From: Auger Eric <eric.auger@redhat.com>
To: "Philippe Mathieu-Daudé" <philmd@redhat.com>, qemu-devel@nongnu.org
Cc: Fam Zheng <fam@euphon.net>, Kevin Wolf <kwolf@redhat.com>,
qemu-block@nongnu.org, Max Reitz <mreitz@redhat.com>,
Alex Williamson <alex.williamson@redhat.com>,
Stefan Hajnoczi <stefanha@redhat.com>
Subject: Re: [PATCH v2 03/19] block/nvme: Introduce device/iommu 'page_size_min' variables
Date: Mon, 26 Oct 2020 18:38:42 +0100 [thread overview]
Message-ID: <81caffeb-832d-ab6f-b733-dc3ea61a9d93@redhat.com> (raw)
In-Reply-To: <20201026105504.4023620-4-philmd@redhat.com>
Hi Philippe,
On 10/26/20 11:54 AM, Philippe Mathieu-Daudé wrote:
> Introduce device/iommu 'page_size_min' variables to make
> the code clearer.
I am unclear how much the device and the iommu page size must equal. For
instance, in [RFC 0/5] NVMe passthrough: Support 64kB page host, I have
a 64kB host page and an MPS set to 4kB.
Thanks
Eric
>
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> block/nvme.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/block/nvme.c b/block/nvme.c
> index aa290996679..5abd7257cac 100644
> --- a/block/nvme.c
> +++ b/block/nvme.c
> @@ -690,6 +690,8 @@ static int nvme_init(BlockDriverState *bs, const char *device, int namespace,
> uint64_t deadline, now;
> Error *local_err = NULL;
> volatile NvmeBar *regs = NULL;
> + size_t device_page_size_min;
> + size_t iommu_page_size_min = 4096;
>
> qemu_co_mutex_init(&s->dma_map_lock);
> qemu_co_queue_init(&s->dma_flush_queue);
> @@ -724,7 +726,8 @@ static int nvme_init(BlockDriverState *bs, const char *device, int namespace,
> goto out;
> }
>
> - s->page_size = MAX(4096, 1u << (12 + NVME_CAP_MPSMIN(cap)));
> + device_page_size_min = 1u << (12 + NVME_CAP_MPSMIN(cap));
> + s->page_size = MAX(iommu_page_size_min, device_page_size_min);
> s->doorbell_scale = (4 << NVME_CAP_DSTRD(cap)) / sizeof(uint32_t);
> bs->bl.opt_mem_alignment = s->page_size;
> bs->bl.request_alignment = s->page_size;
>
next prev parent reply other threads:[~2020-10-26 18:14 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-10-26 10:54 [PATCH v2 00/19] util/vfio-helpers: Allow using multiple MSIX IRQs Philippe Mathieu-Daudé
2020-10-26 10:54 ` [PATCH v2 01/19] block/nvme: Correct minimum device page size Philippe Mathieu-Daudé
2020-10-26 17:57 ` Auger Eric
2020-10-26 10:54 ` [PATCH v2 02/19] block/nvme: Set request_alignment at initialization Philippe Mathieu-Daudé
2020-10-26 17:38 ` Auger Eric
2020-10-27 9:24 ` Stefan Hajnoczi
2020-10-26 10:54 ` [PATCH v2 03/19] block/nvme: Introduce device/iommu 'page_size_min' variables Philippe Mathieu-Daudé
2020-10-26 17:38 ` Auger Eric [this message]
2020-10-27 9:32 ` Stefan Hajnoczi
2020-10-26 10:54 ` [PATCH v2 04/19] block/nvme: Trace controller capabilities Philippe Mathieu-Daudé
2020-10-26 17:57 ` Auger Eric
2020-10-27 9:41 ` Stefan Hajnoczi
2020-10-26 10:54 ` [PATCH v2 05/19] util/vfio-helpers: Improve reporting unsupported IOMMU type Philippe Mathieu-Daudé
2020-10-26 10:54 ` [PATCH v2 06/19] util/vfio-helpers: Trace PCI I/O config accesses Philippe Mathieu-Daudé
2020-10-26 18:02 ` Auger Eric
2020-10-26 10:54 ` [PATCH v2 07/19] util/vfio-helpers: Trace PCI BAR region info Philippe Mathieu-Daudé
2020-10-26 18:06 ` Auger Eric
2020-10-26 10:54 ` [PATCH v2 08/19] util/vfio-helpers: Trace where BARs are mapped Philippe Mathieu-Daudé
2020-10-26 10:54 ` [PATCH v2 09/19] util/vfio-helpers: Improve DMA trace events Philippe Mathieu-Daudé
2020-10-26 10:54 ` [PATCH v2 10/19] util/vfio-helpers: Convert vfio_dump_mapping to " Philippe Mathieu-Daudé
2020-10-26 10:54 ` [PATCH v2 11/19] util/vfio-helpers: Let qemu_vfio_dma_map() propagate Error Philippe Mathieu-Daudé
2020-10-26 19:58 ` Auger Eric
2020-10-26 10:54 ` [PATCH v2 12/19] util/vfio-helpers: Let qemu_vfio_do_mapping() " Philippe Mathieu-Daudé
2020-10-26 20:02 ` Auger Eric
2020-10-26 10:54 ` [PATCH v2 13/19] util/vfio-helpers: Let qemu_vfio_verify_mappings() use error_report() Philippe Mathieu-Daudé
2020-10-26 10:54 ` [PATCH v2 14/19] util/vfio-helpers: Pass minimum page size to qemu_vfio_open_pci() Philippe Mathieu-Daudé
2020-10-27 9:50 ` Stefan Hajnoczi
2020-10-26 10:55 ` [PATCH v2 15/19] util/vfio-helpers: Report error when IOMMU page size is not supported Philippe Mathieu-Daudé
2020-10-26 16:12 ` Auger Eric
2020-10-26 10:55 ` [PATCH v2 16/19] util/vfio-helpers: Introduce qemu_vfio_pci_msix_init_irqs() Philippe Mathieu-Daudé
2020-10-26 20:24 ` Auger Eric
2020-10-26 10:55 ` [PATCH v2 17/19] util/vfio-helpers: Introduce qemu_vfio_pci_msix_set_irq() Philippe Mathieu-Daudé
2020-10-26 20:28 ` Auger Eric
2020-10-26 10:55 ` [PATCH v2 18/19] block/nvme: Switch to using the MSIX API Philippe Mathieu-Daudé
2020-10-26 20:32 ` Auger Eric
2020-10-27 9:55 ` Philippe Mathieu-Daudé
2020-10-26 10:55 ` [PATCH v2 19/19] util/vfio-helpers: Remove now unused qemu_vfio_pci_init_irq() Philippe Mathieu-Daudé
2020-10-27 9:55 ` [PATCH v2 00/19] util/vfio-helpers: Allow using multiple MSIX IRQs Stefan Hajnoczi
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=81caffeb-832d-ab6f-b733-dc3ea61a9d93@redhat.com \
--to=eric.auger@redhat.com \
--cc=alex.williamson@redhat.com \
--cc=fam@euphon.net \
--cc=kwolf@redhat.com \
--cc=mreitz@redhat.com \
--cc=philmd@redhat.com \
--cc=qemu-block@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).