linux-pci.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Heiner Kallweit <hkallweit1@gmail.com>
To: Bjorn Helgaas <helgaas@kernel.org>
Cc: Bjorn Helgaas <bhelgaas@google.com>,
	"linux-pci@vger.kernel.org" <linux-pci@vger.kernel.org>,
	Qian Cai <quic_qiancai@quicinc.com>
Subject: Re: [PATCH] PCI/VPD: Fix stack overflow caused by pci_read_vpd_any()
Date: Wed, 13 Oct 2021 21:12:14 +0200	[thread overview]
Message-ID: <ff35d592-af37-2975-0b8f-9a8e2616d38a@gmail.com> (raw)
In-Reply-To: <20211013185353.GA1909717@bhelgaas>

On 13.10.2021 20:53, Bjorn Helgaas wrote:
> On Wed, Oct 13, 2021 at 08:19:59PM +0200, Heiner Kallweit wrote:
>> Recent bug fix 00e1a5d21b4f ("PCI/VPD: Defer VPD sizing until first
>> access") interferes with the original change, resulting in a stack
>> overflow. The following fix has been successfully tested by Qian
>> and myself.
> 
> What does "the original change" refer to?  80484b7f8db1?  I guess the
> stack overflow is an unintended recursion?  Is there a URL to Qian's
> bug report with more details that we can include here?
> 

1. yes
2. yes
3. https://lore.kernel.org/netdev/e89087c5-c495-c5ca-feb1-54cf3a8775c5@quicinc.com/

>> Fixes: 80484b7f8db1 ("PCI/VPD: Use pci_read_vpd_any() in pci_vpd_size()")
>> Reported-by: Qian Cai <quic_qiancai@quicinc.com>
>> Tested-by: Qian Cai <quic_qiancai@quicinc.com>
>> Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
>> ---
>>  drivers/pci/vpd.c | 18 +++++++++++-------
>>  1 file changed, 11 insertions(+), 7 deletions(-)
>>
>> diff --git a/drivers/pci/vpd.c b/drivers/pci/vpd.c
>> index 5108bbd20..a4fc4d069 100644
>> --- a/drivers/pci/vpd.c
>> +++ b/drivers/pci/vpd.c
>> @@ -96,14 +96,14 @@ static size_t pci_vpd_size(struct pci_dev *dev)
>>  	return off ?: PCI_VPD_SZ_INVALID;
>>  }
>>  
>> -static bool pci_vpd_available(struct pci_dev *dev)
>> +static bool pci_vpd_available(struct pci_dev *dev, bool check_size)
>>  {
>>  	struct pci_vpd *vpd = &dev->vpd;
>>  
>>  	if (!vpd->cap)
>>  		return false;
>>  
>> -	if (vpd->len == 0) {
>> +	if (vpd->len == 0 && check_size) {
>>  		vpd->len = pci_vpd_size(dev);
>>  		if (vpd->len == PCI_VPD_SZ_INVALID) {
>>  			vpd->cap = 0;
>> @@ -156,17 +156,19 @@ static ssize_t pci_vpd_read(struct pci_dev *dev, loff_t pos, size_t count,
>>  			    void *arg, bool check_size)
>>  {
>>  	struct pci_vpd *vpd = &dev->vpd;
>> -	unsigned int max_len = check_size ? vpd->len : PCI_VPD_MAX_SIZE;
>> +	unsigned int max_len;
>>  	int ret = 0;
>>  	loff_t end = pos + count;
>>  	u8 *buf = arg;
>>  
>> -	if (!pci_vpd_available(dev))
>> +	if (!pci_vpd_available(dev, check_size))
>>  		return -ENODEV;
>>  
>>  	if (pos < 0)
>>  		return -EINVAL;
>>  
>> +	max_len = check_size ? vpd->len : PCI_VPD_MAX_SIZE;
>> +
>>  	if (pos >= max_len)
>>  		return 0;
>>  
>> @@ -218,17 +220,19 @@ static ssize_t pci_vpd_write(struct pci_dev *dev, loff_t pos, size_t count,
>>  			     const void *arg, bool check_size)
>>  {
>>  	struct pci_vpd *vpd = &dev->vpd;
>> -	unsigned int max_len = check_size ? vpd->len : PCI_VPD_MAX_SIZE;
>> +	unsigned int max_len;
>>  	const u8 *buf = arg;
>>  	loff_t end = pos + count;
>>  	int ret = 0;
>>  
>> -	if (!pci_vpd_available(dev))
>> +	if (!pci_vpd_available(dev, check_size))
>>  		return -ENODEV;
>>  
>>  	if (pos < 0 || (pos & 3) || (count & 3))
>>  		return -EINVAL;
>>  
>> +	max_len = check_size ? vpd->len : PCI_VPD_MAX_SIZE;
>> +
>>  	if (end > max_len)
>>  		return -EINVAL;
>>  
>> @@ -312,7 +316,7 @@ void *pci_vpd_alloc(struct pci_dev *dev, unsigned int *size)
>>  	void *buf;
>>  	int cnt;
>>  
>> -	if (!pci_vpd_available(dev))
>> +	if (!pci_vpd_available(dev, true))
>>  		return ERR_PTR(-ENODEV);
>>  
>>  	len = dev->vpd.len;
>> -- 
>> 2.33.0
>>


  reply	other threads:[~2021-10-13 19:12 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-13 18:19 [PATCH] PCI/VPD: Fix stack overflow caused by pci_read_vpd_any() Heiner Kallweit
2021-10-13 18:53 ` Bjorn Helgaas
2021-10-13 19:12   ` Heiner Kallweit [this message]
2021-10-25 20:57 ` Bjorn Helgaas
2021-10-26  0:22   ` Bjorn Helgaas

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=ff35d592-af37-2975-0b8f-9a8e2616d38a@gmail.com \
    --to=hkallweit1@gmail.com \
    --cc=bhelgaas@google.com \
    --cc=helgaas@kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=quic_qiancai@quicinc.com \
    /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).