From: Mika Kuoppala <mika.kuoppala@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Cc: matthew.auld@intel.com
Subject: Re: [PATCH 2/5] drm/i915/gtt: Add some range asserts
Date: Wed, 21 Aug 2019 19:24:05 +0300 [thread overview]
Message-ID: <875zmqv656.fsf@gaia.fi.intel.com> (raw)
In-Reply-To: <20190821155728.2839-2-chris@chris-wilson.co.uk>
Chris Wilson <chris@chris-wilson.co.uk> writes:
> These should have been validated in the upper layers, but for sanity's
> sake, repeat them.
>
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> ---
> drivers/gpu/drm/i915/i915_gem_gtt.c | 16 +++++++++++-----
> 1 file changed, 11 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
> index b06d1d9054ba..0b81e0b64393 100644
> --- a/drivers/gpu/drm/i915/i915_gem_gtt.c
> +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
> @@ -965,8 +965,10 @@ static u64 __gen8_ppgtt_clear(struct i915_address_space * const vm,
> const struct i915_page_scratch * const scratch = &vm->scratch[lvl];
> unsigned int idx, len;
>
> + GEM_BUG_ON(end > vm->total >> GEN8_PTE_SHIFT);
I am having thoughts to clean all underlying handling to
use page indexes consistently...
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
> +
> len = gen8_pd_range(start, end, lvl--, &idx);
> - DBG("%s(%p):{ lvl:%d, start:%llx, end:%llx, idx:%d, len:%d, used:%d}\n",
> + DBG("%s(%p):{ lvl:%d, start:%llx, end:%llx, idx:%d, len:%d, used:%d }\n",
> __func__, vm, lvl + 1, start, end,
> idx, len, atomic_read(px_used(pd)));
> GEM_BUG_ON(!len || len >= atomic_read(px_used(pd)));
> @@ -992,7 +994,7 @@ static u64 __gen8_ppgtt_clear(struct i915_address_space * const vm,
> u64 *vaddr;
>
> count = gen8_pt_count(start, end);
> - DBG("%s(%p):{ lvl:%d, start:%llx, end:%llx, idx:%d, len:%d, used:%d} removing pte\n",
> + DBG("%s(%p):{ lvl:%d, start:%llx, end:%llx, idx:%d, len:%d, used:%d } removing pte\n",
> __func__, vm, lvl, start, end,
> gen8_pd_index(start, 0), count,
> atomic_read(&pt->used));
> @@ -1020,6 +1022,7 @@ static void gen8_ppgtt_clear(struct i915_address_space *vm,
> {
> GEM_BUG_ON(!IS_ALIGNED(start, BIT_ULL(GEN8_PTE_SHIFT)));
> GEM_BUG_ON(!IS_ALIGNED(length, BIT_ULL(GEN8_PTE_SHIFT)));
> + GEM_BUG_ON(range_overflows(start, length, vm->total));
>
> start >>= GEN8_PTE_SHIFT;
> length >>= GEN8_PTE_SHIFT;
> @@ -1031,15 +1034,17 @@ static void gen8_ppgtt_clear(struct i915_address_space *vm,
>
> static int __gen8_ppgtt_alloc(struct i915_address_space * const vm,
> struct i915_page_directory * const pd,
> - u64 * const start, u64 end, int lvl)
> + u64 * const start, const u64 end, int lvl)
> {
> const struct i915_page_scratch * const scratch = &vm->scratch[lvl];
> struct i915_page_table *alloc = NULL;
> unsigned int idx, len;
> int ret = 0;
>
> + GEM_BUG_ON(end > vm->total >> GEN8_PTE_SHIFT);
> +
> len = gen8_pd_range(*start, end, lvl--, &idx);
> - DBG("%s(%p):{lvl:%d, start:%llx, end:%llx, idx:%d, len:%d, used:%d}\n",
> + DBG("%s(%p):{ lvl:%d, start:%llx, end:%llx, idx:%d, len:%d, used:%d }\n",
> __func__, vm, lvl + 1, *start, end,
> idx, len, atomic_read(px_used(pd)));
> GEM_BUG_ON(!len || (idx + len - 1) >> gen8_pd_shift(1));
> @@ -1105,7 +1110,7 @@ static int __gen8_ppgtt_alloc(struct i915_address_space * const vm,
> } else {
> unsigned int count = gen8_pt_count(*start, end);
>
> - DBG("%s(%p):{lvl:%d, start:%llx, end:%llx, idx:%d, len:%d, used:%d} inserting pte\n",
> + DBG("%s(%p):{ lvl:%d, start:%llx, end:%llx, idx:%d, len:%d, used:%d } inserting pte\n",
> __func__, vm, lvl, *start, end,
> gen8_pd_index(*start, 0), count,
> atomic_read(&pt->used));
> @@ -1131,6 +1136,7 @@ static int gen8_ppgtt_alloc(struct i915_address_space *vm,
>
> GEM_BUG_ON(!IS_ALIGNED(start, BIT_ULL(GEN8_PTE_SHIFT)));
> GEM_BUG_ON(!IS_ALIGNED(length, BIT_ULL(GEN8_PTE_SHIFT)));
> + GEM_BUG_ON(range_overflows(start, length, vm->total));
>
> start >>= GEN8_PTE_SHIFT;
> length >>= GEN8_PTE_SHIFT;
> --
> 2.23.0.rc1
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2019-08-21 16:24 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-08-21 15:57 [PATCH 1/5] drm/i915/execlists: Set priority hint prior to submission Chris Wilson
2019-08-21 15:57 ` [PATCH 2/5] drm/i915/gtt: Add some range asserts Chris Wilson
2019-08-21 16:24 ` Mika Kuoppala [this message]
2019-08-21 16:53 ` Chris Wilson
2019-08-21 15:57 ` [PATCH 3/5] drm/i915: Track ggtt fence reservations under its own mutex Chris Wilson
2019-08-21 20:42 ` Matthew Auld
2019-08-21 15:57 ` [PATCH 4/5] drm/i915: Pull obj->userfault tracking under the ggtt->mutex Chris Wilson
2019-08-21 20:44 ` Matthew Auld
2019-08-21 15:57 ` [PATCH 5/5] drm/i915: Replace i915_vma_put_fence() Chris Wilson
2019-08-21 21:05 ` Matthew Auld
2019-08-21 21:27 ` Chris Wilson
2019-08-21 16:18 ` [PATCH 1/5] drm/i915/execlists: Set priority hint prior to submission Mika Kuoppala
2019-08-21 18:04 ` ✗ Fi.CI.BAT: failure for series starting with [1/5] " 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=875zmqv656.fsf@gaia.fi.intel.com \
--to=mika.kuoppala@linux.intel.com \
--cc=chris@chris-wilson.co.uk \
--cc=intel-gfx@lists.freedesktop.org \
--cc=matthew.auld@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.