Merge branch 'master' into no_guard_sc

This commit is contained in:
ismagom 2016-02-01 11:07:02 +01:00
commit f8df593e5d
3 changed files with 18 additions and 11 deletions

5
CHANGELOG Normal file
View File

@ -0,0 +1,5 @@
Change Log for Releases
==============================
## 001.001.000
* Added support for BladeRF

View File

@ -5,6 +5,9 @@ srsLTE is a free and open-source LTE library for SDR UE and eNodeB. The library
**srsLTE is used by srsUE, a full stack (PHY to IP) implementation of an LTE UE. srsUE is available at https://github.com/srslte/srsue**
*News*: We have just released srsUE 1.1 now supporting the bladeRF hardware. We have also added Paging and improved general stability.
The srsLTE software license is AGPLv3.
Current Features:
@ -28,9 +31,12 @@ Missing Features:
Hardware
========
The library currently uses Ettus Universal Hardware Driver (UHD). Thus, any hardware supported by UHD can be used. There is no sampling rate conversion, therefore the hardware should support 30.72 MHz clock in order to work correctly with LTE sampling frequencies and decode signals from live LTE base stations. We are using the B210 USRP.
The library currently supports the Ettus Universal Hardware Driver (UHD) and the bladeRF driver. Thus, any hardware supported by UHD or bladeRF can be used. There is no sampling rate conversion, therefore the hardware should support 30.72 MHz clock in order to work correctly with LTE sampling frequencies and decode signals from live LTE base stations.
*News*: We have recently added support for the bladeRF in the `next` branch. It's still experimental so any feedback will be very appreciated.
We have tested the following hardware:
* USRP B210
* USRP X300
* bladeRF
Download & Install Instructions
=================================

View File

@ -309,22 +309,18 @@ void srslte_vec_deinterleave_real_cf(cf_t *x, float *real, uint32_t len) {
#endif
}
/* Note: We align memory to 32 bytes (for AVX compatibility)
* because in some cases volk can incorrectly detect the architecture.
* This could be inefficient for SSE or non-SIMD platforms but shouldn't
* be a huge problem.
*/
void *srslte_vec_malloc(uint32_t size) {
#ifndef HAVE_VOLK
void *ptr;
if (posix_memalign(&ptr,32,size)) {
return NULL;
} else {
return ptr;
}
#else
void *ptr;
if (posix_memalign(&ptr,volk_get_alignment(),size)) {
return NULL;
} else {
return ptr;
}
#endif
}
void *srslte_vec_realloc(void *ptr, uint32_t old_size, uint32_t new_size) {