> But in the case of an application which fits in main memory, and
> has been running for a while (so all pages are present and
> dirty), all you'd really have to do is verify the page tables are
> in the proper state and skip the TLB flush, right?
We
But in the case of an application which fits in main memory, and
has been running for a while (so all pages are present and
dirty), all you'd really have to do is verify the page tables are
in the proper state and skip the TLB flush, right?
We really
> Then there's the interrupt problem, which someone will have to solve
> before they start shipping 10gigE NICs that use 1500-byte frames, 85
> interrupts/s without mitigation. Wh
In this situations polling helps rather than interrupt driven IO. When there
is heavy IO(read more
Then there's the interrupt problem, which someone will have to solve
before they start shipping 10gigE NICs that use 1500-byte frames, 85
interrupts/s without mitigation. Wh
In this situations polling helps rather than interrupt driven IO. When there
is heavy IO(read more
>> technology is Infiniband . In Infiniband, the hardware supports
IPv6 . For
>> this type of devices there is no need for software TCP/IP. But
for
>> networking application, which mostly uses sockets, there is a
performance
>> penalty with using software TCP/IP
> For the case where the routing will be external. Thats conveniently
> something
> you can deduce in advance. In theory nothing stops you implementing this.
> Conventionally you would do that with BSD sockets by implementing a new
> socket family PF_INFINIBAND. You might then choose to make the
> Doesn't this bypass all of the network security controls? Granted
- it is
> completely reasonable in a dedicated environment, but I would
think the
> security loss would prevent it from being used for most usage.
Direct Sockets makes sense only in clustering
> Define 'direct sockets' firstly.
Direct Sockets is the ablity by which the application(using sockets)
can use the hardwares features to provide connection, flow control,
etc.,instead of the TCP and IP software module. A typical hardware
technology is Infiniband . In Infiniband,
Can you provide the dmesg output of your system after loading the driver
either successfully or unsucessfully.
- Original Message -
From: "Rafael Martinez" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, May 03, 2001 11:37 AM
Subject: Solution - AMI megaraid driver doesn't
Can you provide the dmesg output of your system after loading the driver
either successfully or unsucessfully.
- Original Message -
From: Rafael Martinez [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Sent: Thursday, May 03, 2001 11:37 AM
Subject: Solution - AMI megaraid driver doesn't work
Define 'direct sockets' firstly.
Direct Sockets is the ablity by which the application(using sockets)
can use the hardwares features to provide connection, flow control,
etc.,instead of the TCP and IP software module. A typical hardware
technology is Infiniband . In Infiniband,
Doesn't this bypass all of the network security controls? Granted
- it is
completely reasonable in a dedicated environment, but I would
think the
security loss would prevent it from being used for most usage.
Direct Sockets makes sense only in clustering
For the case where the routing will be external. Thats conveniently
something
you can deduce in advance. In theory nothing stops you implementing this.
Conventionally you would do that with BSD sockets by implementing a new
socket family PF_INFINIBAND. You might then choose to make the
technology is Infiniband . In Infiniband, the hardware supports
IPv6 . For
this type of devices there is no need for software TCP/IP. But
for
networking application, which mostly uses sockets, there is a
performance
penalty with using software TCP/IP over this
> > VMS does this. It at least used to have a great tendency to crash
> > itself, because it swapped out something that was called from a driver
> > that was called by the swapper -- resulting in deadlock. You need
> > iron discipline for this to work right in all circumstances.
>
> Actually,
VMS does this. It at least used to have a great tendency to crash
itself, because it swapped out something that was called from a driver
that was called by the swapper -- resulting in deadlock. You need
iron discipline for this to work right in all circumstances.
Actually, VMS doesn't
> It seems to me that not doing #ifdef CONFIG_HIGHMEM right now is a
> bug... I think it's the megaraid driver that wants to set dma_addr_t to
> a 64-bit mask.
MegaRAID driver:
Only if the flag __LP64__ is defined, a 64 bit mask is set , otherwise only
a 32 bit mask is used instead. However
It seems to me that not doing #ifdef CONFIG_HIGHMEM right now is a
bug... I think it's the megaraid driver that wants to set dma_addr_t to
a 64-bit mask.
MegaRAID driver:
Only if the flag __LP64__ is defined, a 64 bit mask is set , otherwise only
a 32 bit mask is used instead. However check
Hi,
I am interested in using linux driver for 3com ADSL PCI Modem Card. 3Com
does not have the linux driver for it(according to thier websites and as far
as i have searched). I wish to write the driver for it (ofcourse under GPL).
Any pointers where i could find tech specification for it?
Thanks
Hi,
I am interested in using linux driver for 3com ADSL PCI Modem Card. 3Com
does not have the linux driver for it(according to thier websites and as far
as i have searched). I wish to write the driver for it (ofcourse under GPL).
Any pointers where i could find tech specification for it?
Thanks
Enterprise customers are beginning to see the value of having
source available, and MS is doing this as a half-baked
solution to give decition makers one less reason for switching
to Open Source.
Microsoft such attempts can be viewed as either
1. Trying to make
My initial thought after seeing this article was that microsoft was testing
its waters on open sourcing. If i have 1500 licenses then i would get the
source. If i find any bug in thier source , i would report to microsoft or
send a patch and they would put it in thier next version. Is this not
Please check out this article. Looks like microsoft know open source is the
thing of the future. I would consider that it is a begining step for full
blown GPL
http://www.zdnet.com/enterprise/stories/main/0,10228,2692987,00.html
-
To unsubscribe from this list: send the line "unsubscribe
Please check out this article. Looks like microsoft know open source is the
thing of the future. I would consider that it is a begining step for full
blown GPL
http://www.zdnet.com/enterprise/stories/main/0,10228,2692987,00.html
-
To unsubscribe from this list: send the line "unsubscribe
My initial thought after seeing this article was that microsoft was testing
its waters on open sourcing. If i have 1500 licenses then i would get the
source. If i find any bug in thier source , i would report to microsoft or
send a patch and they would put it in thier next version. Is this not
Enterprise customers are beginning to see the value of having
source available, and MS is doing this as a half-baked
solution to give decition makers one less reason for switching
to Open Source.
Microsoft such attempts can be viewed as either
1. Trying to make
Hi,
I have submitted the latest 1.14g driver patch for both 2.4.1pre and
2.2.19pre series. Hopefully waiting for Linus and Alan to integrate into the
standard tree.
Thanks
Venkatesh
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL
Hi,
I have submitted the latest 1.14g driver patch for both 2.4.1pre and
2.2.19pre series. Hopefully waiting for Linus and Alan to integrate into the
standard tree.
Thanks
Venkatesh
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL
This is because the 2.2.XX contains the a later version of the
driver when compared to 2.4.1 kernel. I have submitted to Linus and Alan the
1.14g version of the driver which works for 2.2.XX and contains special code
for 2.4.XX kernels.
As the file is too large for me post to LKML
This is because the 2.2.XX contains the a later version of the
driver when compared to 2.4.1 kernel. I have submitted to Linus and Alan the
1.14g version of the driver which works for 2.2.XX and contains special code
for 2.4.XX kernels.
As the file is too large for me post to LKML
driver development is a part of kernel development. Kernel is a bigger
entity which contains scheduler, io subsystem, memory subsystem etc
Drivers comes under the IO subsystem.
- Original Message -
From: "Shyam Iyengar" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday,
driver development is a part of kernel development. Kernel is a bigger
entity which contains scheduler, io subsystem, memory subsystem etc
Drivers comes under the IO subsystem.
- Original Message -
From: "Shyam Iyengar" [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Sent: Wednesday, January
This driver replaces version 1.07b present in the standard 2.4.0 kernel. The
driver features are
1. Supports more than 4 GB Addressing (full 64 bit addressing)
2. IA32 / IA64 platform support
3. Supports 2.2.XX & 2.4.XX kernels
<>
m114b.tar.gz
This driver replaces version 1.07b present in the standard 2.4.0 kernel. The
driver features are
1. Supports more than 4 GB Addressing (full 64 bit addressing)
2. IA32 / IA64 platform support
3. Supports 2.2.XX 2.4.XX kernels
m114b.tar.gz
m114b.tar.gz
> You seem to be full of things that "we" can implement. So I just have
> to wonder: do you by any chance have some prototype code somewhere to
> figure out, reliably, which SCSI cards have BIOS extensions enabled,
> and the order they hook in?
>
[Venkat] It would be a very bad idea
are adding hardware into their systems this is
> by no means beyond their capabilities)
>
> at least with physical device nodes i can say 'computer, you will mount
> this partition on this mountpoint!' and be done with it.
[Venkatesh Ramamurthy] You are getting my point exactly. This was
my
than one scsi adapter into his
> system, let him read some documentation on how to do so. (is this even a
> documented feature? if not, i think it should be added to the docs...)
[Venkatesh Ramamurthy] Dont you think that mounting and booting
based on disk label names is better, then
art of the disk.
[Venkatesh Ramamurthy] The LILO boot loader and the LILO command
line utility should be changed for this. There are some issues when we have
different set of labels names for file system and partition table. The LILO
command line utility should make use of the disk labels of
gt; case. The hard case is what happens if someone installs on /dev/sdg. Now
>
> they boot up with a disk array turned off. Is the mid-layer going to
> figure
> out that what is now /dev/sda suppose to be /dev/sdg? Or they install to
> /dev/sdb and /dev/sda goes bad so they p
> The scsi host numbers will be allocated to the HBAs in
> the order shown starting at 0. This method does not
> distinguish between the two advansys controllers, luckily
> swapping their positions on the PCI bus does.
[Venkatesh Ramamurthy] Just think an end-user
> > Is someone handling this already?
>
> "mount by uuid"?
>
> Amiga's Rigid Disk Block?
[Venkatesh Ramamurthy] Something like this is better. The problem
is where do we store this info. Last sector is one of the options. Does
anyone know where NT stores th
> This is due to the fixed ordering of the scsi drivers. You can change the
> order of the scsi hosts with the "scsihosts" kernel parameter. See
> linux/drivers/scsi/scsi.c
[Venkatesh Ramamurthy] I think it would be a nice idea if we can
make this process automatic , w
you control the
> initialization order to your liking.
[Venkatesh Ramamurthy] I think there should be a better way to
handle this , compiling is one of the options, but an end-user should not
think of compiling. The end user needs to put an another card and connect
drives and get his system up and run
> Like the ext2 labels? (man e2label)
[Venkatesh Ramamurthy] This re-ordering of the scsi drives should
be done by SCSI ML , so is incorporating ext2 fs data structure knowledge on
the SCSI ML a good idea?.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> It should be possible to read the BIOS setting for this option and
> behave accordingly. Please give full details of how to read and interpret
> the information stored in the CMOS for all versions of AMI BIOS, and I'll
> take a look at this.
[Venkatesh Ramamurthy] When
Hi,
I have one issue which requires fix from the linux kernel.
Initially i put a SCSI controller and install the OS on the drive connected
to it. After installing the OS (on sda), the customer puts another SCSI
controller. The BIOS for the first controller has BIOS enabled and for the
second
.
[Venkatesh Ramamurthy] The LILO boot loader and the LILO command
line utility should be changed for this. There are some issues when we have
different set of labels names for file system and partition table. The LILO
command line utility should make use of the disk labels of the file system
You seem to be full of things that "we" can implement. So I just have
to wonder: do you by any chance have some prototype code somewhere to
figure out, reliably, which SCSI cards have BIOS extensions enabled,
and the order they hook in?
[Venkat] It would be a very bad idea for the
turned off. Is the mid-layer going to
figure
out that what is now /dev/sda suppose to be /dev/sdg? Or they install to
/dev/sdb and /dev/sda goes bad so they pull it out?
[Venkatesh Ramamurthy] If we can truly go for label based mouting
and lilo'ing this would so
their capabilities)
at least with physical device nodes i can say 'computer, you will mount
this partition on this mountpoint!' and be done with it.
[Venkatesh Ramamurthy] You are getting my point exactly. This was
my argument from the first, we should have a efficient mechanism to mount
It should be possible to read the BIOS setting for this option and
behave accordingly. Please give full details of how to read and interpret
the information stored in the CMOS for all versions of AMI BIOS, and I'll
take a look at this.
[Venkatesh Ramamurthy] When i meant BIOS setting
Hi,
I have one issue which requires fix from the linux kernel.
Initially i put a SCSI controller and install the OS on the drive connected
to it. After installing the OS (on sda), the customer puts another SCSI
controller. The BIOS for the first controller has BIOS enabled and for the
second
In article 1355693A51C0D211B55A00105ACCFE64E9518C@ATL_MS1 you wrote:
we need some kind of signature being written in the drive, which the
kernel
will use for determining the boot drive and later re-order drives, if
required.
Like the ext2 labels? (man e2label)
[Venkatesh
order to your liking.
[Venkatesh Ramamurthy] I think there should be a better way to
handle this , compiling is one of the options, but an end-user should not
think of compiling. The end user needs to put an another card and connect
drives and get his system up and running. He should
m, let him read some documentation on how to do so. (is this even a
documented feature? if not, i think it should be added to the docs...)
[Venkatesh Ramamurthy] Dont you think that mounting and booting
based on disk label names is better, then relying on device nodes which can
change when
Is someone handling this already?
"mount by uuid"?
Amiga's Rigid Disk Block?
[Venkatesh Ramamurthy] Something like this is better. The problem
is where do we store this info. Last sector is one of the options. Does
anyone know where NT stores this info?
-
To unsubs
This is due to the fixed ordering of the scsi drivers. You can change the
order of the scsi hosts with the "scsihosts" kernel parameter. See
linux/drivers/scsi/scsi.c
[Venkatesh Ramamurthy] I think it would be a nice idea if we can
make this process automatic , with out u
The scsi host numbers will be allocated to the HBAs in
the order shown starting at 0. This method does not
distinguish between the two advansys controllers, luckily
swapping their positions on the PCI bus does.
[Venkatesh Ramamurthy] Just think an end-user fuguring out
I had sent a beta release few hours back to kernel and scsi list, but never
made it? Is there any latency time in sending an attachment. Can the list
administrator look into this ?
Thanks
> -Original Message-
> From: Venkatesh Ramamurthy
> Sent: Wednesday, January 10, 200
I had sent a beta release few hours back to kernel and scsi list, but never
made it? Is there any latency time in sending an attachment. Can the list
administrator look into this ?
Thanks
-Original Message-
From: Venkatesh Ramamurthy
Sent: Wednesday, January 10, 2001 4:37 PM
Hi,
I am the linux developer for this card. This is the problem we have fixed
long back and the version of linux driver that comes with 2.4.0 is way too
old. A new driver will be released in a weeks time.
Thanks
Venkat
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
w. For instance, the Qlogic ISP
> SCSI
> > cards can do 64-bit addressing, as can many others. Has anybody taken a
> look
> > at enabling this ?
[Venkatesh Ramamurthy] Our AMI RAID controller and MegaRAID driver
( part of kernel) supports full blown 64 bit and we have succ
> Problem is that it needs a driver interface change and cooperation from
> the
> drivers.
[Venkatesh Ramamurthy] Atleast the spec for this new interface,
that the driver has to support be prepared? Once this is done we can port
driver by driver to this new standard.
B so never need to be buffered.
[Venkatesh Ramamurthy] If anything over 1GB is bounce buffered than
what is the purpose of setting the pci_dev->dma_mask field. On a IA32
system we set it to 32 1's and IA64 to 64 1's - Venkat
>
>
-
To unsubscribe from this list: send the line &
Problem is that it needs a driver interface change and cooperation from
the
drivers.
[Venkatesh Ramamurthy] Atleast the spec for this new interface,
that the driver has to support be prepared? Once this is done we can port
driver by driver to this new standard.
-Andi
cards can do 64-bit addressing, as can many others. Has anybody taken a
look
at enabling this ?
[Venkatesh Ramamurthy] Our AMI RAID controller and MegaRAID driver
( part of kernel) supports full blown 64 bit and we have successfully tested
with more than 4 GB RAM under IA32 (uses bounce
Hi,
I am the linux developer for this card. This is the problem we have fixed
long back and the version of linux driver that comes with 2.4.0 is way too
old. A new driver will be released in a weeks time.
Thanks
Venkat
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> Max. RAM size:64 GB (any slowness
accessing RAM over 4 GB
* with 32 bit machines ?)
Imore than 4GB in RAM is bounce buffered, so there is performance
penalty as the data have to be copied into the 4GB RAM area
-
To unsubscribe from this list:
Max. RAM size:64 GB (any slowness
accessing RAM over 4 GB
* with 32 bit machines ?)
Imore than 4GB in RAM is bounce buffered, so there is performance
penalty as the data have to be copied into the 4GB RAM area
-
To unsubscribe from this list: send
69 matches
Mail list logo