powerpc/crashkernel: Take "mem=" option into account
authorPingfan Liu <kernelfans@gmail.com>
Wed, 1 Apr 2020 14:00:44 +0000 (22:00 +0800)
committerMichael Ellerman <mpe@ellerman.id.au>
Tue, 2 Jun 2020 10:59:05 +0000 (20:59 +1000)
commitbe5470e0c285a68dc3afdea965032f5ddc8269d7
tree7661dabaeee89965efff8c8563fde8bdff5493f9
parentef3534a94fdbdeab4c89d18d0164be2ad5d6dbb7
powerpc/crashkernel: Take "mem=" option into account

'mem=" option is an easy way to put high pressure on memory during
some test. Hence after applying the memory limit, instead of total
mem, the actual usable memory should be considered when reserving mem
for crashkernel. Otherwise the boot up may experience OOM issue.

E.g. it would reserve 4G prior to the change and 512M afterward, if
passing
crashkernel="2G-4G:384M,4G-16G:512M,16G-64G:1G,64G-128G:2G,128G-:4G",
and mem=5G on a 256G machine.

This issue is powerpc specific because it puts higher priority on
fadump and kdump reservation than on "mem=". Referring the following
code:
    if (fadump_reserve_mem() == 0)
            reserve_crashkernel();
    ...
    /* Ensure that total memory size is page-aligned. */
    limit = ALIGN(memory_limit ?: memblock_phys_mem_size(), PAGE_SIZE);
    memblock_enforce_memory_limit(limit);

While on other arches, the effect of "mem=" takes a higher priority
and pass through memblock_phys_mem_size() before calling
reserve_crashkernel().

Signed-off-by: Pingfan Liu <kernelfans@gmail.com>
Reviewed-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/1585749644-4148-1-git-send-email-kernelfans@gmail.com
arch/powerpc/kexec/core.c