From: Mika Kuoppala <mika.kuoppala@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 2/5] drm/i915: Do not access stolen memory directly by the CPU, even for error capture
Date: Fri, 15 Aug 2014 14:11:12 +0300 [thread overview]
Message-ID: <87r40it3cf.fsf@gaia.fi.intel.com> (raw)
In-Reply-To: <1407870351-6064-2-git-send-email-chris@chris-wilson.co.uk>
Chris Wilson <chris@chris-wilson.co.uk> writes:
> For stolen pages, since it is verboten to access them directly on many
> architectures, we have to read them through the GTT aperture. If they
> are not accessible through the aperture, then we have to abort.
>
> This was complicated by
>
> commit 8b6124a633d8095b0c8364f585edff9c59568a96
> Author: Chris Wilson <chris@chris-wilson.co.uk>
> Date: Thu Jan 30 14:38:16 2014 +0000
>
> drm/i915: Don't access snooped pages through the GTT (even for error capture)
>
> and the desire to use stolen memory for ringbuffers, contexts and
> batches in the future.
>
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> ---
> drivers/gpu/drm/i915/i915_gpu_error.c | 50 ++++++++++++++++++++++-------------
> 1 file changed, 31 insertions(+), 19 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
> index 35e70d5..6d280c07 100644
> --- a/drivers/gpu/drm/i915/i915_gpu_error.c
> +++ b/drivers/gpu/drm/i915/i915_gpu_error.c
> @@ -561,10 +561,11 @@ static struct drm_i915_error_object *
> i915_error_object_create_sized(struct drm_i915_private *dev_priv,
> struct drm_i915_gem_object *src,
> struct i915_address_space *vm,
> - const int num_pages)
> + int num_pages)
> {
> struct drm_i915_error_object *dst;
> - int i;
> + bool use_ggtt;
> + int i = 0;
> u32 reloc_offset;
>
> if (src == NULL || src->pages == NULL)
> @@ -574,8 +575,32 @@ i915_error_object_create_sized(struct drm_i915_private *dev_priv,
> if (dst == NULL)
> return NULL;
>
> - reloc_offset = dst->gtt_offset = i915_gem_obj_offset(src, vm);
> - for (i = 0; i < num_pages; i++) {
> + dst->gtt_offset = i915_gem_obj_offset(src, vm);
> +
> + reloc_offset = dst->gtt_offset;
> + use_ggtt = (src->cache_level == I915_CACHE_NONE &&
Take this cache level check out so that we end up doing the
snoopable check for non stolen objects too?
-Mika
> + i915_is_ggtt(vm) &&
> + src->has_global_gtt_mapping &&
> + reloc_offset + num_pages * PAGE_SIZE <= dev_priv->gtt.mappable_end);
> +
> + /* Cannot access stolen address directly, try to use the aperture */
> + if (src->stolen) {
> + use_ggtt = true;
> +
> + if (!src->has_global_gtt_mapping)
> + goto unwind;
> +
> + reloc_offset = i915_gem_obj_ggtt_offset(src);
> + if (reloc_offset + num_pages * PAGE_SIZE > dev_priv->gtt.mappable_end)
> + goto unwind;
> + }
> +
> + /* Cannot access snooped pages through the aperture */
> + if (use_ggtt && src->cache_level != I915_CACHE_NONE && !HAS_LLC(dev_priv->dev))
> + goto unwind;
> +
> + dst->page_count = num_pages;
> + while (num_pages--) {
> unsigned long flags;
> void *d;
>
> @@ -584,10 +609,7 @@ i915_error_object_create_sized(struct drm_i915_private *dev_priv,
> goto unwind;
>
> local_irq_save(flags);
> - if (src->cache_level == I915_CACHE_NONE &&
> - reloc_offset < dev_priv->gtt.mappable_end &&
> - src->has_global_gtt_mapping &&
> - i915_is_ggtt(vm)) {
> + if (use_ggtt) {
> void __iomem *s;
>
> /* Simply ignore tiling or any overlapping fence.
> @@ -599,14 +621,6 @@ i915_error_object_create_sized(struct drm_i915_private *dev_priv,
> reloc_offset);
> memcpy_fromio(d, s, PAGE_SIZE);
> io_mapping_unmap_atomic(s);
> - } else if (src->stolen) {
> - unsigned long offset;
> -
> - offset = dev_priv->mm.stolen_base;
> - offset += src->stolen->start;
> - offset += i << PAGE_SHIFT;
> -
> - memcpy_fromio(d, (void __iomem *) offset, PAGE_SIZE);
> } else {
> struct page *page;
> void *s;
> @@ -623,11 +637,9 @@ i915_error_object_create_sized(struct drm_i915_private *dev_priv,
> }
> local_irq_restore(flags);
>
> - dst->pages[i] = d;
> -
> + dst->pages[i++] = d;
> reloc_offset += PAGE_SIZE;
> }
> - dst->page_count = num_pages;
>
> return dst;
>
> --
> 1.9.1
>
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2014-08-15 11:11 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-08-12 19:05 [PATCH 1/5] drm/i915: Print captured bo for all VM in error state Chris Wilson
2014-08-12 19:05 ` [PATCH 2/5] drm/i915: Do not access stolen memory directly by the CPU, even for error capture Chris Wilson
2014-08-14 14:51 ` Mika Kuoppala
2014-08-14 19:35 ` Chris Wilson
2014-08-15 11:11 ` Mika Kuoppala [this message]
2014-08-15 18:07 ` Mika Kuoppala
2014-08-12 19:05 ` [PATCH 3/5] drm/i915: Remove num_pages parameter to i915_error_object_create() Chris Wilson
2014-08-15 18:07 ` Mika Kuoppala
2014-08-12 19:05 ` [PATCH 4/5] drm/i915: Suppress a WARN on reading an object back for a GPU hang Chris Wilson
2014-08-15 18:09 ` Mika Kuoppala
2014-08-25 21:27 ` Daniel Vetter
2014-08-12 19:05 ` [PATCH 5/5] drm/i915: s/seqno/request/ tracking inside objects Chris Wilson
2014-08-27 9:55 ` Daniel Vetter
2014-08-27 10:39 ` Chris Wilson
2014-09-02 10:06 ` John Harrison
2014-09-06 9:12 ` Chris Wilson
2014-08-13 14:50 ` [PATCH 1/5] drm/i915: Print captured bo for all VM in error state Mika Kuoppala
2014-08-14 6:50 ` Chris Wilson
2014-08-14 10:18 ` Mika Kuoppala
2014-08-14 15:03 ` Daniel Vetter
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=87r40it3cf.fsf@gaia.fi.intel.com \
--to=mika.kuoppala@linux.intel.com \
--cc=chris@chris-wilson.co.uk \
--cc=intel-gfx@lists.freedesktop.org \
/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.