Hi,
I picked up some trivial patches and fixed some stuff here and there...
The result is available at
<URL:ftp://luckynet.dynu.com/pub/linux/2.5.20-ct1/linux-2.5.20-ct1.patch.bz2>
The full ChangeLog is available at
<URL:ftp://luckynet.dynu.com/pub/linux/2.5.20-ct1/ChangeLog-2.5.20-ct1>
The single patches are available at
<URL:ftp://luckynet.dynu.com/pub/linux/2.5.20-ct1/single-patches/>
Here is the ChangeLog:
Summary of changes from v2.5.20 to v2.5.20-ct1
==============================================
<[email protected]>:
o Trivial: must be __KERNEL__ for byteorder/generic.h
<[email protected]>:
o Scalable CPU bitmasks
o Scalable phys_cpu_present_map
<[email protected]>:
o remove mixture of non-atomic operations with page->flags which requires atomic operations to access
o repetitive reinitialization of active_list and inactive_list in free_area_init_core()
o make balance_classzone() use list.h
o complete comment regarding inner workings of buddy system
o duplicate declaration of rq in sched_init()
o Re: forget_pte()
o remove antiquated comment from page_alloc.c
o convert page_alloc.c bugchecks to BUG_ON()
o remove MARK_USED() macros
o remove memlist_* macros from page_alloc.c
o correct inaccurate comment regarding zone_table's usage
<[email protected]>:
o bluesmoke merge
Adam J. Richter <[email protected]>:
o Trivial patch: linux-2.5.20/Rules.make cleanup
Andreas Dilger <[email protected]>:
o Re: another -pre
Andrew Morton <[email protected]>:
o PCI device matching fix
Anton Blanchard <[email protected]>:
o Fix for recent swap changes on 64 bit archs
Arnaldo Carvalho de Melo <[email protected]>:
o more copy_{to,from}_user fixes
Brad Hards <[email protected]>:
o General options - begone
[email protected] <[email protected]>:
o agppart SiS 745 Patch - did it wrong before
James Simmons <[email protected]>:
o Fbdev updates and fixes
Keith Owens <[email protected]>:
o kbuild-2.5 (slightly older version (didn't have a certain bug))
Maksim Krasnyanskiy <[email protected]>:
o Fix non-modular Bluetooth compilation
o Bluetooth PCMCIA drivers update
Martin Dalecki <[email protected]>:
o airo wireless - "I can't get no, compilation..."
o 2.5.20 IDE 83
Pavel Machek <[email protected]>:
o Cleanup swsusp in 2.5.20
o Fix suspend-to-RAM in 2.5.20
o Re: Fix suspend-to-RAM in 2.5.20
Randy Hron <[email protected]>:
o remove space in cache names
Russell King <[email protected]>:
o fix 2.5.20 ramdisk
Rusty Russell <[email protected]>:
o TRIVIAL: TAGS creation should go into arch dirs
Thunder from the hill <[email protected]>:
o Remove the special handling for hidden files in fs/isofs/namei.c and fs/isofs/dir.c
o Make mount_block_root() wait up to 60 seconds before panic() in case we don't find root fs.
Trond Myklebust <[email protected]>:
o Re: [2.5.20-BUG] 3c59x + highmem + acpi + nfs -> kernel panic
--
Lightweight patch manager using pine. If you have any objections, tell me.
On Mon, Jun 03, 2002 at 09:58:32PM -0600, Lightweight patch manager wrote:
> <[email protected]>:
> o remove mixture of non-atomic operations with page->flags which requires atomic operations to access
> o repetitive reinitialization of active_list and inactive_list in free_area_init_core()
> o make balance_classzone() use list.h
> o complete comment regarding inner workings of buddy system
> o duplicate declaration of rq in sched_init()
> o Re: forget_pte()
> o remove antiquated comment from page_alloc.c
> o convert page_alloc.c bugchecks to BUG_ON()
> o remove MARK_USED() macros
> o remove memlist_* macros from page_alloc.c
> o correct inaccurate comment regarding zone_table's usage
There were discussions about a number of these patches resulting in
changes, would you mind letting me know what versions of these things
you're pushing upstream and let me hand you updates?
Thanks,
Bill
Hi,
On Mon, 3 Jun 2002, William Lee Irwin III wrote:
> There were discussions about a number of these patches resulting in
> changes, would you mind letting me know what versions of these things
> you're pushing upstream and let me hand you updates?
Since you gave it no version number, it's exactly the version which is
saved at
<URL:ftp://luckynet.dynu.com/pub/linux/2.5.20-ct1/single-patches/>
Usually the latest available.
Regards,
Thunder
--
ship is leaving right on time | Thunder from the hill at ngforever
empty harbour, wave goodbye |
evacuation of the isle | free inhabitant not directly
caveman's paintings drowning | belonging anywhere
On Mon, 3 Jun 2002, William Lee Irwin III wrote:
>> There were discussions about a number of these patches resulting in
>> changes, would you mind letting me know what versions of these things
>> you're pushing upstream and let me hand you updates?
On Mon, Jun 03, 2002 at 10:30:18PM -0600, Thunder from the hill wrote:
> Since you gave it no version number, it's exactly the version which is
> saved at
> <URL:ftp://luckynet.dynu.com/pub/linux/2.5.20-ct1/single-patches/>
> Usually the latest available.
Please discard the atomic update patch altogether; there were enough
eyebrows raised that this cannot qualify as a simple cleanup.
Thanks,
Bill
On Mon, 3 Jun 2002, William Lee Irwin III wrote:
>> There were discussions about a number of these patches resulting in
>> changes, would you mind letting me know what versions of these things
>> you're pushing upstream and let me hand you updates?
On Mon, Jun 03, 2002 at 10:30:18PM -0600, Thunder from the hill wrote:
> Since you gave it no version number, it's exactly the version which is
> saved at
> <URL:ftp://luckynet.dynu.com/pub/linux/2.5.20-ct1/single-patches/>
> Usually the latest available.
Rusty Russell gave me copious assistance in clarifying and verifying the
effectiveness of the explanation given in updated buddy comment patches.
Please replace the version you've provided with the following.
Thanks,
Bill
===== mm/page_alloc.c 1.63 vs edited =====
--- 1.63/mm/page_alloc.c Tue May 28 16:57:49 2002
+++ edited/mm/page_alloc.c Mon Jun 3 15:21:55 2002
@@ -82,10 +82,13 @@
* at the bottom level available, and propagating the changes upward
* as necessary, plus some accounting needed to play nicely with other
* parts of the VM system.
- *
- * TODO: give references to descriptions of buddy system allocators,
- * describe precisely the silly trick buddy allocators use to avoid
- * storing an extra bit, utilizing entry point information.
+ * At each level, we keep one bit for each pair of blocks, which
+ * is set to 1 iff only one of the pair is allocated. So when we
+ * are allocating or freeing one, we can derive the state of the
+ * other. That is, if we allocate a small block, and both were
+ * free, the remainder of the region must be split into blocks.
+ * If a block is freed, and its buddy is also free, then this
+ * triggers coalescing into a block of larger size.
*
* -- wli
*/
On Mon, 3 Jun 2002, William Lee Irwin III wrote:
>> There were discussions about a number of these patches resulting in
>> changes, would you mind letting me know what versions of these things
>> you're pushing upstream and let me hand you updates?
On Mon, Jun 03, 2002 at 10:30:18PM -0600, Thunder from the hill wrote:
> Since you gave it no version number, it's exactly the version which is
> saved at
> <URL:ftp://luckynet.dynu.com/pub/linux/2.5.20-ct1/single-patches/>
> Usually the latest available.
Please discard the balance_classzone() patch. I will resubmit after
other cleanups have made it to mainline as it alters some of the same
lines affected by the others.
Thanks,
Bill
Hi,
On Mon, 3 Jun 2002, William Lee Irwin III wrote:
> Please discard the atomic update patch altogether; there were enough
> eyebrows raised that this cannot qualify as a simple cleanup.
Is there something serious to add about them? Is it sure that they won't
work or such? Otherwise I'd suggest just getting them tested.
Regards,
Thunder
--
ship is leaving right on time | Thunder from the hill at ngforever
empty harbour, wave goodbye |
evacuation of the isle | free inhabitant not directly
caveman's paintings drowning | belonging anywhere
On Mon, 3 Jun 2002, William Lee Irwin III wrote:
>> There were discussions about a number of these patches resulting in
>> changes, would you mind letting me know what versions of these things
>> you're pushing upstream and let me hand you updates?
On Mon, Jun 03, 2002 at 10:30:18PM -0600, Thunder from the hill wrote:
> Since you gave it no version number, it's exactly the version which is
> saved at
> <URL:ftp://luckynet.dynu.com/pub/linux/2.5.20-ct1/single-patches/>
> Usually the latest available.
Please replace the forget_pte() patch with the following:
Thanks,
Bill
===== mm/memory.c 1.70 vs edited =====
--- 1.70/mm/memory.c Fri May 31 18:18:07 2002
+++ edited/mm/memory.c Sun Jun 2 22:37:17 2002
@@ -310,17 +310,6 @@
return -ENOMEM;
}
-/*
- * Return indicates whether a page was freed so caller can adjust rss
- */
-static inline void forget_pte(pte_t page)
-{
- if (!pte_none(page)) {
- printk("forget_pte: old mapping existed!\n");
- BUG();
- }
-}
-
static void zap_pte_range(mmu_gather_t *tlb, pmd_t * pmd, unsigned long address, unsigned long size)
{
unsigned long offset;
@@ -779,7 +768,8 @@
pte_t zero_pte = pte_wrprotect(mk_pte(ZERO_PAGE(address), prot));
pte_t oldpage = ptep_get_and_clear(pte);
set_pte(pte, zero_pte);
- forget_pte(oldpage);
+ /* PTE's must be unmapped */
+ BUG_ON(!pte_none(oldpage));
address += PAGE_SIZE;
pte++;
} while (address && (address < end));
@@ -857,7 +847,8 @@
if (!pfn_valid(pfn) || PageReserved(pfn_to_page(pfn)))
set_pte(pte, pfn_pte(pfn, prot));
- forget_pte(oldpage);
+ /* PTE's must be unmapped */
+ BUG_ON(!pte_none(oldpage));
address += PAGE_SIZE;
pfn++;
pte++;
On Mon, 3 Jun 2002, William Lee Irwin III wrote:
>> Please discard the atomic update patch altogether; there were enough
>> eyebrows raised that this cannot qualify as a simple cleanup.
On Mon, Jun 03, 2002 at 10:41:15PM -0600, Thunder from the hill wrote:
> Is there something serious to add about them? Is it sure that they won't
> work or such? Otherwise I'd suggest just getting them tested.
The original patch as posted is incorrect due to a misreading on my
part of what the flags clearing did. One of the few remotely close
to correct alternatives follows, but I will not endorse it as a
candidate for inclusion, but give it only as an illustration of how
incorrect the originally posted patch was.
Cheers,
Bill
===== mm/page_alloc.c 1.63 vs edited =====
--- 1.63/mm/page_alloc.c Tue May 28 16:57:49 2002
+++ edited/mm/page_alloc.c Mon Jun 3 16:27:41 2002
@@ -111,7 +111,7 @@
if (PageWriteback(page))
BUG();
ClearPageDirty(page);
- page->flags &= ~(1<<PG_referenced);
+ __clear_bit(PG_referenced, &page->flags);
if (current->flags & PF_FREE_PAGES)
goto local_freelist;
On Mon, 2002-06-03 at 21:41, Thunder from the hill wrote:
> On Mon, 3 Jun 2002, William Lee Irwin III wrote:
> > Please discard the atomic update patch altogether; there were enough
> > eyebrows raised that this cannot qualify as a simple cleanup.
>
> Is there something serious to add about them? Is it sure that they won't
> work or such? Otherwise I'd suggest just getting them tested.
Maybe it is not wise to integrate a tree with patches you do to know are
correct or not?
Robert Love
Hi,
On 4 Jun 2002, Robert Love wrote:
> Maybe it is not wise to integrate a tree with patches you do to know are
> correct or not?
I was just wondering why my testal machine wasn't yet burning...
Regards,
Thunder
--
ship is leaving right on time | Thunder from the hill at ngforever
empty harbour, wave goodbye |
evacuation of the isle | free inhabitant not directly
caveman's paintings drowning | belonging anywhere