drm/nouveau/disp/nv50-: implement a common supervisor 2.2

This makes use of all the additional routing and state added in previous
commits, making it possible to deal with GM20x macro link routing, while
also sharing code between the NV50 and GF119 implementations.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

1 files changed, 134 insertions, 117 deletions

diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c
index ebe7657..f21e0e9 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c
@@ -130,6 +130,62 @@ nv50_disp_super_iedt(struct nvkm_head *head, struct nvkm_outp *outp,
 }
 
 static void
+nv50_disp_super_ied_on(struct nvkm_head *head,
+		       struct nvkm_ior *ior, int id, u32 khz)
+{
+	struct nvkm_subdev *subdev = &head->disp->engine.subdev;
+	struct nvkm_bios *bios = subdev->device->bios;
+	struct nvkm_outp *outp = ior->asy.outp;
+	struct nvbios_ocfg iedtrs;
+	struct nvbios_outp iedt;
+	u8  ver, hdr, cnt, len, flags = 0x00;
+	u32 data;
+
+	if (!outp) {
+		IOR_DBG(ior, "nothing to attach");
+		return;
+	}
+
+	/* Lookup IED table for the device. */
+	data = nv50_disp_super_iedt(head, outp, &ver, &hdr, &cnt, &len, &iedt);
+	if (!data)
+		return;
+
+	/* Lookup IEDT runtime settings for the current configuration. */
+	if (ior->type == SOR) {
+		if (ior->asy.proto == LVDS) {
+			if (head->asy.or.depth == 24)
+				flags |= 0x02;
+		}
+		if (ior->asy.link == 3)
+			flags |= 0x01;
+	}
+
+	data = nvbios_ocfg_match(bios, data, ior->asy.proto_evo, flags,
+				 &ver, &hdr, &cnt, &len, &iedtrs);
+	if (!data) {
+		OUTP_DBG(outp, "missing IEDT RS for %02x:%02x",
+			 ior->asy.proto_evo, flags);
+		return;
+	}
+
+	/* Execute the OnInt[23] script for the current frequency. */
+	data = nvbios_oclk_match(bios, iedtrs.clkcmp[id], khz);
+	if (!data) {
+		OUTP_DBG(outp, "missing IEDT RSS %d for %02x:%02x %d khz",
+			 id, ior->asy.proto_evo, flags, khz);
+		return;
+	}
+
+	nvbios_init(subdev, data,
+		init.outp = &outp->info;
+		init.or   = ior->id;
+		init.link = ior->asy.link;
+		init.head = head->id;
+	);
+}
+
+static void
 nv50_disp_super_ied_off(struct nvkm_head *head, struct nvkm_ior *ior, int id)
 {
 	struct nvkm_outp *outp = ior->arm.outp;
@@ -155,6 +211,20 @@ nv50_disp_super_ied_off(struct nvkm_head *head, struct nvkm_ior *ior, int id)
 }
 
 static struct nvkm_ior *
+nv50_disp_super_ior_asy(struct nvkm_head *head)
+{
+	struct nvkm_ior *ior;
+	list_for_each_entry(ior, &head->disp->ior, head) {
+		if (ior->asy.head & (1 << head->id)) {
+			HEAD_DBG(head, "to %s", ior->name);
+			return ior;
+		}
+	}
+	HEAD_DBG(head, "nothing to attach");
+	return NULL;
+}
+
+static struct nvkm_ior *
 nv50_disp_super_ior_arm(struct nvkm_head *head)
 {
 	struct nvkm_ior *ior;
@@ -327,58 +397,40 @@ nv50_disp_intr_unk40_0(struct nv50_disp *disp, int head)
 }
 
 static void
-nv50_disp_intr_unk20_2_dp(struct nv50_disp *disp, int head,
-			  struct dcb_output *outp, u32 pclk)
+nv50_disp_super_2_2_dp(struct nvkm_head *head, struct nvkm_ior *ior)
 {
-	struct nvkm_subdev *subdev = &disp->base.engine.subdev;
-	struct nvkm_device *device = subdev->device;
-	const int link = !(outp->sorconf.link & 1);
-	const int   or = ffs(outp->or) - 1;
-	const u32 soff = (  or * 0x800);
-	const u32 loff = (link * 0x080) + soff;
-	const u32 ctrl = nvkm_rd32(device, 0x610794 + (or * 8));
-	const u32 symbol = 100000;
-	const s32 vactive = nvkm_rd32(device, 0x610af8 + (head * 0x540)) & 0xffff;
-	const s32 vblanke = nvkm_rd32(device, 0x610ae8 + (head * 0x540)) & 0xffff;
-	const s32 vblanks = nvkm_rd32(device, 0x610af0 + (head * 0x540)) & 0xffff;
-	u32 dpctrl = nvkm_rd32(device, 0x61c10c + loff);
-	u32 clksor = nvkm_rd32(device, 0x614300 + soff);
+	struct nvkm_subdev *subdev = &head->disp->engine.subdev;
+	const u32      khz = head->asy.hz / 1000;
+	const u32 linkKBps = ior->dp.bw * 27000;
+	const u32   symbol = 100000;
 	int bestTU = 0, bestVTUi = 0, bestVTUf = 0, bestVTUa = 0;
 	int TU, VTUi, VTUf, VTUa;
 	u64 link_data_rate, link_ratio, unk;
 	u32 best_diff = 64 * symbol;
-	u32 link_nr, link_bw, bits;
-	u64 value;
-
-	link_bw = (clksor & 0x000c0000) ? 270000 : 162000;
-	link_nr = hweight32(dpctrl & 0x000f0000);
+	u64 h, v;
 
 	/* symbols/hblank - algorithm taken from comments in tegra driver */
-	value = vblanke + vactive - vblanks - 7;
-	value = value * link_bw;
-	do_div(value, pclk);
-	value = value - (3 * !!(dpctrl & 0x00004000)) - (12 / link_nr);
-	nvkm_mask(device, 0x61c1e8 + soff, 0x0000ffff, value);
+	h = head->asy.hblanke + head->asy.htotal - head->asy.hblanks - 7;
+	h = h * linkKBps;
+	do_div(h, khz);
+	h = h - (3 * ior->dp.ef) - (12 / ior->dp.nr);
 
 	/* symbols/vblank - algorithm taken from comments in tegra driver */
-	value = vblanks - vblanke - 25;
-	value = value * link_bw;
-	do_div(value, pclk);
-	value = value - ((36 / link_nr) + 3) - 1;
-	nvkm_mask(device, 0x61c1ec + soff, 0x00ffffff, value);
+	v = head->asy.vblanks - head->asy.vblanke - 25;
+	v = v * linkKBps;
+	do_div(v, khz);
+	v = v - ((36 / ior->dp.nr) + 3) - 1;
 
-	/* watermark / activesym */
-	if      ((ctrl & 0xf0000) == 0x60000) bits = 30;
-	else if ((ctrl & 0xf0000) == 0x50000) bits = 24;
-	else                                  bits = 18;
+	ior->func->dp.audio_sym(ior, head->id, h, v);
 
-	link_data_rate = (pclk * bits / 8) / link_nr;
+	/* watermark / activesym */
+	link_data_rate = (khz * head->asy.or.depth / 8) / ior->dp.nr;
 
 	/* calculate ratio of packed data rate to link symbol rate */
 	link_ratio = link_data_rate * symbol;
-	do_div(link_ratio, link_bw);
+	do_div(link_ratio, linkKBps);
 
-	for (TU = 64; TU >= 32; TU--) {
+	for (TU = 64; ior->func->dp.activesym && TU >= 32; TU--) {
 		/* calculate average number of valid symbols in each TU */
 		u32 tu_valid = link_ratio * TU;
 		u32 calc, diff;
@@ -429,9 +481,15 @@ nv50_disp_intr_unk20_2_dp(struct nv50_disp *disp, int head,
 		}
 	}
 
-	if (!bestTU) {
-		nvkm_error(subdev, "unable to find suitable dp config\n");
-		return;
+	if (ior->func->dp.activesym) {
+		if (!bestTU) {
+			nvkm_error(subdev, "unable to determine dp config\n");
+			return;
+		}
+		ior->func->dp.activesym(ior, head->id, bestTU,
+					bestVTUa, bestVTUf, bestVTUi);
+	} else {
+		bestTU = 64;
 	}
 
 	/* XXX close to vbios numbers, but not right */
@@ -441,102 +499,61 @@ nv50_disp_intr_unk20_2_dp(struct nv50_disp *disp, int head,
 	do_div(unk, symbol);
 	unk += 6;
 
-	nvkm_mask(device, 0x61c10c + loff, 0x000001fc, bestTU << 2);
-	nvkm_mask(device, 0x61c128 + loff, 0x010f7f3f, bestVTUa << 24 |
-						   bestVTUf << 16 |
-						   bestVTUi << 8 | unk);
+	ior->func->dp.watermark(ior, head->id, unk);
 }
 
-static void
-nv50_disp_intr_unk20_2(struct nv50_disp *disp, int head)
+void
+nv50_disp_super_2_2(struct nv50_disp *disp, struct nvkm_head *head)
 {
-	struct nvkm_device *device = disp->base.engine.subdev.device;
-	struct nvkm_output *outp;
-	u32 pclk = nvkm_rd32(device, 0x610ad0 + (head * 0x540)) & 0x3fffff;
-	u32 hval, hreg = 0x614200 + (head * 0x800);
-	u32 oval, oreg;
-	u32 mask, conf;
+	const u32 khz = head->asy.hz / 1000;
+	struct nvkm_outp *outp;
+	struct nvkm_ior *ior;
 
-	outp = exec_clkcmp(disp, head, 0xff, pclk, &conf);
-	if (!outp)
+	/* Determine which OR, if any, we're attaching from the head. */
+	HEAD_DBG(head, "supervisor 2.2");
+	ior = nv50_disp_super_ior_asy(head);
+	if (!ior)
 		return;
 
-	/* we allow both encoder attach and detach operations to occur
-	 * within a single supervisor (ie. modeset) sequence.  the
-	 * encoder detach scripts quite often switch off power to the
-	 * lanes, which requires the link to be re-trained.
-	 *
-	 * this is not generally an issue as the sink "must" (heh)
-	 * signal an irq when it's lost sync so the driver can
-	 * re-train.
+	/* For some reason, NVIDIA decided not to:
 	 *
-	 * however, on some boards, if one does not configure at least
-	 * the gpu side of the link *before* attaching, then various
-	 * things can go horribly wrong (PDISP disappearing from mmio,
-	 * third supervisor never happens, etc).
+	 * A) Give dual-link LVDS a separate EVO protocol, like for TMDS.
+	 *  and
+	 * B) Use SetControlOutputResource.PixelDepth on LVDS.
 	 *
-	 * the solution is simply to retrain here, if necessary.  last
-	 * i checked, the binary driver userspace does not appear to
-	 * trigger this situation (it forces an UPDATE between steps).
+	 * Override the values we usually read from HW with the same
+	 * data we pass though an ioctl instead.
 	 */
-	if (outp->info.type == DCB_OUTPUT_DP) {
-		u32 soff = (ffs(outp->info.or) - 1) * 0x08;
-		u32 ctrl, datarate;
-
-		if (outp->info.location == 0) {
-			ctrl = nvkm_rd32(device, 0x610794 + soff);
-			soff = 1;
-		} else {
-			ctrl = nvkm_rd32(device, 0x610b80 + soff);
-			soff = 2;
-		}
-
-		switch ((ctrl & 0x000f0000) >> 16) {
-		case 6: datarate = pclk * 30; break;
-		case 5: datarate = pclk * 24; break;
-		case 2:
-		default:
-			datarate = pclk * 18;
-			break;
-		}
-
-		if (nvkm_output_dp_train(outp, datarate / soff))
-			OUTP_ERR(outp, "link not trained before attach");
+	if (ior->type == SOR && ior->asy.proto == LVDS) {
+		head->asy.or.depth = (disp->sor.lvdsconf & 0x0200) ? 24 : 18;
+		ior->asy.link      = (disp->sor.lvdsconf & 0x0100) ? 3  : 1;
 	}
 
-	exec_clkcmp(disp, head, 0, pclk, &conf);
+	/* Handle any link training, etc. */
+	if ((outp = ior->asy.outp) && outp->func->acquire)
+		outp->func->acquire(outp);
 
-	if (!outp->info.location && outp->info.type == DCB_OUTPUT_ANALOG) {
-		oreg = 0x614280 + (ffs(outp->info.or) - 1) * 0x800;
-		oval = 0x00000000;
-		hval = 0x00000000;
-		mask = 0xffffffff;
-	} else
-	if (!outp->info.location) {
-		if (outp->info.type == DCB_OUTPUT_DP)
-			nv50_disp_intr_unk20_2_dp(disp, head, &outp->info, pclk);
-		oreg = 0x614300 + (ffs(outp->info.or) - 1) * 0x800;
-		oval = (conf & 0x0100) ? 0x00000101 : 0x00000000;
-		hval = 0x00000000;
-		mask = 0x00000707;
-	} else {
-		oreg = 0x614380 + (ffs(outp->info.or) - 1) * 0x800;
-		oval = 0x00000001;
-		hval = 0x00000001;
-		mask = 0x00000707;
-	}
+	/* Execute OnInt2 IED script. */
+	nv50_disp_super_ied_on(head, ior, 0, khz);
+
+	/* Program RG clock divider. */
+	head->func->rgclk(head, ior->asy.rgdiv);
 
-	nvkm_mask(device, hreg, 0x0000000f, hval);
-	nvkm_mask(device, oreg, mask, oval);
+	/* Mode-specific internal DP configuration. */
+	if (ior->type == SOR && ior->asy.proto == DP)
+		nv50_disp_super_2_2_dp(head, ior);
 
-	nv50_disp_dptmds_war_2(disp, &outp->info);
+	/* OR-specific handling. */
+	ior->func->clock(ior);
+	if (ior->func->war_2)
+		ior->func->war_2(ior);
 }
 
 void
 nv50_disp_super_2_1(struct nv50_disp *disp, struct nvkm_head *head)
 {
 	struct nvkm_devinit *devinit = disp->base.engine.subdev.device->devinit;
-	u32 khz = head->asy.hz / 1000;
+	const u32 khz = head->asy.hz / 1000;
 	HEAD_DBG(head, "supervisor 2.1 - %d khz", khz);
 	if (khz)
 		nvkm_devinit_pll_set(devinit, PLL_VPLL0 + head->id, khz);
@@ -636,7 +653,7 @@ nv50_disp_super(struct work_struct *work)
 		list_for_each_entry(head, &disp->base.head, head) {
 			if (!(super & (0x00000080 << head->id)))
 				continue;
-			nv50_disp_intr_unk20_2(disp, head->id);
+			nv50_disp_super_2_2(disp, head);
 		}
 	} else
 	if (disp->super & 0x00000040) {