| Message ID | 1526818839-8497-1-git-send-email-chenzhou10@huawei.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On 20/05/18 13:20, Chen Zhou wrote: > From: Mao Wenan <maowenan@huawei.com> > > On arm64 kdump capture kernel, ACPI runtime code and data are outside > of usable memory range and __alloc_bootmem_low may alloc memory below > 4G. If we use "crashkernel=Y[@X]" and the start address is above 4G, > there will be out of memory. > > In this patch, we get max_zone_dma_phys by memblock_start_of_DRAM > and __pa_symbol(_end). Entering the crash kernel doesn't magically reconfigure interconnects such that 32-bit DMA masters can suddenly address whatever physical address range the kernel image happens to be loaded at. This change is fundamentally incorrect. Robin. > Signed-off-by: Chen Zhou <chenzhou10@huawei.com> > --- > arch/arm64/mm/init.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c > index 1b18b47..df4d212 100644 > --- a/arch/arm64/mm/init.c > +++ b/arch/arm64/mm/init.c > @@ -223,7 +223,8 @@ static void __init reserve_elfcorehdr(void) > */ > static phys_addr_t __init max_zone_dma_phys(void) > { > - phys_addr_t offset = memblock_start_of_DRAM() & GENMASK_ULL(63, 32); > + phys_addr_t offset = max(memblock_start_of_DRAM(), __pa_symbol(_end)) & > + GENMASK_ULL(63, 32); > return min(offset + (1ULL << 32), memblock_end_of_DRAM()); > } > >
diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index 1b18b47..df4d212 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -223,7 +223,8 @@ static void __init reserve_elfcorehdr(void) */ static phys_addr_t __init max_zone_dma_phys(void) { - phys_addr_t offset = memblock_start_of_DRAM() & GENMASK_ULL(63, 32); + phys_addr_t offset = max(memblock_start_of_DRAM(), __pa_symbol(_end)) & + GENMASK_ULL(63, 32); return min(offset + (1ULL << 32), memblock_end_of_DRAM()); }