67fd6be0b
Denys Vlasenko
2022-05-03
Subject
seedrng: do not hash in a constant string, it's not adding entropy
Binary
No binary (build failed)
Architecture & Compiler Flags
—
Build Log
(log yok)
Commit Diff GitHub →
miscutils/seedrng.c+8 -8
| @@ -169,7 +169,7 @@ int seedrng_main(int argc UNUSED_PARAM, char **argv) | |
| uint8_t new_seed[MAX_SEED_LEN]; | |
| size_t new_seed_len; | |
| bool new_seed_creditable; | |
| - | struct timespec timestamp; |
| + | struct timespec timestamp[2]; |
| sha256_ctx_t hash; | |
| enum { | |
| @@ -197,19 +197,19 @@ int seedrng_main(int argc UNUSED_PARAM, char **argv) | |
| * Avoid concurrent runs by taking a blocking lock on the directory. | |
| * Not checking for errors. Looking at manpage, | |
| * ENOLCK "The kernel ran out of memory for allocating lock records" | |
| - | * seems to be the only one which is likely - and if that happens, |
| + | * seems to be the only one which is possible - and if that happens, |
| * machine is OOMing (much worse problem than inability to lock...). | |
| * Also, typically configured Linux machines do not fail GFP_KERNEL | |
| * allocations (they trigger memory reclaim instead). | |
| */ | |
| - | flock(dfd, LOCK_EX); /* would block while another copy runs */ |
| + | flock(dfd, LOCK_EX); /* blocks while another instance runs */ |
| sha256_begin(&hash); | |
| - | sha256_hash(&hash, "SeedRNG v1 Old+New Prefix", 25); |
| - | clock_gettime(CLOCK_REALTIME, ×tamp); |
| - | sha256_hash(&hash, ×tamp, sizeof(timestamp)); |
| - | clock_gettime(CLOCK_BOOTTIME, ×tamp); |
| - | sha256_hash(&hash, ×tamp, sizeof(timestamp)); |
| + | //Hashing in a constant string doesn't add any entropy |
| + | // sha256_hash(&hash, "SeedRNG v1 Old+New Prefix", 25); |
| + | clock_gettime(CLOCK_REALTIME, ×tamp[0]); |
| + | clock_gettime(CLOCK_BOOTTIME, ×tamp[1]); |
| + | sha256_hash(&hash, timestamp, sizeof(timestamp)); |
| for (i = 0; i <= 1; i++) { | |
| seed_from_file_if_exists( | |