All of lore.kernel.org
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Tejas Upadhyay" <tejas.upadhyay@intel.com>
Cc: intel-xe@lists.freedesktop.org
Subject: Re: [PATCH V19 07/15] drm/xe/vram: Extract buddy alloc and free helpers
Date: Mon, 31 Aug 2026 08:08:47 +0000	[thread overview]
Message-ID: <20260831080848.30F171F000E9@smtp.kernel.org> (raw)
In-Reply-To: <20260831064942.315720-24-tejas.upadhyay@intel.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Uninitialized output parameter `used_visible` in `xe_ttm_vram_buddy_alloc()`.
--

commit 9483caeb9aea9c04580bdd071e565c0158f680fa
Author: Tejas Upadhyay <tejas.upadhyay@intel.com>

drm/xe/vram: Extract buddy alloc and free helpers

This commit factors out block allocation and freeing into reusable helpers
xe_ttm_vram_buddy_alloc() and xe_ttm_vram_buddy_free(). These helpers
consolidate block allocation with visible-size tracking for upcoming VRAM
page offline reservation paths.

> diff --git a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
> index 51e983ee3bad4..16ecea4977809 100644
> --- a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
> +++ b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
[ ... ]
> +static int xe_ttm_vram_buddy_alloc(struct xe_ttm_vram_mgr *mgr, u64 start,
> +				   u64 end, u64 size, u64 min_page_size,
> +				   struct list_head *blocks, unsigned long flags,
> +				   void *priv, u64 *used_visible)
> +{
[ ... ]
> +	if (end <= mgr->visible_size) {
> +		*used_visible = size;
> +	} else {
> +		list_for_each_entry(block, blocks, link) {
> +			u64 blk_start = gpu_buddy_block_offset(block);
> +
> +			if (blk_start < mgr->visible_size) {
> +				u64 blk_end = blk_start + gpu_buddy_block_size(mm, block);
> +
> +				*used_visible += min(blk_end, mgr->visible_size) - blk_start;

[Severity: Medium]
Does this accumulate into an uninitialized variable?

In this else branch, *used_visible is incremented via += without being
explicitly zeroed first. If a caller does not explicitly initialize the
memory pointed to by used_visible before calling xe_ttm_vram_buddy_alloc(),
this will accumulate garbage values and corrupt the VRAM visible memory
accounting in mgr->visible_avail.

While current callers seem to pass pointers to zero-allocated memory,
could *used_visible be explicitly initialized to 0 before the loop to
prevent a trap for future callers?

> +			}
> +		}
> +	}
> +
> +	mgr->visible_avail -= *used_visible;
> +	return 0;
> +}

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260831064942.315720-17-tejas.upadhyay@intel.com?part=7

  reply	other threads:[~2026-08-31  8:08 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-08-31  6:49 [PATCH V19 00/15] Add memory page offlining support Tejas Upadhyay
2026-08-31  6:49 ` [PATCH V19 01/15] drm/xe: Link VRAM object with gpu buddy Tejas Upadhyay
2026-08-31  6:49 ` [PATCH V19 02/15] drm/xe: Link LRC BO and its execution Queue Tejas Upadhyay
2026-08-31  7:24   ` sashiko-bot
2026-08-31  6:49 ` [PATCH V19 03/15] drm/xe: Export xe_ttm_bo_purge() Tejas Upadhyay
2026-09-02  0:33   ` Matthew Brost
2026-08-31  6:49 ` [PATCH V19 04/15] drm/xe: Handle NULL resource and allow purging of VRAM pages Tejas Upadhyay
2026-08-31  6:52   ` Ghimiray, Himal Prasad
2026-08-31  7:42   ` sashiko-bot
2026-08-31  6:49 ` [PATCH V19 05/15] drm/xe/bo: Make xe_bo_is_user() public Tejas Upadhyay
2026-08-31  6:49 ` [PATCH V19 06/15] drm/xe: Guard teardown paths against purged BOs Tejas Upadhyay
2026-08-31  6:53   ` Ghimiray, Himal Prasad
2026-08-31  8:03   ` sashiko-bot
2026-09-01  5:05     ` Upadhyay, Tejas
2026-08-31  6:49 ` [PATCH V19 07/15] drm/xe/vram: Extract buddy alloc and free helpers Tejas Upadhyay
2026-08-31  8:08   ` sashiko-bot [this message]
2026-09-01  5:10     ` Upadhyay, Tejas
2026-08-31  6:49 ` [PATCH V19 08/15] drm/xe/vram: Add page offline data structures and lifecycle Tejas Upadhyay
2026-08-31  6:49 ` [PATCH V19 09/15] drm/xe/vram: Add VRAM page offline fault handler Tejas Upadhyay
2026-08-31  6:57   ` Ghimiray, Himal Prasad
2026-08-31  8:37   ` sashiko-bot
2026-09-01  5:48     ` Upadhyay, Tejas
2026-08-31  6:49 ` [PATCH V19 10/15] drm/xe/configfs: Add disable_vram_page_offline attribute Tejas Upadhyay
2026-08-31  8:45   ` sashiko-bot
2026-09-01  5:25     ` Upadhyay, Tejas
2026-08-31  6:49 ` [PATCH V19 11/15] drm/xe/ras: Cache disable_vram_page_offline policy at init Tejas Upadhyay
2026-08-31  9:02   ` sashiko-bot
2026-09-01  5:22     ` Upadhyay, Tejas
2026-08-31  6:49 ` [PATCH V19 12/15] drm/xe/vram: Check disable_vram_page_offline policy in fault handler Tejas Upadhyay
2026-08-31  6:49 ` [PATCH V19 13/15] drm/xe: Expose bad VRAM pages via debugfs Tejas Upadhyay
2026-08-31  6:55   ` Ghimiray, Himal Prasad
2026-08-31  9:18   ` sashiko-bot
2026-09-01  5:13     ` Upadhyay, Tejas
2026-08-31  6:49 ` [PATCH V19 14/15] drm/xe/uapi: Expose ban reason in EXEC_QUEUE_GET_PROPERTY_BAN Tejas Upadhyay
2026-08-31  9:40   ` sashiko-bot
2026-09-01  9:40     ` Upadhyay, Tejas
2026-08-31  6:49 ` [PATCH V19 15/15] drm/xe: Add fault-inject based VRAM page offline injection Tejas Upadhyay
2026-08-31  6:54   ` Ghimiray, Himal Prasad
2026-08-31 11:53 ` ✓ CI.KUnit: success for Add memory page offlining support (rev22) Patchwork
2026-08-31 13:28 ` ✓ Xe.CI.BAT: " Patchwork
2026-08-31 13:58 ` ✗ Xe.CI.FULL: failure " Patchwork
2026-08-31 14:04 ` [PATCH V19 00/15] Add memory page offlining support Rodrigo Vivi
2026-08-31 14:58   ` Matthew Brost
2026-09-01  4:09   ` Upadhyay, Tejas
2026-08-31 17:44 ` ✓ CI.KUnit: success for Add memory page offlining support (rev23) Patchwork
2026-08-31 18:32 ` ✓ Xe.CI.BAT: " Patchwork
2026-08-31 22:22 ` ✗ Xe.CI.FULL: failure " 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=20260831080848.30F171F000E9@smtp.kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=sashiko-reviews@lists.linux.dev \
    --cc=tejas.upadhyay@intel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.