summaryrefslogtreecommitdiffstats
path: root/sys/i386/isa/npx.c
Commit message (Expand)AuthorAgeFilesLines
* MFC r320307:kib2017-06-271-1/+1
* MFC r320051:kib2017-06-241-3/+8
* MFC r313154:kib2017-02-241-79/+1
* MFC r313109:kib2017-02-091-6/+4
* Adjust i386 comment to match amd64 one after r294311.kib2016-01-191-5/+6
* Convert between abridged (from FXSAVE) and unabridged (from FSAVE)kib2015-07-101-6/+13
* Now that aesni won't reuse fpu contexts (D3016), add seatbelts to thejmg2015-07-081-2/+10
* Provide npx_get_fsave(9) and npx_set_fsave(9) functions to obtain andkib2015-06-291-0/+96
* Remove support for Xen PV domU kernels. Support for HVM domU kernelsjhb2015-04-301-9/+0
* Remove write-only variable.kib2015-03-121-2/+0
* Improve support for XSAVE with debuggers.jhb2014-11-211-0/+7
* Move all of the XSAVE support under #ifdef I686_CPU and fix the buildjhb2014-11-031-3/+34
* MFamd64: Add support for extended FPU states on i386. This includesjhb2014-11-021-158/+361
* MFamd64: Explicitly initialize the mxcsr during npxinit().jhb2014-11-021-0/+9
* Save and restore FPU state across suspend and resume. In earlier revisionsjhb2014-08-301-0/+37
* Add FPU_KERN_KTHR flag to fpu_kern_enter(9), which avoids saving FPUkib2014-06-231-0/+7
* Implement vector callback for PVHVM and unify event channel implementationsgibbs2013-08-291-1/+1
* Change (unused) prototype for stmxcsr() to match reality.kib2012-07-301-1/+1
* MFamd64 r238623:kib2012-07-261-15/+14
* MFCamd64 r238598:kib2012-07-211-2/+25
* MFamd64 r238668:kib2012-07-211-16/+13
* MFamd64 r238597:kib2012-07-211-0/+2
* Add a clts() wrapper around the 'clts' instruction to <machine/cpufunc.h>jhb2012-07-091-11/+8
* Add support for the extended FPU states on amd64, both for nativekib2012-01-211-1/+45
* In fpudna()/npxdna(), mark FPU context initialized and optionallykib2010-12-121-1/+3
* Remove npxgetregs(), npxsetregs(), fpugetregs() and fpusetregs()kib2010-11-261-72/+22
* Use unambiguous inline assembly to load a float variable. GNU asdim2010-11-251-1/+1
* Use 'saveintr' instead of 'savecrit' or 'eflags' to hold the state returnedjhb2010-10-251-3/+3
* Simplify fldcw() macro. There is no reason to use pointer here. No objectjkim2010-07-261-5/+5
* Remove an unused macro since r189418.jkim2010-07-261-1/+0
* Reduce diff against fenv.h:jkim2010-07-261-12/+14
* FNSTSW instruction can use AX register as an operand.jkim2010-07-261-1/+1
* After the FPU use requires #MF working due to INT13 FPU exception handlingkib2010-06-231-35/+27
* Remove the support for int13 FPU exception reporting on i386. It iskib2010-06-231-132/+31
* Remove unused i586 optimized bcopy/bzero/etc implementations that utilizekib2010-06-231-54/+1
* Introduce the x86 kernel interfaces to allow kernel code to usekib2010-06-051-19/+158
* Fix a few nits in the earlier changes to prevent local information leakagejhb2009-03-251-4/+8
* Rename (fpu|npx)_cleanstate to (fpu|npx)_initialstate to better reflectjhb2009-03-251-13/+13
* A better fix for handling different FPU initial control words for differentjhb2009-03-051-0/+11
* Some cleanups to the i386 FPU support:jhb2009-03-051-51/+34
* Change the functions to ANSI in those cases where it breaks promotionrdivacky2009-02-241-2/+1
* Some whitespace and style fixes.jhb2009-02-231-5/+4
* merge 186535, 186537, and 186538 from releng_7_xenkmacy2008-12-291-1/+1
* Call in to xen for fpu handling when XEN is setkmacy2008-08-151-1/+10
* Commit 14/14 of sched_lock decomposition.jeff2007-06-051-2/+2
* o break newbus api: add a new argument of type driver_filter_t topiso2007-02-231-4/+5
* Use the method described in IA-32 Intel Architecture Software Developer'sdavidxu2006-05-301-0/+9
* Backout changes trying to inherit floating-point environment, althoughdavidxu2006-05-291-35/+0
* If parent thread never used FPU, the only work is to clear flagdavidxu2006-05-281-1/+6
* When creating a new thread, inherit floating-point environment fromdavidxu2006-05-281-0/+30
OpenPOWER on IntegriCloud