| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <95db9ba2-ffbb-ca92-6a70-1ee401920eed@amd.com>
Date: Tue, 21 Jul 2020 12:15:13 -0500
From: Tom Lendacky <thomas.lendacky@....com>
To: Vaibhav Gupta <vaibhavgupta40@...il.com>
Cc: Bjorn Helgaas <helgaas@...nel.org>,
Bjorn Helgaas <bhelgaas@...gle.com>,
Bjorn Helgaas <bjorn@...gaas.com>,
Vaibhav Gupta <vaibhav.varodek@...il.com>,
"David S. Miller" <davem@...emloft.net>,
Herbert Xu <herbert@...dor.apana.org.au>,
linux-kernel@...r.kernel.org,
linux-kernel-mentees@...ts.linuxfoundation.org,
Shuah Khan <skhan@...uxfoundation.org>,
linux-crypto@...r.kernel.org
Subject: Re: [PATCH v1] crypto: ccp: sp-pci: use generic power management
On 7/21/20 11:30 AM, Vaibhav Gupta wrote:
> On Tue, Jul 21, 2020 at 10:19:33AM -0500, Tom Lendacky wrote:
>> On 7/21/20 7:31 AM, Vaibhav Gupta wrote:
>>> Drivers using legacy power management .suspen()/.resume() callbacks
>>> have to manage PCI states and device's PM states themselves. They also
>>> need to take care of standard configuration registers.
>>>
>>> Switch to generic power management framework using a single
>>> "struct dev_pm_ops" variable to take the unnecessary load from the driver.
>>> This also avoids the need for the driver to directly call most of the PCI
>>> helper functions and device power state control functions as through
>>> the generic framework, PCI Core takes care of the necessary operations,
>>> and drivers are required to do only device-specific jobs.
>>>
>>> Signed-off-by: Vaibhav Gupta <vaibhavgupta40@...il.com>
>>> ---
>>> drivers/crypto/ccp/ccp-dev.c | 8 +++-----
>>> drivers/crypto/ccp/sp-dev.c | 6 ++----
>>> drivers/crypto/ccp/sp-dev.h | 6 +++---
>>> drivers/crypto/ccp/sp-pci.c | 17 ++++++-----------
>>> drivers/crypto/ccp/sp-platform.c | 2 +-
>>> 5 files changed, 15 insertions(+), 24 deletions(-)
>>>
>>> diff --git a/drivers/crypto/ccp/ccp-dev.c b/drivers/crypto/ccp/ccp-dev.c
>>> index 19ac509ed76e..8ae26d3cffff 100644
>>> --- a/drivers/crypto/ccp/ccp-dev.c
>>> +++ b/drivers/crypto/ccp/ccp-dev.c
>>> @@ -531,8 +531,7 @@ int ccp_trng_read(struct hwrng *rng, void *data, size_t max, bool wait)
>>> return len;
>>> }
>>>
>>> -#ifdef CONFIG_PM
>>> -bool ccp_queues_suspended(struct ccp_device *ccp)
>>> +bool __maybe_unused ccp_queues_suspended(struct ccp_device *ccp)
>>
>> These aren't static functions, so is the __maybe_unused necessary?
>>
>> (Same comment on the other non-static functions changed below)
>>
>>> {
>>> unsigned int suspended = 0;
>>> unsigned long flags;
>>> @@ -549,7 +548,7 @@ bool ccp_queues_suspended(struct ccp_device *ccp)
>>> return ccp->cmd_q_count == suspended;
>>> }
>>>
>>> -int ccp_dev_suspend(struct sp_device *sp, pm_message_t state)
>>> +int __maybe_unused ccp_dev_suspend(struct sp_device *sp)
>>> {
>>> struct ccp_device *ccp = sp->ccp_data;
>>> unsigned long flags;
>>> @@ -577,7 +576,7 @@ int ccp_dev_suspend(struct sp_device *sp, pm_message_t state)
>>> return 0;
>>> }
>>>
>>> -int ccp_dev_resume(struct sp_device *sp)
>>> +int __maybe_unused ccp_dev_resume(struct sp_device *sp)
>>> {
>>> struct ccp_device *ccp = sp->ccp_data;
>>> unsigned long flags;
>>> @@ -601,7 +600,6 @@ int ccp_dev_resume(struct sp_device *sp)
>>>
>>> return 0;
>>> }
>>> -#endif
>>>
>>> int ccp_dev_init(struct sp_device *sp)
>>> {
>>> diff --git a/drivers/crypto/ccp/sp-dev.c b/drivers/crypto/ccp/sp-dev.c
>>> index ce42675d3274..6284a15e5047 100644
>>> --- a/drivers/crypto/ccp/sp-dev.c
>>> +++ b/drivers/crypto/ccp/sp-dev.c
>>> @@ -211,13 +211,12 @@ void sp_destroy(struct sp_device *sp)
>>> sp_del_device(sp);
>>> }
>>>
>>> -#ifdef CONFIG_PM
>>> -int sp_suspend(struct sp_device *sp, pm_message_t state)
>>> +int sp_suspend(struct sp_device *sp)
>>> {
>>> int ret;
>>>
>>> if (sp->dev_vdata->ccp_vdata) {
>>> - ret = ccp_dev_suspend(sp, state);
>>> + ret = ccp_dev_suspend(sp);
>>> if (ret)
>>> return ret;
>>> }
>>> @@ -237,7 +236,6 @@ int sp_resume(struct sp_device *sp)
>>>
>>> return 0;
>>> }
>>> -#endif
>>>
>>> struct sp_device *sp_get_psp_master_device(void)
>>> {
>>> diff --git a/drivers/crypto/ccp/sp-dev.h b/drivers/crypto/ccp/sp-dev.h
>>> index f913f1494af9..0218d0670eee 100644
>>> --- a/drivers/crypto/ccp/sp-dev.h
>>> +++ b/drivers/crypto/ccp/sp-dev.h
>>> @@ -119,7 +119,7 @@ int sp_init(struct sp_device *sp);
>>> void sp_destroy(struct sp_device *sp);
>>> struct sp_device *sp_get_master(void);
>>>
>>> -int sp_suspend(struct sp_device *sp, pm_message_t state);
>>> +int sp_suspend(struct sp_device *sp);
>>> int sp_resume(struct sp_device *sp);
>>> int sp_request_ccp_irq(struct sp_device *sp, irq_handler_t handler,
>>> const char *name, void *data);
>>> @@ -134,7 +134,7 @@ struct sp_device *sp_get_psp_master_device(void);
>>> int ccp_dev_init(struct sp_device *sp);
>>> void ccp_dev_destroy(struct sp_device *sp);
>>>
>>> -int ccp_dev_suspend(struct sp_device *sp, pm_message_t state);
>>> +int ccp_dev_suspend(struct sp_device *sp);
>>> int ccp_dev_resume(struct sp_device *sp);
>>>
>>> #else /* !CONFIG_CRYPTO_DEV_SP_CCP */
>>> @@ -145,7 +145,7 @@ static inline int ccp_dev_init(struct sp_device *sp)
>>> }
>>> static inline void ccp_dev_destroy(struct sp_device *sp) { }
>>>
>>> -static inline int ccp_dev_suspend(struct sp_device *sp, pm_message_t state)
>>> +static inline int ccp_dev_suspend(struct sp_device *sp)
>>> {
>>> return 0;
>>> }
>>> diff --git a/drivers/crypto/ccp/sp-pci.c b/drivers/crypto/ccp/sp-pci.c
>>> index cb6cb47053f4..f471dbaef1fb 100644
>>> --- a/drivers/crypto/ccp/sp-pci.c
>>> +++ b/drivers/crypto/ccp/sp-pci.c
>>> @@ -252,23 +252,19 @@ static void sp_pci_remove(struct pci_dev *pdev)
>>> sp_free_irqs(sp);
>>> }
>>>
>>> -#ifdef CONFIG_PM
>>> -static int sp_pci_suspend(struct pci_dev *pdev, pm_message_t state)
>>> +static int __maybe_unused sp_pci_suspend(struct device *dev)
>>> {
>>> - struct device *dev = &pdev->dev;
>>> struct sp_device *sp = dev_get_drvdata(dev);
>>>
>>> - return sp_suspend(sp, state);
>>> + return sp_suspend(sp);
>>> }
>>>
>>> -static int sp_pci_resume(struct pci_dev *pdev)
>>> +static int __maybe_unused sp_pci_resume(struct device *dev)
>>> {
>>> - struct device *dev = &pdev->dev;
>>> struct sp_device *sp = dev_get_drvdata(dev);
>>>
>>> return sp_resume(sp);
>>> }
>>> -#endif
>>>
>>> #ifdef CONFIG_CRYPTO_DEV_SP_PSP
>>> static const struct sev_vdata sevv1 = {
>>> @@ -365,15 +361,14 @@ static const struct pci_device_id sp_pci_table[] = {
>>> };
>>> MODULE_DEVICE_TABLE(pci, sp_pci_table);
>>>
>>> +static SIMPLE_DEV_PM_OPS(sp_pci_pm_ops, sp_pci_suspend, sp_pci_resume);
>>> +
>>> static struct pci_driver sp_pci_driver = {
>>> .name = "ccp",
>>> .id_table = sp_pci_table,
>>> .probe = sp_pci_probe,
>>> .remove = sp_pci_remove,
>>> -#ifdef CONFIG_PM
>>> - .suspend = sp_pci_suspend,
>>> - .resume = sp_pci_resume,
>>> -#endif
>>> + .driver.pm = &sp_pci_pm_ops,
>>> };
>>>
>>> int sp_pci_init(void)
>>> diff --git a/drivers/crypto/ccp/sp-platform.c b/drivers/crypto/ccp/sp-platform.c
>>> index 831aac1393a2..9dba52fbee99 100644
>>> --- a/drivers/crypto/ccp/sp-platform.c
>>> +++ b/drivers/crypto/ccp/sp-platform.c
>>
>> This file use the same "#ifdef CONFIG_PM" to define the suspend and resume
>> functions (see sp_platform_driver variable). Doesn't that need to be
>> updated similar to sp-pci.c? Not sure how this compile tested successfully
>> unless your kernel config didn't have CONFIG_PM defined?
> I am not sure but my .config file has "CONFIG_PM=y"
Ok, my miss on that, you didn't update the sp_platform_suspend signature,
so no issue there.
>
> The motive is update PM of sp-pci. Months ago, we decided to remove
> "#ifdef CONFIG_PM" container and mark the callbacks as __maybe_unused.
Is this being done only for struct pci_driver structures then? Or will
there be a follow on effort for struct platform_driver structures?
I can see the need for the __maybe_unused on the sp_pci_suspend() and
sp_pci_resume() functions since those are static functions that may cause
warnings depending on whether CONFIG_PM_SLEEP is defined or not.
The ccp_dev_suspend() and ccp_dev_resume() functions, though, are external
functions that I would think shouldn't require the __may_unused attribute
because the compiler wouldn't know if they are invoked or not within that
file (similar to how the sp_suspend() and sp_resume() weren't modified to
include the __maybe_unused attribute). Can you try a compile test without
changing those functions and without CONFIG_PM=y and see if you run into
issues?
Thanks,
Tom
> Hence, I did similar changes to all files on which sp-pci is dependent.
>
> This caused change in function defintion and declaration of sp_suspend().
>
> sp-pci is not depended upon sp-platform. sp-platform was modified only because
> it also invoked sp_suspend() which got modified.
>
> Thus, I didn't made any other changes to sp-platofrm.
>
> Thanks
> Vaibhav Gupta
>>
>> Thanks,
>> Tom
>>
>>> @@ -207,7 +207,7 @@ static int sp_platform_suspend(struct platform_device *pdev,
>>> struct device *dev = &pdev->dev;
>>> struct sp_device *sp = dev_get_drvdata(dev);
>>>
>>> - return sp_suspend(sp, state);
>>> + return sp_suspend(sp);
>>> }
>>>
>>> static int sp_platform_resume(struct platform_device *pdev)
>>>
Powered by blists - more mailing lists