* Re: [RFC PATCH 16/35] subarch support for interrupt and exception gates
From: Andi Kleen @ 2006-05-09 11:09 UTC (permalink / raw)
To: virtualization; +Cc: Chris Wright, Ian Pratt, xen-devel, linux-kernel
In-Reply-To: <20060509085154.441800000@sous-sol.org>
> +/*
> + * This needs to use 'idt_table' rather than 'idt', and
> + * thus use the _nonmapped_ version of the IDT, as the
> + * Pentium F0 0F bugfix can have resulted in the mapped
> + * IDT being write-protected.
> + */
> +void set_intr_gate(unsigned int n, void *addr)
> +{
> + _set_gate(idt_table+n,14,0,addr,__KERNEL_CS);
> +}
No need to duplicate the various set_*_gate functions into the subarchs.
> +static void __init set_task_gate(unsigned int n, unsigned int gdt_entry)
> +{
> + /* _set_gate(n, 5, 0, 0, (gdt_entry<<3)); */
> +}
Looks weird, but can be handled in the low level function.
-Andi
^ permalink raw reply
* [U-Boot-Users] How to support multiple flash sizes ?
From: Joao, Nuno @ 2006-05-09 11:10 UTC (permalink / raw)
To: u-boot
Hello,
I've also a (MPC8541E based) system where I can have 64M or 128M
flash. I'm using CFI, and have this config:
/*
* FLASH on the Local Bus.
* Boot from BR0/OR0 bank.
*/
#define CFG_FLASH_MEM_BASE 0xF8000000 /* Flash start at 128M from end of addr space */
#define CFG_FLASH_MEM_BASE2 0xFC000000 /* Flash start at 64M from end of addr space */
#define CFG_FLASH_BASE (0 - flash_info[0].size - flash_info[1].size)
#define CFG_FLASH_BANKS_LIST { CFG_FLASH_MEM_BASE2, CFG_FLASH_MEM_BASE }
#define CFG_LOWEST_FLASH_BASE CFG_FLASH_MEM_BASE /* MUST have the lowest possible flash start addr */
/* probe for 64M at CFG_FLASH_MEM_BASE2 and 128M at CFG_FLASH_MEM_BASE */
#define CFG_FLASH_BANKS_SIZES_LIST { (64 * 1024*1024), (128 * 1024*1024) }
#define CONFIG_FLASH_PREDEFINED_PROBING /* enable predefined flash probing */
and I did the following changes ("CCEP" is our board's name):
drivers/cfi_flash.c:
170a171,173
> #ifdef CONFIG_FLASH_PREDEFINED_PROBING
> static ulong bank_expected_size[CFG_MAX_FLASH_BANKS] = CFG_FLASH_BANKS_SIZES_LIST;
> #endif
338a342,356
> #ifdef CONFIG_FLASH_PREDEFINED_PROBING
> else
> {
> if (flash_info[i].size != bank_expected_size[i])
> {
> flash_info[i].flash_id = FLASH_UNKNOWN;
> size -= flash_info[i].size;
> flash_info[i].size = 0; /* don't remove this! */
> }
> else
> {
> printf("found flash @ 0x%08X: ", bank_base[i]);
> }
> }
> #endif
341,346c359,371
< /* Monitor protection ON by default */
< #if (CFG_MONITOR_BASE >= CFG_FLASH_BASE)
< flash_protect (FLAG_PROTECT_SET,
< CFG_MONITOR_BASE,
< CFG_MONITOR_BASE + monitor_flash_len - 1,
< flash_get_info(CFG_MONITOR_BASE));
---
> #define ENABLE_MONITOR_PROTECTION() \
> /* Monitor protection ON by default */ \
> flash_protect (FLAG_PROTECT_SET, \
> CFG_MONITOR_BASE, \
> CFG_MONITOR_BASE + monitor_flash_len - 1, \
> flash_get_info(CFG_MONITOR_BASE))
>
> #if defined(CONFIG_CCEP)
> ENABLE_MONITOR_PROTECTION();
> #else
> # if (CFG_MONITOR_BASE >= CFG_FLASH_BASE)
> ENABLE_MONITOR_PROTECTION();
> # endif
This is based on u-boot 1.1.3. You will get some "unkown" banks for
the smaller flashes.
It's certainly not the best solution, but it's a solution.
Best Regards
-----Original Message-----
From: u-boot-users-admin@lists.sourceforge.net [mailto:u-boot-users-admin at lists.sourceforge.net] On Behalf Of Laurent Pinchart
Sent: ter?a-feira, 9 de Maio de 2006 10:08
To: u-boot-users at lists.sourceforge.net
Subject: [U-Boot-Users] How to support multiple flash sizes ?
Hi,
I'm trying to find a way for U-Boot to automatically detect the flash base
address.
Our custom board can be mounted with different flash options, from 16MB to
128MB. I would like to avoid programming the board with a different boot
loader for each flash size, so I was trying a way to create a
"one-fits-them-all" U-Boot version.
(...)
^ permalink raw reply
* [U-Boot-Users] How to support multiple flash sizes ?
From: Wolfgang Denk @ 2006-05-09 11:11 UTC (permalink / raw)
To: u-boot
In-Reply-To: <200605091250.45978.laurent.pinchart@tbox.biz>
In message <200605091250.45978.laurent.pinchart@tbox.biz> you wrote:
>
> The only place where U-Boot handles flash memory mapping is in cpu_init_f (in
> cpu/mpc8260/cpu_init.c). The flash is mapped using CFG_OR0_PRELIM and
> CFG_BR0_PRELIM which are compile-time constants. I can't find any flash
> remapping code.
Please check again. Guess what the "PRELIM" means. BTW: This is
documented in the README.
> Where does it remap the flash after detecting its size ?
In the flash driver.
> > Why do you define a TEXT_BASE+256kB location for your environment then?
>
> That was copied from another board's configuration file. I'd like to put the
Never copy anything without really understanding why you need it, and
what it means for your baord.
> other boards. The monitor must be located at 0xFFF00000 and is thus at a
> known constant offset from the *end* of the flash. As I don't know the flash
You are wrong. You can, for example, map your flash at 0x40000000 as
well. The location of the reset entry point has nothing to do with
your final flash mapping.
BTW: You may consider switching to use a low-boot configuration (i.
e. 0x100 reset vector). This usually has several advantages, easier
mapping and better flash use being just the most obvious.
> size at compile time, I can't define the environment address as an offset
> from the beginning of the flash.
This can be done, too, if really necessary (which it rarely is). See
for example the definition of CFG_FLASH_BASE for the PCU_E board. But
really: this is just meant as a demonstration that such a thing is
possibe. I did NOT say that it was a good idea. Just use a sane flash
mapping. Say, you map it at 0x40000000, ok?
> Let's assume I'm using a 16MB flash chip, with options up to 128MB.
>
> To accommodate space for the largest possible flash, I must define
> CFG_FLASH_BASE to 0xF8000000 with a 128MB size. TEXT_BASE *must* be defined
No. You can use any other base address as well. Let's say, you use
0xF0000000, or 0xE0000000.
> to 0xFFF00000, as the processor's interrupt vectors are located at that
> address. The flash is mapped from 0xF8000000 to 0xFFFFFFFF, but physically
The reset vector's location has NOTHING to do with the final flash
mapping.
> At bootup, the processor starts running at 0xFFF00100. This is inside the 8th
> flash copy in its mapping, and the monitor is stored in flash at the correct
> offset, so everything works correctly.
No. At reset, there is no mapping at all, and the defaults of the
processor apply. Please re-read the relevant section in the User's
manual. At this point, no code has been run, so it is obviously
completely irrelevant if you will later map the flash to this address
or any other one (if this will work is a different story, but here it
is irrelevant).
> When U-Boot wants to write the environment, it will try to erase sector
> 0xFFF40000. This leads to an error, as U-Boot knows the flash is only 16MB
> large, and 0xFFF40000 is thus outside the flash.
This is a misconfiguration of your board. Fix it.
> Is this clearer, or is there still something you don't understand ?
Ummm... I don't think it's me who doesn't understand ;-)
Best regards,
Wolfgang Denk
--
Software Engineering: Embedded and Realtime Systems, Embedded Linux
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
The project was large enough and management communication poor enough
to prompt many members of the team to see themselves as contestants
making brownie points, rather than as builders making programming
products. Each suboptimized his piece to meet his targets; few
stopped to think about the total effect on the customer.
- Fred Brooks, "The Mythical Man Month"
^ permalink raw reply
* Re: [LARTC] Matching interface using U32(?)
From: Nataniel Klug @ 2006-05-09 11:11 UTC (permalink / raw)
To: lartc
In-Reply-To: <445B5F2C.604@cnett.com.br>
Andy
I make it work using iptables mark... I just had to change some parts of
the script and it is, now, working fine.
Att,
Nataniel Klug
Andy Furniss escreveu:
> Nataniel Klug wrote:
>> Hello all,
>>
>> I am trying to make a little bit more complex QoS/Shapping form
>> and I need to shape a PPPoE conection that I serve to my clients. So
>> this is the scope:
>>
>> client connects using pppoe so it gets an IP address (from pppoe
>> pool) and open an interface into my linux box
>> interface for this client is ppp0
>> client has got an ip 1.1.1.2/32 and it is poiting to pppoe-server
>> 1.1.1.1
>>
>> So, if i want to shape download for this client its really easy
>> and I make a htb rule that shapes all traffic into ppp0... My problem
>> is when I have to shape upload traffic. All my traffic goes for eth0
>> to the world. I tryed to make this rule:
>>
>> $TC class add dev eth0 parent 1:1 classid 1:500 htb rate 200Kbit ceil
>> 200Kbit
>> $TC filter add dev eth0 parent 1:0 protocol ip prio 1 u32 match ip
>> src 1.1.1.2/32 flowid 1:500
>
> protocol ip on the filter is the problem - Without trying I am not
> sure what will work but in the case of vlans you can just sat protocol
> 8021q.
>
> If pppoe doesn't work try its' ethertype number - you could also use
> protocol all and match the ethertype protocol number with u32 and a
> negative offset.
>
>>
>> But this rule is not matching the upload connection. If i put the
>> same rule to an IP (i set an ip into client) this rule works but into
>> pppoe conection it did not work properly.
>>
>> Someone has some tip to match the interface that the conection is
>> coming from? Like I wanna match all traffic coming from ppp0 and
>> going throw eth0 to be shapped?! There is anyway to make this work?
>
> If you are running >2.6.16 then you could use IFB and attach a queue
> to each ppp.
>
> There is a kernel config option for u32 to mach indev - I don't know
> the syntax,though.
>
> Andy.
>
>
_______________________________________________
LARTC mailing list
LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/cgi-bin/mailman/listinfo/lartc
^ permalink raw reply
* [U-Boot-Users] AT91RM9200 Dataflash op error with U-Boot1.1.4
From: KylongMu @ 2006-05-09 11:11 UTC (permalink / raw)
To: u-boot
In-Reply-To: <20060508171505.09348FE89@sc8-sf-spam2.sourceforge.net>
Dear Groups:
My board contains: AT91RM9200 DataFlash:AT45DB642 (without nand
flash)
Crosstool: ELDK4.0 for arm
UBoot: Get from www.denx.de with git, and update to most recent
code.
UBoot Config file: at91rm9200dk_config
At fist, I compiled the uboot with at91rm9200dk config, but it halt
after boot and without prompt. It's work after I add some definition in the
file include/configs/at91rm9200dk.h like this:
----------------------------------------------------------------------------
#define CONFIG_ENV_OVERWRITE
#define CONFIG_SKIP_LOWLEVEL_INIT
#define CFG_ENV_IS_IN_DATAFLASH
----------------------------------------------------------------------------
I think this is very confuse for a uboot user, it waste me many time
to find out the correct way to make it work, I think it should be with a
simple define to make sure if you boot from dataflash.
The good message is my board can show uboot prompt now, the bad
message is when I try to erase the DataFlash, it shows error like this:
----------------------------------------------------------------------------
*--------------------------------------*
DataFlash[15Mhz]:AT45DB642
Nb pages: 8192
Page Size: 1056
Size= 8650752 bytes
Logical address: 0xC0000000
*--------------------------------------*
1: Download Dataflash [addr]
2: Read Dataflash [addr]
3: Start UBOOT [C0008000 => 20000000]
*--------------------------------------*
Enter:
Load UBOOT from dataflash[c0008000] to SDRAM[21f00000]
PLLA[180MHz], MCK[60Mhz] ==> Start UBOOT
U-Boot 1.1.4 (May 8 2006 - 20:12:59)
DRAM: 64 MB
Atmel: Flash: 0 kB
DataFlash:AT45DB642
Nb pages: 8192
Page Size: 1056
Size= 8650752 bytes
Logical address: 0xC0000000
Area 0: C0000000 to C0007FFF (RO)
Area 1: C0008000 to C001FFFF (RO)
Area 2: C0020000 to C0027FFF
Area 3: C0028000 to C083FFFF
*** Warning - bad CRC, using default environment
In: serial
Out: serial
Err: serial
U-Boot> erase c0020000 c0027fff
Error: start address not on sector boundary
U-Boot>
----------------------------------------------------------------------------
Does anybody know how to solve with this?
Thanks a lot!
KylongMu
^ permalink raw reply
* [U-Boot-Users] (no subject)
From: Wolfgang Denk @ 2006-05-09 11:13 UTC (permalink / raw)
To: u-boot
In-Reply-To: <E7E846FF4F1FE947AABFF850350820780CC09B@PTLISI051MSX.PT001.SIEMENS.NET>
In message <E7E846FF4F1FE947AABFF850350820780CC09B@PTLISI051MSX.PT001.SIEMENS.NET> you wrote:
>
> Hello,
>
> I've had a similar problem when booting an MPC8541E custom board.
...
> -----Original Message-----
Grrrrgh.. No subject, even though replying to a previous message, and
top-post/full-quote?
PLEASE read the netiquette, see
http://www.netmeister.org/news/learn2quote.html
Best regards,
Wolfgang Denk
--
Software Engineering: Embedded and Realtime Systems, Embedded Linux
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
If a man had a child who'd gone anti-social, killed perhaps, he'd
still tend to protect that child.
-- McCoy, "The Ultimate Computer", stardate 4731.3
^ permalink raw reply
* [Bridge] I want to realize spanning tree , can somebody help me?
From: zhangliang @ 2006-05-09 11:14 UTC (permalink / raw)
To: bridge
[-- Attachment #1: Type: text/plain, Size: 829 bytes --]
Hello, everybody:
Now my system’s structure as follows:
One mips cpu + one switch(which has five ports), switch’s one port
connect witch cpu with MII, and
The switch’s other four port connect with RJ45 which is used for connecting
with other pc or switch.
Now I want to realize stp in this system, what I want to know is that is it
realizable? One cpu + one switch
With five ports , thanks!
Another system is that one cpu + one switch which has five pors, this
cpu has 2 MAC connected
With switch’s port 3 & port 4, then port0, port1, port2 connected with
RJ45, now I want to realize stp in
This system, is it realizable?
Thanks!
此致
敬礼
张亮
浙江大学数字所
[-- Attachment #2: Type: text/html, Size: 6197 bytes --]
^ permalink raw reply
* Re: [RFC PATCH 16/35] subarch support for interrupt and exception gates
From: Andi Kleen @ 2006-05-09 11:09 UTC (permalink / raw)
To: virtualization; +Cc: Chris Wright, linux-kernel, xen-devel, Ian Pratt
In-Reply-To: <20060509085154.441800000@sous-sol.org>
> +/*
> + * This needs to use 'idt_table' rather than 'idt', and
> + * thus use the _nonmapped_ version of the IDT, as the
> + * Pentium F0 0F bugfix can have resulted in the mapped
> + * IDT being write-protected.
> + */
> +void set_intr_gate(unsigned int n, void *addr)
> +{
> + _set_gate(idt_table+n,14,0,addr,__KERNEL_CS);
> +}
No need to duplicate the various set_*_gate functions into the subarchs.
> +static void __init set_task_gate(unsigned int n, unsigned int gdt_entry)
> +{
> + /* _set_gate(n, 5, 0, 0, (gdt_entry<<3)); */
> +}
Looks weird, but can be handled in the low level function.
-Andi
^ permalink raw reply
* [KJ] Re: the CtALlkS
From: Gavino Hackley @ 2006-05-09 11:16 UTC (permalink / raw)
To: kernel-janitors
[-- Attachment #1.1.1: Type: text/plain, Size: 1001 bytes --]
web site <http://ca.geocities.com/HalvdovMayesdwin/>
still forward after dusk and into the night beneath the moon. As the
light faded Bilbo thought he saw away to the right, or to the left, the
shadowy form of a great bear prowling along in the same direction. But
if he dared to mention it to Gandalf, the wizard only said: Hush! Take
no notice! Next day they started before dawn, though their night had
been short. As soon as it was light they could see the forest coming as
it were to meet them, or waiting for them like a black and frowning wall
before them. The land began to slope up and up, and it seemed to the
hobbit that a silence began to draw in upon them. Birds began to sing
less. There were no more deer; not even rabbits were to be seen. By the
afternoon they had reached the eaves of Mirkwood, and were resting
almost beneath the great overhanging boughs of its outer trees. Their
trunks were huge and gnarled, their branches twisted, their leaves were
[-- Attachment #1.1.2: Type: text/html, Size: 1495 bytes --]
[-- Attachment #1.2: Type: image/gif, Size: 2414 bytes --]
[-- Attachment #2: Type: text/plain, Size: 168 bytes --]
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/kernel-janitors
^ permalink raw reply
* [ALSA - lib 0002095]: Choppy sound using default dmix config for some applications but not others.
From: bugtrack @ 2006-05-09 11:20 UTC (permalink / raw)
To: alsa-devel
A NOTE has been added to this issue.
======================================================================
<https://bugtrack.alsa-project.org/alsa-bug/view.php?id=2095>
======================================================================
Reported By: BigBaaadBob
Assigned To: perex
======================================================================
Project: ALSA - lib
Issue ID: 2095
Category: 1_pcm_d - direct PCM plugins (dmix+)
Reproducibility: always
Severity: major
Priority: normal
Status: assigned
======================================================================
Date Submitted: 05-03-2006 01:09 CEST
Last Modified: 05-09-2006 13:20 CEST
======================================================================
Summary: Choppy sound using default dmix config for some
applications but not others.
Description:
As reported in http://forums.gentoo.org/viewtopic-t-457951.html
alsa 1.0.11 causes skipping sound in some applications but not others.
Sound skips in rhytymbox and totem, but not xmms, realplayer, or mplayer.
This occurs with default dmix setup (i.e. with no .asoundrc). Creating an
.asoundrc like this causes the skipping to stop for those applications
that skip.
pcm.!default {
type hw
card 0
}
This has been reported using the cmipci driver, the Intel ICH4 driver, the
i915GM driver, and on a ppc iBook G4.
On user reports that fiddling with the .asoundrc parameters can lessen the
problem, but not eliminate the problem.
======================================================================
----------------------------------------------------------------------
razamatan - 05-09-06 07:23
----------------------------------------------------------------------
actually, my experience is similar but different.
i have a abit mobo using the via chipset (via82xx), and i use a 2.6.16-r6
gentoo-sources kernel and the driver is compiled from the kernel source.
my sound gets severly garbled (not choppy, but sounds like cat /dev/random
> /dev/audio) at odd times when i try using ogg123, mpg321, aplay.
however, when i use anything backended by xine or madplay everything's
fine.
when i use a asoundrc that looks like:
pcm.!default {
type hw
card 0
}
things sorta work... nothing is garbled, but things play in roughly 2x
speed.
using a asoundrc that looks like:
defaults.pcm.dmix_max_periods -1
doesn't do anything at all (garbled audio using mpg321, ogg123, aplay).
i've since reverted to alsa-lib-1.0.10 w/ no asoundrc and things work
peachy like it should.
----------------------------------------------------------------------
dsandras - 05-09-06 13:20
----------------------------------------------------------------------
Many Ekiga users are also complaining about this problem. Reverting to an
older libasound2 makes it disappear.
It also crashes the Ekiga ALSA plugin:
#3 0xb6376fe5 in free () from /lib/tls/libc.so.6
#4 0xb58b6f70 in snd_pcm_adpcm_decode () from /usr/lib/libasound.so.2
#5 0xb58955b9 in snd_pcm_hw_free () from /usr/lib/libasound.so.2
#6 0xb58bc3ed in _snd_pcm_rate_linear_open () from
/usr/lib/libasound.so.2
#7 0xb58955b9 in snd_pcm_hw_free () from /usr/lib/libasound.so.2
#8 0xb589fc2e in snd_pcm_close () from /usr/lib/libasound.so.2
I think that the person who will fix that problem should check everything
works OK with Ekiga and contact me personally in case of problem.
(The backtrace above is obtained after a export MALLOC_CHECK_=2 to force
crashes in case of memory allocation problems).
Issue History
Date Modified Username Field Change
======================================================================
05-03-06 01:09 BigBaaadBob New Issue
05-03-06 01:27 wellwhoopdedoooNote Added: 0009580
05-03-06 01:28 wellwhoopdedoooIssue Monitored: wellwhoopdedooo
05-03-06 01:29 wellwhoopdedoooNote Edited: 0009580
05-03-06 01:31 BigBaaadBob Note Added: 0009581
05-03-06 11:45 tiwai Note Added: 0009586
05-03-06 19:27 tiwai Note Added: 0009601
05-05-06 22:49 BigBaaadBob Note Added: 0009632
05-09-06 06:48 razamatan Note Added: 0009651
05-09-06 06:49 razamatan Issue Monitored: razamatan
05-09-06 06:55 razamatan Note Edited: 0009651
05-09-06 07:23 razamatan Note Edited: 0009651
05-09-06 13:20 dsandras Note Added: 0009654
======================================================================
-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
^ permalink raw reply
* Re: Umount call getting stuck, hanging nfs?
From: Ian Kent @ 2006-05-09 11:21 UTC (permalink / raw)
To: Mike Marion; +Cc: autofs
In-Reply-To: <20060508233437.GG9873@cornholio.qualcomm.com>
On Mon, 8 May 2006, Mike Marion wrote:
> On Fri, May 05, 2006 at 09:59:14AM +0800, Ian Kent wrote:
>
> > I think that might be the other way around.
>
> You mean rpciod is already hung before the umount, and the umount is
> then hung due to the rpciod?
My mistake. We don't actually know which way around it is.
>
> > > /bin/umount //usr/local/projects/dsp/qdsp6
> > >
> > > root 6270 0.0 0.0 0 0 ? D Apr28 3:17 [rpciod]
> > >
> > > unfortunately, once this happens, any new mounts will fail. Can't even
> > > stat the path above via df. Basically the whole NFS layer is stuck.
> >
> > Tell us what the maps look like.
>
> The noted path is like this:
> /prj/dsp/qdsp6 -rw,acdirmin=1,acdirmax=5,acregmin=1,acregmax=5,rsize=32768,wsize=32768,noquota
> western:/vol/eng_aus_0004/qdsp6
>
> And the rest of the file is very similar. The huge amount of options
> came from trial and error with performance problems we were having, and
> they're defaulting to tcp mounts. The odd thing is that we basically
> never have mounts hanging off this /prj tree in San Diego, or any other
> office except for one. And it's only hanging when talking to their 2
> local NetApp filers.
This problem does appear a bit like an NFS problem.
Have you checked the versions for each of the subsystems involved and
compared working and not working?
Ian
^ permalink raw reply
* Re: Implementing branch attributes in git config
From: Johannes Schindelin @ 2006-05-09 11:21 UTC (permalink / raw)
To: Jakub Narebski; +Cc: git
In-Reply-To: <e3p5om$djs$1@sea.gmane.org>
Hi,
On Tue, 9 May 2006, Jakub Narebski wrote:
> Linus Torvalds wrote:
>
> > I would suggest a much more readable format:
> >
> > [core]
> > ...
> >
> > [branch "core"]
> > ...
> >
> > [remote "core"]
> > ...
> >
> > and yes, enforce the <space>+<quoted name> format. We'd turn it internally
> > into some random internal string format (probably replacing the space with
> > a dot, and removing the quotes, so it would become "remote.core.<key>").
>
> I'm all for it. Nice compromise of [branch."miXedCaps"] and ["miXedCaps"],
> human readable end editable, and easy parsable.
Okay, to summarize what people proposed (and that I remember):
1) [branch."AnY+String"]
2) multiple [branch]
3) magic <space>+<quoted>
4) [branch.just/allow-slashes/and-dashes]
5) the " for " notation
Of all these, only (5) is backwards compatible, but it is also the only
one where you have to type the branch name over and over again.
However, the old gits do not really know what to do with the [branch]
section anyway, so you could consider (2) (and (4), if you do not have
branch names with slashes and/or dashes) backwards-compatible, because git
will continue to work -- ignoring the funny entries.
(1) and (3) definitely would make an old git choke.
Now, for the ease of use:
(1), (3) and (4) are in the same league of easiness (except maybe that you
have to keep in mind to extra-quote in shell scripts with (1) and (3)),
(2) is especially good for people with a database mindset, and (5) is
annoying as hell.
Now, for the ease of implementation:
(1) and (3) are in the same league, they have to change the way the config
is parsed as well as make downcasing conditional in repo-config. (2) is
obviously hardest of all. (4) is very easy (one line in config.c), and (5)
easiest (nothing to do).
Now, for the versatility, i.e. what you can express with the syntax:
The same for all (except for (4) which has very weak restrictions on the
branch name).
Oh, I completely forgot about another proposal: (6) subkeys (something
like "url[branchname] = blablabla"). It has about the same effects as (1)
or (3).
Another thing: I completely ignored the case sensitivity. Because it is
irrelevant. Why? Because you do not have two branches which are only
different by case-ness. It is confusing, and that's why. And you don't
need to handle the case specially, because the comparison is done by
downcasing anyway.
Obviously, I deem (4) the best solution ATM, because it has all the
expressability needed, while being the simplest.
Ciao,
Dscho
^ permalink raw reply
* Re: [PATCH -mm] swsusp: support creating bigger images
From: Pavel Machek @ 2006-05-09 11:22 UTC (permalink / raw)
To: Rafael J. Wysocki; +Cc: Andrew Morton, linux-kernel, nickpiggin
In-Reply-To: <200605091219.17386.rjw@sisk.pl>
Hi!
> > EXTRA_PAGES is not a well-chosen identifier. Please choose something
> > within the swsusp "namespace", if there's such a thing.
>
> Unfortunately there's not any, but I'll try to invent a better
> name. :-)
PM_EXTRA_PAGES would probably be acceptable, as would
SUSPEND_EXTRA_PAGES be...
Pavel
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
^ permalink raw reply
* Re: Any reason for passing "tlb" to "free_pgtables()" by address?
From: Hugh Dickins @ 2006-05-09 11:23 UTC (permalink / raw)
To: Zoltan Menyhart; +Cc: Zoltan Menyhart, linux-mm
In-Reply-To: <445FBD1B.6080404@free.fr>
On Mon, 8 May 2006, Zoltan Menyhart wrote:
>
> Could you please explain what your plans are?
Nick already answered, and I agree with his answer;
but I get the feeling you'd like to hear it from me too.
My plan (as his) is to rework the mmu_gather TLB flush batching
so that it can be done without disabling preemption, so that it
does not add to latency.
Nick uses the pagetables themselves as buffering, I allocate a
temporary buffer: in each case we abandon the per-cpu arrays which
need preemption disabled. But neither patch is good enough yet.
> How much do you think it is worth to optimize "free_pgtables()",
> knowing that:
> - PTE, PMD and PUD pages are freed seldom (wrt. the leaf pages)
> - The number of these pages is much more less than
> that of the leaf pages.
Not worth much at all in comparison: it just falls automatically
out of the mmu_gathering rework which covers the leaf pages too.
But free_pgtables is where I got caught out inadvertently adding
latency in 2.6.15, so I've a responsibility to correct that case.
Hugh
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply
* Re: [PATCH 7/14] random: Remove SA_SAMPLE_RANDOM from network drivers
From: Pavel Machek @ 2006-05-09 11:23 UTC (permalink / raw)
To: Matt Mackall; +Cc: David S. Miller, tytso, mrmacman_g4, akpm, linux-kernel
In-Reply-To: <20060508172726.GH15445@waste.org>
On Po 08-05-06 12:27:26, Matt Mackall wrote:
> On Mon, May 08, 2006 at 05:21:12PM +0000, Pavel Machek wrote:
> > Hi!
> >
> > > > What do other platforms without a TSC do?
> > >
> > > Using get_cycles() for /dev/random is new as of 2.6. Before that, we
> > > were directly calling rdtsc on x86 alone. 10msec resolution is fine
> > > for plenty of sources.
> >
> > For what devices are timestamps still 'random/unobservable' in 10msec
> > range?
> >
> > Maybe keyboard... but no, keyboard has autorepeat and can be observed
> > remotely with 10msec accuracy in many cases. (telnet to bbs?)
>
> Please go look at the code.
Ok, so perhaps it got autorepeat right. Still you claimed "plenty" of
usable sources, and keyboard is remotely observable. Please show me
"plenty".
Pavel
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
^ permalink raw reply
* [U-Boot-Users] u-boot fails at cpu_init_f
From: Charles J Gillan @ 2006-05-09 11:24 UTC (permalink / raw)
To: u-boot
In-Reply-To: <20060509025819.33828.qmail@web60014.mail.yahoo.com>
Gordon,
The piece of code which you mention:
fff82e5c: 80 01 00 14 lwz r0,20(r1)
fff82e60: 38 21 00 10 addi r1,r1,16
is part of the "return" process from the routine back
to its caller. r1 is pointing at the C run-time stack.
The code that sets up this stack is in the assembler routine
in_flash
in file start.S for the 85xx cpus.
Usually, the 8540 is configured so that the initial runtime
stack consists of 16 KB of memory locked into L1 cache meaning
that the L1 cache is in effect providing the actual RAM.
You might need to check the configuration in init.S for your
board to see that this is being set up properly and then check that
the address range is configured into TLB0 during execution.
Charles.
-----Original Message-----
From: u-boot-users-admin@lists.sourceforge.net
[mailto:u-boot-users-admin at lists.sourceforge.net] On Behalf Of shangway
Sent: 09 May 2006 03:58
To: u-boot-users at lists.sourceforge.net
Subject: [U-Boot-Users] u-boot fails at cpu_init_f
Hi all,
I'm trying to get u-boot to run on a proprietary 8540
based board. The board is similar to the MPC8540ADS
which configuration I used to build u-boot. I am using
eldk 4.0. I managed to download u-boot to flash, but I
am unable to run it. By stepping through the code, I
found that it fails in the first c-function
"cpu_init_f".
This is the assembly code for cpu_init_f:
fff82e10 <cpu_init_f>:
fff82e10: 94 21 ff f0 stwu r1,-16(r1)
fff82e14: 3c 60 e4 01 lis r3,-7167
fff82e18: 7c 08 02 a6 mflr r0
fff82e1c: 60 63 3f 80 ori r3,r3,16256
fff82e20: 38 80 00 00 li r4,0
fff82e24: 38 a0 00 30 li r5,48
fff82e28: 90 01 00 14 stw r0,20(r1)
fff82e2c: 48 00 59 b1 bl fff887dc <memset>
fff82e30: 3c 00 ff 00 lis r0,-256
fff82e34: 3d 20 e0 00 lis r9,-8192
fff82e38: 60 00 18 01 ori r0,r0,6145
fff82e3c: 3d 60 ff 00 lis r11,-256
fff82e40: 90 09 50 00 stw r0,20480(r9)
fff82e44: 61 6b 6f f7 ori r11,r11,28663
fff82e48: 38 00 e1 f1 li r0,-7695
fff82e4c: 91 69 50 04 stw r11,20484(r9)
fff82e50: 3d 40 f8 00 lis r10,-2048
fff82e54: 90 09 50 24 stw r0,20516(r9)
fff82e58: 61 4a 08 01 ori r10,r10,2049
fff82e5c: 80 01 00 14 lwz r0,20(r1)
fff82e60: 38 21 00 10 addi r1,r1,16
fff82e64: 91 49 50 20 stw r10,20512(r9)
fff82e68: 7c 08 03 a6 mtlr r0
fff82e6c: 4e 80 00 20 blr
fff82e70: 00 02 86 0c .long 0x2860c
The code runs fine (at least that's what I think)
until 0xfff82e5c. The cpu hangs when I try to execute
that instruction. This is the first time, that there
is a load from an address which is not in flash, so I
suspect, that setting up the cache as RAM failed.
I also tried to manually write data to that location
with gdb, but I read only 0x0.
As this is still in the initialization process of the
cpu, I don't think that it is related to a possibly
wrong board configuration, or am I wrong.
Any ideas?
Thanks in advance and best regards,
Gordon.
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job
easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
U-Boot-Users mailing list
U-Boot-Users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/u-boot-users
^ permalink raw reply
* Re: [PATCH] fix can_share_swap_page() when !CONFIG_SWAP
From: Hugh Dickins @ 2006-05-09 11:25 UTC (permalink / raw)
To: Nick Piggin; +Cc: Hua Zhong, linux-kernel, akpm, linux-mm
In-Reply-To: <445FF78B.9060803@yahoo.com.au>
On Tue, 9 May 2006, Nick Piggin wrote:
> Hugh Dickins wrote:
> >
> >True; but I think Hua's patch is good as is for now, to fix
> >that inefficiency. I do agree (as you know) that there's scope for
> >cleanup there, and that that function is badly named; but I'm still
> >unprepared to embark on the cleanup, so let's just get the fix in.
>
> Sure. Queue it up for 2.6.18?
I'd be perfectly happy for Hua's one-liner to go into 2.6.17;
but that's up to Andrew.
Hugh
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply
* [ALSA - driver 0001833]: oops in snd_mpu401_uart
From: bugtrack @ 2006-05-09 11:26 UTC (permalink / raw)
To: alsa-devel
A NOTE has been added to this issue.
======================================================================
<https://bugtrack.alsa-project.org/alsa-bug/view.php?id=1833>
======================================================================
Reported By: ponchick
Assigned To:
======================================================================
Project: ALSA - driver
Issue ID: 1833
Category: GENERIC - mpu401
Reproducibility: always
Severity: major
Priority: normal
Status: new
Distribution: gentoo
Kernel Version: gentoo-sources-2.6.15-r4
======================================================================
Date Submitted: 02-09-2006 08:33 CET
Last Modified: 05-09-2006 13:26 CEST
======================================================================
Summary: oops in snd_mpu401_uart
Description:
I don't know what exactly happend but when I run wine or winecfg it
segfaulted. dmesg shows following:
Unable to handle kernel paging request at 0000000010228800 RIP:
<ffffffff884ca00c>{:snd_mpu401_uart:mpu401_write_mmio+0}
PGD 6e57067 PUD 0
Oops: 0002 [1] PREEMPT SMP
CPU 1
Modules linked in: snd_rtctimer snd_seq_midi snd_pcm_oss snd_mixer_oss
snd_seq_oss snd_seq_midi_event snd_seq snd_au8830 snd_ac97_codec snd_pcm
snd_timer snd_page_alloc snd_ac97_bus snd_mpu401_uart snd_rawmidi
snd_seq_device snd soundcore nvidia sky2
Pid: 11110, comm: wine-preloader Tainted: P 2.6.15-gentoo-r4 #1
RIP: 0010:[<ffffffff884ca00c>]
<ffffffff884ca00c>{:snd_mpu401_uart:mpu401_write_mmio+0}
RSP: 0018:ffff81003564f8f0 EFLAGS: 00010046
RAX: 0000000000000013 RBX: ffff81003cb4e580 RCX: 0000000000000000
RDX: 0000000010228800 RSI: 0000000000000000 RDI: ffff81003cb4e580
RBP: ffff81003eb60740 R08: 0000000000000000 R09: ffff8100279794c0
R10: 0000000000000000 R11: 0000000000000000 R12: ffff81003cb4fe00
R13: 0000000000000001 R14: ffff81003cb4e5f8 R15: 0000000000000246
FS: 000000005556c000(006b) GS:ffffffff8057b880(0063)
knlGS:00000000556cb920
CS: 0010 DS: 002b ES: 002b CR0: 000000008005003b
CR2: 0000000010228800 CR3: 00000000375ad000 CR4: 00000000000006e0
Process wine-preloader (pid: 11110, threadinfo ffff81003564e000, task
ffff810002178440)
Stack: ffffffff884ca19f ffff81003ffef500 0000000000000046
ff00000000000286
ffff81003cb4e580 ffff81003eb60740 ffff81003cb4fe00
ffff81003eb606c0
0000000000000000 ffff81003d7c3bd0
Call Trace:<ffffffff884ca19f>{:snd_mpu401_uart:snd_mpu401_uart_cmd+65}
<ffffffff884ca2f4>{:snd_mpu401_uart:snd_mpu401_uart_input_open+73}
<ffffffff884c36b3>{:snd_rawmidi:snd_rawmidi_kernel_open+644}
<ffffffff885471f0>{:snd_seq_midi:midisynth_subscribe+53}
<ffffffff80397b49>{_spin_lock+27}
<ffffffff8851fcc2>{:snd_seq:subscribe_port+214}
<ffffffff8851ff8f>{:snd_seq:snd_seq_port_connect+433}
<ffffffff8851a9c8>{:snd_seq:snd_seq_ioctl_subscribe_port+226}
<ffffffff8851c19a>{:snd_seq:snd_seq_kernel_client_ctl+92}
<ffffffff8852eb1b>{:snd_seq_oss:snd_seq_oss_midi_open+267}
<ffffffff8852eb83>{:snd_seq_oss:snd_seq_oss_midi_open_all+35}
<ffffffff8852b89a>{:snd_seq_oss:snd_seq_oss_open+1152}
<ffffffff8852d20c>{:snd_seq_oss:snd_seq_oss_event_input+0}
<ffffffff8852b90a>{:snd_seq_oss:free_devinfo+0}
<ffffffff8018075f>{__link_path_walk+3332}
<ffffffff80397b0c>{__lock_text_start+20}
<ffffffff8852b034>{:snd_seq_oss:odev_open+52}
<ffffffff884ac6e4>{:soundcore:soundcore_open+513}
<ffffffff8017ab11>{chrdev_open+454}
<ffffffff8017a94b>{chrdev_open+0}
<ffffffff80171031>{__dentry_open+310}
<ffffffff8017118d>{nameidata_to_filp+34}
<ffffffff801711d6>{filp_open+58}
<ffffffff801713d4>{get_unused_fd+234}
<ffffffff80171533>{do_sys_open+81}
<ffffffff8011e57f>{sysenter_do_call+27}
Code: 40 88 32 c3 0f b6 06 0f b6 c0 c3 48 83 ec 28 4c 89 64 24 10
RIP <ffffffff884ca00c>{:snd_mpu401_uart:mpu401_write_mmio+0} RSP
<ffff81003564f8f0>
CR2: 0000000010228800
<6>note: wine-preloader[11110] exited with preempt_count 1
I use standalone alsa-driver (kernel one is out of date and doesn't
support au8830 on 64-bit platforms) with gentoo-sources-2.6.14-r5,
gentoo-sources-2.6.15-r1 and gentoo-sources-2.6.15-r4.
======================================================================
----------------------------------------------------------------------
rlrevell - 05-07-06 19:23
----------------------------------------------------------------------
It means "the hardware doesn't have one". How could it be any clearer?
----------------------------------------------------------------------
Raymond - 05-09-06 13:26
----------------------------------------------------------------------
> A hacked diff is attached which shows the affected bits of
> code, and additional writes and reads in au88x0_mpu401 which demonstrate
the
> behaviour. Result is: Temp1 0xfe, Temp2 0xfe, Temp3 0xf9, Temp4 0xfe,
Temp5
> 0xf9.
au8830 (Subsystem 008812eb)
Temp1 0xfe, Temp2 0xfe, Temp3 0xd1, Temp4 0xfe, Temp5 0xd1.
aplaymidi --port=16:0 test.mid
Using alsa-driver-1.0.11, it does not hang even there is no midi keyboard
or wavetable daugther card
> I have made up a joystick port MIDI OUT lead for testing. MIDI sent to
the
> au8820 port shown in aplaymidi -l goes to both the wavetable card and
the
> joystick midi at the same time; the DB50XG is then generating sound at
the
> same time as an external keyboard/synth connected from the joystick
> connector. So it would seem the card has a single port feeding both.
http://www.3dsoundsurge.com/faq/4.3.html
Issue History
Date Modified Username Field Change
======================================================================
02-09-06 08:33 ponchick New Issue
02-09-06 08:33 ponchick Distribution => gentoo
02-09-06 08:33 ponchick Kernel Version =>
gentoo-sources-2.6.15-r4
02-09-06 09:12 Raymond Note Added: 0007971
02-09-06 09:32 Raymond Note Added: 0007973
02-09-06 11:57 ponchick Note Added: 0007975
02-09-06 12:32 Raymond Note Added: 0007976
02-09-06 12:35 Raymond Note Edited: 0007976
02-10-06 03:06 Raymond Note Added: 0007982
02-10-06 03:11 rlrevell Note Added: 0007983
02-10-06 03:49 Raymond Note Added: 0007984
02-10-06 05:08 ponchick Note Added: 0007986
02-10-06 05:36 Raymond Note Added: 0007988
02-10-06 06:03 Raymond Note Added: 0007990
02-10-06 10:09 ponchick Note Added: 0007997
02-10-06 11:30 Raymond Note Added: 0008001
02-12-06 03:32 Raymond Note Added: 0008024
02-12-06 14:14 Raymond Note Deleted: 0007990
02-13-06 07:08 Raymond Note Edited: 0008024
02-13-06 16:44 Raymond Note Edited: 0008001
02-14-06 10:41 Raymond Note Edited: 0008024
02-16-06 04:40 Raymond Note Added: 0008084
02-16-06 06:33 ponchick Note Added: 0008086
02-16-06 09:57 Raymond Note Added: 0008089
02-16-06 10:16 ponchick Note Added: 0008091
02-16-06 10:38 Raymond Note Added: 0008092
02-16-06 10:40 Raymond Note Edited: 0008092
02-16-06 13:08 Raymond Note Edited: 0008089
02-25-06 13:15 Raymond Note Added: 0008189
03-21-06 18:02 Raymond Note Edited: 0008089
04-06-06 19:13 Raymond Note Added: 0009126
04-06-06 19:23 rlrevell Note Added: 0009127
04-11-06 05:09 Raymond Note Added: 0009226
04-11-06 06:08 ponchick Note Added: 0009227
04-11-06 09:06 Raymond Note Added: 0009230
04-12-06 10:28 Raymond Note Added: 0009240
04-12-06 11:36 ponchick Note Added: 0009241
04-12-06 15:16 Raymond Note Added: 0009243
04-13-06 07:16 Raymond Note Added: 0009253
04-13-06 10:05 Raymond Note Added: 0009255
04-13-06 10:37 Raymond Note Edited: 0009253
04-14-06 03:30 Raymond Note Added: 0009279
04-17-06 22:57 gineera Note Added: 0009322
04-18-06 04:53 ponchick Note Added: 0009323
04-18-06 05:11 ponchick Note Added: 0009324
04-18-06 05:32 Raymond Note Added: 0009325
04-18-06 06:04 Raymond Note Edited: 0009325
04-19-06 05:42 Raymond Issue Monitored: alien999999999
04-19-06 05:42 Raymond Note Added: 0009349
04-20-06 04:27 Raymond Issue Monitored: penguinparty
04-20-06 04:27 Raymond Note Added: 0009364
04-20-06 05:05 Raymond Note Edited: 0009325
04-20-06 08:55 alien999999999 Note Added: 0009366
04-20-06 09:27 Raymond Note Added: 0009367
04-20-06 10:25 Raymond Note Edited: 0009367
04-21-06 08:11 alien999999999 Note Added: 0009408
04-21-06 16:41 rlrevell Note Added: 0009414
04-21-06 18:00 Raymond Note Added: 0009420
04-21-06 18:13 Raymond Note Added: 0009422
04-21-06 18:23 Raymond Note Added: 0009425
04-21-06 18:28 Raymond Note Added: 0009426
04-21-06 18:54 rlrevell Note Added: 0009427
04-22-06 02:56 Raymond Note Edited: 0009422
04-22-06 02:57 Raymond Note Edited: 0009425
04-22-06 06:58 Raymond Note Edited: 0009426
04-24-06 03:20 Raymond Note Added: 0009447
04-24-06 05:43 ponchick Note Added: 0009448
04-24-06 11:34 Raymond Note Added: 0009450
04-25-06 10:08 Raymond Note Added: 0009485
04-25-06 23:30 gineera Note Added: 0009496
04-26-06 04:26 Raymond Note Added: 0009500
04-26-06 04:33 Raymond Note Added: 0009501
04-26-06 04:35 Raymond Note Edited: 0009501
04-26-06 04:50 Raymond Note Added: 0009502
04-26-06 05:06 Raymond Note Added: 0009503
04-26-06 07:13 Raymond Note Edited: 0009501
04-26-06 16:38 Raymond Note Added: 0009515
04-26-06 17:20 Raymond Note Added: 0009518
04-26-06 17:34 Raymond Note Edited: 0009518
04-26-06 18:41 Raymond Issue Monitored: perex
04-26-06 18:41 Raymond Note Added: 0009521
04-27-06 13:36 Raymond Note Added: 0009529
05-01-06 10:58 Raymond Note Added: 0009554
05-01-06 18:12 Raymond Note Edited: 0009554
05-02-06 04:38 ponchick Note Added: 0009555
05-02-06 06:52 Raymond Issue Monitored: tiwai
05-02-06 06:52 Raymond Note Added: 0009556
05-02-06 06:54 Raymond Note Edited: 0009556
05-02-06 17:55 tiwai Note Added: 0009561
05-03-06 11:23 Raymond Note Added: 0009583
05-03-06 11:29 ponchick Note Added: 0009584
05-03-06 15:23 Raymond Issue Monitored: mjander
05-03-06 15:23 Raymond Note Added: 0009593
05-03-06 18:31 Raymond Note Added: 0009596
05-04-06 05:08 Raymond Note Edited: 0009596
05-04-06 07:11 Raymond Note Edited: 0009596
05-04-06 12:18 Raymond Note Added: 0009614
05-05-06 04:35 Raymond Note Edited: 0009529
05-05-06 07:10 ponchick Note Added: 0009623
05-05-06 15:20 Raymond Note Added: 0009625
05-06-06 04:05 ponchick Note Added: 0009635
05-06-06 05:10 Raymond Note Added: 0009636
05-06-06 05:22 Raymond Note Edited: 0009636
05-07-06 19:17 Raymond Note Added: 0009642
05-07-06 19:21 Raymond Note Edited: 0009642
05-07-06 19:23 rlrevell Note Added: 0009643
05-09-06 13:26 Raymond Note Added: 0009655
======================================================================
-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
^ permalink raw reply
* [NET] linkwatch: Handle jiffies wrap-around
From: Herbert Xu @ 2006-05-09 11:26 UTC (permalink / raw)
To: Stefan Rompf; +Cc: David S. Miller, netdev
In-Reply-To: <200605082028.57328.stefan@loplof.de>
[-- Attachment #1: Type: text/plain, Size: 1118 bytes --]
On Mon, May 08, 2006 at 06:28:56PM +0000, Stefan Rompf wrote:
>
> time_after() and friends can handle jiffies wrapping, however they require the
> difference between compared times to be less than 0x80000000 jiffies (about
> 24 days on HZ=1000) to work reliably on 32bit architectures. So if the
> network is stable for 24 days, events generated within days 25-49 will suffer
> a *huge* false delay.
How about something like this?
[NET] linkwatch: Handle jiffies wrap-around
The test used in the linkwatch does not handle wrap-arounds correctly.
Since the intention of the code is to eliminate bursts of messages we
can afford to delay things up to a second. Using that fact we can
easily handle wrap-arounds by making sure that we don't delay things
by more than one second.
This is based on diagnosis and a patch by Stefan Rompf.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Cheers,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
[-- Attachment #2: linkwatch-wrap.patch --]
[-- Type: text/plain, Size: 717 bytes --]
diff --git a/net/core/link_watch.c b/net/core/link_watch.c
index 341de44..646937c 100644
--- a/net/core/link_watch.c
+++ b/net/core/link_watch.c
@@ -170,13 +170,13 @@
spin_unlock_irqrestore(&lweventlist_lock, flags);
if (!test_and_set_bit(LW_RUNNING, &linkwatch_flags)) {
- unsigned long thisevent = jiffies;
+ unsigned long delay = linkwatch_nextevent - jiffies;
- if (thisevent >= linkwatch_nextevent) {
+ /* If we wrap around we'll delay it by at most HZ. */
+ if (!delay || delay > HZ)
schedule_work(&linkwatch_work);
- } else {
- schedule_delayed_work(&linkwatch_work, linkwatch_nextevent - thisevent);
- }
+ else
+ schedule_delayed_work(&linkwatch_work, delay);
}
}
}
^ permalink raw reply related
* Re: Implementing branch attributes in git config
From: Martin Waitz @ 2006-05-09 11:26 UTC (permalink / raw)
To: Linus Torvalds; +Cc: sean, junkio, git
In-Reply-To: <Pine.LNX.4.64.0605082007100.3718@g5.osdl.org>
[-- Attachment #1: Type: text/plain, Size: 816 bytes --]
hoi :)
On Mon, May 08, 2006 at 08:08:41PM -0700, Linus Torvalds wrote:
> > What's the advantage of section quotation marks over just allowing these
> > characters in regular section names? To be specific, what is wrong with:
> >
> > [jc/show-branch-dense]
>
> This would _suck_
>
> What if you have a branch called "core"? Not all all unlikely.
>
> Think about what a section like
>
> [core]
>
> really means.
>
> Plus I really want to not be case sensitive by default. Case sensitivity
> really is _not_ normal for this kind of config file syntax.
So why is everybody trying to munch all branch related data into
one .ini style config file?
why not simply use the mechanisms we use elsewhere and build something
like our remotes or the new HEAD file?
--
Martin Waitz
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]
^ permalink raw reply
* Re: [PATCH 1/3] zone init check and report unaligned zone boundries
From: Nick Piggin @ 2006-05-09 11:28 UTC (permalink / raw)
To: Andy Whitcroft
Cc: Dave Hansen, Bob Picco, Ingo Molnar, Martin J. Bligh, Andi Kleen,
linux-kernel, Andrew Morton, Linux Memory Management
In-Reply-To: <20060509110520.GA9634@shadowen.org>
Patchset looks nice, Andy. Page allocator stuff looks fine, just
a couple of questions about the setup and alignment stuff:
Andy Whitcroft wrote:
> zone init check and report unaligned zone boundries
>
> We have a number of strict constraints on the layout of struct
> page's for use with the buddy allocator. One of which is that zone
> boundries must occur at MAX_ORDER page boundries. Add a check for
> this during init.
>
> Signed-off-by: Andy Whitcroft <apw@shadowen.org>
> ---
> include/linux/mmzone.h | 5 +++++
> mm/page_alloc.c | 4 ++++
> 2 files changed, 9 insertions(+)
> diff -upN reference/include/linux/mmzone.h current/include/linux/mmzone.h
> --- reference/include/linux/mmzone.h
> +++ current/include/linux/mmzone.h
> @@ -388,6 +388,11 @@ static inline int is_dma(struct zone *zo
> return zone == zone->zone_pgdat->node_zones + ZONE_DMA;
> }
>
> +static inline unsigned long zone_boundry_align_pfn(unsigned long pfn)
> +{
> + return pfn & ~((1 << MAX_ORDER) - 1);
> +}
> +
> /* These two functions are used to setup the per zone pages min values */
> struct ctl_table;
> struct file;
> diff -upN reference/mm/page_alloc.c current/mm/page_alloc.c
> --- reference/mm/page_alloc.c
> +++ current/mm/page_alloc.c
> @@ -2078,6 +2078,10 @@ static void __init free_area_init_core(s
> struct zone *zone = pgdat->node_zones + j;
> unsigned long size, realsize;
>
> + if (zone_boundry_align_pfn(zone_start_pfn) != zone_start_pfn)
> + printk(KERN_CRIT "node %d zone %s missaligned "
> + "start pfn\n", nid, zone_names[j]);
> +
We also need to align the end of the zone I think? I think we should
try to force alignment by reducing the size here, or if that is not
possible, then panic?
If we don't either fix it or panic, then we're allowing users' machines
to run in an unstable state.
--
SUSE Labs, Novell Inc.
Send instant messages to your online friends http://au.messenger.yahoo.com
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply
* Re: [RFC][SECMARK 05/08] Add new packet controls to SELinux
From: Stephen Smalley @ 2006-05-09 11:28 UTC (permalink / raw)
To: James Morris; +Cc: selinux, Daniel J Walsh
In-Reply-To: <Pine.LNX.4.64.0605081947400.19850@d.namei>
On Mon, 2006-05-08 at 19:53 -0400, James Morris wrote:
> On Mon, 8 May 2006, Stephen Smalley wrote:
>
> > Should we provide alternative ways of specifying the initial value for
> > selinux_compat_net as well, as we do with checkreqprot (it can be
> > initialized via kernel config option or via boot parameter as well as
> > modified via selinuxfs)? That way a distribution can choose to build
> > with compatibility enabled by default without patching, and users can
> > boot a kernel with a particular mode of operation if desired.
>
> It doesn't seem warranted as a boot option and/or kernel config option as
> the desired behavior can be trivially selected during early boot.
I was thinking it would be helpful for distributions (example: Fedora
Core 4 and 5 re-base to 2.6.18, want the compatibility mode enabled by
default, do not wish to update/add some new userspace component to
trigger the compatibility mode). And also for users to deal with
mismatches between their kernels and userlands (example: kernel
developer using Fedora Core 5 as his development platform builds an
upstream kernel and boots it, but FC5 userland has no knowledge of
either the new checks or the compatibility mode, so it would be nice if
he could enable the compatibility mode in his kernel config or select it
trivially via boot param). But I have no strong opinion; if no one else
thinks it is necessary, that is fine.
--
Stephen Smalley
National Security Agency
--
This message was distributed to subscribers of the selinux mailing list.
If you no longer wish to subscribe, send mail to majordomo@tycho.nsa.gov with
the words "unsubscribe selinux" without quotes as the message.
^ permalink raw reply
* [PATCH] paging_enabled and non-HVM guests
From: Simon Kagstrom @ 2006-05-09 11:31 UTC (permalink / raw)
To: xen-devel list
I had a problem with the GDB-server crashing on connections in
xen_ptrace.c:map_domain_va(). paging_enabled() should only be checked
for HVM guests, and the patch adds a check for that.
Signed-off-by: Simon Kagstrom <ska@bth.se>
diff -r 4501d60d6add tools/libxc/xc_ptrace.c
--- a/tools/libxc/xc_ptrace.c Tue May 9 09:57:05 2006
+++ b/tools/libxc/xc_ptrace.c Tue May 9 13:26:14 2006
@@ -374,7 +374,7 @@
if (fetch_regs(xc_handle, cpu, NULL))
return NULL;
- if (!paging_enabled(&ctxt[cpu])) {
+ if ( (ctxt[cpu].flags & VGCF_HVM_GUEST) && !paging_enabled(&ctxt[cpu])) {
static void * v;
unsigned long page;
^ permalink raw reply
* [U-Boot-Users] AT91RM9200 Dataflash op error with U-Boot1.1.4
From: Andrey P. Vasilyev @ 2006-05-09 11:32 UTC (permalink / raw)
To: u-boot
In-Reply-To: <005a01c67359$63cf6e60$b7899d0a@first>
On Tue, May 09, 2006 at 07:11:52PM +0800, KylongMu wrote:
> Dear Groups:
> My board contains: AT91RM9200 DataFlash:AT45DB642 (without nand
> flash)
...
> U-Boot> erase c0020000 c0027fff
> Error: start address not on sector boundary
> U-Boot>
> ----------------------------------------------------------------------------
>
> Does anybody know how to solve with this?
> Thanks a lot!
Perhaps, this patch will help you:
--- u-boot-1.1.3/common/cmd_mem.c 2005-09-14 22:45:12.000000000 +0400
+++ u-boot-1.1.3-new/common/cmd_mem.c 2005-09-14 22:47:03.000000000 +0400
@@ -423,7 +423,7 @@
/* check if we are copying to Flash */
if ( (addr2info(dest) != NULL)
#ifdef CONFIG_HAS_DATAFLASH
- && (!addr_dataflash(addr))
+ && (!addr_dataflash(addr)) && (!addr_dataflash(dest))
#endif
) {
int rc;
If it will not solve the problem, try this:
--- u-boot-1.1.4/common/flash.c 2005-08-14 03:53:35.000000000 +0400
+++ u-boot-1.1.4-new/common/flash.c 2005-09-15 01:15:58.000000000 +0400
@@ -105,6 +105,8 @@
int i;
for (i=0, info=&flash_info[0]; i<CFG_MAX_FLASH_BANKS; ++i, ++info) {
+ if (info->size == 0)
+ continue;
if (info->flash_id != FLASH_UNKNOWN &&
addr >= info->start[0] &&
/* WARNING - The '- 1' is needed if the flash
--
With best regards,
Andrey Vasilyev
^ permalink raw reply
* Re: Intel HDA Mixer on Intel Mac Mini
From: Thomas Glanzmann @ 2006-05-09 11:32 UTC (permalink / raw)
To: Takashi Iwai, Michael Gernoth; +Cc: alsa-devel, Jonathan Woithe, Sam Revitch
In-Reply-To: <s5hd5f3xgqr.wl%tiwai@suse.de>
[-- Attachment #1: Type: text/plain, Size: 124 bytes --]
Hello Takashi,
another Mac Mini User has workaround for the notworking audio. I
attached the eMail.
Thanks,
Thomas
[-- Attachment #2: Type: message/rfc822, Size: 9214 bytes --]
[-- Attachment #2.1.1: Type: text/plain, Size: 1309 bytes --]
Hey again,
Regarding the audio problem, I spent some time on the issue and came
up with two changes to make to the HDA driver.
One, the patch_sigmatel module needs to support mute-via-GPIO. The
Mac Mini's internal speaker and line-out jack have two mute switches
in series, and one of them that the driver is unaware of is controlled
by a GPIO pin. There isn't a secondary mute on the line-in jack,
hence it was usable in line-out mode without driver support for the
GPIO mute.
Two, if only one usable input is detected, the patch_sigmatel module
won't switch the mux sources to it, and nothing else seems to do so
either. Easy enough to fix.
Still unresolved is the issue of accessing the internal speaker. The
driver enumerates it, but this patch doesn't fix the driver to
actually create a new PCM playback device for it, or do jack detection
and automatically switch the primary output to the internal speaker
when nothing is plugged in to the line-out.
This patch should work on an Intel Mac Mini. It might work on other
Intel Macs. It's rough, in need of more testing, and might break the
driver for other systems with STAC922X codecs.
If the patch gets filtered out by the mail list, go to:
http://lsb.blogdns.net/files/mini_hda.patch
Thanks.
-Sam Revitch
[-- Attachment #2.1.2: mini_hda.patch --]
[-- Type: text/x-patch, Size: 2747 bytes --]
--- alsa-driver-1.0.11-orig/alsa-kernel/pci/hda/patch_sigmatel.c 2006-04-10 03:31:03.000000000 -0700
+++ alsa-driver-1.0.11/alsa-kernel/pci/hda/patch_sigmatel.c 2006-05-09 02:45:46.000000000 -0700
@@ -52,6 +52,7 @@
unsigned int mic_switch: 1;
unsigned int alt_switch: 1;
unsigned int hp_detect: 1;
+ unsigned int gpio_mute: 1;
/* playback */
struct hda_multi_out multiout;
@@ -841,6 +842,19 @@
}
}
+ if (imux->num_items == 1) {
+ /*
+ * Set the current input for the muxes.
+ * The STAC9221 has two input muxes with identical source
+ * NID lists. Hopefully this won't get confused.
+ */
+ for (i = 0; i < spec->num_muxes; i++) {
+ snd_hda_codec_write(codec, spec->mux_nids[i], 0,
+ AC_VERB_SET_CONNECT_SEL,
+ imux->items[0].index);
+ }
+ }
+
return 0;
}
@@ -946,6 +960,46 @@
return 1;
}
+/*
+ * Early 2006 Intel Macintoshes with STAC9220X5 codecs seem to have a
+ * funky external mute control using GPIO pins.
+ */
+
+static void stac922x_gpio_mute(struct hda_codec *codec, int pin, int muted)
+{
+ unsigned int gpiostate, gpiomask, gpiodir;
+
+ gpiostate = snd_hda_codec_read(codec, codec->afg, 0,
+ AC_VERB_GET_GPIO_DATA, 0);
+
+ if (!muted) {
+ gpiostate |= (1 << pin);
+ } else {
+ gpiostate &= ~(1 << pin);
+ }
+
+ gpiomask = snd_hda_codec_read(codec, codec->afg, 0,
+ AC_VERB_GET_GPIO_MASK, 0);
+ gpiomask |= (1 << pin);
+
+ gpiodir = snd_hda_codec_read(codec, codec->afg, 0,
+ AC_VERB_GET_GPIO_DIRECTION, 0);
+ gpiodir |= (1 << pin);
+
+ /* AppleHDA seems to do this -- WTF is this verb?? */
+ snd_hda_codec_write(codec, codec->afg, 0, 0x7e7, 0);
+
+ snd_hda_codec_write(codec, codec->afg, 0,
+ AC_VERB_SET_GPIO_MASK, gpiomask);
+ snd_hda_codec_write(codec, codec->afg, 0,
+ AC_VERB_SET_GPIO_DIRECTION, gpiodir);
+
+ udelay(1000);
+
+ snd_hda_codec_write(codec, codec->afg, 0,
+ AC_VERB_SET_GPIO_DATA, gpiostate);
+}
+
static int stac92xx_init(struct hda_codec *codec)
{
struct sigmatel_spec *spec = codec->spec;
@@ -982,6 +1036,11 @@
stac92xx_auto_set_pinctl(codec, cfg->dig_in_pin,
AC_PINCTL_IN_EN);
+ if (spec->gpio_mute) {
+ stac922x_gpio_mute(codec, 0, 0);
+ stac922x_gpio_mute(codec, 1, 0);
+ }
+
return 0;
}
@@ -1154,6 +1213,18 @@
return err;
}
+ {
+ u16 subsystem_vendor, subsystem_device;
+ pci_read_config_word(codec->bus->pci, PCI_SUBSYSTEM_VENDOR_ID, &subsystem_vendor);
+ pci_read_config_word(codec->bus->pci, PCI_SUBSYSTEM_ID, &subsystem_device);
+
+ /* Need a better way of detecting an Intel Macintosh */
+ if ((subsystem_vendor == 0x8384) &&
+ (subsystem_device == 0x7680)) {
+ spec->gpio_mute = 1;
+ }
+ }
+
codec->patch_ops = stac92xx_patch_ops;
return 0;
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.