From patchwork Thu Jun 13 15:18:36 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Tvrtko Ursulin X-Patchwork-Id: 10992199 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id D9DAE76 for ; Thu, 13 Jun 2019 15:18:45 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id C74A1204FA for ; Thu, 13 Jun 2019 15:18:45 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id BB378205FC; Thu, 13 Jun 2019 15:18:45 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-5.2 required=2.0 tests=BAYES_00,MAILING_LIST_MULTI, RCVD_IN_DNSWL_MED autolearn=ham version=3.3.1 Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher DHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.wl.linuxfoundation.org (Postfix) with ESMTPS id 2901E204FA for ; Thu, 13 Jun 2019 15:18:45 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 8B7BF89B20; Thu, 13 Jun 2019 15:18:44 +0000 (UTC) X-Original-To: Intel-gfx@lists.freedesktop.org Delivered-To: Intel-gfx@lists.freedesktop.org Received: from mga01.intel.com (mga01.intel.com [192.55.52.88]) by gabe.freedesktop.org (Postfix) with ESMTPS id EE7CD89B20 for ; Thu, 13 Jun 2019 15:18:42 +0000 (UTC) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga004.jf.intel.com ([10.7.209.38]) by fmsmga101.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 13 Jun 2019 08:18:42 -0700 X-ExtLoop1: 1 Received: from yyosef-mobl.ger.corp.intel.com (HELO localhost.localdomain) ([10.252.23.42]) by orsmga004.jf.intel.com with ESMTP; 13 Jun 2019 08:18:40 -0700 From: Tvrtko Ursulin To: Intel-gfx@lists.freedesktop.org Date: Thu, 13 Jun 2019 16:18:36 +0100 Message-Id: <20190613151836.16093-1-tvrtko.ursulin@linux.intel.com> X-Mailer: git-send-email 2.20.1 In-Reply-To: <20190613133539.12620-1-tvrtko.ursulin@linux.intel.com> References: <20190613133539.12620-1-tvrtko.ursulin@linux.intel.com> MIME-Version: 1.0 Subject: [Intel-gfx] [RFC 22/28] drm/i915: Convert i915_gem_flush_ggtt_writes to intel_gt 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" X-Virus-Scanned: ClamAV using ClamSMTP From: Tvrtko Ursulin Having introduced struct intel_gt (named the anonymous structure in i915) we can start using it to compartmentalize our code better. It makes more sense logically to have the code internally like this and it will also help with future split between gt and display in i915. Signed-off-by: Tvrtko Ursulin --- drivers/gpu/drm/i915/gem/i915_gem_object.c | 5 +-- .../drm/i915/gem/selftests/i915_gem_mman.c | 3 +- drivers/gpu/drm/i915/gt/intel_gt.c | 41 +++++++++++++++++++ drivers/gpu/drm/i915/gt/intel_gt.h | 2 + drivers/gpu/drm/i915/i915_drv.h | 2 - drivers/gpu/drm/i915/i915_gem.c | 40 ------------------ drivers/gpu/drm/i915/i915_vma.c | 3 +- drivers/gpu/drm/i915/selftests/i915_gem_gtt.c | 2 +- 8 files changed, 50 insertions(+), 48 deletions(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c index 36b76c6a0a9d..9ae7743348f2 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c @@ -27,6 +27,7 @@ #include "i915_gem_context.h" #include "i915_gem_object.h" #include "i915_globals.h" +#include "gt/intel_gt.h" #include "intel_frontbuffer.h" static struct i915_global_object { @@ -367,7 +368,6 @@ void i915_gem_object_flush_write_domain(struct drm_i915_gem_object *obj, unsigned int flush_domains) { - struct drm_i915_private *dev_priv = to_i915(obj->base.dev); struct i915_vma *vma; assert_object_held(obj); @@ -377,8 +377,6 @@ i915_gem_object_flush_write_domain(struct drm_i915_gem_object *obj, switch (obj->write_domain) { case I915_GEM_DOMAIN_GTT: - i915_gem_flush_ggtt_writes(dev_priv); - intel_fb_obj_flush(obj, fb_write_origin(obj, I915_GEM_DOMAIN_GTT)); @@ -386,6 +384,7 @@ i915_gem_object_flush_write_domain(struct drm_i915_gem_object *obj, if (vma->iomap) continue; + intel_gt_flush_ggtt_writes(vma->vm->gt); i915_vma_unset_ggtt_write(vma); } break; diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c index b92809418729..b46d57967bfa 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c @@ -6,6 +6,7 @@ #include +#include "gt/intel_gt.h" #include "gt/intel_gt_pm.h" #include "huge_gem_object.h" #include "i915_selftest.h" @@ -143,7 +144,7 @@ static int check_partial_mapping(struct drm_i915_gem_object *obj, if (offset >= obj->base.size) continue; - i915_gem_flush_ggtt_writes(to_i915(obj->base.dev)); + intel_gt_flush_ggtt_writes(&to_i915(obj->base.dev)->gt); p = i915_gem_object_get_page(obj, offset >> PAGE_SHIFT); cpu = kmap(p) + offset_in_page(offset); diff --git a/drivers/gpu/drm/i915/gt/intel_gt.c b/drivers/gpu/drm/i915/gt/intel_gt.c index c6a67393ee72..7bf01365573a 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt.c +++ b/drivers/gpu/drm/i915/gt/intel_gt.c @@ -188,3 +188,44 @@ void intel_gt_init_swizzling(struct intel_gt *gt) else MISSING_CASE(INTEL_GEN(i915)); } + +void intel_gt_flush_ggtt_writes(struct intel_gt *gt) +{ + struct drm_i915_private *i915 = gt->i915; + intel_wakeref_t wakeref; + + /* + * No actual flushing is required for the GTT write domain for reads + * from the GTT domain. Writes to it "immediately" go to main memory + * as far as we know, so there's no chipset flush. It also doesn't + * land in the GPU render cache. + * + * However, we do have to enforce the order so that all writes through + * the GTT land before any writes to the device, such as updates to + * the GATT itself. + * + * We also have to wait a bit for the writes to land from the GTT. + * An uncached read (i.e. mmio) seems to be ideal for the round-trip + * timing. This issue has only been observed when switching quickly + * between GTT writes and CPU reads from inside the kernel on recent hw, + * and it appears to only affect discrete GTT blocks (i.e. on LLC + * system agents we cannot reproduce this behaviour, until Cannonlake + * that was!). + */ + + wmb(); + + if (INTEL_INFO(i915)->has_coherent_ggtt) + return; + + i915_gem_chipset_flush(i915); + + with_intel_runtime_pm(i915, wakeref) { + struct intel_uncore *uncore = gt->uncore; + + spin_lock_irq(&uncore->lock); + intel_uncore_posting_read_fw(uncore, + RING_HEAD(RENDER_RING_BASE)); + spin_unlock_irq(&uncore->lock); + } +} diff --git a/drivers/gpu/drm/i915/gt/intel_gt.h b/drivers/gpu/drm/i915/gt/intel_gt.h index e026b2dc1115..700bb56d1e9b 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt.h +++ b/drivers/gpu/drm/i915/gt/intel_gt.h @@ -20,4 +20,6 @@ void intel_gt_clear_error_registers(struct intel_gt *gt, void intel_gt_init_swizzling(struct intel_gt *gt); +void intel_gt_flush_ggtt_writes(struct intel_gt *gt); + #endif /* __INTEL_GT_H__ */ diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 1eb203fdee60..4987a048b3d3 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -2648,8 +2648,6 @@ int __must_check i915_gem_evict_for_node(struct i915_address_space *vm, unsigned int flags); int i915_gem_evict_vm(struct i915_address_space *vm); -void i915_gem_flush_ggtt_writes(struct drm_i915_private *dev_priv); - /* belongs in i915_gem_gtt.h */ static inline void i915_gem_chipset_flush(struct drm_i915_private *dev_priv) { diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index b7f88e2bd7df..4f9aac62a8a4 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -233,46 +233,6 @@ i915_gem_create_ioctl(struct drm_device *dev, void *data, &args->size, &args->handle); } -void i915_gem_flush_ggtt_writes(struct drm_i915_private *dev_priv) -{ - intel_wakeref_t wakeref; - - /* - * No actual flushing is required for the GTT write domain for reads - * from the GTT domain. Writes to it "immediately" go to main memory - * as far as we know, so there's no chipset flush. It also doesn't - * land in the GPU render cache. - * - * However, we do have to enforce the order so that all writes through - * the GTT land before any writes to the device, such as updates to - * the GATT itself. - * - * We also have to wait a bit for the writes to land from the GTT. - * An uncached read (i.e. mmio) seems to be ideal for the round-trip - * timing. This issue has only been observed when switching quickly - * between GTT writes and CPU reads from inside the kernel on recent hw, - * and it appears to only affect discrete GTT blocks (i.e. on LLC - * system agents we cannot reproduce this behaviour, until Cannonlake - * that was!). - */ - - wmb(); - - if (INTEL_INFO(dev_priv)->has_coherent_ggtt) - return; - - i915_gem_chipset_flush(dev_priv); - - with_intel_runtime_pm(dev_priv, wakeref) { - struct intel_uncore *uncore = &dev_priv->uncore; - - spin_lock_irq(&uncore->lock); - intel_uncore_posting_read_fw(uncore, - RING_HEAD(RENDER_RING_BASE)); - spin_unlock_irq(&uncore->lock); - } -} - static int shmem_pread(struct page *page, int offset, int len, char __user *user_data, bool needs_clflush) diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c index cb341e4acf99..4f4695c351af 100644 --- a/drivers/gpu/drm/i915/i915_vma.c +++ b/drivers/gpu/drm/i915/i915_vma.c @@ -23,6 +23,7 @@ */ #include "gt/intel_engine.h" +#include "gt/intel_gt.h" #include "i915_vma.h" @@ -408,7 +409,7 @@ void i915_vma_flush_writes(struct i915_vma *vma) if (!i915_vma_has_ggtt_write(vma)) return; - i915_gem_flush_ggtt_writes(vma->vm->i915); + intel_gt_flush_ggtt_writes(vma->vm->gt); i915_vma_unset_ggtt_write(vma); } diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c index 2093d08a7569..a67f0e9b4d5f 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c @@ -1195,7 +1195,7 @@ static int igt_ggtt_page(void *arg) iowrite32(n, vaddr + n); io_mapping_unmap_atomic(vaddr); } - i915_gem_flush_ggtt_writes(i915); + intel_gt_flush_ggtt_writes(ggtt->vm.gt); i915_random_reorder(order, count, &prng); for (n = 0; n < count; n++) {