2022-05-08 19:06:40

by Baoquan He

[permalink] [raw]
Subject: Re: [PATCH v23 3/6] arm64: kdump: Reimplement crashkernel=X

On 05/06/22 at 11:22am, Leizhen (ThunderTown) wrote:
......
> >> @@ -118,8 +159,7 @@ static void __init reserve_crashkernel(void)
> >> if (crash_base)
> >> crash_max = crash_base + crash_size;
> >>
> >> - /* Current arm64 boot protocol requires 2MB alignment */
> >> - crash_base = memblock_phys_alloc_range(crash_size, SZ_2M,
> >> + crash_base = memblock_phys_alloc_range(crash_size, CRASH_ALIGN,
> >> crash_base, crash_max);
> >> if (!crash_base) {
> >> pr_warn("cannot allocate crashkernel (size:0x%llx)\n",
> >
> > I personally like this but let's see how the other thread goes. I guess
>
> Me too. This fallback complicates code logic more than just a little.
> I'm not sure why someone would rather add fallback than change the bootup
> options to crashkernel=X,[high|low]. Perhaps fallback to high/low is a better
> compatible and extended mode when crashkernel=X fails to reserve memory. And
> the code logic will be much clearer.

The fallback does complicates code, while it was not made at the
beginning, but added later. The original crahskernel=xM can only reserve
low memory under 896M on x86 to be back compatible with the case in which
normal kernel is x86_64, while kdump kernel could be i386. Then customer
complained why crashkernel=xM can't be put anywhere so that they don't
need to know the details of limited low memory and huge high memory fact
in system.

The implementation of fallback is truly complicated, but its use is
quite simple. And it makes crashkernel reservation setting simple.
Most of users don't need to know crashkernel=,high, ,low things, unless
the crashkernel region is too big. Nobody wants to take away 1G or more
from low memory for kdump just in case bad thing happens, while normal
kernel itself is seriously impacted by limited low memory.

>
> //parse crashkernel=X //To simplify the discussion, Ignore [@offset]
> crash_base = memblock_phys_alloc_range()
> if (!crash_base || /* crashkernel=X is not specified */) {
> //parse crashkernel=X,[high,low]
> //reserve high/low memory
> }
>
> So that, the following three modes are supported:
> 1) crashkernel=X[@offset]
> 2) crashkernel=X,high crashkernel=X,low
> 3) crashkernel=X[@offset] crashkernel=X,high [crashkernel=Y,low]
>
> For case 3), try "crashkernel=X[@offset]" first, if it can not work, fallback
> to "crashkernel=X,high crashkernel=X,low". This looks better than the old "crashkernel=X"
> fallback ---- Select a region under 4G first, and fall back to reserve region above 4G.

Don't get it. Aren't they the same?

>
> Note: when the X of crashkernel=X and crashkernel=X,high are the same, It's equivalent
> to the old "crashkernel=X" fallback.
>
> > if we want a fallback, it would come just before the check the above:
> >
> > if (!crash_base && crash_max != CRASH_ADDR_HIGH_MAX) {
> > /* attempt high allocation with default low */
> > if (!crash_low_size)
> > crash_low_size = some default;
> > crash_max = CRASH_ADDR_LOW_MAX;
>
> crash_max = CRASH_ADDR_HIGH_MAX; We should fallback to high memory now.
>
> > crash_base = memblock_phys_alloc_range();
> > }
> >
> > Well, I guess we end up with your earlier proposal but I think I
> > understand it better now ;).
> >
>
> --
> Regards,
> Zhen Lei
>



2022-05-09 04:41:44

by Catalin Marinas

[permalink] [raw]
Subject: Re: [PATCH v23 3/6] arm64: kdump: Reimplement crashkernel=X

On Fri, May 06, 2022 at 09:16:08PM +0800, Baoquan He wrote:
> On 05/06/22 at 11:22am, Leizhen (ThunderTown) wrote:
> ......
> > >> @@ -118,8 +159,7 @@ static void __init reserve_crashkernel(void)
> > >> if (crash_base)
> > >> crash_max = crash_base + crash_size;
> > >>
> > >> - /* Current arm64 boot protocol requires 2MB alignment */
> > >> - crash_base = memblock_phys_alloc_range(crash_size, SZ_2M,
> > >> + crash_base = memblock_phys_alloc_range(crash_size, CRASH_ALIGN,
> > >> crash_base, crash_max);
> > >> if (!crash_base) {
> > >> pr_warn("cannot allocate crashkernel (size:0x%llx)\n",
> > >
> > > I personally like this but let's see how the other thread goes. I guess
> >
> > Me too. This fallback complicates code logic more than just a little.
> > I'm not sure why someone would rather add fallback than change the bootup
> > options to crashkernel=X,[high|low]. Perhaps fallback to high/low is a better
> > compatible and extended mode when crashkernel=X fails to reserve memory. And
> > the code logic will be much clearer.
>
> The fallback does complicates code, while it was not made at the
> beginning, but added later. The original crahskernel=xM can only reserve
> low memory under 896M on x86 to be back compatible with the case in which
> normal kernel is x86_64, while kdump kernel could be i386. Then customer
> complained why crashkernel=xM can't be put anywhere so that they don't
> need to know the details of limited low memory and huge high memory fact
> in system.
>
> The implementation of fallback is truly complicated, but its use is
> quite simple. And it makes crashkernel reservation setting simple.
> Most of users don't need to know crashkernel=,high, ,low things, unless
> the crashkernel region is too big. Nobody wants to take away 1G or more
> from low memory for kdump just in case bad thing happens, while normal
> kernel itself is seriously impacted by limited low memory.

IIUC, that's exactly what happens even on x86, it may take away a
significant chunk of the low memory. Let's say we have 1.2GB of 'low'
memory (below 4GB) on an arm64 platform. A crashkernel=1G would succeed
in a low allocation, pretty much affecting the whole system. It would
only fall back to 'high' _if_ you pass something like crashkernel=1.2G
so that the low allocation fails. So if I got this right, I find the
fall-back from crashkernel=X pretty useless, we shouldn't even try it.

It makes more sense if crashkernel=X,high is a hint to attempt a high
allocation first with a default low (overridden by a ,low option) or
even fall-back to low if there's no memory above 4GB.

Could you please have a look at Zhen Lei's latest series without any
fall-backs? I'd like to queue that if you are happy with it. We can then
look at adding some fall-back options on top.

IMO, we should only aim for:

crashkernel=X ZONE_DMA allocation, no fall-back
crashkernel=X,high hint for high allocation, small default
low, fall back to low if alloc fails
crashkernel=X,low control the default low allocation, only
high is passed

With the above, I'd expect admins to just go for crashkernel=X,high on
modern hardware with up to date kexec tools and it does the right thing.
The crashkernel=X can lead to unexpected results if it eats up all the
low memory. Let's say this option is for backwards compatibility only.

Thanks.

--
Catalin