mirror of
https://github.com/Evolution-X-Devices/kernel_google_b1c1
synced 2026-01-27 17:26:47 +00:00
e7a8a5f04c
This is a backport from Linux 5.2-rc1 of a patch series to greatly enhance vmalloc's performance especially on embedded systems, plus all of its dependencies that were missing in kernel 4.9. For all the informations, refer to LKML: https://lkml.org/lkml/2018/10/19/786 Brief informations: Currently an allocation of the new VA area is done over busy list iteration until a suitable hole is found between two busy areas. Therefore each new allocation causes the list being grown. Due to long list and different permissive parameters an allocation can take a long time on embedded devices(milliseconds). This patch organizes the vmalloc memory layout into free areas of the VMALLOC_START-VMALLOC_END range. It uses a red-black tree that keeps blocks sorted by their offsets in pair with linked list keeping the free space in order of increasing addresses. Quote Phoronix: With this patch from Uladzislau Rezki, calling vmalloc() can take up to 67% less time compared to the behavior on Linux 5.1 and prior, at least with tests done by the developer under QEMU. Personal tests are showing that the device is more responsive when memory pressure is high and when huge allocations are to be done, it's also noticeably faster in this case, like when starting Chrome with more than 100 opened tabs after a system reboot (so, an uncached complete load of it). Shameless kanged from: https://github.com/sonyxperiadev/kernel / Pull Request 2016