Skip to main content

Browse the archive

Revision 832a3aad1e2927b1684e7369c9f36a370e0b95da authored by Chris Wilson on 18 March 2015, 18:19:22 UTC, committed by Jani Nikula on 26 March 2015, 09:05:54 UTC 
drm/i915: Keep ring->active_list and ring->requests_list consistent
 
If we retire requests last, we may use a later seqno and so clear
the requests lists without clearing the active list, leading to
confusion. Hence we should retire requests first for consistency with
the early return. The order used to be important as the lifecycle for
the object on the active list was determined by request->seqno. However,
the requests themselves are now reference counted removing the
constraint from the order of retirement.

Fixes regression from

commit 1b5a433a4dd967b125131da42b89b5cc0d5b1f57
Author: John Harrison <John.C.Harrison@Intel.com>
Date:   Mon Nov 24 18:49:42 2014 +0000

    drm/i915: Convert 'i915_seqno_passed' calls into 'i915_gem_request_completed
'

and a

	WARNING: CPU: 0 PID: 1383 at drivers/gpu/drm/i915/i915_gem_evict.c:279 i915_gem_evict_vm+0x10c/0x140()
	WARN_ON(!list_empty(&vm->active_list))

Identified by updating WATCH_LISTS:

	[drm:i915_verify_lists] *ERROR* blitter ring: active list not empty, but no requests
	WARNING: CPU: 0 PID: 681 at drivers/gpu/drm/i915/i915_gem.c:2751 i915_gem_retire_requests_ring+0x149/0x230()
	WARN_ON(i915_verify_lists(ring->dev))

Note that this is only a problem in evict_vm where the following happens
after a retire_request has cleaned out all requests, but not all active
bo:
- intel_ring_idle called from i915_gpu_idle notices that no requests are
  outstanding and immediately returns.
- i915_gem_retire_requests_ring called from i915_gem_retire_requests also
  immediately returns when there's no request, still leaving the bo on the
  active list.
- evict_vm hits the WARN_ON(!list_empty(&vm->active_list)) after evicting
  all active objects that there's still stuff left that shouldn't be
  there.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: John Harrison <John.C.Harrison@Intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
1 parent 59a58cb
Raw File
gfp-translate 
#!/bin/bash
# Translate the bits making up a GFP mask
# (c) 2009, Mel Gorman <mel@csn.ul.ie>
# Licensed under the terms of the GNU GPL License version 2
SOURCE=
GFPMASK=none

# Helper function to report failures and exit
die() {
	echo ERROR: $@
	if [ "$TMPFILE" != "" ]; then
		rm -f $TMPFILE
	fi
	exit -1
}

usage() {
	echo "usage: gfp-translate [-h] [ --source DIRECTORY ] gfpmask"
	exit 0
}

# Parse command-line arguments
while [ $# -gt 0 ]; do
	case $1 in
		--source)
			SOURCE=$2
			shift 2
			;;
		-h)
			usage
			;;
		--help)
			usage
			;;
		*)
			GFPMASK=$1
			shift
			;;
	esac
done

# Guess the kernel source directory if it's not set. Preference is in order of
# o current directory
# o /usr/src/linux
if [ "$SOURCE" = "" ]; then
	if [ -r "/usr/src/linux/Makefile" ]; then
		SOURCE=/usr/src/linux
	fi
	if [ -r "`pwd`/Makefile" ]; then
		SOURCE=`pwd`
	fi
fi

# Confirm that a source directory exists
if [ ! -r "$SOURCE/Makefile" ]; then
	die "Could not locate kernel source directory or it is invalid"
fi

# Confirm that a GFP mask has been specified
if [ "$GFPMASK" = "none" ]; then
	usage
fi

# Extract GFP flags from the kernel source
TMPFILE=`mktemp -t gfptranslate-XXXXXX` || exit 1
grep -q ___GFP $SOURCE/include/linux/gfp.h
if [ $? -eq 0 ]; then
	grep "^#define ___GFP" $SOURCE/include/linux/gfp.h | sed -e 's/u$//' | grep -v GFP_BITS > $TMPFILE
else
	grep "^#define __GFP" $SOURCE/include/linux/gfp.h | sed -e 's/(__force gfp_t)//' | sed -e 's/u)/)/' | grep -v GFP_BITS | sed -e 's/)\//) \//' > $TMPFILE
fi

# Parse the flags
IFS="
"
echo Source: $SOURCE
echo Parsing: $GFPMASK
for LINE in `cat $TMPFILE`; do
	MASK=`echo $LINE | awk '{print $3}'`
	if [ $(($GFPMASK&$MASK)) -ne 0 ]; then
		echo $LINE
	fi
done

rm -f $TMPFILE
exit 0
back to top