[PATCH 0/2] MIPS: SiByte: Handle PCI DMA with 64-bit memory addressing

[PATCH 0/2] MIPS: SiByte: Handle PCI DMA with 64-bit memory addressing

[Maciej W. Rozycki]

Hi,

 This mini patch series enables correct support for DMA in the presence of 
memory outside the 32-bit address range with the Broadcom SiByte SOCs and 
the relevant development boards.

 There is a quirk in the BCM1250, BCM1125 and BCM1125H SOCs in that their 
onchip 32-bit PCI host bridge does not support DAC, however the HT link 
(where available) does support 40-bit addressing as per the HT spec.  
Therefore the first patch sets the bus mask accordingly, and then the 
second patch enables swiotlb.  See individual change descriptions for 
additional details; there's also a further discussion alongside.

 This has been verified with a Broadcom SWARM board equipped with 3200MiB 
of RAM (2176MiB of which the address decoder in the SOC maps above 4GiB), 
a pair of DEFPA FDDI adapters and an XHCI USB adapter.  There were also 
some other PCI and PCIe devices present in the system, though not actively 
used beyond being probed at boot, and none has shown any symptoms of 
breakage.

 Thanks to Christoph for making this change so easy with his recent work!

 Please apply.

  Maciej

[PATCH 1/2] MIPS: SiByte: Set 32-bit bus mask for BCM1250 PCI

[Maciej W. Rozycki]

The Broadcom SiByte BCM1250, BCM1125H and BCM1125 SOCs have an onchip 
32-bit PCI host bridge, and the two former SOCs also have an onchip HT 
host bridge.  The HT host bridge, where present, appears in the PCI 
configuration space as if it was a device on the 32-bit PCI bus behind 
the PCI host bridge, however at the hardware level its signals are 
routed separately, so these two devices are actually peer host bridges.

As documented[1] and observed in reality the 32-bit PCI host bridge does 
not support 64-bit addressing as it does not support the Dual Address 
Cycle (DAC) PCI command, and naturally, being 32-bit only, it has no 
means to carry the high 32 address bits otherwise.  However the DRAM 
controller also included in the SOC supports memory amounts of up to 
16GiB, and due to how the address decoder has been wired in the SOC any 
memory beyond 1GiB is actually mapped starting from 4GiB physical up, 
that is beyond the 32-bit addressable limit.  Consequently if the 
maximum amount of memory has been installed, then it will span up to 
19GiB.

Contrariwise, the HT host bridge does support full 40-bit addressing 
defined by the HyperTransport (formerly LDT) specification the bridge 
adheres to, depending on the peripherals revision of the SOC[2] either 
revision 0.17[3] or revision 1.03[4].  This allows addressing any and 
all memory installed, and well beyond.

Set the bus mask then to limit DMA addressing to 32 bits for all the 
devices down the 32-bit PCI host bridge, excluding however any devices 
that are down the HT host bridge.

References:

[1] "BCM1250/BCM1125/BCM1125H User Manual", Revision 1250_1125-UM100-R, 
    Broadcom Corporation, 21 Oct 2002, Section 8: "PCI Bus and 
    HyperTransport Fabric", "Introduction", p. 190

[2] same, Table 140: "HyperTransport Configuration Header (Type 1)", p. 
    245

[3] "Lightning Data Transport IO Specification", Revision 0.17, Advanced 
    Micro Devices, 21 Jan 2000, Section 3.2.1.2 "Command Packet", p. 8

[4] "HyperTransport I/O Link Specification", Revision 1.03, 
    HyperTransport Technology Consortium, 10 Oct 2001, Section 3.2.1.2 
    "Request Packet", pp. 27-28

Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
---
Hi,

 This has been verified with a Broadcom SWARM board and an XHCI USB 32-bit 
PCI option board plugged to one of the mainboard's 32-bit slots wired to 
the PCI host bridge, and then a flash storage device plugged to adapter's 
USB socket.

 With 2/2 applied first so that the bus mask is respected and a diagnostic 
patch for `dma_direct_alloc' made to debug an earlier issue also applied
the system shows these messages upon boot:

xhci_hcd 0000:03:00.0: assign IRQ: got 56
xhci_hcd 0000:03:00.0: enabling bus mastering
xhci_hcd 0000:03:00.0: xHCI Host Controller
xhci_hcd 0000:03:00.0: new USB bus registered, assigned bus number 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b48000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b4c000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b54000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b58000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b5c000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b60000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b64000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b68000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b6c000
xhci_hcd 0000:03:00.0: hcc params 0x014051c7 hci version 0x100 quirks 0x0000000100000090
xhci_hcd 0000:03:00.0: enabling Mem-Wr-Inval
usb usb1: New USB device found, idVendor=1d6b, idProduct=0002, bcdDevice= 4.19
usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1
usb usb1: Product: xHCI Host Controller
usb usb1: Manufacturer: Linux 4.19.0 xhci-hcd
usb usb1: SerialNumber: 0000:03:00.0
hub 1-0:1.0: USB hub found
hub 1-0:1.0: 2 ports detected
xhci_hcd 0000:03:00.0: xHCI Host Controller
xhci_hcd 0000:03:00.0: new USB bus registered, assigned bus number 2
xhci_hcd 0000:03:00.0: Host supports USB 3.0  SuperSpeed
xhci_hcd 0000:03:00.0: Host took too long to start, waited 16000 microseconds.
xhci_hcd 0000:03:00.0: startup error -19
xhci_hcd 0000:03:00.0: USB bus 2 deregistered
xhci_hcd 0000:03:00.0: remove, state 1
usb usb1: USB disconnect, device number 1
xhci_hcd 0000:03:00.0: USB bus 1 deregistered
usbcore: registered new interface driver usb-storage

As you can see `dma_direct_alloc' hands out addresses outside the 32-bit 
physical range and then the USB host controller cannot be communicated to.  
Also some memory has likely got corrupted (or random MMIO poked at).

 Then with this change applied the messages change to these:

xhci_hcd 0000:03:00.0: assign IRQ: got 56
xhci_hcd 0000:03:00.0: enabling bus mastering
xhci_hcd 0000:03:00.0: xHCI Host Controller
xhci_hcd 0000:03:00.0: new USB bus registered, assigned bus number 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca050000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca054000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca058000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca05c000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca060000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca064000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca068000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca06c000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca070000
xhci_hcd 0000:03:00.0: hcc params 0x014051c7 hci version 0x100 quirks 0x0000000100000090
xhci_hcd 0000:03:00.0: enabling Mem-Wr-Inval
usb usb1: New USB device found, idVendor=1d6b, idProduct=0002, bcdDevice= 4.19
usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1
usb usb1: Product: xHCI Host Controller
usb usb1: Manufacturer: Linux 4.19.0 xhci-hcd
usb usb1: SerialNumber: 0000:03:00.0
hub 1-0:1.0: USB hub found
hub 1-0:1.0: 2 ports detected
xhci_hcd 0000:03:00.0: xHCI Host Controller
xhci_hcd 0000:03:00.0: new USB bus registered, assigned bus number 2
xhci_hcd 0000:03:00.0: Host supports USB 3.0  SuperSpeed
usb usb2: We don't know the algorithms for LPM for this host, disabling LPM.
usb usb2: New USB device found, idVendor=1d6b, idProduct=0003, bcdDevice= 4.19
usb usb2: New USB device strings: Mfr=3, Product=2, SerialNumber=1
usb usb2: Product: xHCI Host Controller
usb usb2: Manufacturer: Linux 4.19.0 xhci-hcd
usb usb2: SerialNumber: 0000:03:00.0
hub 2-0:1.0: USB hub found
hub 2-0:1.0: 2 ports detected
usbcore: registered new interface driver usb-storage

and then later on these:

xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca074000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca078000
usb 2-1: new SuperSpeed Gen 1 USB device number 2 using xhci_hcd
usb 2-1: New USB device found, idVendor=0781, idProduct=5583, bcdDevice= 1.00
usb 2-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
usb 2-1: Product: Ultra Fit
usb 2-1: Manufacturer: SanDisk
usb 2-1: SerialNumber: 4C531001340112110513
usb-storage 2-1:1.0: USB Mass Storage device detected
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca07c000
scsi host5: usb-storage 2-1:1.0
scsi 5:0:0:0: Direct-Access     SanDisk  Ultra Fit        1.00 PQ: 0 ANSI: 6
sd 5:0:0:0: Attached scsi generic sg2 type 0
sd 5:0:0:0: [sdc] 120127488 512-byte logical blocks: (61.5 GB/57.3 GiB)
sd 5:0:0:0: [sdc] Write Protect is off
sd 5:0:0:0: [sdc] Mode Sense: 43 00 00 00
sd 5:0:0:0: [sdc] Write cache: disabled, read cache: enabled, doesn't support DPO or FUA
 sdc: sdc1 sdc2
sd 5:0:0:0: [sdc] Attached SCSI removable disk

moving addresses handed out by `dma_direct_alloc' to the 32-bit physical 
range and consequently both the USB host controller and the storage device 
behind responsive.

 If moved to one of the PCI slots down the HT link it works correctly with 
64-bit addressing:

xhci_hcd 0000:03:00.0: assign IRQ: got 58
xhci_hcd 0000:03:00.0: enabling bus mastering
xhci_hcd 0000:03:00.0: xHCI Host Controller
xhci_hcd 0000:03:00.0: new USB bus registered, assigned bus number 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185aec000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185af0000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185af8000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185afc000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b00000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b04000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b08000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b0c000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b10000
xhci_hcd 0000:03:00.0: hcc params 0x014051c7 hci version 0x100 quirks 0x0000000100000090
xhci_hcd 0000:03:00.0: enabling Mem-Wr-Inval
usb usb1: New USB device found, idVendor=1d6b, idProduct=0002, bcdDevice= 4.20
usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1
usb usb1: Product: xHCI Host Controller
usb usb1: Manufacturer: Linux 4.20.0-rc1 xhci-hcd
usb usb1: SerialNumber: 0000:03:00.0
hub 1-0:1.0: USB hub found
hub 1-0:1.0: 2 ports detected
xhci_hcd 0000:03:00.0: xHCI Host Controller
xhci_hcd 0000:03:00.0: new USB bus registered, assigned bus number 2
xhci_hcd 0000:03:00.0: Host supports USB 3.0  SuperSpeed
usb usb2: We don't know the algorithms for LPM for this host, disabling LPM.
usb usb2: New USB device found, idVendor=1d6b, idProduct=0003, bcdDevice= 4.20
usb usb2: New USB device strings: Mfr=3, Product=2, SerialNumber=1
usb usb2: Product: xHCI Host Controller
usb usb2: Manufacturer: Linux 4.20.0-rc1 xhci-hcd
usb usb2: SerialNumber: 0000:03:00.0
hub 2-0:1.0: USB hub found
hub 2-0:1.0: 2 ports detected
usbcore: registered new interface driver usb-storage

and:

xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000001830d4000
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000001830d8000
usb 2-1: new SuperSpeed Gen 1 USB device number 2 using xhci_hcd
usb 2-1: New USB device found, idVendor=0781, idProduct=5583, bcdDevice= 1.00
usb 2-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
usb 2-1: Product: Ultra Fit
usb 2-1: Manufacturer: SanDisk
usb 2-1: SerialNumber: 4C531001340112110513
usb-storage 2-1:1.0: USB Mass Storage device detected
xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000001830dc000
scsi host5: usb-storage 2-1:1.0
scsi 5:0:0:0: Direct-Access     SanDisk  Ultra Fit        1.00 PQ: 0 ANSI: 6
sd 5:0:0:0: Attached scsi generic sg2 type 0
sd 5:0:0:0: [sdc] 120127488 512-byte logical blocks: (61.5 GB/57.3 GiB)
sd 5:0:0:0: [sdc] Write Protect is off
sd 5:0:0:0: [sdc] Mode Sense: 43 00 00 00
sd 5:0:0:0: [sdc] Write cache: disabled, read cache: enabled, doesn't support DPO or FUA

confirming that indeed the devices down the PCI-HT bridge don't need the 
bus mask to be set.

 NB the reason for the bus address of the XHCI PCI device remaining the 
same (0000:03:00.0) in the logs above regardless of the slot it has been 
plugged in is that the option board actually includes a PCI-PCIe bridge 
the actual XHCI device is behind, and that secondary PCIe link happens to 
get the same bus number in enumeration.

 The BCM1480 SOC may or may not require a similar quirk, but I have no 
documentation nor hardware to check with, so let's leave it for someone 
who can verify it.  My understanding is it implements a PCI-X rather than 
a plain PCI host bridge, so the limitation may well have been lifted in 
the redesign.

  Maciej
---
 arch/mips/pci/fixup-sb1250.c |   48 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 48 insertions(+)

linux-mips-sibyte-sb1250-bus-dma-mask.diff
Index: linux-20181104-swarm64-eb/arch/mips/pci/fixup-sb1250.c
===================================================================
--- linux-20181104-swarm64-eb.orig/arch/mips/pci/fixup-sb1250.c
+++ linux-20181104-swarm64-eb/arch/mips/pci/fixup-sb1250.c
@@ -1,6 +1,7 @@
 /*
  *	Copyright (C) 2004, 2006  MIPS Technologies, Inc.  All rights reserved.
  *	    Author:	Maciej W. Rozycki <macro@mips.com>
+ *	Copyright (C) 2018  Maciej W. Rozycki
  *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
@@ -8,6 +9,7 @@
  *	2 of the License, or (at your option) any later version.
  */
 
+#include <linux/dma-mapping.h>
 #include <linux/pci.h>
 
 /*
@@ -22,6 +24,52 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_SI
 			quirk_sb1250_pci);
 
 /*
+ * The BCM1250, etc. PCI host bridge does not support DAC on its 32-bit
+ * bus, so we set the bus's DMA mask accordingly.  However the HT link
+ * down the artificial PCI-HT bridge supports 40-bit addressing and the
+ * SP1011 HT-PCI bridge downstream supports both DAC and a 64-bit bus
+ * width, so we record the PCI-HT bridge's secondary and subordinate bus
+ * numbers and do not set the mask for devices present in the inclusive
+ * range of those.
+ */
+struct sb1250_bus_dma_mask_exclude {
+	bool set;
+	unsigned char start;
+	unsigned char end;
+};
+
+static int sb1250_bus_dma_mask(struct pci_dev *dev, void *data)
+{
+	struct sb1250_bus_dma_mask_exclude *exclude = data;
+
+	if (!exclude->set && (dev->vendor == PCI_VENDOR_ID_SIBYTE &&
+			      dev->device == PCI_DEVICE_ID_BCM1250_HT)) {
+		exclude->start = dev->subordinate->number;
+		exclude->end = pci_bus_max_busnr(dev->subordinate);
+		exclude->set = true;
+		dev_dbg(&dev->dev, "not disabling DAC for [bus %02x-%02x]",
+			exclude->start, exclude->end);
+	} else if (!exclude->set ||
+		   (exclude->set && (dev->bus->number < exclude->start ||
+				     dev->bus->number > exclude->end))) {
+		dev_dbg(&dev->dev, "disabling DAC for device");
+		dev->dev.bus_dma_mask = DMA_BIT_MASK(32);
+	} else {
+		dev_dbg(&dev->dev, "not disabling DAC for device");
+	}
+	return 0;
+}
+
+static void quirk_sb1250_pci_dac(struct pci_dev *dev)
+{
+	struct sb1250_bus_dma_mask_exclude exclude = { .set = false };
+
+	pci_walk_bus(dev->bus, sb1250_bus_dma_mask, &exclude);
+}
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_SIBYTE, PCI_DEVICE_ID_BCM1250_PCI,
+			quirk_sb1250_pci_dac);
+
+/*
  * The BCM1250, etc. PCI/HT bridge reports as a host bridge.
  */
 static void quirk_sb1250_ht(struct pci_dev *dev)

Re: [PATCH 1/2] MIPS: SiByte: Set 32-bit bus mask for BCM1250 PCI

[Christoph Hellwig]

On Wed, Nov 07, 2018 at 12:08:23AM +0000, Maciej W. Rozycki wrote:
> The Broadcom SiByte BCM1250, BCM1125H and BCM1125 SOCs have an onchip 
> 32-bit PCI host bridge, and the two former SOCs also have an onchip HT 
> host bridge.  The HT host bridge, where present, appears in the PCI 
> configuration space as if it was a device on the 32-bit PCI bus behind 
> the PCI host bridge, however at the hardware level its signals are 
> routed separately, so these two devices are actually peer host bridges.
> 
> As documented[1] and observed in reality the 32-bit PCI host bridge does 
> not support 64-bit addressing as it does not support the Dual Address 
> Cycle (DAC) PCI command, and naturally, being 32-bit only, it has no 
> means to carry the high 32 address bits otherwise.  However the DRAM 
> controller also included in the SOC supports memory amounts of up to 
> 16GiB, and due to how the address decoder has been wired in the SOC any 
> memory beyond 1GiB is actually mapped starting from 4GiB physical up, 
> that is beyond the 32-bit addressable limit.  Consequently if the 
> maximum amount of memory has been installed, then it will span up to 
> 19GiB.
> 
> Contrariwise, the HT host bridge does support full 40-bit addressing 
> defined by the HyperTransport (formerly LDT) specification the bridge 
> adheres to, depending on the peripherals revision of the SOC[2] either 
> revision 0.17[3] or revision 1.03[4].  This allows addressing any and 
> all memory installed, and well beyond.
> 
> Set the bus mask then to limit DMA addressing to 32 bits for all the 
> devices down the 32-bit PCI host bridge, excluding however any devices 
> that are down the HT host bridge.
> 
> References:
> 
> [1] "BCM1250/BCM1125/BCM1125H User Manual", Revision 1250_1125-UM100-R, 
>     Broadcom Corporation, 21 Oct 2002, Section 8: "PCI Bus and 
>     HyperTransport Fabric", "Introduction", p. 190
> 
> [2] same, Table 140: "HyperTransport Configuration Header (Type 1)", p. 
>     245
> 
> [3] "Lightning Data Transport IO Specification", Revision 0.17, Advanced 
>     Micro Devices, 21 Jan 2000, Section 3.2.1.2 "Command Packet", p. 8
> 
> [4] "HyperTransport I/O Link Specification", Revision 1.03, 
>     HyperTransport Technology Consortium, 10 Oct 2001, Section 3.2.1.2 
>     "Request Packet", pp. 27-28
> 
> Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
> ---
> Hi,
> 
>  This has been verified with a Broadcom SWARM board and an XHCI USB 32-bit 
> PCI option board plugged to one of the mainboard's 32-bit slots wired to 
> the PCI host bridge, and then a flash storage device plugged to adapter's 
> USB socket.
> 
>  With 2/2 applied first so that the bus mask is respected and a diagnostic 
> patch for `dma_direct_alloc' made to debug an earlier issue also applied
> the system shows these messages upon boot:
> 
> xhci_hcd 0000:03:00.0: assign IRQ: got 56
> xhci_hcd 0000:03:00.0: enabling bus mastering
> xhci_hcd 0000:03:00.0: xHCI Host Controller
> xhci_hcd 0000:03:00.0: new USB bus registered, assigned bus number 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b48000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b4c000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b54000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b58000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b5c000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b60000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b64000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b68000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b6c000
> xhci_hcd 0000:03:00.0: hcc params 0x014051c7 hci version 0x100 quirks 0x0000000100000090
> xhci_hcd 0000:03:00.0: enabling Mem-Wr-Inval
> usb usb1: New USB device found, idVendor=1d6b, idProduct=0002, bcdDevice= 4.19
> usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1
> usb usb1: Product: xHCI Host Controller
> usb usb1: Manufacturer: Linux 4.19.0 xhci-hcd
> usb usb1: SerialNumber: 0000:03:00.0
> hub 1-0:1.0: USB hub found
> hub 1-0:1.0: 2 ports detected
> xhci_hcd 0000:03:00.0: xHCI Host Controller
> xhci_hcd 0000:03:00.0: new USB bus registered, assigned bus number 2
> xhci_hcd 0000:03:00.0: Host supports USB 3.0  SuperSpeed
> xhci_hcd 0000:03:00.0: Host took too long to start, waited 16000 microseconds.
> xhci_hcd 0000:03:00.0: startup error -19
> xhci_hcd 0000:03:00.0: USB bus 2 deregistered
> xhci_hcd 0000:03:00.0: remove, state 1
> usb usb1: USB disconnect, device number 1
> xhci_hcd 0000:03:00.0: USB bus 1 deregistered
> usbcore: registered new interface driver usb-storage
> 
> As you can see `dma_direct_alloc' hands out addresses outside the 32-bit 
> physical range and then the USB host controller cannot be communicated to.  
> Also some memory has likely got corrupted (or random MMIO poked at).
> 
>  Then with this change applied the messages change to these:
> 
> xhci_hcd 0000:03:00.0: assign IRQ: got 56
> xhci_hcd 0000:03:00.0: enabling bus mastering
> xhci_hcd 0000:03:00.0: xHCI Host Controller
> xhci_hcd 0000:03:00.0: new USB bus registered, assigned bus number 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca050000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca054000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca058000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca05c000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca060000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca064000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca068000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca06c000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca070000
> xhci_hcd 0000:03:00.0: hcc params 0x014051c7 hci version 0x100 quirks 0x0000000100000090
> xhci_hcd 0000:03:00.0: enabling Mem-Wr-Inval
> usb usb1: New USB device found, idVendor=1d6b, idProduct=0002, bcdDevice= 4.19
> usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1
> usb usb1: Product: xHCI Host Controller
> usb usb1: Manufacturer: Linux 4.19.0 xhci-hcd
> usb usb1: SerialNumber: 0000:03:00.0
> hub 1-0:1.0: USB hub found
> hub 1-0:1.0: 2 ports detected
> xhci_hcd 0000:03:00.0: xHCI Host Controller
> xhci_hcd 0000:03:00.0: new USB bus registered, assigned bus number 2
> xhci_hcd 0000:03:00.0: Host supports USB 3.0  SuperSpeed
> usb usb2: We don't know the algorithms for LPM for this host, disabling LPM.
> usb usb2: New USB device found, idVendor=1d6b, idProduct=0003, bcdDevice= 4.19
> usb usb2: New USB device strings: Mfr=3, Product=2, SerialNumber=1
> usb usb2: Product: xHCI Host Controller
> usb usb2: Manufacturer: Linux 4.19.0 xhci-hcd
> usb usb2: SerialNumber: 0000:03:00.0
> hub 2-0:1.0: USB hub found
> hub 2-0:1.0: 2 ports detected
> usbcore: registered new interface driver usb-storage
> 
> and then later on these:
> 
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca074000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca078000
> usb 2-1: new SuperSpeed Gen 1 USB device number 2 using xhci_hcd
> usb 2-1: New USB device found, idVendor=0781, idProduct=5583, bcdDevice= 1.00
> usb 2-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
> usb 2-1: Product: Ultra Fit
> usb 2-1: Manufacturer: SanDisk
> usb 2-1: SerialNumber: 4C531001340112110513
> usb-storage 2-1:1.0: USB Mass Storage device detected
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000000ca07c000
> scsi host5: usb-storage 2-1:1.0
> scsi 5:0:0:0: Direct-Access     SanDisk  Ultra Fit        1.00 PQ: 0 ANSI: 6
> sd 5:0:0:0: Attached scsi generic sg2 type 0
> sd 5:0:0:0: [sdc] 120127488 512-byte logical blocks: (61.5 GB/57.3 GiB)
> sd 5:0:0:0: [sdc] Write Protect is off
> sd 5:0:0:0: [sdc] Mode Sense: 43 00 00 00
> sd 5:0:0:0: [sdc] Write cache: disabled, read cache: enabled, doesn't support DPO or FUA
>  sdc: sdc1 sdc2
> sd 5:0:0:0: [sdc] Attached SCSI removable disk
> 
> moving addresses handed out by `dma_direct_alloc' to the 32-bit physical 
> range and consequently both the USB host controller and the storage device 
> behind responsive.
> 
>  If moved to one of the PCI slots down the HT link it works correctly with 
> 64-bit addressing:
> 
> xhci_hcd 0000:03:00.0: assign IRQ: got 58
> xhci_hcd 0000:03:00.0: enabling bus mastering
> xhci_hcd 0000:03:00.0: xHCI Host Controller
> xhci_hcd 0000:03:00.0: new USB bus registered, assigned bus number 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185aec000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185af0000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185af8000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185afc000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b00000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b04000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b08000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b0c000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a800000185b10000
> xhci_hcd 0000:03:00.0: hcc params 0x014051c7 hci version 0x100 quirks 0x0000000100000090
> xhci_hcd 0000:03:00.0: enabling Mem-Wr-Inval
> usb usb1: New USB device found, idVendor=1d6b, idProduct=0002, bcdDevice= 4.20
> usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1
> usb usb1: Product: xHCI Host Controller
> usb usb1: Manufacturer: Linux 4.20.0-rc1 xhci-hcd
> usb usb1: SerialNumber: 0000:03:00.0
> hub 1-0:1.0: USB hub found
> hub 1-0:1.0: 2 ports detected
> xhci_hcd 0000:03:00.0: xHCI Host Controller
> xhci_hcd 0000:03:00.0: new USB bus registered, assigned bus number 2
> xhci_hcd 0000:03:00.0: Host supports USB 3.0  SuperSpeed
> usb usb2: We don't know the algorithms for LPM for this host, disabling LPM.
> usb usb2: New USB device found, idVendor=1d6b, idProduct=0003, bcdDevice= 4.20
> usb usb2: New USB device strings: Mfr=3, Product=2, SerialNumber=1
> usb usb2: Product: xHCI Host Controller
> usb usb2: Manufacturer: Linux 4.20.0-rc1 xhci-hcd
> usb usb2: SerialNumber: 0000:03:00.0
> hub 2-0:1.0: USB hub found
> hub 2-0:1.0: 2 ports detected
> usbcore: registered new interface driver usb-storage
> 
> and:
> 
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000001830d4000
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000001830d8000
> usb 2-1: new SuperSpeed Gen 1 USB device number 2 using xhci_hcd
> usb 2-1: New USB device found, idVendor=0781, idProduct=5583, bcdDevice= 1.00
> usb 2-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
> usb 2-1: Product: Ultra Fit
> usb 2-1: Manufacturer: SanDisk
> usb 2-1: SerialNumber: 4C531001340112110513
> usb-storage 2-1:1.0: USB Mass Storage device detected
> xhci_hcd 0000:03:00.0: dma_direct_alloc: coherent: 1
> xhci_hcd 0000:03:00.0: dma_direct_alloc: returned: a8000001830dc000
> scsi host5: usb-storage 2-1:1.0
> scsi 5:0:0:0: Direct-Access     SanDisk  Ultra Fit        1.00 PQ: 0 ANSI: 6
> sd 5:0:0:0: Attached scsi generic sg2 type 0
> sd 5:0:0:0: [sdc] 120127488 512-byte logical blocks: (61.5 GB/57.3 GiB)
> sd 5:0:0:0: [sdc] Write Protect is off
> sd 5:0:0:0: [sdc] Mode Sense: 43 00 00 00
> sd 5:0:0:0: [sdc] Write cache: disabled, read cache: enabled, doesn't support DPO or FUA
> 
> confirming that indeed the devices down the PCI-HT bridge don't need the 
> bus mask to be set.
> 
>  NB the reason for the bus address of the XHCI PCI device remaining the 
> same (0000:03:00.0) in the logs above regardless of the slot it has been 
> plugged in is that the option board actually includes a PCI-PCIe bridge 
> the actual XHCI device is behind, and that secondary PCIe link happens to 
> get the same bus number in enumeration.
> 
>  The BCM1480 SOC may or may not require a similar quirk, but I have no 
> documentation nor hardware to check with, so let's leave it for someone 
> who can verify it.  My understanding is it implements a PCI-X rather than 
> a plain PCI host bridge, so the limitation may well have been lifted in 
> the redesign.
> 
>   Maciej
> ---
>  arch/mips/pci/fixup-sb1250.c |   48 +++++++++++++++++++++++++++++++++++++++++++
>  1 file changed, 48 insertions(+)
> 
> linux-mips-sibyte-sb1250-bus-dma-mask.diff
> Index: linux-20181104-swarm64-eb/arch/mips/pci/fixup-sb1250.c
> ===================================================================
> --- linux-20181104-swarm64-eb.orig/arch/mips/pci/fixup-sb1250.c
> +++ linux-20181104-swarm64-eb/arch/mips/pci/fixup-sb1250.c
> @@ -1,6 +1,7 @@
>  /*
>   *	Copyright (C) 2004, 2006  MIPS Technologies, Inc.  All rights reserved.
>   *	    Author:	Maciej W. Rozycki <macro@mips.com>
> + *	Copyright (C) 2018  Maciej W. Rozycki
>   *
>   *	This program is free software; you can redistribute it and/or
>   *	modify it under the terms of the GNU General Public License
> @@ -8,6 +9,7 @@
>   *	2 of the License, or (at your option) any later version.
>   */
>  
> +#include <linux/dma-mapping.h>
>  #include <linux/pci.h>
>  
>  /*
> @@ -22,6 +24,52 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_SI
>  			quirk_sb1250_pci);
>  
>  /*
> + * The BCM1250, etc. PCI host bridge does not support DAC on its 32-bit
> + * bus, so we set the bus's DMA mask accordingly.  However the HT link
> + * down the artificial PCI-HT bridge supports 40-bit addressing and the
> + * SP1011 HT-PCI bridge downstream supports both DAC and a 64-bit bus
> + * width, so we record the PCI-HT bridge's secondary and subordinate bus
> + * numbers and do not set the mask for devices present in the inclusive
> + * range of those.
> + */
> +struct sb1250_bus_dma_mask_exclude {
> +	bool set;
> +	unsigned char start;
> +	unsigned char end;
> +};
> +
> +static int sb1250_bus_dma_mask(struct pci_dev *dev, void *data)
> +{
> +	struct sb1250_bus_dma_mask_exclude *exclude = data;
> +
> +	if (!exclude->set && (dev->vendor == PCI_VENDOR_ID_SIBYTE &&
> +			      dev->device == PCI_DEVICE_ID_BCM1250_HT)) {
> +		exclude->start = dev->subordinate->number;
> +		exclude->end = pci_bus_max_busnr(dev->subordinate);
> +		exclude->set = true;
> +		dev_dbg(&dev->dev, "not disabling DAC for [bus %02x-%02x]",
> +			exclude->start, exclude->end);
> +	} else if (!exclude->set ||
> +		   (exclude->set && (dev->bus->number < exclude->start ||
> +				     dev->bus->number > exclude->end))) {
> +		dev_dbg(&dev->dev, "disabling DAC for device");
> +		dev->dev.bus_dma_mask = DMA_BIT_MASK(32);
> +	} else {
> +		dev_dbg(&dev->dev, "not disabling DAC for device");
> +	}
> +	return 0;

Hmm, these conditions look very hard to read to me.  Wouldn't this
have the same effect?

	if (exclude->set)
		return;

	if (dev->vendor == PCI_VENDOR_ID_SIBYTE &&
	    dev->device == PCI_DEVICE_ID_BCM1250_HT) {
		exclude->start = dev->subordinate->number;
		exclude->end = pci_bus_max_busnr(dev->subordinate);
		exclude->set = true;
		dev_dbg(&dev->dev, "not disabling DAC for [bus %02x-%02x]",
			exclude->start, exclude->end);
		return;
	}

	if (dev->bus->number < exclude->start ||
	    dev->bus->number > exclude->end))) {
		dev_dbg(&dev->dev, "disabling DAC for device");
		dev->dev.bus_dma_mask = DMA_BIT_MASK(32);
		return;
	}

	dev_dbg(&dev->dev, "not disabling DAC for device");
	return 0;

Otherwise this looks fine to me:

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH 1/2] MIPS: SiByte: Set 32-bit bus mask for BCM1250 PCI

[Maciej W. Rozycki]

On Wed, 7 Nov 2018, Christoph Hellwig wrote:

> > +static int sb1250_bus_dma_mask(struct pci_dev *dev, void *data)
> > +{
> > +	struct sb1250_bus_dma_mask_exclude *exclude = data;
> > +
> > +	if (!exclude->set && (dev->vendor == PCI_VENDOR_ID_SIBYTE &&
> > +			      dev->device == PCI_DEVICE_ID_BCM1250_HT)) {
> > +		exclude->start = dev->subordinate->number;
> > +		exclude->end = pci_bus_max_busnr(dev->subordinate);
> > +		exclude->set = true;
> > +		dev_dbg(&dev->dev, "not disabling DAC for [bus %02x-%02x]",
> > +			exclude->start, exclude->end);
> > +	} else if (!exclude->set ||
> > +		   (exclude->set && (dev->bus->number < exclude->start ||
> > +				     dev->bus->number > exclude->end))) {
> > +		dev_dbg(&dev->dev, "disabling DAC for device");
> > +		dev->dev.bus_dma_mask = DMA_BIT_MASK(32);
> > +	} else {
> > +		dev_dbg(&dev->dev, "not disabling DAC for device");
> > +	}
> > +	return 0;
> 
> Hmm, these conditions look very hard to read to me.  Wouldn't this
> have the same effect?
> 
> 	if (exclude->set)
> 		return;

 Nope, `exclude->set' only means we already know what range to exclude 
(and that gets set mid-way through scanning as the HT bridge is 
encountered).  Then if it's unset, we know we are (still) outside that 
range.

 Maybe I can flatten the conditions at the small cost of executing some 
code unnecessarily.  But that won't be a big deal as this stuff is only 
executed once at boot and isn't performance critical.

 It'll have to wait until next week though as I'll be travelling 
throughout the rest of this and won't be able to test anything.

  Maciej