2.5.71-bk2-wli-1

2.5.71-bk2-wli-1

[William Lee Irwin III]

Available from:
ftp://ftp.kernel.org/pub/linux/kernel/people/wli/kernels/2.5.71/linux-2.5.71-bk2-wli-1.bz2

applies incrementally atop the bk2 snapshot.

So, where does one get a fanbase, anyway? NFI. Here's at least an
attempt to acquire one, with desktop-oriented goodies to speed up
top(1), apparently to the point where top(1) barely even registers
on its own cpu usage reports even with 1s refreshes (900MHz P-III).
It might also vaguely help people instrument heavy multitasking loads.

For what it's worth, the O(1) proc_pid_statm() patch also restores some
of the old 2.4.x fields whose support was removed from 2.5.x.


Merged:
- flow.c compilefix
	Well, I didn't really do any of the work, but at least I don't
	have to carry it around anymore.


Changes since 2.5.71-wli-1:

+ O(1) proc_pid_readdir()
	originally due to Manfred Spraul; figures out its position from
	a small pid hashtable rearrangement

+ O(1) proc_pid_statm()
	originally due to Ben LaHaise (I think); keeps count of the various
	proc_pid_statm() counters whenever twiddling ptes.

+ non-i386 highpmd fixes
	write stubs for non-i386 pmd twiddling functions

+ highpmd fixes
	a few mm/memory.c functions forgot to pass pmd pointers by reference

+ O(1) buffered_rmqueue() fixes
	correct small typo in CONFIG_SOFTWARE_SUSPEND code touched by it

+ i386 pagetable cache fixes
	Correct failure to reset tlb->freed, which resulted in wildly wrong
	mm->rss counts. Also define FREE_PTE_NR so CONFIG_PREEMPT compiles.


All 12 patches:

O(1) rmqueue_bulk()
	Implement deferred coalescing with list-of-lists -structured order 0
	deferred queues so buffered_rmqueue() in O(1) expected time.

lowmem_page_address() microoptimization
	use page_to_pfn() to inherit its arch-specific microoptimizations

highpmd
	shove pmd's into highmem, by brute force

trivial /proc/ BKL removals
	Kill off some blatantly unnecessary BKL-grabbing in /proc/

i386 pagetable cache
	Resurrect the i386 pagetable cache

pgd_ctor
	Use slab ctors for i386 pgd's, and be safe with AGP and highpmd

O(1) proc_pid_readdir()
	originally due to Manfred Spraul; figures out its position from
	a small pid hashtable rearrangement

O(1) proc_pid_statm()
	originally due to Ben LaHaise (I think); keeps count of the various
	proc_pid_statm() counters whenever twiddling ptes.

non-i386 highpmd fixes
	write stubs for non-i386 pmd twiddling functions

highpmd fixes
	a few mm/memory.c functions forgot to pass pmd pointers by reference

O(1) buffered_rmqueue() fixes
	correct small typo in CONFIG_SOFTWARE_SUSPEND code touched by it

i386 pagetable cache fixes
	Correct failure to reset tlb->freed, which resulted in wildly wrong
	mm->rss counts. Also define FREE_PTE_NR so CONFIG_PREEMPT compiles.


-- wli

Re: 2.5.71-bk2-wli-1

[William Lee Irwin III]

On Mon, Jun 16, 2003 at 05:58:07PM -0700, William Lee Irwin III wrote:
> + highpmd fixes
> 	a few mm/memory.c functions forgot to pass pmd pointers by reference

Woops, I missed one:


diff -prauN wli-2.5.71-bk2-8/mm/memory.c wli-2.5.71-bk2-9/mm/memory.c
--- wli-2.5.71-bk2-8/mm/memory.c	2003-06-16 14:59:11.000000000 -0700
+++ wli-2.5.71-bk2-9/mm/memory.c	2003-06-16 17:45:05.000000000 -0700
@@ -1002,7 +1002,7 @@ int remap_page_range(struct vm_area_stru
 		error = remap_pmd_range(vma, &pmd, from, end - from, phys_addr + from, prot);
 		if (error)
 			break;
-		pmd_unmap(pmd);
+		pmd_unmap(pmd - 1);
 		from = (from + PGDIR_SIZE) & PGDIR_MASK;
 		dir++;
 	} while (from && (from < end));