| Message ID | 20220928165018.73496-1-Jason@zx2c4.com (mailing list archive) |
|---|---|
| State | Not Applicable |
| Delegated to: | Herbert Xu |
| Headers | show |
| Series | random: add 8-bit and 16-bit batches | expand |
On Thu, Sep 29, 2022 at 1:33 AM Jason A. Donenfeld <Jason@zx2c4.com> wrote: > > There are numerous places in the kernel that would be sped up by having > smaller batches. ... > void get_random_bytes(void *buf, size_t len); > +u8 get_random_u8(void); > +u16 get_random_u16(void); > u32 get_random_u32(void); > u64 get_random_u64(void); > static inline unsigned int get_random_int(void) To me, the 32-bit & 64-bit functions look like an obviously good idea. However, I cannot see that the 8-bit or 16-bit functions are needed. Even library functions like getchar() return an int & whatever you return, it is going to be handled as an int-sized item if it goes in a register, so what's the point?
From: Sandy Harris > Sent: 25 November 2022 04:11 > > On Thu, Sep 29, 2022 at 1:33 AM Jason A. Donenfeld <Jason@zx2c4.com> wrote: > > > > There are numerous places in the kernel that would be sped up by having > > smaller batches. ... > > > void get_random_bytes(void *buf, size_t len); > > +u8 get_random_u8(void); > > +u16 get_random_u16(void); > > u32 get_random_u32(void); > > u64 get_random_u64(void); > > static inline unsigned int get_random_int(void) > > To me, the 32-bit & 64-bit functions look like an > obviously good idea. However, I cannot see > that the 8-bit or 16-bit functions are needed. > > Even library functions like getchar() return an > int & whatever you return, it is going to be > handled as an int-sized item if it goes in a > register, so what's the point? They avoid 'using up' random values the callers wont use. This can save expensive re-hashing (etc). OTOH the return types should probably be u32 even though the domain is smaller. David - Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK Registration No: 1397386 (Wales)
diff --git a/drivers/char/random.c b/drivers/char/random.c index f2aa3ab1b458..64ee16ffb8b7 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -506,6 +506,8 @@ EXPORT_SYMBOL(get_random_ ##type); DEFINE_BATCHED_ENTROPY(u64) DEFINE_BATCHED_ENTROPY(u32) +DEFINE_BATCHED_ENTROPY(u16) +DEFINE_BATCHED_ENTROPY(u8) #ifdef CONFIG_SMP /* diff --git a/include/linux/random.h b/include/linux/random.h index a9e6e16f9774..2c130f8f18e5 100644 --- a/include/linux/random.h +++ b/include/linux/random.h @@ -38,6 +38,8 @@ static inline int unregister_random_vmfork_notifier(struct notifier_block *nb) { #endif void get_random_bytes(void *buf, size_t len); +u8 get_random_u8(void); +u16 get_random_u16(void); u32 get_random_u32(void); u64 get_random_u64(void); static inline unsigned int get_random_int(void)
There are numerous places in the kernel that would be sped up by having smaller batches. Currently those callsites do `get_random_u32() & 0xff` or similar. Since these are pretty spread out, and will require patches to multiple different trees, let's get ahead of the curve and lay the foundation for `get_random_u8()` and `get_random_u16()`, so that it's then possible to start submitting conversion patches leisurely. Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com> --- drivers/char/random.c | 2 ++ include/linux/random.h | 2 ++ 2 files changed, 4 insertions(+)