From 3db7a4eeb231955ec9b7c9f42c0da566629301aa Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Thu, 23 Apr 2020 09:27:53 +0100
Subject: [PATCH] pci/msi: Stop warning for MSI enabling failure
If the MSI is already enabled, trying to enable it again results in an
-EINVAL and on the first attempt a WARN. That WARN causes our CI to
abort the run [on each first attempt to suspend]:
<4> [463.142025] WARNING: CPU: 0 PID: 2225 at drivers/pci/msi.c:1074 __pci_enable_msi_range+0x3cb/0x420
<4> [463.142026] Modules linked in: snd_hda_intel i915 snd_hda_codec_hdmi snd_hda_codec_realtek snd_hda_codec_generic mei_hdcp x86_pkg_temp_thermal coretemp crct10dif_pclmul crc32_pclmul snd_intel_dspcfg ghash_clmulni_intel snd_hda_codec btusb btrtl btbcm btintel e1000e bluetooth snd_hwdep snd_hda_core ptp ecdh_generic snd_pcm ecc pps_core mei_me mei prime_numbers [last unloaded: i915]
<4> [463.142045] CPU: 0 PID: 2225 Comm: kworker/u8:14 Tainted: G U 5.7.0-rc2-CI-CI_DRM_8350+ #1
<4> [463.142046] Hardware name: Intel Corporation NUC7i5BNH/NUC7i5BNB, BIOS BNKBL357.86A.0060.2017.1214.2013 12/14/2017
<4> [463.142049] Workqueue: events_unbound async_run_entry_fn
<4> [463.142051] RIP: 0010:__pci_enable_msi_range+0x3cb/0x420
<4> [463.142053] Code: 76 58 49 8d 56 48 48 89 df e8 31 73 fd ff e9 20 fe ff ff 31 f6 48 89 df e8 c2 e9 fd ff e9 d6 fe ff ff 45 89 fc e9 1a ff ff ff <0f> 0b 41 bc ea ff ff ff e9 0d ff ff ff 41 bc ea ff ff ff e9 02 ff
<4> [463.142054] RSP: 0018:ffffc90000593cd0 EFLAGS: 00010202
<4> [463.142056] RAX: 0000000000000010 RBX: ffff888274051000 RCX: 0000000000000000
<4> [463.142057] RDX: 0000000000000001 RSI: 0000000000000001 RDI: ffff888274051000
<4> [463.142058] RBP: ffff888238aa1018 R08: 0000000000000001 R09: 0000000000000001
<4> [463.142060] R10: ffffc90000593d90 R11: 00000000c79cdfd5 R12: ffff8882740510b0
<4> [463.142061] R13: 0000000000000001 R14: 0000000000000000 R15: 0000000000000001
<4> [463.142062] FS: 0000000000000000(0000) GS:ffff888276c00000(0000) knlGS:0000000000000000
<4> [463.142064] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
<4> [463.142065] CR2: 000055706f347d80 CR3: 0000000005610003 CR4: 00000000003606f0
<4> [463.142066] Call Trace:
<4> [463.142073] pci_enable_msi+0x11/0x20
<4> [463.142077] azx_resume+0x1ab/0x200 [snd_hda_intel]
<4> [463.142080] ? pci_pm_thaw+0x80/0x80
<4> [463.142084] dpm_run_callback+0x64/0x280
<4> [463.142089] device_resume+0xd4/0x1c0
<4> [463.142093] ? dpm_watchdog_set+0x60/0
While this would appear to be a bug in snd-hda, it does appear
inconsequential, at least for gfx-ci.
Downgrade the warning to an info, like the other already-enabled error
for MSI-X.
References: https://gitlab.freedesktop.org/drm/intel/-/issues/8041
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1687
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20200423082753.3899018-1-chris@chris-wilson.co.uk
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
drivers/pci/msi/msi.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/pci/msi/msi.c b/drivers/pci/msi/msi.c
index 2f647cac4cae3..a8d54aa21505f 100644
--- a/drivers/pci/msi/msi.c
+++ b/drivers/pci/msi/msi.c
@@ -430,8 +430,10 @@ int __pci_enable_msi_range(struct pci_dev *dev, int minvec, int maxvec,
if (maxvec < minvec)
return -ERANGE;
- if (WARN_ON_ONCE(dev->msi_enabled))
+ if (dev->msi_enabled) {
+ pci_info(dev, "can't enable MSI, already enabled\n");
return -EINVAL;
+ }
/* Test for the availability of MSI support */
if (!pci_msi_domain_supports(dev, 0, ALLOW_LEGACY))
--
GitLab