)

[PATCH v4 1/4] mm,page_alloc: Bail out earlier on -ENOMEM in alloc_contig_migrate_range

2021-03-10 Thread Oscar Salvador
Currently, __alloc_contig_migrate_range can generate -EINTR, -ENOMEM or -EBUSY, and report them down the chain. The problem is that when migrate_pages() reports -ENOMEM, we keep going till we exhaust all the try-attempts (5 at the moment) instead of bailing out. migrate_pages bails out right away