On Tue, Oct 27, 2020 at 12:05:56AM -0500, Eric Blake wrote:
> Anywhere we create a list of just one item or by prepending items
> (typically because order doesn't matter), we can use the now-public
> macro. But places where we must keep the list in order by appending
> remain open-coded.
>
> Sign
strace shows the problem:
add_key("user", "qemu_test_secret", "Test Payload", 12,
KEY_SPEC_PROCESS_KEYRING) = -1 EPERM (Operation not permitted)
It appears systemd-nspawn containers don't have CAP_SYS_ADMIN which is
apparently needed for the keyring stuff to work. Fair enough.
But the underlying
marcandre.lur...@redhat.com writes:
> From: Marc-André Lureau
>
> The function is going to be called from a coroutine, and may yields.
s/yields/yield/
> Let's ensure our image reference doesn't change over time (due to resize
> etc) by keeping a ref.
>
> Signed-off-by: Marc-André Lureau
Revie
On Tue, 27 Oct 2020 12:56:40 +1100
David Gibson wrote:
> On Mon, Oct 26, 2020 at 03:46:47PM +0100, Greg Kurz wrote:
> > On Mon, 26 Oct 2020 14:43:08 +0100
> > Philippe Mathieu-Daudé wrote:
> >
> > > On 10/26/20 1:40 PM, Greg Kurz wrote:
> > > > qemu_memalign() aborts if OOM. Drop some dead code
The following changes since commit a46e72710566eea0f90f9c673a0f02da0064acce:
Merge remote-tracking branch 'remotes/cohuck/tags/s390x-20201026' into
staging (2020-10-26 14:50:03 +)
are available in the Git repository at:
git://github.com/mdroth/qemu.git tags/qga-pull-2020-10-27-tag
for
From: Markus Armbruster
guest-get-devices returns 'driver-date' as string in the format
-MM-DD. Goes back to recent commit 2e4211cee4 "qga: add command
guest-get-devices for reporting VirtIO devices".
We should avoid use of multiple encodings for the same kind of data.
Especially string enc
From: Marc-André Lureau
I prefer 'reset' over 'clear', since 'clear' and keys may have some
other relations or meaning.
Signed-off-by: Marc-André Lureau
Signed-off-by: Michael Roth
---
qga/commands-posix-ssh.c | 53
qga/qapi-schema.json | 3 ++-
2
From: Marc-André Lureau
The glib function was introduced in 2.64. It's a safer version of
getpwnam, and also simpler to use than getpwnam_r.
Currently, it's only use by the next patch in qemu-ga, which doesn't
(well well...) need the thread safety guarantees. Since the fallback
version is still
From: Marc-André Lureau
Signed-off-by: Marc-André Lureau
Signed-off-by: Michael Roth
---
qga/meson.build | 7 +--
1 file changed, 1 insertion(+), 6 deletions(-)
diff --git a/qga/meson.build b/qga/meson.build
index 6315bb357e..8340892139 100644
--- a/qga/meson.build
+++ b/qga/meson.build
@
From: Marc-André Lureau
Add new commands to add and remove SSH public keys from
~/.ssh/authorized_keys.
I took a different approach for testing, including the unit tests right
with the code. I wanted to overwrite the function to get the user
details, I couldn't easily do that over QMP. Furthermo
From: Marc-André Lureau
Signed-off-by: Marc-André Lureau
Signed-off-by: Michael Roth
---
qga/commands-posix-ssh.c | 66
qga/meson.build | 11 +--
qga/qapi-schema.json | 31 +++
3 files changed, 106 insertions(+), 2 delet
From: Markus Armbruster
Member 'address' is union GuestDeviceAddress with a single branch
GuestDeviceAddressPCI, containing PCI vendor ID and device ID. This
is not a PCI address. Type GuestPCIAddress is. Messed up in recent
commit 2e4211cee4 "qga: add command guest-get-devices for reporting
V
From: Markus Armbruster
Simple unions are simpler than flat unions in the schema, but more
complicated in C and on the QMP wire: there's extra indirection in C
and extra nesting on the wire, both pointless. They should be avoided
in new code.
GuestDeviceId was recently added for guest-get-devic
I've recently switched employers and the current email address is out
of date.
Signed-off-by: Michael Roth
---
MAINTAINERS | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index ef6f5c7399..c72af7c2f5 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@
From: Markus Armbruster
The Error ** argument must be NULL, &error_abort, &error_fatal, or a
pointer to a variable containing NULL. Passing an argument of the
latter kind twice without clearing it in between is wrong: if the
first call sets an error, it no longer points to NULL for the second
ca
From: Tomáš Golembiovský
Add API and stubs for new guest-get-disks command.
The command guest-get-fsinfo can be used to list information about disks
and partitions but it is limited only to mounted disks with filesystem.
This new command should allow listing information about disks of the VM
reg
From: Tomáš Golembiovský
The command lists all disks (real and virtual) as well as disk
partitions. For each disk the list of dependent disks is also listed and
/dev path is used as a handle so it can be matched with "name" field of
other returned disk entries. For disk partitions the "dependents
From: Tomáš Golembiovský
The command lists all the physical disk drives. Unlike for Linux
partitions and virtual volumes are not listed.
Example output:
{
"return": [
{
"name": ".\\PhysicalDrive0",
"partition": false,
"address": {
"serial": "QM1",
Patchew URL:
https://patchew.org/QEMU/20201027055317.351868-1-michael.r...@amd.com/
Hi,
This series seems to have some coding style problems. See output below for
more information:
Type: series
Message-id: 20201027055317.351868-1-michael.r...@amd.com
Subject: [PULL 00/12] qemu-ga patch queue
> case CHR_EVENT_OPENED:
> -if (!s->dev.attached) {
> +if (!s->always_plugged && !s->dev.attached) {
> usb_device_attach(&s->dev, &error_abort);
> }
Not needed (but doesn't hurt either).
> break;
> case CHR_EVENT_CLOSED:
> -if (s->d
The commit 88fc107956a5812649e5918e0c092d3f78bb28ad disabled remote
posix locks by default. But the --help message still says it is enabled
by default. So fix it to output no_posix_lock.
Signed-off-by: Jiachen Zhang
---
tools/virtiofsd/helper.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(
On Mon, Oct 26, 2020 at 6:07 PM Philippe Mathieu-Daudé wrote:
>
> ping?
Oops wrong series, these patches are already merged, sorry.
>
> On 10/17/20 8:07 PM, Philippe Mathieu-Daudé wrote:
> > Add trace event for IRQ from CPU/GPU,
> > use definitions for IRQ numbers.
> >
> > Philippe Mathieu-Daudé
On Mon, 26 Oct 2020 10:30:25 -0400
Paolo Bonzini wrote:
> Cc: Thomas Huth
> Signed-off-by: Paolo Bonzini
> ---
> hw/s390x/ipl.c | 8 ++--
> hw/s390x/s390-virtio-ccw.c | 3 ++-
> 2 files changed, 4 insertions(+), 7 deletions(-)
Reviewed-by: Cornelia Huck
On 26.10.20 15:30, Paolo Bonzini wrote:
> Cc: Thomas Huth
> Signed-off-by: Paolo Bonzini
> ---
> hw/s390x/ipl.c | 8 ++--
> hw/s390x/s390-virtio-ccw.c | 3 ++-
> 2 files changed, 4 insertions(+), 7 deletions(-)
>
> diff --git a/hw/s390x/ipl.c b/hw/s390x/ipl.c
> index 3d2652d7
marcandre.lur...@redhat.com writes:
> From: Marc-André Lureau
>
> Thanks to the monitors coroutine support, the screendump handler can
monitors'
Suggest to add (merge commit b7092cda1b3) right before the comma.
> trigger a graphic_hw_update(), yield and let the main loop run until
> update is
On 10/27/20 9:15 AM, Jiachen Zhang wrote:
> The commit 88fc107956a5812649e5918e0c092d3f78bb28ad disabled remote
> posix locks by default. But the --help message still says it is enabled
> by default. So fix it to output no_posix_lock.
>
> Signed-off-by: Jiachen Zhang
Reviewed-by: Philippe Mathie
On 10/26/20 3:47 PM, Greg Kurz wrote:
> On Mon, 26 Oct 2020 14:49:34 +0100
> Philippe Mathieu-Daudé wrote:
>
>> On 10/26/20 1:40 PM, Greg Kurz wrote:
>>> spapr_reallocate_hpt() has three users, two of which pass &error_fatal
>>> and the third one, htab_load(), passes &local_err, uses it to detect
Hi Peter,
On 10/19/20 9:31 PM, Peter Maydell wrote:
> On Mon, 19 Oct 2020 at 16:45, Peter Maydell wrote:
>>
>> On Sat, 10 Oct 2020 at 14:57, Luc Michel wrote:
>>>
>>> v2 -> v3:
>>> - patch 03: moved clock_new definition to hw/core/clock.c [Phil]
>>> - patch 03: commit message typo [Clement]
On 10/26/20 9:33 AM, Mark Cave-Ayland wrote:
> Signed-off-by: Mark Cave-Ayland
> ---
> hw/usb/dev-serial.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
Reviewed-by: Philippe Mathieu-Daudé
* Greg Kurz (gr...@kaod.org) wrote:
> On Mon, 26 Oct 2020 13:48:37 +0100
> Christian Schoenebeck wrote:
>
> > On Montag, 26. Oktober 2020 11:33:42 CET Peter Maydell wrote:
> > > On Fri, 23 Oct 2020 at 12:46, Christian Schoenebeck
> > >
> > > wrote:
> > > > The following changes since commit
>
On 10/26/20 9:33 AM, Mark Cave-Ayland wrote:
> Signed-off-by: Mark Cave-Ayland
> ---
> hw/usb/dev-serial.c | 28 ++--
> hw/usb/trace-events | 8
> 2 files changed, 22 insertions(+), 14 deletions(-)
>
> diff --git a/hw/usb/dev-serial.c b/hw/usb/dev-serial.c
> ind
On 10/26/20 9:33 AM, Mark Cave-Ayland wrote:
> Signed-off-by: Mark Cave-Ayland
> ---
> hw/usb/dev-serial.c | 3 +++
> hw/usb/trace-events | 2 ++
> 2 files changed, 5 insertions(+)
>
> diff --git a/hw/usb/dev-serial.c b/hw/usb/dev-serial.c
> index abc316c7bf..badf8785db 100644
> --- a/hw/usb/dev
On 10/26/20 9:33 AM, Mark Cave-Ayland wrote:
> The DeviceOutVendor and DeviceInVendor macros can be replaced with their
> equivalent VendorDeviceOutRequest and VendorDeviceRequest macros from usb.h.
>
> Signed-off-by: Mark Cave-Ayland
> ---
> hw/usb/dev-serial.c | 25 ++---
>
On Mon, Oct 26, 2020 at 11:54:47AM +0100, Philippe Mathieu-Daudé wrote:
> When introducing this driver in commit bdd6a90a9e5
> ("block: Add VFIO based NVMe driver") we correctly
> set the request_alignment in nvme_refresh_limits()
> but forgot to set it at initialization. Do it now.
This patch is
Hi Peter,
On 10/19/20 9:26 PM, Peter Maydell wrote:
> On Wed, 14 Oct 2020 at 20:36, Dr. David Alan Gilbert (git)
> wrote:
>>
>> From: "Dr. David Alan Gilbert"
>>
>> Use of 0x%d - make up our mind as 0x%x
>>
>> Signed-off-by: Dr. David Alan Gilbert
>> ---
>> hw/arm/trace-events | 2 +-
>> 1 fil
On 10/26/20 2:51 PM, Paolo Bonzini wrote:
> Apple's nm implementation includes empty lines in the output that are not
> found in GNU binutils. This confuses scripts/undefsym.py, though it did
> not confuse the scripts/undefsym.sh script that it replaced. To fix
> this, ignore lines that do not ha
On Mon, Oct 26, 2020 at 11:54:48AM +0100, Philippe Mathieu-Daudé wrote:
> Introduce device/iommu 'page_size_min' variables to make
> the code clearer.
>
> Signed-off-by: Philippe Mathieu-Daudé
> ---
> block/nvme.c | 5 -
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/bloc
On 25/10/2020 11:41, Philippe Mathieu-Daudé wrote:
On 10/25/20 12:11 PM, Mark Cave-Ayland wrote:
On 21/10/2020 12:43, Mark Cave-Ayland wrote:
Store the child object directly within the sabre object rather than using
link properties.
Signed-off-by: Mark Cave-Ayland
---
hw/pci-host/sabre.c
John Snow writes:
> There's no reason to keep this here; the versions described are
> ancient. Everything here is still mirrored on
> https://wiki.qemu.org/ChangeLog/old if anyone is curious; otherwise, use
> the git history.
>
> Signed-off-by: John Snow
Reviewed-by: Alex Bennée
--
Alex Be
On Mon, Oct 26, 2020 at 11:54:49AM +0100, Philippe Mathieu-Daudé wrote:
> Controllers have different capabilities and report them in the
> CAP register. We are particularly interested by the page size
> limits.
>
> Signed-off-by: Philippe Mathieu-Daudé
> ---
> block/nvme.c | 10 ++
On Mon, Oct 26, 2020 at 11:54:59AM +0100, Philippe Mathieu-Daudé wrote:
> @@ -737,6 +738,17 @@ static int nvme_init(BlockDriverState *bs, const char
> *device, int namespace,
> }
>
> device_page_size_min = 1u << (12 + NVME_CAP_MPSMIN(cap));
> +device_page_size_max = 1u << (12 + NVM
Le 26/10/2020 à 20:51, John Snow a écrit :
> There's no reason to keep this here; the versions described are
> ancient. Everything here is still mirrored on
> https://wiki.qemu.org/ChangeLog/old if anyone is curious; otherwise, use
> the git history.
>
> Signed-off-by: John Snow
> ---
>
> V2: Ad
On 10/26/20 9:32 PM, Auger Eric wrote:
> Hi Philippe,
>
> On 10/26/20 11:55 AM, Philippe Mathieu-Daudé wrote:
>> In preparation of using multiple IRQs, switch to using the recently
>> introduced MSIX API. Instead of allocating and assigning IRQ in
>> a single step, we now have to use two distinct
On Mon, Oct 26, 2020 at 11:54:45AM +0100, Philippe Mathieu-Daudé wrote:
> This series allow using multiple MSIX IRQs
> We currently share a single IRQ between 2 NVMe queues
> (ADMIN and I/O). This series still uses 1 shared IRQ
> but prepare for using multiple ones.
>
> Since v1:
> - Addressed Ste
From: Daniel P. Berrangé
The error message was supposed to mention the input revision list start
point, not the branch flag.
Signed-off-by: Daniel P. Berrangé
Signed-off-by: Alex Bennée
Message-Id: <20201019143537.283094-3-berra...@redhat.com>
Message-Id: <20201021163136.27324-6-alex.ben...@li
From: Ani Sinha
Ani is an individual contributor into qemu project. Adding my email into the
correct file to reflect so.
Signed-off-by: Ani Sinha
Signed-off-by: Alex Bennée
Reviewed-by: Philippe Mathieu-Daudé
Message-Id: <20201007161940.1478-1-...@anisinha.ca>
Message-Id: <20201021163136.2732
Signed-off-by: Alex Bennée
Reviewed-by: Philippe Mathieu-Daudé
Acked-by: Subbaraya Sundeep
Acked-by: Michael Rolnik
Acked-by: Thomas Huth
Acked-by: James Hogan
Acked-by: Artyom Tarasenko
Message-id: <20201004182506.2038515-1-f4...@amsat.org>
Message-Id: <20201021163136.27324-3-alex.ben...@li
The following changes since commit 1dc887329a10903940501b43e8c0cc67af7c06d5:
Merge remote-tracking branch 'remotes/philmd-gitlab/tags/sd-next-20201026'
into staging (2020-10-26 17:19:26 +)
are available in the Git repository at:
https://github.com/stsquad/qemu.git tags/pull-testing-and-
Currently the test randomly fails if you are using a shared machine
due to contention on the well known port 1234. We can ameliorate this
a bit by picking a random non-ephemeral port although it doesn't
totally avoid the problem. While we could use a totally unique socket
address for debugging it i
From: Philippe Mathieu-Daudé
Let GitLab runners use GitLab repository directly.
Suggested-by: Paolo Bonzini
Signed-off-by: Philippe Mathieu-Daudé
Reviewed-by: Thomas Huth
Message-Id: <20201022123302.2884788-1-phi...@redhat.com>
Signed-off-by: Alex Bennée
diff --git a/.gitlab-ci.yml b/.gitla
From: Thomas Huth
We do not support SDL1 in QEMU anymore. Use SDL2 instead.
Signed-off-by: Thomas Huth
Signed-off-by: Alex Bennée
Tested-by: Philippe Mathieu-Daudé
Reviewed-by: Daniel P. Berrangé
Reviewed-by: Philippe Mathieu-Daudé
Message-Id: <20201021072308.9224-1-th...@redhat.com>
Messag
From: Daniel P. Berrangé
If the current branch is synced to the current upstream git master,
there are no commits that need checking. This causes checkpatch.pl
to print an error that it found no commits. We need to avoid calling
checkpatch.pl in this case.
Signed-off-by: Daniel P. Berrangé
Sign
From: Paolo Bonzini
Use $(findstring) instead of $(filter) to detect -n/-k
as different versions of MAKE fill in $(MAKEFLAGS) differently.
Do not bother running ninja at all if -nq is passed.
Tested-by: Alex Bennée
Signed-off-by: Paolo Bonzini
Signed-off-by: Alex Bennée
Message-Id: <202010261
On 10/26/20 3:28 PM, Gerd Hoffmann wrote:
> There is no actual code in the CONFIG_VIRGL=n case. So building is
> (a) pointless and (b) makes macos ranlib complain.
>
> Reported-by: Paolo Bonzini
> Signed-off-by: Gerd Hoffmann
> ---
> hw/display/virtio-gpu-3d.c | 4
> hw/display/meson.buil
27.10.2020 08:05, Eric Blake wrote:
There is no need to rely on the verbosity of the gcc/clang compiler
extension of g_new(typeof(X), 1) when we can instead use the standard
g_malloc(sizeof(X)). In general, we like g_new over g_malloc for
returning type X rather than void* to let the compiler ca
On 10/27/20 6:30 AM, Brad Smith wrote:
> tests/vm: update openbsd to release 6.8
>
> A double dash at the end of a package name removes ambiguity
> when the intent is to install a non-FLAVORed package.
>
> Signed-off-by: Brad Smith
> Reviewed-by: Gerd Hoffmann
> Tested-by: Gerd Hoffmann
> Revi
Eric Blake writes:
> Anywhere we create a list of just one item or by prepending items
> (typically because order doesn't matter), we can use the now-public
> macro. But places where we must keep the list in order by appending
> remain open-coded.
Should we rename the macro to QAPI_LIST_PREPEND
Daniele Buono writes:
> LLVM/Clang, supports runtime checks for forward-edge Control-Flow
> Integrity (CFI).
>
> CFI on indirect function calls (cfi-icall) ensures that, in indirect
> function calls, the function called is of the right signature for the
> pointer type defined at compile time.
>
From: John Snow
This script has not seen a patch that was specifically for this script
since it was moved to this location in 2013, and I doubt it is used. It
uses "man qmp" for its help message, which does not exist. It also
presumes there is a manual page for qmp-XXX, for each defined qmp
comma
The following changes since commit 1dc887329a10903940501b43e8c0cc67af7c06d5:
Merge remote-tracking branch 'remotes/philmd-gitlab/tags/sd-next-20201026'
into staging (2020-10-26 17:19:26 +)
are available in the Git repository at:
git://repo.or.cz/qemu/armbru.git tags/pull-qmp-2020-10-27
From: David Edmondson
If the user selects pretty-printing (-p) the contents of any
dictionaries in the output are sorted by key.
Signed-off-by: David Edmondson
Message-Id: <20201013141414.18398-1-david.edmond...@oracle.com>
Reviewed-by: Philippe Mathieu-Daudé
Signed-off-by: Markus Armbruster
On Dienstag, 27. Oktober 2020 10:06:53 CET Dr. David Alan Gilbert wrote:
> * Greg Kurz (gr...@kaod.org) wrote:
> > On Mon, 26 Oct 2020 13:48:37 +0100
> >
> > Christian Schoenebeck wrote:
> > > On Montag, 26. Oktober 2020 11:33:42 CET Peter Maydell wrote:
> > > > On Fri, 23 Oct 2020 at 12:46, Chri
From: Alexander von Gluck IV
Signed-off-by: Alexander von Gluck IV
[PMD: Avoid recreating the image each time]
Signed-off-by: Philippe Mathieu-Daudé
---
tests/keys/vagrant| 27 +
tests/keys/vagrant.pub| 1 +
tests/vm/Makefile.include | 3 +-
tests/vm/basevm.py|
Intent to not get Alexander's patch lost.
Since v3:
- Grammar fix (eblake)
Since v2:
- No change, posted as single patch
Since v1:
- Addressed Thomas Huth review comments
Alexander von Gluck IV (1):
tests/vm: Add Haiku test based on their vagrant images
tests/keys/vagrant| 27 +
On Tue, Oct 27, 2020 at 11:05:20AM +0100, Philippe Mathieu-Daud?? wrote:
> On 10/27/20 6:30 AM, Brad Smith wrote:
> > tests/vm: update openbsd to release 6.8
> >
> > A double dash at the end of a package name removes ambiguity
> > when the intent is to install a non-FLAVORed package.
> >
> > Sign
On Mon, 26 Oct 2020 at 12:05, Eric Blake wrote:
>
> The following changes since commit 4c5b97bfd0dd54dc27717ae8d1cd10e14eef1430:
>
> Merge remote-tracking branch
> 'remotes/kraxel/tags/modules-20201022-pull-request' into staging (2020-10-22
> 12:33:21 +0100)
>
> are available in the Git reposi
* Christian Schoenebeck (qemu_...@crudebyte.com) wrote:
> On Dienstag, 27. Oktober 2020 10:06:53 CET Dr. David Alan Gilbert wrote:
> > * Greg Kurz (gr...@kaod.org) wrote:
> > > On Mon, 26 Oct 2020 13:48:37 +0100
> > >
> > > Christian Schoenebeck wrote:
> > > > On Montag, 26. Oktober 2020 11:33:42
"Minus" features are applied after "plus" features, so ensure that a
later "plus" feature causes an earlier "minus" feature to be removed.
This has no effect on the existing "-feature,feature=on" backward
compatibility code (which warns and turns the feature off).
Signed-off-by: David Edmondson
27.10.2020 08:05, Eric Blake wrote:
Since 'block-export-add' is new to 5.2, we can still tweak the
interface; there, allowing 'bitmaps':['str'] is nicer than
'bitmap':'str'. This wires up the qapi and qemu-nbd changes to permit
passing multiple bitmaps as distinct metadata contexts that the NBD
27.10.2020 08:05, Eric Blake wrote:
Rather than open-code the count of negotiated contexts at several
sites, embed it directly into the struct. This will make it easier
for upcoming commits to support even more simultaneous contexts.
Signed-off-by: Eric Blake
Reviewed-by: Vladimir Sementsov-O
Am 10.10.2020 um 22:41 hat marcandre.lur...@redhat.com geschrieben:
> From: Marc-André Lureau
>
> The assert() was added in commit b681a1c73e15 ("block: Repair the
> throttling code."), when the qemu_co_queue_do_restart() function
> required to be running in a coroutine. It was later made unnece
On Tue, 27 Oct 2020 at 09:24, Auger Eric wrote:
>
> Hi Peter,
>
> On 10/19/20 9:26 PM, Peter Maydell wrote:
> > Eric, do we want to use hex here, or should we go for
> > decimal the way we do with (almost) all the other
> > tracing of stream IDs (eg mmuv3_cmdq_cfgi_ste in the line before)?
> >
> >
Hi Peter,
On 10/27/20 11:38 AM, Peter Maydell wrote:
> OK; I'll apply this patch to target-arm.next; feel free to send
> a patch updating the other tracepoints to hex.
sure, I will.
Thanks
Eric
From: Klaus Jensen
Hi Peter,
The following changes since commit 1dc887329a10903940501b43e8c0cc67af7c06d5:
Merge remote-tracking branch 'remotes/philmd-gitlab/tags/sd-next-20201026'
into staging (2020-10-26 17:19:26 +)
are available in the Git repository at:
git://git.infradead.org/qe
From: Klaus Jensen
Some devices might want to know the return value of dma_memory_rw, so
pass it along instead of ignoring it.
There are no existing users of the return value, so this patch should be
safe.
Signed-off-by: Klaus Jensen
Reviewed-by: Philippe Mathieu-Daudé
Reviewed-by: Michael S.
From: Klaus Jensen
Handling DMA errors gracefully is required for the device to pass the
block/011 test ("disable PCI device while doing I/O") in the blktests
suite.
With this patch the device sets the Controller Fatal Status bit in the
CSTS register when failing to read from a submission queue
From: Klaus Jensen
Style fixes.
Signed-off-by: Klaus Jensen
Reviewed-by: Philippe Mathieu-Daudé
Reviewed-by: Keith Busch
---
hw/block/nvme.c | 25 +
1 file changed, 13 insertions(+), 12 deletions(-)
diff --git a/hw/block/nvme.c b/hw/block/nvme.c
index 0ac9d856632e..d
From: Klaus Jensen
Move common error handling to a label.
Signed-off-by: Klaus Jensen
Reviewed-by: Keith Busch
---
hw/block/nvme.c | 16 +---
1 file changed, 9 insertions(+), 7 deletions(-)
diff --git a/hw/block/nvme.c b/hw/block/nvme.c
index 7d328c08b894..0ac9d856632e 100644
---
From: Klaus Jensen
Fix a typo in the sq doorbell trace event.
Signed-off-by: Klaus Jensen
Reviewed-by: Philippe Mathieu-Daudé
Reviewed-by: Keith Busch
---
hw/block/trace-events | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/block/trace-events b/hw/block/trace-events
i
From: Klaus Jensen
Add the nvme_l2b helper and use it for converting NLB and SLBA to byte
counts and offsets.
Signed-off-by: Klaus Jensen
Reviewed-by: Keith Busch
---
hw/block/nvme.h | 6 ++
hw/block/nvme.c | 12
2 files changed, 10 insertions(+), 8 deletions(-)
diff --git
From: Klaus Jensen
For now, support the Data Block, Segment and Last Segment descriptor
types.
See NVM Express 1.3d, Section 4.4 ("Scatter Gather List (SGL)").
Signed-off-by: Klaus Jensen
Reviewed-by: Keith Busch
---
include/block/nvme.h | 6 +-
hw/block/nvme.c | 329 +++
From: Klaus Jensen
The raw NLB field is a 16 bit value, so use le16_to_cpu instead of
le32_to_cpu and cast to uint32_t before incrementing the value to not
wrap around.
Signed-off-by: Klaus Jensen
Reviewed-by: Keith Busch
Reviewed-by: Philippe Mathieu-Daudé
---
hw/block/nvme.c | 4 ++--
1 fi
From: Klaus Jensen
This pulls block layer aio submission/completion to common functions.
For completions, additionally map an AIO error to the Unrecovered Read
and Write Fault status codes.
Signed-off-by: Klaus Jensen
Reviewed-by: Keith Busch
---
hw/block/nvme.h | 14 +
hw/block/n
From: Klaus Jensen
Add the symbolic command name to the pci_nvme_{io,admin}_cmd and
pci_nvme_rw trace events.
Signed-off-by: Klaus Jensen
Reviewed-by: Philippe Mathieu-Daudé
Reviewed-by: Keith Busch
---
hw/block/nvme.h | 28
hw/block/nvme.c | 8 +
From: Klaus Jensen
There are two reasons for changing this:
1. The nvme device currently uses an internal Intel device id.
2. Since commits "nvme: fix write zeroes offset and count" and "nvme:
support multiple namespaces" the controller device no longer has
the quirks that the Lin
From: Klaus Jensen
Make the default request status NVME_SUCCESS so only error status codes
have to be set.
Signed-off-by: Klaus Jensen
Reviewed-by: Philippe Mathieu-Daudé
Reviewed-by: Keith Busch
---
hw/block/nvme.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/hw/bl
From: Klaus Jensen
This adds support for multiple namespaces by introducing a new 'nvme-ns'
device model. The nvme device creates a bus named from the device name
('id'). The nvme-ns devices then connect to this and registers
themselves with the nvme device.
This changes how an nvme device is cr
From: Klaus Jensen
Prepare to support inactive namespaces.
Signed-off-by: Klaus Jensen
Reviewed-by: Maxim Levitsky
Reviewed-by: Keith Busch
---
hw/block/nvme.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/hw/block/nvme.c b/hw/block/nvme.c
index 4f08f55a76a0..92
From: Keith Busch
The code switches on the opcode to invoke a function specific to that
opcode. There's no point in consolidating back to a common function that
just switches on that same opcode without any actual common code.
Restore the opcode specific behavior without going back through anothe
From: Klaus Jensen
Include the namespace id in the pci_nvme_{get,set}feat trace events.
Signed-off-by: Klaus Jensen
Signed-off-by: Keith Busch
---
hw/block/nvme.c | 4 ++--
hw/block/trace-events | 4 ++--
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/hw/block/nvme.c b/h
From: Klaus Jensen
Since the controller has only supported PRPs so far it has not been
required to check the ending address (addr + len - 1) of the CMB access
for validity since it has been guaranteed to be in range of the CMB.
This changes when the controller adds support for SGLs (next patch),
From: Klaus Jensen
If the user does not specify an nsid parameter on the nvme-ns device,
nvme_register_namespace will find the first free namespace id and assign
that.
This fix makes sure the assigned id is saved.
Signed-off-by: Klaus Jensen
Reviewed-by: Dmitry Fomichev
---
hw/block/nvme.c |
From: Keith Busch
Signed-off-by: Keith Busch
Signed-off-by: Klaus Jensen
---
include/block/nvme.h | 3 ++-
hw/block/nvme.c | 1 +
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/include/block/nvme.h b/include/block/nvme.h
index bc20a2ba5e87..521533fd2a10 100644
--- a/include
From: Gollu Appalanaidu
This adds support for SGL descriptor type 0x1 (bit bucket descriptor).
See the NVM Express v1.3d specification, Section 4.4 ("Scatter Gather
List (SGL)").
Signed-off-by: Gollu Appalanaidu
Signed-off-by: Klaus Jensen
Reviewed-by: Keith Busch
---
hw/block/nvme.c | 33 ++
From: Klaus Jensen
If the host sets CC.CSS to 111b, all commands submitted to I/O queues
should be completed with status Invalid Command Opcode.
Note that this is technically a v1.4 feature, but it does not hurt to
implement before we finally bump the reported version implemented.
Reviewed-by:
From: Keith Busch
Fail to start the controller if the user requests a command set that the
controller does not support.
Signed-off-by: Keith Busch
Signed-off-by: Klaus Jensen
---
include/block/nvme.h | 4
hw/block/nvme.c | 6 +-
hw/block/trace-events | 1 +
3 files changed, 10
From: Gollu Appalanaidu
Replace the Invalid Field in Command with the Invalid PRP Offset status
code in the nvme_create_{cq,sq} functions. Also, allow PRP1 to be
address 0x0.
Also replace the Completion Queue Invalid status code returned in
nvme_create_cq when the the queue identifier is invalid
From: Gollu Appalanaidu
Address 0 is not an invalid address. Remove those invalikd checks.
Unaligned PRP2 and PRP list entries should result in Invalid PRP Offset
status code and not Invalid Field. Fix that.
See NVMe Express v1.3d, Section 4.3 ("Physical Region Page Entry and
List").
Suggested
From: Klaus Jensen
The emulated nvme device (hw/block/nvme.c) is currently using an
internal Intel device id.
Prepare to change that by allocating a device id under the 1b36 (Red
Hat, Inc.) vendor id.
Signed-off-by: Klaus Jensen
Acked-by: Gerd Hoffmann
Reviewed-by: Maxim Levitsky
Reviewed-by
From: Keith Busch
Return error if the requested offset starts after the size of the log
being returned. Also, move the check for earlier in the function so
we're not doing unnecessary calculations.
Reviewed-by: Philippe Mathieu-Daudé
Reviewed- by: Dmitry Fomichev
Signed-off-by: Keith Busch
Si
1 - 100 of 592 matches
Mail list logo
