On May 17 17:52, Changqi Lu wrote:
> This commit enables the rescap function in the
> namespace by detecting the supported reservation
> function in the backend driver.
>
> Signed-off-by: Changqi Lu
> Signed-off-by: zhenwei pi
> ---
> hw/nvme/ns.c | 8
> 1 file changed, 8 insertions(+)
On May 17 17:52, Changqi Lu wrote:
> This commit enables ONCS to support the reservation
> function at the controller level. It also lays the
> groundwork for detecting and enabling the reservation
> function on a per-namespace basis in RESCAP.
>
> Signed-off-by: Changqi Lu
> Signed-off-by: zhenw
On May 17 17:52, Changqi Lu wrote:
> Add constants for the NVMe persistent command protocol.
> The constants include the reservation command opcode and
> reservation type values defined in section 7 of the NVMe
> 2.0 specification.
>
> Signed-off-by: Changqi Lu
> Signed-off-by: zhenwei pi
> ---
On May 17 17:52, Changqi Lu wrote:
> This commit enables the rescap function in the
> namespace by detecting the supported reservation
> function in the backend driver.
>
> Signed-off-by: Changqi Lu
> Signed-off-by: zhenwei pi
> ---
> hw/nvme/ns.c | 8
> 1 file changed, 8 insertions(+)
false, nvme_misc_cb, req);
> +break;
> +case NVME_RESV_ACQUIRE_ACTION_PREEMPT_AND_ABORT:
> +req->aiocb = blk_aio_pr_preempt(ns->blkconf.blk, key_info.cr_key,
> +key_info.nr_key, type, true,
> +
On May 8 09:36, Vincent Fu wrote:
> On 5/7/24 10:05, Vincent Fu wrote:
> > On 5/6/24 04:06, Klaus Jensen wrote:
> > > The Management Operation field of I/O Management Send is only 8 bits,
> > > not 16.
> > >
> > > Fixes: 73064edfb864 ("hw/nvme
exible=$nr_vq,sriov_vi_flexible=$nr_vi
> \
>
> To fix this issue, this patch modifies the calculation of BAR size in
> the PF and VF initialization by using different elements:
>
> PF: `max_ioqpairs + 1` with `msix_qsize`
> VF: VQFRSM with VIFRSM
>
> Signed
nvme: separate identify data for sec. ctrl list
> hw/nvme: Allocate sec-ctrl-list as a dynamic array
> hw/nvme: Expand VI/VQ resource to uint32
>
> hw/nvme/ctrl.c | 59 +++-----
> hw/nvme/nvme.h | 19 +++---
> hw/nvme/subsys.c | 10 +---
> include/block/nvme.h | 1 +
> 4 files changed, 54 insertions(+), 35 deletions(-)
>
> --
> 2.34.1
>
Looks good Minwoo!
Grabbing for nvme-next.
Reviewed-by: Klaus Jensen
signature.asc
Description: PGP signature
de/block/nvme.h
> index 8b125f7769..9b9eaeb3a7 100644
> --- a/include/block/nvme.h
> +++ b/include/block/nvme.h
> @@ -1251,7 +1251,7 @@ enum NvmeIdCtrlOncs {
> NVME_ONCS_DSM = 1 << 2,
> NVME_ONCS_WRITE_ZEROES = 1 << 3,
> NVME_ONCS_FEATURES = 1 <&l
AP_WR_EX_RO : 0;
> +res |= (block_pr_cap & BLK_PR_CAP_EX_AC_RO) ?
> + NVME_PR_CAP_EX_AC_RO : 0;
> +res |= (block_pr_cap & BLK_PR_CAP_WR_EX_AR) ?
> + NVME_PR_CAP_WR_EX_AR : 0;
> +res |= (block_pr_cap & BLK_PR_CAP_EX_AC_AR) ?
> + NVME_PR_CAP_EX_AC_AR : 0;
> +
> +return res;
> +}
> +
> typedef struct NvmeSQueue {
> struct NvmeCtrl *ctrl;
> uint16_tsqid;
> --
> 2.20.1
>
Reviewed-by: Klaus Jensen
--
One of us - No more doubt, silence or taboo about mental illness.
signature.asc
Description: PGP signature
; @@ -2592,6 +2592,7 @@ next:
> done:
> iocb->aiocb = NULL;
> iocb->common.cb(iocb->common.opaque, iocb->ret);
> +g_free(iocb->range);
> qemu_aio_unref(iocb);
> }
>
> --
> 2.34.1
>
Thanks! LGTM
Reviewed-by: Klaus Jensen
Fixes: d7d1474fd85d ("hw/nvme: reimplement dsm to allow cancellation")
Cc: qemu-sta...@nongnu.org
signature.asc
Description: PGP signature
From: Vincent Fu
The number of PIDs is in the upper 16 bits of cdw10. So we need to
right-shift by 16 bits instead of only a single bit.
Fixes: 73064edfb864 ("hw/nvme: flexible data placement emulation")
Cc: qemu-sta...@nongnu.org
Signed-off-by: Vincent Fu
Reviewed-by: Klaus Jensen
From: Minwoo Im
To prevent further bumping up the number of maximum VF te support, this
patch allocates a dynamic array (NvmeCtrl *)->sec_ctrl_list based on
number of VF supported by sriov_max_vfs property.
Reviewed-by: Klaus Jensen
Signed-off-by: Minwoo Im
Signed-off-by: Klaus Jen
data structure when the
command comes in.
Reviewed-by: Klaus Jensen
Signed-off-by: Minwoo Im
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 21 ++---
hw/nvme/nvme.h | 14 --
hw/nvme/subsys.c | 8
3 files changed, 22 insertions(+), 21 deletions(-)
diff --
controllers.
Reviewed-by: Klaus Jensen
Signed-off-by: Minwoo Im
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 8
hw/nvme/nvme.h | 4 ++--
2 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/hw/nvme/ctrl.c b/hw/nvme/ctrl.c
index 1e50b57707ba..5b1b0cabcfc3 100644
--- a/hw
From: Klaus Jensen
Hi,
The following changes since commit 59084feb256c617063e0dbe7e64821ae8852d7cf:
Merge tag 'pull-aspeed-20240709' of https://github.com/legoater/qemu into
staging (2024-07-09 07:13:55 -0700)
are available in the Git repository at:
https://gitlab.com/birkelun
property is not provided then the default mqes will still be
0x7ff (the queue size is 2048 entries).
Signed-off-by: John Berg
Reviewed-by: Keith Busch
Reviewed-by: Klaus Jensen
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 8 +++-
hw/nvme/nvme.h | 1 +
2 files changed, 8 insertions
List
(CNS 19h). But, yes, only just for the endurance group #1. This patch
allows host applications to ask for which endurance group is available
and utilize FDP through that endurance group.
Reviewed-by: Klaus Jensen
Signed-off-by: Minwoo Im
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c
initialization by using different elements:
PF: `max_ioqpairs + 1` with `msix_qsize`
VF: VQFRSM with VIFRSM
Signed-off-by: Minwoo Im
Reviewed-by: Klaus Jensen
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 19 +++
1 file changed, 15 insertions(+), 4 deletions(-)
diff
in a persistent manner.
> Signed-off-by: Changqi Lu
> Signed-off-by: zhenwei pi
> Acked-by: Klaus Jensen
> ---
> hw/nvme/ctrl.c | 318 +++
> hw/nvme/nvme.h | 4 +
> include/block/nvme.h | 37 +
> 3 files changed
eq)
> {
> -NvmeCmd *cmd = (NvmeCmd *)&req->cmd;
> +NvmeCmd *cmd = &req->cmd;
> NvmeNamespace *ns = req->ns;
> /* cdw12 is zero-based number of dwords to return. Convert to bytes */
> uint32_t data_size = (le32_to_cpu(cmd->cdw12) + 1) <<
From: Yao Xingtao
The type of req->cmd is NvmeCmd, cast the pointer of this type to
NvmeCmd* is useless.
Signed-off-by: Yao Xingtao
Reviewed-by: Klaus Jensen
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/nvme/ctrl.
From: Arun Kumar
Extend copy command to copy user data across different namespaces via
support for specifying a namespace for each source range
Signed-off-by: Arun Kumar
Reviewed-by: Klaus Jensen
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 369
to allow cancellation")
Signed-off-by: Zheyu Ma
Reviewed-by: Klaus Jensen
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/hw/nvme/ctrl.c b/hw/nvme/ctrl.c
index 5b1b0cabcfc3..f7b4e4627dd2 100644
--- a/hw/nvme/ctrl.c
+++ b/hw/nvme/ctrl
From: Ayush Mishra
Abort was not implemented previously, but we can implement it for AERs
and asynchrnously for I/O.
Signed-off-by: Ayush Mishra
Reviewed-by: Klaus Jensen
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 32
1 file changed, 32 insertions
From: Klaus Jensen
Hi,
The following changes since commit a7ddb48bd1363c8bcdf42776d320289c42191f01:
Merge tag 'pull-aspeed-20240721' of https://github.com/legoater/qemu into
staging (2024-07-22 07:52:05 +1000)
are available in the Git repository at:
https://gitlab.com/birkelun
On Jul 25 19:42, 卢长奇 wrote:
> Hi,
>
> ```
> 2685 nvme_status->regctl_ds[i].cntlid = nvme_ctrl(req)->cntlid;
> 2686 nvme_status->regctl_ds[i].rkey = keys_info->keys[i];
> 2687 nvme_status->regctl_ds[i].rcsts = keys_info->keys[i] ==
> 2688 reservation->key ? 1 : 0;
> 2689 /* hostid is not supported
> nvme resv-register /dev/nvme0n1 --crkey 3 --nrkey 5 --rrega 2
> > #release
> > nvme resv-release /dev/nvme0n1 --crkey 5 --rtype 1 --rrela 0
> > #clear
> > nvme resv-release /dev/nvme0n1 --crkey 5 --rtype 1 --rrela 1
> > #reserve
> > nvme resv-acquire /dev/nvm
From: Klaus Jensen
Yutaro Shimizu from the Cyber Defense Institute discovered a bug in the
NVMe emulation that leaks contents of an uninitialized heap buffer if
subsystem and FDP emulation are enabled.
Cc: qemu-sta...@nongnu.org
Reported-by: Yutaro Shimizu
Signed-off-by: Klaus Jensen
---
hw
From: Klaus Jensen
Hi,
The following changes since commit 48e4ba59a3756aad743982da16bf9b5120d91a0c:
Merge tag 'pull-riscv-to-apply-20240819-1' of
https://github.com/alistair23/qemu into staging (2024-08-19 14:55:23 +1000)
are available in the Git repository at:
https://
On Aug 20 12:30, Philippe Mathieu-Daudé wrote:
> Hi Klaus,
>
> On 20/8/24 06:45, Klaus Jensen wrote:
> > From: Klaus Jensen
> >
> > Yutaro Shimizu from the Cyber Defense Institute discovered a bug in the
> > NVMe emulation that leaks contents of an uninitialized
e0n1 --crkey 3 --nrkey 5 --rrega 2
> #release
> nvme resv-release /dev/nvme0n1 --crkey 5 --rtype 1 --rrela 0
> #clear
> nvme resv-release /dev/nvme0n1 --crkey 5 --rtype 1 --rrela 1
> #reserve
> nvme resv-acquire /dev/nvme0n1 --crkey 3 --rtype 1 --racqa 0
> #premmpt
> nvme r
in a persistent manner.
> Signed-off-by: Changqi Lu
> Signed-off-by: zhenwei pi
> Acked-by: Klaus Jensen
> ---
> hw/nvme/ctrl.c | 318 +++
> hw/nvme/nvme.h | 4 +
> include/block/nvme.h | 37 +
> 3 files change
in a persistent manner.
> Signed-off-by: Changqi Lu
> Signed-off-by: zhenwei pi
> Acked-by: Klaus Jensen
> ---
> hw/nvme/ctrl.c | 318 +++
> hw/nvme/nvme.h | 4 +
> include/block/nvme.h | 37 +
> 3 files chang
On Aug 28 00:20, 卢长奇 wrote:
> Hi,
>
> I want to know if I understand it correctly.
>
> ```
> static void nvme_aio_err(NvmeRequest *req, int ret)
> {
> uint16_t status = NVME_SUCCESS;
> Error *local_err = NULL;
>
> switch (req->cmd.opcode) {
> case NVME_CMD_READ:
> case NVME_CMD_RESV_REPORT:
> st
in a persistent manner.
> Signed-off-by: Changqi Lu
> Signed-off-by: zhenwei pi
> Acked-by: Klaus Jensen
> ---
> hw/nvme/ctrl.c | 330 +++
> hw/nvme/nvme.h | 4 +
> include/block/nvme.h | 37 +
> 3 files change
---
> hw/nvme/ctrl.c | 8
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
Reviewed-by: Klaus Jensen
> set to the AWUN value. If Disable Normal is set, the maximum atomic write
> size is set to AWUPF.
>
> Signed-off-by: Alan Adamson
LGTM. I can fix up the few nits below.
Keith, I'll let this linger a few days for you to comment :)
Revi
On Aug 20 09:11, Alan Adamson wrote:
> Since there is work in the Linux NVMe Driver community to add Atomic Write
> support, it would be desirable to be able to test it with qemu nvme emulation.
>
> This patch will focus on supporting NVMe controller atomic write parameters
> (AWUN and
> AWUPF)
From: Łukasz Gieryk
With the new command one can:
- assign flexible resources (queues, interrupts) to primary and
secondary controllers,
- toggle the online/offline state of given controller.
Signed-off-by: Łukasz Gieryk
Acked-by: Michael S. Tsirkin
Reviewed-by: Klaus Jensen
Signed-off
enable SR-IOV support.
Signed-off-by: Lukasz Maniak
Acked-by: Michael S. Tsirkin
Reviewed-by: Klaus Jensen
Signed-off-by: Klaus Jensen
---
docs/system/devices/nvme.rst | 82
1 file changed, 82 insertions(+)
diff --git a/docs/system/devices/nvme.rst b/docs
uninitialized queue.
While it’s an interesting question whether a VF should support AER in
the first place, I don’t think it must be answered today.
Signed-off-by: Łukasz Gieryk
Reviewed-by: Klaus Jensen
Acked-by: Michael S. Tsirkin
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 3 +--
1 file changed
posed fix is to make the PCI ACPI code aware of SR/IOV.
Signed-off-by: Łukasz Gieryk
Acked-by: Michael S. Tsirkin
Reviewed-by: Michael S. Tsirkin
Signed-off-by: Klaus Jensen
---
hw/acpi/pcihp.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/hw/acpi/pcihp.c b/hw/acpi/p
From: Klaus Jensen
This reverts commit d97eee64fef35655bd06f5c44a07fdb83a6274ae.
The emulated controller correctly accounts for not including bit buckets
in the controller-to-host data transfer, however it doesn't correctly
account for the holes for the on-disk data offsets.
Reported-by:
From: Klaus Jensen
The SRIOV series exposed an issued with how CC register writes are
handled and how CSTS is set in response to that. Specifically, after
applying the SRIOV series, the controller could end up in a state with
CC.EN set to '1' but with CSTS.RDY cleared to '0'
On Jun 27 14:39, Niklas Cassel wrote:
> The serial prop on the controller is actually describing the nvme
> subsystem serial, which has to be identical for all controllers within
> the same nvme subsystem.
>
> This is enforced since commit a859eb9f8f64 ("hw/nvme: enforce common
> serial per subsys
On Jun 27 13:47, Niklas Cassel wrote:
> Since we verify that "ready_delay" parameter has to be smaller than CRWMT,
> we know that the namespace will always become ready.
> Therefore the "Namespace Not Ready" status code will never have the DNR
> bit set.
>
> Add a new parameter "never_ready" that
On Jun 27 10:59, Christoph Hellwig wrote:
> On Mon, Jun 27, 2022 at 01:47:28PM +0200, Niklas Cassel via wrote:
> > CRMS.CRWMS bit shall be set to 1 on controllers compliant with versions
> > later than NVMe 1.4.
> >
> > The first version later than NVMe 1.4 is NVMe 2.0
> >
> > Let's claim complia
On Jun 27 18:34, Niklas Cassel wrote:
> Hello qemu-nvme maintainers,
>
> I noticed that since commit
> 916b0f0b5264 ("hw/nvme: change nvme-ns 'shared' default")
>
> The default value of ns param 'shared' is true, regardless
> if there is a nvme-subsys node or not.
>
> It probably doesn't matter
ams params;
> +NvmeSubsystem *subsys;
>
> struct {
> uint32_t err_rec;
> --
> 2.36.1
>
No problemo.
Reviewed-by: Klaus Jensen
signature.asc
Description: PGP signature
On Jun 27 13:47, Niklas Cassel wrote:
> TP4084 adds a new mode, CC.CRIME, that can be used to mark a namespace
> as ready independently from the controller.
>
> When CC.CRIME is 0 (default), things behave as before, all namespaces
> are ready when CSTS.RDY gets set to 1.
>
> When CC.CRIME is 1, t
On Jun 28 13:26, Daniel P. Berrangé wrote:
> On Mon, Jun 27, 2022 at 02:39:57PM +0200, Niklas Cassel via wrote:
> > The serial prop on the controller is actually describing the nvme
> > subsystem serial, which has to be identical for all controllers within
> > the same nvme subsystem.
>
> Given th
On Jun 28 14:22, Niklas Cassel wrote:
> Since commit 916b0f0b5264 ("hw/nvme: change nvme-ns 'shared' default")
> the default value of nvme-ns param 'shared' is set to true, regardless
> if there is a nvme-subsys node or not.
>
> On a system without a nvme-subsys node, a namespace will never be abl
tions(-)
>
> Klaus: ping
>
> Reviewed-by: Stefan Hajnoczi
Sorry,
Thanks Darren, applied to nvme-next!
Reviewed-by: Klaus Jensen
signature.asc
Description: PGP signature
From: Klaus Jensen
While testing Jinhaos ioeventfd patch I found it useful with a couple of
additional trace events since we no longer see the mmio events.
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 8
hw/nvme/trace-events | 4
2 files changed, 12 insertions(+)
diff
On Jun 23 23:18, Klaus Jensen wrote:
> From: Klaus Jensen
>
> This series includes a couple of misc fixes as well as some cleanup
> pertaining to the aio handling in flush, dsm, copy and zone reset. As
> Jinhao gets around to iothread stuff, it might come in handy to have
> thi
bell buffer for the admin queue. Copying the doorbell register
value to the shadow doorbell buffer allows us to support these hosts
as well as spec-compliant hosts that use shadow doorbell buffer for
the admin queue.
Signed-off-by: Jinhao Fan
Reviewed-by: Klaus Jensen
Reviewed-by: Keith Busch
[k.je
From: Klaus Jensen
Hi,
The following changes since commit 8482ab545e52f50facacfe1118b22b97462724ab:
Merge tag 'qga-win32-pull-2022-07-13' of github.com:kostyanf14/qemu into
staging (2022-07-14 14:52:16 +0100)
are available in the Git repository at:
git://git.infradead.org/qem
Buffer Config command.
Signed-off-by: Jinhao Fan
Reviewed-by: Klaus Jensen
Reviewed-by: Keith Busch
[k.jensen: rebased]
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 5 +
hw/nvme/trace-events | 5 +
2 files changed, 10 insertions(+)
diff --git a/hw/nvme/ctrl.c b/hw/nvme
.
Fix the documentation, so that people copying the qemu command line
example won't get an error on qemu start.
Signed-off-by: Niklas Cassel
Signed-off-by: Klaus Jensen
---
docs/system/devices/nvme.rst | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/docs/system/devices/nvme.rs
-by: Klaus Jensen
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 113 -
hw/nvme/nvme.h | 5 +++
2 files changed, 117 insertions(+), 1 deletion(-)
diff --git a/hw/nvme/ctrl.c b/hw/nvme/ctrl.c
index 55cb0ba1d591..533ad14e7a61 100644
--- a/hw/nvme/ct
more than one controller anyway.
Signed-off-by: Niklas Cassel
Reviewed-by: Klaus Jensen
Signed-off-by: Klaus Jensen
---
hw/nvme/ns.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/hw/nvme/ns.c b/hw/nvme/ns.c
index 870c3ca1a2f0..62a1f97be010 100644
--- a/hw/nvme/ns.c
+++ b/hw/n
& FLAG_TEST
which will result in a different result to:
(x ? a: b) & FLAG_TEST.
The macros should wrap each of the parameters in brackets to ensure the
correct result on expansion.
Signed-off-by: Darren Kenny
Reviewed-by: Stefan Hajnoczi
Reviewed-by: Klaus Jensen
Signed-off-by: Klau
On Jul 20 10:47, Jinhao Fan wrote:
> at 10:41 PM, Jinhao Fan wrote:
>
> > at 1:34 PM, Klaus Jensen wrote:
> >
> >> From: Klaus Jensen
> >>
> >> While testing Jinhaos ioeventfd patch I found it useful with a couple of
> >> addition
From: Klaus Jensen
A set of fixes/changes to the ioeventfd support.
Klaus Jensen (3):
hw/nvme: skip queue processing if notifier is cleared
hw/nvme: unregister the event notifier handler on the main loop
hw/nvme: do not enable ioeventfd by default
hw/nvme/ctrl.c | 12 +---
1
From: Klaus Jensen
While it is safe to process the queues when they are empty, skip it if
the event notifier callback was invoked spuriously.
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/hw/nvme/ctrl.c b/hw/nvme
From: Klaus Jensen
Make sure the notifier handler is unregistered in the main loop prior to
cleaning it up.
Fixes: 2e53b0b45024 ("hw/nvme: Use ioeventfd to handle doorbell updates")
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 2 ++
1 file changed, 2 insertions(+)
diff --git
From: Klaus Jensen
Do not enable ioeventfd by default. Let the feature mature a bit before
we consider enabling it by default.
Fixes: 2e53b0b45024 ("hw/nvme: Use ioeventfd to handle doorbell updates")
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 2 +-
1 file changed, 1 inser
From: Klaus Jensen
While it is safe to process the queues when they are empty, skip it if
the event notifier callback was invoked spuriously.
Reviewed-by: Keith Busch
Reviewed-by: Jinhao Fan
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 8 ++--
1 file changed, 6 insertions(+), 2
From: Klaus Jensen
Make sure the notifier handler is unregistered in the main loop prior to
cleaning it up.
Fixes: 2e53b0b45024 ("hw/nvme: Use ioeventfd to handle doorbell updates")
Reviewed-by: Keith Busch
Reviewed-by: Jinhao Fan
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c
From: Klaus Jensen
Hi,
The following changes since commit 3916603e0c1d909e14e09d5ebcbdaa9c9e21adf3:
Merge tag 'pull-la-20220729' of https://gitlab.com/rth7680/qemu into staging
(2022-07-29 17:39:17 -0700)
are available in the Git repository at:
git://git.infradead.org/qemu-nvm
From: Klaus Jensen
Do not enable ioeventfd by default. Let the feature mature a bit before
we consider enabling it by default.
Fixes: 2e53b0b45024 ("hw/nvme: Use ioeventfd to handle doorbell updates")
Reviewed-by: Keith Busch
Reviewed-by: Jinhao Fan
Signed-off-by: Klaus Jensen
--
On Jul 28 16:07, Jinhao Fan wrote:
> With the introduction of shadow doorbell and ioeventfd, we need to do
> frequent conversion between qid and its doorbell offset. The original
> hard-coded calculation is confusing and error-prone. Add several helper
> functions to do this task.
>
> Signed-off-b
On Aug 1 19:22, Stephen Bates wrote:
> Hi
>
> I am interested in exploring the impact of increased storage IO delay
> on applications and was thinking of doing this by adding a delay
> argument to the NVMe device model in QEMU. Is this something that
> others would be interested in? If so I woul
On Aug 2 09:03, Klaus Jensen wrote:
> In that case we would need to introduce a timer delay in
> QEMU.
I meant to type "[...] introduce a timer delay in the nvme device
model", not "[...] in QEMU".
Brain fart ;)
signature.asc
Description: PGP signature
On Aug 2 16:31, Jinhao Fan wrote:
> at 2:02 PM, Klaus Jensen wrote:
>
> > On Jul 28 16:07, Jinhao Fan wrote:
> >> With the introduction of shadow doorbell and ioeventfd, we need to do
> >> frequent conversion between qid and its doorbell offset. The original
On Aug 2 10:54, Klaus Jensen wrote:
> On Aug 2 16:31, Jinhao Fan wrote:
> > at 2:02 PM, Klaus Jensen wrote:
> >
> > > On Jul 28 16:07, Jinhao Fan wrote:
> > >> With the introduction of shadow doorbell and ioeventfd, we need to do
> > >> frequen
;
> Signed-off-by: Jinhao Fan
Looks good,
Reviewed-by: Klaus Jensen
signature.asc
Description: PGP signature
On Aug 11 23:37, Jinhao Fan wrote:
> When irqfd is enabled, we bypass QEMU's irq emulation and let KVM to
> directly assert the irq. However, KVM is not aware of the device's MSI-x
> masking status. Add MSI-x mask bookkeeping in NVMe emulation and
> detach the corresponding irqfd when the certain v
On Aug 11 23:37, Jinhao Fan wrote:
> When the new option 'irq-eventfd' is turned on, the IO emulation code
> signals an eventfd when it want to (de)assert an irq. The main loop
> eventfd handler does the actual irq (de)assertion. This paves the way
> for iothread support since QEMU's interrupt emu
On Aug 11 23:37, Jinhao Fan wrote:
> When the new option 'irq-eventfd' is turned on, the IO emulation code
> signals an eventfd when it want to (de)assert an irq. The main loop
> eventfd handler does the actual irq (de)assertion. This paves the way
> for iothread support since QEMU's interrupt emu
On Aug 11 23:37, Jinhao Fan wrote:
> When irqfd is enabled, we bypass QEMU's irq emulation and let KVM to
> directly assert the irq. However, KVM is not aware of the device's MSI-x
> masking status. Add MSI-x mask bookkeeping in NVMe emulation and
> detach the corresponding irqfd when the certain v
amp;& format != 0x1) {
> +if ((ns->pif == 0x0 && format != 0x0) || (ns->pif && format != 0x1)) {
> status = NVME_INVALID_FORMAT | NVME_DNR;
> goto invalid;
> }
> --
> 2.25.1
>
Thanks, looks good.
Reviewed-by: Klaus Jensen
signature.asc
Description: PGP signature
Hi Michael,
Thanks for your interest in contributing this!
(CC: +Naveen, +Keith, -linux-nvme)
Some observations and commets on your patch submission below.
Firstly, your mail delivery is broken - this is not getting through to
qemu-devel. Also, I don't think kernel developers have that much
int
On Aug 23 22:43, Jinhao Fan wrote:
> On 8/16/2022 6:46 PM, Klaus Jensen wrote:
> > Did qtest work out for you for testing? If so, it would be nice to add a
> > simple test case as well.
>
> Since MSI-x masking handlers are only implemented for IO queues, if we want
>
From: Klaus Jensen
Even if the host is somehow using compare to do compare-and-write, the
host should be notified immediately about the compare failure and not
have to wait for the driver to potentially retry the command.
Reported-by: Jim Harris
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c
On Aug 25 15:47, Jinhao Fan wrote:
> When the new option 'irq-eventfd' is turned on, the IO emulation code
> signals an eventfd when it want to (de)assert an irq. The main loop
> eventfd handler does the actual irq (de)assertion. This paves the way
> for iothread support since QEMU's interrupt emu
On Aug 25 19:16, Jinhao Fan wrote:
> On 8/25/2022 5:33 PM, Klaus Jensen wrote:
> > I'm still a bit perplexed by this issue, so I just tried moving
> > nvme_init_irq_notifier() to the end of nvme_init_cq() and removing this
> > first_io_cqe thing. I did not obs
On Aug 25 13:56, Klaus Jensen wrote:
> On Aug 25 19:16, Jinhao Fan wrote:
> > On 8/25/2022 5:33 PM, Klaus Jensen wrote:
> > > I'm still a bit perplexed by this issue, so I just tried moving
> > > nvme_init_irq_notifier() to the end of nvme_init_cq() and removing
On Aug 25 21:09, Jinhao Fan wrote:
>
>
>
> > 在 2022年8月25日，20:39，Klaus Jensen 写道：
> >
> > On Aug 25 13:56, Klaus Jensen wrote:
> >>> On Aug 25 19:16, Jinhao Fan wrote:
> >>> On 8/25/2022 5:33 PM, Klaus Jensen wrote:
> >>>&g
From: Klaus Jensen
This is a re-spin of Jinhao's irqfd support series that fixes msix
vector masking/unmasking to work correctly.
I kept being bugged out about that msi route not getting updated, so I
hit the code into submission with a hammer.
I finally noticed the core issue:
1
o Fan
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 145 ++-
hw/nvme/nvme.h | 3 +
hw/nvme/trace-events | 3 +
3 files changed, 149 insertions(+), 2 deletions(-)
diff --git a/hw/nvme/ctrl.c b/hw/nvme/ctrl.c
index 51792f395597..396f3f0cddbd 1
32 106 240 364
Signed-off-by: Jinhao Fan
Signed-off-by: Klaus Jensen
---
hw/nvme/ctrl.c | 120 ++---
hw/nvme/nvme.h | 3 ++
2 files changed, 106 insertions(+), 17 deletions(-)
diff --git a/hw/nvme/ctrl.c b/hw/nvme/ctrl.c
index 87aeba056499..51792f3955
On Aug 26 10:03, Jinhao Fan wrote:
> at 4:14 AM, Klaus Jensen wrote:
>
> > From: Klaus Jensen
> >
> > This is a re-spin of Jinhao's irqfd support series that fixes msix
> > vector masking/unmasking to work correctly.
> >
> > I kept being bugged o
nt notifiers since handler callback are called
> > in the same system call as irqfd write.
> >
> > Signed-off-by: Jinhao Fan
> > Signed-off-by: Klaus Jensen
>
> No idea what's going on here... This one is causing the following assert
> failure with --en
On Aug 26 09:54, Keith Busch wrote:
> On Fri, Aug 26, 2022 at 05:45:21PM +0200, Klaus Jensen wrote:
> > On Aug 26 09:34, Keith Busch wrote:
> > > On Fri, Aug 26, 2022 at 11:12:04PM +0800, Jinhao Fan wrote:
> > > > Use KVM's irqfd to send interrupts when possib
On Aug 12 13:01, Niklas Cassel wrote:
> The intention of the Zoned Namespace Command Set Specification was
> never to make an automatic zone transition optional.
>
> Excerpt from the nvmexpress.org zns mailing list:
> """
> A question came up internally on the differences between ZNS and ZAC/ZBC
>
On Sep 10 13:27, Sam Li wrote:
> Add a new zoned_host_device BlockDriver. The zoned_host_device option
> accepts only zoned host block devices. By adding zone management
> operations in this new BlockDriver, users can use the new block
> layer APIs including Report Zone and four zone management ope
On Sep 21 13:44, Damien Le Moal wrote:
> On 9/20/22 17:51, Klaus Jensen wrote:
> > On Sep 10 13:27, Sam Li wrote:
> > > Add a new zoned_host_device BlockDriver. The zoned_host_device option
> > > accepts only zoned host block devices. By adding zone management
&
On Okt 10 19:01, Daniel Wagner wrote:
> On Tue, May 11, 2021 at 06:12:47PM +0200, Hannes Reinecke wrote:
> > On 5/11/21 6:03 PM, Klaus Jensen wrote:
> > > On May 11 16:54, Hannes Reinecke wrote:
> > > > On 5/11/21 3:37 PM, Klaus Jensen wrote:
> > > >
1 - 100 of 2378 matches
Mail list logo
