From patchwork Tue Sep 11 05:24:49 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Hui Wang X-Patchwork-Id: 968415 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Authentication-Results: ozlabs.org; spf=none (mailfrom) smtp.mailfrom=lists.ubuntu.com (client-ip=91.189.94.19; helo=huckleberry.canonical.com; envelope-from=kernel-team-bounces@lists.ubuntu.com; receiver=) Authentication-Results: ozlabs.org; dmarc=fail (p=none dis=none) header.from=canonical.com Received: from huckleberry.canonical.com (huckleberry.canonical.com [91.189.94.19]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ozlabs.org (Postfix) with ESMTPS id 428YJk6SFKz9s4s; Tue, 11 Sep 2018 15:25:10 +1000 (AEST) Received: from localhost ([127.0.0.1] helo=huckleberry.canonical.com) by huckleberry.canonical.com with esmtp (Exim 4.86_2) (envelope-from ) id 1fzbAe-0005vq-7L; Tue, 11 Sep 2018 05:25:04 +0000 Received: from youngberry.canonical.com ([91.189.89.112]) by huckleberry.canonical.com with esmtps (TLS1.0:DHE_RSA_AES_128_CBC_SHA1:128) (Exim 4.86_2) (envelope-from ) id 1fzbAb-0005ua-Af for kernel-team@lists.ubuntu.com; Tue, 11 Sep 2018 05:25:01 +0000 Received: from [125.35.49.90] (helo=hwang4-Lenovo-V480c.bluefin) by youngberry.canonical.com with esmtpsa (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.76) (envelope-from ) id 1fzbAa-0003kH-1g for kernel-team@lists.ubuntu.com; Tue, 11 Sep 2018 05:25:01 +0000 From: Hui Wang To: kernel-team@lists.ubuntu.com Subject: [SRU][Bionic][B-OEM][PATCH 1/1] drm/nouveau/gr/gf100-: virtualise tpc_mask + apply fixes from traces Date: Tue, 11 Sep 2018 13:24:49 +0800 Message-Id: <1536643489-6238-2-git-send-email-hui.wang@canonical.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1536643489-6238-1-git-send-email-hui.wang@canonical.com> References: <1536643489-6238-1-git-send-email-hui.wang@canonical.com> X-BeenThere: kernel-team@lists.ubuntu.com X-Mailman-Version: 2.1.20 Precedence: list List-Id: Kernel team discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Errors-To: kernel-team-bounces@lists.ubuntu.com Sender: "kernel-team" From: Ben Skeggs BugLink: http://bugs.launchpad.net/bugs/1791569 We weren't placing higher TPC IDs in the right place on some configurations. [Due to the context difference, the ctxgm200.c and ctxgp100.c are changed a bit against the original patch, after this change, they have the same logic as the original patch. -- Hui's comment] Signed-off-by: Ben Skeggs (backported from commit fc36076441bae141893bd79899d19aa1b5fdf524) Signed-off-by: Hui Wang Acked-by: Stefan Bader Acked-by: Kleber Sacilotto de Souza --- drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c | 6 ++++++ drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h | 12 +++++++----- drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm200.c | 17 ++++++++++++++--- drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm20b.c | 2 +- drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp100.c | 17 +++++++++-------- drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp102.c | 2 ++ drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp107.c | 2 ++ drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h | 2 ++ drivers/gpu/drm/nouveau/nvkm/engine/gr/gm200.c | 1 + drivers/gpu/drm/nouveau/nvkm/engine/gr/gp100.c | 2 ++ drivers/gpu/drm/nouveau/nvkm/engine/gr/gp102.c | 2 ++ drivers/gpu/drm/nouveau/nvkm/engine/gr/gp107.c | 2 ++ drivers/gpu/drm/nouveau/nvkm/engine/gr/gp10b.c | 2 ++ 13 files changed, 52 insertions(+), 17 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c index 8810150..1540fde 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c @@ -1200,6 +1200,7 @@ void gf100_grctx_generate_r406800(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; + const struct gf100_grctx_func *func = gr->func->grctx; u64 tpc_mask = 0, tpc_set = 0; u8 tpcnr[GPC_MAX]; int gpc, tpc; @@ -1228,6 +1229,11 @@ gf100_grctx_generate_r406800(struct gf100_gr *gr) nvkm_wr32(device, 0x406c04 + (i * 0x20), upper_32_bits(tpc_set ^ tpc_mask)); } } + + if (func->tpc_mask) + func->tpc_mask(gr); + if (func->smid_config) + func->smid_config(gr); } void diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h index 5199e5a..74a4bd1 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h @@ -48,6 +48,8 @@ struct gf100_grctx_func { u32 attrib_nr; u32 alpha_nr_max; u32 alpha_nr; + void (*tpc_mask)(struct gf100_gr *); + void (*smid_config)(struct gf100_gr *); }; extern const struct gf100_grctx_func gf100_grctx; @@ -83,10 +85,6 @@ void gk104_grctx_generate_pagepool(struct gf100_grctx *); void gk104_grctx_generate_unkn(struct gf100_gr *); void gk104_grctx_generate_r418bb8(struct gf100_gr *); -void gm107_grctx_generate_bundle(struct gf100_grctx *); -void gm107_grctx_generate_pagepool(struct gf100_grctx *); -void gm107_grctx_generate_attrib(struct gf100_grctx *); - extern const struct gf100_grctx_func gk110_grctx; extern const struct gf100_grctx_func gk110b_grctx; extern const struct gf100_grctx_func gk208_grctx; @@ -95,16 +93,20 @@ extern const struct gf100_grctx_func gm107_grctx; void gm107_grctx_generate_bundle(struct gf100_grctx *); void gm107_grctx_generate_pagepool(struct gf100_grctx *); void gm107_grctx_generate_attrib(struct gf100_grctx *); +void gm107_grctx_generate_sm_id(struct gf100_gr *, int, int, int); extern const struct gf100_grctx_func gm200_grctx; + void gm200_grctx_generate_tpcid(struct gf100_gr *); -void gm200_grctx_generate_405b60(struct gf100_gr *); +void gm200_grctx_generate_tpc_mask(struct gf100_gr *); +void gm200_grctx_generate_smid_config(struct gf100_gr *); extern const struct gf100_grctx_func gm20b_grctx; extern const struct gf100_grctx_func gp100_grctx; void gp100_grctx_generate_main(struct gf100_gr *, struct gf100_grctx *); void gp100_grctx_generate_pagepool(struct gf100_grctx *); +void gp100_grctx_generate_smid_config(struct gf100_gr *); extern const struct gf100_grctx_func gp102_grctx; void gp102_grctx_generate_attrib(struct gf100_grctx *); diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm200.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm200.c index db209d3..93fae26 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm200.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm200.c @@ -46,7 +46,7 @@ gm200_grctx_generate_tpcid(struct gf100_gr *gr) } void -gm200_grctx_generate_405b60(struct gf100_gr *gr) +gm200_grctx_generate_smid_config(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; const u32 dist_nr = DIV_ROUND_UP(gr->tpc_total, 4); @@ -77,6 +77,15 @@ gm200_grctx_generate_405b60(struct gf100_gr *gr) nvkm_wr32(device, 0x405ba0 + (i * 4), gpcs[i]); } +void +gm200_grctx_generate_tpc_mask(struct gf100_gr *gr) +{ + u32 tmp, i; + for (tmp = 0, i = 0; i < gr->gpc_nr; i++) + tmp |= ((1 << gr->tpc_nr[i]) - 1) << (i * gr->func->tpc_nr); + nvkm_wr32(gr->base.engine.subdev.device, 0x4041c4, tmp); +} + static void gm200_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info) { @@ -105,10 +114,10 @@ gm200_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info) nvkm_wr32(device, 0x405b00, (gr->tpc_total << 8) | gr->gpc_nr); for (tmp = 0, i = 0; i < gr->gpc_nr; i++) - tmp |= ((1 << gr->tpc_nr[i]) - 1) << (i * 4); + tmp |= ((1 << gr->tpc_nr[i]) - 1) << (i * gr->func->tpc_nr); nvkm_wr32(device, 0x4041c4, tmp); - gm200_grctx_generate_405b60(gr); + gm200_grctx_generate_smid_config(gr); gf100_gr_icmd(gr, gr->fuc_bundle); nvkm_wr32(device, 0x404154, idle_timeout); @@ -133,4 +142,6 @@ gm200_grctx = { .attrib_nr = 0x400, .alpha_nr_max = 0x1800, .alpha_nr = 0x1000, + .tpc_mask = gm200_grctx_generate_tpc_mask, + .smid_config = gm200_grctx_generate_smid_config, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm20b.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm20b.c index e5702e3..26b2866 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm20b.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm20b.c @@ -68,7 +68,7 @@ gm20b_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info) tmp |= ((1 << gr->tpc_nr[i]) - 1) << (i * 4); nvkm_wr32(device, 0x4041c4, tmp); - gm200_grctx_generate_405b60(gr); + gm200_grctx_generate_smid_config(gr); gf100_gr_wait_idle(gr); diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp100.c index 88ea322..850a1b10 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp100.c @@ -89,13 +89,12 @@ gp100_grctx_generate_attrib(struct gf100_grctx *info) mmio_wr32(info, 0x41befc, 0x00000000); } -static void -gp100_grctx_generate_405b60(struct gf100_gr *gr) +void +gp100_grctx_generate_smid_config(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; const u32 dist_nr = DIV_ROUND_UP(gr->tpc_total, 4); - u32 dist[TPC_MAX / 4] = {}; - u32 gpcs[GPC_MAX * 2] = {}; + u32 dist[TPC_MAX / 4] = {}, gpcs[16] = {}; u8 tpcnr[GPC_MAX]; int tpc, gpc, i; @@ -112,12 +111,12 @@ gp100_grctx_generate_405b60(struct gf100_gr *gr) tpc = gr->tpc_nr[gpc] - tpcnr[gpc]--; dist[i / 4] |= ((gpc << 4) | tpc) << ((i % 4) * 8); - gpcs[gpc + (gr->gpc_nr * (tpc / 4))] |= i << (tpc * 8); + gpcs[gpc + (gr->func->gpc_nr * (tpc / 4))] |= i << (tpc * 8); } for (i = 0; i < dist_nr; i++) nvkm_wr32(device, 0x405b60 + (i * 4), dist[i]); - for (i = 0; i < gr->gpc_nr * 2; i++) + for (i = 0; i < ARRAY_SIZE(gpcs); i++) nvkm_wr32(device, 0x405ba0 + (i * 4), gpcs[i]); } @@ -149,10 +148,10 @@ gp100_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info) nvkm_wr32(device, 0x405b00, (gr->tpc_total << 8) | gr->gpc_nr); for (tmp = 0, i = 0; i < gr->gpc_nr; i++) - tmp |= ((1 << gr->tpc_nr[i]) - 1) << (i * 5); + tmp |= ((1 << gr->tpc_nr[i]) - 1) << (i * gr->func->tpc_nr); nvkm_wr32(device, 0x4041c4, tmp); - gp100_grctx_generate_405b60(gr); + gp100_grctx_generate_smid_config(gr); gf100_gr_icmd(gr, gr->fuc_bundle); nvkm_wr32(device, 0x404154, idle_timeout); @@ -174,4 +173,6 @@ gp100_grctx = { .attrib_nr = 0x440, .alpha_nr_max = 0xc00, .alpha_nr = 0x800, + .tpc_mask = gm200_grctx_generate_tpc_mask, + .smid_config = gp100_grctx_generate_smid_config, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp102.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp102.c index 7a66b4c..234b0a8 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp102.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp102.c @@ -94,4 +94,6 @@ gp102_grctx = { .attrib_nr = 0x320, .alpha_nr_max = 0xc00, .alpha_nr = 0x800, + .tpc_mask = gm200_grctx_generate_tpc_mask, + .smid_config = gp100_grctx_generate_smid_config, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp107.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp107.c index 8da91a0..62443c9 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp107.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgp107.c @@ -44,4 +44,6 @@ gp107_grctx = { .attrib_nr = 0x540, .alpha_nr_max = 0xc00, .alpha_nr = 0x800, + .tpc_mask = gm200_grctx_generate_tpc_mask, + .smid_config = gp100_grctx_generate_smid_config, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h index d7c2adb..b9ed3aa 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h @@ -135,6 +135,8 @@ struct gf100_gr_func { struct gf100_gr_ucode *ucode; } gpccs; int (*rops)(struct gf100_gr *); + int gpc_nr; + int tpc_nr; int ppc_nr; const struct gf100_grctx_func *grctx; struct nvkm_sclass sclass[]; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm200.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm200.c index 6435f12..c5cca42 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm200.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm200.c @@ -213,6 +213,7 @@ gm200_gr = { .init_rop_active_fbps = gm200_gr_init_rop_active_fbps, .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, .rops = gm200_gr_rops, + .tpc_nr = 4, .ppc_nr = 2, .grctx = &gm200_grctx, .sclass = { diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp100.c index 867a5f7..b87d865 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp100.c @@ -164,6 +164,8 @@ gp100_gr = { .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, .init_num_active_ltcs = gp100_gr_init_num_active_ltcs, .rops = gm200_gr_rops, + .gpc_nr = 6, + .tpc_nr = 5, .ppc_nr = 2, .grctx = &gp100_grctx, .sclass = { diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp102.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp102.c index 61e3a0b..8d1b09b 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp102.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp102.c @@ -49,6 +49,8 @@ gp102_gr = { .init_swdx_pes_mask = gp102_gr_init_swdx_pes_mask, .init_num_active_ltcs = gp100_gr_init_num_active_ltcs, .rops = gm200_gr_rops, + .gpc_nr = 6, + .tpc_nr = 5, .ppc_nr = 3, .grctx = &gp102_grctx, .sclass = { diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp107.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp107.c index f727232..7ca037e 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp107.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp107.c @@ -35,6 +35,8 @@ gp107_gr = { .init_swdx_pes_mask = gp102_gr_init_swdx_pes_mask, .init_num_active_ltcs = gp100_gr_init_num_active_ltcs, .rops = gm200_gr_rops, + .gpc_nr = 2, + .tpc_nr = 3, .ppc_nr = 1, .grctx = &gp107_grctx, .sclass = { diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp10b.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp10b.c index 5f3d161..775c4cf 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp10b.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gp10b.c @@ -41,6 +41,8 @@ gp10b_gr = { .init_ppc_exceptions = gk104_gr_init_ppc_exceptions, .init_num_active_ltcs = gp10b_gr_init_num_active_ltcs, .rops = gm200_gr_rops, + .gpc_nr = 1, + .tpc_nr = 2, .ppc_nr = 1, .grctx = &gp102_grctx, .sclass = {