Re: [PATCH 04/12] mips: Reserve memory for the kernel image resources
From: Geert Uytterhoeven
Date: Wed May 22 2019 - 09:47:42 EST
Hi Serge,
On Wed, May 22, 2019 at 3:34 PM Serge Semin <fancer.lancer@xxxxxxxxx> wrote:
> On Wed, May 22, 2019 at 10:14:47AM +0200, Geert Uytterhoeven wrote:
> > On Wed, May 22, 2019 at 10:08 AM Mike Rapoport <rppt@xxxxxxxxxxxxx> wrote:
> > > On Wed, May 22, 2019 at 09:47:04AM +0200, Geert Uytterhoeven wrote:
> > > > On Tue, May 21, 2019 at 5:53 PM Mike Rapoport <rppt@xxxxxxxxxxxxx> wrote:
> > > > > On Tue, May 21, 2019 at 04:56:39PM +0200, Geert Uytterhoeven wrote:
> > > > > > On Wed, Apr 24, 2019 at 12:50 AM Serge Semin <fancer.lancer@xxxxxxxxx> wrote:
> > > > > > > The reserved_end variable had been used by the bootmem_init() code
> > > > > > > to find a lowest limit of memory available for memmap blob. The original
> > > > > > > code just tried to find a free memory space higher than kernel was placed.
> > > > > > > This limitation seems justified for the memmap ragion search process, but
> > > > > > > I can't see any obvious reason to reserve the unused space below kernel
> > > > > > > seeing some platforms place it much higher than standard 1MB. Moreover
> > > > > > > the RELOCATION config enables it to be loaded at any memory address.
> > > > > > > So lets reserve the memory occupied by the kernel only, leaving the region
> > > > > > > below being free for allocations. After doing this we can now discard the
> > > > > > > code freeing a space between kernel _text and VMLINUX_LOAD_ADDRESS symbols
> > > > > > > since it's going to be free anyway (unless marked as reserved by
> > > > > > > platforms).
> > > > > > >
> > > > > > > Signed-off-by: Serge Semin <fancer.lancer@xxxxxxxxx>
> > > > > >
> > > > > > This is now commit b93ddc4f9156205e ("mips: Reserve memory for the kernel
> > > > > > image resources") in v5.2-rc1, which causes rbtx4927 to crash during boot:
> > > > > >
> > > > > > VFS: Mounted root (nfs filesystem) on device 0:13.
> > > > > > devtmpfs: mounted
> > > > > > BUG: Bad page state in process swapper pfn:00001
> > > > > > page:804b7820 refcount:0 mapcount:-128 mapping:00000000 index:0x1
> > > > > > flags: 0x0()
> > > > > > raw: 00000000 00000100 00000200 00000000 00000001 00000000 ffffff7f 00000000
> > > > > > page dumped because: nonzero mapcount
> > > > > > Modules linked in:
> > > > > > CPU: 0 PID: 1 Comm: swapper Not tainted
> > > > > > 5.2.0-rc1-rbtx4927-00468-g3c05ea3d4077b756-dirty #137
> > > > > > Stack : 00000000 10008400 8040dd2c 87c1b974 8044af63 8040dd2c
> > > > > > 00000001 804a3490
> > > > > > 00000001 81000000 0030f231 80148558 00000003 10008400
> > > > > > 87c1dd80 7599ee13
> > > > > > 00000000 00000000 804b0000 00000000 00000007 00000000
> > > > > > 00000085 00000000
> > > > > > 62722d31 00000084 804b0000 39347874 00000000 804b7820
> > > > > > 8040cef8 81000010
> > > > > > 00000001 00000007 00000001 81000000 00000008 8021de24
> > > > > > 00000000 804a0000
> > > > > > ...
> > > > > > Call Trace:
> > > > > > [<8010adec>] show_stack+0x74/0x104
> > > > > > [<801a5e44>] bad_page+0x130/0x138
> > > > > > [<801a654c>] free_pcppages_bulk+0x17c/0x3b0
> > > > > > [<801a789c>] free_unref_page+0x40/0x68
> > > > > > [<801120f4>] free_init_pages+0xec/0x104
> > > > > > [<803bdde8>] free_initmem+0x10/0x58
> > > > > > [<803bdb8c>] kernel_init+0x20/0x100
> > > > > > [<801057c8>] ret_from_kernel_thread+0x14/0x1c
> > > > > > Disabling lock debugging due to kernel taint
> > > > > > BUG: Bad page state in process swapper pfn:00002
> > > > > > [...]
> > > > > >
> > > > > > CONFIG_RELOCATABLE is not set, so the only relevant part is the
> > > > > > change quoted below.
> > > > > >
> > > > > > > --- a/arch/mips/kernel/setup.c
> > > > > > > +++ b/arch/mips/kernel/setup.c
> > > > > > > @@ -371,7 +371,6 @@ static void __init bootmem_init(void)
> > > > > > >
> > > > > > > static void __init bootmem_init(void)
> > > > > > > {
> > > > > > > - unsigned long reserved_end;
> > > > > > > phys_addr_t ramstart = PHYS_ADDR_MAX;
> > > > > > > int i;
> > > > > > >
> > > > > > > @@ -382,10 +381,10 @@ static void __init bootmem_init(void)
> > > > > > > * will reserve the area used for the initrd.
> > > > > > > */
> > > > > > > init_initrd();
> > > > > > > - reserved_end = (unsigned long) PFN_UP(__pa_symbol(&_end));
> > > > > > >
> > > > > > > - memblock_reserve(PHYS_OFFSET,
> > > > > > > - (reserved_end << PAGE_SHIFT) - PHYS_OFFSET);
> > > > > > > + /* Reserve memory occupied by kernel. */
> > > > > > > + memblock_reserve(__pa_symbol(&_text),
> > > > > > > + __pa_symbol(&_end) - __pa_symbol(&_text));
> > > > > > >
> > > > > > > /*
> > > > > > > * max_low_pfn is not a number of pages. The number of pages
> >
> > > > > > Hence the old code reserved 1 MiB extra at the beginning.
> > > > > >
> > > > > > Note that the new code also dropped the rounding up of the memory block
> > > > > > size to a multiple of PAGE_SIZE. I'm not sure the latter actually
> > > > > > matters or not.
> > > > >
> > > > > I'd say that bad page state for pfn 1 is caused by "freeing" the first 1M.
> > > >
> > > > Yes, by prom_free_prom_memory(), as pointed out by Serge.
> > >
> > > I wonder how other MIPS variants would react to the fact that the memory
> > > below the kernel is not reserved ;-)
> >
> > Exactly...
> >
> > Looks like at least arch/mips/dec/prom/memory.c needs a similar but\
> > more complicated fix, due to declance handling...
> >
>
> The problem might be fixed there by the next patch:
> ---
> diff --git a/arch/mips/dec/prom/memory.c b/arch/mips/dec/prom/memory.c
> index 5073d2ed78bb..5a0c734b5d04 100644
> --- a/arch/mips/dec/prom/memory.c
> +++ b/arch/mips/dec/prom/memory.c
> @@ -91,29 +91,14 @@ void __init prom_meminit(u32 magic)
> pmax_setup_memory_region();
> else
> rex_setup_memory_region();
> -}
> -
> -void __init prom_free_prom_memory(void)
> -{
> - unsigned long end;
> -
> - /*
> - * Free everything below the kernel itself but leave
> - * the first page reserved for the exception handlers.
> - */
>
> #if IS_ENABLED(CONFIG_DECLANCE)
> /*
> - * Leave 128 KB reserved for Lance memory for
> - * IOASIC DECstations.
> + * Reserve 128 KB for Lance memory for IOASIC DECstations.
> *
> * XXX: save this address for use in dec_lance.c?
> */
> if (IOASIC)
> - end = __pa(&_text) - 0x00020000;
> - else
> + memblock_reserve(__pa_symbol(&_text), 0x00020000);
Shouldn't that be
memblock_reserve(__pa_symbol(&_text) - 0x00020000, 0x00020000);
?
> #endif
> - end = __pa(&_text);
> -
> - free_init_pages("unused PROM memory", PAGE_SIZE, end);
> }
> ---
>
> Didn't wanna use prom_FREE_prom_memory to actually reserve a memory
> chunk, so I moved the reservation into the prom_meminit() method.
I guess Maciej will test this on real hardware, eventually...
> Regarding the first page for the exception handlers. We don't need
> to reserve it here, since it is already done in arch/mips/kernel/traps.c .
Thanks for checking! That was actually something I was still wondering
about...
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds