yuzu/unicorn
1
0
Fork 0
mirror of https://github.com/yuzu-emu/unicorn.git synced 2025-02-25 13:26:58 +00:00
Code Issues Packages Projects Releases Wiki Activity

tcg: Adjust CODE_GEN_AVG_BLOCK_SIZE

Browse source 
At present, the "average" guestimate of TB size is way too small, leading
to many unused entries in the pre-allocated TB array. For a guest with 1GB
ram, we're currently allocating 256MB for the array.

Survey arm, alpha, aarch64, ppc, sparc, i686, x86_64 guests running on
x86_64 and ppc64 hosts and select a new average. The size of the array
drops to 81MB with no more flushing than before.

Backports commit 126d89e8cdfa3be15d51f76906eaccbcd0023f98 from qemu
This commit is contained in:
Richard Henderson 2018-02-16 14:28:17 -05:00 committed by Lioncash
parent bdf667fd4e
commit a276496ebc
No known key found for this signature in database
GPG key ID: 4E3C3CC1031BA9C7
1 changed files with 6 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

11
qemu/include/exec/exec-all.h
View file

@ -160,13 +160,14 @@ static inline void tlb_flush_by_mmuidx(CPUState *cpu, ...)
#define CODE_GEN_PHYS_HASH_BITS 15 #define CODE_GEN_PHYS_HASH_BITS 15
#define CODE_GEN_PHYS_HASH_SIZE (1 << CODE_GEN_PHYS_HASH_BITS) #define CODE_GEN_PHYS_HASH_SIZE (1 << CODE_GEN_PHYS_HASH_BITS)
/* estimated block size for TB allocation */ /* Estimated block size for TB allocation. */
/* XXX: use a per code average code fragment size and modulate it /* ??? The following is based on a 2015 survey of x86_64 host output.
according to the host CPU */ Better would seem to be some sort of dynamically sized TB array,
adapting to the block sizes actually being produced. */
#if defined(CONFIG_SOFTMMU) #if defined(CONFIG_SOFTMMU)
#define CODE_GEN_AVG_BLOCK_SIZE 128 #define CODE_GEN_AVG_BLOCK_SIZE 400
#else #else
#define CODE_GEN_AVG_BLOCK_SIZE 64 #define CODE_GEN_AVG_BLOCK_SIZE 150
#endif #endif
#if defined(__arm__) || defined(_ARCH_PPC) \ #if defined(__arm__) || defined(_ARCH_PPC) \