Re: [mm 4.15-rc8] Random oopses under memory pressure.

[Al Viro <viro@ZenIV.linux.org.uk>]

On Sat, Jan 20, 2018 at 02:02:37AM +0000, Al Viro wrote:

> Note that those sizes are rather sensitive to lockdep, spinlock debugging, etc.

That they certainly are: on one of the testing .config I'm using it gave this:
   1104 sizeof struct page = 56
     81 sizeof struct cpufreq_frequency_table = 12
     32 sizeof struct Indirect = 24
      7 sizeof struct zone = 1400
      7 sizeof struct hstate = 152
      6 sizeof struct lock_class = 336
      6 sizeof struct hpet_dev = 152
      6 sizeof struct ext4_extent = 12
      4 sizeof struct ext4_extent_idx = 12
      3 sizeof struct mbox_chan = 456
      2 sizeof struct strip_zone = 24
      2 sizeof struct kobj_attribute = 48
      2 sizeof struct kernel_param = 40
      2 sizeof struct exception_table_entry = 12
      1 sizeof struct vif_device = 104
      1 sizeof struct unixware_slice = 12
      1 sizeof struct svc_pool = 152
      1 sizeof struct srcu_node = 152
      1 sizeof struct r5worker_group = 56
      1 sizeof struct pebs_record_core = 144
      1 sizeof struct netdev_queue = 384
      1 sizeof struct mirror = 40
      1 sizeof struct mif_device = 56
      1 sizeof struct e1000_tx_ring = 48
      1 sizeof struct dx_frame = 24
      1 sizeof struct bts_record = 24
      1 sizeof struct ata_device = 2560
      1 sizeof struct acpi_processor_cx = 52

--
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>