From patchwork Thu Nov 24 04:01:32 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Kumar, Mahesh" X-Patchwork-Id: 9444735 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id 879E66071C for ; Thu, 24 Nov 2016 04:00:33 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 795EC27E15 for ; Thu, 24 Nov 2016 04:00:33 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 6E78227E5A; Thu, 24 Nov 2016 04:00:33 +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=-4.2 required=2.0 tests=BAYES_00, 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 E2C0F27E15 for ; Thu, 24 Nov 2016 04:00:32 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id A54F66E29C; Thu, 24 Nov 2016 04:00:31 +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 D3EC46E29C for ; Thu, 24 Nov 2016 04:00:29 +0000 (UTC) Received: from orsmga005.jf.intel.com ([10.7.209.41]) by fmsmga101.fm.intel.com with ESMTP; 23 Nov 2016 20:00:29 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.31,690,1473145200"; d="scan'208";a="34969845" Received: from kumarmah-desk.iind.intel.com ([10.223.26.44]) by orsmga005.jf.intel.com with ESMTP; 23 Nov 2016 20:00:27 -0800 From: Mahesh Kumar To: intel-gfx@lists.freedesktop.org Date: Thu, 24 Nov 2016 09:31:32 +0530 Message-Id: <20161124040135.5517-6-mahesh1.kumar@intel.com> X-Mailer: git-send-email 2.10.1 In-Reply-To: <20161124040135.5517-1-mahesh1.kumar@intel.com> References: <20161124040135.5517-1-mahesh1.kumar@intel.com> Cc: paulo.r.zanoni@intel.com, maarten.lankhorst@intel.com Subject: [Intel-gfx] [PATCH v6 5/8] drm/i915/skl+: change WM calc to fixed point 16.16 X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" X-Virus-Scanned: ClamAV using ClamSMTP This patch changes Watermak calculation to fixed point calculation. Problem with current calculation is during plane_blocks_per_line calculation we divide intermediate blocks with min_scanlines and takes floor of the result because of integer operation. hence we end-up assigning less blocks than required. Which leads to flickers. Changes since V1: - Add fixed point data type as per Paulo's review Changes since V1: - use fixed_point instead of fp_16_16 Signed-off-by: Mahesh Kumar --- drivers/gpu/drm/i915/i915_drv.h | 84 +++++++++++++++++++++++++++++++++++++++++ drivers/gpu/drm/i915/intel_pm.c | 69 +++++++++++++++++++-------------- 2 files changed, 124 insertions(+), 29 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 6691a4e..e332f04 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -119,6 +119,90 @@ bool __i915_inject_load_failure(const char *func, int line); #define i915_inject_load_failure() \ __i915_inject_load_failure(__func__, __LINE__) +typedef struct { + uint32_t val; +} uint_fixed_16_16_t; + +#define FP_16_16_MAX ({ \ + uint_fixed_16_16_t fp; \ + fp.val = UINT_MAX; \ + fp; \ +}) + +static inline uint_fixed_16_16_t u32_to_fixed_point(uint32_t val) +{ + uint_fixed_16_16_t fp; + + WARN_ON(val >> 16); + + fp.val = val << 16; + return fp; +} + +static inline uint32_t fixed_point_to_u32_round_up(uint_fixed_16_16_t fp) +{ + return DIV_ROUND_UP(fp.val, 1 << 16); +} + +static inline uint32_t fixed_point_to_u32(uint_fixed_16_16_t fp) +{ + return (fp.val / (1 << 16)); +} + +static inline uint_fixed_16_16_t min_fixed_point(uint_fixed_16_16_t min1, + uint_fixed_16_16_t min2) +{ + uint_fixed_16_16_t min; + + min.val = min(min1.val, min2.val); + return min; +} + +static inline uint_fixed_16_16_t max_fixed_point(uint_fixed_16_16_t max1, + uint_fixed_16_16_t max2) +{ + uint_fixed_16_16_t max; + + max.val = max(max1.val, max2.val); + return max; +} + +static inline uint_fixed_16_16_t fixed_point_div_round_up(uint32_t val, + uint32_t d) +{ + uint_fixed_16_16_t fp, res; + + fp = u32_to_fixed_point(val); + res.val = DIV_ROUND_UP(fp.val, d); + return res; +} + +static inline uint_fixed_16_16_t fixed_point_div_round_up_u64(uint32_t val, + uint32_t d) +{ + uint_fixed_16_16_t res; + uint64_t interm_val; + + interm_val = (uint64_t)val << 16; + interm_val = DIV_ROUND_UP_ULL(interm_val, d); + WARN_ON(interm_val >> 32); + res.val = (uint32_t) interm_val; + + return res; +} + +static inline uint_fixed_16_16_t mul_fixed_point_u32(uint32_t val, + uint_fixed_16_16_t mul) +{ + uint64_t intermediate_val; + uint_fixed_16_16_t fp; + + intermediate_val = (uint64_t) val * mul.val; + WARN_ON(intermediate_val >> 32); + fp.val = (uint32_t) intermediate_val; + return fp; +} + static inline const char *yesno(bool v) { return v ? "yes" : "no"; diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index df6d231..3e2dd8f 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -3516,32 +3516,35 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate, * should allow pixel_rate up to ~2 GHz which seems sufficient since max * 2xcdclk is 1350 MHz and the pixel rate should never exceed that. */ -static uint32_t skl_wm_method1(uint32_t pixel_rate, uint8_t cpp, uint32_t latency) +static uint_fixed_16_16_t skl_wm_method1(uint32_t pixel_rate, uint8_t cpp, + uint32_t latency) { - uint32_t wm_intermediate_val, ret; + uint32_t wm_intermediate_val; + uint_fixed_16_16_t ret; if (latency == 0) - return UINT_MAX; - - wm_intermediate_val = latency * pixel_rate * cpp / 512; - ret = DIV_ROUND_UP(wm_intermediate_val, 1000); + return FP_16_16_MAX; + wm_intermediate_val = latency * pixel_rate * cpp; + ret = fixed_point_div_round_up_u64(wm_intermediate_val, 1000 * 512); return ret; } -static uint32_t skl_wm_method2(uint32_t pixel_rate, uint32_t pipe_htotal, - uint32_t latency, uint32_t plane_blocks_per_line) +static uint_fixed_16_16_t skl_wm_method2(uint32_t pixel_rate, + uint32_t pipe_htotal, + uint32_t latency, + uint_fixed_16_16_t plane_blocks_per_line) { - uint32_t ret; uint32_t wm_intermediate_val; + uint_fixed_16_16_t ret; if (latency == 0) - return UINT_MAX; + return FP_16_16_MAX; wm_intermediate_val = latency * pixel_rate; - ret = DIV_ROUND_UP(wm_intermediate_val, pipe_htotal * 1000) * - plane_blocks_per_line; - + wm_intermediate_val = DIV_ROUND_UP(wm_intermediate_val, + pipe_htotal * 1000); + ret = mul_fixed_point_u32(wm_intermediate_val, plane_blocks_per_line); return ret; } @@ -3581,14 +3584,17 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, struct drm_plane_state *pstate = &intel_pstate->base; struct drm_framebuffer *fb = pstate->fb; uint32_t latency = dev_priv->wm.skl_latency[level]; - uint32_t method1, method2; - uint32_t plane_bytes_per_line, plane_blocks_per_line; + uint_fixed_16_16_t method1, method2; + uint_fixed_16_16_t plane_blocks_per_line; + uint_fixed_16_16_t selected_result; + uint32_t interm_pbpl; + uint32_t plane_bytes_per_line; uint32_t res_blocks, res_lines; - uint32_t selected_result; uint8_t cpp; uint32_t width = 0, height = 0; uint32_t plane_pixel_rate; - uint32_t y_tile_minimum, y_min_scanlines; + uint_fixed_16_16_t y_tile_minimum; + uint32_t y_min_scanlines; struct intel_atomic_state *state = to_intel_atomic_state(cstate->base.state); bool apply_memory_bw_wa = skl_needs_memory_bw_wa(state); @@ -3647,14 +3653,16 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, plane_bytes_per_line = width * cpp; if (y_tiled) { + interm_pbpl = DIV_ROUND_UP(plane_bytes_per_line * + y_min_scanlines, 512); plane_blocks_per_line = - DIV_ROUND_UP(plane_bytes_per_line * y_min_scanlines, 512); - plane_blocks_per_line /= y_min_scanlines; + fixed_point_div_round_up(interm_pbpl, y_min_scanlines); } else if (x_tiled) { - plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512); + interm_pbpl = DIV_ROUND_UP(plane_bytes_per_line, 512); + plane_blocks_per_line = u32_to_fixed_point(interm_pbpl); } else { - plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512) - + 1; + interm_pbpl = DIV_ROUND_UP(plane_bytes_per_line, 512) + 1; + plane_blocks_per_line = u32_to_fixed_point(interm_pbpl); } method1 = skl_wm_method1(plane_pixel_rate, cpp, latency); @@ -3663,26 +3671,29 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, latency, plane_blocks_per_line); - y_tile_minimum = plane_blocks_per_line * y_min_scanlines; + y_tile_minimum = mul_fixed_point_u32(y_min_scanlines, + plane_blocks_per_line); if (y_tiled) { - selected_result = max(method2, y_tile_minimum); + selected_result = max_fixed_point(method2, y_tile_minimum); } else { if ((cpp * cstate->base.adjusted_mode.crtc_htotal / 512 < 1) && (plane_bytes_per_line / 512 < 1)) selected_result = method2; - else if ((ddb_allocation / plane_blocks_per_line) >= 1) - selected_result = min(method1, method2); + else if ((ddb_allocation / + fixed_point_to_u32_round_up(plane_blocks_per_line)) >= 1) + selected_result = min_fixed_point(method1, method2); else selected_result = method1; } - res_blocks = selected_result + 1; - res_lines = DIV_ROUND_UP(selected_result, plane_blocks_per_line); + res_blocks = fixed_point_to_u32_round_up(selected_result) + 1; + res_lines = DIV_ROUND_UP(selected_result.val, + plane_blocks_per_line.val); if (level >= 1 && level <= 7) { if (y_tiled) { - res_blocks += y_tile_minimum; + res_blocks += fixed_point_to_u32_round_up(y_tile_minimum); res_lines += y_min_scanlines; } else { res_blocks++;