Rik van Riel <[EMAIL PROTECTED]> writes:
> On 7 Jan 2001, Zlatko Calusic wrote:
>
> > Things go berzerk if you have one big process whose working set
> > is around your physical memory size.
>
> "go berzerk" in what way? Does the system cause lots of extra
> swap IO and does it make the
Rik van Riel <[EMAIL PROTECTED]> writes:
> On 7 Jan 2001, Zlatko Calusic wrote:
>
> > OK, maybe I was too fast in concluding with that change. I'm
> > still trying to find out why is MM working bad in some
> > circumstances (see my other email to the list).
> >
> > Anyway, I would than suggest
Alan Cox wrote:
>
> > (Could this code have been written by someone who was confused between
> > MSR 0x8001 and CPUID 0x8001?)
>
> It looks like thats what happened. The docs say it has 3dnow and mmx but
> I think your diagnosis is correct
Especially since it's bit 31 in EDX. I don't
Rich Baum wrote:-
> This patch should fix the rest of the warnings about #endif
> statements when using the 20001225 gcc snapshot. Thanks to
> Keith Owens for providing a script to automate this process. It got
> the job done sooner and found warnings to fix for non x86 platforms.
Rik van Riel <[EMAIL PROTECTED]> writes:
> On 7 Jan 2001, Zlatko Calusic wrote:
>
> > The following patch cleans up some obsolete structures from the
> > mm & proc code.
> >
> > Beside that it also fixes what I think is a bug:
> >
> > if ((rw == WRITE) && atomic_read(_async_pages) >
>
Hi,
I just got this bounce message when sending my response
to [EMAIL PROTECTED] his question. John, could you set
your spam filters to something more reasonable and make
sure you don't include whole countries ???
[or ... why are you asking questions here if you don't
want an answer?]
(my
for what it's worth:
this afternoon i conducted an experiment: i copied everything that
was newer from
[xfree-4.02-sourcedir]/xc/programs/Xserver/hw/xfree86/os-support/linux/drm/kernel
to my [linux-2.4.0-sourcedir]/drivers/char/drm and then built a
monolithic kernel containing agpgart, dri,
On Sun, 7 Jan 2001, John O'Donnell wrote:
> What does this message mean in my dmesg output?
>
> __alloc_pages: 3-order allocation failed.
It means something in the kernel is trying to allocate an
area of 8 physically contiguous pages, but that wasn't
available so the allocation failed...
This
On Fri, Jan 05, 2001 at 12:34:11AM +0200, Eric Lammerts wrote:
> I had the same problem with my 80Gb Maxtor. (Asus P2L97, works with
> 60Gb but hangs with 80Gb :-/) After clipping the drive with ibmsetmax
> (http://www.uwsg.indiana.edu/hypermail/linux/kernel/0012.1/0249.html)
> and removing the
What does this message mean in my dmesg output?
__alloc_pages: 3-order allocation failed.
__alloc_pages: 3-order allocation failed.
__alloc_pages: 3-order allocation failed.
__alloc_pages: 3-order allocation failed.
__alloc_pages: 3-order allocation failed.
reset_xmit_timer sk=c5b3a680 1
> (Could this code have been written by someone who was confused between
> MSR 0x8001 and CPUID 0x8001?)
It looks like thats what happened. The docs say it has 3dnow and mmx but
I think your diagnosis is correct
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel"
Followup to: <[EMAIL PROTECTED]>
By author:Alan Cox <[EMAIL PROTECTED]>
In newsgroup: linux.dev.kernel
>
> >3DNOW extensions for Cyrix III via rdmsr from 0x8001. This
> >fails with an exception, that is not handled and thus we oops
> >on boot.
>
> Interesting. Ok. We can
On 7 Jan 2001, Zlatko Calusic wrote:
> Things go berzerk if you have one big process whose working set
> is around your physical memory size.
"go berzerk" in what way? Does the system cause lots of extra
swap IO and does it make the system thrash where 2.2 didn't
even touch the disk ?
> Final
I get, with XFree86 4.0.1 and an ATI Rage Millenium card:
> (EE) r128(0): R128DRIScreenInit failed (DRM version = 2.1.2, expected 1.0.x).
> Disabling DRI.
Jeff Hartmann ([EMAIL PROTECTED]) says:
> XFree 4.0.2 will fix this
OK, so I'll give a try at building 4.0.2 the Slackware way. While
On 7 Jan 2001, Zlatko Calusic wrote:
> Anyway, I would than suggest to introduce another /proc entry and call
> it appropriately: max_async_pages. Because that is what we care about,
> anyway. I'll send another patch.
Anton Blanchard already did a patch for this. Sent to the list
on Thu, 7 Dec
This patch should fix the rest of the warnings about #endif
statements when using the 20001225 gcc snapshot. Thanks to
Keith Owens for providing a script to automate this process. It got
the job done sooner and found warnings to fix for non x86 platforms.
Rich
diff -urN -X dontdiff
> Sounds like a job for ... ... tmpfs!!
>
> (and yes, I share your opinion that ramfs is nice _because_
> it's an easy example for filesystem code teaching)
The resource tracking ramfs isnt that much uglier to be honest. One that went
off using backing store would be, but ramfs with limits
On 7 Jan 2001, Zlatko Calusic wrote:
> OK, maybe I was too fast in concluding with that change. I'm
> still trying to find out why is MM working bad in some
> circumstances (see my other email to the list).
>
> Anyway, I would than suggest to introduce another /proc entry
> and call it
On 7 Jan 2001, Zlatko Calusic wrote:
> The following patch cleans up some obsolete structures from the
> mm & proc code.
>
> Beside that it also fixes what I think is a bug:
>
> if ((rw == WRITE) && atomic_read(_async_pages) >
>pager_daemon.swap_cluster * (1 <<
Nate Straz of the Linux Test Project at SGI ([EMAIL PROTECTED]) said:
> The Linux Test Project (http://oss.sgi.com/projects/ltp/) was set up to
> create a set of automated tests for Linux.
Nate,
This is most excellent news!
I'd like you to look at http://linuxquality.sunsite.dk/
The
On Sun, 7 Jan 2001, Linus Torvalds wrote:
> On Sun, 7 Jan 2001, Alan Cox wrote:
>
> > -ac has the rather extended ramfs with resource limits and stuff. That one
> > also has rather more extended bugs 8). AFAIK none of those are in the vanilla
> > ramfs code
> This is actually where I agree with
Marcelo Tosatti <[EMAIL PROTECTED]> writes:
> On 7 Jan 2001, Zlatko Calusic wrote:
>
> > The following patch cleans up some obsolete structures from the mm &
> > proc code.
> >
> > Beside that it also fixes what I think is a bug:
> >
> > if ((rw == WRITE) && atomic_read(_async_pages)
On Sun, Jan 07, 2001 at 01:55:15PM +, Alan Cox wrote:
> > + return -EPERM;
>
> To stop a case where the fs gets corrupted otherwise. You can change that to
> return 0 which is more correct but most not remove it.
While I suppose "0" is covered under "the result is
Hi,
I have an old ISA-PNP soundcard whose driver doesn't currently use the
ISA-PNP services. I have therefore been activating this card by
writing to the /proc/isapnp interface. And in an idle moment, I tried
an "auto" instruction:
cat > /proc/isapnp << EOF
card 0 ENS3081
dev 0 ENS
auto
EOF
On 7 Jan 2001, Zlatko Calusic wrote:
> The following patch cleans up some obsolete structures from the mm &
> proc code.
>
> Beside that it also fixes what I think is a bug:
>
> if ((rw == WRITE) && atomic_read(_async_pages) >
>pager_daemon.swap_cluster * (1
I'm trying to get more familiar with the MM code in 2.4.0, as can be
seen from lots of questions I have on the subject. I discovered nasty
mm behaviour under even moderate load (2.2 didn't have troubles).
Things go berzerk if you have one big process whose working set is
around your physical
> Fsck discovered an error it wasn't able to fix. This error never
> appeared before and my Seagate HD actually should be alright.
Umm the error says not
> hda: dma_intr: status=0x51 { DriveReady SeekCompleteError }
> hda: dma_intr: error=0x40 { UncorrectableError } LBAsect = 2421754, sector
>
On Mon, 8 Jan 2001, Chris Wedgwood wrote:
> I would like to determine the banwidth the card is getting from
> the network.
> /proc/net/dev exports counters; you can monitor those -- I'm sure
> there are perfomance program that do exactly this.
I have this little script for monitoring
Hi,
recently I was on the internet with kernel 2.4.0-prerelease.
Suddenly Netscape hung and I couldn't help hard rebooting.
Fsck discovered an error it wasn't able to fix. This error never
appeared before and my Seagate HD actually should be alright.
The following error message appears
The following patch cleans up some obsolete structures from the mm &
proc code.
Beside that it also fixes what I think is a bug:
if ((rw == WRITE) && atomic_read(_async_pages) >
pager_daemon.swap_cluster * (1 << page_cluster))
In that (swapout logic) it
Alan Cox wrote:
> -ac has the rather extended ramfs with resource limits and stuff. That one
> also has rather more extended bugs 8). AFAIK none of those are in the vanilla
> ramfs code
Nifty stuff, too; it's nice for a ramfs mount to show up in 'df' with
useful figures. Shame I can't put
> Could XFree86 4.0.2 fix this? I had been waiting until the binary packages were
> available from ftp.slackware.com because Patrick Volkerding lays out the
> directories in a slightly different manner that he argues pretty convincingly is
> preferable, but it would be a drag for me to
I've noticed that my Linux boxes take quite a hit in terms of
packets per second rate when I define ipchains rules with
2.2.X kernels. Does the netfilter replacement found in 2.4
kernels improve this performance?
11101101 (Ed)
-
To unsubscribe from this list: send the line "unsubscribe
On Sat, Jan 06, 2001 at 11:21:31PM +, Michael D. Crawford wrote:
> Can you tell me about any ready-to-use test suites, for any software
> package that should run under Linux, that I can build and run to test
> the new kernel?
The Linux Test Project (http://oss.sgi.com/projects/ltp/) was set
On Sun, 7 Jan 2001, Alan Cox wrote:
> > > I'll take a look at the ramfs one. I may have broken something else when fixing
> > > everything else with ramfs (like unlink) crashing
> >
> > Ehh.. Plain 2.4.0 ramfs is fine, assuming you add a "UnlockPage()" to
> > ramfs_writepage(). So what do you
> > I'll take a look at the ramfs one. I may have broken something else when fixing
> > everything else with ramfs (like unlink) crashing
>
> Ehh.. Plain 2.4.0 ramfs is fine, assuming you add a "UnlockPage()" to
> ramfs_writepage(). So what do you mean by "fixing everything else"?
-ac has the
Sandy Harris wrote:
>
> jamal wrote:
>
> > > What problem does this fix?
> > >
> > > If you are mucking with the ifindex, you may be affecting many places
> > > in the rest of the kernel, as well as user-space programs which use
> > > ifindex to bind to raw devices.
> >
> > I am talking about
On Sat, 6 Jan 2001, Thiago Rondon wrote:
>
> At 2.4 too, but the status of file is o+r. Do see any
> problem about this?
>
> -Thiago Rondon
>
Yes. /proc//environ is now unreadable by the owner; similarly for
/proc//fd/ . It makes debugging harder.
It is also a major change in a supposedly
On Sun, 7 Jan 2001, Alan Cox wrote:
> > >ramfs croaks with 'kernel BUG in filemap.c line 2559' anytime I make a
> > >file in ac2 and ac3. Works fine in 2.4.0 vanilla. Should be quite
> > >repeatable...
>
> I'll take a look at the ramfs one. I may have broken something else when fixing
>
On Sun, Jan 07, 2001 at 02:10:52PM -0500, jamal wrote:
> On Sun, 7 Jan 2001, Matti Aarnio wrote:
> > Read what I wrote about the issue to Alan.
> > Ben's code has no problems with receiving VLANs with network
> > cards which have "hardware support" for VLANs.
>
> OK. I suppose an
jamal wrote:
> > What problem does this fix?
> >
> > If you are mucking with the ifindex, you may be affecting many places
> > in the rest of the kernel, as well as user-space programs which use
> > ifindex to bind to raw devices.
>
> I am talking about 2.5 possibilities now that 2.4 is out. I
On Sun, 7 Jan 2001, Matti Aarnio wrote:
> Read what I wrote about the issue to Alan.
> Ben's code has no problems with receiving VLANs with network
> cards which have "hardware support" for VLANs.
>
OK. I suppose an skb->vlan_tag is passed to the driver and it will know
what
On Sun, 7 Jan 2001, Gleb Natapov wrote:
> And what about bonding device? What major number should they use?
Would that include several ifindeces?
use standards. 802.3ad(?). Didnt Intel release some code on this or
are they still playing the big bad corporation? Normaly standards will
take
On Sun, Jan 07, 2001 at 01:21:11PM -0500, jamal wrote:
> On Sun, 7 Jan 2001, Ben Greear wrote:
> > > Question: How do devices with hardware vlan support fit into your model ?
> > I don't know of any, and I'm not sure how they would be supported.
>
> erm, this is a MUST. You MUST factor the
On Sun, 7 Jan 2001, Ben Greear wrote:
> jamal wrote:
> >
> > erm, this is a MUST. You MUST factor the hardware VLANs and be totaly
> > 802.1q compliant. Also of interest is 802.1P and D. We must have full
> > compliance, not some toy emulation.
>
> I have seen neither hardware nor spec sheets
On Sun, Jan 07, 2001 at 06:06:37PM +, Alan Cox wrote:
> > Um, what about people running their box as just a VLAN router/firewall?
> > That seems to be one of the principle uses so far. Actually, in that case
> > both VLAN and IP traffic would come through, so it would be a tie if VLAN
> >
On Sun, Jan 07, 2001 at 01:29:51PM -0500, jamal wrote:
>
>
> On Sun, 7 Jan 2001, Ben Greear wrote:
>
> > > My thought was to have the vlan be attached on the interface ifa list and
> > > just give it a different label since it is a "virtual interface" on top
> > > of the "physical interface".
jamal wrote:
> So instead of depending what ifconfig does, maybe a better test for Ben is
> to measure the kernel level improvement in the lookup for example from
> 2..6000 devices.
In the benchmark I gave, the performance increase was in the kernel,
not user space, and it was more than 10
Hi all,
With the release of the 2.4.0 kernel, there was a bit of confusion over
a last minute patch that made it in, which effected the modutils program
and the USB drivers hotplug ability. Realizing that this problem
shouldn't have happened with better communication between developers of
> > according to sportster.c:get_io_range, this appears to be perfectly
> > intentional, request_regioning 64x8 byte from 0x268 in 1024byte-steps.
>
> AFAIK, this is because the hardware is stupid and does decode the higher
> address lines. Therefore, the IO ports are mirrored every 1024 bytes
On Sat, 6 Jan 2001, Daniel Stodden wrote:
> --- linux-2.4/drivers/isdn/hisax/Makefile.orig Sat Jan 6 02:47:31 2001
> +++ linux-2.4/drivers/isdn/hisax/Makefile Sat Jan 6 02:21:22 2001
> @@ -34,7 +34,7 @@
> hisax-objs-$(CONFIG_HISAX_ASUSCOM) += asuscom.o isac.o arcofi.o hscx.o
>
jamal wrote:
>
> On Sun, 7 Jan 2001, Ben Greear wrote:
>
> > > Question: How do devices with hardware vlan support fit into your model ?
> >
> > I don't know of any, and I'm not sure how they would be supported.
> >
>
> erm, this is a MUST. You MUST factor the hardware VLANs and be totaly
>
On Sun, 7 Jan 2001, Ben Greear wrote:
> > My thought was to have the vlan be attached on the interface ifa list and
> > just give it a different label since it is a "virtual interface" on top
> > of the "physical interface". Now that you mention the SNMP requirement,
> > maybe an idea of
> Thats what it already does, if I understand correctly. Of course, if VLAN
> is loaded as a module, then it will be in the hash before IP, right?
Thats fine. I think it'll be a different hash bucket anyway. The point of
having vlan first is that if its not registered or the interface isnt
>3DNOW extensions for Cyrix III via rdmsr from 0x8001. This
>fails with an exception, that is not handled and thus we oops
>on boot.
Interesting. Ok. We can set the bit unconditionally it seems.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the
>
>> [root@eagle linux]# ld -v
>> GNU ld version 2.10.1 (with BFD 2.10.1)
>
>Historically kernel is built with hjl's binutils - try 2.10.1.0.4
> ftp://ftp.valinux.com/pub/support/hjl/binutils
THat got it...though the stock 2.4.0 won't find my IDE drives to boot
from...using Alan Cox's ac3 seems
Alan Cox wrote:
>
> > Um, what about people running their box as just a VLAN router/firewall?
> > That seems to be one of the principle uses so far. Actually, in that case
> > both VLAN and IP traffic would come through, so it would be a tie if VLAN
> > came first, but non-vlan traffic would
Hi,
I have found something weird in kernel 2.2.17.
After installation on the Pentium PRO equipped machine,
I have moved hard disk to another one, but equipped
with AMD-K5 and after encountering problems I moved again
this disk to machine equipped with Intel Pentium MMX.
On all machines except
On Sun, 7 Jan 2001, Ben Greear wrote:
> > Question: How do devices with hardware vlan support fit into your model ?
>
> I don't know of any, and I'm not sure how they would be supported.
>
erm, this is a MUST. You MUST factor the hardware VLANs and be totaly
802.1q compliant. Also of interest
Hi,
I reported the crash on boot with a Winchip (which was actually
an Cyrix III) since test12-pre8.
I couldn't access the machine and debug the problem until now.
[1.] One line summary of the problem:
Cyrix III doesn't boot, because of illegal rdmsr to 8001
[2.] Full description
jamal wrote:
> A very good reason why you would want them to have separate ifindices.
> Essentially, vlans have to be separate interfaces today. Other "virtual"
> interfaces such as aliased devices are not going to work with route
> daemons today since they dont meet this requirement.
>
> Not
On Sun, 7 Jan 2001, Rik van Riel wrote:
> The patch is available at this URL:
>
> http://www.surriel.com/patches/2.4/2.4.0-tunevm+rss
I have one improvement on top of your patch.
Now its not more "rare" (as the comment on the code stated) to have
pages with page->age == 0 being called
Alan Cox wrote:
>
> > Suppose I bind a raw socket to device vlan4001 (ie I have 4k in the list
> > before that one!!). Currently, that means a linear search on all devices,
> > right? In that extreme example, I would expect the hash to be very
> > useful.
>
> At this point you have to ask
> Um, what about people running their box as just a VLAN router/firewall?
> That seems to be one of the principle uses so far. Actually, in that case
> both VLAN and IP traffic would come through, so it would be a tie if VLAN
> came first, but non-vlan traffic would suffer worse.
Why would
this looks like a typo and fixes a compile error in
2.4.0-ac3.
--- drivers/video/vesafb.c.old Sun Jan 7 12:18:13 2001
+++ drivers/video/vesafb.c Sun Jan 7 12:18:23 2001
@@ -637,7 +637,7 @@
int temp_size = video_size;
/* Find the largest power-of-two */
On Sun, 7 Jan 2001, Gleb Natapov wrote:
> > One could have the route daemon take charge of management of these
> > devices, a master device like "eth0" and a attached device like "vlan0".
> > They both share the same ifindex but different have labels.
> > Basically, i dont think there would be
Alan Cox wrote:
>
> > + * NOTE: That is no longer true with the addition of VLAN tags. Not
> > + * sure which should go first, but I bet it won't make much
> > + * difference if we are running VLANs. The good news is that
>
> It makes a lot of difference tha the
> I don't have a problem with the rtc driver delaying 500ms
I still haven't looked at things, but two points:
(i) is the behaviour constant on all architectures?
(ii) instead of waiting, isn't it much easier to redefine
what it means to access rtc?
(If you read a certain value then on average
Em Sun, Jan 07, 2001 at 01:37:15PM +, Alan Cox escreveu:
> > kmalloc and the comment: the buffer is used for DMA but the kmalloc doesn't
> > has GFP_DMA, maybe I'm missing something here, its about time for me to
>
> It should be kmalloc (or 2.4 wise pci_alloc_* I guess eventually). Its
Alan Olsen said once upon a time (Sat, 6 Jan 2001):
> On Sat, 6 Jan 2001, Michael D. Crawford wrote:
>
> > AGP, VIA support, DRM, and r128 DRM are all compiled in statically rather than
> > as modules.
>
> AGPGART doe *not* work if compiled statically. Compile it as a module.
> You will be much
Le dim, 07 jan 2001 18:37:11, Greg KH a écrit :
> On Sun, Jan 07, 2001 at 02:04:40PM +0100, Nicolas Mailhot wrote:
> > Here is a patchlet to stop people searching for the
> > mysteriously hidden USB Mass Storage driver (in case they
> > didn't make the connection with SCSI at once like me).
>
I think atomic lock is needed for SMP code.
diff -ur linux-2.2.19pre6/drivers/block/ll_rw_blk.c
linux-2.2.19pre6+lockfix/drivers/block/ll_rw_blk.c
--- linux-2.2.19pre6/drivers/block/ll_rw_blk.c Mon Dec 11 09:49:41 2000
+++ linux-2.2.19pre6+lockfix/drivers/block/ll_rw_blk.c Mon Jan 8
On Sun, Jan 07, 2001 at 11:56:26AM -0500, jamal wrote:
>
>
> On Sun, 7 Jan 2001, Chris Wedgwood wrote:
>
> > That said, if this was done -- how would things like routing daemons
> > and bind cope?
>
> I dont know of any routing daemons that are taking advantage of the
> alias interfaces
Hi,
I posted a patch for the 2.4.0 VM subsystem today which
includes the following things:
- implement RSS ulimit enforcement
- make the page aging strategy sysctl tunable
(no aging, exponential decay, linear decay)
- don't use the page age in try_to_swap_out(), since that
function
On Sun, Jan 07, 2001 at 02:04:40PM +0100, Nicolas Mailhot wrote:
> [Re-send without the mime stuff, sorry about this one]
>
> Hi,
>
> Here is a patchlet to stop people searching for the
> mysteriously hidden USB Mass Storage driver (in case they
> didn't make the connection with SCSI at once
For some reason whenever i boot linux (i'm using
debian), it assigns IRQ 9 to the network card, the video card AND the keyboard.
whenever i startx the keyboard and mouse completely lock up. I don't have this
problem in windows. I have tried moving the cards around, to no avail. Any
> Hmm... How should we respond to that sort of thing? In principle the
> NFS layer supposes that if we have a hard mount, then unreachable
> ports etc are a temporary problem, and we should wait them out. (In
> fact, I've made an RPC 'ping' routine that improves on that behaviour
> but which
On Sun, Jan 07, 2001 at 04:46:14PM +, Alan Cox wrote:
> But talking between two vlans on the same physical lan you will go in and back
> out via the switch and you wont
So ?
If your box is routing in between VLANs, you are using it wrong way, IMO.
On the other hand, I could very well
> " " == Alan Cox <[EMAIL PROTECTED]> writes:
>> > This is caused by 2.3/2.4 changes in the network code error
>> > reporting of unreachables with UDP I suspect. It looks like
>> > the NFS code hasn't yet caught up with the error notification
>> > stuff
>>
>> No. It
Sure enough, when I changed the processor type from
Pentium-Pro/Celeron/Pentium-II
to
Pentium-III
(which is the type of processor in the machine) it works.
At 08:43 AM 01/07/2001 -0800, Scott Laird wrote:
>It works if you compile the kernel with the processor type set to Pentium
>II or higher,
On Sun, 7 Jan 2001, Richard B. Johnson wrote:
> On Sun, 7 Jan 2001, Mike Galbraith wrote:
>
> > Seriously though, the constraints look fine to me (and the register
> > name is there in the output constraint). I'd say you have a busted
> > compiler. None of the named compilers gripe.
> >
>
>
On Sun, 7 Jan 2001, Chris Wedgwood wrote:
> That said, if this was done -- how would things like routing daemons
> and bind cope?
I dont know of any routing daemons that are taking advantage of the
alias interfaces today. This being said, i think that the fact that a
lot of protocols that
On Sun, 7 Jan 2001, Jim Olsen wrote:
> Hi... I have a question or two that would help me clear up a bit of the fuzz
> I have relating to the VM: do_try_to_free_pages issue.
> My question is, exactly which kernel should I use in order to
> rid my server of this VM issue?
2.2: 2.2.19pre2 and
On Sun, 7 Jan 2001, david wrote:
> can i rewrite the mm system in kernel's 2.2.18 to add new and
> needed functions or may be it can be a compile option old (mm
> system or new mm system) ?
Upgrade to 2.4.0 ;)
But yes, you can rewrite 2.2.18 VM all you want, that's
what the GPL is for...
> Ok. Good point.
> But remember that parsing /proc for an embedded system is also not the
> most healthy thing.
I dont compile in /proc either. SIOCGIFCONF is enough for an embedded box.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Alan Cox wrote:
>
> > +
> > + /* FAT cannot truncate to a longer file */
> > + if (attr->ia_valid & ATTR_SIZE) {
> > + if (attr->ia_size > inode->i_size)
> > + return -EPERM;
> > + }
> >
> > error = inode_change_ok(inode, attr);
> >
2.2.18, SMP.
It seems that anything I write in /proc/sys/vm/bdflush does not make
any difference. Metadata is always flushed every 5 seconds.
Bye.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
Please read the FAQ at
> I just tried to pull data from another machine, which
> is on normal port thru VLAN trunking port to receiving
> machine, and got fast-ether at wire speed. (As near as
> ncftp's 11.11 MB/sec is wirespeed..)
But talking between two vlans on the same physical lan you will
In article <[EMAIL PROTECTED]>,
Alan Cox <[EMAIL PROTECTED]> wrote:
>At this point you have to ask 'why is vlan4001 an interface'. Would it not
>be cleaner to add the vlan id to the entries in the list of addresses per
>interface ?
Not all the world is IP - what if you want to bridge between
It works if you compile the kernel with the processor type set to Pentium
II or higher, or disable RAID5. I've been meaning to report this one, but
2.4.0 was released before I had time to test the last prerelease, and I
haven't had time to test the final release yet.
Scott
On Sun, 7 Jan
david wrote:
>
> hi all
>
> i now need to read a file from in the kernel 2.2.x
> dose any one know how to this ?
Look at open_exec and kernel_read, but also consider whether you could
solve your problem more elegantly with help from user space. (This
should be in the FAQ.)
--
Daniel
-
To
On 6 Jan 2001, Linus Torvalds wrote:
> In short, releasing 2.4.0 does not open up the floor to just
> about anything. In fact, to some degree it will probably make
> patches _less_ likely to be accepted than before, at least for a
> while.
I think this is an excellent idea. To help with this
On Sun, 7 Jan 2001, Gleb Natapov wrote:
> > I used to be against VLANS being devices, i am withdrawing that comment; it's
> > a lot easier to look on them as devices if you want to run IP on them. And
> > in this case, it makes sense the possibilirt of over a thousand devices
> > is good.
> >
On Sun, Jan 07, 2001 at 10:56:23AM -0500, jamal wrote:
[snip]
>
> I used to be against VLANS being devices, i am withdrawing that comment; it's
> a lot easier to look on them as devices if you want to run IP on them. And
> in this case, it makes sense the possibilirt of over a thousand devices
I have been trying out the new 2.4.0 kernel and am unable to get
raid5 to work. When I install the raid5 module with
modprobe raid5
I get a segmentation fault and the following error appears in the dmesg output:
raid5: measuring checksumming speed
8regs : 806.577 MB/sec
32regs:
Alan Cox <[EMAIL PROTECTED]> writes:
> > Putting the LFS checks, & max filesize checks into the VFS sounds
> > right for 2.4.x because it fixes lots of filesystems, with just a
> > couple of lines of code.
>
> Rather more than that, and it only fixes those using generic_file_*
True. But it
On Sun, 7 Jan 2001, Alan Cox wrote:
> Why. Its bad enough that the networking layer doesnt let you configure out
> stuff like SACK and the big routing hashes. Please don't make it even worse
> for the embedded world. 99.9% of Linux boxes probably have less than 5 routing
> table entries
Ok.
On Sat, 6 Jan 2001, David S. Miller wrote:
> I could not agree more. This reminds me to do something I could not
> justify before, making netlink be enabled in the kernel and
> non-configurable.
I always use netlink and friends for something or the other. Route
protocols, traffic control
On 7 Jan 2001, Eric W. Biederman wrote:
> Rik van Riel <[EMAIL PROTECTED]> writes:
>
> > The following bugs _could_ be fixed ... I'm not 100% certain
> > but they're probably gone (could somebody confirm/deny?):
> >
> > * mm->rss is modified in some places without holding the
> >
On Sun, Jan 07, 2001 at 01:42:50PM +, Alan Cox wrote:
> > + * NOTE: That is no longer true with the addition of VLAN tags. Not
> > + * sure which should go first, but I bet it won't make much
> > + * difference if we are running VLANs. The good news is that
>
101 - 200 of 536 matches
Mail list logo
