Skip to content

RFC THP support for PPC64

Aneesh Kumar K.V linuxppc-dev 1 version New

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 RFC Latest17 patches 0 repliesNew

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/17powerpc: Don't hard code the size of pte pagearch/powerpc/include/asm/pgtable.h arch/powerpc/mm/hash_low_64.S+8 -20 replies
2/17arch/powerpc: Reduce the PTE_INDEX_SIZEarch/powerpc/include/asm/pgtable-ppc64-64k.h+2 -20 replies
3/17powerpc: Reduce PTE table memory wastagearch/powerpc/include/asm/mmu-book3e.h arch/powerpc/include/asm/mmu-hash64.h arch/powerpc/include/asm/page.h +6 more+377 -740 replies
4/17mm/THP: Add pmd args to pgtable deposit and withdraw APIsarch/s390/include/asm/pgtable.h arch/s390/mm/pgtable.c arch/sparc/include/asm/pgtable_64.h +4 more+24 -180 replies
5/17powerpc: Add size argument to pgtable_cache_addarch/powerpc/include/asm/pgtable-ppc64.h arch/powerpc/mm/init_64.c+14 -90 replies
6/17powerpc/mm: Decode the pte-lp-encoding bits correctly.arch/powerpc/include/asm/machdep.h arch/powerpc/include/asm/mmu-hash64.h arch/powerpc/mm/hash_low_64.S +3 more+163 -820 replies
7/17powerpc: Update tlbie/tlbiel as per ISA docarch/powerpc/mm/hash_native_64.c+40 -120 replies
8/17powerpc: print both base and actual page size on hash failurearch/powerpc/include/asm/mmu-hash64.h arch/powerpc/mm/hash_utils_64.c arch/powerpc/mm/hugetlbpage-hash64.c+10 -70 replies
9/17powerpc/mm: Use encode avpn where we need only avpn valuesarch/powerpc/include/asm/mmu-hash64.h arch/powerpc/mm/hash_native_64.c arch/powerpc/platforms/pseries/lpar.c+10 -100 replies
10/17powerpc/mm: Fix hpte_decode to use the correct decoding for page sizesarch/powerpc/mm/hash_native_64.c+16 -100 replies
11/17powerpc: Print page size info during bootarch/powerpc/mm/hash_utils_64.c+5 -50 replies
12/17powerpc/THP: Implement transparent huge pages for ppc64arch/powerpc/include/asm/page.h arch/powerpc/include/asm/pgtable-ppc64-64k.h arch/powerpc/include/asm/pgtable-ppc64.h +4 more+610 -30 replies
13/17powerpc/THP: Add code to handle HPTE faults for large pagesarch/powerpc/include/asm/mmu-hash64.h arch/powerpc/include/asm/pgtable-ppc64.h arch/powerpc/kernel/io-workarounds.c +10 more+248 -410 replies
14/17powerpc: support for zerout withdraw.arch/powerpc/include/asm/pgtable.h arch/powerpc/mm/pgtable.c mm/huge_memory.c+22 -100 replies
15/17powerpc: hypervisor require few WIMG bit setarch/powerpc/mm/largepage-hash64.c arch/powerpc/platforms/pseries/lpar.c+3 -10 replies
16/17powerpc: get_user_pages_fast changesarch/powerpc/mm/gup.c+74 -20 replies
17/17powerpc: Save DAR and DSISR in pt_regs on MCEarch/powerpc/kernel/exceptions-64s.S arch/powerpc/xmon/xmon.c+10 -10 replies

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.

Read the v1 thread

Cover letter [RFC PATCH 00/17] THP support for PPC64

Hi,

This is an early RFC version adding transparent huge page support for PPC64. I am sharing the changes, so that we can have early review on the approach taken. The TODOs include

*) Compile issues with different config option
*) HugeTLBfs is disabled now (mostly compile issues)
*) PPC32 and other sub architecture details need to be worked out. *) 4K page size details need to be worked out
*) Closer review of PMD* flags.

Some numbers:

The latency measurements code from Anton found at
http://ozlabs.org/~anton/junkcode/latency2001.c

THP disabled 64K page size
------------------------
[root@llmp24l02 ~]# ./latency2001 8G
8589934592 731.73 cycles 205.77 ns
[root@llmp24l02 ~]# ./latency2001 8G
8589934592 743.39 cycles 209.05 ns
[root@llmp24l02 ~]#

THP disabled large page via hugetlbfs
-------------------------------------
[root@llmp24l02 ~]# ./latency2001 -l 8G
8589934592 416.09 cycles 117.01 ns
[root@llmp24l02 ~]# ./latency2001 -l 8G
8589934592 415.74 cycles 116.91 ns

THP enabled 64K page size.
----------------
[root@llmp24l02 ~]# ./latency2001 8G
8589934592 405.07 cycles 113.91 ns
[root@llmp24l02 ~]# ./latency2001 8G
8589934592 411.82 cycles 115.81 ns
[root@llmp24l02 ~]#

We are close to hugetlbfs in latency and we can achieve this with zero config/page reservation. Most of the allocations above are fault allocated. I haven't really measured the collapse alloc impact.

Another test that does 50000000 random access over 1GB area goes from 2.65 seconds to 1.07 seconds with this patchset.

Thanks,
-aneesh