linux/include
Hugh Dickins e0da382c92 [PATCH] freepgt: free_pgtables use vma list
Recent woes with some arches needing their own pgd_addr_end macro; and 4-level
clear_page_range regression since 2.6.10's clear_page_tables; and its
long-standing well-known inefficiency in searching throughout the higher-level
page tables for those few entries to clear and free: all can be blamed on
ignoring the list of vmas when we free page tables.

Replace exit_mmap's clear_page_range of the total user address space by
free_pgtables operating on the mm's vma list; unmap_region use it in the same
way, giving floor and ceiling beyond which it may not free tables.  This
brings lmbench fork/exec/sh numbers back to 2.6.10 (unless preempt is enabled,
in which case latency fixes spoil unmap_vmas throughput).

Beware: the do_mmap_pgoff driver failure case must now use unmap_region
instead of zap_page_range, since a page table might have been allocated, and
can only be freed while it is touched by some vma.

Move free_pgtables from mmap.c to memory.c, where its lower levels are adapted
from the clear_page_range levels.  (Most of free_pgtables' old code was
actually for a non-existent case, prev not properly set up, dating from before
hch gave us split_vma.) Pass mmu_gather** in the public interfaces, since we
might want to add latency lockdrops later; but no attempt to do so yet, going
by vma should itself reduce latency.

But what if is_hugepage_only_range?  Those ia64 and ppc64 cases need careful
examination: put that off until a later patch of the series.

What of x86_64's 32bit vdso page __map_syscall32 maps outside any vma?

And the range to sparc64's flush_tlb_pgtables?  It's less clear to me now that
we need to do more than is done here - every PMD_SIZE ever occupied will be
flushed, do we really have to flush every PGDIR_SIZE ever partially occupied? 
A shame to complicate it unnecessarily.

Special thanks to David Miller for time spent repairing my ceilings.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-19 13:29:15 -07:00
..
acpi Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
asm-alpha Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
asm-arm [PATCH] ARM: fix debug macros 2005-04-17 16:28:31 +01:00
asm-arm26 Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
asm-cris Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
asm-frv Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
asm-generic [PATCH] add Big Endian variants of ioread/iowrite 2005-04-16 15:25:54 -07:00
asm-h8300 [PATCH] h8300 header update 2005-04-16 15:25:21 -07:00
asm-i386 [PATCH] x86_64: add support for Intel dual-core detection and displaying 2005-04-16 15:25:15 -07:00
asm-ia64 Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
asm-m32r Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
asm-m68k [PATCH] ppc32: fix compilation error in include/asm-m68k/setup.h 2005-04-16 15:24:27 -07:00
asm-m68knommu Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
asm-mips [PATCH] consolidate timeout defintions in scsi.h 2005-04-18 12:31:52 -05:00
asm-parisc Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
asm-ppc [PATCH] u32 vs. pm_message_t in ppc and radeon 2005-04-16 15:25:34 -07:00
asm-ppc64 [PATCH] ppc64: no prefetch for NULL pointers 2005-04-16 15:24:38 -07:00
asm-s390 Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
asm-sh [PATCH] fix u32 vs. pm_message_t in rest of the tree 2005-04-16 15:25:37 -07:00
asm-sh64 Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
asm-sparc [PATCH] sparc64: Fix stat 2005-04-18 15:13:15 -07:00
asm-sparc64 [PATCH] sparc64: Fix stat 2005-04-18 15:13:15 -07:00
asm-um Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
asm-v850 Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
asm-x86_64 [PATCH] x86_64: Switch SMP bootup over to new CPU hotplug state machine 2005-04-16 15:25:19 -07:00
linux [PATCH] freepgt: free_pgtables use vma list 2005-04-19 13:29:15 -07:00
math-emu Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
media Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
mtd Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
net [PATCH] x86_64: Make IRDA devices are not really ISA devices not depend on CONFIG_ISA 2005-04-16 15:24:56 -07:00
pcmcia Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
rxrpc Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
scsi merge by hand (scsi_device.h) 2005-04-18 13:45:00 -05:00
sound Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
video Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00