From: Andrew Cooper <andrew.cooper3@citrix.com> To: Xen-devel <xen-devel@lists.xenproject.org> Cc: "Andrew Cooper" <andrew.cooper3@citrix.com>, "Wei Liu" <wei.liu2@citrix.com>, "Jan Beulich" <JBeulich@suse.com>, "Roger Pau Monné" <roger.pau@citrix.com> Subject: [PATCH] libx86: Elide more empty CPUID leaves when serialising a policy Date: Wed, 22 May 2019 16:50:30 +0100 [thread overview] Message-ID: <1558540230-26612-1-git-send-email-andrew.cooper3@citrix.com> (raw) x86_cpuid_copy_to_buffer() currently serialises the full content of the various subleaf unions. While leaves 4, 0xb and 0xd don't have a concrete max_subleaf field, they do have well defined upper bounds. Diffing the results of `xen-cpuid -p` shows the resutling saving: @@ -1,5 +1,5 @@ Xen reports there are maximum 114 leaves and 1 MSRs -Raw policy: 93 leaves, 1 MSRs +Raw policy: 38 leaves, 1 MSRs CPUID: leaf subleaf -> eax ebx ecx edx 00000000:ffffffff -> 00000016:756e6547:6c65746e:49656e69 @@ -32,7 +32,7 @@ Raw policy: 93 leaves, 1 MSRs MSRs: index -> value 000000ce -> 0000000080000000 -Host policy: 93 leaves, 1 MSRs +Host policy: 33 leaves, 1 MSRs CPUID: leaf subleaf -> eax ebx ecx edx 00000000:ffffffff -> 0000000d:756e6547:6c65746e:49656e69 which is mostly due to no longer writing out 64 leaves for xstate when (on this CoffeeLake system) 8 will do. Extend the unit tests to cover empty and partially filled subleaf unions. Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com> --- CC: Jan Beulich <JBeulich@suse.com> CC: Wei Liu <wei.liu2@citrix.com> CC: Roger Pau Monné <roger.pau@citrix.com> --- tools/tests/cpu-policy/test-cpu-policy.c | 71 ++++++++++++++++++++++++++++++++ xen/lib/x86/cpuid.c | 24 ++++++++++- 2 files changed, 94 insertions(+), 1 deletion(-) diff --git a/tools/tests/cpu-policy/test-cpu-policy.c b/tools/tests/cpu-policy/test-cpu-policy.c index beced5e..fd96c0b 100644 --- a/tools/tests/cpu-policy/test-cpu-policy.c +++ b/tools/tests/cpu-policy/test-cpu-policy.c @@ -65,6 +65,77 @@ static void test_cpuid_serialise_success(void) .name = "empty policy", .nr_leaves = 4, }, + + /* Leaf 4 serialisation stops at the first subleaf with type 0. */ + { + .name = "empty leaf 4", + .p = { + .basic.max_leaf = 4, + }, + .nr_leaves = 4 + 4, + }, + { + .name = "partial leaf 4", + .p = { + .basic.max_leaf = 4, + .cache.subleaf[0].type = 1, + }, + .nr_leaves = 4 + 4 + 1, + }, + + /* Leaf 7 serialisation stops at max_subleaf. */ + { + .name = "empty leaf 7", + .p = { + .basic.max_leaf = 7, + }, + .nr_leaves = 4 + 7, + }, + { + .name = "partial leaf 7", + .p = { + .basic.max_leaf = 7, + .feat.max_subleaf = 1, + }, + .nr_leaves = 4 + 7 + 1, + }, + + /* Leaf 0xb serialisation stops at the first subleaf with type 0. */ + { + .name = "empty leaf 0xb", + .p = { + .basic.max_leaf = 0xb, + }, + .nr_leaves = 4 + 0xb, + }, + { + .name = "partial leaf 0xb", + .p = { + .basic.max_leaf = 0xb, + .topo.subleaf[0].type = 1, + }, + .nr_leaves = 4 + 0xb + 1, + }, + + /* + * Leaf 0xd serialisation automatically has two leaves, and stops the + * highest bit set in {xcr0,xss}_{high,low}. + */ + { + .name = "empty leaf 0xd", + .p = { + .basic.max_leaf = 0xd, + }, + .nr_leaves = 4 + 0xd + 1, + }, + { + .name = "partial 0xd", + .p = { + .basic.max_leaf = 0xd, + .xstate.xcr0_low = 7, + }, + .nr_leaves = 4 + 0xd + 1 + 1, + }, }; printf("Testing CPUID serialise success:\n"); diff --git a/xen/lib/x86/cpuid.c b/xen/lib/x86/cpuid.c index 23619c7..dcab1e7 100644 --- a/xen/lib/x86/cpuid.c +++ b/xen/lib/x86/cpuid.c @@ -242,7 +242,12 @@ int x86_cpuid_copy_to_buffer(const struct cpuid_policy *p, { case 0x4: for ( subleaf = 0; subleaf < ARRAY_SIZE(p->cache.raw); ++subleaf ) + { COPY_LEAF(leaf, subleaf, &p->cache.raw[subleaf]); + + if ( p->cache.subleaf[subleaf].type == 0 ) + break; + } break; case 0x7: @@ -254,13 +259,30 @@ int x86_cpuid_copy_to_buffer(const struct cpuid_policy *p, case 0xb: for ( subleaf = 0; subleaf < ARRAY_SIZE(p->topo.raw); ++subleaf ) + { COPY_LEAF(leaf, subleaf, &p->topo.raw[subleaf]); + + if ( p->topo.subleaf[subleaf].type == 0 ) + break; + } break; case 0xd: - for ( subleaf = 0; subleaf < ARRAY_SIZE(p->xstate.raw); ++subleaf ) + { + uint64_t xstates; + + COPY_LEAF(leaf, 0, &p->xstate.raw[0]); + COPY_LEAF(leaf, 1, &p->xstate.raw[1]); + + xstates = ((uint64_t)(p->xstate.xcr0_high | p->xstate.xss_high) << 32); + xstates |= (p->xstate.xcr0_low | p->xstate.xss_low); + + for ( xstates >>= 2, subleaf = 2; + xstates && subleaf < ARRAY_SIZE(p->xstate.raw); + xstates >>= 1, ++subleaf ) COPY_LEAF(leaf, subleaf, &p->xstate.raw[subleaf]); break; + } default: COPY_LEAF(leaf, XEN_CPUID_NO_SUBLEAF, &p->basic.raw[leaf]); -- 2.1.4 _______________________________________________ Xen-devel mailing list Xen-devel@lists.xenproject.org https://lists.xenproject.org/mailman/listinfo/xen-devel
WARNING: multiple messages have this Message-ID (diff)
From: Andrew Cooper <andrew.cooper3@citrix.com> To: Xen-devel <xen-devel@lists.xenproject.org> Cc: "Andrew Cooper" <andrew.cooper3@citrix.com>, "Wei Liu" <wei.liu2@citrix.com>, "Jan Beulich" <JBeulich@suse.com>, "Roger Pau Monné" <roger.pau@citrix.com> Subject: [Xen-devel] [PATCH] libx86: Elide more empty CPUID leaves when serialising a policy Date: Wed, 22 May 2019 16:50:30 +0100 [thread overview] Message-ID: <1558540230-26612-1-git-send-email-andrew.cooper3@citrix.com> (raw) Message-ID: <20190522155030.m6W1GeugvjnR9jBZfhGkl8hOlySIMYz6wmkbeFAgMgw@z> (raw) x86_cpuid_copy_to_buffer() currently serialises the full content of the various subleaf unions. While leaves 4, 0xb and 0xd don't have a concrete max_subleaf field, they do have well defined upper bounds. Diffing the results of `xen-cpuid -p` shows the resutling saving: @@ -1,5 +1,5 @@ Xen reports there are maximum 114 leaves and 1 MSRs -Raw policy: 93 leaves, 1 MSRs +Raw policy: 38 leaves, 1 MSRs CPUID: leaf subleaf -> eax ebx ecx edx 00000000:ffffffff -> 00000016:756e6547:6c65746e:49656e69 @@ -32,7 +32,7 @@ Raw policy: 93 leaves, 1 MSRs MSRs: index -> value 000000ce -> 0000000080000000 -Host policy: 93 leaves, 1 MSRs +Host policy: 33 leaves, 1 MSRs CPUID: leaf subleaf -> eax ebx ecx edx 00000000:ffffffff -> 0000000d:756e6547:6c65746e:49656e69 which is mostly due to no longer writing out 64 leaves for xstate when (on this CoffeeLake system) 8 will do. Extend the unit tests to cover empty and partially filled subleaf unions. Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com> --- CC: Jan Beulich <JBeulich@suse.com> CC: Wei Liu <wei.liu2@citrix.com> CC: Roger Pau Monné <roger.pau@citrix.com> --- tools/tests/cpu-policy/test-cpu-policy.c | 71 ++++++++++++++++++++++++++++++++ xen/lib/x86/cpuid.c | 24 ++++++++++- 2 files changed, 94 insertions(+), 1 deletion(-) diff --git a/tools/tests/cpu-policy/test-cpu-policy.c b/tools/tests/cpu-policy/test-cpu-policy.c index beced5e..fd96c0b 100644 --- a/tools/tests/cpu-policy/test-cpu-policy.c +++ b/tools/tests/cpu-policy/test-cpu-policy.c @@ -65,6 +65,77 @@ static void test_cpuid_serialise_success(void) .name = "empty policy", .nr_leaves = 4, }, + + /* Leaf 4 serialisation stops at the first subleaf with type 0. */ + { + .name = "empty leaf 4", + .p = { + .basic.max_leaf = 4, + }, + .nr_leaves = 4 + 4, + }, + { + .name = "partial leaf 4", + .p = { + .basic.max_leaf = 4, + .cache.subleaf[0].type = 1, + }, + .nr_leaves = 4 + 4 + 1, + }, + + /* Leaf 7 serialisation stops at max_subleaf. */ + { + .name = "empty leaf 7", + .p = { + .basic.max_leaf = 7, + }, + .nr_leaves = 4 + 7, + }, + { + .name = "partial leaf 7", + .p = { + .basic.max_leaf = 7, + .feat.max_subleaf = 1, + }, + .nr_leaves = 4 + 7 + 1, + }, + + /* Leaf 0xb serialisation stops at the first subleaf with type 0. */ + { + .name = "empty leaf 0xb", + .p = { + .basic.max_leaf = 0xb, + }, + .nr_leaves = 4 + 0xb, + }, + { + .name = "partial leaf 0xb", + .p = { + .basic.max_leaf = 0xb, + .topo.subleaf[0].type = 1, + }, + .nr_leaves = 4 + 0xb + 1, + }, + + /* + * Leaf 0xd serialisation automatically has two leaves, and stops the + * highest bit set in {xcr0,xss}_{high,low}. + */ + { + .name = "empty leaf 0xd", + .p = { + .basic.max_leaf = 0xd, + }, + .nr_leaves = 4 + 0xd + 1, + }, + { + .name = "partial 0xd", + .p = { + .basic.max_leaf = 0xd, + .xstate.xcr0_low = 7, + }, + .nr_leaves = 4 + 0xd + 1 + 1, + }, }; printf("Testing CPUID serialise success:\n"); diff --git a/xen/lib/x86/cpuid.c b/xen/lib/x86/cpuid.c index 23619c7..dcab1e7 100644 --- a/xen/lib/x86/cpuid.c +++ b/xen/lib/x86/cpuid.c @@ -242,7 +242,12 @@ int x86_cpuid_copy_to_buffer(const struct cpuid_policy *p, { case 0x4: for ( subleaf = 0; subleaf < ARRAY_SIZE(p->cache.raw); ++subleaf ) + { COPY_LEAF(leaf, subleaf, &p->cache.raw[subleaf]); + + if ( p->cache.subleaf[subleaf].type == 0 ) + break; + } break; case 0x7: @@ -254,13 +259,30 @@ int x86_cpuid_copy_to_buffer(const struct cpuid_policy *p, case 0xb: for ( subleaf = 0; subleaf < ARRAY_SIZE(p->topo.raw); ++subleaf ) + { COPY_LEAF(leaf, subleaf, &p->topo.raw[subleaf]); + + if ( p->topo.subleaf[subleaf].type == 0 ) + break; + } break; case 0xd: - for ( subleaf = 0; subleaf < ARRAY_SIZE(p->xstate.raw); ++subleaf ) + { + uint64_t xstates; + + COPY_LEAF(leaf, 0, &p->xstate.raw[0]); + COPY_LEAF(leaf, 1, &p->xstate.raw[1]); + + xstates = ((uint64_t)(p->xstate.xcr0_high | p->xstate.xss_high) << 32); + xstates |= (p->xstate.xcr0_low | p->xstate.xss_low); + + for ( xstates >>= 2, subleaf = 2; + xstates && subleaf < ARRAY_SIZE(p->xstate.raw); + xstates >>= 1, ++subleaf ) COPY_LEAF(leaf, subleaf, &p->xstate.raw[subleaf]); break; + } default: COPY_LEAF(leaf, XEN_CPUID_NO_SUBLEAF, &p->basic.raw[leaf]); -- 2.1.4 _______________________________________________ Xen-devel mailing list Xen-devel@lists.xenproject.org https://lists.xenproject.org/mailman/listinfo/xen-devel
next reply other threads:[~2019-05-22 15:50 UTC|newest] Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top 2019-05-22 15:50 Andrew Cooper [this message] 2019-05-22 15:50 ` [Xen-devel] [PATCH] libx86: Elide more empty CPUID leaves when serialising a policy Andrew Cooper 2019-05-23 8:33 ` Jan Beulich 2019-05-23 8:33 ` [Xen-devel] " Jan Beulich 2019-05-23 10:38 ` Andrew Cooper 2019-05-23 10:38 ` [Xen-devel] " Andrew Cooper 2019-05-23 10:27 ` [PATCH] libx86: Introduce wrappers for extracting XCR0/XSS from a cpuid policy Andrew Cooper 2019-05-23 10:27 ` [Xen-devel] " Andrew Cooper 2019-05-23 11:52 ` Jan Beulich 2019-05-23 11:52 ` [Xen-devel] " Jan Beulich 2019-05-23 11:59 ` Andrew Cooper 2019-05-23 11:59 ` [Xen-devel] " Andrew Cooper 2019-05-23 12:08 ` Jan Beulich 2019-05-23 12:08 ` [Xen-devel] " Jan Beulich
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=1558540230-26612-1-git-send-email-andrew.cooper3@citrix.com \ --to=andrew.cooper3@citrix.com \ --cc=JBeulich@suse.com \ --cc=roger.pau@citrix.com \ --cc=wei.liu2@citrix.com \ --cc=xen-devel@lists.xenproject.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).