From: sashiko-bot@kernel.org
To: "Piórkowski, Piotr" <piotr.piorkowski@intel.com>,
intel-xe@lists.freedesktop.org
Subject: Re: [PATCH v9 3/5] drm/xe/ggtt: Split GGTT into usable and shareable pools
Date: Mon, 16 Nov 2026 09:25:52 +0000 [thread overview]
Message-ID: <sashiko-outbox-216610@kernel.org> (raw)
In-Reply-To: <20261116090053.2466938-4-piotr.piorkowski@intel.com>
Thank you for your contribution! Sashiko AI review found 1 potential
issue to consider:
- [Medium] drm/xe/ggtt: u32 size truncation in
xe_ggtt_insert_node_shareable()
--
commit 77f9ae0aec9ac84bb1acae228f0f6fa5525c45d5
Author: Piotr Piórkowski <piotr.piorkowski@intel.com>
drm/xe/ggtt: Split GGTT into usable and shareable pools
This commit splits the GGTT into a usable pool for driver-owned allocations
and a shareable pool for SR-IOV VF provisioning, adding
xe_ggtt_insert_node_shareable() for the shareable range.
> diff --git a/drivers/gpu/drm/xe/xe_ggtt.c b/drivers/gpu/drm/xe/xe_ggtt.c
> index 73c45eecae2be..8bc888f6274cc 100644
> --- a/drivers/gpu/drm/xe/xe_ggtt.c
> +++ b/drivers/gpu/drm/xe/xe_ggtt.c
[ ... ]
> @@ -652,24 +687,65 @@ static void xe_ggtt_invalidate_engine(struct xe_ggtt *ggtt)
[ ... ]
> -static int xe_ggtt_insert_node_locked(struct xe_ggtt_node *node,
> - u32 size, u32 align, u32 mm_flags)
> +static int ggtt_insert_node_in_range_locked(struct xe_ggtt_node *node, u32 size,
> + u32 align, u64 range_start,
> + u64 range_size, u32 mm_flags)
> {
> - return drm_mm_insert_node_generic(&node->ggtt->mm, &node->base, size, align, 0,
> - mm_flags);
> + struct xe_ggtt *ggtt = node->ggtt;
> + u64 range_end = range_start + range_size;
> +
> + lockdep_assert_held(&ggtt->lock);
> +
> + if (!range_size || range_end <= range_start)
> + return -EINVAL;
> +
> + if (range_end > ggtt->full_size)
> + return -ERANGE;
> +
> + if (size > range_size)
> + return -ENOSPC;
> +
> + return drm_mm_insert_node_in_range(&ggtt->mm, &node->base, size, align, 0,
> + range_start, range_end, mm_flags);
> +}
[ ... ]
> @@ -703,8 +779,45 @@ struct xe_ggtt_node *xe_ggtt_insert_node(struct xe_ggtt *ggtt, u32 size, u32 ali
[ ... ]
> +#ifdef CONFIG_PCI_IOV
> +/**
> + * xe_ggtt_insert_node_shareable() - Insert a new node in the shareable GGTT range
> + * @ggtt: the &xe_ggtt into which the node should be inserted.
> + * @size: size of the node
> + * @align: alignment constrain of the node
> + *
> + * Allocations always start from the top (DRM_MM_INSERT_HIGH).
> + *
> + * Return: &xe_ggtt_node on success or an ERR_PTR on failure.
> + */
> +struct xe_ggtt_node *xe_ggtt_insert_node_shareable(struct xe_ggtt *ggtt, u32 size, u32 align)
> +{
> + struct xe_ggtt_node *node;
> + int ret;
> +
> + if (!ggtt->shareable_size)
> + return ERR_PTR(-ENOSPC);
> +
> + node = ggtt_node_init(ggtt);
> + if (IS_ERR(node))
> + return node;
> +
> + guard(mutex)(&ggtt->lock);
> +
> + ret = ggtt_insert_node_in_range_locked(node, size, align,
> + ggtt_shareable_offset(ggtt),
> + ggtt->shareable_size,
> + DRM_MM_INSERT_HIGH);
> if (ret) {
> ggtt_node_fini(node);
> return ERR_PTR(ret);
> }
>
> return node;
> }
> +#endif
[Severity: Medium]
Should size and align in xe_ggtt_insert_node_shareable(),
xe_ggtt_insert_node_locked() and ggtt_insert_node_in_range_locked() be u64
instead of u32?
pf_provision_vf_ggtt() takes a u64 size (for example from the u64
ggtt_quota debugfs attribute) and passes it to
xe_ggtt_insert_node_shareable(). A 64-bit size >= 4 GiB with non-zero
lower 32 bits (such as SZ_4G + SZ_64M) is truncated to u32 before the
size > range_size check in ggtt_insert_node_in_range_locked(), so instead
of returning -ENOSPC, pf_provision_vf_ggtt() allocates a node for the
truncated 32-bit size and then passes the untruncated 64-bit size to GuC
via pf_distribute_config_ggtt(). Similarly, xe_ggtt_insert_node_transform()
accepts a u64 size and passes it to xe_ggtt_insert_node_locked().
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261116090053.2466938-1-piotr.piorkowski@intel.com?part=3
next prev parent reply other threads:[~2026-11-16 9:26 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-16 9:00 [PATCH v9 0/5] Separate GGTT pools for submissions and VFs provisioning Piórkowski, Piotr
2026-11-16 9:00 ` [PATCH v9 1/5] drm/xe/ggtt: Assert lower bound of GGTT PTE address Piórkowski, Piotr
2026-11-16 9:00 ` [PATCH v9 2/5] drm/xe/ggtt: Introduce full GGTT range size Piórkowski, Piotr
2026-11-16 9:00 ` [PATCH v9 3/5] drm/xe/ggtt: Split GGTT into usable and shareable pools Piórkowski, Piotr
2026-11-16 9:25 ` sashiko-bot [this message]
2026-11-16 9:00 ` [PATCH v9 4/5] drm/xe/ggtt: Initialize GGTT pools by SR-IOV mode Piórkowski, Piotr
2026-11-16 9:00 ` [PATCH v9 5/5] drm/xe/ggtt: Add KUnit tests for usable and shareable pools Piórkowski, Piotr
2026-11-16 10:02 ` ✗ CI.checkpatch: warning for Separate GGTT pools for submissions and VFs provisioning (rev9) Patchwork
2026-11-16 10:04 ` ✓ CI.KUnit: success " Patchwork
2026-11-16 10:37 ` ✓ Xe.CI.BAT: " Patchwork
2026-11-16 14:35 ` ✓ Xe.CI.FULL: " Patchwork
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=sashiko-outbox-216610@kernel.org \
--to=sashiko-bot@kernel.org \
--cc=intel-xe@lists.freedesktop.org \
--cc=piotr.piorkowski@intel.com \
--cc=sashiko-reviews@lists.linux.dev \
/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