summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2020-09-25leon3,shm: update default base address (for GR740 MP config)rcc-v1.3.0rcc-1.3Daniel Hellstrom
The shared memory base is normally set by the user application. It defaults to 0x40000000 for LEON3 (start of memory) since before, with GR740 having its memory at 0x00000000 a better base address the new default is to based it on the _RAM_START taken from the linker script.
2020-09-09leon,mptests: add BSP specific linkflags for MP testsDaniel Hellstrom
The RTEMS MP test-suite was originally designed to two identical nodes, however the LEON devices is often operating several instances of the RTEMS MP configuration wihtin the same device sharing memory. Therefore this places the RTEMS MP nodes at different RAM location. The node1.exe and node2.exe can be started from GRMON like this on a GR712RC with 4MiB SRAM: grmon> cpu act 0 grmon> load build/sparc-gaisler-rtems5/c/gr712rc_mp/testsuites/mptests/mp01_node1.exe cpu0 grmon> stack 0x401ffff0 cpu0 grmon> load build/sparc-gaisler-rtems5/c/gr712rc_mp/testsuites/mptests/mp01_node2.exe cpu1 grmon> stack 0x403ffff0 cpu1 grmon> run
2020-09-09leon,gr1553rt: avoid warning of undefined ISR prototypeDaniel Hellstrom
2020-09-09leon3: avoid warning in setvec()Daniel Hellstrom
2020-09-09leon3,gr712rc: added GR712RC AMP BSP configurationDaniel Hellstrom
2020-09-09leon3,gr740: added GR740 AMP BSP configurationDaniel Hellstrom
2020-09-09grlib: remove old non-SMP GRSPW device driverDaniel Hellstrom
There is no need for two device drivers for the same IP. Old applications should migrate to using the newer GRSPW packet driver (grspw_pkt.c) which has a much more flexible interface, more functions with a better zero-copy API and is actively maintained. All LEON devices and SpW IP from Gaisler are supported by the GRSPW Packet driver.
2020-09-09leon3: add BSP config for future LEON3 multi-processor (AMP) BSPDaniel Hellstrom
2020-09-09leon3: add BSP specific default RTEMS configDaniel Hellstrom
Provide a BSP specific default RTEMS configuration mainly intended to simplify for POSIX targets bring-up
2020-09-02leon3,printk: avoid optimizing away preinit printk bufferDaniel Hellstrom
2020-09-02leon,apbuart: sanity check name prefix before usingDaniel Hellstrom
2020-09-02leon,uart: if dev selected sys console force fs-nameDaniel Hellstrom
2020-09-02leon,ckinit: fixes build issue introduced by bb99cd0dDaniel Hellstrom
2020-09-02leon,tn-0018: work around GRLIB-TN-0018 errataDaniel Hellstrom
The errata is worked around in the kernel without requiring toolchain modifications. It is triggered the JMPL/RETT return from trap instruction sequence never generated by the compiler and. There are also other conditions that must must be true to trigger the errata, for example the instruction that the trap returns to has to be a JMPL instruction. The errata can only be triggered if certain data is corrected by ECC (inflicted by radiation), thus it can not be triggered under normal operation. For more information see: www.gaisler.com/notes Affected RTEMS target BSPs: * GR712RC * UT699 * UT700/699E The work around is enabled by defining __FIX_LEON3_TN0018 at build time. In general there are two approaches that the workaround uses: A) avoid ECC restarting the RETT instruction B) avoid returning from trap to a JMPL instruction Where A) comes at a higher performance cost than B), so B) is used where posssible. B) can be achived for certain returns from trap handlers if trap entry is controlled by assembly, such as system calls. A) A special JMPL/RETT sequence where instruction cache is disabled temporarily to avoid RETT containing ECC errors, and reading of RETT source registers to "clean" them from incorrect ECC just before RETT is executed. B) The work around prevents JMPL after system calls (TA instruction) and modifies assembly code on return from traps jumping back to application code. Note that for some traps the trapped instruction is always re-executed and can therefore not trigger the errata, for example the SAVE instruction causing window overflow or an float instruction causing FPU disabled trap. RTEMS SPARC traps workaround implementation: NAME NOTE TRAP COMMENT * window overflow 1 - 0x05 always returns to a SAVE * window underflow 1 - 0x06 always returns to a RESTORE * interrupt traps 2 - 0x10..1f special rett sequence workaround * syscall 3 - 0x80 shutdown system - never returns * ABI flush windows 2 - 0x83 special rett sequence workaround * syscall_irqdis 4 - 0x89 * syscall_irqen 4 - 0x8A * syscall_irqdis_fp 1 - 0x8B always jumps back to FP instruction * syscall_lazy_fp_switch 5 - 0x04 A) jumps back to FP instruction, or to B) _Internal_error() starting with SAVE Notes: 1) no workaround needed because trap always returns to non-JMPL instruction 2) workaround implemented by special rett sequence 3) no workaround needed because system call never returns 4) workaround implemented by inserting NOP in system call generation. Thus fall into 1) when workaround is enabled and no trap handler fix needed. 5) trap handler branches into both 1) and returning to _Internal_error() which starts with a SAVE and besides since it shuts down the system that RETT should never be in cache (only executed once) so fix not necessary in this case. Any custom trap handlers may also have to be updated. To simplify that, helper work around assembly code in macros are available in a separate include file <libcpu/grlib-tn-0018.h>.
2020-09-02rcc: add a RCC version string to LEON2/3 BSPsDaniel Hellstrom
2020-09-02sparc,leon: avoid triggering LEON3FT errata TN-0009Daniel Hellstrom
2020-09-02sparc,leon: avoid triggering TN-0009 bad sequenceDaniel Hellstrom
2020-09-02grlib,grspw_pkt: correct link state enum numberingDaniel Hellstrom
Not used by the driver itself, but shuold be correct if used by application.
2020-09-02leon, l2cache: prevent unused diagnostic accessMartin Aberg
2020-09-02leon, ahbstat: register definitions for AHBSTAT version 1Martin Aberg
2020-09-02bsps/leon3: Do not invalidate the entry pointMartin Aberg
This commit fixes an issue when booting SMP application: entry point can become invalid for secondary processors. The first CPU does the following in the boot_card(): 1. Release other CPUs from power-down (but does not wait here) 2. Some other stuff 3. Overwrite trap entry 0 with spurious interrupt handler. 4. The rest It means that the entry point is guaranteed to be valid for the first CPU entering the RTEMS kernel. But 1. and 3. above gives a race. CPU1.. will either enter the kernel properly or end up in the spurious interrupt handler depending on how soon it reaches the entry point it is. One example where this has been an issue is when secondary processors run self-tests in a ROM boot loader before entering the RTEMS entry point. "start" is trap entry 0 in the trap table and directly jumps to "hard_reset".
2020-09-02leon, grspw_router: added router_port_link_div()Martin Aberg
Allows user to set SpaceWire run clock divisor for an individual port.
2020-09-02sparc: fix bad register alignment for 64 bit storeDaniel Hellstrom
2020-09-02llvm: build dl08 without depending on gccDaniel Hellstrom
2020-09-02leon: disable dl10 because it requires inputDaniel Hellstrom
dl10 test actually works when the appropriate input is given, however the test frame work is not capable of doing this and check agains .scn file.
2020-09-02leon: restart and load timer counter at initializationDaniel Hellstrom
Without this smp05 and smpthreadlife01 tests may fail depending on how the boot loader initialized the GPTIMER. Before the time counter stopped counting when reaching zero, but tests could work since it could take 2^32 us before stopping. The timer driver will potentially overwrite this, but it happens later due to the initialization order having RTEMS_SYSINIT_CPU_COUNTER very early.
2020-09-02leon,ckinit: avoid assuming 1MHz timer pre-scaler clockDaniel Hellstrom
2020-09-02leon3: avoid dependency on apbuart/timer driverDaniel Hellstrom
Moves drvmgr_drivers[] from amba.c to a separate file in order to avoid the dependecy on APBUART/GPTIMER drivers. This has an effect when user configured not to use timer or uart in their project.
2020-09-02leon,greth: added support for variable sized descriptor table sizesDaniel Hellstrom
The descriptor table size is equal to its alignment and set when configuring the HW IP through VHDL generics. This SW patch simply probes the HW how large the RX/TX descriptor tables are and adjusts accordingly. The number of descriptors actual used are controlled by other settings (rxDescs and txDescs) controlled by the user.
2020-09-02leon,grcan: split out GRCAN non-FD specific support in separate fileDaniel Hellstrom
2020-09-02leon,grcanfd: split out GRCANFD specific support in separate fileDaniel Hellstrom
2020-09-02leon,grcan: added support for GRCANFDDaniel Hellstrom
The new GRCAN_FD IP supports CAN FD standard and is mostly backwards compatible with GRCAN SW interface. The GRCAN driver have been extended to support the GRCANFD IP using the same driver. Additional functions have been added that uses a new CAN FD frame format and read/write/baud-rate functions that supports both GRCANFD and GRCAN. To keep the SW API fully backwards compatible with GRCAN, the old functions remain.
2020-09-02grlib: added 64-bit read no-cache functionDaniel Hellstrom
2020-09-02leon,occan: use common CAN baud-rate calculation routineDaniel Hellstrom
2020-09-02leon,grcan: use common CAN baud-rate calculation routineDaniel Hellstrom
2020-09-02leon,can: introduce common CAN baud-rate calculation functionDaniel Hellstrom
Reimplemented the baud-rate algorithm from scratch to cope with GRCAN, GRCANFD and OC_CAN devices.
2020-09-02grlib,ambapp: added new IP core IDsDaniel Hellstrom
2020-09-02leon,grspw: fix for SET_PACKET_SIZEDaniel Hellstrom
When the DMA table has been allocated dynamically, the IOCTL_SET_PACKETSIZE will trigger an issue where pDev->rx and pDev->tx are not updated with the new DMA tables base address. Instead the old pointers are used. There is no point in reallocting the DMA tables because there is no configuration option to it. Therefore the DMA tables allocation is moved to a separate function never called from SET_PACKETSIZE.
2020-09-02leon3v7: added new bsp configurationDaniel Hellstrom
2020-09-02leon,gr1553b: Only align allocated memory. Verify alignment of memory.Arvid
2020-09-02leon,gr1553b: set codec versionArvid
This is enables the updated codec for GR740 and is backwards compatible with all other versions of the IP.
2020-09-02leon,gr1553rt: Fixed spinlock unlockArvid
2020-09-02leon,gr1553rt: Fixed memory leakArvid
2020-09-02testsuite,clang: fix malloc04 and malloctest for on clangDaniel Hellstrom
malloc04 and malloctest tests from the rtems test-suite fails when checking the return value of malloc(). The check is optimized away and always fails.
2020-09-02sparc,clang: disable openmp01 on SMP BSPs since unsupported by clang toolchainDaniel Hellstrom
It seems not to be possible to disable tests based on toolchain or any other if condition or configure (--disable-test-N or similar). Even though it is working on GCC it is temporarily disabled her.
2020-09-02score: Un-constification of arguments to atomic_loadJacob Hansen
The C11 standard definition of atomic_load do not accept const pointers. This appear to be a deficiency in the C11 standard. It looks like a fix is underway/has applied as per: http://www.open-std.org/jtc1/sc22/wg14/www/docs/dr_459.htm While GCC do allow const pointers to atomic_load Clang does not. To compile with Clang we therefore need to remove the const qualifiers.
2020-09-02[DO-NOT-UPSTREAM] Speed up bootstrappingDaniel Hellstrom
2020-09-02gr712/740,leon3_sf/flat: temporarily disable dl06 test until bin2c rebuiltDaniel Hellstrom
Disable dl06 test that does not currently build
2020-09-02leon3: added soft-float BSPDaniel Hellstrom
For compatibility with non-FPU targets.
2020-09-02leon3: added leon3_flat BSP configurationDaniel Hellstrom
The flat register window model activated by -mflat is ABI compatible with the standard SPARC ABI. SAVE/RESTORE instructions are not generated by the compiler.
generated by cgit v1.2.3 (git 2.25.1) at 2024-05-20 20:05:03 +0000