Re: [PATCH 6/6] crypto: eip93: handle request ID exhaustion

[Jihong Min]

Hi Aleksander,

On 5/25/26 06:30, Aleksander Jan Bajkowski wrote:
> Hi Jihong,
> 
> On 24/05/2026 21:45, Jihong Min wrote:
>> diff --git a/drivers/crypto/inside-secure/eip93/eip93-main.h b/drivers/crypto/inside-secure/eip93/eip93-main.h
>> index 990c2401b7ce..5237b75bba62 100644
>> --- a/drivers/crypto/inside-secure/eip93/eip93-main.h
>> +++ b/drivers/crypto/inside-secure/eip93/eip93-main.h
>> @@ -13,11 +13,13 @@
>>  #include <crypto/internal/skcipher.h>
>>  #include <linux/bitfield.h>
>>  #include <linux/interrupt.h>
>> +#include <linux/limits.h>
>>  
>>  #define EIP93_RING_BUSY_DELAY		500
>>  
>>  #define EIP93_RING_NUM			512
>>  #define EIP93_RING_BUSY			32
>> +#define EIP93_REQUEST_IDR_LIMIT		(U16_MAX + 1)
> 
> This looks suspicious. You are now overflowing the 16-bit field
> EIP93_PE_USER_ID_CRYPTO_IDR. Did you mean (U16_MAX - 1)? Best regards,
> Aleksander
> 
U16_MAX + 1 is intentional here because it is passed to idr_alloc() as
the exclusive end value, not stored in EIP93_PE_USER_ID_CRYPTO_IDR.

So this allocates IDs 0..U16_MAX inclusive, and the value 0x10000 is
never written to the 16-bit descriptor field.

That said, the name is confusing. I will rename it to something like
EIP93_REQUEST_IDR_END and add a short comment if you prefer.


Sincerely,
Jihong Min