WebCrypto uses deprecated CCKeyDerivationHMac

[Kate Cheney]

WebCrypto uses deprecated CCKeyDerivationHMac

[Kate Cheney]

Created attachment 436287 [details]
Patch

[Kate Cheney]

rdar://48896021

[Kate Cheney]

Created attachment 436291 [details]
Patch

[Brent Fulgham]

Comment on attachment 436291 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=436291&action=review

r=me

> Source/WebCore/crypto/mac/CryptoUtilitiesCocoa.cpp:119
> +        return rv;

Does an unsuccessful call to CCKDFParametersCreateHkdf guarantee that the params are properly cleaned up?

> Source/WebCore/crypto/mac/CryptoUtilitiesCocoa.cpp:133
> +    if (keyDerivationHMAC(digestAlgorithm, key, keySize, info, infoSize, salt, saltSize, result.data(), result.size()))

Should this be a check for != kCCSuccess?

[Kate Cheney]

(In reply to Brent Fulgham from comment #4)
> Comment on attachment 436291 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=436291&action=review
> 
> r=me
> 
> > Source/WebCore/crypto/mac/CryptoUtilitiesCocoa.cpp:119
> > +        return rv;
> 
> Does an unsuccessful call to CCKDFParametersCreateHkdf guarantee that the
> params are properly cleaned up?
> 

Yes, params are not set in CCKDFParametersCreateHkdf unless it is returning kCCSuccess.

> > Source/WebCore/crypto/mac/CryptoUtilitiesCocoa.cpp:133
> > +    if (keyDerivationHMAC(digestAlgorithm, key, keySize, info, infoSize, salt, saltSize, result.data(), result.size()))
> 
> Should this be a check for != kCCSuccess?

Yes, probably easier to read that way. I'll fix before landing.

Thanks for the review!

[Kate Cheney]

Created attachment 436314 [details]
Patch

[EWS]

Committed r281554 (240921@main): <https://commits.webkit.org/240921@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 436314 [details].