linux_dsm_epyc7002/arch/powerpc/platforms
Alexey Kardashevskiy a68bd1267b powerpc/powernv/ioda: Allocate indirect TCE levels on demand
At the moment we allocate the entire TCE table, twice (hardware part and
userspace translation cache). This normally works as we normally have
contigous memory and the guest will map entire RAM for 64bit DMA.

However if we have sparse RAM (one example is a memory device), then
we will allocate TCEs which will never be used as the guest only maps
actual memory for DMA. If it is a single level TCE table, there is nothing
we can really do but if it a multilevel table, we can skip allocating
TCEs we know we won't need.

This adds ability to allocate only first level, saving memory.

This changes iommu_table::free() to avoid allocating of an extra level;
iommu_table::set() will do this when needed.

This adds @alloc parameter to iommu_table::exchange() to tell the callback
if it can allocate an extra level; the flag is set to "false" for
the realmode KVM handlers of H_PUT_TCE hcalls and the callback returns
H_TOO_HARD.

This still requires the entire table to be counted in mm::locked_vm.

To be conservative, this only does on-demand allocation when
the usespace cache table is requested which is the case of VFIO.

The example math for a system replicating a powernv setup with NVLink2
in a guest:
16GB RAM mapped at 0x0
128GB GPU RAM window (16GB of actual RAM) mapped at 0x244000000000

the table to cover that all with 64K pages takes:
(((0x244000000000 + 0x2000000000) >> 16)*8)>>20 = 4556MB

If we allocate only necessary TCE levels, we will only need:
(((0x400000000 + 0x400000000) >> 16)*8)>>20 = 4MB (plus some for indirect
levels).

Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-07-16 22:53:11 +10:00
..
4xx treewide: kzalloc() -> kcalloc() 2018-06-12 16:19:22 -07:00
8xx powerpc/8xx: Remove RTC clock on 88x 2018-06-04 00:39:18 +10:00
40x License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
44x powerpc/44x/fsp2: Add irq error handlers 2017-12-11 13:03:34 +11:00
52xx powerpc/mpc52xx_gpt: make use of raw_spinlock variants 2018-01-27 21:12:04 +11:00
82xx License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
83xx powerpc: remove retired sbc834x support 2018-05-03 22:32:30 +10:00
85xx powerpc/64: Use array of paca pointers and allocate pacas individually 2018-03-30 23:34:23 +11:00
86xx powerpc: Use pr_warn instead of pr_warning 2017-12-04 11:54:34 +11:00
512x powerpc: Use pr_warn instead of pr_warning 2017-12-04 11:54:34 +11:00
amigaone License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
cell powerpc updates for 4.18 2018-06-07 10:23:33 -07:00
chrp powerpc/chrp/time: Make some functions static, add missing header include 2018-05-25 12:04:45 +10:00
embedded6xx powerpc/embedded6xx: Remove C2K board support 2018-06-04 00:39:22 +10:00
maple powerpc: use time64_t in read_persistent_clock 2018-06-03 20:43:33 +10:00
pasemi powerpc: use time64_t in read_persistent_clock 2018-06-03 20:43:33 +10:00
powermac powerpc: fix spelling mistake: "Usupported" -> "Unsupported" 2018-06-05 11:33:43 +10:00
powernv powerpc/powernv/ioda: Allocate indirect TCE levels on demand 2018-07-16 22:53:11 +10:00
ps3 powerpc: remove unused to_tm() helper 2018-06-03 20:43:34 +10:00
pseries powerpc updates for 4.18 2018-06-07 10:23:33 -07:00
fsl_uli1575.c of/irq: Refactor interrupt-map parsing 2013-10-24 11:43:04 +01:00
Kconfig powerpc/cell: Remove axonram driver 2018-01-28 17:00:36 +11:00
Kconfig.cputype powerpc updates for 4.18 2018-06-07 10:23:33 -07:00
Makefile License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00