On Apr 16 09:22, Gollu Appalanaidu wrote:
Use lower case hexadecimal format for the constants and in
comments use the same format as used in Spec. ("h")
Signed-off-by: Gollu Appalanaidu
---
-v3: Add Suggestions (Philippe)
Describe the NVMe subsystem style in nvme.c header
-v2: Address
A customer reported that running
qemu-img convert -t none -O qcow2 -f qcow2 input.qcow2 output.qcow2
fails for them with the following error message when the images are
stored on a GPFS file system:
qemu-img: error while writing sector 0: Invalid argument
After analyzing the strace output, it
Use lower case hexadecimal format for the constants and in
comments use the same format as used in Spec. ("h")
Signed-off-by: Gollu Appalanaidu
---
-v3: Add Suggestions (Philippe)
Describe the NVMe subsystem style in nvme.c header
-v2: Address review comments (Klaus)
use lower case hex
On Thu, Apr 15, 2021 at 07:18:50PM +0200, Klaus Jensen wrote:
On Apr 15 15:13, Philippe Mathieu-Daudé wrote:
On 4/15/21 2:00 PM, Gollu Appalanaidu wrote:
Make uniform hexadecimal numbers format.
Signed-off-by: Gollu Appalanaidu
---
-v2: Address review comments (Klaus)
use lower case hexa form
On 4/15/21 8:58 AM, Daniel P. Berrangé wrote:
I spent a while debugging a tricky migration failure today which was
ultimately caused by fdatasync() getting EACCESS. The existing probes
were not sufficient to diagnose this, so I had to resort to GDB. This
improves probes and block error reporting
Hi all!
Recently I've implemented fast-cancelling of mirror job: do
bdrv_cancel_in_flight() in mirror_cancel().
Now I'm in doubt: is it a correct thing? I heard, that mirror-cancel is a kind
of valid mirror completion..
Looking at documentation:
# Note that if you issue 'block-job-cancel' af
On Apr 15 15:13, Philippe Mathieu-Daudé wrote:
On 4/15/21 2:00 PM, Gollu Appalanaidu wrote:
Make uniform hexadecimal numbers format.
Signed-off-by: Gollu Appalanaidu
---
-v2: Address review comments (Klaus)
use lower case hexa format for the code and in comments
use the same format as used in
* Daniel P. Berrangé (berra...@redhat.com) wrote:
> A flush failure is a critical failure scenario for some operations.
> For example, it will prevent migration from completing, as it will
> make vm_stop() report an error. Thus it is important to have a
> trace point present for debugging.
>
> Sig
* Daniel P. Berrangé (berra...@redhat.com) wrote:
> Signed-off-by: Daniel P. Berrangé
Reviewed-by: Dr. David Alan Gilbert
> ---
> block/file-posix.c | 2 --
> 1 file changed, 2 deletions(-)
>
> diff --git a/block/file-posix.c b/block/file-posix.c
> index 6aafeda44f..2538e43299 100644
> --- a/
* Daniel P. Berrangé (berra...@redhat.com) wrote:
> This is a critical failure scenario for migration that is hard to
> diagnose from existing probes. Most likely it is caused by an error
> from bdrv_flush(), but we're not logging the errno anywhere, hence
> this new probe.
>
> Signed-off-by: Dani
* Daniel P. Berrangé (berra...@redhat.com) wrote:
> The VM stop process has to flush outstanding I/O and this is a critical
> failure scenario that is hard to diagnose. Add a probe point that
> records the flush return code.
>
> Signed-off-by: Daniel P. Berrangé
Reviewed-by: Dr. David Alan Gilbe
In order to avoid RMW cycles, is_allocated_sectors() treats zeroed areas
like non-zero data if the end of the checked area isn't aligned. This
can improve the efficiency of the conversion and was introduced in
commit 8dcd3c9b91a.
However, it comes with a correctness problem: qemu-img convert is
su
This demonstrates what happens when the block status changes in
sub-min_sparse granularity, but all of the parts are zeroed out. The
alignment logic in is_allocated_sectors() prevents that the target image
remains fully sparse as expected, but turns it into a data cluster of
explicit zeros.
Signed
Peter, three years ago you changed 'qemu-img convert' to sacrifice some
sparsification in order to get aligned requests on the target image. At
the time, I thought the impact would be small, but it turns out that
this can end up wasting gigabytes of storagee (like converting a fully
zeroed 10 GB im
Signed-off-by: Daniel P. Berrangé
---
block/file-posix.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/block/file-posix.c b/block/file-posix.c
index 6aafeda44f..2538e43299 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -106,8 +106,6 @@
#include
#endif
-#include "trace.h"
A flush failure is a critical failure scenario for some operations.
For example, it will prevent migration from completing, as it will
make vm_stop() report an error. Thus it is important to have a
trace point present for debugging.
Signed-off-by: Daniel P. Berrangé
---
block/file-posix.c | 2 ++
When fdatasync() fails on a file backend we set a flag that
short-circuits any future attempts to call fdatasync(). The
first failure returns the true errno, but the later short-
circuited calls return a generic EIO. The latter is unhelpful
because fdatasync() can return a variety of errnos, includ
The VM stop process has to flush outstanding I/O and this is a critical
failure scenario that is hard to diagnose. Add a probe point that
records the flush return code.
Signed-off-by: Daniel P. Berrangé
---
softmmu/cpus.c | 7 ++-
softmmu/trace-events | 3 +++
2 files changed, 9 insert
This is a critical failure scenario for migration that is hard to
diagnose from existing probes. Most likely it is caused by an error
from bdrv_flush(), but we're not logging the errno anywhere, hence
this new probe.
Signed-off-by: Daniel P. Berrangé
---
migration/migration.c | 1 +
migration/t
I spent a while debugging a tricky migration failure today which was
ultimately caused by fdatasync() getting EACCESS. The existing probes
were not sufficient to diagnose this, so I had to resort to GDB. This
improves probes and block error reporting to make future diagnosis
possible without GDB.
On 4/15/21 2:00 PM, Gollu Appalanaidu wrote:
> Make uniform hexadecimal numbers format.
>
> Signed-off-by: Gollu Appalanaidu
> ---
> -v2: Address review comments (Klaus)
> use lower case hexa format for the code and in comments
> use the same format as used in Spec. ("h")
^ This comment
Make uniform hexadecimal numbers format.
Signed-off-by: Gollu Appalanaidu
---
-v2: Address review comments (Klaus)
use lower case hexa format for the code and in comments
use the same format as used in Spec. ("h")
hw/block/nvme-ns.c | 2 +-
hw/block/nvme.c | 40 +++
Patchew URL:
https://patchew.org/QEMU/20210415124307.428203-1-pbonz...@redhat.com/
Hi,
This series seems to have some coding style problems. See output below for
more information:
Type: series
Message-id: 20210415124307.428203-1-pbonz...@redhat.com
Subject: [PATCH 0/3] file-posix: fix refresh
I/O to a disk via read/write is not limited by the number of segments allowed
by the host adapter; the kernel can split requests if needed, and the limit
imposed by the host adapter can be very low (256k or so) to avoid that SG_IO
returns EINVAL if memory is heavily fragmented.
Since this value is
refresh_limits is not doing anything for block devices, and is retrieving
the maximum number of s/g list entries incorrectly for character devices.
Patches 2-3 fix these problems, while patch 1 is a small improvement to
avoid making the BlockLimits unnecessarily restrictive when SG_IO is not
in us
bs->sg is only true for character devices, but block devices can also
be used with scsi-block and scsi-generic. Unfortunately BLKSECTGET
returns bytes for /dev/sgN devices and sectors for block devices, so
account for that in the code.
The maximum transfer also need not be a power of 2 (for examp
Even though it was only called for devices that have bs->sg set (which
must be character devices),
sg_get_max_segments looked at /sys/dev/block which only works for
block devices.
On Linux the sg driver has its own way to provide the maximum number of
iovecs in a scatter/gather list.
---
block/fi
Some machines use floppy controllers via the SysBus interface,
and don't need to pull in all the ISA code.
Extract the ISA specific code to a new unit: fdc-isa.c, and
add a new Kconfig symbol: "FDC_ISA".
Signed-off-by: Philippe Mathieu-Daudé
---
hw/block/fdc-isa.c | 313 +++
We want to extract ISA/SysBus code from the generic fdc.c file.
First, declare the prototypes we will access from the new units
into a new local header: "fdc-internal.h".
Signed-off-by: Philippe Mathieu-Daudé
---
hw/block/fdc-internal.h | 155
hw/block/fd
Some machines use floppy controllers via the SysBus interface,
and don't need to pull in all the SysBus code.
Extract the SysBus specific code to a new unit: fdc-sysbus.c,
and add a new Kconfig symbol: "FDC_SYSBUS".
Signed-off-by: Philippe Mathieu-Daudé
---
hw/block/fdc-sysbus.c | 252 ++
Hi,
The floppy disc controllers pulls in irrelevant devices (sysbus in
an ISA-only machine, ISA bus + isa devices on a sysbus-only machine).
This series clean that by extracting each device in its own file,
adding the corresponding Kconfig symbols: FDC_ISA and FDC_SYSBUS.
Regards,
Phil.
Philip
Signed-off-by: Philippe Mathieu-Daudé
---
hw/block/fdc.c| 7 +--
hw/block/trace-events | 1 +
2 files changed, 2 insertions(+), 6 deletions(-)
diff --git a/hw/block/fdc.c b/hw/block/fdc.c
index a825c2acbae..1d3a0473678 100644
--- a/hw/block/fdc.c
+++ b/hw/block/fdc.c
@@ -1242,12 +124
On 4/14/21 10:02 PM, Stefan Hajnoczi wrote:
> It can be difficult to debug issues with BHs in production environments.
> Although BHs can usually be identified by looking up their ->cb()
> function pointer, this requires debug information for the program. It is
> also not possible to print human-re
On 2021-04-14 21:02, Stefan Hajnoczi wrote:
> Eric Ernst and I debugged a BH leak and it was more involved than it should
> be.
> The problem is that BHs don't have a human-readable identifier, so low-level
> debugging techniques and inferences about the code are required to figure out
> which BH
34 matches
Mail list logo
