Commit Graph

926 Commits

Author SHA1 Message Date
Andrew Turner
41b84341f5 Use the correct address when creating pci resources
When the PCI and CPU physical addresses are identical it doesn't matter
which is used to create the resources, however on some systems, e.g.
qemu armv7 virt, they are different. This leads to a panic as we try to
map the wrong physical address into the kernel address space.

Reported by:	Jenkins via trasz
Sponsored by:	Innovate UK
2020-06-19 18:00:20 +00:00
Michal Meloun
7f8437c353 Adapt ARMADA8k PCIe driver to newly imported 5.7 DT.
- temporarily disable handling with phy, we don't have driver for it yet
- always clear cause for administartive interrupt.
While I'm in, fix style(9) (mainly whitespace).

MFC after:	4 weeks
2020-06-19 17:33:54 +00:00
Michal Meloun
224c5a9ff3 Revert r362389, it was committed with <patch>.diff instead of <patch>.txt as
commit log.
2020-06-19 17:32:50 +00:00
Michal Meloun
7a5750fd2d diff --git a/sys/dev/pci/pci_dw_mv.c b/sys/dev/pci/pci_dw_mv.c
index 06a29fefbdd..571fc00f6c1 100644
--- a/sys/dev/pci/pci_dw_mv.c
+++ b/sys/dev/pci/pci_dw_mv.c
@@ -64,15 +64,11 @@ __FBSDID("$FreeBSD$");

 #define MV_GLOBAL_CONTROL_REG		0x8000
 #define PCIE_APP_LTSSM_EN		(1 << 2)
-//#define PCIE_DEVICE_TYPE_SHIFT		4
-//#define PCIE_DEVICE_TYPE_MASK		0xF
-//#define PCIE_DEVICE_TYPE_RC		0x4/

 #define MV_GLOBAL_STATUS_REG		0x8008
 #define	 MV_STATUS_RDLH_LINK_UP			(1 << 1)
 #define  MV_STATUS_PHY_LINK_UP			(1 << 9)

-
 #define MV_INT_CAUSE1			0x801C
 #define MV_INT_MASK1			0x8020
 #define  INT_A_ASSERT_MASK			(1 <<  9)
@@ -90,11 +86,7 @@ __FBSDID("$FreeBSD$");
 #define MV_ARUSER_REG			0x805C
 #define MV_AWUSER_REG			0x8060

-
-
 #define	MV_MAX_LANES	8
-
-
 struct pci_mv_softc {
 	struct pci_dw_softc	dw_sc;
 	device_t		dev;
@@ -112,7 +104,6 @@ static struct ofw_compat_data compat_data[] = {
 	{NULL,		 	  0},
 };

-
 static int
 pci_mv_phy_init(struct pci_mv_softc *sc)
 {
@@ -121,18 +112,23 @@ pci_mv_phy_init(struct pci_mv_softc *sc)
 	for (i = 0; i < MV_MAX_LANES; i++) {
 		rv =  phy_get_by_ofw_idx(sc->dev, sc->node, i, &(sc->phy[i]));
 		if (rv != 0 && rv != ENOENT) {
-	  		device_printf(sc->dev, "Cannot get phy[%d]\n", i);
-	  		goto fail;
-	  	}
-	  	if (sc->phy[i] == NULL)
-	  		continue;
-	  	rv = phy_enable(sc->phy[i]);
-	  	if (rv != 0) {
-	  		device_printf(sc->dev, "Cannot enable phy[%d]\n", i);
-	  		goto fail;
-	  	}
-	  }
-	  return (0);
+			device_printf(sc->dev, "Cannot get phy[%d]\n", i);
+/* XXX revert when phy driver will be implemented */
+#if 0
+		goto fail;
+#else
+		continue;
+#endif
+		}
+		if (sc->phy[i] == NULL)
+			continue;
+		rv = phy_enable(sc->phy[i]);
+		if (rv != 0) {
+			device_printf(sc->dev, "Cannot enable phy[%d]\n", i);
+			goto fail;
+		}
+	}
+	return (0);

 fail:
 	for (i = 0; i < MV_MAX_LANES; i++) {
@@ -173,13 +169,14 @@ pci_mv_init(struct pci_mv_softc *sc)
 	/* Enable local interrupts */
 	pci_dw_dbi_wr4(sc->dev, DW_MSI_INTR0_MASK, 0xFFFFFFFF);
 	pci_dw_dbi_wr4(sc->dev, MV_INT_MASK1, 0xFFFFFFFF);
-	pci_dw_dbi_wr4(sc->dev, MV_INT_MASK2, 0xFFFFFFFF);
+	pci_dw_dbi_wr4(sc->dev, MV_INT_MASK2, 0xFFFFFFFD);
 	pci_dw_dbi_wr4(sc->dev, MV_INT_CAUSE1, 0xFFFFFFFF);
 	pci_dw_dbi_wr4(sc->dev, MV_INT_CAUSE2, 0xFFFFFFFF);

 	/* Errors have own interrupt, not yet populated in DTt */
 	pci_dw_dbi_wr4(sc->dev, MV_ERR_INT_MASK, 0);
 }
+
 static int pci_mv_intr(void *arg)
 {
 	struct pci_mv_softc *sc = arg;
@@ -188,8 +185,6 @@ static int pci_mv_intr(void *arg)
 	/* Ack all interrups */
 	cause1 = pci_dw_dbi_rd4(sc->dev, MV_INT_CAUSE1);
 	cause2 = pci_dw_dbi_rd4(sc->dev, MV_INT_CAUSE2);
-	if (cause1 == 0 || cause2 == 0)
-		return(FILTER_STRAY);

 	pci_dw_dbi_wr4(sc->dev, MV_INT_CAUSE1, cause1);
 	pci_dw_dbi_wr4(sc->dev, MV_INT_CAUSE2, cause2);
2020-06-19 17:25:54 +00:00
Michal Meloun
1f446a117e Improve DesignWare PCIe driver:
- only normal memory window is mandatory, prefetchable memory and
  I/O windows should be optional
- full PCIe configuration space is supported
- remove duplicated check from function for accessing configuration space.
  It is already contained in pci_dw_check_dev()

MFC after:	2 weeks
2020-06-19 16:15:06 +00:00
Andrew Turner
c794cdc0a2 Stop assuming we can print rman_res_t with %lx
This is not the case on armv6 and armv7, where we also build this driver.
Fix by casting through uintmax_t and using %jx.

Sponsored by:	Innovate UK
2020-06-18 06:21:00 +00:00
Andrew Turner
9a7053ce96 Clean up the pci host generic driver
- Support Prefetchable Memory.
 - Use the correct rman when allocating memory and ioports.
 - Translate PCI addresses in bus_alloc_resource to allow physical
   addresses that are different than pci addresses.

Reviewed by:	Robert Crowston <crowston_protonmail.com>
Sponsored by:	Innovate UK
Differential Revision:	https://reviews.freebsd.org/D25121
2020-06-17 19:56:17 +00:00
Chuck Tuffli
f14f005113 pci: loosen PCIe hot-plug requirements
The original PCIe hot-plug code required a couple of things which cause
PCI probing errors on the QEMU Q35 system and possibly physical systems
(Dell R6515).

Allocate the hot-plug interrupt as shared to support INTx interrupts.
The hot-plug interrupt mechanism should normally be MSI as PCIe mandates
MSI support, but QEMU's Q35 bridge only provides INTx interrupts.

Second, the code required the Electromechanical Interlock (Slot Status
EIS) to be engaged if present (Slot Capability EIP). Some platforms
including QEMU Q35 set EIP but not EIS. Fix by deleting the check.

Reviewed by: imp, mav, jhb
MFC after:	2 weeks
Differential Revision: https://reviews.freebsd.org/D24877
2020-06-10 20:12:45 +00:00
Andriy Gapon
cffd37da23 do not enable pci bridge decoding on resume until I/O windows are restored
PCI bus driver restores most but not all of a child PCI-PCI bridge
configuration.  The bridge's I/O windows are restored by pcib driver and
that happens later in time.  This can be problematic because the Command
register is restored before the windows are restored.  If the firmware
programs the windows incorrectly or even does not program them at all,
then the bridge can start claiming I/O cycles that are not intended for
it.  This will continue until the correct windows are restored.

I have observed this problem with a buggy BIOS where after resuming from
S3 an I/O port window of a PCI-PCI bridge was configured with zero base
and limit causing the bridge to claim 0x0 - 0xFFF port range.  That
interfered with ACPI port access including ACPI PM Timer at port 0x808,
thus wreaking havoc in the time keeping.

The solution is to restore the Command register of PCI-PCI bridges after
the windows are restored in pcib driver.  While here, I decided that for
other PCI device types (normal and cardbus) it's better to restore the
Command register after their BARs are restored.

To do: per jhb's suggestion, move the window handling to pci driver.

Reviewed by:	imp, jhb, kib
MFC after:	2 weeks
Differential Revision: https://reviews.freebsd.org/D25028
2020-05-29 07:50:55 +00:00
Ruslan Bukin
43843cc281 Rename dmar_get_dma_tag() to acpi_iommu_get_dma_tag().
This is needed for a new IOMMU controller support.

Reviewed by:	kib
Differential Revision:	https://reviews.freebsd.org/D24943
2020-05-26 16:40:40 +00:00
Andrew Turner
ed97f6b30f Remove PCI_IO_WINDOW_OFFSET from the pci host generic fdt attachment.
It doesn't seem to be needed, and breaks booting under bhyve/arm64.

Discussed with:	br
MFC after:	2 weeks
Sponsored by:	Innovate UK
2020-04-24 11:03:15 +00:00
Navdeep Parhar
a8f48cf82f Remove spurious warning about invalid VPD data.
The warning used to be displayed for valid VPDs about 512B or above in
size.  Fix the size check and add a break while here so that the routine
stops if if detects any problem.

Tested with "pciconf -lV"

Reviewed by:	kib@, jhb@
MFC after:	1 week
Sponsored by:	Chelsio Communications
Differential Revision:	https://reviews.freebsd.org/D23679
2020-03-18 01:09:40 +00:00
Konstantin Belousov
b08bdabee4 Add more values for PCI capabilities, PCIe extended capabilities, and subclasses.
Taken from
https://pcisig.com/sites/default/files/files/PCI_Code-ID_r_1_11__v24_Jan_2019.pdf

Submitted by:	Dmitry Luhtionov <dmitryluhtionov@gmail.com>
MFC after:	1 week
2020-02-20 17:08:52 +00:00
Konstantin Belousov
eca86ffaa1 Fix typo.
Sponsored by:	The FreeBSD Foundation
MFC after:	3 days
2020-02-17 13:26:36 +00:00
Pawel Biernacki
59e4be225c Mark more nodes as CTLFLAG_MPSAFE or CTLFLAG_NEEDGIANT (5 of many)
r357614 added CTLFLAG_NEEDGIANT to make it easier to find nodes that are
still not MPSAFE (or already are but aren’t properly marked). Use it in
preparation for a general review of all nodes.
This is non-functional change that adds annotations to SYSCTL_NODE and
SYSCTL_PROC nodes using one of the soon-to-be-required flags.

Reviewed by:	imp, kib
Approved by:	kib (mentor)
Differential Revision:	https://reviews.freebsd.org/D23633
2020-02-16 17:11:54 +00:00
Ruslan Bukin
667c3fc0f6 Add PCI Express driver for the ARM Neoverse N1 System Development
Platform (N1SDP).

Neoverse N1 is a high-performance ARM microarchitecture designed
by the ARM Holdings for the server market.

The PCI part on N1SDP was shipped untested and suffers from some
integration issues.

For instance accessing to not existing BDFs causes System Error
(SError) exception. To mitigate this, the firmware scans the bus,
catches SErrors and creates a table with valid BDFs. That allows
us to filter-out accesses to invalid BDFs in this driver.

Also the root complex config space (BDF == 0) has an unusual
location in memory map, so remapping accesses to it is required.

Finally, the config space is restricted to 32-bit accesses only.

This was tested on the ARM boxes kindly provided by the ARM Ltd
to the DARPA CHERI Project.

In collaboration with:	andrew
Reviewed by:	andrew
Sponsored by:	DARPA, AFRL
Differential Revision:	https://reviews.freebsd.org/D23349
2020-02-11 15:12:09 +00:00
Ruslan Bukin
c344a95134 o Move the software context struct to a header file.
o Make the pci_host_generic_acpi_attach() globally visible.
o Declare a new driver class.

These will be used by a new PCI root complex driver.

Sponsored by:	DARPA, AFRL
2020-01-24 16:43:49 +00:00
Ruslan Bukin
9a82a56bee Move the ECAM macroses to the header file.
These will be used by other PCI root complex drivers.

Sponsored by:	DARPA, AFRL
2020-01-24 16:08:06 +00:00
Scott Long
13d700adec Abstract the locking for PCIe hotplug. It still uses Giant so there's
no functional change yet.
2019-12-26 21:00:06 +00:00
Mark Peek
484651cd1b Remove VMware MSI-X from the PCI blacklist.
First reported against ESXi 5.0, PCI passthrough was not working due to
MSI-X issues. However, this issue was fixed via patch releases against
ESXi 5.5 and 6.0 in 2016. Given ESXi 5.5 and earlier have been EOL, this
patch removes the VMware MSI-X blacklist entries in the quirk table.

PR:	203874
Reviewed by:	imp, jhb
MFC after:	1 month
Sponsored by:	VMware
Differential Revision: https://reviews.freebsd.org/D22819
2019-12-18 23:00:56 +00:00
Scott Long
97faa4c470 Add accessors for the Vendor Specific Extended Capability (VSEC)
Parse out the VSEC.  If the user invokes a second -c command line option,
do a hex dump of the vendor data.

Reviewed by:	imp
MFC after:	3 days
Sponsored by:	Intel
Differential Revision:	http://reviews.freebsd.org/D22808
2019-12-13 23:46:59 +00:00
Stefan Eßer
fce5d19d64 Remove "All rights reserved" phrase from copyright notes.
With the ratification of the Berne Convention in 2000, it became obsolete.

I have removed that phrase and the "(c)" only from files without copyright
claims by other parties. There are 2 files (pci.c, pci_private.h) that are
also claimed by Michael Smith <msmith@freebsd.org> and by BSDi, which have
therefore not been included in this commit.

When all member nations of the Buenos Aires Convention adopted the Berne
Convention, the phrase "All rights reserved" became unnecessary to assert
copyright. Remove it from files under my copyright.

There are 2 files (pci.c, pci_private.h) that) that do also bear msmith's
and BSDi's copyright. I have left them unchanged for now, since I do not
know whether they (or the legal successor in case of BSDi) would agree.
2019-12-04 08:03:33 +00:00
Warner Losh
96b506a57c Hoist locking giant back up into the ioctl handler
Move the locking back into the ioctl handler. This "fixes" the race where we hve
a hot plug event just after the dropping of Giant in pci_find_dbsf, assuming the
driver doesn't then call anything that drops and picks up Giant again... It's a
little safer since don't think it doesn't, but we lack the tools to know for
sure.
2019-11-24 15:37:14 +00:00
Warner Losh
dd615d09c4 Push Giant down one layer
The /dev/pci device doesn't need GIANT, per se. However, one routine
that it calls, pci_find_dbsf implicitly does. It walks a list that can
change when PCI scans a new bus. With hotplug, this means we could
have a race with that scanning. To prevent that, take out Giant around
scanning the list.

However, given that we have places in the tree that drop giant, if
held when we call into them, the whole use of Giant to protect newbus
may be less effective that we desire, so add a comment about why we're
talking it out, and we'll address the issue when we lock newbus with
something other than Giant.
2019-11-23 23:43:52 +00:00
Michal Meloun
26abae3f17 Add driver for DesignWare PCIE core, and its Armada 8K specific attachement.
MFC after:	3 weeks
2019-10-20 11:11:32 +00:00
John Baldwin
21d3196209 Export pci_attach() and pci_detach().
Reviewed by:	imp
MFC after:	1 week
Differential Revision:	https://reviews.freebsd.org/D21948
2019-10-15 18:58:01 +00:00
Ed Maste
aa91d4b3a9 pcie: return an error if a matching resource is not found
Submitted by:	markj
Reviewed by:	manu
Event:		vBSDCon FreeBSD hackathon
Sponsored by:	The FreeBSD Foundation
Differential Revision:	https://reviews.freebsd.org/D20884
2019-09-05 15:45:21 +00:00
Alexander Motin
21e51c82fb Don't consider PCIe hot-plug command timeout fatal.
According to my tests and errata to several generations of Intel CPUs,
PCIe hot-plug command completion reporting is not very reliable thing.
At least on my Supermicro X11DPi-NT board I never saw it reported.
Before this change timeout code detached devices and tried to disable
the slot, that in my case resulted in hot-plugged device being detached
just a second after it was successfully detected and attached.  This
change removes that, so in case of timeout it just prints the error and
continue operation.  Linux does the same.

MFC after:	1 week
Sponsored by:	iXsystems, Inc.
2019-08-26 17:27:36 +00:00
Justin Hibbits
a83733119d Fix the build after r350570.
Signed char cannot compare to values above 0x7f.  Use unsigned instead.
2019-08-04 19:47:02 +00:00
Justin Hibbits
937a05ba81 Add necessary bits for Linux KPI to work correctly on powerpc
PowerPC, and possibly other architectures, use different address ranges for
PCI space vs physical address space, which is only mapped at resource
activation time, when the BAR gets written.  The DRM kernel modules do not
activate the rman resources, soas not to waste KVA, instead only mapping
parts of the PCI memory at a time.  This introduces a
BUS_TRANSLATE_RESOURCE() method, implemented in the Open Firmware/FDT PCI
driver, to perform this necessary translation without activating the
resource.

In addition to system KPI changes, LinuxKPI is updated to handle a
big-endian host, by adding proper endian swaps to the I/O functions.

Submitted by:	mmacy
Reported by:	hselasky
Differential Revision:	https://reviews.freebsd.org/D21096
2019-08-04 19:28:10 +00:00
Warner Losh
e64f3dee49 Work around devices which return all zeros for reads of existing MSI-X
table VCTRL registers.

Unconditionally program the MSI-X vector control Mask field for MSI-X
table entries without regarud for Mask's previous value. Some devices
return all zeros on reads of the VCTRL registers, which would cause us
to skip disabling interrupts. This fixes the Samsung SM961/PM961 SSDs
which are return zero starting from offset 0x3084 within the memory
region specified by BAR0, even when they are active MSI-X vectors.

The Illumos kernel writes these unconditionally to 0 or 1. However,
section 6.8.2.9 of the PCI Local Bus 3.0 spec (dated Feb 3, 2004)
states for bits 31::01:
	After reset, the state of these bits must be 0. However, for
	potential future use, software must preserve the value of
	these reserved bits when modifying the value of other Vector
	Control bits. If software modifies the value of these reserved
	bits, the result is undefined."
so we always set or clear the Mask bit, but otherwise preserves the
old value.

PR: https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=211713
Reviewed By: imp, jhb
Submitted by: Ka Ho Ng
MFC After: 1 week
Differential Revision: https://reviews.freebsd.org/D20873
2019-07-08 19:38:49 +00:00
Hans Petter Selasky
f48c41accd Need to apply the PCIM_BAR_MEM_BASE mask to the physical memory
address before returning it to the user. Some of the least significant
bits have special meaning and should be masked away.

Discussed with:	kib@
MFC after:	3 days
Sponsored by:	Mellanox Technologies
2019-06-28 22:28:51 +00:00
Colin Percival
e0235fd34a Only respond to the PCIe Attention Button if a device is already plugged in.
Prior to this commit, if PCIEM_SLOT_STA_ABP and PCIEM_SLOT_STA_PDC are
asserted simultaneously, FreeBSD sets a 5 second "hardware going away" timer
and then processes the "presence detect" change. In the (physically
challenging) case that someone presses the "attention button" and inserts
a new PCIe device at exactly the same moment, this results in FreeBSD
recognizing that the device is present, attaching it, and then detaching it
5 seconds later.

On EC2 "bare metal" hardware this is the precise sequence of events which
takes place when a new EBS volume is attached; virtual machines have no
difficulty effecting physically implausible simultaneity.

This patch changes the handling of PCIEM_SLOT_STA_ABP to only detach a
device if the presence of a device was detected *before* the interrupt
which reports the Attention Button push.

Reported by:	Matt Wilson
Reviewed by:	jhb
MFC after:	1 week
Sponsored by:	https://www.patreon.com/cperciva
Differential Revision:	https://reviews.freebsd.org/D20499
2019-06-05 04:58:42 +00:00
Ruslan Bukin
7b4ec8d2fc Pass pci_base address instead of physical address to rman_manage_region().
This should had been part of r347930 ("pci: ecam: Correctly parse memory
and IO region").

Sponsored by:	DARPA, AFRL
2019-05-29 15:53:33 +00:00
Conrad Meyer
e2e050c8ef Extract eventfilter declarations to sys/_eventfilter.h
This allows replacing "sys/eventfilter.h" includes with "sys/_eventfilter.h"
in other header files (e.g., sys/{bus,conf,cpu}.h) and reduces header
pollution substantially.

EVENTHANDLER_DECLARE and EVENTHANDLER_LIST_DECLAREs were moved out of .c
files into appropriate headers (e.g., sys/proc.h, powernv/opal.h).

As a side effect of reduced header pollution, many .c files and headers no
longer contain needed definitions.  The remainder of the patch addresses
adding appropriate includes to fix those files.

LOCK_DEBUG and LOCK_FILE_LINE_ARG are moved to sys/_lock.h, as required by
sys/mutex.h since r326106 (but silently protected by header pollution prior
to this change).

No functional change (intended).  Of course, any out of tree modules that
relied on header pollution for sys/eventhandler.h, sys/lock.h, or
sys/mutex.h inclusion need to be fixed.  __FreeBSD_version has been bumped.
2019-05-20 00:38:23 +00:00
Emmanuel Vadot
ec55b6c5f5 pci: ecam: Correctly parse memory and IO region
When activating a resource do not compare the resource id to the adress.
Treat IO region as MEMORY region too.

Submitted by:	Tuan Phan <tphan@amperecomputing.com> (Original Version)
Sponsored by:	Ampere Computing, LLC
Differential Revision:	https://reviews.freebsd.org/D20214
2019-05-17 17:05:16 +00:00
Emmanuel Vadot
aa95512f35 pci: ecam: Do not warn on mismatch of bus_end
We cannot know the bus end number before parsing the MCFG table
so don't set the bus_end before that. If the MCFG table doesn't
exist we will set the configuration base address based on the _CBA
value and set the bus_end to the maximal number allowed by PCI.

Sponsored by: Ampere Computing, LLC

Differential Revision:	https://reviews.freebsd.org/D20213
2019-05-17 17:04:01 +00:00
Tycho Nightingale
96ca24dc32 remove the 4GB boundary requirement on PCI DMA segments
Reviewed by:	kib
Discussed with:	jhb
Sponsored by:	Dell EMC Isilon
Differential Revision:	https://reviews.freebsd.org/D19867
2019-04-19 13:43:33 +00:00
Konstantin Belousov
ca19084f36 Give new home to the comment from ppt_pci_reset(), explaining a nuance
of power reset.

Noted by:	soralx@cydem.org
Sponsored by:	Mellanox Technologies
MFC after:	12 days
2019-04-07 08:58:09 +00:00
Konstantin Belousov
5db2a4a812 Implement resets for PCI buses and PCIe bridges.
For PCI device (i.e. child of a PCI bus), reset tries FLR if
implemented and worked, and falls to power reset otherwise.

For PCIe bus (child of a PCIe bridge or root port), reset
disables PCIe link and then re-trains it, performing what is known as
link-level reset.

Reviewed by:	imp (previous version), jhb (previous version)
Sponsored by:	Mellanox Technologies
MFC after:	2 weeks
Differential revision:	https://reviews.freebsd.org/D19646
2019-04-05 19:25:26 +00:00
Tycho Nightingale
b2bbb74489 Devices behind downstream bridges should still get DMAR protection.
Reviewed by:	kib
Sponsored by:	Dell EMC Isilon
Differential Revision:	https://reviews.freebsd.org/D19717
2019-04-01 19:08:05 +00:00
Warner Losh
312437815a Fix two typos: an -> and; the the -> the
And justify the paragraph after the change (and set fill column to 80
instead of 70).

Noticed by: rpokala@, vangyzen@
2019-03-19 21:46:21 +00:00
Warner Losh
35613d10dc Add comment about why we bother to use endian macros here, and why we
must use bitfields.
2019-03-19 15:03:20 +00:00
John Baldwin
2e43efd0bb Drop "All rights reserved" from my copyright statements.
Reviewed by:	rgrimes
MFC after:	1 month
Differential Revision:	https://reviews.freebsd.org/D19485
2019-03-06 22:11:45 +00:00
John Baldwin
967b2dce02 Enable PCI BAR reallocation by default.
When pci_realloc_bars was first added, the intention was to eventually
enable it by default, but it was left disabled to preserve existing
behavior.  The setting is pretty conservative in that it does not
attempt to allocate resources for BARs that the BIOS/firmware leaves
disabled.  It only attempts to reallocate resources for a BAR that the
firmware programmed during boot but that conflicts with another
resource during the kernel's device scan.

PR 221350 is an example of a machine that this knob fixes.

Reviewed by:	imp
Differential Revision:	https://reviews.freebsd.org/D18965
2019-02-11 20:47:09 +00:00
Jayachandran C.
13607f6db5 pci_host_generic_acpi: use IORT data for MSI/MSI-X
Use the information from IORT parsing to translate the PCI RID to
GIC ITS device ID. And similarly, use the information to find the
PIC XREF identifier to be used for PCI devices.

Reviewed by:	andrew
Differential Revision:	https://reviews.freebsd.org/D18004
2019-02-07 04:50:16 +00:00
Andriy Gapon
82a5a27527 add support for marking interrupt handlers as suspended
The goal of this change is to fix a problem with PCI shared interrupts
during suspend and resume.

I have observed a couple of variations of the following scenario.
Devices A and B are on the same PCI bus and share the same interrupt.
Device A's driver is suspended first and the device is powered down.
Device B generates an interrupt. Interrupt handlers of both drivers are
called. Device A's interrupt handler accesses registers of the powered
down device and gets back bogus values (I assume all 0xff). That data is
interpreted as interrupt status bits, etc. So, the interrupt handler
gets confused and may produce some noise or enter an infinite loop, etc.

This change affects only PCI devices.  The pci(4) bus driver marks a
child's interrupt handler as suspended after the child's suspend method
is called and before the device is powered down.  This is done only for
traditional PCI interrupts, because only they can be shared.

At the moment the change is only for x86.

Notable changes in core subsystems / interfaces:
- BUS_SUSPEND_INTR and BUS_RESUME_INTR methods are added to bus
  interface along with convenience functions bus_suspend_intr and
  bus_resume_intr;
- rman_set_irq_cookie and rman_get_irq_cookie functions are added to
  provide a way to associate an interrupt resource with an interrupt
  cookie;
- intr_event_suspend_handler and intr_event_resume_handler functions
  are added to the MI interrupt handler interface.

I added two new interrupt handler flags, IH_SUSP and IH_CHANGED, to
implement the new intr_event functions.  IH_SUSP marks a suspended
interrupt handler.  IH_CHANGED is used to implement a barrier that
ensures that a change to the interrupt handler's state is visible
to future interrupts.
While there, I fixed some whitespace issues in comments and changed a
couple of logically boolean variables to be bool.

MFC after:	1 month (maybe)
Differential Revision: https://reviews.freebsd.org/D15755
2018-12-17 17:11:00 +00:00
Greg Lehey
867bb99bbb Work around BIOS quirks on HPE Proliant MicroServer Gen10
PR:		221350
Submitted by:	Bob Bishop
Reported by:	Rafal Lukawiecki
Reviewed by:	jhb
MFC after:	2 weeks
2018-12-17 07:09:46 +00:00
Justin Hibbits
bfed756af6 Sprinkle EARLY_DRIVER_MODULE around the tree
Mark some buses as BUS_PASS_BUS, and some resources as BUS_PASS_RESOURCE.
This also decouples some resource attachment orderings from being races by
device tree ordering, instead relying on the bus pass to provide the
ordering.

This was originally intended to support multipass suspend/resume, but it's
also needed on PowerMacs when using fdt, as the device tree seems to get
created in reverse of the OFW tree.
Reviewed by:	nwhitehorn (long ago)
Differential Revision:	https://reviews.freebsd.org/D918
2018-12-04 04:55:49 +00:00
Jayachandran C.
bd158cddc4 pci_host_generic : move activate/release to generic code
Now that the ACPI and FDT implementations for activating and
deactivating resources are the same, we can move it to
pci_host_generic.c.  No functional changes.

Reviewed by:	andrew
Differential Revision:	https://reviews.freebsd.org/D17793
2018-11-19 03:43:10 +00:00