Netdev Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Heiner Kallweit <hkallweit1@gmail.com>
To: Qian Cai <quic_qiancai@quicinc.com>,
	Bjorn Helgaas <bhelgaas@google.com>,
	Jakub Kicinski <kuba@kernel.org>,
	David Miller <davem@davemloft.net>,
	Raju Rangoju <rajur@chelsio.com>
Cc: "linux-pci@vger.kernel.org" <linux-pci@vger.kernel.org>,
	"netdev@vger.kernel.org" <netdev@vger.kernel.org>
Subject: Re: [PATCH 1/5] PCI/VPD: Add pci_read/write_vpd_any()
Date: Tue, 12 Oct 2021 22:26:24 +0200	[thread overview]
Message-ID: <ca805454-6ec5-303b-d39f-d505cad6b338@gmail.com> (raw)
In-Reply-To: <e89087c5-c495-c5ca-feb1-54cf3a8775c5@quicinc.com>

On 12.10.2021 20:59, Qian Cai wrote:
> 
> 
> On 9/10/2021 2:22 AM, Heiner Kallweit wrote:
>> In certain cases we need a variant of pci_read_vpd()/pci_write_vpd() that
>> does not check against dev->vpd.len. Such cases are:
>> - reading VPD if dev->vpd.len isn't set yet (in pci_vpd_size())
>> - devices that map non-VPD information to arbitrary places in VPD address
>>   space (example: Chelsio T3 EEPROM write-protect flag)
>> Therefore add function variants that check against PCI_VPD_MAX_SIZE only.
>>
>> Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com
> Reverting this series fixed a hang or stack overflow while reading things like,
> 
> /sys/devices/pci0000:00/0000:00:00.0/0000:01:00.0/vpd
> 
> [  125.797082] Insufficient stack space to handle exception!
> [  125.797091] ESR: 0x96000047 -- DABT (current EL)
> [  125.797095] FAR: 0xffff80002433ffc0
> [  125.797096] Task stack:     [0xffff800024340000..0xffff800024350000]
> [  125.797099] IRQ stack:      [0xffff8000101c0000..0xffff8000101d0000]
> [  125.797102] Overflow stack: [0xffff009b675b02b0..0xffff009b675b12b0]
> [  125.797106] CPU: 14 PID: 1550 Comm: lsbug Not tainted 5.15.0-rc5-next-20211012 #143
> [  125.797110] Hardware name: MiTAC RAPTOR EV-883832-X3-0001/RAPTOR, BIOS 1.6 06/28/2020
> [  125.797114] pstate: 10000005 (nzcV daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> [  125.797118] pc : pci_vpd_size+0xc/0x1f8
> [  125.797128] lr : pci_vpd_read+0x2ec/0x420
> [  125.797132] sp : ffff800024340060
> [  125.797133] x29: ffff800024340060 x28: ffff00001a54cbcc x27: 0000000000000000
> [  125.797142] x26: ffff800024340210 x25: 0000000000000004 x24: 1fffe000034a9979
> [  125.797148] x23: ffff00001a54cbc8 x22: ffff00001a54cb38 x21: 0000000000008000
> [  125.797153] x20: 1fffe000034a9979 x19: ffff00001a54c000 x18: 0000000000000000
> [  125.797158] x17: 0000000000000000 x16: 0000000000000000 x15: dfff800000000000
> [  125.797163] x14: ffff800019ab0560 x13: 1fffe00110f9272f x12: ffff60010e945be1
> [  125.797168] x11: 1fffe0010e945be0 x10: 1ffff00004868022 x9 : ffff800010d1a38c
> [  125.797174] x8 : ffff700004868022 x7 : dfff800000000000 x6 : 0000000000000000
> [  125.797179] x5 : ffff000887c93540 x4 : 0000000000000000 x3 : ffff800024340210
> [  125.797184] x2 : 0000000000000001 x1 : 0000000000000003 x0 : ffff00001a54c000
> [  125.797190] Kernel panic - not syncing: kernel stack overflow
> [  125.797193] CPU: 14 PID: 1550 Comm: lsbug Not tainted 5.15.0-rc5-next-20211012 #143
> [  125.797197] Hardware name: MiTAC RAPTOR EV-883832-X3-0001/RAPTOR, BIOS 1.6 06/28/2020
> [  125.797199] Call trace:
> [  125.797201]  dump_backtrace+0x0/0x3b8
> [  125.797208]  show_stack+0x20/0x30
> [  125.797212]  dump_stack_lvl+0x8c/0xb8
> [  125.797216]  dump_stack+0x1c/0x38
> [  125.797219]  panic+0x2b0/0x538
> [  125.797224]  add_taint+0x0/0xe8
> [  125.797229]  panic_bad_stack+0x1e4/0x230
> [  125.797233]  handle_bad_stack+0x38/0x50
> [  125.797237]  __bad_stack+0x88/0x8c
> [  125.797241]  pci_vpd_size+0xc/0x1f8
> [  125.797244]  __pci_read_vpd+0x114/0x158
> [  125.797247]  pci_vpd_size+0xa0/0x1f8
> [  125.797251]  pci_vpd_read+0x2ec/0x420
> [  125.797254]  __pci_read_vpd+0x114/0x158
> [  125.797258]  pci_vpd_size+0xa0/0x1f8
> [  125.797261]  pci_vpd_read+0x2ec/0x420
> ...
> [  125.798534]  __pci_read_vpd+0x114/0x158
> [  125.798538]  pci_vpd_size+0xa0/0x1f8
> [  125.798541]  pci_vpd_read+0x2ec/0x420
> [  125.798545]  __pci_read_vpd+0x114/0x158
> __pci_read_vpd at /usr/src/linux-next/drivers/pci/vpd.c:398
> [  125.798548]  vpd_read+0x28/0x38
> vpd_read at /usr/src/linux-next/drivers/pci/vpd.c:276
> [  125.798551]  sysfs_kf_bin_read+0x120/0x218
> [  125.798556]  kernfs_fop_read_iter+0x244/0x4a8
> [  125.798559]  new_sync_read+0x2bc/0x4e8
> [  125.798564]  vfs_read+0x18c/0x390
> [  125.798567]  ksys_read+0xf8/0x1e0
> [  125.798570]  __arm64_sys_read+0x74/0xa8
> [  125.798574]  invoke_syscall.constprop.0+0xdc/0x1d8
> [  125.798578]  do_el0_svc+0xe4/0x298
> [  125.798582]  el0_svc+0x64/0x130
> [  125.798586]  el0t_64_sync_handler+0xb0/0xb8
> [  125.798590]  el0t_64_sync+0x180/0x184
> [  125.798598] ------------[ cut here ]------------
> [  125.798600] WARNING: CPU: -32 PID: 1550 at include/linux/cpumask.h:108 smp_send_stop+0x4a4/0x5e8
> [  125.798607] Modules linked in: loop cppc_cpufreq efivarfs ip_tables x_tables ext4 mbcache jbd2 dm_mod igb i2c_algo_bit nvme mlx5_core i2c_core nvme_core firmware_class
> [  125.798632] CPU: 791961908 PID: 1550 Comm: lsbug Not tainted 5.15.0-rc5-next-20211012 #143
> [  125.798637] Hardware name: MiTAC RAPTOR EV-883832-X3-0001/RAPTOR, BIOS 1.6 06/28/2020
> [  125.798639] pstate: a00003c5 (NzCv DAIF -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> [  125.798643] pc : smp_send_stop+0x4a4/0x5e8
> [  125.798647] lr : panic+0x2b8/0x538
> [  125.798650] sp : ffff009b675b0c70
> [  125.798652] x29: ffff009b675b0c70 x28: ffff000887c92ec0 x27: 0000000000000000
> [  125.798658] x26: 0000000000000025 x25: ffff809b55bf0000 x24: ffff800011eeb4d0
> [  125.798663] x23: ffff800011426680 x22: ffff800019393000 x21: ffff800019dfa000
> [  125.798668] x20: 00000000ffffffe0 x19: ffff8000119c0000 x18: 0000000000000000
> [  125.798673] x17: 0000000000000000 x16: 0000000000000002 x15: 0000000000000000
> [  125.798678] x14: 0000000000000000 x13: 000000000000000f x12: ffff7000023ef669
> [  125.798683] x11: 1ffff000023ef668 x10: ffff7000023ef668 x9 : ffff80001133f2cc
> [  125.798688] x8 : 0000000000000003 x7 : 0000000000000001 x6 : ffff800011f7b340
> [  125.798693] x5 : 1fffe0136ceb619e x4 : 0000000041b58ab3 x3 : 1fffe0136ceb6000
> [  125.798698] x2 : 1ffff000023dd69a x1 : 0000000000000000 x0 : 0000000000000020
> [  125.798704] Call trace:
> [  125.798705]  smp_send_stop+0x4a4/0x5e8
> [  125.798709]  panic+0x2b8/0x538
> [  125.798713]  add_taint+0x0/0xe8
> [  125.798717]  panic_bad_stack+0x1e4/0x230
> [  125.798720]  handle_bad_stack+0x38/0x50
> [  125.798724]  __bad_stack+0x88/0x8c
> [  125.798727]  pci_vpd_size+0xc/0x1f8
> [  125.798731]  __pci_read_vpd+0x114/0x158
> [  125.798734]  pci_vpd_size+0xa0/0x1f8
> [  125.798738]  pci_vpd_read+0x2ec/0x420
> [  125.798741]  __pci_read_vpd+0x114/0x158
> [  125.798744]  pci_vpd_size+0xa0/0x1f8
> [  125.798748]  pci_vpd_read+0x2ec/0x420  
> 

Thanks for the report! I could reproduce the issue, the following fixes
it for me. Could you please test whether it fixes the issue for you as well?
Thank you.


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:[~2021-10-12 20:26 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-10  6:20 [PATCH 0/5] PCI/VPD: Add and use pci_read/write_vpd_any() Heiner Kallweit
2021-09-10  6:22 ` [PATCH 1/5] PCI/VPD: Add pci_read/write_vpd_any() Heiner Kallweit
2021-10-12 18:59   ` Qian Cai
2021-10-12 20:26     ` Heiner Kallweit [this message]
2021-10-13 14:22       ` Qian Cai
2021-10-13 18:08         ` Heiner Kallweit
2021-09-10  6:22 ` [PATCH 2/5] PCI/VPD: Use pci_read_vpd_any() in pci_vpd_size() Heiner Kallweit
2021-10-27 11:01   ` Jon Hunter
2021-10-27 11:32     ` Heiner Kallweit
2021-10-28  8:45       ` Jon Hunter
2021-09-10  6:24 ` [PATCH 3/5] cxgb3: Remove t3_seeprom_read and use VPD API Heiner Kallweit
2021-09-10  6:25 ` [PATCH 4/5] cxgb3: Use VPD API in t3_seeprom_wp() Heiner Kallweit
2021-09-10  6:27 ` [PATCH 5/5] cxgb3: Remove seeprom_write and use VPD API Heiner Kallweit
2021-10-08 22:53 ` [PATCH 0/5] PCI/VPD: Add and use pci_read/write_vpd_any() Bjorn Helgaas
2021-10-08 23:42   ` Jakub Kicinski
2021-10-11 20:43     ` 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=ca805454-6ec5-303b-d39f-d505cad6b338@gmail.com \
    --to=hkallweit1@gmail.com \
    --cc=bhelgaas@google.com \
    --cc=davem@davemloft.net \
    --cc=kuba@kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=quic_qiancai@quicinc.com \
    --cc=rajur@chelsio.com \
    --subject='Re: [PATCH 1/5] PCI/VPD: Add pci_read/write_vpd_any()' \
    /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

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).