aboutsummaryrefslogtreecommitdiff
path: root/target-sparc
AgeCommit message (Collapse)Author
2010-05-20target-sparc: Inline some generation of carry for ADDX/SUBX.Richard Henderson
Computing carry is trivial for some inputs. By avoiding an external function call, we generate near-optimal code for the common cases of add+addx (double-word arithmetic) and cmp+addx (a setcc pattern). Signed-off-by: Richard Henderson <rth@twiddle.net> Acked-by: Artyom Tarasenko <atar4qemu@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-05-19target-sparc: Simplify ICC generation.Richard Henderson
Use int32 types instead of target_ulong when computing ICC. This simplifies the generated code for 32-bit host and 64-bit guest. Use the same simplified expressions for ICC as were already used for XCC in carry flag generation. Simplify the ADD carry generation to not consider a possible carry-in. Use the more complex carry computation for ADDX only. Use the same carry algorithm for the XCC result of ADDX. Similarly for SUB/SUBX. Use the ADD carry generation functions for TADD/TADDTV. Similarly for SUB and TSUB/TSUBTV. Tidy the code with respect to CODING_STYLE. Signed-off-by: Richard Henderson <rth@twiddle.net> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-05-19target-sparc: Fix compilation with --enable-debug.Richard Henderson
Return a target_ulong from compute_C_icc to match the width of the users. Signed-off-by: Richard Henderson <rth@twiddle.net> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-05-16sparc: move DT and QT defines to op_helper.cBlue Swirl
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-05-16sparc64: fix TT_WOTHER valueIgor V. Kovalenko
- fix off by one error in spill trap number bit for other window (must be bit 5) - fixes invalid instruction issue with HelenOS Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-05-16sparc64: fix mmu demap operand typoIgor V. Kovalenko
- must use store address operand to demap, not store value Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-05-12target-sparc: Fix wrong printf argumentStefan Weil
cpu_get_ccr() returns a target_ulong, so a type cast is needed to avoid wrong output on big endian hosts. We could also use TARGET_FMT_lx, but that would print 8 instead of 2 digits. Cc: Blue Swirl <blauwirbel@gmail.com> Signed-off-by: Stefan Weil <weil@mail.berlios.de> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-05-09sparc: Fix lazy flag calculation on interrupts, refactorBlue Swirl
Recalculate Sparc64 CPU flags on interrupts, otherwise some earlier flags could be stored to pstate. Refactor PSR/CCR/CWP handling: concentrate the actual functions to op_helper.c. Thanks to Igor Kovalenko for reporting. Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-05-09sparc: lazy C flag calculationBlue Swirl
Calculate only the carry flag for ADDX/SUBX instead of full set of flags. Thanks to Igor Kovalenko for spotting a bug with an earlier version. Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-05-07sparc64: fix build with older gccsBlue Swirl
Fix errors missed in 2065061ede22d401aae2ce995c3af54db9d28639: CC sparc64-softmmu/helper.o cc1: warnings being treated as errors /src/qemu/target-sparc/helper.c: In function 'get_physical_address': /src/qemu/target-sparc/helper.c:426: warning: 'context' may be used uninitialized in this function /src/qemu/target-sparc/helper.c:426: note: 'context' was declared here Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-05-06sparc64: handle asi referencing nucleus and secondary MMU contextsIgor V. Kovalenko
- increase max supported MMU modes to 6 - handle nucleus context asi - handle secondary context asi - handle non-faulting loads from secondary context Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-05-06sparc64: implement global translation table entries v1Igor V. Kovalenko
- match global tte against any context - show global tte in MMU dump v0->v1: added default case to switch statement in demap_tlb - should fix gcc warning about uninitialized context variable Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-04-26target-sparc: Fix -singlestep.Richard Henderson
Single-stepping was not properly updating npc, resulting in some instructions being executed twice. In addition, we were emitting dead code at the end of the TB. Fix both by teaching gen_goto_tb to avoid goto_tb for single-step and removing the special-case code in gen_intermediate_code_internal. Signed-off-by: Richard Henderson <rth@twiddle.net> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-04-23target-sparc: Fix address masking in ldqf and stqf.Richard Henderson
Use address_mask on both addr and addr+8 in both these routines, rather than explicit masking with 0xffffffff. Reformulate address_mask to return a result, rather than masking a pass-by-reference argument. Signed-off-by: Richard Henderson <rth@twiddle.net> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-04-18Fix harmless if statements with empty body, spotted by clangBlue Swirl
These clang errors are harmless but worth fixing: CC ppc-softmmu/usb-ohci.o /src/qemu/hw/usb-ohci.c:1104:59: error: if statement has empty body [-Wempty-body] ohci->ctrl_head, ohci->ctrl_cur); /src/qemu/hw/usb-ohci.c:1371:57: error: if statement has empty body [-Wempty-body] DPRINTF("usb-ohci: port %d: SUSPEND\n", portnum); CC sparc64-softmmu/translate.o /src/qemu/target-sparc/translate.c:3173:37: error: if statement has empty body [-Wempty-body] ; // XXX Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-04-17target-sparc: Free instruction temporaries.Richard Henderson
Rather than creating new temporaries for constants, use the ones created in disas_sparc_insn. Remember the temps created there so that they can be freed at the end of the function. Profile data collected by TCG while booting sparc-test kernel: -avg temps/TB 70.61 max=421 +avg temps/TB 62.75 max=66 Signed-off-by: Richard Henderson <rth@twiddle.net> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-04-17target-sparc: Fix TARGET_{PHYS,VIRT}_ADDR_SPACE_BITS.Richard Henderson
The 32 and 64-bit definitions were swapped in the ifdef. Signed-off-by: Richard Henderson <rth@twiddle.net> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-04-17Sparc: fix PC/NPC during FPU trapsBlue Swirl
All FPU instructions can trap, so save PC/NPC state before executing them. Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-04-11Sparc: fix exceptions in delay slotBlue Swirl
Fix a case where an exception happens with the instruction in the delay slot. Recovery of branch condition in the exception handling code was not converted to TCG. Because the condition was bogus, wrong NPC could be selected from the two candidates. A nice bug report with a test case can be found in: https://bugs.launchpad.net/qemu/+bug/551814 Fix based on patch by Fabrice Bellard. Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-04-08remove TARGET_* defines from translate-all.cPaolo Bonzini
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
2010-03-17Large page TLB flushPaul Brook
QEMU uses a fixed page size for the CPU TLB. If the guest uses large pages then we effectively split these into multiple smaller pages, and populate the corresponding TLB entries on demand. When the guest invalidates the TLB by virtual address we must invalidate all entries covered by the large page. However the address used to invalidate the entry may not be present in the QEMU TLB, so we do not know which regions to clear. Implementing a full vaiable size TLB is hard and slow, so just keep a simple address/mask pair to record which addresses may have been mapped by large pages. If the guest invalidates this region then flush the whole TLB. Signed-off-by: Paul Brook <paul@codesourcery.com>
2010-03-12Target specific usermode cleanupPaul Brook
Disable various target specific code that is only relevant to system emulation. Signed-off-by: Paul Brook <paul@codesourcery.com>
2010-03-12Remove cpu_get_phys_page_debug from userspace emulationPaul Brook
cpu_get_phys_page_debug makes no sense for userspace emulation, so remove it. Signed-off-by: Paul Brook <paul@codesourcery.com>
2010-03-12Move TARGET_PHYS_ADDR_SPACE_BITS to target-*/cpu.h.Richard Henderson
Removes a set of ifdefs from exec.c. Introduce TARGET_VIRT_ADDR_SPACE_BITS for all targets other than Alpha. This will be used for page_find_alloc, which is supposed to be using virtual addresses in the first place. Signed-off-by: Richard Henderson <rth@twiddle.net>
2010-02-25target-sparc: fix --enable-debug build for 64 bit hostStefan Weil
b551ec04ca45d1925417dd2ec7c1b7f115c84f1d fixed the compilation for 32 bit hosts, but introduced a new error for 64 bit hosts: tcg_temp_new_ptr needs a matching tcg_temp_free_ptr. Signed-off-by: Stefan Weil <weil@mail.berlios.de> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-02-20target-sparc: fix --enable-debug buildJay Foad
Use 32-bit arithmetic for the address offset calculation to fix a build failure on 32-bit hosts. Signed-off-by: Jay Foad <jay.foad@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-31sparc32 don't mark page dirty when failingArtyom Tarasenko
if the access check fails, the page can not be modified and shouldn't be marked dirty. The patch fixes the "hsfs_putpage: dirty HSFS page" error in Solaris guests. Signed-off-by: Artyom Tarasenko <atar4qemu@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-27sparc64: reimplement tick timers v4Igor V. Kovalenko
sparc64 timer has tick counter which can be set and read, and tick compare value used as deadline to fire timer interrupt. The timer is not used as periodic timer, instead deadline is set each time new timer interrupt is needed. v3 -> v4: - coding style v2 -> v3: - added missing timer debug output macro - CPUTimer struct and typedef moved to cpu.h - change CPU_SAVE_VERSION to 6, older save formats not supported v1 -> v2: - new conversion helpers cpu_to_timer_ticks and timer_to_cpu_ticks - save offset from clock source to implement cpu_tick_set_count - renamed struct sun4u_timer to CPUTimer - load and save cpu timers v0 -> v1: - coding style Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-27sparc64: correct write extra bits to cwpIgor V. Kovalenko
- correctly fit to cwp if provided window number is out of range Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-23sparc32 fix np dereference in do_unassigned_accessArtyom Tarasenko
fix a potential null pointer dereference introduced in commit 576c2cdc767ab9e2dc038fa4c99f22e53287a3de Signed-off-by: Artyom Tarasenko <atar4qemu@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-19kill regs_to_env and env_to_regsPaolo Bonzini
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
2010-01-17Sparc: improve CPU register dumpBlue Swirl
Common: * Remove unnecessary 0x prefix * Print %y * Fix NZVC flag print order to match CPU bit order Sparc64 specific: * Print registers without line wrapping * Print %f40-%f63 * Pretty print CCR flags * Print %fsr and %fprs in full precision * More consistent formatting Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-15sparc32 do_unassigned_access overhaul v2Artyom Tarasenko
According to pages 9-31 - 9-34 of "SuperSPARC & MultiCache Controller User's Manual": 1. "A lower priority fault may not overwrite the MFSR status of a higher priority fault." 2. The MFAR is overwritten according to the policy defined for the MFSR 3. The overwrite bit is asserted if the fault status register (MFSR) has been written more than once by faults of the same class 4. SuperSPARC will never place instruction fault addresses in the MFAR. Implementation of points 1-3 allows booting Solaris 2.6 and 2.5.1. v2: CODING_STYLE fixes Signed-off-by: Artyom Tarasenko <atar4qemu@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-13Sparc32: remove unused variable, spotted by clangBlue Swirl
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-08sparc64: interrupt trap handlingIgor V. Kovalenko
cpu_check_irqs - handle SOFTINT register TICK and STICK timer bits - only check interrupt levels greater than PIL value - handle preemption by higher level traps cpu_exec - handle CPU_INTERRUPT_HARD only if interrupts are enabled - PIL 15 is not special level on sparcv9 Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-08sparc64: move cpu_interrupts_enabled to cpu.hIgor V. Kovalenko
- to be used by cpu_check_irqs Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-08sparc64: add macros to deal with softint and timer interruptIgor V. Kovalenko
Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-08sparc64: check for pending irq when pil, pstate or softint is changedIgor V. Kovalenko
Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-08sparc64: use helper_wrpil to check pending irq on writeIgor V. Kovalenko
Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-08sparc64: add PIL to cpu state dumpIgor V. Kovalenko
Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-08sparc64: trace pstate and global register set changesIgor V. Kovalenko
Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-08sparc64: change_pstate should have 32bit argumentIgor V. Kovalenko
- pstate is 32bit variable, no need to pass 64bit value around Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-07Sparc32: clear exception_index with -1 valueBlue Swirl
See also 821b19fe923ac49a24cdb4af902584fdd019cee6. Spotted by Artyom Tarasenko and Igor Kovalenko. Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-06sparc64: clear exception_index with -1 valueIgor V. Kovalenko
Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-03pass env to raise_exception if called outside of op_helper codeIgor V. Kovalenko
- this fixes stepping with gdb, where do_unassigned_access may be called from gdb handler, outside of generated code Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-01-03sparc64: switch to MMU global registers in more MMU related trapsIgor V. Kovalenko
- extended range of MMU related traps which use MMU global registers, as listed in Ultrasparc-IIi document - no visible changes, since emulation do not cause added traps Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2009-12-05Sparc64: handle MMU global bit and nucleus contextBlue Swirl
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2009-12-04Sparc64: fix compilation with DEBUG_MMUBlue Swirl
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2009-11-07user: move CPU reset call to main.c for x86/PPC/SparcBlue Swirl
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2009-11-07sparc32 (mostly): remove unneeded calls to device resetBlue Swirl
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>