RE: [char-misc-next] mei: hook mei_device on class device

Usyskin, Alexander posted 1 patch 3 months, 1 week ago
drivers/misc/mei/pci-me.c       | 13 ++++++-------
drivers/misc/mei/pci-txe.c      | 13 ++++++-------
drivers/misc/mei/platform-vsc.c | 11 +++++------
3 files changed, 17 insertions(+), 20 deletions(-)
RE: [char-misc-next] mei: hook mei_device on class device
Posted by Usyskin, Alexander 3 months, 1 week ago
> Subject: Re: [char-misc-next] mei: hook mei_device on class device
> 
> Hi,
> 
> On Tue, Aug 26, 2025 at 03:56:17PM +0300, Alexander Usyskin wrote:
> > mei_device lifetime was managed by devm procedure of parent device.
> > But such memory is freed on device_del.
> > Mei_device object is used by client object that may be alive after
> > parent device is removed.
> > It may lead to use-after-free if discrete graphics driver unloads
> > mei_gsc auxiliary device while user-space holds open handle to mei
> > character device.
> >
> > Connect mei_device structure lifteme to mei class device lifetime
> > by adding mei_device free to class device remove callback.
> >
> > Move exising parent device pointer to separate field in mei_device
> > to avoid misuse.
> >
> > Allocate character device dynamically and allow to control its own
> > lifetime as it may outlive mei_device structure while character
> > device closes after parent device is removed from the system.
> >
> > Leave power management on parent device as we overwrite pci runtime
> > pm procedure and user-space is expecting it there.
> >
> > Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/14201
> > Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
> 
> This patch results in:
> 
> [   20.260342] mei mei0: wait hw ready failed
> [   20.264530] mei mei0: hw_start failed ret = -62 fw status = 00070355
> 002F0006 00000000 00000000 00000000 00000000
> [   22.308353] mei mei0: wait hw ready failed
> [   22.312489] mei mei0: hw_start failed ret = -62 fw status = 00070355
> 002F0006 00000000 00000000 00000000 00000000
> [   24.356433] mei mei0: wait hw ready failed
> [   24.360577] mei mei0: hw_start failed ret = -62 fw status = 00070355
> 002F0006 00000000 00000000 00000000 00000000
> [   24.370911] mei mei0: reset: reached maximal consecutive resets: disabling
> the device
> [   24.378787] mei mei0: reset failed ret = -19
> [   24.383079] mei mei0: link layer initialization failed.
> [   24.388329] mei_me 0000:00:16.0: init hw failure.
> [   51.219835] watchdog: BUG: soft lockup - CPU#0 stuck for 26s!
> [kworker/0:4:838]
> [   79.219833] watchdog: BUG: soft lockup - CPU#0 stuck for 52s!
> [kworker/0:4:838]
> [  107.219832] watchdog: BUG: soft lockup - CPU#0 stuck for 78s!
> [kworker/0:4:838]
> [  135.219831] watchdog: BUG: soft lockup - CPU#0 stuck for 104s!
> [kworker/0:4:838]
> 
> when trying to run v6.18-rc1/2/3 on a Skylaker server with non-functional
> MEI support.
> The problem is only seen if various debug options are enabled. Reverting the
> MEI
> patches since 6.17 fixes the problem (reverting this patch alone is not
> possible).
> 
> Bisect log and more verbose backtracec attached for reference.
> 
> Please let me know if there is anything I can do to help tracking down the
> problem.
> 
> Thanks,
> Guenter
> 
> ----
> Backtrace:
> 
> [   93.187907] watchdog: BUG: soft lockup - CPU#0 stuck for 52s!
> [kworker/0:2:834]
> [   93.187909] Modules linked in:
> [   93.187909] irq event stamp: 523739110
> [   93.187910] hardirqs last  enabled at (523739109): [<ffffffffa43975d7>]
> work_grab_pending+0x1a7/0x360
> [   93.187912] hardirqs last disabled at (523739110): [<ffffffffa5203b9e>]
> sysvec_apic_timer_interrupt+0xe/0x90
> [   93.187914] softirqs last  enabled at (523364938): [<ffffffffa4378d9b>]
> __irq_exit_rcu+0x6b/0x140
> [   93.187916] softirqs last disabled at (523364933): [<ffffffffa4378d9b>]
> __irq_exit_rcu+0x6b/0x140
> [   93.187918] CPU: 0 UID: 0 PID: 834 Comm: kworker/0:2 Tainted: G             L
> 6.18.0-dbg-DEV #2 NONE
> [   93.187920] Tainted: [L]=SOFTLOCKUP
> [   93.187920] Hardware name: Google LLC Indus/Indus_QC_03, BIOS
> 30.116.4 08/29/2025
> [   93.187922] Workqueue: events work_for_cpu_fn
> [   93.187923] RIP: 0010:work_grab_pending+0x36/0x360
> [   93.187925] Code: 41 54 53 48 83 ec 38 49 89 d7 89 34 24 48 89 fb 65 48
> 8b 05 cc eb 66 03 48 89 44 24 30 48 8d 47 48 48 89 44 24 08 eb 02 f3 90
> <9c> 8f 44 24 28 48 8b 44 24 28 fa 49 89 07 a9 00 02 00 00 74 05 e8
> [   93.187927] RSP: 0000:ffffb51d9b59fc60 EFLAGS: 00000202
> [   93.187928] RAX: 94f926a301ea6700 RBX: ffff9afc8231b418 RCX:
> 0000000000000002
> [   93.187928] RDX: 0000000000000006 RSI: ffff9afc81b70b20 RDI:
> ffffffffa43975d7
> [   93.187929] RBP: 0000000000000002 R08: 00000000000e00c8 R09:
> ffffffffffffffff
> [   93.187930] R10: ffffffffa43974b2 R11: 0000000000000000 R12:
> ffffffffa66e0848
> [   93.187930] R13: ffffffffa439de1d R14: ffffffffa43974b2 R15:
> ffffb51d9b59fcd0
> [   93.187931] FS:  0000000000000000(0000) GS:ffff9b59d8812000(0000)
> knlGS:0000000000000000
> [   93.187932] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [   93.187932] CR2: ffff9bba9ffff000 CR3: 0000005c0b44c001 CR4:
> 00000000007706f0
> [   93.187933] PKRU: 55555554
> [   93.187934] Call Trace:
> [   93.187934]  <TASK>
> [   93.187936]  ? lock_release+0x100/0x340
> [   93.187939]  ? process_scheduled_works+0x26d/0x630
> [   93.187941]  __cancel_work+0x29/0xf0
> [   93.187945]  cancel_work_sync+0x18/0x80
> [   93.187947]  mei_cancel_work+0x19/0x40
> [   93.187950]  mei_me_probe+0x2bf/0x330
> [   93.187952]  ? process_scheduled_works+0x26d/0x630
> [   93.187954]  local_pci_probe+0x45/0x90
> [   93.187957]  work_for_cpu_fn+0x1b/0x30
> [   93.187959]  process_scheduled_works+0x2d3/0x630
> [   93.187965]  worker_thread+0x1e8/0x2f0
> [   93.187969]  kthread+0x21f/0x240
> [   93.187972]  ? __pfx_worker_thread+0x10/0x10
> [   93.187974]  ? lock_release+0x100/0x340
> [   93.187975]  ? ret_from_fork+0x2d/0x2b0
> [   93.187979]  ? __pfx_kthread+0x10/0x10
> [   93.187982]  ret_from_fork+0x197/0x2b0
> [   93.187984]  ? __pfx_kthread+0x10/0x10
> [   93.187986]  ret_from_fork_asm+0x1a/0x30
> [   93.187994]  </TASK>
> 
> ---
> Bisect log:
> 
> # bad: [3a8660878839faadb4f1a6dd72c3179c1df56787] Linux 6.18-rc1
> # good: [e5f0a698b34ed76002dc5cff3804a61c80233a7a] Linux 6.17
> git bisect start 'HEAD' 'v6.17'
> # good: [58809f614e0e3f4e12b489bddf680bfeb31c0a20] Merge tag 'drm-
> next-2025-10-01' of https://gitlab.freedesktop.org/drm/kernel
> git bisect good 58809f614e0e3f4e12b489bddf680bfeb31c0a20
> # good: [bed0653fe2aacb0ca8196075cffc9e7062e74927] Merge tag 'iommu-
> updates-v6.18' of git://git.kernel.org/pub/scm/linux/kernel/git/iommu/linux
> git bisect good bed0653fe2aacb0ca8196075cffc9e7062e74927
> # bad: [6a74422b9710e987c7d6b85a1ade7330b1e61626] Merge tag
> 'mips_6.18' of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux
> git bisect bad 6a74422b9710e987c7d6b85a1ade7330b1e61626
> # good: [b66451723c45b791fd2824d1b8f62fe498989e23] Merge tag
> 'platform-drivers-x86-v6.18-1' of
> git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86
> git bisect good b66451723c45b791fd2824d1b8f62fe498989e23
> # good: [59697e061f6aec86d5738cd4752e16520f1d60dc] Merge tag
> 'staging-6.18-rc1' of
> git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging
> git bisect good 59697e061f6aec86d5738cd4752e16520f1d60dc
> # good: [561285d048053fec8a3d6d1e3ddc60df11c393a0] MAINTAINERS:
> Support ROHM BD79112 ADC
> git bisect good 561285d048053fec8a3d6d1e3ddc60df11c393a0
> # bad: [eafedbc7c050c44744fbdf80bdf3315e860b7513] rust_binder: add
> Rust Binder driver
> git bisect bad eafedbc7c050c44744fbdf80bdf3315e860b7513
> # good: [0c82fd9609a1e4bf1db84b0fd56bc3b2773da179] ibmasm: Replace
> kzalloc() + copy_from_user() with memdup_user_nul()
> git bisect good 0c82fd9609a1e4bf1db84b0fd56bc3b2773da179
> # bad: [ef509269d93d9832b366005f9626b44e38cc0ca7] Merge tag
> 'counter-updates-for-6.18' of
> ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/wbg/counter into char-
> misc-next
> git bisect bad ef509269d93d9832b366005f9626b44e38cc0ca7
> # bad: [ae0de6333368fd8c4535f5dbdfe1b2660438e089] slimbus:
> messaging: fix "transfered"->"transferred"
> git bisect bad ae0de6333368fd8c4535f5dbdfe1b2660438e089
> # bad: [7704e6be4ed2835832c445807cdcb2d56d8a8430] mei: hook
> mei_device on class device
> git bisect bad 7704e6be4ed2835832c445807cdcb2d56d8a8430
> # good: [ceda408c0d1d41094ad125332c6fb1d488e61c0c] misc: remove
> ineffective WARN_ON() check from misc_deregister()
> git bisect good ceda408c0d1d41094ad125332c6fb1d488e61c0c
> # good: [76254bc489d39dae9a3427f0984fe64213d20548] cdx: Fix device
> node reference leak in cdx_msi_domain_init
> git bisect good 76254bc489d39dae9a3427f0984fe64213d20548
> # first bad commit: [7704e6be4ed2835832c445807cdcb2d56d8a8430] mei:
> hook mei_device on class device


Seems I've missed the error flow in probe (my test machines always have an ME in a good state...).

Below patch should fix the problem, can you confirm?

From c58f311df60f26df2efe1e0f9fc523bfa4b93936 Mon Sep 17 00:00:00 2001
From: Alexander Usyskin <alexander.usyskin@intel.com>
Date: Sun, 2 Nov 2025 10:57:22 +0200
Subject: [PATCH] mei: fix error flow in probe
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Dismantle class device last in probe error flow to avoid accessing freed memory like:

[   87.926774] WARNING: CPU: 9 PID: 518 at kernel/workqueue.c:4234
__flush_work+0x340/0x390
...
[   87.926912] Workqueue: async async_run_entry_fn
[   87.926918] RIP: e030:__flush_work+0x340/0x390
[   87.926923] Code: 26 9d 05 00 65 48 8b 15 26 3c ca 02 48 85 db 48 8b
04 24 48 89 54 24 58 0f 85 de fe ff ff e9 f6 fd ff ff 0f 0b e9 77 ff ff
ff <0f> 0b e9 70 ff ff ff 0f 0b e9 19 ff ff ff e8 7d 8b 0e 01 48 89 de
[   87.926931] RSP: e02b:ffffc900412ebc00 EFLAGS: 00010246
[   87.926936] RAX: 0000000000000000 RBX: ffff888103e55090 RCX: 0000000000000000
[   87.926941] RDX: 000fffffffe00000 RSI: 0000000000000001 RDI: ffffc900412ebc60
[   87.926945] RBP: ffff888103e55090 R08: ffffffffc1266ec8 R09: ffff8881109076e8
[   87.926949] R10: 0000000080040003 R11: 0000000000000000 R12: ffff888103e54000
[   87.926953] R13: ffffc900412ebc18 R14: 0000000000000001 R15: 0000000000000000
[   87.926962] FS:  0000000000000000(0000) GS:ffff888233238000(0000) knlGS:0000000000000000
[   87.926967] CS:  e030 DS: 0000 ES: 0000 CR0: 0000000080050033
[   87.926971] CR2: 00007e7923b32708 CR3: 00000001088df000 CR4: 0000000000050660
[   87.926977] Call Trace:
[   87.926981]  <TASK>
[   87.926987]  ? __call_rcu_common.constprop.0+0x11e/0x310
[   87.926993]  cancel_work_sync+0x5e/0x80
[   87.926999]  mei_cancel_work+0x19/0x40 [mei]
[   87.927051]  mei_me_probe+0x273/0x2b0 [mei_me]
[   87.927060]  local_pci_probe+0x45/0x90
[   87.927066]  pci_call_probe+0x5b/0x180
[   87.927070]  pci_device_probe+0x95/0x140
[   87.927074]  ? driver_sysfs_add+0x57/0xc0
[   87.927079]  really_probe+0xde/0x340
[   87.927083]  ? pm_runtime_barrier+0x54/0x90
[   87.927087]  __driver_probe_device+0x78/0x110
[   87.927092]  driver_probe_device+0x1f/0xa0
[   87.927095]  __driver_attach_async_helper+0x5e/0xe0
[   87.927100]  async_run_entry_fn+0x34/0x130
[   87.927104]  process_one_work+0x18d/0x340
[   87.927108]  worker_thread+0x256/0x3a0
[   87.927111]  ? __pfx_worker_thread+0x10/0x10
[   87.927115]  kthread+0xfc/0x240
[   87.927120]  ? __pfx_kthread+0x10/0x10
[   87.927124]  ? __pfx_kthread+0x10/0x10
[   87.927127]  ret_from_fork+0xf5/0x110
[   87.927132]  ? __pfx_kthread+0x10/0x10
[   87.927136]  ret_from_fork_asm+0x1a/0x30
[   87.927141]  </TASK>

Reported-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Reported-by: Guenter Roeck <linux@roeck-us.net>
Fixes: 7704e6be4ed2 ("mei: hook mei_device on class device")
Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
---
 drivers/misc/mei/pci-me.c       | 13 ++++++-------
 drivers/misc/mei/pci-txe.c      | 13 ++++++-------
 drivers/misc/mei/platform-vsc.c | 11 +++++------
 3 files changed, 17 insertions(+), 20 deletions(-)

diff --git a/drivers/misc/mei/pci-me.c b/drivers/misc/mei/pci-me.c
index b017ff29dbd1..73cad914be9f 100644
--- a/drivers/misc/mei/pci-me.c
+++ b/drivers/misc/mei/pci-me.c
@@ -223,6 +223,10 @@ static int mei_me_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	hw->mem_addr = pcim_iomap_table(pdev)[0];
 	hw->read_fws = mei_me_read_fws;
 
+	err = mei_register(dev, &pdev->dev);
+	if (err)
+		goto end;
+
 	pci_enable_msi(pdev);
 
 	hw->irq = pdev->irq;
@@ -237,13 +241,9 @@ static int mei_me_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (err) {
 		dev_err(&pdev->dev, "request_threaded_irq failure. irq = %d\n",
 		       pdev->irq);
-		goto end;
+		goto deregister;
 	}
 
-	err = mei_register(dev, &pdev->dev);
-	if (err)
-		goto release_irq;
-
 	if (mei_start(dev)) {
 		dev_err(&pdev->dev, "init hw failure.\n");
 		err = -ENODEV;
@@ -283,11 +283,10 @@ static int mei_me_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	return 0;
 
 deregister:
-	mei_deregister(dev);
-release_irq:
 	mei_cancel_work(dev);
 	mei_disable_interrupts(dev);
 	free_irq(pdev->irq, dev);
+	mei_deregister(dev);
 end:
 	dev_err(&pdev->dev, "initialization failed.\n");
 	return err;
diff --git a/drivers/misc/mei/pci-txe.c b/drivers/misc/mei/pci-txe.c
index 06b55a891c6b..98d1bc2c7f4b 100644
--- a/drivers/misc/mei/pci-txe.c
+++ b/drivers/misc/mei/pci-txe.c
@@ -87,6 +87,10 @@ static int mei_txe_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	hw = to_txe_hw(dev);
 	hw->mem_addr = pcim_iomap_table(pdev);
 
+	err = mei_register(dev, &pdev->dev);
+	if (err)
+		goto end;
+
 	pci_enable_msi(pdev);
 
 	/* clear spurious interrupts */
@@ -106,13 +110,9 @@ static int mei_txe_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (err) {
 		dev_err(&pdev->dev, "mei: request_threaded_irq failure. irq = %d\n",
 			pdev->irq);
-		goto end;
+		goto deregister;
 	}
 
-	err = mei_register(dev, &pdev->dev);
-	if (err)
-		goto release_irq;
-
 	if (mei_start(dev)) {
 		dev_err(&pdev->dev, "init hw failure.\n");
 		err = -ENODEV;
@@ -145,11 +145,10 @@ static int mei_txe_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	return 0;
 
 deregister:
-	mei_deregister(dev);
-release_irq:
 	mei_cancel_work(dev);
 	mei_disable_interrupts(dev);
 	free_irq(pdev->irq, dev);
+	mei_deregister(dev);
 end:
 	dev_err(&pdev->dev, "initialization failed.\n");
 	return err;
diff --git a/drivers/misc/mei/platform-vsc.c b/drivers/misc/mei/platform-vsc.c
index 288e7b72e942..9787b9cee71c 100644
--- a/drivers/misc/mei/platform-vsc.c
+++ b/drivers/misc/mei/platform-vsc.c
@@ -362,28 +362,27 @@ static int mei_vsc_probe(struct platform_device *pdev)
 
 	ret = mei_register(mei_dev, dev);
 	if (ret)
-		goto err_dereg;
+		goto err;
 
 	ret = mei_start(mei_dev);
 	if (ret) {
 		dev_err_probe(dev, ret, "init hw failed\n");
-		goto err_cancel;
+		goto err;
 	}
 
 	pm_runtime_enable(mei_dev->parent);
 
 	return 0;
 
-err_dereg:
-	mei_deregister(mei_dev);
-
-err_cancel:
+err:
 	mei_cancel_work(mei_dev);
 
 	vsc_tp_register_event_cb(tp, NULL, NULL);
 
 	mei_disable_interrupts(mei_dev);
 
+	mei_deregister(mei_dev);
+
 	return ret;
 }
 
-- 
2.43.0




Re: [char-misc-next] mei: hook mei_device on class device
Posted by Guenter Roeck 3 months, 1 week ago
On Sun, Nov 2, 2025 at 7:00 AM Usyskin, Alexander <linux@roeck-us.net> wrote:
...
>
> Seems I've missed the error flow in probe (my test machines always have an ME in a good state...).
>
> Below patch should fix the problem, can you confirm?
>
Yes, it does. With this patch applied, the log output is:

...
[   16.406790] mtdoops: mtd device (mtddev=name/number) must be supplied
[   18.380400] mei mei0: wait hw ready failed
[   18.384594] mei mei0: hw_start failed ret = -62 fw status =
00070355 002F0006 00000000 00000000 00000000 00000000
[   20.428395] mei mei0: wait hw ready failed
[   20.432542] mei mei0: hw_start failed ret = -62 fw status =
00070355 002F0006 00000000 00000000 00000000 00000000
[   22.476390] mei mei0: wait hw ready failed
[   22.480533] mei mei0: hw_start failed ret = -62 fw status =
00070355 002F0006 00000000 00000000 00000000 00000000
[   22.490847] mei mei0: reset: reached maximal consecutive resets:
disabling the device
[   22.498718] mei mei0: reset failed ret = -19
[   22.503012] mei mei0: link layer initialization failed.
[   22.508261] mei_me 0000:00:16.0: init hw failure.
[   22.513662] mei_me 0000:00:16.0: initialization failed.
initramfs(out): Mounting system filesystems for initramfs init.d
...

Official Tested-by: tag below. Note that it is from my Google address
since this is where I tested the fix.

 Thanks a lot for the quick turnaround!

Guenter

> From c58f311df60f26df2efe1e0f9fc523bfa4b93936 Mon Sep 17 00:00:00 2001
> From: Alexander Usyskin <alexander.usyskin@intel.com>
> Date: Sun, 2 Nov 2025 10:57:22 +0200
> Subject: [PATCH] mei: fix error flow in probe
> MIME-Version: 1.0
> Content-Type: text/plain; charset=UTF-8
> Content-Transfer-Encoding: 8bit
>
> Dismantle class device last in probe error flow to avoid accessing freed memory like:
>
> [   87.926774] WARNING: CPU: 9 PID: 518 at kernel/workqueue.c:4234
> __flush_work+0x340/0x390
> ...
> [   87.926912] Workqueue: async async_run_entry_fn
> [   87.926918] RIP: e030:__flush_work+0x340/0x390
> [   87.926923] Code: 26 9d 05 00 65 48 8b 15 26 3c ca 02 48 85 db 48 8b
> 04 24 48 89 54 24 58 0f 85 de fe ff ff e9 f6 fd ff ff 0f 0b e9 77 ff ff
> ff <0f> 0b e9 70 ff ff ff 0f 0b e9 19 ff ff ff e8 7d 8b 0e 01 48 89 de
> [   87.926931] RSP: e02b:ffffc900412ebc00 EFLAGS: 00010246
> [   87.926936] RAX: 0000000000000000 RBX: ffff888103e55090 RCX: 0000000000000000
> [   87.926941] RDX: 000fffffffe00000 RSI: 0000000000000001 RDI: ffffc900412ebc60
> [   87.926945] RBP: ffff888103e55090 R08: ffffffffc1266ec8 R09: ffff8881109076e8
> [   87.926949] R10: 0000000080040003 R11: 0000000000000000 R12: ffff888103e54000
> [   87.926953] R13: ffffc900412ebc18 R14: 0000000000000001 R15: 0000000000000000
> [   87.926962] FS:  0000000000000000(0000) GS:ffff888233238000(0000) knlGS:0000000000000000
> [   87.926967] CS:  e030 DS: 0000 ES: 0000 CR0: 0000000080050033
> [   87.926971] CR2: 00007e7923b32708 CR3: 00000001088df000 CR4: 0000000000050660
> [   87.926977] Call Trace:
> [   87.926981]  <TASK>
> [   87.926987]  ? __call_rcu_common.constprop.0+0x11e/0x310
> [   87.926993]  cancel_work_sync+0x5e/0x80
> [   87.926999]  mei_cancel_work+0x19/0x40 [mei]
> [   87.927051]  mei_me_probe+0x273/0x2b0 [mei_me]
> [   87.927060]  local_pci_probe+0x45/0x90
> [   87.927066]  pci_call_probe+0x5b/0x180
> [   87.927070]  pci_device_probe+0x95/0x140
> [   87.927074]  ? driver_sysfs_add+0x57/0xc0
> [   87.927079]  really_probe+0xde/0x340
> [   87.927083]  ? pm_runtime_barrier+0x54/0x90
> [   87.927087]  __driver_probe_device+0x78/0x110
> [   87.927092]  driver_probe_device+0x1f/0xa0
> [   87.927095]  __driver_attach_async_helper+0x5e/0xe0
> [   87.927100]  async_run_entry_fn+0x34/0x130
> [   87.927104]  process_one_work+0x18d/0x340
> [   87.927108]  worker_thread+0x256/0x3a0
> [   87.927111]  ? __pfx_worker_thread+0x10/0x10
> [   87.927115]  kthread+0xfc/0x240
> [   87.927120]  ? __pfx_kthread+0x10/0x10
> [   87.927124]  ? __pfx_kthread+0x10/0x10
> [   87.927127]  ret_from_fork+0xf5/0x110
> [   87.927132]  ? __pfx_kthread+0x10/0x10
> [   87.927136]  ret_from_fork_asm+0x1a/0x30
> [   87.927141]  </TASK>
>
> Reported-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
> Reported-by: Guenter Roeck <linux@roeck-us.net>
> Fixes: 7704e6be4ed2 ("mei: hook mei_device on class device")
> Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>

Tested-by: Guenter Roeck <groeck@google.com>

>
> ---
>  drivers/misc/mei/pci-me.c       | 13 ++++++-------
>  drivers/misc/mei/pci-txe.c      | 13 ++++++-------
>  drivers/misc/mei/platform-vsc.c | 11 +++++------
>  3 files changed, 17 insertions(+), 20 deletions(-)
>
> diff --git a/drivers/misc/mei/pci-me.c b/drivers/misc/mei/pci-me.c
> index b017ff29dbd1..73cad914be9f 100644
> --- a/drivers/misc/mei/pci-me.c
> +++ b/drivers/misc/mei/pci-me.c
> @@ -223,6 +223,10 @@ static int mei_me_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
>         hw->mem_addr = pcim_iomap_table(pdev)[0];
>         hw->read_fws = mei_me_read_fws;
>
> +       err = mei_register(dev, &pdev->dev);
> +       if (err)
> +               goto end;
> +
>         pci_enable_msi(pdev);
>
>         hw->irq = pdev->irq;
> @@ -237,13 +241,9 @@ static int mei_me_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
>         if (err) {
>                 dev_err(&pdev->dev, "request_threaded_irq failure. irq = %d\n",
>                        pdev->irq);
> -               goto end;
> +               goto deregister;
>         }
>
> -       err = mei_register(dev, &pdev->dev);
> -       if (err)
> -               goto release_irq;
> -
>         if (mei_start(dev)) {
>                 dev_err(&pdev->dev, "init hw failure.\n");
>                 err = -ENODEV;
> @@ -283,11 +283,10 @@ static int mei_me_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
>         return 0;
>
>  deregister:
> -       mei_deregister(dev);
> -release_irq:
>         mei_cancel_work(dev);
>         mei_disable_interrupts(dev);
>         free_irq(pdev->irq, dev);
> +       mei_deregister(dev);
>  end:
>         dev_err(&pdev->dev, "initialization failed.\n");
>         return err;
> diff --git a/drivers/misc/mei/pci-txe.c b/drivers/misc/mei/pci-txe.c
> index 06b55a891c6b..98d1bc2c7f4b 100644
> --- a/drivers/misc/mei/pci-txe.c
> +++ b/drivers/misc/mei/pci-txe.c
> @@ -87,6 +87,10 @@ static int mei_txe_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
>         hw = to_txe_hw(dev);
>         hw->mem_addr = pcim_iomap_table(pdev);
>
> +       err = mei_register(dev, &pdev->dev);
> +       if (err)
> +               goto end;
> +
>         pci_enable_msi(pdev);
>
>         /* clear spurious interrupts */
> @@ -106,13 +110,9 @@ static int mei_txe_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
>         if (err) {
>                 dev_err(&pdev->dev, "mei: request_threaded_irq failure. irq = %d\n",
>                         pdev->irq);
> -               goto end;
> +               goto deregister;
>         }
>
> -       err = mei_register(dev, &pdev->dev);
> -       if (err)
> -               goto release_irq;
> -
>         if (mei_start(dev)) {
>                 dev_err(&pdev->dev, "init hw failure.\n");
>                 err = -ENODEV;
> @@ -145,11 +145,10 @@ static int mei_txe_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
>         return 0;
>
>  deregister:
> -       mei_deregister(dev);
> -release_irq:
>         mei_cancel_work(dev);
>         mei_disable_interrupts(dev);
>         free_irq(pdev->irq, dev);
> +       mei_deregister(dev);
>  end:
>         dev_err(&pdev->dev, "initialization failed.\n");
>         return err;
> diff --git a/drivers/misc/mei/platform-vsc.c b/drivers/misc/mei/platform-vsc.c
> index 288e7b72e942..9787b9cee71c 100644
> --- a/drivers/misc/mei/platform-vsc.c
> +++ b/drivers/misc/mei/platform-vsc.c
> @@ -362,28 +362,27 @@ static int mei_vsc_probe(struct platform_device *pdev)
>
>         ret = mei_register(mei_dev, dev);
>         if (ret)
> -               goto err_dereg;
> +               goto err;
>
>         ret = mei_start(mei_dev);
>         if (ret) {
>                 dev_err_probe(dev, ret, "init hw failed\n");
> -               goto err_cancel;
> +               goto err;
>         }
>
>         pm_runtime_enable(mei_dev->parent);
>
>         return 0;
>
> -err_dereg:
> -       mei_deregister(mei_dev);
> -
> -err_cancel:
> +err:
>         mei_cancel_work(mei_dev);
>
>         vsc_tp_register_event_cb(tp, NULL, NULL);
>
>         mei_disable_interrupts(mei_dev);
>
> +       mei_deregister(mei_dev);
> +
>         return ret;
>  }
>
> --
> 2.43.0
>
>
>
>
Re: [char-misc-next] mei: hook mei_device on class device
Posted by Guenter Roeck 3 months, 1 week ago
On 11/2/25 07:37, Guenter Roeck wrote:
> On Sun, Nov 2, 2025 at 7:00 AM Usyskin, Alexander <linux@roeck-us.net> wrote:

That got messed up. Sorry for that. Really sending to Alexander this time.

Guenter

> ...
>>
>> Seems I've missed the error flow in probe (my test machines always have an ME in a good state...).
>>
>> Below patch should fix the problem, can you confirm?
>>
> Yes, it does. With this patch applied, the log output is:
> 
> ...
> [   16.406790] mtdoops: mtd device (mtddev=name/number) must be supplied
> [   18.380400] mei mei0: wait hw ready failed
> [   18.384594] mei mei0: hw_start failed ret = -62 fw status =
> 00070355 002F0006 00000000 00000000 00000000 00000000
> [   20.428395] mei mei0: wait hw ready failed
> [   20.432542] mei mei0: hw_start failed ret = -62 fw status =
> 00070355 002F0006 00000000 00000000 00000000 00000000
> [   22.476390] mei mei0: wait hw ready failed
> [   22.480533] mei mei0: hw_start failed ret = -62 fw status =
> 00070355 002F0006 00000000 00000000 00000000 00000000
> [   22.490847] mei mei0: reset: reached maximal consecutive resets:
> disabling the device
> [   22.498718] mei mei0: reset failed ret = -19
> [   22.503012] mei mei0: link layer initialization failed.
> [   22.508261] mei_me 0000:00:16.0: init hw failure.
> [   22.513662] mei_me 0000:00:16.0: initialization failed.
> initramfs(out): Mounting system filesystems for initramfs init.d
> ...
> 
> Official Tested-by: tag below. Note that it is from my Google address
> since this is where I tested the fix.
> 
>   Thanks a lot for the quick turnaround!
> 
> Guenter
> 
>>  From c58f311df60f26df2efe1e0f9fc523bfa4b93936 Mon Sep 17 00:00:00 2001
>> From: Alexander Usyskin <alexander.usyskin@intel.com>
>> Date: Sun, 2 Nov 2025 10:57:22 +0200
>> Subject: [PATCH] mei: fix error flow in probe
>> MIME-Version: 1.0
>> Content-Type: text/plain; charset=UTF-8
>> Content-Transfer-Encoding: 8bit
>>
>> Dismantle class device last in probe error flow to avoid accessing freed memory like:
>>
>> [   87.926774] WARNING: CPU: 9 PID: 518 at kernel/workqueue.c:4234
>> __flush_work+0x340/0x390
>> ...
>> [   87.926912] Workqueue: async async_run_entry_fn
>> [   87.926918] RIP: e030:__flush_work+0x340/0x390
>> [   87.926923] Code: 26 9d 05 00 65 48 8b 15 26 3c ca 02 48 85 db 48 8b
>> 04 24 48 89 54 24 58 0f 85 de fe ff ff e9 f6 fd ff ff 0f 0b e9 77 ff ff
>> ff <0f> 0b e9 70 ff ff ff 0f 0b e9 19 ff ff ff e8 7d 8b 0e 01 48 89 de
>> [   87.926931] RSP: e02b:ffffc900412ebc00 EFLAGS: 00010246
>> [   87.926936] RAX: 0000000000000000 RBX: ffff888103e55090 RCX: 0000000000000000
>> [   87.926941] RDX: 000fffffffe00000 RSI: 0000000000000001 RDI: ffffc900412ebc60
>> [   87.926945] RBP: ffff888103e55090 R08: ffffffffc1266ec8 R09: ffff8881109076e8
>> [   87.926949] R10: 0000000080040003 R11: 0000000000000000 R12: ffff888103e54000
>> [   87.926953] R13: ffffc900412ebc18 R14: 0000000000000001 R15: 0000000000000000
>> [   87.926962] FS:  0000000000000000(0000) GS:ffff888233238000(0000) knlGS:0000000000000000
>> [   87.926967] CS:  e030 DS: 0000 ES: 0000 CR0: 0000000080050033
>> [   87.926971] CR2: 00007e7923b32708 CR3: 00000001088df000 CR4: 0000000000050660
>> [   87.926977] Call Trace:
>> [   87.926981]  <TASK>
>> [   87.926987]  ? __call_rcu_common.constprop.0+0x11e/0x310
>> [   87.926993]  cancel_work_sync+0x5e/0x80
>> [   87.926999]  mei_cancel_work+0x19/0x40 [mei]
>> [   87.927051]  mei_me_probe+0x273/0x2b0 [mei_me]
>> [   87.927060]  local_pci_probe+0x45/0x90
>> [   87.927066]  pci_call_probe+0x5b/0x180
>> [   87.927070]  pci_device_probe+0x95/0x140
>> [   87.927074]  ? driver_sysfs_add+0x57/0xc0
>> [   87.927079]  really_probe+0xde/0x340
>> [   87.927083]  ? pm_runtime_barrier+0x54/0x90
>> [   87.927087]  __driver_probe_device+0x78/0x110
>> [   87.927092]  driver_probe_device+0x1f/0xa0
>> [   87.927095]  __driver_attach_async_helper+0x5e/0xe0
>> [   87.927100]  async_run_entry_fn+0x34/0x130
>> [   87.927104]  process_one_work+0x18d/0x340
>> [   87.927108]  worker_thread+0x256/0x3a0
>> [   87.927111]  ? __pfx_worker_thread+0x10/0x10
>> [   87.927115]  kthread+0xfc/0x240
>> [   87.927120]  ? __pfx_kthread+0x10/0x10
>> [   87.927124]  ? __pfx_kthread+0x10/0x10
>> [   87.927127]  ret_from_fork+0xf5/0x110
>> [   87.927132]  ? __pfx_kthread+0x10/0x10
>> [   87.927136]  ret_from_fork_asm+0x1a/0x30
>> [   87.927141]  </TASK>
>>
>> Reported-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
>> Reported-by: Guenter Roeck <linux@roeck-us.net>
>> Fixes: 7704e6be4ed2 ("mei: hook mei_device on class device")
>> Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
> 
> Tested-by: Guenter Roeck <groeck@google.com>
> 
>>
>> ---
>>   drivers/misc/mei/pci-me.c       | 13 ++++++-------
>>   drivers/misc/mei/pci-txe.c      | 13 ++++++-------
>>   drivers/misc/mei/platform-vsc.c | 11 +++++------
>>   3 files changed, 17 insertions(+), 20 deletions(-)
>>
>> diff --git a/drivers/misc/mei/pci-me.c b/drivers/misc/mei/pci-me.c
>> index b017ff29dbd1..73cad914be9f 100644
>> --- a/drivers/misc/mei/pci-me.c
>> +++ b/drivers/misc/mei/pci-me.c
>> @@ -223,6 +223,10 @@ static int mei_me_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
>>          hw->mem_addr = pcim_iomap_table(pdev)[0];
>>          hw->read_fws = mei_me_read_fws;
>>
>> +       err = mei_register(dev, &pdev->dev);
>> +       if (err)
>> +               goto end;
>> +
>>          pci_enable_msi(pdev);
>>
>>          hw->irq = pdev->irq;
>> @@ -237,13 +241,9 @@ static int mei_me_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
>>          if (err) {
>>                  dev_err(&pdev->dev, "request_threaded_irq failure. irq = %d\n",
>>                         pdev->irq);
>> -               goto end;
>> +               goto deregister;
>>          }
>>
>> -       err = mei_register(dev, &pdev->dev);
>> -       if (err)
>> -               goto release_irq;
>> -
>>          if (mei_start(dev)) {
>>                  dev_err(&pdev->dev, "init hw failure.\n");
>>                  err = -ENODEV;
>> @@ -283,11 +283,10 @@ static int mei_me_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
>>          return 0;
>>
>>   deregister:
>> -       mei_deregister(dev);
>> -release_irq:
>>          mei_cancel_work(dev);
>>          mei_disable_interrupts(dev);
>>          free_irq(pdev->irq, dev);
>> +       mei_deregister(dev);
>>   end:
>>          dev_err(&pdev->dev, "initialization failed.\n");
>>          return err;
>> diff --git a/drivers/misc/mei/pci-txe.c b/drivers/misc/mei/pci-txe.c
>> index 06b55a891c6b..98d1bc2c7f4b 100644
>> --- a/drivers/misc/mei/pci-txe.c
>> +++ b/drivers/misc/mei/pci-txe.c
>> @@ -87,6 +87,10 @@ static int mei_txe_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
>>          hw = to_txe_hw(dev);
>>          hw->mem_addr = pcim_iomap_table(pdev);
>>
>> +       err = mei_register(dev, &pdev->dev);
>> +       if (err)
>> +               goto end;
>> +
>>          pci_enable_msi(pdev);
>>
>>          /* clear spurious interrupts */
>> @@ -106,13 +110,9 @@ static int mei_txe_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
>>          if (err) {
>>                  dev_err(&pdev->dev, "mei: request_threaded_irq failure. irq = %d\n",
>>                          pdev->irq);
>> -               goto end;
>> +               goto deregister;
>>          }
>>
>> -       err = mei_register(dev, &pdev->dev);
>> -       if (err)
>> -               goto release_irq;
>> -
>>          if (mei_start(dev)) {
>>                  dev_err(&pdev->dev, "init hw failure.\n");
>>                  err = -ENODEV;
>> @@ -145,11 +145,10 @@ static int mei_txe_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
>>          return 0;
>>
>>   deregister:
>> -       mei_deregister(dev);
>> -release_irq:
>>          mei_cancel_work(dev);
>>          mei_disable_interrupts(dev);
>>          free_irq(pdev->irq, dev);
>> +       mei_deregister(dev);
>>   end:
>>          dev_err(&pdev->dev, "initialization failed.\n");
>>          return err;
>> diff --git a/drivers/misc/mei/platform-vsc.c b/drivers/misc/mei/platform-vsc.c
>> index 288e7b72e942..9787b9cee71c 100644
>> --- a/drivers/misc/mei/platform-vsc.c
>> +++ b/drivers/misc/mei/platform-vsc.c
>> @@ -362,28 +362,27 @@ static int mei_vsc_probe(struct platform_device *pdev)
>>
>>          ret = mei_register(mei_dev, dev);
>>          if (ret)
>> -               goto err_dereg;
>> +               goto err;
>>
>>          ret = mei_start(mei_dev);
>>          if (ret) {
>>                  dev_err_probe(dev, ret, "init hw failed\n");
>> -               goto err_cancel;
>> +               goto err;
>>          }
>>
>>          pm_runtime_enable(mei_dev->parent);
>>
>>          return 0;
>>
>> -err_dereg:
>> -       mei_deregister(mei_dev);
>> -
>> -err_cancel:
>> +err:
>>          mei_cancel_work(mei_dev);
>>
>>          vsc_tp_register_event_cb(tp, NULL, NULL);
>>
>>          mei_disable_interrupts(mei_dev);
>>
>> +       mei_deregister(mei_dev);
>> +
>>          return ret;
>>   }
>>
>> --
>> 2.43.0
>>
>>
>>
>>

RE: [char-misc-next] mei: hook mei_device on class device
Posted by Usyskin, Alexander 3 months, 1 week ago
> Subject: Re: [char-misc-next] mei: hook mei_device on class device
> 
> On 11/2/25 07:37, Guenter Roeck wrote:
> > On Sun, Nov 2, 2025 at 7:00 AM Usyskin, Alexander <linux@roeck-us.net>
> wrote:
> 
> That got messed up. Sorry for that. Really sending to Alexander this time.
> 
> Guenter

Thanks for testing, sending the patch to Greg for next -rc

- - 
Thanks,
Sasha


> 
> > ...
> >>
> >> Seems I've missed the error flow in probe (my test machines always have
> an ME in a good state...).
> >>
> >> Below patch should fix the problem, can you confirm?
> >>
> > Yes, it does. With this patch applied, the log output is:
> >
> > ...
> > [   16.406790] mtdoops: mtd device (mtddev=name/number) must be
> supplied
> > [   18.380400] mei mei0: wait hw ready failed
> > [   18.384594] mei mei0: hw_start failed ret = -62 fw status =
> > 00070355 002F0006 00000000 00000000 00000000 00000000
> > [   20.428395] mei mei0: wait hw ready failed
> > [   20.432542] mei mei0: hw_start failed ret = -62 fw status =
> > 00070355 002F0006 00000000 00000000 00000000 00000000
> > [   22.476390] mei mei0: wait hw ready failed
> > [   22.480533] mei mei0: hw_start failed ret = -62 fw status =
> > 00070355 002F0006 00000000 00000000 00000000 00000000
> > [   22.490847] mei mei0: reset: reached maximal consecutive resets:
> > disabling the device
> > [   22.498718] mei mei0: reset failed ret = -19
> > [   22.503012] mei mei0: link layer initialization failed.
> > [   22.508261] mei_me 0000:00:16.0: init hw failure.
> > [   22.513662] mei_me 0000:00:16.0: initialization failed.
> > initramfs(out): Mounting system filesystems for initramfs init.d
> > ...
> >
> > Official Tested-by: tag below. Note that it is from my Google address
> > since this is where I tested the fix.
> >
> >   Thanks a lot for the quick turnaround!
> >
> > Guenter
> >
> >>  From c58f311df60f26df2efe1e0f9fc523bfa4b93936 Mon Sep 17
> 00:00:00 2001
> >> From: Alexander Usyskin <alexander.usyskin@intel.com>
> >> Date: Sun, 2 Nov 2025 10:57:22 +0200
> >> Subject: [PATCH] mei: fix error flow in probe
> >> MIME-Version: 1.0
> >> Content-Type: text/plain; charset=UTF-8
> >> Content-Transfer-Encoding: 8bit
> >>
> >> Dismantle class device last in probe error flow to avoid accessing freed
> memory like:
> >>
> >> [   87.926774] WARNING: CPU: 9 PID: 518 at kernel/workqueue.c:4234
> >> __flush_work+0x340/0x390
> >> ...
> >> [   87.926912] Workqueue: async async_run_entry_fn
> >> [   87.926918] RIP: e030:__flush_work+0x340/0x390
> >> [   87.926923] Code: 26 9d 05 00 65 48 8b 15 26 3c ca 02 48 85 db 48 8b
> >> 04 24 48 89 54 24 58 0f 85 de fe ff ff e9 f6 fd ff ff 0f 0b e9 77 ff ff
> >> ff <0f> 0b e9 70 ff ff ff 0f 0b e9 19 ff ff ff e8 7d 8b 0e 01 48 89 de
> >> [   87.926931] RSP: e02b:ffffc900412ebc00 EFLAGS: 00010246
> >> [   87.926936] RAX: 0000000000000000 RBX: ffff888103e55090 RCX:
> 0000000000000000
> >> [   87.926941] RDX: 000fffffffe00000 RSI: 0000000000000001 RDI:
> ffffc900412ebc60
> >> [   87.926945] RBP: ffff888103e55090 R08: ffffffffc1266ec8 R09:
> ffff8881109076e8
> >> [   87.926949] R10: 0000000080040003 R11: 0000000000000000 R12:
> ffff888103e54000
> >> [   87.926953] R13: ffffc900412ebc18 R14: 0000000000000001 R15:
> 0000000000000000
> >> [   87.926962] FS:  0000000000000000(0000)
> GS:ffff888233238000(0000) knlGS:0000000000000000
> >> [   87.926967] CS:  e030 DS: 0000 ES: 0000 CR0: 0000000080050033
> >> [   87.926971] CR2: 00007e7923b32708 CR3: 00000001088df000 CR4:
> 0000000000050660
> >> [   87.926977] Call Trace:
> >> [   87.926981]  <TASK>
> >> [   87.926987]  ? __call_rcu_common.constprop.0+0x11e/0x310
> >> [   87.926993]  cancel_work_sync+0x5e/0x80
> >> [   87.926999]  mei_cancel_work+0x19/0x40 [mei]
> >> [   87.927051]  mei_me_probe+0x273/0x2b0 [mei_me]
> >> [   87.927060]  local_pci_probe+0x45/0x90
> >> [   87.927066]  pci_call_probe+0x5b/0x180
> >> [   87.927070]  pci_device_probe+0x95/0x140
> >> [   87.927074]  ? driver_sysfs_add+0x57/0xc0
> >> [   87.927079]  really_probe+0xde/0x340
> >> [   87.927083]  ? pm_runtime_barrier+0x54/0x90
> >> [   87.927087]  __driver_probe_device+0x78/0x110
> >> [   87.927092]  driver_probe_device+0x1f/0xa0
> >> [   87.927095]  __driver_attach_async_helper+0x5e/0xe0
> >> [   87.927100]  async_run_entry_fn+0x34/0x130
> >> [   87.927104]  process_one_work+0x18d/0x340
> >> [   87.927108]  worker_thread+0x256/0x3a0
> >> [   87.927111]  ? __pfx_worker_thread+0x10/0x10
> >> [   87.927115]  kthread+0xfc/0x240
> >> [   87.927120]  ? __pfx_kthread+0x10/0x10
> >> [   87.927124]  ? __pfx_kthread+0x10/0x10
> >> [   87.927127]  ret_from_fork+0xf5/0x110
> >> [   87.927132]  ? __pfx_kthread+0x10/0x10
> >> [   87.927136]  ret_from_fork_asm+0x1a/0x30
> >> [   87.927141]  </TASK>
> >>
> >> Reported-by: Marek Marczykowski-Górecki
> <marmarek@invisiblethingslab.com>
> >> Reported-by: Guenter Roeck <linux@roeck-us.net>
> >> Fixes: 7704e6be4ed2 ("mei: hook mei_device on class device")
> >> Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
> >
> > Tested-by: Guenter Roeck <groeck@google.com>
> >
> >>
> >> ---
> >>   drivers/misc/mei/pci-me.c       | 13 ++++++-------
> >>   drivers/misc/mei/pci-txe.c      | 13 ++++++-------
> >>   drivers/misc/mei/platform-vsc.c | 11 +++++------
> >>   3 files changed, 17 insertions(+), 20 deletions(-)
> >>
> >> diff --git a/drivers/misc/mei/pci-me.c b/drivers/misc/mei/pci-me.c
> >> index b017ff29dbd1..73cad914be9f 100644
> >> --- a/drivers/misc/mei/pci-me.c
> >> +++ b/drivers/misc/mei/pci-me.c
> >> @@ -223,6 +223,10 @@ static int mei_me_probe(struct pci_dev *pdev,
> const struct pci_device_id *ent)
> >>          hw->mem_addr = pcim_iomap_table(pdev)[0];
> >>          hw->read_fws = mei_me_read_fws;
> >>
> >> +       err = mei_register(dev, &pdev->dev);
> >> +       if (err)
> >> +               goto end;
> >> +
> >>          pci_enable_msi(pdev);
> >>
> >>          hw->irq = pdev->irq;
> >> @@ -237,13 +241,9 @@ static int mei_me_probe(struct pci_dev *pdev,
> const struct pci_device_id *ent)
> >>          if (err) {
> >>                  dev_err(&pdev->dev, "request_threaded_irq failure. irq = %d\n",
> >>                         pdev->irq);
> >> -               goto end;
> >> +               goto deregister;
> >>          }
> >>
> >> -       err = mei_register(dev, &pdev->dev);
> >> -       if (err)
> >> -               goto release_irq;
> >> -
> >>          if (mei_start(dev)) {
> >>                  dev_err(&pdev->dev, "init hw failure.\n");
> >>                  err = -ENODEV;
> >> @@ -283,11 +283,10 @@ static int mei_me_probe(struct pci_dev *pdev,
> const struct pci_device_id *ent)
> >>          return 0;
> >>
> >>   deregister:
> >> -       mei_deregister(dev);
> >> -release_irq:
> >>          mei_cancel_work(dev);
> >>          mei_disable_interrupts(dev);
> >>          free_irq(pdev->irq, dev);
> >> +       mei_deregister(dev);
> >>   end:
> >>          dev_err(&pdev->dev, "initialization failed.\n");
> >>          return err;
> >> diff --git a/drivers/misc/mei/pci-txe.c b/drivers/misc/mei/pci-txe.c
> >> index 06b55a891c6b..98d1bc2c7f4b 100644
> >> --- a/drivers/misc/mei/pci-txe.c
> >> +++ b/drivers/misc/mei/pci-txe.c
> >> @@ -87,6 +87,10 @@ static int mei_txe_probe(struct pci_dev *pdev, const
> struct pci_device_id *ent)
> >>          hw = to_txe_hw(dev);
> >>          hw->mem_addr = pcim_iomap_table(pdev);
> >>
> >> +       err = mei_register(dev, &pdev->dev);
> >> +       if (err)
> >> +               goto end;
> >> +
> >>          pci_enable_msi(pdev);
> >>
> >>          /* clear spurious interrupts */
> >> @@ -106,13 +110,9 @@ static int mei_txe_probe(struct pci_dev *pdev,
> const struct pci_device_id *ent)
> >>          if (err) {
> >>                  dev_err(&pdev->dev, "mei: request_threaded_irq failure. irq =
> %d\n",
> >>                          pdev->irq);
> >> -               goto end;
> >> +               goto deregister;
> >>          }
> >>
> >> -       err = mei_register(dev, &pdev->dev);
> >> -       if (err)
> >> -               goto release_irq;
> >> -
> >>          if (mei_start(dev)) {
> >>                  dev_err(&pdev->dev, "init hw failure.\n");
> >>                  err = -ENODEV;
> >> @@ -145,11 +145,10 @@ static int mei_txe_probe(struct pci_dev *pdev,
> const struct pci_device_id *ent)
> >>          return 0;
> >>
> >>   deregister:
> >> -       mei_deregister(dev);
> >> -release_irq:
> >>          mei_cancel_work(dev);
> >>          mei_disable_interrupts(dev);
> >>          free_irq(pdev->irq, dev);
> >> +       mei_deregister(dev);
> >>   end:
> >>          dev_err(&pdev->dev, "initialization failed.\n");
> >>          return err;
> >> diff --git a/drivers/misc/mei/platform-vsc.c b/drivers/misc/mei/platform-
> vsc.c
> >> index 288e7b72e942..9787b9cee71c 100644
> >> --- a/drivers/misc/mei/platform-vsc.c
> >> +++ b/drivers/misc/mei/platform-vsc.c
> >> @@ -362,28 +362,27 @@ static int mei_vsc_probe(struct platform_device
> *pdev)
> >>
> >>          ret = mei_register(mei_dev, dev);
> >>          if (ret)
> >> -               goto err_dereg;
> >> +               goto err;
> >>
> >>          ret = mei_start(mei_dev);
> >>          if (ret) {
> >>                  dev_err_probe(dev, ret, "init hw failed\n");
> >> -               goto err_cancel;
> >> +               goto err;
> >>          }
> >>
> >>          pm_runtime_enable(mei_dev->parent);
> >>
> >>          return 0;
> >>
> >> -err_dereg:
> >> -       mei_deregister(mei_dev);
> >> -
> >> -err_cancel:
> >> +err:
> >>          mei_cancel_work(mei_dev);
> >>
> >>          vsc_tp_register_event_cb(tp, NULL, NULL);
> >>
> >>          mei_disable_interrupts(mei_dev);
> >>
> >> +       mei_deregister(mei_dev);
> >> +
> >>          return ret;
> >>   }
> >>
> >> --
> >> 2.43.0
> >>
> >>
> >>
> >>

Re: [char-misc-next] mei: hook mei_device on class device
Posted by Guenter Roeck 3 months, 1 week ago
On Sun, Nov 02, 2025 at 10:12:56AM +0000, Usyskin, Alexander wrote:
...
> 
> Seems I've missed the error flow in probe (my test machines always have an ME in a good state...).
> 
> Below patch should fix the problem, can you confirm?

testing.

Thanks!

Guenter