| Message ID | 20240813142119.29012-3-alejandro.vallejo@cloud.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | x86: FPU handling cleanup | expand |
On 13.08.2024 16:21, Alejandro Vallejo wrote: > It was trying to do too many things at once and there was no clear way of > defining what it was meant to do. This commit splits the function in three. > > 1. A function to return the FPU to power-on reset values. > 2. A function to return the FPU to default values. > 3. A x87/SSE state loader (equivalent to the old function when it took a data > pointer). > > While at it, make sure the abridged tag is consistent with the manuals and > start as 0xFF. > > Signed-off-by: Alejandro Vallejo <alejandro.vallejo@cloud.com> Reviewed-by: Jan Beulich <jbeulich@suse.com> > --- > v3: > * Adjust commit message, as the split is now in 3. > * Remove bulky comment, as the rationale for it turned out to be > unsubstantiated. I can't find proof in xen-devel of the stream > operating the way I claimed, and at that point having the comment > at all is pointless So you deliberately removed the comment altogether, not just point 3 of it? Jan
On Tue Aug 13, 2024 at 3:32 PM BST, Jan Beulich wrote: > On 13.08.2024 16:21, Alejandro Vallejo wrote: > > It was trying to do too many things at once and there was no clear way of > > defining what it was meant to do. This commit splits the function in three. > > > > 1. A function to return the FPU to power-on reset values. > > 2. A function to return the FPU to default values. > > 3. A x87/SSE state loader (equivalent to the old function when it took a data > > pointer). > > > > While at it, make sure the abridged tag is consistent with the manuals and > > start as 0xFF. > > > > Signed-off-by: Alejandro Vallejo <alejandro.vallejo@cloud.com> > > Reviewed-by: Jan Beulich <jbeulich@suse.com> > > > --- > > v3: > > * Adjust commit message, as the split is now in 3. > > * Remove bulky comment, as the rationale for it turned out to be > > unsubstantiated. I can't find proof in xen-devel of the stream > > operating the way I claimed, and at that point having the comment > > at all is pointless > > So you deliberately removed the comment altogether, not just point 3 of it? > > Jan Yes. The other two cases can be deduced pretty trivially from the conditional, I reckon. I commented them more heavily in order to properly introduce (3), but seeing how it was all a midsummer dream might as well reduce clutter. I got as far as the original implementation of XSAVE in Xen and it seems to have been tested against many combinations of src and dst, none of which was that ficticious "xsave enabled + xsave context missing". I suspect the xsave_enabled(v) was merely avoiding writing to the XSAVE buffer just for efficiency (however minor effect it might have had). I just reverse engineering it wrong. Which reminds me. Thanks for mentioning that, because it was really just guesswork on my part. Cheers, Alejandro
Hi, On Tue Aug 13, 2024 at 5:33 PM BST, Alejandro Vallejo wrote: > On Tue Aug 13, 2024 at 3:32 PM BST, Jan Beulich wrote: > > On 13.08.2024 16:21, Alejandro Vallejo wrote: > > > It was trying to do too many things at once and there was no clear way of > > > defining what it was meant to do. This commit splits the function in three. > > > > > > 1. A function to return the FPU to power-on reset values. > > > 2. A function to return the FPU to default values. > > > 3. A x87/SSE state loader (equivalent to the old function when it took a data > > > pointer). > > > > > > While at it, make sure the abridged tag is consistent with the manuals and > > > start as 0xFF. > > > > > > Signed-off-by: Alejandro Vallejo <alejandro.vallejo@cloud.com> > > > > Reviewed-by: Jan Beulich <jbeulich@suse.com> > > > > > --- > > > v3: > > > * Adjust commit message, as the split is now in 3. > > > * Remove bulky comment, as the rationale for it turned out to be > > > unsubstantiated. I can't find proof in xen-devel of the stream > > > operating the way I claimed, and at that point having the comment > > > at all is pointless > > > > So you deliberately removed the comment altogether, not just point 3 of it? > > > > Jan > > Yes. The other two cases can be deduced pretty trivially from the conditional, > I reckon. I commented them more heavily in order to properly introduce (3), but > seeing how it was all a midsummer dream might as well reduce clutter. > > I got as far as the original implementation of XSAVE in Xen and it seems to > have been tested against many combinations of src and dst, none of which was > that ficticious "xsave enabled + xsave context missing". I suspect the > xsave_enabled(v) was merely avoiding writing to the XSAVE buffer just for > efficiency (however minor effect it might have had). I just reverse engineering > it wrong. > > Which reminds me. Thanks for mentioning that, because it was really just > guesswork on my part. > > Cheers, > Alejandro Playing around with the FPU I noticed this patch wasn't committed, did it fall under the cracks or is there a specific reason? Cheers, Alejandro
On 13/08/2024 3:21 pm, Alejandro Vallejo wrote: > It was trying to do too many things at once and there was no clear way of > defining what it was meant to do. This commit splits the function in three. > > 1. A function to return the FPU to power-on reset values. > 2. A function to return the FPU to default values. > 3. A x87/SSE state loader (equivalent to the old function when it took a data > pointer). > > While at it, make sure the abridged tag is consistent with the manuals and > start as 0xFF. > > Signed-off-by: Alejandro Vallejo <alejandro.vallejo@cloud.com> fpu_setup_fpu() definitely needs splitting, and it was doing many things before, but... > diff --git a/xen/arch/x86/include/asm/i387.h b/xen/arch/x86/include/asm/i387.h > index a783549db991..7a69577de45b 100644 > --- a/xen/arch/x86/include/asm/i387.h > +++ b/xen/arch/x86/include/asm/i387.h > @@ -31,10 +31,30 @@ void vcpu_restore_fpu_nonlazy(struct vcpu *v, bool need_stts); > void vcpu_restore_fpu_lazy(struct vcpu *v); > void vcpu_save_fpu(struct vcpu *v); > void save_fpu_enable(void); > - > int vcpu_init_fpu(struct vcpu *v); > -struct xsave_struct; > -void vcpu_setup_fpu(struct vcpu *v, struct xsave_struct *xsave_area, > - const void *data, unsigned int fcw_default); > void vcpu_destroy_fpu(struct vcpu *v); > + > +/* > + * Restore v's FPU to power-on reset values > + * > + * @param v vCPU containing the FPU > + */ > +void vcpu_reset_fpu(struct vcpu *v); > + > +/* > + * Restore v's FPU to default values > + * > + * @param v vCPU containing the FPU > + */ > +void vcpu_default_fpu(struct vcpu *v); ... whether this is an accurate transform of the logic or not, we oughtn't to have both. AFAICT, these two functions differ only in the choice of FCW constant, and whether FTW gets a nonzero value or not. The x86 architecture has #RESET and #INIT states (and Xen is especially bad at these right now). They're well defined in the SDM/APM; #RESET zeroes most things but leaves FCW=0x40 and FTW=0x5555 (for which the FXSAVE mapping is 0xFF). #INIT leaves everything unmodified. [After a very long manual-diving session. Thankyou Christian] The 80287's #RESET and F(N)INIT instructions were broadly similar. They differed on whether they changed the addressing mode[1] but both cases set FCW=0x37f and FTW=0x5555. The 80387 intentionally diverged #RESET and FNINIT, with #RESET setting FCW=0x40. This had a side effect of asserting #ERROR, and software is required to execute FNINIT which sets FCW=0x37f and de-asserts #ERROR Why? From the 387's programmers reference: ---%<--- 6.2.2 Hardware Recognition of the NPX The 80386 identifies the type of its coprocessor (80287 or 80387) by sampling its ERROR# input some time after the falling edge of RESET and before executing the first ESC instruction. The 80287 keeps its ERROR# output in inactive state after hardware reset; the 80387 keeps its ERROR# output in active state after hardware reset. The 80386 records this difference in the ET bit of control register zero (CR0). The 80386 subsequently uses ET to control its interface with the coprocessor. If ET is set, it employs the 32-bit protocol of the 80387; if ET is not set, it employs the 16-bit protocol of the 80287. <snip> 6.2.5 Initializing the 80387 Initializing the 80387 NPX simply means placing the NPX in a known state unaffected by any activity performed earlier. A single FNINIT instruction performs this initialization. All the error masks are set, all registers are tagged empty, TOP is set to zero, and default rounding and precision controls are set. Table 6-1 shows the state of the 80387 NPX following FINIT or FNINIT. This state is compatible with that of the 80287 after FINIT or after hardware RESET. ---%<--- The 486 gets even more complicated, but I've been writing this email for long enough. So, 0x40 is the correct reset value for the 386/387 and later, hence why that's what the SDM/APM say these days. As for Xen. For HVM guests, setting FCW=0x37f is definitely wrong. For PV guests, things are ill-defined, but software has been required to issue FNINIT for 4 decades already (even MiniOS does!) so I find myself dis-interested in trying to maintain compatibility for coprocessor which predates Xen by ~15y. You're already playing with FTW too (vs the old logic), so please drop vcpu_default_fpu() and just use vcpu_reset_fpu() instead. As a bonus, it lets you delete yet-more code, but it also needs to come with a rework of the commit message. I'd aim for less "split in 3" and more "delete and write something sane". > + > +/* > + * Load x87/SSE state into v's FPU > + * > + * Overrides the XSAVE header to set the state components to be x87 and SSE. > + * > + * @param v vCPU containing the FPU > + * @param data 512-octet blob for x87/SSE state > + */ > +void vcpu_setup_fpu(struct vcpu *v, const void *data); > #endif /* __ASM_I386_I387_H */ > diff --git a/xen/arch/x86/include/asm/xstate.h b/xen/arch/x86/include/asm/xstate.h > index ebeb2a3dcaf9..6144ed6f8551 100644 > --- a/xen/arch/x86/include/asm/xstate.h > +++ b/xen/arch/x86/include/asm/xstate.h > @@ -14,6 +14,7 @@ > > #define FCW_DEFAULT 0x037f > #define FCW_RESET 0x0040 > +#define FTW_RESET 0xFF This isn't a tag word reset value. The value for that would be 0x5555. I'd suggest: #define FXSAVE_FTW_RESET 0xff /* Abridged Tag Word format */ for want of anything better. At least this name makes it clear it's specifically for the FXSAVE format. ~Andrew [1] the 286, (in)famously couldn't leave protected mode, and neither could the 287 it seems.
On 03.10.2024 15:54, Alejandro Vallejo wrote: > On Tue Aug 13, 2024 at 5:33 PM BST, Alejandro Vallejo wrote: >> On Tue Aug 13, 2024 at 3:32 PM BST, Jan Beulich wrote: >>> On 13.08.2024 16:21, Alejandro Vallejo wrote: >>>> It was trying to do too many things at once and there was no clear way of >>>> defining what it was meant to do. This commit splits the function in three. >>>> >>>> 1. A function to return the FPU to power-on reset values. >>>> 2. A function to return the FPU to default values. >>>> 3. A x87/SSE state loader (equivalent to the old function when it took a data >>>> pointer). >>>> >>>> While at it, make sure the abridged tag is consistent with the manuals and >>>> start as 0xFF. >>>> >>>> Signed-off-by: Alejandro Vallejo <alejandro.vallejo@cloud.com> >>> >>> Reviewed-by: Jan Beulich <jbeulich@suse.com> >>> >>>> --- >>>> v3: >>>> * Adjust commit message, as the split is now in 3. >>>> * Remove bulky comment, as the rationale for it turned out to be >>>> unsubstantiated. I can't find proof in xen-devel of the stream >>>> operating the way I claimed, and at that point having the comment >>>> at all is pointless >>> >>> So you deliberately removed the comment altogether, not just point 3 of it? >>> >>> Jan >> >> Yes. The other two cases can be deduced pretty trivially from the conditional, >> I reckon. I commented them more heavily in order to properly introduce (3), but >> seeing how it was all a midsummer dream might as well reduce clutter. >> >> I got as far as the original implementation of XSAVE in Xen and it seems to >> have been tested against many combinations of src and dst, none of which was >> that ficticious "xsave enabled + xsave context missing". I suspect the >> xsave_enabled(v) was merely avoiding writing to the XSAVE buffer just for >> efficiency (however minor effect it might have had). I just reverse engineering >> it wrong. >> >> Which reminds me. Thanks for mentioning that, because it was really just >> guesswork on my part. >> >> Cheers, >> Alejandro > > Playing around with the FPU I noticed this patch wasn't committed, did it fall > under the cracks or is there a specific reason? Well, it's patch 2 in a series with no statement that it's independent of patch 1, and patch 1 continues to lack an ack (based on earlier comments of mine you probably have inferred that I'm not intending to ack it in this shape, while at the same time - considering the arguments you gave - I also don't mean to stand in the way of it going in with someone else's ack). Jan
Hi, On Fri Oct 4, 2024 at 7:08 AM BST, Jan Beulich wrote: > On 03.10.2024 15:54, Alejandro Vallejo wrote: > > On Tue Aug 13, 2024 at 5:33 PM BST, Alejandro Vallejo wrote: > >> On Tue Aug 13, 2024 at 3:32 PM BST, Jan Beulich wrote: > >>> On 13.08.2024 16:21, Alejandro Vallejo wrote: > >>>> It was trying to do too many things at once and there was no clear way of > >>>> defining what it was meant to do. This commit splits the function in three. > >>>> > >>>> 1. A function to return the FPU to power-on reset values. > >>>> 2. A function to return the FPU to default values. > >>>> 3. A x87/SSE state loader (equivalent to the old function when it took a data > >>>> pointer). > >>>> > >>>> While at it, make sure the abridged tag is consistent with the manuals and > >>>> start as 0xFF. > >>>> > >>>> Signed-off-by: Alejandro Vallejo <alejandro.vallejo@cloud.com> > >>> > >>> Reviewed-by: Jan Beulich <jbeulich@suse.com> > >>> > >>>> --- > >>>> v3: > >>>> * Adjust commit message, as the split is now in 3. > >>>> * Remove bulky comment, as the rationale for it turned out to be > >>>> unsubstantiated. I can't find proof in xen-devel of the stream > >>>> operating the way I claimed, and at that point having the comment > >>>> at all is pointless > >>> > >>> So you deliberately removed the comment altogether, not just point 3 of it? > >>> > >>> Jan > >> > >> Yes. The other two cases can be deduced pretty trivially from the conditional, > >> I reckon. I commented them more heavily in order to properly introduce (3), but > >> seeing how it was all a midsummer dream might as well reduce clutter. > >> > >> I got as far as the original implementation of XSAVE in Xen and it seems to > >> have been tested against many combinations of src and dst, none of which was > >> that ficticious "xsave enabled + xsave context missing". I suspect the > >> xsave_enabled(v) was merely avoiding writing to the XSAVE buffer just for > >> efficiency (however minor effect it might have had). I just reverse engineering > >> it wrong. > >> > >> Which reminds me. Thanks for mentioning that, because it was really just > >> guesswork on my part. > >> > >> Cheers, > >> Alejandro > > > > Playing around with the FPU I noticed this patch wasn't committed, did it fall > > under the cracks or is there a specific reason? > > Well, it's patch 2 in a series with no statement that it's independent of patch I meant the series as a whole, rather than this specific patch. They are indeed not independent. > 1, and patch 1 continues to lack an ack (based on earlier comments of mine you > probably have inferred that I'm not intending to ack it in this shape, while at > the same time - considering the arguments you gave - I also don't mean to stand > in the way of it going in with someone else's ack). I didn't infer that at all, I'm afraid. I merely thought you had been busy and forgot about it. Is the "in this shape" about the overallocation that you mentioned in v1? > > Jan Cheers, Alejandro
On 07.10.2024 17:59, Alejandro Vallejo wrote: > On Fri Oct 4, 2024 at 7:08 AM BST, Jan Beulich wrote: >> On 03.10.2024 15:54, Alejandro Vallejo wrote: >>> On Tue Aug 13, 2024 at 5:33 PM BST, Alejandro Vallejo wrote: >>>> On Tue Aug 13, 2024 at 3:32 PM BST, Jan Beulich wrote: >>>>> On 13.08.2024 16:21, Alejandro Vallejo wrote: >>>>>> It was trying to do too many things at once and there was no clear way of >>>>>> defining what it was meant to do. This commit splits the function in three. >>>>>> >>>>>> 1. A function to return the FPU to power-on reset values. >>>>>> 2. A function to return the FPU to default values. >>>>>> 3. A x87/SSE state loader (equivalent to the old function when it took a data >>>>>> pointer). >>>>>> >>>>>> While at it, make sure the abridged tag is consistent with the manuals and >>>>>> start as 0xFF. >>>>>> >>>>>> Signed-off-by: Alejandro Vallejo <alejandro.vallejo@cloud.com> >>>>> >>>>> Reviewed-by: Jan Beulich <jbeulich@suse.com> >>>>> >>>>>> --- >>>>>> v3: >>>>>> * Adjust commit message, as the split is now in 3. >>>>>> * Remove bulky comment, as the rationale for it turned out to be >>>>>> unsubstantiated. I can't find proof in xen-devel of the stream >>>>>> operating the way I claimed, and at that point having the comment >>>>>> at all is pointless >>>>> >>>>> So you deliberately removed the comment altogether, not just point 3 of it? >>>>> >>>>> Jan >>>> >>>> Yes. The other two cases can be deduced pretty trivially from the conditional, >>>> I reckon. I commented them more heavily in order to properly introduce (3), but >>>> seeing how it was all a midsummer dream might as well reduce clutter. >>>> >>>> I got as far as the original implementation of XSAVE in Xen and it seems to >>>> have been tested against many combinations of src and dst, none of which was >>>> that ficticious "xsave enabled + xsave context missing". I suspect the >>>> xsave_enabled(v) was merely avoiding writing to the XSAVE buffer just for >>>> efficiency (however minor effect it might have had). I just reverse engineering >>>> it wrong. >>>> >>>> Which reminds me. Thanks for mentioning that, because it was really just >>>> guesswork on my part. >>>> >>>> Cheers, >>>> Alejandro >>> >>> Playing around with the FPU I noticed this patch wasn't committed, did it fall >>> under the cracks or is there a specific reason? >> >> Well, it's patch 2 in a series with no statement that it's independent of patch > > I meant the series as a whole, rather than this specific patch. They are indeed > not independent. > >> 1, and patch 1 continues to lack an ack (based on earlier comments of mine you >> probably have inferred that I'm not intending to ack it in this shape, while at >> the same time - considering the arguments you gave - I also don't mean to stand >> in the way of it going in with someone else's ack). > > I didn't infer that at all, I'm afraid. I merely thought you had been busy and > forgot about it. Is the "in this shape" about the overallocation that you > mentioned in v1? Yes. Jan
diff --git a/xen/arch/x86/domain.c b/xen/arch/x86/domain.c index d977ec71ca20..5af9e3e7a8b4 100644 --- a/xen/arch/x86/domain.c +++ b/xen/arch/x86/domain.c @@ -1186,9 +1186,10 @@ int arch_set_info_guest( is_pv_64bit_domain(d) ) v->arch.flags &= ~TF_kernel_mode; - vcpu_setup_fpu(v, v->arch.xsave_area, - flags & VGCF_I387_VALID ? &c.nat->fpu_ctxt : NULL, - FCW_DEFAULT); + if ( flags & VGCF_I387_VALID ) + vcpu_setup_fpu(v, &c.nat->fpu_ctxt); + else + vcpu_default_fpu(v); if ( !compat ) { diff --git a/xen/arch/x86/hvm/hvm.c b/xen/arch/x86/hvm/hvm.c index 76bbb645b77a..95d66e68a849 100644 --- a/xen/arch/x86/hvm/hvm.c +++ b/xen/arch/x86/hvm/hvm.c @@ -1165,10 +1165,10 @@ static int cf_check hvm_load_cpu_ctxt(struct domain *d, hvm_domain_context_t *h) seg.attr = ctxt.ldtr_arbytes; hvm_set_segment_register(v, x86_seg_ldtr, &seg); - /* Cover xsave-absent save file restoration on xsave-capable host. */ - vcpu_setup_fpu(v, xsave_enabled(v) ? NULL : v->arch.xsave_area, - ctxt.flags & XEN_X86_FPU_INITIALISED ? ctxt.fpu_regs : NULL, - FCW_RESET); + if ( ctxt.flags & XEN_X86_FPU_INITIALISED ) + vcpu_setup_fpu(v, &ctxt.fpu_regs); + else + vcpu_reset_fpu(v); v->arch.user_regs.rax = ctxt.rax; v->arch.user_regs.rbx = ctxt.rbx; @@ -4008,9 +4008,7 @@ void hvm_vcpu_reset_state(struct vcpu *v, uint16_t cs, uint16_t ip) v->arch.guest_table = pagetable_null(); } - if ( v->arch.xsave_area ) - v->arch.xsave_area->xsave_hdr.xstate_bv = 0; - vcpu_setup_fpu(v, v->arch.xsave_area, NULL, FCW_RESET); + vcpu_reset_fpu(v); arch_vcpu_regs_init(v); v->arch.user_regs.rip = ip; diff --git a/xen/arch/x86/i387.c b/xen/arch/x86/i387.c index fbb9d3584a3d..f7a9dcd162ba 100644 --- a/xen/arch/x86/i387.c +++ b/xen/arch/x86/i387.c @@ -303,41 +303,37 @@ int vcpu_init_fpu(struct vcpu *v) return xstate_alloc_save_area(v); } -void vcpu_setup_fpu(struct vcpu *v, struct xsave_struct *xsave_area, - const void *data, unsigned int fcw_default) +void vcpu_reset_fpu(struct vcpu *v) { - fpusse_t *fpu_sse = &v->arch.xsave_area->fpu_sse; - - ASSERT(!xsave_area || xsave_area == v->arch.xsave_area); - - v->fpu_initialised = !!data; + v->fpu_initialised = false; + *v->arch.xsave_area = (struct xsave_struct) { + .fpu_sse = { + .mxcsr = MXCSR_DEFAULT, + .fcw = FCW_RESET, + .ftw = FTW_RESET, + }, + .xsave_hdr.xstate_bv = X86_XCR0_X87, + }; +} - if ( data ) - { - memcpy(fpu_sse, data, sizeof(*fpu_sse)); - if ( xsave_area ) - xsave_area->xsave_hdr.xstate_bv = XSTATE_FP_SSE; - } - else if ( xsave_area && fcw_default == FCW_DEFAULT ) - { - xsave_area->xsave_hdr.xstate_bv = 0; - fpu_sse->mxcsr = MXCSR_DEFAULT; - } - else - { - memset(fpu_sse, 0, sizeof(*fpu_sse)); - fpu_sse->fcw = fcw_default; - fpu_sse->mxcsr = MXCSR_DEFAULT; - if ( v->arch.xsave_area ) - { - v->arch.xsave_area->xsave_hdr.xstate_bv &= ~XSTATE_FP_SSE; - if ( fcw_default != FCW_DEFAULT ) - v->arch.xsave_area->xsave_hdr.xstate_bv |= X86_XCR0_X87; - } - } +void vcpu_default_fpu(struct vcpu *v) +{ + v->fpu_initialised = false; + *v->arch.xsave_area = (struct xsave_struct) { + .fpu_sse = { + .mxcsr = MXCSR_DEFAULT, + .fcw = FCW_DEFAULT, + }, + }; +} - if ( xsave_area ) - xsave_area->xsave_hdr.xcomp_bv = 0; +void vcpu_setup_fpu(struct vcpu *v, const void *data) +{ + v->fpu_initialised = true; + *v->arch.xsave_area = (struct xsave_struct) { + .fpu_sse = *(const fpusse_t*)data, + .xsave_hdr.xstate_bv = XSTATE_FP_SSE, + }; } /* Free FPU's context save area */ diff --git a/xen/arch/x86/include/asm/i387.h b/xen/arch/x86/include/asm/i387.h index a783549db991..7a69577de45b 100644 --- a/xen/arch/x86/include/asm/i387.h +++ b/xen/arch/x86/include/asm/i387.h @@ -31,10 +31,30 @@ void vcpu_restore_fpu_nonlazy(struct vcpu *v, bool need_stts); void vcpu_restore_fpu_lazy(struct vcpu *v); void vcpu_save_fpu(struct vcpu *v); void save_fpu_enable(void); - int vcpu_init_fpu(struct vcpu *v); -struct xsave_struct; -void vcpu_setup_fpu(struct vcpu *v, struct xsave_struct *xsave_area, - const void *data, unsigned int fcw_default); void vcpu_destroy_fpu(struct vcpu *v); + +/* + * Restore v's FPU to power-on reset values + * + * @param v vCPU containing the FPU + */ +void vcpu_reset_fpu(struct vcpu *v); + +/* + * Restore v's FPU to default values + * + * @param v vCPU containing the FPU + */ +void vcpu_default_fpu(struct vcpu *v); + +/* + * Load x87/SSE state into v's FPU + * + * Overrides the XSAVE header to set the state components to be x87 and SSE. + * + * @param v vCPU containing the FPU + * @param data 512-octet blob for x87/SSE state + */ +void vcpu_setup_fpu(struct vcpu *v, const void *data); #endif /* __ASM_I386_I387_H */ diff --git a/xen/arch/x86/include/asm/xstate.h b/xen/arch/x86/include/asm/xstate.h index ebeb2a3dcaf9..6144ed6f8551 100644 --- a/xen/arch/x86/include/asm/xstate.h +++ b/xen/arch/x86/include/asm/xstate.h @@ -14,6 +14,7 @@ #define FCW_DEFAULT 0x037f #define FCW_RESET 0x0040 +#define FTW_RESET 0xFF #define MXCSR_DEFAULT 0x1f80 extern uint32_t mxcsr_mask;
It was trying to do too many things at once and there was no clear way of defining what it was meant to do. This commit splits the function in three. 1. A function to return the FPU to power-on reset values. 2. A function to return the FPU to default values. 3. A x87/SSE state loader (equivalent to the old function when it took a data pointer). While at it, make sure the abridged tag is consistent with the manuals and start as 0xFF. Signed-off-by: Alejandro Vallejo <alejandro.vallejo@cloud.com> --- v3: * Adjust commit message, as the split is now in 3. * Remove bulky comment, as the rationale for it turned out to be unsubstantiated. I can't find proof in xen-devel of the stream operating the way I claimed, and at that point having the comment at all is pointless I suspect the rationale for xsave_vcpu(v) was merely to skip writing the XSAVE header when it would be rewritten later on. Whatever it might be the current logic does the right thing and is several orders of magnitude clearer about its objective and its intent. --- xen/arch/x86/domain.c | 7 ++-- xen/arch/x86/hvm/hvm.c | 12 +++---- xen/arch/x86/i387.c | 60 +++++++++++++++---------------- xen/arch/x86/include/asm/i387.h | 28 ++++++++++++--- xen/arch/x86/include/asm/xstate.h | 1 + 5 files changed, 62 insertions(+), 46 deletions(-)