On Dec 12 12:44, Klaus Jensen wrote:
> From: Klaus Jensen <k.jensen@samsung.com>
>
> Guenter reports that hw/nvme is broken on riscv64[1] and big endian
> platforms[2].
>
> This is a regression since 7.1, so this does not warrent an rc5 for 7.2.
> I'm sure Guenter can carry this patch in his tree, and maybe we can get
> this out in a stable release.
>
> On riscv, the issue is a missing cq eventidx update. I really wonder why
> this issue only shows up on riscv64. We have not observed this on other
> platforms (yet).
>
> Further, Guenter also reported problems on big-endian platforms. The
> issue here is missing endian conversions which patch 3 addresses. This
> also requires a fix for the Linux kernel that I am posting separately
> (can't link to it, chicken and egg problem).
>
> [1]: https://lore.kernel.org/qemu-devel/20221207174918.GA1151796@roeck-us.net/
> [2]: https://lore.kernel.org/qemu-devel/20221209110022.GA3396194@roeck-us.net/
>
> v4:
> - screwed up the rebase (Philippe)
>
> v3:
> - add patch to fix big-endian platforms
>
> v2:
> - use QOM accessor (Philippe)
> - added some cleanup patches in front
>
> Klaus Jensen (4):
> hw/nvme: use QOM accessors
> hw/nvme: rename shadow doorbell related trace events
> hw/nvme: fix missing endian conversions for doorbell buffers
> hw/nvme: fix missing cq eventidx update
>
> hw/nvme/ctrl.c | 121 ++++++++++++++++++++++++++-----------------
> hw/nvme/trace-events | 8 +--
> 2 files changed, 78 insertions(+), 51 deletions(-)
>
> --
> 2.38.1
>
Applied to nvme-next. Thansk for the reviews!