All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] PCI: quirks: Fix out-of-bounds MMIO read in nvme_disable_and_flr()
@ 2026-08-17  9:24 Mohamad Raizudeen
  2026-08-17  9:34 ` sashiko-bot
  0 siblings, 1 reply; 2+ messages in thread
From: Mohamad Raizudeen @ 2026-08-17  9:24 UTC (permalink / raw)
  To: bhelgaas; +Cc: alex, skhan, jkoolstra, linux-pci, linux-kernel,
	Mohamad Raizudeen

In nvme_disable_and_flr(), the PCI bar is mapped using
NVME_REG_CC + sizeof(cfg) which is (0x14 + 4 = 0x18 bytes)

However, the function later reads the controller status from
NVME_REG_CSTS - offset 0x1C, which is outside the mapped 0x18 byte
boundary and it can cause a page fault or kernel panic on architectures
that enforce strict MMIO boundaries.

Fix this by increasing the mapping size to include NVME_REG_CSTS.

Fixes: ffb0863426eb9 ("PCI: Disable Samsung SM961/PM961 NVMe before FLR")
Signed-off-by: Mohamad Raizudeen <raizudeen.kerneldev@gmail.com>
---
 drivers/pci/quirks.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index b09f27f7846f..ed03892cc960 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -4090,7 +4090,7 @@ static int nvme_disable_and_flr(struct pci_dev *dev, bool probe)
 	if (probe)
 		return 0;
 
-	bar = pci_iomap(dev, 0, NVME_REG_CC + sizeof(cfg));
+	bar = pci_iomap(dev, 0, NVME_REG_CSTS + sizeof(cfg));
 	if (!bar)
 		return -ENOTTY;
 
-- 
2.53.0


^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [PATCH] PCI: quirks: Fix out-of-bounds MMIO read in nvme_disable_and_flr()
  2026-08-17  9:24 [PATCH] PCI: quirks: Fix out-of-bounds MMIO read in nvme_disable_and_flr() Mohamad Raizudeen
@ 2026-08-17  9:34 ` sashiko-bot
  0 siblings, 0 replies; 2+ messages in thread
From: sashiko-bot @ 2026-08-17  9:34 UTC (permalink / raw)
  To: Mohamad Raizudeen; +Cc: linux-pci

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] This problem wasn't introduced by this patch, but missing checks for PCI device disconnect (0xFFFFFFFF) in MMIO reads cause a 128-second kernel thread stall.
--

--- Patch [1]: [PATCH] PCI: quirks: Fix out-of-bounds MMIO read in nvme_disable_and_flr() ---
commit df33f68f6dbf217384a1bdde080cb5742ca0e99d
Author: Mohamad Raizudeen <raizudeen.kerneldev@gmail.com>

PCI: quirks: Fix out-of-bounds MMIO read in nvme_disable_and_flr()

This commit fixes an out-of-bounds MMIO read in nvme_disable_and_flr() by
increasing the mapped area size to include the NVME_REG_CSTS register.

> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
> index b09f27f7846fc..ed03892cc960e 100644
> --- a/drivers/pci/quirks.c
> +++ b/drivers/pci/quirks.c
> @@ -4090,7 +4090,7 @@ static int nvme_disable_and_flr(struct pci_dev *dev, bool probe)
>  	if (probe)
>  		return 0;
>  
> -	bar = pci_iomap(dev, 0, NVME_REG_CC + sizeof(cfg));
> +	bar = pci_iomap(dev, 0, NVME_REG_CSTS + sizeof(cfg));
>  	if (!bar)
>  		return -ENOTTY;
>  
> 	pci_read_config_word(dev, PCI_COMMAND, &cmd);
> 	pci_write_config_word(dev, PCI_COMMAND, cmd | PCI_COMMAND_MEMORY);
> 
> 	cfg = readl(bar + NVME_REG_CC);
> 
> 	/* Disable controller if enabled */
> 	if (cfg & NVME_CC_ENABLE) {
> 		u32 cap = readl(bar + NVME_REG_CAP);
> 		unsigned long timeout;

[ ... ]

> 		timeout = ((NVME_CAP_TIMEOUT(cap) + 1) * HZ / 2) + jiffies;
> 
> 		for (;;) {
> 			u32 status = readl(bar + NVME_REG_CSTS);
> 
> 			/* Ready status becomes zero on disable complete */
> 			if (!(status & NVME_CSTS_RDY))
> 				break;
> 
> 			msleep(100);

[Severity: High]
This is a pre-existing issue, but does this function properly handle PCI
device disconnects?

If a PCIe device drops off the bus or encounters a fatal error, MMIO reads
such as readl(bar + NVME_REG_CC) will return 0xFFFFFFFF. Because 0xFFFFFFFF
passes the cfg & NVME_CC_ENABLE check, the code will erroneously enter this
block.

It will then read 0xFFFFFFFF from NVME_REG_CAP, resulting in a maximum
timeout of 128 seconds. The loop continuously reads 0xFFFFFFFF for
NVME_REG_CSTS, which fails the NVME_CSTS_RDY check, causing the kernel
thread to sleep for 100ms in a loop for the full 128 seconds.

Could this cause system management threads like VFIO teardown or PCI hotplug
workers to stall?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260817092448.4395-1-raizudeen.kerneldev@gmail.com?part=1

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2026-08-17  9:34 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-08-17  9:24 [PATCH] PCI: quirks: Fix out-of-bounds MMIO read in nvme_disable_and_flr() Mohamad Raizudeen
2026-08-17  9:34 ` sashiko-bot

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.