similar to: [PATCH] fast_flush_area in blkback.c still broken after 55194bd55b86

In dispatch_rw_block_io after a call to HYPERVISOR_grant_table_op, there is the following code which calls fast_flush_area and breaks out of the loop early if one of the handles returned from HYPERVISOR_grant_table_op is negative: for (i = 0; i < nseg; i++) { if (unlikely(map[i].handle < 0)) { DPRINTK("invalid buffer -- could not remap it\n"); fast_flush_area(pending_idx,

Moving grant ref handles from blkbk to pending_req will allow us to get rid of the shared blkbk structure. Signed-off-by: Roger Pau Monné <roger.pau@citrix.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: xen-devel@lists.xen.org --- drivers/block/xen-blkback/blkback.c | 16 ++++------------ 1 files changed, 4 insertions(+), 12 deletions(-) diff --git

As usual, written and (build-)tested on 2.6.27.4 and made apply to the 2.6.18 tree without further testing. Signed-off-by: Jan Beulich <jbeulich@novell.com> Index: head-2008-11-04/drivers/xen/blktap/blktap.c =================================================================== --- head-2008-11-04.orig/drivers/xen/blktap/blktap.c 2008-10-01 16:35:04.000000000 +0200 +++

This patch implements a local buffer reference type for kernel virtual address space and a remote buffer reference type for buffers up to 16 pages based on grant tables. These types are used by the USB driver for bulk data transfer. Signed-off-by: Harry Butterworth <butterwo@uk.ibm.com> _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com

This patch picks up all the fixes to the previous 13 patches that were required to get the code working now that the last four patches are reformatted and I''m able to test the code again. These fixes will go into the correct patches if I do another revision. Aside from the fixes, this patch defines the USB interdomain protocol which consists of a number of transactions and a message

This patch implements the xenidc_channel interface using grant tables but leaves the channel bring-up and tear-down sequencing to a derived class (xenidc_gnttab_channel implements that). This code is part of the implementation of the xenidc_endpoint used for interdomain communication by the USB driver. Signed-off-by: Harry Butterworth <butterwo@uk.ibm.com>

Lots of xm-test tests are failing with console timeouts on some machines at the moment. I reproduced the problem and found it was due to some probing activity in the boot process of the -xen kernel. The behaviour of the current xm-test code is to assume that the boot process has finished after performing three one-second waits for input---when the probing activity introduces more than three

Hi all, I add a blkback QoS patch. You can config(dynamic/static) different I/O speed for different VM disk by this patch. ---------------------------------------------------------------------------- diff -urNp blkback/blkback.c blkback-qos/blkback.c --- blkback/blkback.c 2011-06-22 07:54:19.000000000 +0800 +++ blkback-qos/blkback.c 2011-06-22 07:53:18.000000000 +0800 @@ -44,6 +44,11 @@

Hi all, I add a blkback QoS patch. You can config(dynamic/static) different I/O speed for different VM disk by this patch. ---------------------------------------------------------------------------- diff -urNp blkback/blkback.c blkback-qos/blkback.c --- blkback/blkback.c 2011-06-22 07:54:19.000000000 +0800 +++ blkback-qos/blkback.c 2011-06-22 07:53:18.000000000 +0800 @@ -44,6 +44,11 @@

Hi have 4 dom0 with debian lenny running xen. When I run ps axf|grep xvd I see it''s different on diferent dom0s. dom0-A 15677 ? S< 0:00 \_ [blkback.139.xvd] 15678 ? S< 0:38 \_ [blkback.139.xvd] 17015 ? S< 0:00 \_ [blkback.140.xvd] 17016 ? S< 2:34 \_ [blkback.140.xvd] 21309 ? S< 0:00 \_ [blkback.142.xvd] 21310 ?

Hi, One of our colleagues spotted a problem in xentop. Sometimes the VBD_WSECT value suddenly becomes unreasonably high, and it turned out xentop reads /sys/bus/xen-backend/devices/vbd-(domid)-(devID)/statistics/wr_sect into an unsigned long long. That value is exposed by blkback, and among other stat counters, it''s a signed integer: drivers/block/xen-blkback/common.h struct

Patrick, there following patches fix xenpaging for me. Granttable handling is incomplete. If a page is gone, a GNTST_eagain should be returned to the caller to inidcate the hypercall has to be retried after a while, until the page is available again. Please review. Olaf _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com

Also remove unused xenbus_{,un}map_ring(), adjust types, and clean up header inclusion. Signed-off-by: Jan Beulich <jbeulich@novell.com> --- a/drivers/xen/blkback/blkback.c +++ b/drivers/xen/blkback/blkback.c @@ -41,6 +41,8 @@ #include <linux/list.h> #include <linux/delay.h> #include <xen/balloon.h> +#include <xen/evtchn.h> +#include <xen/gnttab.h>

Is exporting a block device from one DomU to another supported? ''xm block-attach'' appears to support it based on the command line options displayed but maybe it''s one of those things that doesn''t work because nobody ever uses it? I''m wondering about accessing a VSS snapshot from a windows DomU in Dom0 (or another DomU). In theory, GPLPV should be able

Hi Everyone, When you want to use ionice to limit the amount of disk a DomU has, due to have to run ionice on every blkback process? Incidently, what is the format of the blkback process? I see the following in ps aux: blkback.xx.xvda blkback.xx.xvda1 blkback.xx.xvd where xx appears to be the domain ID. I''m curious as to the last few letters mean? Thanks

Hey Roger, I tested the feature-persistent patches and they rolled out succesfully out of the test system. This is a bootup test with 32/64 dom0, 32/64 PV/PVHVM domU, with new and old OS-es. I am thinking to send a git pull on Friday to Jens so that they will be part of Linux v3.8.

I just noticed that the linux block backend does not copy the full 64 bits of the id field of a request to the associated response. The blkif.h states that this is echoed in the response. An unsigned long in make_response is used as an intermediary. This would sure be helpful to me if the full 64 bits were echoed. - Bruce Rogers _______________________________________________ Xen-devel

Hi all, I want to use ionice command change the priority of threads of blkback(in domain0) for each corresponding blkfront . However, I do not know how to recognize which is the thread for the corresponding blkfront using ps aux. Anybody can give me some advice? Thanks in advance! -- View this message in context:

Hi,all In the vbd blkback driver(linux/drivers/xen/blkback/blkback.c), when function dispatch_rw_block_io() try to do the real I/O job, it will do a sanity check on I/O request sent from DomU in the following code fragment: ... 430 for (i = 0; i < nseg; i++) { 431 uint32_t flags; 432 433 seg[i].nsec = req->seg[i].last_sect -

In case of failing due to lack of memory in the system, blkif_init potentially leaves a leak if some of the allocations was sucessfully suceeded. This patch frees the memory allocated, avoiding the leak. Signed-off-by: Glauber de Oliveira Costa <glommer@br.ibm.com> -- glommer _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com