[PULL 076/153] i386/cpu: Make ExtSaveArea store an array of dependencies

Paolo Bonzini posted 153 patches 1 week, 4 days ago
Maintainers: "Alex Bennée" <alex.bennee@linaro.org>, "Philippe Mathieu-Daudé" <philmd@linaro.org>, Thomas Huth <thuth@redhat.com>, Yonggang Luo <luoyonggang@gmail.com>, Richard Henderson <richard.henderson@linaro.org>, Paolo Bonzini <pbonzini@redhat.com>, Cameron Esfahani <dirty@apple.com>, Roman Bolshakov <rbolshakov@ddn.com>, Phil Dennis-Jordan <phil@philjordan.eu>, Mads Ynddal <mads@ynddal.dk>, Magnus Kulke <magnus.kulke@linux.microsoft.com>, Wei Liu <wei.liu@kernel.org>, Stefano Stabellini <sstabellini@kernel.org>, Anthony PERARD <anthony@xenproject.org>, Paul Durrant <paul@xen.org>, "Edgar E. Iglesias" <edgar.iglesias@gmail.com>, David Hildenbrand <david@kernel.org>, Igor Mammedov <imammedo@redhat.com>, Alistair Francis <alistair.francis@wdc.com>, Stefan Berger <stefanb@linux.vnet.ibm.com>, John Snow <jsnow@redhat.com>, Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>, Kevin Wolf <kwolf@redhat.com>, Hanna Reitz <hreitz@redhat.com>, Stefan Hajnoczi <stefanha@redhat.com>, Fam Zheng <fam@euphon.net>, Aarushi Mehta <mehta.aaru20@gmail.com>, Julia Suvorova <jusual@redhat.com>, Stefano Garzarella <sgarzare@redhat.com>, Stefan Weil <sw@weilnetz.de>, Samuel Thibault <samuel.thibault@ens-lyon.org>, "Marc-André Lureau" <marcandre.lureau@redhat.com>, Zhao Liu <zhao1.liu@intel.com>, Peter Maydell <peter.maydell@linaro.org>, Mauro Carvalho Chehab <mchehab+huawei@kernel.org>, "Daniel P. Berrangé" <berrange@redhat.com>, Luc Michel <luc@lmichel.fr>, Damien Hedde <damien.hedde@dahe.fr>, Peter Xu <peterx@redhat.com>, Fabiano Rosas <farosas@suse.de>, Markus Armbruster <armbru@redhat.com>, Eduardo Habkost <eduardo@habkost.net>, Ani Sinha <anisinha@redhat.com>, Christian Schoenebeck <qemu_oss@crudebyte.com>, Greg Kurz <groug@kaod.org>, "Michael S. Tsirkin" <mst@redhat.com>, Dongjiu Geng <gengdongjiu1@gmail.com>, Aurelien Jarno <aurelien@aurel32.net>, Marcel Apfelbaum <marcel.apfelbaum@gmail.com>, "Cédric Le Goater" <clg@kaod.org>, Steven Lee <steven_lee@aspeedtech.com>, Troy Lee <leetroy@gmail.com>, Jamin Lin <jamin_lin@aspeedtech.com>, Andrew Jeffery <andrew@codeconstruct.com.au>, Joel Stanley <joel@jms.id.au>, Tyrone Ting <kfting@nuvoton.com>, Hao Wu <wuhaotsh@google.com>, Beniamino Galvani <b.galvani@gmail.com>, Strahinja Jankovic <strahinja.p.jankovic@gmail.com>, Niek Linnenbank <nieklinnenbank@gmail.com>, Samuel Tardieu <sam@rfc1149.net>, Antony Pavlov <antonynpavlov@gmail.com>, Igor Mitsyanko <i.mitsyanko@gmail.com>, Jean-Christophe Dubois <jcd@tribudubois.net>, Andrey Smirnov <andrew.smirnov@gmail.com>, Bernhard Beschow <shentey@gmail.com>, Rob Herring <robh@kernel.org>, Subbaraya Sundeep <sundeep.lkml@gmail.com>, Jan Kiszka <jan.kiszka@web.de>, Felipe Balbi <balbi@kernel.org>, Radoslaw Biernacki <rad@semihalf.com>, Leif Lindholm <leif.lindholm@oss.qualcomm.com>, Eric Auger <eric.auger@redhat.com>, Alexandre Iooss <erdnaxe@crans.org>, Shannon Zhao <shannon.zhaosl@gmail.com>, Gerd Hoffmann <kraxel@redhat.com>, Laurent Vivier <laurent@vivier.eu>, Michael Rolnik <mrolnik@gmail.com>, Raphael Norwitz <raphael@enfabrica.net>, Helge Deller <deller@gmx.de>, "Clément Chigot" <chigot@adacore.com>, Frederic Konrad <konrad.frederic@yahoo.fr>, Alberto Garcia <berto@igalia.com>, Yoshinori Sato <yoshinori.sato@nifty.com>, Magnus Damm <magnus.damm@gmail.com>, Palmer Dabbelt <palmer@dabbelt.com>, Weiwei Li <liwei1518@gmail.com>, Daniel Henrique Barboza <dbarboza@ventanamicro.com>, Liu Zhiwei <zhiwei_liu@linux.alibaba.com>, Halil Pasic <pasic@linux.ibm.com>, Christian Borntraeger <borntraeger@linux.ibm.com>, Jason Herne <jjherne@linux.ibm.com>, Vijai Kumar K <vijai@behindbytes.com>, Nicholas Piggin <npiggin@gmail.com>, Harsh Prateek Bora <harshpb@linux.ibm.com>, "Collin L. Walling" <walling@linux.ibm.com>, Amit Shah <amit@kernel.org>, Yanan Wang <wangyanan55@huawei.com>, Riku Voipio <riku.voipio@iki.fi>, Paul Burton <paulburton@kernel.org>, Aleksandar Rikalo <arikalo@gmail.com>, Jonathan Cameron <jonathan.cameron@huawei.com>, Fan Ni <fan.ni@samsung.com>, Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>, "Hervé Poussineau" <hpoussin@reactos.org>, BALATON Zoltan <balaton@eik.bme.hu>, Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>, Dmitry Osipenko <dmitry.osipenko@collabora.com>, Dmitry Fleytman <dmitry.fleytman@gmail.com>, Ninad Palsule <ninad@linux.ibm.com>, Glenn Miles <milesg@linux.ibm.com>, "Maciej S. Szmigiero" <maciej.szmigiero@oracle.com>, Corey Minyard <cminyard@mvista.com>, Patrick Leis <venture@google.com>, Alejandro Jimenez <alejandro.j.jimenez@oracle.com>, Sairaj Kodilkar <sarunkod@amd.com>, Jason Wang <jasowang@redhat.com>, Yi Liu <yi.l.liu@intel.com>, "Clément Mathieu--Drif" <clement.mathieu--drif@eviden.com>, David Woodhouse <dwmw2@infradead.org>, Sergio Lopez <slp@redhat.com>, Alexander Graf <graf@amazon.com>, Dorjoy Chowdhury <dorjoychy111@gmail.com>, Song Gao <gaosong@loongson.cn>, Bibo Mao <maobibo@loongson.cn>, Jiaxun Yang <jiaxun.yang@flygoat.com>, Huacai Chen <chenhuacai@kernel.org>, Jia Liu <proljc@gmail.com>, Gautam Menghani <gautam@linux.ibm.com>, Aditya Gupta <adityag@linux.ibm.com>, Xiao Guangrong <xiaoguangrong.eric@gmail.com>, Alexander Bulekov <alxndr@bu.edu>, Bandan Das <bsd@redhat.com>, Darren Kenny <darren.kenny@oracle.com>, Qiuhao Li <Qiuhao.Li@outlook.com>, Artyom Tarasenko <atar4qemu@gmail.com>, Gustavo Romero <gustavo.romero@linaro.org>, Francisco Iglesias <francisco.iglesias@amd.com>, Pavel Pisa <pisa@cmp.felk.cvut.cz>, Vikram Garhwal <vikram.garhwal@bytedance.com>, Sriram Yagnaraman <sriram.yagnaraman@ericsson.com>, Max Filippov <jcmvbkbc@gmail.com>, Jiri Pirko <jiri@resnulli.us>, Sven Schnelle <svens@stackframe.org>, Stafford Horne <shorne@gmail.com>, Elena Ufimtseva <elena.ufimtseva@oracle.com>, Jagannathan Raman <jag.raman@oracle.com>, Chinmay Rath <rathc@linux.ibm.com>, Sai Pavan Boddu <sai.pavan.boddu@amd.com>, Ran Wang <wangran@bosc.ac.cn>, Anup Patel <anup.patel@wdc.com>, Eric Farman <farman@linux.ibm.com>, Matthew Rosato <mjrosato@linux.ibm.com>, Ilya Leoshkevich <iii@linux.ibm.com>, Tony Krowiak <akrowiak@linux.ibm.com>, Farhan Ali <alifm@linux.ibm.com>, Nina Schoetterl-Glausch <nsg@linux.ibm.com>, Jared Rossi <jrossi@linux.ibm.com>, Zhuoying Cai <zycai@linux.ibm.com>, Cornelia Huck <cohuck@redhat.com>, Hannes Reinecke <hare@suse.com>, Bin Meng <bmeng.cn@gmail.com>, Titus Rwantare <titusr@google.com>, Bastian Koppelmann <kbastian@mail.uni-paderborn.de>, Jeuk Kim <jeuk20.kim@samsung.com>, "Hongren (Zenithal) Zheng" <i@zenithal.me>, "Canokeys.org" <contact@canokeys.org>, John Levon <john.levon@nutanix.com>, Thanos Makatos <thanos.makatos@nutanix.com>, Alex Williamson <alex@shazbot.org>, Tomita Moeko <tomitamoeko@gmail.com>, Viresh Kumar <viresh.kumar@linaro.org>, Mathieu Poirier <mathieu.poirier@linaro.org>, Manos Pitsidianakis <manos.pitsidianakis@linaro.org>, "Gonglei (Arei)" <arei.gonglei@huawei.com>, Eric Blake <eblake@redhat.com>, Keith Busch <kbusch@kernel.org>, Klaus Jensen <its@irrelevant.dk>, Jesper Devantier <foss@defmacro.it>, Huai-Cheng Kuo <hchkuo@avery-design.com.tw>, Chris Browy <cbrowy@avery-design.com>, Zhenzhong Duan <zhenzhong.duan@intel.com>, "Dr. David Alan Gilbert" <dave@treblig.org>, Zhang Chen <zhangckid@gmail.com>, Li Zhijian <lizhijian@fujitsu.com>, Mahmoud Mandour <ma.mandourr@gmail.com>, Pierrick Bouvier <pierrick.bouvier@linaro.org>, Cleber Rosa <crosa@redhat.com>, Michael Roth <michael.roth@amd.com>, Maksim Davydov <davydov-max@yandex-team.ru>, David Gibson <david@gibson.dropbear.id.au>, Hyman Huang <yong.huang@smartx.com>, Brian Cain <brian.cain@oss.qualcomm.com>, Marcelo Tosatti <mtosatti@redhat.com>, Pedro Barbuda <pbarbuda@microsoft.com>, Mohamed Mediouni <mohamed@unpredictable.fr>, Coiby Xu <Coiby.Xu@gmail.com>
[PULL 076/153] i386/cpu: Make ExtSaveArea store an array of dependencies
Posted by Paolo Bonzini 1 week, 3 days ago
From: Zhao Liu <zhao1.liu@intel.com>

Some XSAVE components depend on multiple features. For example, Opmask/
ZMM_Hi256/Hi16_ZMM depend on avx512f OR avx10, and for CET (which will
be supported later), cet_u/cet_s will depend on shstk OR ibt.

Although previously there's the special check for the dependencies of
AVX512F OR AVX10 on their respective XSAVE components (in
cpuid_has_xsave_feature()), to make the code more general and avoid
adding more special cases, make ExtSaveArea store a features array
instead of a single feature, so that it can describe multiple
dependencies.

Tested-by: Farrah Chen <farrah.chen@intel.com>
Signed-off-by: Zhao Liu <zhao1.liu@intel.com>
Link: https://lore.kernel.org/r/20251211060801.3600039-5-zhao1.liu@intel.com
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 target/i386/cpu.h |  9 +++++-
 target/i386/cpu.c | 78 ++++++++++++++++++++++++++++++++++-------------
 2 files changed, 65 insertions(+), 22 deletions(-)

diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index a183394eca7..3d74afc5a8e 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -1769,9 +1769,16 @@ QEMU_BUILD_BUG_ON(sizeof(XSaveXTILECFG) != 0x40);
 QEMU_BUILD_BUG_ON(sizeof(XSaveXTILEDATA) != 0x2000);
 
 typedef struct ExtSaveArea {
-    uint32_t feature, bits;
     uint32_t offset, size;
     uint32_t ecx;
+    /*
+     * The dependencies in the array work as OR relationships, which
+     * means having just one of those features is enough.
+     *
+     * At most two features are sharing the same xsave area.
+     * Number of features can be adjusted if necessary.
+     */
+    const FeatureMask features[2];
 } ExtSaveArea;
 
 #define XSAVE_STATE_AREA_COUNT (XSTATE_XTILE_DATA_BIT + 1)
diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index de4e5c57746..367396a216e 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -2020,53 +2020,77 @@ static const X86RegisterInfo32 x86_reg_info_32[CPU_NB_REGS32] = {
 ExtSaveArea x86_ext_save_areas[XSAVE_STATE_AREA_COUNT] = {
     [XSTATE_FP_BIT] = {
         /* x87 FP state component is always enabled if XSAVE is supported */
-        .feature = FEAT_1_ECX, .bits = CPUID_EXT_XSAVE,
         .size = sizeof(X86LegacyXSaveArea) + sizeof(X86XSaveHeader),
+        .features = {
+            { FEAT_1_ECX,           CPUID_EXT_XSAVE },
+        },
     },
     [XSTATE_SSE_BIT] = {
         /* SSE state component is always enabled if XSAVE is supported */
-        .feature = FEAT_1_ECX, .bits = CPUID_EXT_XSAVE,
         .size = sizeof(X86LegacyXSaveArea) + sizeof(X86XSaveHeader),
+        .features = {
+            { FEAT_1_ECX,           CPUID_EXT_XSAVE },
+        },
     },
     [XSTATE_YMM_BIT] = {
-        .feature = FEAT_1_ECX, .bits = CPUID_EXT_AVX,
         .size = sizeof(XSaveAVX),
+        .features = {
+            { FEAT_1_ECX,           CPUID_EXT_AVX },
+        },
     },
     [XSTATE_BNDREGS_BIT] = {
-        .feature = FEAT_7_0_EBX, .bits = CPUID_7_0_EBX_MPX,
         .size = sizeof(XSaveBNDREG),
+        .features = {
+            { FEAT_7_0_EBX,         CPUID_7_0_EBX_MPX },
+        },
     },
     [XSTATE_BNDCSR_BIT] = {
-        .feature = FEAT_7_0_EBX, .bits = CPUID_7_0_EBX_MPX,
         .size = sizeof(XSaveBNDCSR),
+        .features = {
+            { FEAT_7_0_EBX,         CPUID_7_0_EBX_MPX },
+        },
     },
     [XSTATE_OPMASK_BIT] = {
-        .feature = FEAT_7_0_EBX, .bits = CPUID_7_0_EBX_AVX512F,
         .size = sizeof(XSaveOpmask),
+        .features = {
+            { FEAT_7_0_EBX,         CPUID_7_0_EBX_AVX512F },
+        },
     },
     [XSTATE_ZMM_Hi256_BIT] = {
-        .feature = FEAT_7_0_EBX, .bits = CPUID_7_0_EBX_AVX512F,
         .size = sizeof(XSaveZMM_Hi256),
+        .features = {
+            { FEAT_7_0_EBX,         CPUID_7_0_EBX_AVX512F },
+        },
     },
     [XSTATE_Hi16_ZMM_BIT] = {
-        .feature = FEAT_7_0_EBX, .bits = CPUID_7_0_EBX_AVX512F,
         .size = sizeof(XSaveHi16_ZMM),
+        .features = {
+            { FEAT_7_0_EBX,         CPUID_7_0_EBX_AVX512F },
+        },
     },
     [XSTATE_PKRU_BIT] = {
-        .feature = FEAT_7_0_ECX, .bits = CPUID_7_0_ECX_PKU,
         .size = sizeof(XSavePKRU),
+        .features = {
+            { FEAT_7_0_ECX,         CPUID_7_0_ECX_PKU },
+        },
     },
     [XSTATE_ARCH_LBR_BIT] = {
-        .feature = FEAT_7_0_EDX, .bits = CPUID_7_0_EDX_ARCH_LBR,
         .size = sizeof(XSaveArchLBR),
+        .features = {
+            { FEAT_7_0_EDX,         CPUID_7_0_EDX_ARCH_LBR },
+        },
     },
     [XSTATE_XTILE_CFG_BIT] = {
-        .feature = FEAT_7_0_EDX, .bits = CPUID_7_0_EDX_AMX_TILE,
         .size = sizeof(XSaveXTILECFG),
+        .features = {
+            { FEAT_7_0_EDX,         CPUID_7_0_EDX_AMX_TILE },
+        },
     },
     [XSTATE_XTILE_DATA_BIT] = {
-        .feature = FEAT_7_0_EDX, .bits = CPUID_7_0_EDX_AMX_TILE,
         .size = sizeof(XSaveXTILEDATA),
+        .features = {
+            { FEAT_7_0_EDX,         CPUID_7_0_EDX_AMX_TILE },
+        },
     },
 };
 
@@ -7131,16 +7155,24 @@ static inline void feat2prop(char *s)
 static const char *x86_cpu_feature_name(FeatureWord w, int bitnr)
 {
     const char *name;
-    /* XSAVE components are automatically enabled by other features,
+    /*
+     * XSAVE components are automatically enabled by other features,
      * so return the original feature name instead
      */
     if (w == FEAT_XSAVE_XCR0_LO || w == FEAT_XSAVE_XCR0_HI) {
         int comp = (w == FEAT_XSAVE_XCR0_HI) ? bitnr + 32 : bitnr;
 
-        if (comp < ARRAY_SIZE(x86_ext_save_areas) &&
-            x86_ext_save_areas[comp].bits) {
-            w = x86_ext_save_areas[comp].feature;
-            bitnr = ctz32(x86_ext_save_areas[comp].bits);
+        if (comp < ARRAY_SIZE(x86_ext_save_areas)) {
+            /*
+             * Present the first feature as the default.
+             * FIXME: select and present the one which is actually enabled
+             * among multiple dependencies.
+             */
+            const FeatureMask *fm = &x86_ext_save_areas[comp].features[0];
+            if (fm->mask) {
+                w = fm->index;
+                bitnr = ctz32(fm->mask);
+            }
         }
     }
 
@@ -8610,11 +8642,15 @@ static bool cpuid_has_xsave_feature(CPUX86State *env, const ExtSaveArea *esa)
         return false;
     }
 
-    if (env->features[esa->feature] & esa->bits) {
-        return true;
+    for (int i = 0; i < ARRAY_SIZE(esa->features); i++) {
+        if (env->features[esa->features[i].index] & esa->features[i].mask) {
+            return true;
+        }
     }
-    if (esa->feature == FEAT_7_0_EBX && esa->bits == CPUID_7_0_EBX_AVX512F
-        && (env->features[FEAT_7_1_EDX] & CPUID_7_1_EDX_AVX10)) {
+
+    if (esa->features[0].index == FEAT_7_0_EBX &&
+        esa->features[0].mask == CPUID_7_0_EBX_AVX512F &&
+        (env->features[FEAT_7_1_EDX] & CPUID_7_1_EDX_AVX10)) {
         return true;
     }
 
-- 
2.52.0