Alexander Motin 15183f36e5 amd64: Stop using REP MOVSB for backward memmove()s. ... Enhanced REP MOVSB feature of CPUs starting from Ivy Bridge makes REP MOVSB the fastest way to copy memory in most of cases. However Intel Optimization Reference Manual says: "setting the DF to force REP MOVSB to copy bytes from high towards low addresses will expe- rience significant performance degradation". Measurements on Intel Cascade Lake and Alder Lake, same as on AMD Zen3 show that it can drop throughput to as low as 2.5-3.5GB/s, comparing to ~10-30GB/s of REP MOVSQ or hand-rolled loop, used for non-ERMS CPUs. This patch keeps ERMS use for forward ordered memory copies, but removes it for backward overlapped moves where it does not work. Reviewed by: mjg MFC after: 2 weeks (cherry picked from commit 6210ac95a1)		2022-06-29 21:15:49 -04:00
..
acpica	amd64 wakeup: rework trampoline page allocation	2021-09-20 03:47:21 +03:00
amd64	amd64: Stop using REP MOVSB for backward memmove()s.	2022-06-29 21:15:49 -04:00
cloudabi32	fork: Allow ABI to specify fork return values for child.	2022-06-17 22:33:28 +03:00
cloudabi64	fork: Allow ABI to specify fork return values for child.	2022-06-17 22:33:28 +03:00
conf	linux(4): Get rid of Linuxulator kernel build options.	2022-06-17 22:30:22 +03:00
ia32	x86: Add a NT_X86_SEGBASES register set.	2022-05-13 09:45:19 -07:00
include	vmm: add tunable to trap WBINVD	2022-06-20 17:05:36 +02:00
linux	linux(4): Properly build argument list for the signal handler	2022-06-17 22:35:38 +03:00
linux32	linux(4): Properly restore the thread signal mask after signal delivery on i386	2022-06-17 22:35:40 +03:00
pci	pci_cfgreg.c: Use io port config access for early boot time.	2019-04-09 18:07:17 +00:00
sgx	Convert remaining cap_rights_init users to cap_rights_init_one	2021-01-12 13:16:10 +00:00
vmm	vmm: move bumping VMEXIT_USERSPACE stat to the right place	2022-06-24 10:40:48 -04:00
Makefile	Bring the tags and links entries for amd64 up to date.	2021-10-07 16:01:46 -07:00