From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Alex Deucher <alexander.deucher@amd.com>,
Bjorn Helgaas <bhelgaas@google.com>
Subject: [PATCH 4.19 25/92] PCI: Mark AMD Navi14 GPU rev 0xc5 ATS as broken
Date: Tue, 28 Jan 2020 15:07:53 +0100 [thread overview]
Message-ID: <20200128135812.295361279@linuxfoundation.org> (raw)
In-Reply-To: <20200128135809.344954797@linuxfoundation.org>
From: Alex Deucher <alexander.deucher@amd.com>
commit 5e89cd303e3a4505752952259b9f1ba036632544 upstream.
To account for parts of the chip that are "harvested" (disabled) due to
silicon flaws, caches on some AMD GPUs must be initialized before ATS is
enabled.
ATS is normally enabled by the IOMMU driver before the GPU driver loads, so
this cache initialization would have to be done in a quirk, but that's too
complex to be practical.
For Navi14 (device ID 0x7340), this initialization is done by the VBIOS,
but apparently some boards went to production with an older VBIOS that
doesn't do it. Disable ATS for those boards.
Link: https://lore.kernel.org/r/20200114205523.1054271-3-alexander.deucher@amd.com
Bug: https://gitlab.freedesktop.org/drm/amd/issues/1015
See-also: d28ca864c493 ("PCI: Mark AMD Stoney Radeon R7 GPU ATS as broken")
See-also: 9b44b0b09dec ("PCI: Mark AMD Stoney GPU ATS as broken")
[bhelgaas: squash into one patch, simplify slightly, commit log]
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
drivers/pci/quirks.c | 19 +++++++++++++------
1 file changed, 13 insertions(+), 6 deletions(-)
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -4891,18 +4891,25 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_SE
#ifdef CONFIG_PCI_ATS
/*
- * Some devices have a broken ATS implementation causing IOMMU stalls.
- * Don't use ATS for those devices.
+ * Some devices require additional driver setup to enable ATS. Don't use
+ * ATS for those devices as ATS will be enabled before the driver has had a
+ * chance to load and configure the device.
*/
-static void quirk_no_ats(struct pci_dev *pdev)
+static void quirk_amd_harvest_no_ats(struct pci_dev *pdev)
{
- pci_info(pdev, "disabling ATS (broken on this device)\n");
+ if (pdev->device == 0x7340 && pdev->revision != 0xc5)
+ return;
+
+ pci_info(pdev, "disabling ATS\n");
pdev->ats_cap = 0;
}
/* AMD Stoney platform GPU */
-DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, 0x98e4, quirk_no_ats);
-DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, 0x6900, quirk_no_ats);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, 0x98e4, quirk_amd_harvest_no_ats);
+/* AMD Iceland dGPU */
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, 0x6900, quirk_amd_harvest_no_ats);
+/* AMD Navi14 dGPU */
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, 0x7340, quirk_amd_harvest_no_ats);
#endif /* CONFIG_PCI_ATS */
/* Freescale PCIe doesn't support MSI in RC mode */
next prev parent reply other threads:[~2020-01-28 14:31 UTC|newest]
Thread overview: 99+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-01-28 14:07 [PATCH 4.19 00/92] 4.19.100-stable review Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 01/92] can, slip: Protect tty->disc_data in write_wakeup and close with RCU Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 02/92] firestream: fix memory leaks Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 03/92] gtp: make sure only SOCK_DGRAM UDP sockets are accepted Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 04/92] ipv6: sr: remove SKB_GSO_IPXIP6 on End.D* actions Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 05/92] net: bcmgenet: Use netif_tx_napi_add() for TX NAPI Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 06/92] net: cxgb3_main: Add CAP_NET_ADMIN check to CHELSIO_GET_MEM Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 07/92] net: ip6_gre: fix moving ip6gre between namespaces Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 08/92] net, ip6_tunnel: fix namespaces move Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 09/92] net, ip_tunnel: " Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 10/92] net: rtnetlink: validate IFLA_MTU attribute in rtnl_create_link() Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 11/92] net_sched: fix datalen for ematch Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 12/92] net-sysfs: Fix reference count leak in rx|netdev_queue_add_kobject Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 13/92] net-sysfs: fix netdev_queue_add_kobject() breakage Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 14/92] net-sysfs: Call dev_hold always in netdev_queue_add_kobject Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 15/92] net-sysfs: Call dev_hold always in rx_queue_add_kobject Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 16/92] net-sysfs: Fix reference count leak Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 17/92] net: usb: lan78xx: Add .ndo_features_check Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 18/92] Revert "udp: do rmem bulk free even if the rx sk queue is empty" Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 19/92] tcp_bbr: improve arithmetic division in bbr_update_bw() Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 20/92] tcp: do not leave dangling pointers in tp->highest_sack Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 21/92] tun: add mutex_unlock() call and napi.skb clearing in tun_get_user() Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 22/92] afs: Fix characters allowed into cell names Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 23/92] hwmon: (adt7475) Make volt2reg return same reg as reg2volt input Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 24/92] hwmon: (core) Do not use device managed functions for memory allocations Greg Kroah-Hartman
2020-01-28 14:07 ` Greg Kroah-Hartman [this message]
2020-01-28 14:07 ` [PATCH 4.19 26/92] tracing: trigger: Replace unneeded RCU-list traversals Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 27/92] Input: keyspan-remote - fix control-message timeouts Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 28/92] Revert "Input: synaptics-rmi4 - dont increment rmiaddr for SMBus transfers" Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 29/92] ARM: 8950/1: ftrace/recordmcount: filter relocation types Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 30/92] mmc: tegra: fix SDR50 tuning override Greg Kroah-Hartman
2020-01-28 14:07 ` [PATCH 4.19 31/92] mmc: sdhci: fix minimum clock rate for v3 controller Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 32/92] Documentation: Document arm64 kpti control Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 33/92] Input: pm8xxx-vib - fix handling of separate enable register Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 34/92] Input: sur40 - fix interface sanity checks Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 35/92] Input: gtco - fix endpoint sanity check Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 36/92] Input: aiptek " Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 37/92] Input: pegasus_notetaker " Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 38/92] Input: sun4i-ts - add a check for devm_thermal_zone_of_sensor_register Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 39/92] netfilter: nft_osf: add missing check for DREG attribute Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 40/92] hwmon: (nct7802) Fix voltage limits to wrong registers Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 41/92] scsi: RDMA/isert: Fix a recently introduced regression related to logout Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 42/92] tracing: xen: Ordered comparison of function pointers Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 43/92] do_last(): fetch directory ->i_mode and ->i_uid before its too late Greg Kroah-Hartman
2020-01-31 10:08 ` Rantala, Tommi T. (Nokia - FI/Espoo)
2020-01-31 12:20 ` Al Viro
2020-01-31 13:57 ` Rantala, Tommi T. (Nokia - FI/Espoo)
2020-01-28 14:08 ` [PATCH 4.19 44/92] net/sonic: Add mutual exclusion for accessing shared state Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 45/92] net/sonic: Clear interrupt flags immediately Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 46/92] net/sonic: Use MMIO accessors Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 47/92] net/sonic: Fix interface error stats collection Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 48/92] net/sonic: Fix receive buffer handling Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 49/92] net/sonic: Avoid needless receive descriptor EOL flag updates Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 50/92] net/sonic: Improve receive descriptor status flag check Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 51/92] net/sonic: Fix receive buffer replenishment Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 52/92] net/sonic: Quiesce SONIC before re-initializing descriptor memory Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 53/92] net/sonic: Fix command register usage Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 54/92] net/sonic: Fix CAM initialization Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 55/92] net/sonic: Prevent tx watchdog timeout Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 56/92] tracing: Use hist triggers var_ref array to destroy var_refs Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 57/92] tracing: Remove open-coding of hist trigger var_ref management Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 58/92] tracing: Fix histogram code when expression has same var as value Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 59/92] sd: Fix REQ_OP_ZONE_REPORT completion handling Greg Kroah-Hartman
2020-01-28 18:02 ` Pavel Machek
2020-01-28 18:15 ` Greg Kroah-Hartman
2020-01-29 1:05 ` Damien Le Moal
2020-01-28 14:08 ` [PATCH 4.19 60/92] crypto: geode-aes - switch to skcipher for cbc(aes) fallback Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 61/92] coresight: etb10: Do not call smp_processor_id from preemptible Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 62/92] coresight: tmc-etf: " Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 63/92] libertas: Fix two buffer overflows at parsing bss descriptor Greg Kroah-Hartman
2020-01-29 22:47 ` Pavel Machek
2020-01-28 14:08 ` [PATCH 4.19 64/92] media: v4l2-ioctl.c: zero reserved fields for S/TRY_FMT Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 65/92] scsi: iscsi: Avoid potential deadlock in iscsi_if_rx func Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 66/92] netfilter: ipset: use bitmap infrastructure completely Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 67/92] netfilter: nf_tables: add __nft_chain_type_get() Greg Kroah-Hartman
2020-01-28 18:13 ` Pavel Machek
2020-01-28 14:08 ` [PATCH 4.19 68/92] net/x25: fix nonblocking connect Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 69/92] mm/memory_hotplug: make remove_memory() take the device_hotplug_lock Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 70/92] mm, sparse: drop pgdat_resize_lock in sparse_add/remove_one_section() Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 71/92] mm, sparse: pass nid instead of pgdat to sparse_add_one_section() Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 72/92] drivers/base/memory.c: remove an unnecessary check on NR_MEM_SECTIONS Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 73/92] mm, memory_hotplug: add nid parameter to arch_remove_memory Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 74/92] mm/memory_hotplug: release memory resource after arch_remove_memory() Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 75/92] drivers/base/memory.c: clean up relics in function parameters Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 76/92] mm, memory_hotplug: update a comment in unregister_memory() Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 77/92] mm/memory_hotplug: make unregister_memory_section() never fail Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 78/92] mm/memory_hotplug: make __remove_section() " Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 79/92] powerpc/mm: Fix section mismatch warning Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 80/92] mm/memory_hotplug: make __remove_pages() and arch_remove_memory() never fail Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 88/92] mm/hotplug: kill is_dev_zone() usage in __remove_pages() Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 89/92] drivers/base/node.c: simplify unregister_memory_block_under_nodes() Greg Kroah-Hartman
2020-01-28 14:08 ` [PATCH 4.19 91/92] mm/memory_hotplug: fix try_offline_node() Greg Kroah-Hartman
2020-01-28 14:09 ` [PATCH 4.19 92/92] mm/memory_hotplug: shrink zones when offlining memory Greg Kroah-Hartman
2020-01-28 23:03 ` [PATCH 4.19 00/92] 4.19.100-stable review shuah
2020-01-29 4:54 ` Naresh Kamboju
2020-01-29 11:31 ` Pavel Machek
2020-01-29 12:57 ` Greg Kroah-Hartman
2020-01-29 13:16 ` Jon Hunter
2020-01-29 14:43 ` Guenter Roeck
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200128135812.295361279@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=alexander.deucher@amd.com \
--cc=bhelgaas@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=stable@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).