xemu-project/xemu
1
0
Fork 0
mirror of https://github.com/xemu-project/xemu.git synced 2024-11-28 14:00:44 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
bffe88d139
xemu/include/qemu/bswap.h
Peter Maydell 77b1757090 include/qemu/bswap.h: Use __builtin_memcpy() in accessor functions 
In the accessor functions ld*_he_p() and st*_he_p() we use memcpy()
to perform a load or store to a pointer which might not be aligned
for the size of the type. We rely on the compiler to optimize this
memcpy() into an efficient load or store instruction where possible.
This is required for good performance, but at the moment it is also
required for correct operation, because some users of these functions
require that the access is atomic if the pointer is aligned, which
will only be the case if the compiler has optimized out the memcpy().
(The particular example where we discovered this is the virtio
vring_avail_idx() which calls virtio_lduw_phys_cached() which
eventually ends up calling lduw_he_p().)

Unfortunately some compile environments, such as the fortify-source
setup used in Alpine Linux, define memcpy() to a wrapper function
in a way that inhibits this compiler optimization.

The correct long-term fix here is to add a set of functions for
doing atomic accesses into AddressSpaces (and to other relevant
families of accessor functions like the virtio_*_phys_cached()
ones), and make sure that callsites which want atomic behaviour
use the correct functions.

In the meantime, switch to using __builtin_memcpy() in the
bswap.h accessor functions. This will make us robust against things
like this fortify library in the short term. In the longer term
it will mean that we don't end up with these functions being really
badly-performing even if the semantics of the out-of-line memcpy()
are correct.

Reported-by: Fernando Casas Schössow <casasfernando@outlook.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20190318112938.8298-1-peter.maydell@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2019-04-09 18:34:21 +02:00

562 lines
14 KiB
C
Raw Blame History

#ifndef BSWAP_H
#define BSWAP_H
#include "fpu/softfloat-types.h"
#ifdef CONFIG_MACHINE_BSWAP_H
# include <sys/endian.h>
# include <machine/bswap.h>
#elif defined(__FreeBSD__)
# include <sys/endian.h>
#elif defined(CONFIG_BYTESWAP_H)
# include <byteswap.h>
static inline uint16_t bswap16(uint16_t x)
{
return bswap_16(x);
}
static inline uint32_t bswap32(uint32_t x)
{
return bswap_32(x);
}
static inline uint64_t bswap64(uint64_t x)
{
return bswap_64(x);
}
# else
static inline uint16_t bswap16(uint16_t x)
{
return (((x & 0x00ff) << 8) |
((x & 0xff00) >> 8));
}
static inline uint32_t bswap32(uint32_t x)
{
return (((x & 0x000000ffU) << 24) |
((x & 0x0000ff00U) << 8) |
((x & 0x00ff0000U) >> 8) |
((x & 0xff000000U) >> 24));
}
static inline uint64_t bswap64(uint64_t x)
{
return (((x & 0x00000000000000ffULL) << 56) |
((x & 0x000000000000ff00ULL) << 40) |
((x & 0x0000000000ff0000ULL) << 24) |
((x & 0x00000000ff000000ULL) << 8) |
((x & 0x000000ff00000000ULL) >> 8) |
((x & 0x0000ff0000000000ULL) >> 24) |
((x & 0x00ff000000000000ULL) >> 40) |
((x & 0xff00000000000000ULL) >> 56));
}
#endif /* ! CONFIG_MACHINE_BSWAP_H */
static inline void bswap16s(uint16_t *s)
{
*s = bswap16(*s);
}
static inline void bswap32s(uint32_t *s)
{
*s = bswap32(*s);
}
static inline void bswap64s(uint64_t *s)
{
*s = bswap64(*s);
}
#if defined(HOST_WORDS_BIGENDIAN)
#define be_bswap(v, size) (v)
#define le_bswap(v, size) glue(bswap, size)(v)
#define be_bswaps(v, size)
#define le_bswaps(p, size) do { *p = glue(bswap, size)(*p); } while(0)
#else
#define le_bswap(v, size) (v)
#define be_bswap(v, size) glue(bswap, size)(v)
#define le_bswaps(v, size)
#define be_bswaps(p, size) do { *p = glue(bswap, size)(*p); } while(0)
#endif
/**
* Endianness conversion functions between host cpu and specified endianness.
* (We list the complete set of prototypes produced by the macros below
* to assist people who search the headers to find their definitions.)
*
* uint16_t le16_to_cpu(uint16_t v);
* uint32_t le32_to_cpu(uint32_t v);
* uint64_t le64_to_cpu(uint64_t v);
* uint16_t be16_to_cpu(uint16_t v);
* uint32_t be32_to_cpu(uint32_t v);
* uint64_t be64_to_cpu(uint64_t v);
*
* Convert the value @v from the specified format to the native
* endianness of the host CPU by byteswapping if necessary, and
* return the converted value.
*
* uint16_t cpu_to_le16(uint16_t v);
* uint32_t cpu_to_le32(uint32_t v);
* uint64_t cpu_to_le64(uint64_t v);
* uint16_t cpu_to_be16(uint16_t v);
* uint32_t cpu_to_be32(uint32_t v);
* uint64_t cpu_to_be64(uint64_t v);
*
* Convert the value @v from the native endianness of the host CPU to
* the specified format by byteswapping if necessary, and return
* the converted value.
*
* void le16_to_cpus(uint16_t *v);
* void le32_to_cpus(uint32_t *v);
* void le64_to_cpus(uint64_t *v);
* void be16_to_cpus(uint16_t *v);
* void be32_to_cpus(uint32_t *v);
* void be64_to_cpus(uint64_t *v);
*
* Do an in-place conversion of the value pointed to by @v from the
* specified format to the native endianness of the host CPU.
*
* void cpu_to_le16s(uint16_t *v);
* void cpu_to_le32s(uint32_t *v);
* void cpu_to_le64s(uint64_t *v);
* void cpu_to_be16s(uint16_t *v);
* void cpu_to_be32s(uint32_t *v);
* void cpu_to_be64s(uint64_t *v);
*
* Do an in-place conversion of the value pointed to by @v from the
* native endianness of the host CPU to the specified format.
*
* Both X_to_cpu() and cpu_to_X() perform the same operation; you
* should use whichever one is better documenting of the function your
* code is performing.
*
* Do not use these functions for conversion of values which are in guest
* memory, since the data may not be sufficiently aligned for the host CPU's
* load and store instructions. Instead you should use the ld*_p() and
* st*_p() functions, which perform loads and stores of data of any
* required size and endianness and handle possible misalignment.
*/
#define CPU_CONVERT(endian, size, type)\
static inline type endian ## size ## _to_cpu(type v)\
{\
return glue(endian, _bswap)(v, size);\
}\
\
static inline type cpu_to_ ## endian ## size(type v)\
{\
return glue(endian, _bswap)(v, size);\
}\
\
static inline void endian ## size ## _to_cpus(type *p)\
{\
glue(endian, _bswaps)(p, size);\
}\
\
static inline void cpu_to_ ## endian ## size ## s(type *p)\
{\
glue(endian, _bswaps)(p, size);\
}
CPU_CONVERT(be, 16, uint16_t)
CPU_CONVERT(be, 32, uint32_t)
CPU_CONVERT(be, 64, uint64_t)
CPU_CONVERT(le, 16, uint16_t)
CPU_CONVERT(le, 32, uint32_t)
CPU_CONVERT(le, 64, uint64_t)
/* len must be one of 1, 2, 4 */
static inline uint32_t qemu_bswap_len(uint32_t value, int len)
{
return bswap32(value) >> (32 - 8 * len);
}
/*
* Same as cpu_to_le{16,32}, except that gcc will figure the result is
* a compile-time constant if you pass in a constant. So this can be
* used to initialize static variables.
*/
#if defined(HOST_WORDS_BIGENDIAN)
# define const_le32(_x) \
((((_x) & 0x000000ffU) << 24) | \
(((_x) & 0x0000ff00U) << 8) | \
(((_x) & 0x00ff0000U) >> 8) | \
(((_x) & 0xff000000U) >> 24))
# define const_le16(_x) \
((((_x) & 0x00ff) << 8) | \
(((_x) & 0xff00) >> 8))
#else
# define const_le32(_x) (_x)
# define const_le16(_x) (_x)
#endif
/* Unions for reinterpreting between floats and integers. */
typedef union {
float32 f;
uint32_t l;
} CPU_FloatU;
typedef union {
float64 d;
#if defined(HOST_WORDS_BIGENDIAN)
struct {
uint32_t upper;
uint32_t lower;
} l;
#else
struct {
uint32_t lower;
uint32_t upper;
} l;
#endif
uint64_t ll;
} CPU_DoubleU;
typedef union {
floatx80 d;
struct {
uint64_t lower;
uint16_t upper;
} l;
} CPU_LDoubleU;
typedef union {
float128 q;
#if defined(HOST_WORDS_BIGENDIAN)
struct {
uint32_t upmost;
uint32_t upper;
uint32_t lower;
uint32_t lowest;
} l;
struct {
uint64_t upper;
uint64_t lower;
} ll;
#else
struct {
uint32_t lowest;
uint32_t lower;
uint32_t upper;
uint32_t upmost;
} l;
struct {
uint64_t lower;
uint64_t upper;
} ll;
#endif
} CPU_QuadU;
/* unaligned/endian-independent pointer access */
/*
* the generic syntax is:
*
* load: ld{type}{sign}{size}_{endian}_p(ptr)
*
* store: st{type}{size}_{endian}_p(ptr, val)
*
* Note there are small differences with the softmmu access API!
*
* type is:
* (empty): integer access
* f : float access
*
* sign is:
* (empty): for 32 or 64 bit sizes (including floats and doubles)
* u : unsigned
* s : signed
*
* size is:
* b: 8 bits
* w: 16 bits
* l: 32 bits
* q: 64 bits
*
* endian is:
* he : host endian
* be : big endian
* le : little endian
* te : target endian
* (except for byte accesses, which have no endian infix).
*
* The target endian accessors are obviously only available to source
* files which are built per-target; they are defined in cpu-all.h.
*
* In all cases these functions take a host pointer.
* For accessors that take a guest address rather than a
* host address, see the cpu_{ld,st}_* accessors defined in
* cpu_ldst.h.
*
* For cases where the size to be used is not fixed at compile time,
* there are
* stn_{endian}_p(ptr, sz, val)
* which stores @val to @ptr as an @endian-order number @sz bytes in size
* and
* ldn_{endian}_p(ptr, sz)
* which loads @sz bytes from @ptr as an unsigned @endian-order number
* and returns it in a uint64_t.
*/
static inline int ldub_p(const void *ptr)
{
return *(uint8_t *)ptr;
}
static inline int ldsb_p(const void *ptr)
{
return *(int8_t *)ptr;
}
static inline void stb_p(void *ptr, uint8_t v)
{
*(uint8_t *)ptr = v;
}
/*
* Any compiler worth its salt will turn these memcpy into native unaligned
* operations. Thus we don't need to play games with packed attributes, or
* inline byte-by-byte stores.
* Some compilation environments (eg some fortify-source implementations)
* may intercept memcpy() in a way that defeats the compiler optimization,
* though, so we use __builtin_memcpy() to give ourselves the best chance
* of good performance.
*/
static inline int lduw_he_p(const void *ptr)
{
uint16_t r;
__builtin_memcpy(&r, ptr, sizeof(r));
return r;
}
static inline int ldsw_he_p(const void *ptr)
{
int16_t r;
__builtin_memcpy(&r, ptr, sizeof(r));
return r;
}
static inline void stw_he_p(void *ptr, uint16_t v)
{
__builtin_memcpy(ptr, &v, sizeof(v));
}
static inline int ldl_he_p(const void *ptr)
{
int32_t r;
__builtin_memcpy(&r, ptr, sizeof(r));
return r;
}
static inline void stl_he_p(void *ptr, uint32_t v)
{
__builtin_memcpy(ptr, &v, sizeof(v));
}
static inline uint64_t ldq_he_p(const void *ptr)
{
uint64_t r;
__builtin_memcpy(&r, ptr, sizeof(r));
return r;
}
static inline void stq_he_p(void *ptr, uint64_t v)
{
__builtin_memcpy(ptr, &v, sizeof(v));
}
static inline int lduw_le_p(const void *ptr)
{
return (uint16_t)le_bswap(lduw_he_p(ptr), 16);
}
static inline int ldsw_le_p(const void *ptr)
{
return (int16_t)le_bswap(lduw_he_p(ptr), 16);
}
static inline int ldl_le_p(const void *ptr)
{
return le_bswap(ldl_he_p(ptr), 32);
}
static inline uint64_t ldq_le_p(const void *ptr)
{
return le_bswap(ldq_he_p(ptr), 64);
}
static inline void stw_le_p(void *ptr, uint16_t v)
{
stw_he_p(ptr, le_bswap(v, 16));
}
static inline void stl_le_p(void *ptr, uint32_t v)
{
stl_he_p(ptr, le_bswap(v, 32));
}
static inline void stq_le_p(void *ptr, uint64_t v)
{
stq_he_p(ptr, le_bswap(v, 64));
}
/* float access */
static inline float32 ldfl_le_p(const void *ptr)
{
CPU_FloatU u;
u.l = ldl_le_p(ptr);
return u.f;
}
static inline void stfl_le_p(void *ptr, float32 v)
{
CPU_FloatU u;
u.f = v;
stl_le_p(ptr, u.l);
}
static inline float64 ldfq_le_p(const void *ptr)
{
CPU_DoubleU u;
u.ll = ldq_le_p(ptr);
return u.d;
}
static inline void stfq_le_p(void *ptr, float64 v)
{
CPU_DoubleU u;
u.d = v;
stq_le_p(ptr, u.ll);
}
static inline int lduw_be_p(const void *ptr)
{
return (uint16_t)be_bswap(lduw_he_p(ptr), 16);
}
static inline int ldsw_be_p(const void *ptr)
{
return (int16_t)be_bswap(lduw_he_p(ptr), 16);
}
static inline int ldl_be_p(const void *ptr)
{
return be_bswap(ldl_he_p(ptr), 32);
}
static inline uint64_t ldq_be_p(const void *ptr)
{
return be_bswap(ldq_he_p(ptr), 64);
}
static inline void stw_be_p(void *ptr, uint16_t v)
{
stw_he_p(ptr, be_bswap(v, 16));
}
static inline void stl_be_p(void *ptr, uint32_t v)
{
stl_he_p(ptr, be_bswap(v, 32));
}
static inline void stq_be_p(void *ptr, uint64_t v)
{
stq_he_p(ptr, be_bswap(v, 64));
}
/* float access */
static inline float32 ldfl_be_p(const void *ptr)
{
CPU_FloatU u;
u.l = ldl_be_p(ptr);
return u.f;
}
static inline void stfl_be_p(void *ptr, float32 v)
{
CPU_FloatU u;
u.f = v;
stl_be_p(ptr, u.l);
}
static inline float64 ldfq_be_p(const void *ptr)
{
CPU_DoubleU u;
u.ll = ldq_be_p(ptr);
return u.d;
}
static inline void stfq_be_p(void *ptr, float64 v)
{
CPU_DoubleU u;
u.d = v;
stq_be_p(ptr, u.ll);
}
static inline unsigned long leul_to_cpu(unsigned long v)
{
#if HOST_LONG_BITS == 32
return le_bswap(v, 32);
#elif HOST_LONG_BITS == 64
return le_bswap(v, 64);
#else
# error Unknown sizeof long
#endif
}
/* Store v to p as a sz byte value in host order */
#define DO_STN_LDN_P(END) \
static inline void stn_## END ## _p(void *ptr, int sz, uint64_t v) \
{ \
switch (sz) { \
case 1: \
stb_p(ptr, v); \
break; \
case 2: \
stw_ ## END ## _p(ptr, v); \
break; \
case 4: \
stl_ ## END ## _p(ptr, v); \
break; \
case 8: \
stq_ ## END ## _p(ptr, v); \
break; \
default: \
g_assert_not_reached(); \
} \
} \
static inline uint64_t ldn_## END ## _p(const void *ptr, int sz) \
{ \
switch (sz) { \
case 1: \
return ldub_p(ptr); \
case 2: \
return lduw_ ## END ## _p(ptr); \
case 4: \
return (uint32_t)ldl_ ## END ## _p(ptr); \
case 8: \
return ldq_ ## END ## _p(ptr); \
default: \
g_assert_not_reached(); \
} \
}
DO_STN_LDN_P(he)
DO_STN_LDN_P(le)
DO_STN_LDN_P(be)
#undef DO_STN_LDN_P
#undef le_bswap
#undef be_bswap
#undef le_bswaps
#undef be_bswaps
#endif /* BSWAP_H */
Reference in New Issue View Git Blame Copy Permalink