netdev - Re: [PATCH net-next 1/8] net: ipa: maintain bitmap of suspend-enabled endpoints

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <b675810d-d06b-454e-9726-c987603453e7@linaro.org>
Date: Thu, 18 Apr 2024 19:46:38 -0500
From: Alex Elder <elder@...aro.org>
To: Bryan O'Donoghue <bryan.odonoghue@...aro.org>, davem@...emloft.net,
 edumazet@...gle.com, kuba@...nel.org, pabeni@...hat.com
Cc: mka@...omium.org, andersson@...nel.org, quic_cpratapa@...cinc.com,
 quic_avuyyuru@...cinc.com, quic_jponduru@...cinc.com,
 quic_subashab@...cinc.com, elder@...nel.org, netdev@...r.kernel.org,
 linux-arm-msm@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH net-next 1/8] net: ipa: maintain bitmap of suspend-enabled
 endpoints

On 4/18/24 6:52 PM, Bryan O'Donoghue wrote:
> On 18/04/2024 21:47, Alex Elder wrote:
>> Keep track of which endpoints have the SUSPEND IPA interrupt enabled
>> in a variable-length bitmap.  This will be used in the next patch to
>> allow the SUSPEND interrupt type to be disabled except when needed.
>>
>> Signed-off-by: Alex Elder <elder@...aro.org>
>> ---
>>   drivers/net/ipa/ipa_interrupt.c | 19 +++++++++++++++++--
>>   1 file changed, 17 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/net/ipa/ipa_interrupt.c 
>> b/drivers/net/ipa/ipa_interrupt.c
>> index c44ec05f71e6f..0e8d4e43275ea 100644
>> --- a/drivers/net/ipa/ipa_interrupt.c
>> +++ b/drivers/net/ipa/ipa_interrupt.c
>> @@ -37,11 +37,13 @@
>>    * @ipa:        IPA pointer
>>    * @irq:        Linux IRQ number used for IPA interrupts
>>    * @enabled:        Mask indicating which interrupts are enabled
>> + * @suspend_enabled:    Bitmap of endpoints with the SUSPEND 
>> interrupt enabled
>>    */
>>   struct ipa_interrupt {
>>       struct ipa *ipa;
>>       u32 irq;
>>       u32 enabled;
>> +    unsigned long *suspend_enabled;
>>   };
>>   /* Clear the suspend interrupt for all endpoints that signaled it */
>> @@ -211,6 +213,7 @@ static void ipa_interrupt_suspend_control(struct 
>> ipa_interrupt *interrupt,
>>           val |= mask;
>>       else
>>           val &= ~mask;
>> +    __change_bit(endpoint_id, interrupt->suspend_enabled);
>>       iowrite32(val, ipa->reg_virt + offset);
>>   }
>> @@ -246,7 +249,16 @@ int ipa_interrupt_config(struct ipa *ipa)
>>       interrupt->ipa = ipa;
>> -    /* Disable all IPA interrupt types */
>> +    /* Initially all IPA interrupt types are disabled */
>> +    interrupt->enabled = 0;
>> +    interrupt->suspend_enabled = bitmap_zalloc(ipa->endpoint_count,
>> +                           GFP_KERNEL);
> 
> why not use devm_bitmap_zalloc() instead and skip managing the cleanup ?

I don't use the devm_*() variants in the IPA driver.

I know I can, but if I'm make the switch I want to
do it everywhere.  Not now.

Thanks for the review.

					-Alex

>> +    if (!interrupt->suspend_enabled) {
>> +        ret = -ENOMEM;
>> +        goto err_kfree;
>> +    }
>> +
>> +    /* Disable IPA interrupt types */
>>       reg = ipa_reg(ipa, IPA_IRQ_EN);
>>       iowrite32(0, ipa->reg_virt + reg_offset(reg));
>> @@ -254,7 +266,7 @@ int ipa_interrupt_config(struct ipa *ipa)
>>                      "ipa", interrupt);
>>       if (ret) {
>>           dev_err(dev, "error %d requesting \"ipa\" IRQ\n", ret);
>> -        goto err_kfree;
>> +        goto err_free_bitmap;
>>       }
>>       ret = dev_pm_set_wake_irq(dev, irq);
>> @@ -270,6 +282,8 @@ int ipa_interrupt_config(struct ipa *ipa)
>>   err_free_irq:
>>       free_irq(interrupt->irq, interrupt);
>> +err_free_bitmap:
>> +    bitmap_free(interrupt->suspend_enabled);
>>   err_kfree:
>>       kfree(interrupt);
> 
> You could also use devm_kzalloc() and do away with the kfree()s you have 
> here on the probe path.
> 
>> @@ -286,6 +300,7 @@ void ipa_interrupt_deconfig(struct ipa *ipa)
>>       dev_pm_clear_wake_irq(dev);
>>       free_irq(interrupt->irq, interrupt);
>> +    bitmap_free(interrupt->suspend_enabled);
>>   }
>>   /* Initialize the IPA interrupt structure */
> 
> Just suggestions though.
> 
> Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@...aro.org>