From patchwork Fri Nov 29 00:30:17 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Chris Wilson X-Patchwork-Id: 11266651 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 8FE2B138C for ; Fri, 29 Nov 2019 00:30:24 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 70977208E4 for ; Fri, 29 Nov 2019 00:30:24 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 70977208E4 Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=chris-wilson.co.uk Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=intel-gfx-bounces@lists.freedesktop.org Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id F0C166E0A1; Fri, 29 Nov 2019 00:30:22 +0000 (UTC) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from fireflyinternet.com (mail.fireflyinternet.com [109.228.58.192]) by gabe.freedesktop.org (Postfix) with ESMTPS id 80BC66E0A1 for ; Fri, 29 Nov 2019 00:30:21 +0000 (UTC) X-Default-Received-SPF: pass (skip=forwardok (res=PASS)) x-ip-name=78.156.65.138; Received: from haswell.alporthouse.com (unverified [78.156.65.138]) by fireflyinternet.com (Firefly Internet (M1)) with ESMTP id 19383623-1500050 for ; Fri, 29 Nov 2019 00:30:18 +0000 From: Chris Wilson To: intel-gfx@lists.freedesktop.org Date: Fri, 29 Nov 2019 00:30:17 +0000 Message-Id: <20191129003017.574505-1-chris@chris-wilson.co.uk> X-Mailer: git-send-email 2.24.0 MIME-Version: 1.0 Subject: [Intel-gfx] [CI] drm/i915/gen7: Re-enable full-ppgtt for ivb & hsw X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" After much hair pulling, resort to preallocating the ppGTT entries on init to circumvent the apparent lack of PD invalidate following the write to PP_DCLV upon switching mm between contexts (and here the same context after binding new objects). However, the details of that PP_DCLV invalidate are still unknown, and it appears we need to reload the mm twice to cover over a timing issue. Worrying. Fixes: 3dc007fe9b2b ("drm/i915/gtt: Downgrade gen7 (ivb, byt, hsw) back to aliasing-ppgtt") Signed-off-by: Chris Wilson --- .../gpu/drm/i915/gt/intel_ring_submission.c | 25 +++++++++---------- drivers/gpu/drm/i915/i915_gem_gtt.c | 14 +++++------ drivers/gpu/drm/i915/i915_pci.c | 2 +- 3 files changed, 20 insertions(+), 21 deletions(-) diff --git a/drivers/gpu/drm/i915/gt/intel_ring_submission.c b/drivers/gpu/drm/i915/gt/intel_ring_submission.c index f25ceccb335e..a3b3a4923e30 100644 --- a/drivers/gpu/drm/i915/gt/intel_ring_submission.c +++ b/drivers/gpu/drm/i915/gt/intel_ring_submission.c @@ -1366,13 +1366,21 @@ static int load_pd_dir(struct i915_request *rq, const struct i915_ppgtt *ppgtt) const struct intel_engine_cs * const engine = rq->engine; u32 *cs; - cs = intel_ring_begin(rq, 6); + cs = intel_ring_begin(rq, 12); if (IS_ERR(cs)) return PTR_ERR(cs); - *cs++ = MI_LOAD_REGISTER_IMM(1); + *cs++ = MI_LOAD_REGISTER_IMM(2); *cs++ = i915_mmio_reg_offset(RING_PP_DIR_DCLV(engine->mmio_base)); *cs++ = PP_DIR_DCLV_2G; + *cs++ = i915_mmio_reg_offset(RING_PP_DIR_DCLV(engine->mmio_base)) + 4; + *cs++ = 0; + + *cs++ = MI_STORE_REGISTER_MEM | MI_SRM_LRM_GLOBAL_GTT; + *cs++ = i915_mmio_reg_offset(RING_PP_DIR_DCLV(engine->mmio_base)); + *cs++ = intel_gt_scratch_offset(rq->engine->gt, + INTEL_GT_SCRATCH_FIELD_DEFAULT); + *cs++ = MI_NOOP; *cs++ = MI_LOAD_REGISTER_IMM(1); *cs++ = i915_mmio_reg_offset(RING_PP_DIR_BASE(engine->mmio_base)); @@ -1579,6 +1587,7 @@ static int switch_context(struct i915_request *rq) { struct intel_context *ce = rq->hw_context; struct i915_address_space *vm = vm_alias(ce); + u32 hw_flags = 0; int ret; GEM_BUG_ON(HAS_EXECLISTS(rq->i915)); @@ -1590,19 +1599,9 @@ static int switch_context(struct i915_request *rq) } if (ce->state) { - u32 hw_flags; - GEM_BUG_ON(rq->engine->id != RCS0); - /* - * The kernel context(s) is treated as pure scratch and is not - * expected to retain any state (as we sacrifice it during - * suspend and on resume it may be corrupted). This is ok, - * as nothing actually executes using the kernel context; it - * is purely used for flushing user contexts. - */ - hw_flags = 0; - if (i915_gem_context_is_kernel(rq->gem_context)) + if (!rq->engine->default_state) hw_flags = MI_RESTORE_INHIBIT; ret = mi_set_context(rq, hw_flags); diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 6239a9adbf14..fb72ceeeb2b3 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -1692,7 +1692,6 @@ static int gen6_alloc_va_range(struct i915_address_space *vm, intel_wakeref_t wakeref; u64 from = start; unsigned int pde; - bool flush = false; int ret = 0; wakeref = intel_runtime_pm_get(&vm->i915->runtime_pm); @@ -1717,11 +1716,6 @@ static int gen6_alloc_va_range(struct i915_address_space *vm, spin_lock(&pd->lock); if (pd->entry[pde] == &vm->scratch[1]) { pd->entry[pde] = pt; - if (i915_vma_is_bound(ppgtt->vma, - I915_VMA_GLOBAL_BIND)) { - gen6_write_pde(ppgtt, pde, pt); - flush = true; - } } else { alloc = pt; pt = pd->entry[pde]; @@ -1732,8 +1726,13 @@ static int gen6_alloc_va_range(struct i915_address_space *vm, } spin_unlock(&pd->lock); - if (flush) + if (i915_vma_is_bound(ppgtt->vma, I915_VMA_GLOBAL_BIND)) { + /* Rewrite them all! Anything less misses an invalidate. */ + gen6_for_all_pdes(pt, pd, pde) + gen6_write_pde(ppgtt, pde, pt); + ioread32(ppgtt->pd_addr + pde - 1); gen6_ggtt_invalidate(vm->gt->ggtt); + } goto out; @@ -1994,6 +1993,7 @@ static struct i915_ppgtt *gen6_ppgtt_create(struct drm_i915_private *i915) err_pd: kfree(ppgtt->base.pd); err_free: + mutex_destroy(&ppgtt->pin_mutex); kfree(ppgtt); return ERR_PTR(err); } diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c index da3e9b5752ac..583e0cd94a6a 100644 --- a/drivers/gpu/drm/i915/i915_pci.c +++ b/drivers/gpu/drm/i915/i915_pci.c @@ -436,7 +436,7 @@ static const struct intel_device_info intel_sandybridge_m_gt2_info = { .has_rc6 = 1, \ .has_rc6p = 1, \ .has_rps = true, \ - .ppgtt_type = INTEL_PPGTT_ALIASING, \ + .ppgtt_type = INTEL_PPGTT_FULL, \ .ppgtt_size = 31, \ IVB_PIPE_OFFSETS, \ IVB_CURSOR_OFFSETS, \