Hello everyone,
On Mon, 2005-03-21 at 15:27 -0800, Andrew Morton wrote:
> > On a four CPU Opteron compiling the Fusion-MPT as module gives much better
> > performance when compiling it in, here some bonnie++ results:
> >
> > Version 1.03 --Sequential Output-- --Sequential Input-
>
Hi, Dmitry.
Dmitry Torokhov wrote:
On Mon, 21 Mar 2005 18:18:46 +0900, Tejun Heo <[EMAIL PROTECTED]> wrote:
Hello, Dmitry, Mochel and James.
I've been looking at sd code and found seemingly bogus 'if (!sdkp)'
tests with /* this can happen */ comment. I've digged changelog and
found out that this
This patch updates the driver for the 3ware 5/6/7/8000 series to do
the following:
- Increase max ioctl buffer size to 512 sectors.
- Make tw_scsi_queue() return 0 for 'Unknown scsi opcode'.
- Fix tw_remove() to free irq handler/unregister_chrdev() before shutting down
the card.
- Change to new
Hi, list
Background: As we know, when the SCSI disk is in failure state, for
example bad sectors appear on disk or disk is surprise-removed, the SCSI
middle-layer will endless retry the failed I/O requests if SCSI
mid-layer can't get the notification from LLDD to stop the retry.
Unfortunately now
I'm in the process of creating a patch that contains Jeff's fixes, and the
change_queue_depth api change from James, etc. I will post it shortly.
-Adam
On Wed, 16 Mar 2005 17:21:33 -0500, James Bottomley
<[EMAIL PROTECTED]> wrote:
> On Wed, 2005-03-16 at 15:31 -0500, Jeff Garzik wrote:
> > More
Maarten de Boer <[EMAIL PROTECTED]> wrote:
>
> Hello,
>
> I am having nasty problems with an Adaptec 29160 and Promise
> Ultratrak100 TX8 external RAID. (kernel 2.6.11)
It would be helpful if you could retest 2.6.12-rc1 or, better,
2.6.12-rc1-mm1.
But afaik this remains unresolved?
> Initially,
Sorry for delay in responding.
We at LSI Logic are looking into these reports
of poor performance coming from the U320.
I will report later.
Eric Moore
LSI Logic
On Monday, March 21, 2005 4:27 PM, Andrew Morton wrote:
>
>
> Holger Kiehl <[EMAIL PROTECTED]> wrote:
> >
> > Hello
> >
> > On a fo
Hello, Jens.
Hello, James.
On Mon, Mar 21, 2005 at 05:57:52PM +0100, Jens Axboe wrote:
> On Mon, Mar 21 2005, James Bottomley wrote:
> > On Mon, 2005-03-21 at 14:26 +0100, Jens Axboe wrote:
> > > scsi_allocate_request() doesn't hold a reference to the device that it
> > > points to, that is not
Holger Kiehl <[EMAIL PROTECTED]> wrote:
>
> Hello
>
> On a four CPU Opteron compiling the Fusion-MPT as module gives much better
> performance when compiling it in, here some bonnie++ results:
>
> Version 1.03 --Sequential Output-- --Sequential Input-
> --Random-
>
Hi,
There has been a running thread for a while on several mailing lists
concerning PCI bus error recovery. Very breifly, some architectures
have PCI error recovery mechanisms built into them (e.g. IBM PowerPC,
also new PCI-Express chips from Intel (and other vendors) and possibly
pa-risc and o
> Thanks for the correction. How does this look?
>
> * Notes: Only required in "hotplug initialization model" after a
> * successful call to scsi_host_alloc(). This function does not
> * scan the bus; this can be done by calling scsi_scan_host() or
> * in some other trans
Sergey, I'm assuming this is a bug in the megaraid driver.
Can you confirm that it's still happening in 2.6.12-rc1?
"Sergey S. Kostyliov" <[EMAIL PROTECTED]> wrote:
>
> On Thursday 10 February 2005 12:49, Sergey S. Kostyliov wrote:
> > On Thursday 10 February 2005 03:36, Andrew Morton wrote:
> >
Mac NCR5380 SCSI: Fix bus error by passing the correct instance pointer to
request_irq()
Signed-off-by: Finn Thain <[EMAIL PROTECTED]>
Signed-off-by: Geert Uytterhoeven <[EMAIL PROTECTED]>
--- linux-2.6.12-rc1/drivers/scsi/mac_scsi.c2005-03-09 22:21:01.369397570
+1100
+++ linux-m68k-2.6.12-r
A patch from Adaptec with byte order changes:
- drop byte swapping on all 0's and all 1's content
- fix up missing swapping directives
- ensure swapping on 16 bit values does not use 32 bit swap
- reconcile the readl/writel auto-swapping in only most of the
configurations.
I tested this on a p
On Mon, Mar 21 2005, James Bottomley wrote:
> On Mon, 2005-03-21 at 14:26 +0100, Jens Axboe wrote:
> > scsi_allocate_request() doesn't hold a reference to the device that it
> > points to, that is not good. This patch fixes that up.
>
> Actually, I don't think this is correct. The reference is ta
On Mon, Mar 21 2005, James Bottomley wrote:
> On Mon, 2005-03-21 at 15:59 +0100, Jens Axboe wrote:
> > This is not even enough, since the queue lock is embedded in sdev
> > structure. Guys, this is a serious issue. Oopsing a kernel is trivial
> > with a hotplug device like a usb stick.
>
> Do you
On Mon, Mar 21, 2005 at 09:50:07AM -0500, [EMAIL PROTECTED] wrote:
> > + * successful call to scsi_host_alloc(). The LLD must call
> > + * scsi_scan_host() in order to scan the bus. The LLD must set up
> > + * the transport template before calling this function and may only
> > + * access th
On Mon, 2005-03-21 at 15:59 +0100, Jens Axboe wrote:
> This is not even enough, since the queue lock is embedded in sdev
> structure. Guys, this is a serious issue. Oopsing a kernel is trivial
> with a hotplug device like a usb stick.
Do you have the instructions to reproduce and a trace ... I've
On Fri, Mar 18 2005, Jens Axboe wrote:
> Hi,
>
> There is a problem with the way sdev is freed currently. The reason is
> really that there is a circular referencing problem: the sdev needs to
> hold on to the queue, but the queue (through the request function) also
> needs to hold on to the sdev.
On Mon, 2005-03-21 at 14:26 +0100, Jens Axboe wrote:
> scsi_allocate_request() doesn't hold a reference to the device that it
> points to, that is not good. This patch fixes that up.
Actually, I don't think this is correct. The reference is taken when
the command is attached to a request in the s
A nit on the use of "must"
In the following:
> /**
> - * scsi_add_host - perform sysfs registration and SCSI bus scan.
> + * scsi_add_host - perform sysfs registration and set up
> transport class
> * @shost: pointer to scsi host instance
> * @dev: pointer to struct device of type
On Mon, 21 Mar 2005 18:18:46 +0900, Tejun Heo <[EMAIL PROTECTED]> wrote:
> Hello, Dmitry, Mochel and James.
>
> I've been looking at sd code and found seemingly bogus 'if (!sdkp)'
> tests with /* this can happen */ comment. I've digged changelog and
> found out that this was to prevent oops which
scsi_mid_low_api doesn't seem to have been updated for the scsi_add_host /
scsi_scan_host split. Here's a patch.
diff -u -p -r1.9 scsi_mid_low_api.txt
--- ./Documentation/scsi/scsi_mid_low_api.txt 12 Jan 2005 20:14:58 -
1.9
+++ ./Documentation/scsi/scsi_mid_low_api.txt 21 Mar 2005
Hi,
scsi_allocate_request() doesn't hold a reference to the device that it
points to, that is not good. This patch fixes that up.
Signed-off-by: Jens Axboe <[EMAIL PROTECTED]>
= drivers/scsi/scsi.c 1.157 vs edited =
--- 1.157/drivers/scsi/scsi.c 2005-03-03 09:22:17 +01:00
+++ edited/dr
The following is v2.2 of the iscsi-initiator-core-tools. This release
contains support for iSCSI Authentication by way of initiator-authd.
This release also adds support for generating unique iSCSI Initiator
Node Names and using the same /etc/initiatorname.iscsi format as
linux-sfnet in order to e
The following is v1.6.1.23 of the iSCSI Initiator Core Stack for
2.6.12-rc1. This release adds support for iSCSI Authentication with
v2.2 of iscsi-initiator-core-tools. This release also fixes issues with
underflow/residual counts that have been reported by various iSCSi
Target vendors. If any i
On Mon, 21 Mar 2005, Jens Axboe wrote:
> On Mon, Mar 21 2005, [EMAIL PROTECTED] wrote:
> > On Mon, 21 Mar 2005, Jens Axboe wrote:
> >
> > > easily doable. I made a suggestion earlier in the thread, you need to do
> > > something ala
> > >
> > > sg_map_each_entry(sglist, entries, sg, ouput_
On Mon, Mar 21 2005, [EMAIL PROTECTED] wrote:
> On Mon, 21 Mar 2005, Jens Axboe wrote:
>
> > On Sun, Mar 20 2005, Guennadi Liakhovetski wrote:
> > >
> > > char *kmap_atomic_sg(struct scatterlist *sg, unsigned int offset, int
> > > *mapped);
> > > void kunmap_atomic_sg(struct scatterlist *sg, int
On Mon, 21 Mar 2005, Jens Axboe wrote:
> On Sun, Mar 20 2005, Guennadi Liakhovetski wrote:
> >
> > char *kmap_atomic_sg(struct scatterlist *sg, unsigned int offset, int
> > *mapped);
> > void kunmap_atomic_sg(struct scatterlist *sg, int mapped);
> >
> > The latter would just call the kunmap_atomi
Hello, Dmitry, Mochel and James.
I've been looking at sd code and found seemingly bogus 'if (!sdkp)'
tests with /* this can happen */ comment. I've digged changelog and
found out that this was to prevent oops which occurs if some driver
gets stuck inside ->probe and the machine goes down and ca
30 matches
Mail list logo
