Skip to contentSkip to conversation

x86 updates

GIT PULL Thomas Gleixner 4 participants 5 messages Feb 9, 2008 – Feb 10, 2008

lore
Share & tools
lore
#0 lore raw

Linus,

please pull the pending x86 updates from:

ssh://master.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86.git master

The update contains:

- a couple of bugfixes
- CPA and DEBUG_PAGEALLOC improvements
- x86 power management consolidation
- GEODE updates
- 32bit boot time page table construction rework
- sparse and compile warning fixes
- trivial cleanups

There are two patches out of x86 scope as well:

- lguest bugfix: x86 broke it, fix is obviously correct and Rusty is away

- randomization docs: resulted out of a x86 randomization discussion

Thanks,

tglx

----

Ahmed S. Darwish (1):
lguest: accept guest _PAGE_PWT page table entries

Andres Salomon (5):
x86: GEODE: MFGPT: Minor cleanups
x86: GEODE: MFGPT: drop module owner usage from MFGPT API
x86: GEODE: MFGPT: replace 'flags' field with 'avail' bit
x86: GEODE: MFGPT: make mfgpt_timer_setup available outside of mfgpt_32.c
x86: GEODE: MFGPT: fix a potential race when disabling a timer

Arnd Hannemann (1):
x86: GEODE: MFGPT: fix typo in printk in mfgpt_timer_setup

Denys Vlasenko (1):
x86: trivial printk optimizations

Harvey Harrison (6):
x86: fix sparse warning in xen/time.c
x86: sparse warning in therm_throt.c
x86: sparse warnings in pageattr.c
x86: fix sparse warning in topology.c
x86: fix sparse warnings in acpi/bus.c
x86, core: remove CONFIG_FORCED_INLINING

Ian Campbell (2):
x86: construct 32-bit boot time page tables in native format.
x86: fix early_ioremap pagetable ops

Ingo Molnar (2):
x86: fixup more paravirt fallout
brk: help text typo fix

Jiri Kosina (1):
brk: document randomize_va_space and CONFIG_COMPAT_BRK (was Re:

Jordan Crouse (2):
x86: GEODE: MFGPT: Use "just-in-time" detection for the MFGPT timers
x86: GEODE: make sure the right MFGPT timer fired the timer tick

Rafael J. Wysocki (4):
x86 PM: move 64-bit hibernation files to arch/x86/power x86 PM: rename 32-bit files in arch/x86/power x86 PM: consolidate suspend and hibernation code x86 PM: update stale comments

Thomas Gleixner (6):
x86: avoid unused variable warning in mm/init_64.c
x86: DEBUG_PAGEALLOC: enable after mem_init()
x86: introduce page pool in cpa
x86: cpa, use page pool
x86: cpa, enable CONFIG_DEBUG_PAGEALLOC on 64-bit
x86: cpa, strict range check in try_preserve_large_page()

Willy Tarreau (1):
x86: GEODE fix MFGPT input clock value

#1 lore raw
Quoted 5 lines · On Sun, 10 Feb 2008 00:24:50 +0100 (CET) Thomas Gleixner wrote:

Linus,

please pull the pending x86 updates from:

ssh://master.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86.git master

Hi Thomas,
can we please get diffstats with git pull requests? (in the future)

Quoted 77 lines

The update contains:

- a couple of bugfixes
- CPA and DEBUG_PAGEALLOC improvements
- x86 power management consolidation
- GEODE updates
- 32bit boot time page table construction rework
- sparse and compile warning fixes
- trivial cleanups

There are two patches out of x86 scope as well:

- lguest bugfix: x86 broke it, fix is obviously correct and Rusty is away

- randomization docs: resulted out of a x86 randomization discussion

Thanks,

tglx

----

Ahmed S. Darwish (1):
lguest: accept guest _PAGE_PWT page table entries

Andres Salomon (5):
x86: GEODE: MFGPT: Minor cleanups
x86: GEODE: MFGPT: drop module owner usage from MFGPT API
x86: GEODE: MFGPT: replace 'flags' field with 'avail' bit
x86: GEODE: MFGPT: make mfgpt_timer_setup available outside of mfgpt_32.c
x86: GEODE: MFGPT: fix a potential race when disabling a timer

Arnd Hannemann (1):
x86: GEODE: MFGPT: fix typo in printk in mfgpt_timer_setup

Denys Vlasenko (1):
x86: trivial printk optimizations

Harvey Harrison (6):
x86: fix sparse warning in xen/time.c
x86: sparse warning in therm_throt.c
x86: sparse warnings in pageattr.c
x86: fix sparse warning in topology.c
x86: fix sparse warnings in acpi/bus.c
x86, core: remove CONFIG_FORCED_INLINING

Ian Campbell (2):
x86: construct 32-bit boot time page tables in native format.
x86: fix early_ioremap pagetable ops

Ingo Molnar (2):
x86: fixup more paravirt fallout
brk: help text typo fix

Jiri Kosina (1):
brk: document randomize_va_space and CONFIG_COMPAT_BRK (was Re:

Jordan Crouse (2):
x86: GEODE: MFGPT: Use "just-in-time" detection for the MFGPT timers
x86: GEODE: make sure the right MFGPT timer fired the timer tick

Rafael J. Wysocki (4):
x86 PM: move 64-bit hibernation files to arch/x86/power x86 PM: rename 32-bit files in arch/x86/power x86 PM: consolidate suspend and hibernation code x86 PM: update stale comments

Thomas Gleixner (6):
x86: avoid unused variable warning in mm/init_64.c
x86: DEBUG_PAGEALLOC: enable after mem_init()
x86: introduce page pool in cpa
x86: cpa, use page pool
x86: cpa, enable CONFIG_DEBUG_PAGEALLOC on 64-bit
x86: cpa, strict range check in try_preserve_large_page()

Quoted 2 lines

Willy Tarreau (1):
x86: GEODE fix MFGPT input clock value

---
~Randy

#2 lore raw
Quoted 10 lines · Randy Dunlap
Quoted 5 lines · On Sun, 10 Feb 2008 00:24:50 +0100 (CET) Thomas Gleixner wrote:

Linus,

please pull the pending x86 updates from:

ssh://master.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86.git master

Hi Thomas,
can we please get diffstats with git pull requests?

Well, not only that, but I actually pulled it once, then ended up undoing my pull, and had to think twice about whether I want to pull it at all.

I do *not* like it when subsystem maintainers start mixing in stuff into their subsystem pull that has absolutely _zero_ to do with that subsystem. In this case, we have commit 9b706aee7d92d6ac3002547aea12e3eaa0a750ae, which calls itself "x86: trivial printk optimizations", but it really has nothing AT ALL to do with x86 except that it also did the same things to the bogus x86 version of those routines.

The thing is, trying to mix up things like that into a subsystem pull means that now I cannot trust the subsystem maintainer as much! And what should be a simple "git pull" becomes not just the pull, but also me having to then scrounge around in the result to see if I really want to do the pull in the first place.

That's the point where the subsystem maintainer just makes me do unnecessary extra and stupid work!

In other words: DO NOT DO THINGS LIKE THAT!

That vsnprintf() optimization looks ok, but it really doesn't have anything what-so-ever to do with x86, and some of it is rather dubious and too clever by half for its own good, for rather little gain. For example, we have:

	+/* Works only for digits and letters, but small and fast */
	+#define TOLOWER(x) ((x) | 0x20)

but then later on we have:

- if (cp[0] == '0' && toupper(cp[1]) == 'X')
+ if (cp[0] == '0' && TOLOWER(cp[1]) == 'x')

where we now use that new TOLOWER() macro on a character that is *not* guaranteed to be a digit or a letter at all!

Does it work? Yes, it does happen to work. If it's a non-character or a non-letter, it will do essentially random things, but it will never turn that non-character/letter into 'x' unless it was 'X' or 'x' before. So yes, it works, but let's face it, that clever trick saved something like two instructions and a branch from a code-path that wasn't really even critical!

In general, I don't disagree with being clever. People enjoy tricks like that, and I don't really disagree with the patch. That's not the problem I have. The problem I have is that I shouldn't be taken by surprise by these things, and start having to worry whether I can trust the person who sends me mis-labeled patches without even warning me.

So I had this thing in my tree, went out to dinner, came back and decided I don't want to pull it after all. Did some other work, and decided I have the time to look through it, and then re-pulled it.

But I was *this* close to just deciding that "ok, I'm ready to release the -rc1 kernel, I don't need this kind of aggravation" and just decided to not bother pulling something dubious.

So Thomas, don't do this. I don't like it. The same way I didn't like seeing Ingo trying to mix in a kgdb pull into his x86 pull. Keep these things separate - git is *really* good at having multiple branches with different lines of development, use it that way (or send odd-ball misc patches just as emails).

But it's merged now.

Linus

PS. And no, maybe I don't always notice. I bet maintainers can slip things by me all the time without me waking up to it. The fact that it sometimes works doesn't mean that it's a good idea, though - it just means that if I catch it, my level of trust goes down.

#3 lore raw
Quoted 7 lines · Linus Torvalds

On Sat, 9 Feb 2008, Randy Dunlap wrote:

So Thomas, don't do this. I don't like it. The same way I didn't like seeing Ingo trying to mix in a kgdb pull into his x86 pull. Keep these things separate - git is *really* good at having multiple branches with different lines of development, use it that way (or send odd-ball misc patches just as emails).

Please accept my apologies.

I went through the content, pointed out the other two non x86 patches and did not notice the vsprintf one. My bad.

Quoted 4 lines

PS. And no, maybe I don't always notice. I bet maintainers can slip things by me all the time without me waking up to it. The fact that it sometimes works doesn't mean that it's a good idea, though - it just means that if I catch it, my level of trust goes down.

I know. And I have no intention to sneak stuff in that way.

Conclusions:
1) add diffstat to the pull request generator
2) add alert about non x86 stuff to it and send it separate
3) postpone pull requests to the morning hours, even if time is short

Thanks,

tglx

#4 lore raw
Quoted 9 lines · Thomas Gleixner
Quoted 7 lines · Linus Torvalds

On Sat, 9 Feb 2008, Randy Dunlap wrote:

So Thomas, don't do this. I don't like it. The same way I didn't like seeing Ingo trying to mix in a kgdb pull into his x86 pull. Keep these things separate - git is *really* good at having multiple branches with different lines of development, use it that way (or send odd-ball misc patches just as emails).

Quoted 4 lines

Please accept my apologies.

I went through the content, pointed out the other two non x86 patches and did not notice the vsprintf one. My bad.

and i'd like to apologize for the misleading Subject line which said "Subject: x86: ...". That i think made us miss the generic impact later on (we remind ourselves about non-arch/x86 patches via Subject line annotations and via looking at the diffstat before pull requests), and the accidentally missing diffstat just removed the last-line defense against such annotation mistakes.

btw., i was -->.<--- this close to removing this patch from x86.git altogether - because it had a bug (which already shows its non-obviousness) and i didnt think this complication was worth it. When that happened i took a really hard look at its correctness and worthiness and narrowly decided to keep it - but enforced an unusually long testing interval for this patch.

Ingo