On Tue, 1 Dec 2020 21:03:59 +0200 Joonas Kylmälä <joonas.kylm...@iki.fi> wrote:
> +#define MD5_DIGEST_LENGTH_ASCII (2 * MD5_DIGEST_LENGTH) > + > char *ipc_nv_data_md5_calculate(struct ipc_client *client, > const char *path, const char *secret, > size_t size, size_t chunk_size) > @@ -112,7 +114,7 @@ int ipc_nv_data_md5_path_check(struct ipc_client > *client) return -1; > } > > - if (st.st_size < 2 * sizeof(char) * MD5_DIGEST_LENGTH) { > + if (st.st_size != MD5_DIGEST_LENGTH_ASCII) { > ipc_client_log(client, "Checking nv_data md5 size Thanks a lot, that looks way cleaner. Besides the very long line in the commit ("rfs.c:115:17: error: comparison of integer expressions of different signedness: ‘__off_t’ {aka ‘long int’} and ‘long unsigned int’ [-Werror=sign-compare]") that needs to be spitted, everything looks good. The warnings are also gone in all the configuration tested by the Guix.scm. I can take care of that right pushing it though if you want. > I couldn't do any real device testing because I don't have any storage > space to build replicant 6 and would have had to delete replicant 11 > work to do this so if this patch looks good to you it would be awesome > if you could give it a spin on a real device. That indeed needs to be tested. I'll try to test it on my main device, however it might take a bit of delay (I'm also in the middle of merging the patchset to remove the Ambient SDK and I need to do a backup of the phone data too). Denis.
pgpF7yobsIrKC.pgp
Description: OpenPGP digital signature
_______________________________________________ Replicant mailing list Replicant@osuosl.org https://lists.osuosl.org/mailman/listinfo/replicant