| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | Expose doreti as a global symbol on amd64 and i386. | jhb | 2016-04-13 | 1 | -0/+1 |
| * | [amd64] dtrace_invop handler is to be called only for kernel exceptions | avg | 2016-04-12 | 1 | -0/+2 |
| * | Remove an unneeded instruction. | markj | 2015-08-28 | 1 | -1/+0 |
| * | The iret instruction may generate #np and #ss fault, besides #gp. | kib | 2014-12-16 | 1 | -2/+7 |
| * | Use a C wrapper for trap() instead of checking and calling the DTrace trap | markj | 2014-07-19 | 1 | -18/+1 |
| * | Invoke the DTrace trap handler before calling trap() on amd64. This matches | markj | 2014-07-14 | 1 | -0/+17 |
| * | When usermode loaded non-default segment selector into the %gs, | kib | 2014-05-29 | 1 | -3/+20 |
| * | - For kernel compiled only with KDTRACE_HOOKS and not any lock debugging | attilio | 2013-11-25 | 1 | -1/+0 |
| * | Some clarifications and updates for the comments, mostly retrieved | kib | 2013-06-19 | 1 | -12/+16 |
| * | When handling an exception from the attempt from loading the faulting | kib | 2013-05-27 | 1 | -0/+24 |
| * | Microoptimize the return path for the fast syscalls on amd64. Arrange | kib | 2011-09-15 | 1 | -10/+10 |
| * | The jump target shall be after the padding, not into it. | kib | 2011-09-11 | 1 | -1/+2 |
| * | Perform amd64-specific microoptimizations for native syscall entry | kib | 2011-09-11 | 1 | -8/+6 |
| * | Disable local interrupts before testing the PCB_FULL_IRET flag. | kib | 2011-04-08 | 1 | -1/+2 |
| * | Fix a bug in the previous change to restore the fast path for syscall | jhb | 2011-04-08 | 1 | -6/+5 |
| * | pcb_flags is an int, so use testl rather than testq. | jhb | 2011-04-07 | 1 | -1/+1 |
| * | If a system call does not request a full interrupt return, use a fast | jhb | 2011-04-07 | 1 | -0/+33 |
| * | Increase size of pcb_flags to four bytes. | jkim | 2010-12-22 | 1 | -6/+6 |
| * | Improve PCB flags handling and make it more robust. Add two new functions | jkim | 2010-12-22 | 1 | -7/+7 |
| * | Retire write-only PCB_FULLCTX pcb flag on amd64. | kib | 2010-12-07 | 1 | -1/+0 |
| * | Register an interrupt vector for DTrace return probes. There is some | rpaulo | 2010-08-28 | 1 | -0/+4 |
| * | Clear DF bit in eflags/rflags on the kernel entry. The i386 and amd64 | kib | 2010-06-23 | 1 | -0/+4 |
| * | Route all returns from the interrupts and faults through the doreti_iret | kib | 2010-05-12 | 1 | -1/+1 |
| * | Style and comment adjustements. | kib | 2010-05-03 | 1 | -34/+45 |
| * | ld_gs_base is executing with stack containing only the frame, | kib | 2010-04-14 | 1 | -1/+0 |
| * | Handle a case when non-canonical address is loaded into the fsbase or | kib | 2010-04-10 | 1 | -2/+28 |
| * | Provide groundwork for 32-bit binary compatibility on non-x86 platforms, | nwhitehorn | 2010-03-11 | 1 | -1/+1 |
| * | Introduce the new kernel sub-tree x86 which should contain all the code | attilio | 2010-02-25 | 1 | -1/+1 |
| * | When amd64 CPU cannot load segment descriptor during trap return to | kib | 2009-07-10 | 1 | -12/+12 |
| * | Restore the segment registers and segment base MSRs for amd64 syscall | kib | 2009-07-09 | 1 | -9/+19 |
| * | Save and restore segment registers on amd64 when entering and leaving | kib | 2009-04-01 | 1 | -50/+147 |
| * | Improve robustness of NMI handling, for NMIs recognized in kernel | jkoshy | 2009-02-03 | 1 | -25/+52 |
| * | Disable interrupts, if they were enabled, before doing swapgs. | kib | 2009-01-14 | 1 | -0/+1 |
| * | Bug fix: %ebx needs to be preserved in the user callchain capture | jkoshy | 2008-12-14 | 1 | -4/+4 |
| * | - Bug fix: prevent a thread from migrating between CPUs between the | jkoshy | 2008-12-13 | 1 | -6/+10 |
| * | The doreti_iret_fault code is always called with gs base MSR containing | kib | 2008-08-18 | 1 | -6/+3 |
| * | Add the DTrace hooks for exception handling (Function boundary trace | jb | 2008-05-24 | 1 | -0/+40 |
| * | Kernel and hwpmc(4) support for callchain capture. | jkoshy | 2007-12-07 | 1 | -0/+69 |
| * | Add support for cross double fault frames in stack traces: | jhb | 2007-11-15 | 1 | -1/+19 |
| * | Fix a panic when rebooting a SMP machine, when option STOP_NMI is used, | davidxu | 2006-12-23 | 1 | -0/+1 |
| * | Newer versions of gcc don't support treating structures passed by value | kmacy | 2006-12-17 | 1 | -0/+2 |
| * | - Special-case NMI handling on the AMD64. | jkoshy | 2005-08-27 | 1 | -2/+76 |
| * | NMI handler should not enable interrupts. | ups | 2005-08-25 | 1 | -1/+1 |
| * | Add a special-case handler for general protection faults. It appears to | peter | 2005-06-30 | 1 | -3/+26 |
| * | Eliminate a source of 'trap xx with interrupts disabled'. I was jumping to | peter | 2005-06-24 | 1 | -1/+1 |
| * | Fix some of the problems Bruce observed with this code. | peter | 2005-05-22 | 1 | -2/+3 |
| * | For non-profiling kernels, there were two symbols assigned to the same | peter | 2005-05-21 | 1 | -2/+4 |
| * | MFi386: whitespace, copyright header, etc updates | peter | 2005-01-21 | 1 | -1/+1 |
| * | Complete 'IA32' -> 'COMPAT_IA32' change for the Linuxulator32. | obrien | 2004-08-16 | 1 | -2/+2 |
| * | Oops, ".align 4" for the data section in the previous commit should | bde | 2004-05-24 | 1 | -3/+3 |
