Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-nextgrafted

Pull networking updates from Jakub Kicinski:
 "Core:

   - Add dedicated kmem_cache for typical/small skb->head, avoid having
     to access struct page at kfree time, and improve memory use.

   - Introduce sysctl to set default RPS configuration for new netdevs.

   - Define Netlink protocol specification format which can be used to
     describe messages used by each family and auto-generate parsers.
     Add tools for generating kernel data structures and uAPI headers.

   - Expose all net/core sysctls inside netns.

   - Remove 4s sleep in netpoll if carrier is instantly detected on
     boot.

   - Add configurable limit of MDB entries per port, and port-vlan.

   - Continue populating drop reasons throughout the stack.

   - Retire a handful of legacy Qdiscs and classifiers.

  Protocols:

   - Support IPv4 big TCP (TSO frames larger than 64kB).

   - Add IP_LOCAL_PORT_RANGE socket option, to control local port range
     on socket by socket basis.

   - Track and report in procfs number of MPTCP sockets used.

   - Support mixing IPv4 and IPv6 flows in the in-kernel MPTCP path
     manager.

   - IPv6: don't check net.ipv6.route.max_size and rely on garbage
     collection to free memory (similarly to IPv4).

   - Support Penultimate Segment Pop (PSP) flavor in SRv6 (RFC8986).

   - ICMP: add per-rate limit counters.

   - Add support for user scanning requests in ieee802154.

   - Remove static WEP support.

   - Support minimal Wi-Fi 7 Extremely High Throughput (EHT) rate
     reporting.

   - WiFi 7 EHT channel puncturing support (client & AP).

  BPF:

   - Add a rbtree data structure following the "next-gen data structure"
     precedent set by recently added linked list, that is, by using
     kfunc + kptr instead of adding a new BPF map type.

   - Expose XDP hints via kfuncs with initial support for RX hash and
     timestamp metadata.

   - Add BPF_F_NO_TUNNEL_KEY extension to bpf_skb_set_tunnel_key to
     better support decap on GRE tunnel devices not operating in collect
     metadata.

   - Improve x86 JIT's codegen for PROBE_MEM runtime error checks.

   - Remove the need for trace_printk_lock for bpf_trace_printk and
     bpf_trace_vprintk helpers.

   - Extend libbpf's bpf_tracing.h support for tracing arguments of
     kprobes/uprobes and syscall as a special case.

   - Significantly reduce the search time for module symbols by
     livepatch and BPF.

   - Enable cpumasks to be used as kptrs, which is useful for tracing
     programs tracking which tasks end up running on which CPUs in
     different time intervals.

   - Add support for BPF trampoline on s390x and riscv64.

   - Add capability to export the XDP features supported by the NIC.

   - Add __bpf_kfunc tag for marking kernel functions as kfuncs.

   - Add cgroup.memory=nobpf kernel parameter option to disable BPF
     memory accounting for container environments.

  Netfilter:

   - Remove the CLUSTERIP target. It has been marked as obsolete for
     years, and we still have WARN splats wrt races of the out-of-band
     /proc interface installed by this target.

   - Add 'destroy' commands to nf_tables. They are identical to the
     existing 'delete' commands, but do not return an error if the
     referenced object (set, chain, rule...) did not exist.

  Driver API:

   - Improve cpumask_local_spread() locality to help NICs set the right
     IRQ affinity on AMD platforms.

   - Separate C22 and C45 MDIO bus transactions more clearly.

   - Introduce new DCB table to control DSCP rewrite on egress.

   - Support configuration of Physical Layer Collision Avoidance (PLCA)
     Reconciliation Sublayer (RS) (802.3cg-2019). Modern version of
     shared medium Ethernet.

   - Support for MAC Merge layer (IEEE 802.3-2018 clause 99). Allowing
     preemption of low priority frames by high priority frames.

   - Add support for controlling MACSec offload using netlink SET.

   - Rework devlink instance refcounts to allow registration and
     de-registration under the instance lock. Split the code into
     multiple files, drop some of the unnecessarily granular locks and
     factor out common parts of netlink operation handling.

   - Add TX frame aggregation parameters (for USB drivers).

   - Add a new attr TCA_EXT_WARN_MSG to report TC (offload) warning
     messages with notifications for debug.

   - Allow offloading of UDP NEW connections via act_ct.

   - Add support for per action HW stats in TC.

   - Support hardware miss to TC action (continue processing in SW from
     a specific point in the action chain).

   - Warn if old Wireless Extension user space interface is used with
     modern cfg80211/mac80211 drivers. Do not support Wireless
     Extensions for Wi-Fi 7 devices at all. Everyone should switch to
     using nl80211 interface instead.

   - Improve the CAN bit timing configuration. Use extack to return
     error messages directly to user space, update the SJW handling,
     including the definition of a new default value that will benefit
     CAN-FD controllers, by increasing their oscillator tolerance.

  New hardware / drivers:

   - Ethernet:
      - nVidia BlueField-3 support (control traffic driver)
      - Ethernet support for imx93 SoCs
      - Motorcomm yt8531 gigabit Ethernet PHY
      - onsemi NCN26000 10BASE-T1S PHY (with support for PLCA)
      - Microchip LAN8841 PHY (incl. cable diagnostics and PTP)
      - Amlogic gxl MDIO mux

   - WiFi:
      - RealTek RTL8188EU (rtl8xxxu)
      - Qualcomm Wi-Fi 7 devices (ath12k)

   - CAN:
      - Renesas R-Car V4H

  Drivers:

   - Bluetooth:
      - Set Per Platform Antenna Gain (PPAG) for Intel controllers.

   - Ethernet NICs:
      - Intel (1G, igc):
         - support TSN / Qbv / packet scheduling features of i226 model
      - Intel (100G, ice):
         - use GNSS subsystem instead of TTY
         - multi-buffer XDP support
         - extend support for GPIO pins to E823 devices
      - nVidia/Mellanox:
         - update the shared buffer configuration on PFC commands
         - implement PTP adjphase function for HW offset control
         - TC support for Geneve and GRE with VF tunnel offload
         - more efficient crypto key management method
         - multi-port eswitch support
      - Netronome/Corigine:
         - add DCB IEEE support
         - support IPsec offloading for NFP3800
      - Freescale/NXP (enetc):
         - support XDP_REDIRECT for XDP non-linear buffers
         - improve reconfig, avoid link flap and waiting for idle
         - support MAC Merge layer
      - Other NICs:
         - sfc/ef100: add basic devlink support for ef100
         - ionic: rx_push mode operation (writing descriptors via MMIO)
         - bnxt: use the auxiliary bus abstraction for RDMA
         - r8169: disable ASPM and reset bus in case of tx timeout
         - cpsw: support QSGMII mode for J721e CPSW9G
         - cpts: support pulse-per-second output
         - ngbe: add an mdio bus driver
         - usbnet: optimize usbnet_bh() by avoiding unnecessary queuing
         - r8152: handle devices with FW with NCM support
         - amd-xgbe: support 10Mbps, 2.5GbE speeds and rx-adaptation
         - virtio-net: support multi buffer XDP
         - virtio/vsock: replace virtio_vsock_pkt with sk_buff
         - tsnep: XDP support

   - Ethernet high-speed switches:
      - nVidia/Mellanox (mlxsw):
         - add support for latency TLV (in FW control messages)
      - Microchip (sparx5):
         - separate explicit and implicit traffic forwarding rules, make
           the implicit rules always active
         - add support for egress DSCP rewrite
         - IS0 VCAP support (Ingress Classification)
         - IS2 VCAP filters (protos, L3 addrs, L4 ports, flags, ToS
           etc.)
         - ES2 VCAP support (Egress Access Control)
         - support for Per-Stream Filtering and Policing (802.1Q,
           8.6.5.1)

   - Ethernet embedded switches:
      - Marvell (mv88e6xxx):
         - add MAB (port auth) offload support
         - enable PTP receive for mv88e6390
      - NXP (ocelot):
         - support MAC Merge layer
         - support for the the vsc7512 internal copper phys
      - Microchip:
         - lan9303: convert to PHYLINK
         - lan966x: support TC flower filter statistics
         - lan937x: PTP support for KSZ9563/KSZ8563 and LAN937x
         - lan937x: support Credit Based Shaper configuration
         - ksz9477: support Energy Efficient Ethernet
      - other:
         - qca8k: convert to regmap read/write API, use bulk operations
         - rswitch: Improve TX timestamp accuracy

   - Intel WiFi (iwlwifi):
      - EHT (Wi-Fi 7) rate reporting
      - STEP equalizer support: transfer some STEP (connection to radio
        on platforms with integrated wifi) related parameters from the
        BIOS to the firmware.

   - Qualcomm 802.11ax WiFi (ath11k):
      - IPQ5018 support
      - Fine Timing Measurement (FTM) responder role support
      - channel 177 support

   - MediaTek WiFi (mt76):
      - per-PHY LED support
      - mt7996: EHT (Wi-Fi 7) support
      - Wireless Ethernet Dispatch (WED) reset support
      - switch to using page pool allocator

   - RealTek WiFi (rtw89):
      - support new version of Bluetooth co-existance

   - Mobile:
      - rmnet: support TX aggregation"

* tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1872 commits)
  page_pool: add a comment explaining the fragment counter usage
  net: ethtool: fix __ethtool_dev_mm_supported() implementation
  ethtool: pse-pd: Fix double word in comments
  xsk: add linux/vmalloc.h to xsk.c
  sefltests: netdevsim: wait for devlink instance after netns removal
  selftest: fib_tests: Always cleanup before exit
  net/mlx5e: Align IPsec ASO result memory to be as required by hardware
  net/mlx5e: TC, Set CT miss to the specific ct action instance
  net/mlx5e: Rename CHAIN_TO_REG to MAPPED_OBJ_TO_REG
  net/mlx5: Refactor tc miss handling to a single function
  net/mlx5: Kconfig: Make tc offload depend on tc skb extension
  net/sched: flower: Support hardware miss to tc action
  net/sched: flower: Move filter handle initialization earlier
  net/sched: cls_api: Support hardware miss to tc action
  net/sched: Rename user cookie and act cookie
  sfc: fix builds without CONFIG_RTC_LIB
  sfc: clean up some inconsistent indentings
  net/mlx4_en: Introduce flexible array to silence overflow warning
  net: lan966x: Fix possible deadlock inside PTP
  net/ulp: Remove redundant ->clone() test in inet_clone_ulp().
  ...

1 files changed, 1204 insertions, 0 deletions

diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c
new file mode 100644
index 000000000..1c236f02b
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_bw.c
@@ -0,0 +1,1204 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#include <drm/drm_atomic_state_helper.h>
+
+#include "i915_drv.h"
+#include "i915_reg.h"
+#include "i915_utils.h"
+#include "intel_atomic.h"
+#include "intel_bw.h"
+#include "intel_cdclk.h"
+#include "intel_display_core.h"
+#include "intel_display_types.h"
+#include "skl_watermark.h"
+#include "intel_mchbar_regs.h"
+#include "intel_pcode.h"
+
+/* Parameters for Qclk Geyserville (QGV) */
+struct intel_qgv_point {
+	u16 dclk, t_rp, t_rdpre, t_rc, t_ras, t_rcd;
+};
+
+struct intel_psf_gv_point {
+	u8 clk; /* clock in multiples of 16.6666 MHz */
+};
+
+struct intel_qgv_info {
+	struct intel_qgv_point points[I915_NUM_QGV_POINTS];
+	struct intel_psf_gv_point psf_points[I915_NUM_PSF_GV_POINTS];
+	u8 num_points;
+	u8 num_psf_points;
+	u8 t_bl;
+	u8 max_numchannels;
+	u8 channel_width;
+	u8 deinterleave;
+};
+
+static int dg1_mchbar_read_qgv_point_info(struct drm_i915_private *dev_priv,
+					  struct intel_qgv_point *sp,
+					  int point)
+{
+	u32 dclk_ratio, dclk_reference;
+	u32 val;
+
+	val = intel_uncore_read(&dev_priv->uncore, SA_PERF_STATUS_0_0_0_MCHBAR_PC);
+	dclk_ratio = REG_FIELD_GET(DG1_QCLK_RATIO_MASK, val);
+	if (val & DG1_QCLK_REFERENCE)
+		dclk_reference = 6; /* 6 * 16.666 MHz = 100 MHz */
+	else
+		dclk_reference = 8; /* 8 * 16.666 MHz = 133 MHz */
+	sp->dclk = DIV_ROUND_UP((16667 * dclk_ratio * dclk_reference) + 500, 1000);
+
+	val = intel_uncore_read(&dev_priv->uncore, SKL_MC_BIOS_DATA_0_0_0_MCHBAR_PCU);
+	if (val & DG1_GEAR_TYPE)
+		sp->dclk *= 2;
+
+	if (sp->dclk == 0)
+		return -EINVAL;
+
+	val = intel_uncore_read(&dev_priv->uncore, MCHBAR_CH0_CR_TC_PRE_0_0_0_MCHBAR);
+	sp->t_rp = REG_FIELD_GET(DG1_DRAM_T_RP_MASK, val);
+	sp->t_rdpre = REG_FIELD_GET(DG1_DRAM_T_RDPRE_MASK, val);
+
+	val = intel_uncore_read(&dev_priv->uncore, MCHBAR_CH0_CR_TC_PRE_0_0_0_MCHBAR_HIGH);
+	sp->t_rcd = REG_FIELD_GET(DG1_DRAM_T_RCD_MASK, val);
+	sp->t_ras = REG_FIELD_GET(DG1_DRAM_T_RAS_MASK, val);
+
+	sp->t_rc = sp->t_rp + sp->t_ras;
+
+	return 0;
+}
+
+static int icl_pcode_read_qgv_point_info(struct drm_i915_private *dev_priv,
+					 struct intel_qgv_point *sp,
+					 int point)
+{
+	u32 val = 0, val2 = 0;
+	u16 dclk;
+	int ret;
+
+	ret = snb_pcode_read(&dev_priv->uncore, ICL_PCODE_MEM_SUBSYSYSTEM_INFO |
+			     ICL_PCODE_MEM_SS_READ_QGV_POINT_INFO(point),
+			     &val, &val2);
+	if (ret)
+		return ret;
+
+	dclk = val & 0xffff;
+	sp->dclk = DIV_ROUND_UP((16667 * dclk) + (DISPLAY_VER(dev_priv) > 11 ? 500 : 0), 1000);
+	sp->t_rp = (val & 0xff0000) >> 16;
+	sp->t_rcd = (val & 0xff000000) >> 24;
+
+	sp->t_rdpre = val2 & 0xff;
+	sp->t_ras = (val2 & 0xff00) >> 8;
+
+	sp->t_rc = sp->t_rp + sp->t_ras;
+
+	return 0;
+}
+
+static int adls_pcode_read_psf_gv_point_info(struct drm_i915_private *dev_priv,
+					    struct intel_psf_gv_point *points)
+{
+	u32 val = 0;
+	int ret;
+	int i;
+
+	ret = snb_pcode_read(&dev_priv->uncore, ICL_PCODE_MEM_SUBSYSYSTEM_INFO |
+			     ADL_PCODE_MEM_SS_READ_PSF_GV_INFO, &val, NULL);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < I915_NUM_PSF_GV_POINTS; i++) {
+		points[i].clk = val & 0xff;
+		val >>= 8;
+	}
+
+	return 0;
+}
+
+int icl_pcode_restrict_qgv_points(struct drm_i915_private *dev_priv,
+				  u32 points_mask)
+{
+	int ret;
+
+	/* bspec says to keep retrying for at least 1 ms */
+	ret = skl_pcode_request(&dev_priv->uncore, ICL_PCODE_SAGV_DE_MEM_SS_CONFIG,
+				points_mask,
+				ICL_PCODE_REP_QGV_MASK | ADLS_PCODE_REP_PSF_MASK,
+				ICL_PCODE_REP_QGV_SAFE | ADLS_PCODE_REP_PSF_SAFE,
+				1);
+
+	if (ret < 0) {
+		drm_err(&dev_priv->drm, "Failed to disable qgv points (%d) points: 0x%x\n", ret, points_mask);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int mtl_read_qgv_point_info(struct drm_i915_private *dev_priv,
+				   struct intel_qgv_point *sp, int point)
+{
+	u32 val, val2;
+	u16 dclk;
+
+	val = intel_uncore_read(&dev_priv->uncore,
+				MTL_MEM_SS_INFO_QGV_POINT_LOW(point));
+	val2 = intel_uncore_read(&dev_priv->uncore,
+				 MTL_MEM_SS_INFO_QGV_POINT_HIGH(point));
+	dclk = REG_FIELD_GET(MTL_DCLK_MASK, val);
+	sp->dclk = DIV_ROUND_UP((16667 * dclk), 1000);
+	sp->t_rp = REG_FIELD_GET(MTL_TRP_MASK, val);
+	sp->t_rcd = REG_FIELD_GET(MTL_TRCD_MASK, val);
+
+	sp->t_rdpre = REG_FIELD_GET(MTL_TRDPRE_MASK, val2);
+	sp->t_ras = REG_FIELD_GET(MTL_TRAS_MASK, val2);
+
+	sp->t_rc = sp->t_rp + sp->t_ras;
+
+	return 0;
+}
+
+static int
+intel_read_qgv_point_info(struct drm_i915_private *dev_priv,
+			  struct intel_qgv_point *sp,
+			  int point)
+{
+	if (DISPLAY_VER(dev_priv) >= 14)
+		return mtl_read_qgv_point_info(dev_priv, sp, point);
+	else if (IS_DG1(dev_priv))
+		return dg1_mchbar_read_qgv_point_info(dev_priv, sp, point);
+	else
+		return icl_pcode_read_qgv_point_info(dev_priv, sp, point);
+}
+
+static int icl_get_qgv_points(struct drm_i915_private *dev_priv,
+			      struct intel_qgv_info *qi,
+			      bool is_y_tile)
+{
+	const struct dram_info *dram_info = &dev_priv->dram_info;
+	int i, ret;
+
+	qi->num_points = dram_info->num_qgv_points;
+	qi->num_psf_points = dram_info->num_psf_gv_points;
+
+	if (DISPLAY_VER(dev_priv) >= 14) {
+		switch (dram_info->type) {
+		case INTEL_DRAM_DDR4:
+			qi->t_bl = 4;
+			qi->max_numchannels = 2;
+			qi->channel_width = 64;
+			qi->deinterleave = 2;
+			break;
+		case INTEL_DRAM_DDR5:
+			qi->t_bl = 8;
+			qi->max_numchannels = 4;
+			qi->channel_width = 32;
+			qi->deinterleave = 2;
+			break;
+		case INTEL_DRAM_LPDDR4:
+		case INTEL_DRAM_LPDDR5:
+			qi->t_bl = 16;
+			qi->max_numchannels = 8;
+			qi->channel_width = 16;
+			qi->deinterleave = 4;
+			break;
+		default:
+			MISSING_CASE(dram_info->type);
+			return -EINVAL;
+		}
+	} else if (DISPLAY_VER(dev_priv) >= 12) {
+		switch (dram_info->type) {
+		case INTEL_DRAM_DDR4:
+			qi->t_bl = is_y_tile ? 8 : 4;
+			qi->max_numchannels = 2;
+			qi->channel_width = 64;
+			qi->deinterleave = is_y_tile ? 1 : 2;
+			break;
+		case INTEL_DRAM_DDR5:
+			qi->t_bl = is_y_tile ? 16 : 8;
+			qi->max_numchannels = 4;
+			qi->channel_width = 32;
+			qi->deinterleave = is_y_tile ? 1 : 2;
+			break;
+		case INTEL_DRAM_LPDDR4:
+			if (IS_ROCKETLAKE(dev_priv)) {
+				qi->t_bl = 8;
+				qi->max_numchannels = 4;
+				qi->channel_width = 32;
+				qi->deinterleave = 2;
+				break;
+			}
+			fallthrough;
+		case INTEL_DRAM_LPDDR5:
+			qi->t_bl = 16;
+			qi->max_numchannels = 8;
+			qi->channel_width = 16;
+			qi->deinterleave = is_y_tile ? 2 : 4;
+			break;
+		default:
+			qi->t_bl = 16;
+			qi->max_numchannels = 1;
+			break;
+		}
+	} else if (DISPLAY_VER(dev_priv) == 11) {
+		qi->t_bl = dev_priv->dram_info.type == INTEL_DRAM_DDR4 ? 4 : 8;
+		qi->max_numchannels = 1;
+	}
+
+	if (drm_WARN_ON(&dev_priv->drm,
+			qi->num_points > ARRAY_SIZE(qi->points)))
+		qi->num_points = ARRAY_SIZE(qi->points);
+
+	for (i = 0; i < qi->num_points; i++) {
+		struct intel_qgv_point *sp = &qi->points[i];
+
+		ret = intel_read_qgv_point_info(dev_priv, sp, i);
+		if (ret)
+			return ret;
+
+		drm_dbg_kms(&dev_priv->drm,
+			    "QGV %d: DCLK=%d tRP=%d tRDPRE=%d tRAS=%d tRCD=%d tRC=%d\n",
+			    i, sp->dclk, sp->t_rp, sp->t_rdpre, sp->t_ras,
+			    sp->t_rcd, sp->t_rc);
+	}
+
+	if (qi->num_psf_points > 0) {
+		ret = adls_pcode_read_psf_gv_point_info(dev_priv, qi->psf_points);
+		if (ret) {
+			drm_err(&dev_priv->drm, "Failed to read PSF point data; PSF points will not be considered in bandwidth calculations.\n");
+			qi->num_psf_points = 0;
+		}
+
+		for (i = 0; i < qi->num_psf_points; i++)
+			drm_dbg_kms(&dev_priv->drm,
+				    "PSF GV %d: CLK=%d \n",
+				    i, qi->psf_points[i].clk);
+	}
+
+	return 0;
+}
+
+static int adl_calc_psf_bw(int clk)
+{
+	/*
+	 * clk is multiples of 16.666MHz (100/6)
+	 * According to BSpec PSF GV bandwidth is
+	 * calculated as BW = 64 * clk * 16.666Mhz
+	 */
+	return DIV_ROUND_CLOSEST(64 * clk * 100, 6);
+}
+
+static int icl_sagv_max_dclk(const struct intel_qgv_info *qi)
+{
+	u16 dclk = 0;
+	int i;
+
+	for (i = 0; i < qi->num_points; i++)
+		dclk = max(dclk, qi->points[i].dclk);
+
+	return dclk;
+}
+
+struct intel_sa_info {
+	u16 displayrtids;
+	u8 deburst, deprogbwlimit, derating;
+};
+
+static const struct intel_sa_info icl_sa_info = {
+	.deburst = 8,
+	.deprogbwlimit = 25, /* GB/s */
+	.displayrtids = 128,
+	.derating = 10,
+};
+
+static const struct intel_sa_info tgl_sa_info = {
+	.deburst = 16,
+	.deprogbwlimit = 34, /* GB/s */
+	.displayrtids = 256,
+	.derating = 10,
+};
+
+static const struct intel_sa_info rkl_sa_info = {
+	.deburst = 8,
+	.deprogbwlimit = 20, /* GB/s */
+	.displayrtids = 128,
+	.derating = 10,
+};
+
+static const struct intel_sa_info adls_sa_info = {
+	.deburst = 16,
+	.deprogbwlimit = 38, /* GB/s */
+	.displayrtids = 256,
+	.derating = 10,
+};
+
+static const struct intel_sa_info adlp_sa_info = {
+	.deburst = 16,
+	.deprogbwlimit = 38, /* GB/s */
+	.displayrtids = 256,
+	.derating = 20,
+};
+
+static const struct intel_sa_info mtl_sa_info = {
+	.deburst = 32,
+	.deprogbwlimit = 38, /* GB/s */
+	.displayrtids = 256,
+	.derating = 20,
+};
+
+static int icl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel_sa_info *sa)
+{
+	struct intel_qgv_info qi = {};
+	bool is_y_tile = true; /* assume y tile may be used */
+	int num_channels = max_t(u8, 1, dev_priv->dram_info.num_channels);
+	int ipqdepth, ipqdepthpch = 16;
+	int dclk_max;
+	int maxdebw;
+	int num_groups = ARRAY_SIZE(dev_priv->display.bw.max);
+	int i, ret;
+
+	ret = icl_get_qgv_points(dev_priv, &qi, is_y_tile);
+	if (ret) {
+		drm_dbg_kms(&dev_priv->drm,
+			    "Failed to get memory subsystem information, ignoring bandwidth limits");
+		return ret;
+	}
+
+	dclk_max = icl_sagv_max_dclk(&qi);
+	maxdebw = min(sa->deprogbwlimit * 1000, dclk_max * 16 * 6 / 10);
+	ipqdepth = min(ipqdepthpch, sa->displayrtids / num_channels);
+	qi.deinterleave = DIV_ROUND_UP(num_channels, is_y_tile ? 4 : 2);
+
+	for (i = 0; i < num_groups; i++) {
+		struct intel_bw_info *bi = &dev_priv->display.bw.max[i];
+		int clpchgroup;
+		int j;
+
+		clpchgroup = (sa->deburst * qi.deinterleave / num_channels) << i;
+		bi->num_planes = (ipqdepth - clpchgroup) / clpchgroup + 1;
+
+		bi->num_qgv_points = qi.num_points;
+		bi->num_psf_gv_points = qi.num_psf_points;
+
+		for (j = 0; j < qi.num_points; j++) {
+			const struct intel_qgv_point *sp = &qi.points[j];
+			int ct, bw;
+
+			/*
+			 * Max row cycle time
+			 *
+			 * FIXME what is the logic behind the
+			 * assumed burst length?
+			 */
+			ct = max_t(int, sp->t_rc, sp->t_rp + sp->t_rcd +
+				   (clpchgroup - 1) * qi.t_bl + sp->t_rdpre);
+			bw = DIV_ROUND_UP(sp->dclk * clpchgroup * 32 * num_channels, ct);
+
+			bi->deratedbw[j] = min(maxdebw,
+					       bw * (100 - sa->derating) / 100);
+
+			drm_dbg_kms(&dev_priv->drm,
+				    "BW%d / QGV %d: num_planes=%d deratedbw=%u\n",
+				    i, j, bi->num_planes, bi->deratedbw[j]);
+		}
+	}
+	/*
+	 * In case if SAGV is disabled in BIOS, we always get 1
+	 * SAGV point, but we can't send PCode commands to restrict it
+	 * as it will fail and pointless anyway.
+	 */
+	if (qi.num_points == 1)
+		dev_priv->display.sagv.status = I915_SAGV_NOT_CONTROLLED;
+	else
+		dev_priv->display.sagv.status = I915_SAGV_ENABLED;
+
+	return 0;
+}
+
+static int tgl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel_sa_info *sa)
+{
+	struct intel_qgv_info qi = {};
+	const struct dram_info *dram_info = &dev_priv->dram_info;
+	bool is_y_tile = true; /* assume y tile may be used */
+	int num_channels = max_t(u8, 1, dev_priv->dram_info.num_channels);
+	int ipqdepth, ipqdepthpch = 16;
+	int dclk_max;
+	int maxdebw, peakbw;
+	int clperchgroup;
+	int num_groups = ARRAY_SIZE(dev_priv->display.bw.max);
+	int i, ret;
+
+	ret = icl_get_qgv_points(dev_priv, &qi, is_y_tile);
+	if (ret) {
+		drm_dbg_kms(&dev_priv->drm,
+			    "Failed to get memory subsystem information, ignoring bandwidth limits");
+		return ret;
+	}
+
+	if (DISPLAY_VER(dev_priv) < 14 &&
+	    (dram_info->type == INTEL_DRAM_LPDDR4 || dram_info->type == INTEL_DRAM_LPDDR5))
+		num_channels *= 2;
+
+	qi.deinterleave = qi.deinterleave ? : DIV_ROUND_UP(num_channels, is_y_tile ? 4 : 2);
+
+	if (num_channels < qi.max_numchannels && DISPLAY_VER(dev_priv) >= 12)
+		qi.deinterleave = max(DIV_ROUND_UP(qi.deinterleave, 2), 1);
+
+	if (DISPLAY_VER(dev_priv) > 11 && num_channels > qi.max_numchannels)
+		drm_warn(&dev_priv->drm, "Number of channels exceeds max number of channels.");
+	if (qi.max_numchannels != 0)
+		num_channels = min_t(u8, num_channels, qi.max_numchannels);
+
+	dclk_max = icl_sagv_max_dclk(&qi);
+
+	peakbw = num_channels * DIV_ROUND_UP(qi.channel_width, 8) * dclk_max;
+	maxdebw = min(sa->deprogbwlimit * 1000, peakbw * 6 / 10); /* 60% */
+
+	ipqdepth = min(ipqdepthpch, sa->displayrtids / num_channels);
+	/*
+	 * clperchgroup = 4kpagespermempage * clperchperblock,
+	 * clperchperblock = 8 / num_channels * interleave
+	 */
+	clperchgroup = 4 * DIV_ROUND_UP(8, num_channels) * qi.deinterleave;
+
+	for (i = 0; i < num_groups; i++) {
+		struct intel_bw_info *bi = &dev_priv->display.bw.max[i];
+		struct intel_bw_info *bi_next;
+		int clpchgroup;
+		int j;
+
+		clpchgroup = (sa->deburst * qi.deinterleave / num_channels) << i;
+
+		if (i < num_groups - 1) {
+			bi_next = &dev_priv->display.bw.max[i + 1];
+
+			if (clpchgroup < clperchgroup)
+				bi_next->num_planes = (ipqdepth - clpchgroup) /
+						       clpchgroup + 1;
+			else
+				bi_next->num_planes = 0;
+		}
+
+		bi->num_qgv_points = qi.num_points;
+		bi->num_psf_gv_points = qi.num_psf_points;
+
+		for (j = 0; j < qi.num_points; j++) {
+			const struct intel_qgv_point *sp = &qi.points[j];
+			int ct, bw;
+
+			/*
+			 * Max row cycle time
+			 *
+			 * FIXME what is the logic behind the
+			 * assumed burst length?
+			 */
+			ct = max_t(int, sp->t_rc, sp->t_rp + sp->t_rcd +
+				   (clpchgroup - 1) * qi.t_bl + sp->t_rdpre);
+			bw = DIV_ROUND_UP(sp->dclk * clpchgroup * 32 * num_channels, ct);
+
+			bi->deratedbw[j] = min(maxdebw,
+					       bw * (100 - sa->derating) / 100);
+
+			drm_dbg_kms(&dev_priv->drm,
+				    "BW%d / QGV %d: num_planes=%d deratedbw=%u\n",
+				    i, j, bi->num_planes, bi->deratedbw[j]);
+		}
+
+		for (j = 0; j < qi.num_psf_points; j++) {
+			const struct intel_psf_gv_point *sp = &qi.psf_points[j];
+
+			bi->psf_bw[j] = adl_calc_psf_bw(sp->clk);
+
+			drm_dbg_kms(&dev_priv->drm,
+				    "BW%d / PSF GV %d: num_planes=%d bw=%u\n",
+				    i, j, bi->num_planes, bi->psf_bw[j]);
+		}
+	}
+
+	/*
+	 * In case if SAGV is disabled in BIOS, we always get 1
+	 * SAGV point, but we can't send PCode commands to restrict it
+	 * as it will fail and pointless anyway.
+	 */
+	if (qi.num_points == 1)
+		dev_priv->display.sagv.status = I915_SAGV_NOT_CONTROLLED;
+	else
+		dev_priv->display.sagv.status = I915_SAGV_ENABLED;
+
+	return 0;
+}
+
+static void dg2_get_bw_info(struct drm_i915_private *i915)
+{
+	unsigned int deratedbw = IS_DG2_G11(i915) ? 38000 : 50000;
+	int num_groups = ARRAY_SIZE(i915->display.bw.max);
+	int i;
+
+	/*
+	 * DG2 doesn't have SAGV or QGV points, just a constant max bandwidth
+	 * that doesn't depend on the number of planes enabled. So fill all the
+	 * plane group with constant bw information for uniformity with other
+	 * platforms. DG2-G10 platforms have a constant 50 GB/s bandwidth,
+	 * whereas DG2-G11 platforms have 38 GB/s.
+	 */
+	for (i = 0; i < num_groups; i++) {
+		struct intel_bw_info *bi = &i915->display.bw.max[i];
+
+		bi->num_planes = 1;
+		/* Need only one dummy QGV point per group */
+		bi->num_qgv_points = 1;
+		bi->deratedbw[0] = deratedbw;
+	}
+
+	i915->display.sagv.status = I915_SAGV_NOT_CONTROLLED;
+}
+
+static unsigned int icl_max_bw(struct drm_i915_private *dev_priv,
+			       int num_planes, int qgv_point)
+{
+	int i;
+
+	/*
+	 * Let's return max bw for 0 planes
+	 */
+	num_planes = max(1, num_planes);
+
+	for (i = 0; i < ARRAY_SIZE(dev_priv->display.bw.max); i++) {
+		const struct intel_bw_info *bi =
+			&dev_priv->display.bw.max[i];
+
+		/*
+		 * Pcode will not expose all QGV points when
+		 * SAGV is forced to off/min/med/max.
+		 */
+		if (qgv_point >= bi->num_qgv_points)
+			return UINT_MAX;
+
+		if (num_planes >= bi->num_planes)
+			return bi->deratedbw[qgv_point];
+	}
+
+	return 0;
+}
+
+static unsigned int tgl_max_bw(struct drm_i915_private *dev_priv,
+			       int num_planes, int qgv_point)
+{
+	int i;
+
+	/*
+	 * Let's return max bw for 0 planes
+	 */
+	num_planes = max(1, num_planes);
+
+	for (i = ARRAY_SIZE(dev_priv->display.bw.max) - 1; i >= 0; i--) {
+		const struct intel_bw_info *bi =
+			&dev_priv->display.bw.max[i];
+
+		/*
+		 * Pcode will not expose all QGV points when
+		 * SAGV is forced to off/min/med/max.
+		 */
+		if (qgv_point >= bi->num_qgv_points)
+			return UINT_MAX;
+
+		if (num_planes <= bi->num_planes)
+			return bi->deratedbw[qgv_point];
+	}
+
+	return dev_priv->display.bw.max[0].deratedbw[qgv_point];
+}
+
+static unsigned int adl_psf_bw(struct drm_i915_private *dev_priv,
+			       int psf_gv_point)
+{
+	const struct intel_bw_info *bi =
+			&dev_priv->display.bw.max[0];
+
+	return bi->psf_bw[psf_gv_point];
+}
+
+void intel_bw_init_hw(struct drm_i915_private *dev_priv)
+{
+	if (!HAS_DISPLAY(dev_priv))
+		return;
+
+	if (DISPLAY_VER(dev_priv) >= 14)
+		tgl_get_bw_info(dev_priv, &mtl_sa_info);
+	else if (IS_DG2(dev_priv))
+		dg2_get_bw_info(dev_priv);
+	else if (IS_ALDERLAKE_P(dev_priv))
+		tgl_get_bw_info(dev_priv, &adlp_sa_info);
+	else if (IS_ALDERLAKE_S(dev_priv))
+		tgl_get_bw_info(dev_priv, &adls_sa_info);
+	else if (IS_ROCKETLAKE(dev_priv))
+		tgl_get_bw_info(dev_priv, &rkl_sa_info);
+	else if (DISPLAY_VER(dev_priv) == 12)
+		tgl_get_bw_info(dev_priv, &tgl_sa_info);
+	else if (DISPLAY_VER(dev_priv) == 11)
+		icl_get_bw_info(dev_priv, &icl_sa_info);
+}
+
+static unsigned int intel_bw_crtc_num_active_planes(const struct intel_crtc_state *crtc_state)
+{
+	/*
+	 * We assume cursors are small enough
+	 * to not not cause bandwidth problems.
+	 */
+	return hweight8(crtc_state->active_planes & ~BIT(PLANE_CURSOR));
+}
+
+static unsigned int intel_bw_crtc_data_rate(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
+	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	unsigned int data_rate = 0;
+	enum plane_id plane_id;
+
+	for_each_plane_id_on_crtc(crtc, plane_id) {
+		/*
+		 * We assume cursors are small enough
+		 * to not not cause bandwidth problems.
+		 */
+		if (plane_id == PLANE_CURSOR)
+			continue;
+
+		data_rate += crtc_state->data_rate[plane_id];
+
+		if (DISPLAY_VER(i915) < 11)
+			data_rate += crtc_state->data_rate_y[plane_id];
+	}
+
+	return data_rate;
+}
+
+/* "Maximum Pipe Read Bandwidth" */
+static int intel_bw_crtc_min_cdclk(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
+	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+
+	if (DISPLAY_VER(i915) < 12)
+		return 0;
+
+	return DIV_ROUND_UP_ULL(mul_u32_u32(intel_bw_crtc_data_rate(crtc_state), 10), 512);
+}
+
+void intel_bw_crtc_update(struct intel_bw_state *bw_state,
+			  const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
+	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+
+	bw_state->data_rate[crtc->pipe] =
+		intel_bw_crtc_data_rate(crtc_state);
+	bw_state->num_active_planes[crtc->pipe] =
+		intel_bw_crtc_num_active_planes(crtc_state);
+
+	drm_dbg_kms(&i915->drm, "pipe %c data rate %u num active planes %u\n",
+		    pipe_name(crtc->pipe),
+		    bw_state->data_rate[crtc->pipe],
+		    bw_state->num_active_planes[crtc->pipe]);
+}
+
+static unsigned int intel_bw_num_active_planes(struct drm_i915_private *dev_priv,
+					       const struct intel_bw_state *bw_state)
+{
+	unsigned int num_active_planes = 0;
+	enum pipe pipe;
+
+	for_each_pipe(dev_priv, pipe)
+		num_active_planes += bw_state->num_active_planes[pipe];
+
+	return num_active_planes;
+}
+
+static unsigned int intel_bw_data_rate(struct drm_i915_private *dev_priv,
+				       const struct intel_bw_state *bw_state)
+{
+	unsigned int data_rate = 0;
+	enum pipe pipe;
+
+	for_each_pipe(dev_priv, pipe)
+		data_rate += bw_state->data_rate[pipe];
+
+	if (DISPLAY_VER(dev_priv) >= 13 && i915_vtd_active(dev_priv))
+		data_rate = DIV_ROUND_UP(data_rate * 105, 100);
+
+	return data_rate;
+}
+
+struct intel_bw_state *
+intel_atomic_get_old_bw_state(struct intel_atomic_state *state)
+{
+	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
+	struct intel_global_state *bw_state;
+
+	bw_state = intel_atomic_get_old_global_obj_state(state, &dev_priv->display.bw.obj);
+
+	return to_intel_bw_state(bw_state);
+}
+
+struct intel_bw_state *
+intel_atomic_get_new_bw_state(struct intel_atomic_state *state)
+{
+	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
+	struct intel_global_state *bw_state;
+
+	bw_state = intel_atomic_get_new_global_obj_state(state, &dev_priv->display.bw.obj);
+
+	return to_intel_bw_state(bw_state);
+}
+
+struct intel_bw_state *
+intel_atomic_get_bw_state(struct intel_atomic_state *state)
+{
+	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
+	struct intel_global_state *bw_state;
+
+	bw_state = intel_atomic_get_global_obj_state(state, &dev_priv->display.bw.obj);
+	if (IS_ERR(bw_state))
+		return ERR_CAST(bw_state);
+
+	return to_intel_bw_state(bw_state);
+}
+
+static bool intel_bw_state_changed(struct drm_i915_private *i915,
+				   const struct intel_bw_state *old_bw_state,
+				   const struct intel_bw_state *new_bw_state)
+{
+	enum pipe pipe;
+
+	for_each_pipe(i915, pipe) {
+		const struct intel_dbuf_bw *old_crtc_bw =
+			&old_bw_state->dbuf_bw[pipe];
+		const struct intel_dbuf_bw *new_crtc_bw =
+			&new_bw_state->dbuf_bw[pipe];
+		enum dbuf_slice slice;
+
+		for_each_dbuf_slice(i915, slice) {
+			if (old_crtc_bw->max_bw[slice] != new_crtc_bw->max_bw[slice] ||
+			    old_crtc_bw->active_planes[slice] != new_crtc_bw->active_planes[slice])
+				return true;
+		}
+
+		if (old_bw_state->min_cdclk[pipe] != new_bw_state->min_cdclk[pipe])
+			return true;
+	}
+
+	return false;
+}
+
+static void skl_plane_calc_dbuf_bw(struct intel_bw_state *bw_state,
+				   struct intel_crtc *crtc,
+				   enum plane_id plane_id,
+				   const struct skl_ddb_entry *ddb,
+				   unsigned int data_rate)
+{
+	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_dbuf_bw *crtc_bw = &bw_state->dbuf_bw[crtc->pipe];
+	unsigned int dbuf_mask = skl_ddb_dbuf_slice_mask(i915, ddb);
+	enum dbuf_slice slice;
+
+	/*
+	 * The arbiter can only really guarantee an
+	 * equal share of the total bw to each plane.
+	 */
+	for_each_dbuf_slice_in_mask(i915, slice, dbuf_mask) {
+		crtc_bw->max_bw[slice] = max(crtc_bw->max_bw[slice], data_rate);
+		crtc_bw->active_planes[slice] |= BIT(plane_id);
+	}
+}
+
+static void skl_crtc_calc_dbuf_bw(struct intel_bw_state *bw_state,
+				  const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
+	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_dbuf_bw *crtc_bw = &bw_state->dbuf_bw[crtc->pipe];
+	enum plane_id plane_id;
+
+	memset(crtc_bw, 0, sizeof(*crtc_bw));
+
+	if (!crtc_state->hw.active)
+		return;
+
+	for_each_plane_id_on_crtc(crtc, plane_id) {
+		/*
+		 * We assume cursors are small enough
+		 * to not cause bandwidth problems.
+		 */
+		if (plane_id == PLANE_CURSOR)
+			continue;
+
+		skl_plane_calc_dbuf_bw(bw_state, crtc, plane_id,
+				       &crtc_state->wm.skl.plane_ddb[plane_id],
+				       crtc_state->data_rate[plane_id]);
+
+		if (DISPLAY_VER(i915) < 11)
+			skl_plane_calc_dbuf_bw(bw_state, crtc, plane_id,
+					       &crtc_state->wm.skl.plane_ddb_y[plane_id],
+					       crtc_state->data_rate[plane_id]);
+	}
+}
+
+/* "Maximum Data Buffer Bandwidth" */
+static int
+intel_bw_dbuf_min_cdclk(struct drm_i915_private *i915,
+			const struct intel_bw_state *bw_state)
+{
+	unsigned int total_max_bw = 0;
+	enum dbuf_slice slice;
+
+	for_each_dbuf_slice(i915, slice) {
+		int num_active_planes = 0;
+		unsigned int max_bw = 0;
+		enum pipe pipe;
+
+		/*
+		 * The arbiter can only really guarantee an
+		 * equal share of the total bw to each plane.
+		 */
+		for_each_pipe(i915, pipe) {
+			const struct intel_dbuf_bw *crtc_bw = &bw_state->dbuf_bw[pipe];
+
+			max_bw = max(crtc_bw->max_bw[slice], max_bw);
+			num_active_planes += hweight8(crtc_bw->active_planes[slice]);
+		}
+		max_bw *= num_active_planes;
+
+		total_max_bw = max(total_max_bw, max_bw);
+	}
+
+	return DIV_ROUND_UP(total_max_bw, 64);
+}
+
+int intel_bw_min_cdclk(struct drm_i915_private *i915,
+		       const struct intel_bw_state *bw_state)
+{
+	enum pipe pipe;
+	int min_cdclk;
+
+	min_cdclk = intel_bw_dbuf_min_cdclk(i915, bw_state);
+
+	for_each_pipe(i915, pipe)
+		min_cdclk = max(bw_state->min_cdclk[pipe], min_cdclk);
+
+	return min_cdclk;
+}
+
+int intel_bw_calc_min_cdclk(struct intel_atomic_state *state,
+			    bool *need_cdclk_calc)
+{
+	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
+	struct intel_bw_state *new_bw_state = NULL;
+	const struct intel_bw_state *old_bw_state = NULL;
+	const struct intel_cdclk_state *cdclk_state;
+	const struct intel_crtc_state *crtc_state;
+	int old_min_cdclk, new_min_cdclk;
+	struct intel_crtc *crtc;
+	int i;
+
+	if (DISPLAY_VER(dev_priv) < 9)
+		return 0;
+
+	for_each_new_intel_crtc_in_state(state, crtc, crtc_state, i) {
+		new_bw_state = intel_atomic_get_bw_state(state);
+		if (IS_ERR(new_bw_state))
+			return PTR_ERR(new_bw_state);
+
+		old_bw_state = intel_atomic_get_old_bw_state(state);
+
+		skl_crtc_calc_dbuf_bw(new_bw_state, crtc_state);
+
+		new_bw_state->min_cdclk[crtc->pipe] =
+			intel_bw_crtc_min_cdclk(crtc_state);
+	}
+
+	if (!old_bw_state)
+		return 0;
+
+	if (intel_bw_state_changed(dev_priv, old_bw_state, new_bw_state)) {
+		int ret = intel_atomic_lock_global_state(&new_bw_state->base);
+		if (ret)
+			return ret;
+	}
+
+	old_min_cdclk = intel_bw_min_cdclk(dev_priv, old_bw_state);
+	new_min_cdclk = intel_bw_min_cdclk(dev_priv, new_bw_state);
+
+	/*
+	 * No need to check against the cdclk state if
+	 * the min cdclk doesn't increase.
+	 *
+	 * Ie. we only ever increase the cdclk due to bandwidth
+	 * requirements. This can reduce back and forth
+	 * display blinking due to constant cdclk changes.
+	 */
+	if (new_min_cdclk <= old_min_cdclk)
+		return 0;
+
+	cdclk_state = intel_atomic_get_cdclk_state(state);
+	if (IS_ERR(cdclk_state))
+		return PTR_ERR(cdclk_state);
+
+	/*
+	 * No need to recalculate the cdclk state if
+	 * the min cdclk doesn't increase.
+	 *
+	 * Ie. we only ever increase the cdclk due to bandwidth
+	 * requirements. This can reduce back and forth
+	 * display blinking due to constant cdclk changes.
+	 */
+	if (new_min_cdclk <= cdclk_state->bw_min_cdclk)
+		return 0;
+
+	drm_dbg_kms(&dev_priv->drm,
+		    "new bandwidth min cdclk (%d kHz) > old min cdclk (%d kHz)\n",
+		    new_min_cdclk, cdclk_state->bw_min_cdclk);
+	*need_cdclk_calc = true;
+
+	return 0;
+}
+
+static u16 icl_qgv_points_mask(struct drm_i915_private *i915)
+{
+	unsigned int num_psf_gv_points = i915->display.bw.max[0].num_psf_gv_points;
+	unsigned int num_qgv_points = i915->display.bw.max[0].num_qgv_points;
+	u16 qgv_points = 0, psf_points = 0;
+
+	/*
+	 * We can _not_ use the whole ADLS_QGV_PT_MASK here, as PCode rejects
+	 * it with failure if we try masking any unadvertised points.
+	 * So need to operate only with those returned from PCode.
+	 */
+	if (num_qgv_points > 0)
+		qgv_points = GENMASK(num_qgv_points - 1, 0);
+
+	if (num_psf_gv_points > 0)
+		psf_points = GENMASK(num_psf_gv_points - 1, 0);
+
+	return ICL_PCODE_REQ_QGV_PT(qgv_points) | ADLS_PCODE_REQ_PSF_PT(psf_points);
+}
+
+static int intel_bw_check_data_rate(struct intel_atomic_state *state, bool *changed)
+{
+	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	const struct intel_crtc_state *new_crtc_state, *old_crtc_state;
+	struct intel_crtc *crtc;
+	int i;
+
+	for_each_oldnew_intel_crtc_in_state(state, crtc, old_crtc_state,
+					    new_crtc_state, i) {
+		unsigned int old_data_rate =
+			intel_bw_crtc_data_rate(old_crtc_state);
+		unsigned int new_data_rate =
+			intel_bw_crtc_data_rate(new_crtc_state);
+		unsigned int old_active_planes =
+			intel_bw_crtc_num_active_planes(old_crtc_state);
+		unsigned int new_active_planes =
+			intel_bw_crtc_num_active_planes(new_crtc_state);
+		struct intel_bw_state *new_bw_state;
+
+		/*
+		 * Avoid locking the bw state when
+		 * nothing significant has changed.
+		 */
+		if (old_data_rate == new_data_rate &&
+		    old_active_planes == new_active_planes)
+			continue;
+
+		new_bw_state = intel_atomic_get_bw_state(state);
+		if (IS_ERR(new_bw_state))
+			return PTR_ERR(new_bw_state);
+
+		new_bw_state->data_rate[crtc->pipe] = new_data_rate;
+		new_bw_state->num_active_planes[crtc->pipe] = new_active_planes;
+
+		*changed = true;
+
+		drm_dbg_kms(&i915->drm,
+			    "[CRTC:%d:%s] data rate %u num active planes %u\n",
+			    crtc->base.base.id, crtc->base.name,
+			    new_bw_state->data_rate[crtc->pipe],
+			    new_bw_state->num_active_planes[crtc->pipe]);
+	}
+
+	return 0;
+}
+
+int intel_bw_atomic_check(struct intel_atomic_state *state)
+{
+	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
+	const struct intel_bw_state *old_bw_state;
+	struct intel_bw_state *new_bw_state;
+	unsigned int data_rate;
+	unsigned int num_active_planes;
+	int i, ret;
+	u16 qgv_points = 0, psf_points = 0;
+	unsigned int max_bw_point = 0, max_bw = 0;
+	unsigned int num_qgv_points = dev_priv->display.bw.max[0].num_qgv_points;
+	unsigned int num_psf_gv_points = dev_priv->display.bw.max[0].num_psf_gv_points;
+	bool changed = false;
+
+	/* FIXME earlier gens need some checks too */
+	if (DISPLAY_VER(dev_priv) < 11)
+		return 0;
+
+	ret = intel_bw_check_data_rate(state, &changed);
+	if (ret)
+		return ret;
+
+	old_bw_state = intel_atomic_get_old_bw_state(state);
+	new_bw_state = intel_atomic_get_new_bw_state(state);
+
+	if (new_bw_state &&
+	    intel_can_enable_sagv(dev_priv, old_bw_state) !=
+	    intel_can_enable_sagv(dev_priv, new_bw_state))
+		changed = true;
+
+	/*
+	 * If none of our inputs (data rates, number of active
+	 * planes, SAGV yes/no) changed then nothing to do here.
+	 */
+	if (!changed)
+		return 0;
+
+	ret = intel_atomic_lock_global_state(&new_bw_state->base);
+	if (ret)
+		return ret;
+
+	data_rate = intel_bw_data_rate(dev_priv, new_bw_state);
+	data_rate = DIV_ROUND_UP(data_rate, 1000);
+
+	num_active_planes = intel_bw_num_active_planes(dev_priv, new_bw_state);
+
+	for (i = 0; i < num_qgv_points; i++) {
+		unsigned int max_data_rate;
+
+		if (DISPLAY_VER(dev_priv) > 11)
+			max_data_rate = tgl_max_bw(dev_priv, num_active_planes, i);
+		else
+			max_data_rate = icl_max_bw(dev_priv, num_active_planes, i);
+		/*
+		 * We need to know which qgv point gives us
+		 * maximum bandwidth in order to disable SAGV
+		 * if we find that we exceed SAGV block time
+		 * with watermarks. By that moment we already
+		 * have those, as it is calculated earlier in
+		 * intel_atomic_check,
+		 */
+		if (max_data_rate > max_bw) {
+			max_bw_point = i;
+			max_bw = max_data_rate;
+		}
+		if (max_data_rate >= data_rate)
+			qgv_points |= BIT(i);
+
+		drm_dbg_kms(&dev_priv->drm, "QGV point %d: max bw %d required %d\n",
+			    i, max_data_rate, data_rate);
+	}
+
+	for (i = 0; i < num_psf_gv_points; i++) {
+		unsigned int max_data_rate = adl_psf_bw(dev_priv, i);
+
+		if (max_data_rate >= data_rate)
+			psf_points |= BIT(i);
+
+		drm_dbg_kms(&dev_priv->drm, "PSF GV point %d: max bw %d"
+			    " required %d\n",
+			    i, max_data_rate, data_rate);
+	}
+
+	/*
+	 * BSpec states that we always should have at least one allowed point
+	 * left, so if we couldn't - simply reject the configuration for obvious
+	 * reasons.
+	 */
+	if (qgv_points == 0) {
+		drm_dbg_kms(&dev_priv->drm, "No QGV points provide sufficient memory"
+			    " bandwidth %d for display configuration(%d active planes).\n",
+			    data_rate, num_active_planes);
+		return -EINVAL;
+	}
+
+	if (num_psf_gv_points > 0 && psf_points == 0) {
+		drm_dbg_kms(&dev_priv->drm, "No PSF GV points provide sufficient memory"
+			    " bandwidth %d for display configuration(%d active planes).\n",
+			    data_rate, num_active_planes);
+		return -EINVAL;
+	}
+
+	/*
+	 * Leave only single point with highest bandwidth, if
+	 * we can't enable SAGV due to the increased memory latency it may
+	 * cause.
+	 */
+	if (!intel_can_enable_sagv(dev_priv, new_bw_state)) {
+		qgv_points = BIT(max_bw_point);
+		drm_dbg_kms(&dev_priv->drm, "No SAGV, using single QGV point %d\n",
+			    max_bw_point);
+	}
+
+	/*
+	 * We store the ones which need to be masked as that is what PCode
+	 * actually accepts as a parameter.
+	 */
+	new_bw_state->qgv_points_mask =
+		~(ICL_PCODE_REQ_QGV_PT(qgv_points) |
+		  ADLS_PCODE_REQ_PSF_PT(psf_points)) &
+		icl_qgv_points_mask(dev_priv);
+
+	/*
+	 * If the actual mask had changed we need to make sure that
+	 * the commits are serialized(in case this is a nomodeset, nonblocking)
+	 */
+	if (new_bw_state->qgv_points_mask != old_bw_state->qgv_points_mask) {
+		ret = intel_atomic_serialize_global_state(&new_bw_state->base);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static struct intel_global_state *
+intel_bw_duplicate_state(struct intel_global_obj *obj)
+{
+	struct intel_bw_state *state;
+
+	state = kmemdup(obj->state, sizeof(*state), GFP_KERNEL);
+	if (!state)
+		return NULL;
+
+	return &state->base;
+}
+
+static void intel_bw_destroy_state(struct intel_global_obj *obj,
+				   struct intel_global_state *state)
+{
+	kfree(state);
+}
+
+static const struct intel_global_state_funcs intel_bw_funcs = {
+	.atomic_duplicate_state = intel_bw_duplicate_state,
+	.atomic_destroy_state = intel_bw_destroy_state,
+};
+
+int intel_bw_init(struct drm_i915_private *dev_priv)
+{
+	struct intel_bw_state *state;
+
+	state = kzalloc(sizeof(*state), GFP_KERNEL);
+	if (!state)
+		return -ENOMEM;
+
+	intel_atomic_global_obj_init(dev_priv, &dev_priv->display.bw.obj,
+				     &state->base, &intel_bw_funcs);
+
+	return 0;
+}