linux-kernel - Re: [PATCH RFC v2 02/18] irq/dev-msi: Add support for a new DEV

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

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 22 Jul 2020 09:50:47 -0700
From:   "Dey, Megha" <megha.dey@...el.com>
To:     Jason Gunthorpe <jgg@...lanox.com>,
        Dave Jiang <dave.jiang@...el.com>
CC:     <vkoul@...nel.org>, <maz@...nel.org>, <bhelgaas@...gle.com>,
        <rafael@...nel.org>, <gregkh@...uxfoundation.org>,
        <tglx@...utronix.de>, <hpa@...or.com>,
        <alex.williamson@...hat.com>, <jacob.jun.pan@...el.com>,
        <ashok.raj@...el.com>, <yi.l.liu@...el.com>, <baolu.lu@...el.com>,
        <kevin.tian@...el.com>, <sanjay.k.kumar@...el.com>,
        <tony.luck@...el.com>, <jing.lin@...el.com>,
        <dan.j.williams@...el.com>, <kwankhede@...dia.com>,
        <eric.auger@...hat.com>, <parav@...lanox.com>,
        <dave.hansen@...el.com>, <netanelg@...lanox.com>,
        <shahafs@...lanox.com>, <yan.y.zhao@...ux.intel.com>,
        <pbonzini@...hat.com>, <samuel.ortiz@...el.com>,
        <mona.hossain@...el.com>, <dmaengine@...r.kernel.org>,
        <linux-kernel@...r.kernel.org>, <x86@...nel.org>,
        <linux-pci@...r.kernel.org>, <kvm@...r.kernel.org>
Subject: Re: [PATCH RFC v2 02/18] irq/dev-msi: Add support for a new DEV_MSI
 irq domain

Hi Jason,

On 7/21/2020 9:13 AM, Jason Gunthorpe wrote:
> On Tue, Jul 21, 2020 at 09:02:28AM -0700, Dave Jiang wrote:
>> From: Megha Dey <megha.dey@...el.com>
>>
>> Add support for the creation of a new DEV_MSI irq domain. It creates a
>> new irq chip associated with the DEV_MSI domain and adds the necessary
>> domain operations to it.
>>
>> Add a new config option DEV_MSI which must be enabled by any
>> driver that wants to support device-specific message-signaled-interrupts
>> outside of PCI-MSI(-X).
>>
>> Lastly, add device specific mask/unmask callbacks in addition to a write
>> function to the platform_msi_ops.
>>
>> Reviewed-by: Dan Williams <dan.j.williams@...el.com>
>> Signed-off-by: Megha Dey <megha.dey@...el.com>
>> Signed-off-by: Dave Jiang <dave.jiang@...el.com>
>>   arch/x86/include/asm/hw_irq.h |    5 ++
>>   drivers/base/Kconfig          |    7 +++
>>   drivers/base/Makefile         |    1
>>   drivers/base/dev-msi.c        |   95 +++++++++++++++++++++++++++++++++++++++++
>>   drivers/base/platform-msi.c   |   45 +++++++++++++------
>>   drivers/base/platform-msi.h   |   23 ++++++++++
>>   include/linux/msi.h           |    8 +++
>>   7 files changed, 168 insertions(+), 16 deletions(-)
>>   create mode 100644 drivers/base/dev-msi.c
>>   create mode 100644 drivers/base/platform-msi.h
>>
>> diff --git a/arch/x86/include/asm/hw_irq.h b/arch/x86/include/asm/hw_irq.h
>> index 74c12437401e..8ecd7570589d 100644
>> +++ b/arch/x86/include/asm/hw_irq.h
>> @@ -61,6 +61,11 @@ struct irq_alloc_info {
>>   			irq_hw_number_t	msi_hwirq;
>>   		};
>>   #endif
>> +#ifdef CONFIG_DEV_MSI
>> +		struct {
>> +			irq_hw_number_t hwirq;
>> +		};
>> +#endif
> 
> Why is this in this patch? I didn't see an obvious place where it is
> used?

Since I have introduced the DEV-MSI domain and related ops, this is 
required in the dev_msi_set_hwirq and dev_msi_set_desc in this patch.

>>   
>> +static void __platform_msi_desc_mask_unmask_irq(struct msi_desc *desc, u32 mask)
>> +{
>> +	const struct platform_msi_ops *ops;
>> +
>> +	ops = desc->platform.msi_priv_data->ops;
>> +	if (!ops)
>> +		return;
>> +
>> +	if (mask) {
>> +		if (ops->irq_mask)
>> +			ops->irq_mask(desc);
>> +	} else {
>> +		if (ops->irq_unmask)
>> +			ops->irq_unmask(desc);
>> +	}
>> +}
>> +
>> +void platform_msi_mask_irq(struct irq_data *data)
>> +{
>> +	__platform_msi_desc_mask_unmask_irq(irq_data_get_msi_desc(data), 1);
>> +}
>> +
>> +void platform_msi_unmask_irq(struct irq_data *data)
>> +{
>> +	__platform_msi_desc_mask_unmask_irq(irq_data_get_msi_desc(data), 0);
>> +}
> 
> This is a bit convoluted, just call the op directly:
> 
> void platform_msi_unmask_irq(struct irq_data *data)
> {
> 	const struct platform_msi_ops *ops = desc->platform.msi_priv_data->ops;
> 
> 	if (ops->irq_unmask)
> 		ops->irq_unmask(desc);
> }
>

Sure, I will update this.

>> diff --git a/include/linux/msi.h b/include/linux/msi.h
>> index 7f6a8eb51aca..1da97f905720 100644
>> +++ b/include/linux/msi.h
>> @@ -323,9 +323,13 @@ enum {
>>   
>>   /*
>>    * platform_msi_ops - Callbacks for platform MSI ops
>> + * @irq_mask:   mask an interrupt source
>> + * @irq_unmask: unmask an interrupt source
>>    * @write_msg:	write message content
>>    */
>>   struct platform_msi_ops {
>> +	unsigned int            (*irq_mask)(struct msi_desc *desc);
>> +	unsigned int            (*irq_unmask)(struct msi_desc *desc);
> 
> Why do these functions return things if the only call site throws it
> away?

Hmmm, fair enough, I will change it to void.

> 
> Jason
>