Skip to contentSkip to conversation

myrb: fixup null pointer access on myrb_cleanup()

PATCH Superseded Hannes Reinecke 2 participants 2 messages May 22, 2022

Subsystems (2)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 myrb: fixup null pointer access on myrb_cleanup()

When myrb_probe() fails the callback might not be set, so we need to validate the 'disable_intr' callback in myrb_cleanup() to not cause a null pointer exception. And while at it do not call myrb_cleanup() if we cannot enable the PCI device at all.

---
 drivers/scsi/myrb.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)
drivers/scsi/myrb.c +7 -3
index 71585528e8db..f460aed2435f 100644
@@ -1239,7 +1239,8 @@ static void myrb_cleanup(struct myrb_hba *cb)
 	myrb_unmap(cb);
 
 	if (cb->mmio_base) {
-		cb->disable_intr(cb->io_base);
+		if (cb->disable_intr)
+			cb->disable_intr(cb->io_base);
 		iounmap(cb->mmio_base);
 	}
 	if (cb->irq)
@@ -3414,8 +3415,11 @@ static struct myrb_hba *myrb_detect(struct pci_dev *pdev,
 	mutex_init(&cb->dma_mutex);
 	cb->pdev = pdev;
 
-	if (pci_enable_device(pdev))
-		goto failure;
+	if (pci_enable_device(pdev)) {
+		dev_err(&pdev->dev, "Failed to enable PCI device\n");
+		scsi_host_put(shost);
+		return NULL;
+	}
 
 	if (privdata->hw_init == DAC960_PD_hw_init ||
 	    privdata->hw_init == DAC960_P_hw_init) {
Signature · 1 line
2.29.2
#1 lore raw
Quoted 41 lines · Hannes Reinecke

When myrb_probe() fails the callback might not be set, so we need to validate the 'disable_intr' callback in myrb_cleanup() to not cause a null pointer exception. And while at it do not call myrb_cleanup() if we cannot enable the PCI device at all.

---
 drivers/scsi/myrb.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)
drivers/scsi/myrb.c +7 -4
index 71585528e8db..f460aed2435f 100644
@@ -1239,7 +1239,8 @@ static void myrb_cleanup(struct myrb_hba *cb)
        myrb_unmap(cb);
 
        if (cb->mmio_base) {
-               cb->disable_intr(cb->io_base);
+               if (cb->disable_intr)
+                       cb->disable_intr(cb->io_base);
                iounmap(cb->mmio_base);
        }
        if (cb->irq)
@@ -3414,8 +3415,11 @@ static struct myrb_hba *myrb_detect(struct pci_dev *pdev,
        mutex_init(&cb->dma_mutex);
        cb->pdev = pdev;
 
-       if (pci_enable_device(pdev))
-               goto failure;
+       if (pci_enable_device(pdev)) {
+               dev_err(&pdev->dev, "Failed to enable PCI device\n");
+               scsi_host_put(shost);
+               return NULL;
+       }
 
        if (privdata->hw_init == DAC960_PD_hw_init ||
            privdata->hw_init == DAC960_P_hw_init) {
--
Quoted 2 lines

2.29.2

Thanks for your patch, it works for me :) But after applying the patch, I got another panic:

[    0.957522] myrb 0000:00:05.0: Unknown Initialization Error 49
[    0.957804] myrb 0000:00:05.0: Failed to initialize Controller
[    0.963581] general protection fault, probably for non-canonical
address 0xdffffc0000000075: 0000 [#1]
[    0.963942] RIP: 0010:kobject_put+0x2f/0x1d0
[    0.963942] Call Trace:
[    0.963942]  <TASK>
[    0.963942]  put_device+0x1b/0x30
[    0.963942]  myrb_probe.cold+0x182/0x12d3

The bug occurs when the driver fails at hw_init() in line 3443 and then goes to myrb_cleanup(). The panic occurs at scsi_host_put().

Thanks,
Zheyu Ma