[Qemu-devel] [PATCH] ACPI/unit-test: Add a testcase for RAM allocation in numa node
As QEMU supports the memory-less node, it is possible that there is
no RAM in the first numa node(also be called as node0). eg:
... \
-m 128,slots=3,maxmem=1G \
-numa node -numa node,mem=128M \
But, this makes it hard for QEMU to build a known-to-work ACPI SRAT
table. Only fixing it is not enough.
Add a testcase for this situation to make sure the ACPI table is
correct for guest.
Suggested-by: Eduardo Habkost
Signed-off-by: Dou Liyang
---
Changelog:
This patch can pass the 'make check' in Peter's machine,
But, in my own and Eduardo's machine, we can pass it.
So I rebase and spilt it independently.
Peter:
could you help me test it in your machine.
then, give me the output files, such as /tmp/asl-6QYK7Y.dsl and
/tmp/asl-1H1I7Y.dsl file.
tests/acpi-test-data/pc/DSDT.numamem | Bin 0 -> 5104 bytes
tests/acpi-test-data/pc/SRAT.numamem | Bin 0 -> 224 bytes
tests/acpi-test-data/q35/DSDT.numamem | Bin 0 -> 7788 bytes
tests/acpi-test-data/q35/SRAT.numamem | Bin 0 -> 224 bytes
tests/bios-tables-test.c | 24
5 files changed, 24 insertions(+)
create mode 100644 tests/acpi-test-data/pc/DSDT.numamem
create mode 100644 tests/acpi-test-data/pc/SRAT.numamem
create mode 100644 tests/acpi-test-data/q35/DSDT.numamem
create mode 100644 tests/acpi-test-data/q35/SRAT.numamem
diff --git a/tests/acpi-test-data/pc/DSDT.numamem
b/tests/acpi-test-data/pc/DSDT.numamem
new file mode 100644
index
..bc703d77ee48720af1b801c518b43511c95c852d
GIT binary patch
literal 5104
zcmb7IZExE~5}qX~n_5aXWizjC8^LKVXmL&SlDFc3gC$b76k9SQ%DEh9oG8b|DMfK1
z(gvt&6i`&ab^ERG;xE-sw(qx*XF!z}jjPX%ajXzq&jTQFq
zw)Zd3`{YZHwS3rmsXyOp`CsgNTR@obq*UN7;@fd>bkHW>7wH}lOw3;x+kz<+;?^`Xya)uvf@QYk*O7srT^929^Z
z(63x0Z^&J@d%=E?%?G|FWMJruoz0sP@PWP*8-tlj#VIsH94;*3Ze&b}dmP>U!{4qE
z3%vBYS9|OW!Z=at1=2Sc+mq77{0X)+xWdGVZ3hlYIS{M^y2LNdbhD6aE$wxA%i+Vt_OfTwcvY}08l4PJ2-Q=9}7b8sV#4=e38#LcV|M*sM#V>#9yxmCRb#$#4_CDp-{qY)9{PBnYsRh0J+mH1
zKs}S1;o4VI5D$`V2fn5`9>Zs)r#)|D%xxO?Y1-|sO=Fmt%;AAdU;&}>q~cmRsk40k
zs~L#PG0akqe;WSnfH51ML2`oJGg3{f;=t!L=AB?>mQFSF$)!L(*L3O*`??)^fz_;D
zq4}Zp;)Hd~-`{LKJ7zr_SkIz=DIGuw@R^_6V|!(JIv?C%;y60xe4>R2=lq2
zd27-a9i_3C|w1tG&gG0m!lrDl#mvgwr8
z(ulMQjkJ+yR%#X12by%d#rHOYDuup;{v`{hUCCs!8S)^!&tpc)Y%Kp(>hXg%?3tNN
z=8;jJ!WveHyO%ewE8=3K7|D04M3d8K%m=S`@nBLx-urykbFZGztgZGvqZ*@#exD&W
zNreoj@*CwD(=lsmR2a;AS0R=mO%QXqd#b^Er&k*f1@5QRp
zE1#qa_VaWqE}!H=IC7mHXftCpr`KF31=~4|IsON`COWuCFqBfleh$@dgp$z
zs!&?t8N&}|D5jR$rv$?!tQHz6jjNzi?}b{eNf}N_0me)dgVHE6Xg~T8Pway#7z>!u
zD|V?_%H(j*g0_pYn>JcsS4b6{^{o(hA
zOd*oHie0wr;nMt?1cN)JPMzd}SMZ7%*!jG(iRPVrb8bpu=roRRH0M+WyFu*pP`XJP
z3PAtU@$$LdYs-HzmqQ2cm8u(<5II6mc&x|t7*#{P4F#GT1}X
z4-mKeu9F=KxZV;N$d2or;7C)@cEqj)LA0tk`~stZF1wu*L;69tFKA
zSh;}mNT^`N#0wqJqM)|~YeGO1kx;>!6wqW8^tND~5zv`Ps9?ngF!Xgc3VKJd>H?}q
zLIrC|KvPlB_XX>mfX+oi1?#+k&PPEDf)z7A49kdw3f40MdL{~56s%_j^lT(lu$~jp
zb5T%Ju$~vt^N~=&dO<)hL_saVdQm_xMnVPaf`BeWK}&-54FP>45-M2V6wo)Lpv!{w
zEdhNi5-M0P3FxKMphkJYx?gKZy~J)C@6tW0b&mo;M**u@D5Oacnk5vlB88d`38YEx
zwl%zn_Z%!MLrEfOq*GAB;xN>at(zkq8N(w!)RDDYBORH;gF4iaz1yUl-ECua#am?0
z!2>5`yhl(z*5hrm=it#1dTbU@KGy3b;~^0GbRZw=mq^BR7Wz&gAM1mNF)@W6FOxmT
z9MeiGn0^Um-1d>aLo(`(OVoJae}w3#J#8W0bsKCru(jB1F=Cg2*ZH@DxXmUE`z~9*2k!U%mXRq
Yf~!ZCL8t>-1O^}2VG2>z!9?-X08OU~0RR91
literal 0
HcmV?d1
diff --git a/tests/acpi-test-data/q35/DSDT.numamem
b/tests/acpi-test-data/q35/DSDT.numamem
new file mode 100644
index
..fcb18d947b28c09c8b5a4fa65692efb41ef3a5d9
GIT binary patch
literal 7788
zcmb7JTW=f38J#7U)M~kumeR`dU4ruvv}j_=PLrZV1CzVFRiZ?Zbew0D2nJQ>iNFm$TLd{i1Cv5m>(Yk5>h?S3es3>yTVbQ<^y`ht
zD}C=ePCt}eX{TR`+QRAIY(!SDHr3zgg!YqrB8+aW4A(RJ+l6`8?=}B<{fDKCH@>;`
zu=MP=|M{Ta~UFtA!C_Y?zOFtJ@Tw5a0ip6LxRcCp`aFWj$JMwWI!W@Or<~
zEr|iO!~<%&{pZ;AB}}cr9Mv;2C#&|v>NTrlpDKVZ)XihFy-#Jsow56+7{tP
zvy>JVe#-b+YjvX(vnnZfk5~&}RYX{-tXp9jUu!LV`8O7?rd36%*4ulf-3?Jx9j2+qnGr}4+bEN1yYJhhJt`fk2veU$zv?KoF<
zxKPE$oUABE8EbUbjk=);IjNh&^kG^!SA6cKO5I#bbmOMfaA;0X>?d`YLO%}raaxVA
zDgv*MUu&>U$p6bb799dV{5M}Nt#-FHxB6gbE@$KH5o_zK6(P4RY*)pcN9}G3pr$|?
zpf2X}ws&-di&^qUh*TiX$bGB>RY6jsf99rG3GN7yQc+Wam7uZ`kdzn=bJMI;BcSt-
zdu3``DnV6{l&T8Od&%W~Y*s3&EtP8UfV$6&;WNWZMa@X18a$xxGi&(FvQkmArcXfK
z=ak`d%J4a5`UKQ{I)+cj@adR70d=3aLufNSZTOrveFEw}UBjnq_;gL5fV$6|;WKCW
z%$Ys`b)UFXX~)wue0ru&K;38F@R>J!=1re~y3ZNI=ZxWV#`FoO`vO`m|e
zPv7wA8$Ny0C!p@LVE8N;J`1K#K;7q@;d9RLIcNF=)P2qyKIaXe^QKQg-DhC<3=E%v
z=@U@*xnTHQFnlhUJ^^)~XAGZb44-FApMbj0MZ@Q!;d9aS38?!#Yxq2C_&jU+1k`<=
zV`XV{Cp^c>^5|xFPAbC}kd)Dgd)`o3aES{KqZ%z2vChm0V=7iqyPm}
zJ`tdj%Sr^OMx_9iR8~@e0xF*fP|0N_0#u_?fJ!PWDL?_0PXwssvJwHRQ7J$rm6a5r
zfXXKVRB~B~0M)1zppwc;3Q$1h69Fo@tVDonR0>c@WhDhDpz?_Tm0VULKs71_sHC!z
z0u)gBM1V>zD-oa?l>$^!SxEs3sC*(oC6|>5P>o6fDygib00mS&5ulRGN(87zr2v&w
zR#JchDxU~Y$z>%1RHIUWN-8TUKmnCc1gPY)5&^1FDL^Hal@y?W$|nL;a#@K0)u`1ynu}pg>xH0_gz?s0S#Z7ND9E0jen}Ks6-=
zsHQ}KYDxsCrlbJXloX(v5&^0y5uloq0#s8{fNDwvsHQ}KYDx-FO-TW&DG{JR>gXas
zfpEllGATd-aTpXNrJ7p=D3F?41SpW2TMAG>HMbO?fRPB!_Yt9D=>lAuf<6zjG!Wnn|%Na49M!E^yzXk5Z=q;F)?cC^rJ>cd_vpq4!6wG
zn+|b617f_`@3cGClkL$Ms64fgs+SYh@mk?0S1)VT%Ur#
Re: [Qemu-devel] [PATCH] ACPI/unit-test: Add a testcase for RAM allocation in numa node
At 10/18/2017 11:48 AM, Dou Liyang wrote:
As QEMU supports the memory-less node, it is possible that there is
no RAM in the first numa node(also be called as node0). eg:
... \
-m 128,slots=3,maxmem=1G \
-numa node -numa node,mem=128M \
But, this makes it hard for QEMU to build a known-to-work ACPI SRAT
table. Only fixing it is not enough.
Add a testcase for this situation to make sure the ACPI table is
correct for guest.
Suggested-by: Eduardo Habkost
Signed-off-by: Dou Liyang
---
Changelog:
This patch can pass the 'make check' in Peter's machine,
Oops, s/can/can not :-).
Thanks,
dou.
But, in my own and Eduardo's machine, we can pass it.
So I rebase and spilt it independently.
Peter:
could you help me test it in your machine.
then, give me the output files, such as /tmp/asl-6QYK7Y.dsl and
/tmp/asl-1H1I7Y.dsl file.
tests/acpi-test-data/pc/DSDT.numamem | Bin 0 -> 5104 bytes
tests/acpi-test-data/pc/SRAT.numamem | Bin 0 -> 224 bytes
tests/acpi-test-data/q35/DSDT.numamem | Bin 0 -> 7788 bytes
tests/acpi-test-data/q35/SRAT.numamem | Bin 0 -> 224 bytes
tests/bios-tables-test.c | 24
5 files changed, 24 insertions(+)
create mode 100644 tests/acpi-test-data/pc/DSDT.numamem
create mode 100644 tests/acpi-test-data/pc/SRAT.numamem
create mode 100644 tests/acpi-test-data/q35/DSDT.numamem
create mode 100644 tests/acpi-test-data/q35/SRAT.numamem
diff --git a/tests/acpi-test-data/pc/DSDT.numamem
b/tests/acpi-test-data/pc/DSDT.numamem
new file mode 100644
index
..bc703d77ee48720af1b801c518b43511c95c852d
GIT binary patch
literal 5104
zcmb7IZExE~5}qX~n_5aXWizjC8^LKVXmL&SlDFc3gC$b76k9SQ%DEh9oG8b|DMfK1
z(gvt&6i`&ab^ERG;xE-sw(qx*XF!z}jjPX%ajXzq&jTQFq
zw)Zd3`{YZHwS3rmsXyOp`CsgNTR@obq*UN7;@fd>bkHW>7wH}lOw3;x+kz<+;?^`Xya)uvf@QYk*O7srT^929^Z
z(63x0Z^&J@d%=E?%?G|FWMJruoz0sP@PWP*8-tlj#VIsH94;*3Ze&b}dmP>U!{4qE
z3%vBYS9|OW!Z=at1=2Sc+mq77{0X)+xWdGVZ3hlYIS{M^y2LNdbhD6aE$wxA%i+Vt_OfTwcvY}08l4PJ2-Q=9}7b8sV#4=e38#LcV|M*sM#V>#9yxmCRb#$#4_CDp-{qY)9{PBnYsRh0J+mH1
zKs}S1;o4VI5D$`V2fn5`9>Zs)r#)|D%xxO?Y1-|sO=Fmt%;AAdU;&}>q~cmRsk40k
zs~L#PG0akqe;WSnfH51ML2`oJGg3{f;=t!L=AB?>mQFSF$)!L(*L3O*`??)^fz_;D
zq4}Zp;)Hd~-`{LKJ7zr_SkIz=DIGuw@R^_6V|!(JIv?C%;y60xe4>R2=lq2
zd27-a9i_3C|w1tG&gG0m!lrDl#mvgwr8
z(ulMQjkJ+yR%#X12by%d#rHOYDuup;{v`{hUCCs!8S)^!&tpc)Y%Kp(>hXg%?3tNN
z=8;jJ!WveHyO%ewE8=3K7|D04M3d8K%m=S`@nBLx-urykbFZGztgZGvqZ*@#exD&W
zNreoj@*CwD(=lsmR2a;AS0R=mO%QXqd#b^Er&k*f1@5QRp
zE1#qa_VaWqE}!H=IC7mHXftCpr`KF31=~4|IsON`COWuCFqBfleh$@dgp$z
zs!&?t8N&}|D5jR$rv$?!tQHz6jjNzi?}b{eNf}N_0me)dgVHE6Xg~T8Pway#7z>!u
zD|V?_%H(j*g0_pYn>JcsS4b6{^{o(hA
zOd*oHie0wr;nMt?1cN)JPMzd}SMZ7%*!jG(iRPVrb8bpu=roRRH0M+WyFu*pP`XJP
z3PAtU@$$LdYs-HzmqQ2cm8u(<5II6mc&x|t7*#{P4F#GT1}X
z4-mKeu9F=KxZV;N$d2or;7C)@cEqj)LA0tk`~stZF1wu*L;69tFKA
zSh;}mNT^`N#0wqJqM)|~YeGO1kx;>!6wqW8^tND~5zv`Ps9?ngF!Xgc3VKJd>H?}q
zLIrC|KvPlB_XX>mfX+oi1?#+k&PPEDf)z7A49kdw3f40MdL{~56s%_j^lT(lu$~jp
zb5T%Ju$~vt^N~=&dO<)hL_saVdQm_xMnVPaf`BeWK}&-54FP>45-M2V6wo)Lpv!{w
zEdhNi5-M0P3FxKMphkJYx?gKZy~J)C@6tW0b&mo;M**u@D5Oacnk5vlB88d`38YEx
zwl%zn_Z%!MLrEfOq*GAB;xN>at(zkq8N(w!)RDDYBORH;gF4iaz1yUl-ECua#am?0
z!2>5`yhl(z*5hrm=it#1dTbU@KGy3b;~^0GbRZw=mq^BR7Wz&gAM1mNF)@W6FOxmT
z9MeiGn0^Um-1d>aLo(`(OVoJae}w3#J#8W0bsKCru(jB1F=Cg2*ZH@DxXmUE`z~9*2k!U%mXRq
Yf~!ZCL8t>-1O^}2VG2>z!9?-X08OU~0RR91
literal 0
HcmV?d1
diff --git a/tests/acpi-test-data/q35/DSDT.numamem
b/tests/acpi-test-data/q35/DSDT.numamem
new file mode 100644
index
..fcb18d947b28c09c8b5a4fa65692efb41ef3a5d9
GIT binary patch
literal 7788
zcmb7JTW=f38J#7U)M~kumeR`dU4ruvv}j_=PLrZV1CzVFRiZ?Zbew0D2nJQ>iNFm$TLd{i1Cv5m>(Yk5>h?S3es3>yTVbQ<^y`ht
zD}C=ePCt}eX{TR`+QRAIY(!SDHr3zgg!YqrB8+aW4A(RJ+l6`8?=}B<{fDKCH@>;`
zu=MP=|M{Ta~UFtA!C_Y?zOFtJ@Tw5a0ip6LxRcCp`aFWj$JMwWI!W@Or<~
zEr|iO!~<%&{pZ;AB}}cr9Mv;2C#&|v>NTrlpDKVZ)XihFy-#Jsow56+7{tP
zvy>JVe#-b+YjvX(vnnZfk5~&}RYX{-tXp9jUu!LV`8O7?rd36%*4ulf-3?Jx9j2+qnGr}4+bEN1yYJhhJt`fk2veU$zv?KoF<
zxKPE$oUABE8EbUbjk=);IjNh&^kG^!SA6cKO5I#bbmOMfaA;0X>?d`YLO%}raaxVA
zDgv*MUu&>U$p6bb799dV{5M}Nt#-FHxB6gbE@$KH5o_zK6(P4RY*)pcN9}G3pr$|?
zpf2X}ws&-di&^qUh*TiX$bGB>RY6jsf99rG3GN7yQc+Wam7uZ`kdzn=bJMI;BcSt-
zdu3``DnV6{l&T8Od&%W~Y*s3&EtP8UfV$6&;WNWZMa@X18a$xxGi&(FvQkmArcXfK
z=ak`d%J4a5`UKQ{I)+cj@adR70d=3aLufNSZTOrveFEw}UBjnq_;gL5fV$6|;WKCW
z%$Ys`b)UFXX~)wue0ru&K;38F@R>J!=1re~y3ZNI=ZxWV#`FoO`vO`m|e
zPv7wA8$Ny0C!p@LVE8N;J`1K#K;7q@;d9RLIcNF=)P2qyKIaXe^QKQg-DhC<3=E%v
z=@U@*xnTHQFnlhUJ^^)~XAGZb44-FApMbj0MZ@Q!;d9aS38?!#Yxq2C_&jU+1k`<=
zV`XV{Cp^c>^5|xFPAbC}kd)Dgd)`o3aES{KqZ%z2vChm0V=7iqyPm}
zJ`tdj%Sr^OMx_9iR8~@e0xF*fP|0N_0#u_?fJ!PWDL?_0PXwssvJwHRQ7J$rm6a5r
zfXXKVRB~B~0M)1zppwc;3Q$1h69Fo@tVDonR0>c@WhDhDpz?_Tm0VULKs71_sHC!z
z0u)gBM1V>zD-oa?l>$^!SxEs3sC*(oC6|>5P>o6fDygib00mS&5ulRGN(87zr2v&w
zR#JchDxU~Y$z>%1RHIUWN-8TUKmnCc1gPY)5&^1FDL^Hal@y?W$|nL;a#@K0)u`1ynu}pg>xH0_gz?s0S#Z7ND9E0jen}Ks6-=
zsHQ}KYDxsCrlbJXloX(v5&^0y5uloq0#s8{fNDwvsHQ}KYDx-FO-TW&DG{JR>gXas
zfpEllGATd-aTpXNrJ7p=D3F?41SpW2TMAG>HMbO?fRPB!_Yt9D=>lA
Re: [Qemu-devel] [PATCH] ACPI/unit-test: Add a testcase for RAM allocation in numa node
On Wed, Oct 18, 2017 at 11:50:32AM +0800, Dou Liyang wrote:
>
> At 10/18/2017 11:48 AM, Dou Liyang wrote:
> > As QEMU supports the memory-less node, it is possible that there is
> > no RAM in the first numa node(also be called as node0). eg:
> > ... \
> > -m 128,slots=3,maxmem=1G \
> > -numa node -numa node,mem=128M \
> >
> > But, this makes it hard for QEMU to build a known-to-work ACPI SRAT
> > table. Only fixing it is not enough.
> >
> > Add a testcase for this situation to make sure the ACPI table is
> > correct for guest.
> >
> > Suggested-by: Eduardo Habkost
> > Signed-off-by: Dou Liyang
> > ---
> > Changelog:
> >
> > This patch can pass the 'make check' in Peter's machine,
>
> Oops, s/can/can not :-).
>
> Thanks,
> dou.
>
> > But, in my own and Eduardo's machine, we can pass it.
> >
> > So I rebase and spilt it independently.
Thanks!
> >
> > Peter:
> >
> > could you help me test it in your machine.
> > then, give me the output files, such as /tmp/asl-6QYK7Y.dsl and
> > /tmp/asl-1H1I7Y.dsl file.
As the failures reported by Peter were about the *.memhp test
case files, I think this patch is unlikely to be the cause of
those failures. I believe the problem is at:
From: Dou Liyang
Date: Fri, 1 Sep 2017 10:10:03 +0800
Message-ID: <1504231805-30957-3-git-send-email-douly.f...@cn.fujitsu.com>
Subject: [PATCH v6 2/4] hw/acpi-build: Make assignment statement
of next_base easy to read
However, I would still like to get test results from Peter before
applying this patch again.
> >
> > tests/acpi-test-data/pc/DSDT.numamem | Bin 0 -> 5104 bytes
> > tests/acpi-test-data/pc/SRAT.numamem | Bin 0 -> 224 bytes
> > tests/acpi-test-data/q35/DSDT.numamem | Bin 0 -> 7788 bytes
> > tests/acpi-test-data/q35/SRAT.numamem | Bin 0 -> 224 bytes
> > tests/bios-tables-test.c | 24
> >
> > 5 files changed, 24 insertions(+) create mode 100644
> > tests/acpi-test-data/pc/DSDT.numamem create mode 100644
> > tests/acpi-test-data/pc/SRAT.numamem create mode 100644
> > tests/acpi-test-data/q35/DSDT.numamem create mode 100644
> > tests/acpi-test-data/q35/SRAT.numamem
> >
> > diff --git a/tests/acpi-test-data/pc/DSDT.numamem
> > b/tests/acpi-test-data/pc/DSDT.numamem new file mode 100644
> > index
> > ..bc703d77ee48720af1b801c518b43511c95c852d
> > GIT binary patch literal 5104
> > zcmb7IZExE~5}qX~n_5aXWizjC8^LKVXmL&SlDFc3gC$b76k9SQ%DEh9oG8b|DMfK1
> > z(gvt&6i`&ab^ERG > zwj8VRzj**QUtBlKPP+KOHZ7cE06=5<)+@>;xE-sw(qx*XF!z}jjPX%ajXzq&jTQFq
> > zw)Zd3`{YZHwS3rmsXyOp`CsgNTR@ > zpf)!u7SpS0h$*RMyVMXMh9Fd<8)dsug#^HNKWm`4&vcUK@0V&;+OAv8v~jEHGz#F;
> > zjOWu->obq*UN7;@fd>bkHW>7wH}lOw3;x+kz<+;?^`Xya)uvf@QYk*O7srT^929^Z
> > z(63x0Z^&J@d%=E?%?G|FWMJruoz0sP@PWP*8-tlj#VIsH94;*3Ze&b}dmP>U!{4qE
> > z3%v > zuodBYS9|OW!Z=at1=2Sc+mq77{0X)+xWdGVZ3hlYIS{M^y2LN > zErL$Kg1T&(qfMM1dbGgLudpBFA7oHg*lYPUF4W=@ysHG<+2u96AU1p1s?4Qz;4!|=
> > zGmIZ{iC@9LIljOL{3HGopXI9BT((N7bKIab9REltxZOXm*^QB}3K-|Zt*0gRdQ-UF
> > zeR!daV%B+bg?%c2Dy!;ZC-A4FnCsQ7SkxE`Gf>dbhD6aE$wxA%i+Vt_Of > za7QC~q$3G=ROHdi>TwcvY}08l4PJ2-Q=9}7b8sV#4=e3 > zMX^?`GffW_B$&n8Mm6oS;+nY%#ueKv-Id(SBn<|Bo0R?zdj#~ldzGg9px5vS6`T4h
> > zr?fXL-n8ot@aLqC=R@mdG5);8JN{=aF2R<+`=qr6T!x1|{oo;j;&^P > z$a<-R7FR~o4s5?jj-&#DQU?#YG@2<6XXMe03ETeXJ6l`_Jh-sG4dCy#(BA$w1Alwp
> > zYr)f2-*cb6eO)GR>8#LcV|M*sM#V>#9yxmCRb#$#4_CDp-{qY)9{PBnYsRh0J+mH1
> > zKs}S1;o4VI5D$`V2fn5`9>Zs)r#)|D%xxO?Y1-|sO=Fmt%;AAdU;&}>q~cmRsk40k
> > zs~L#PG0akqe;WSnfH51ML2`oJGg3{f;=t!L=AB?>mQFSF$)!L(*L3O*`??)^fz_;D
> > zq4}Zp;)Hd~-`{LKJ7zr_SkIz=DIGuw@R^_6V|!(JIv?C%;y60xe4>R2=lq2
> > zd27- > zMZ2r7-wrQ4lW!gueY;*7nc!1FawScBBVB&{iT}~lzo(yk@bTlPPab~oF}TdM*H(w+
> > zH_7=5gF`rE39QkWR6!Lva9i_3C|w1tG&gG0m!lrDl#mvgwr8
> > z(ulMQjkJ+yR%#X12by%d#rHOYDuup;{v`{hUCCs!8S)^!&tpc)Y%Kp(>hXg%?3tNN
> > z=8;jJ!WveHyO%ewE8=3K7|D04M3d8K%m=S`@nBLx-urykbFZGztgZGvqZ*@#exD&W
> > zNreoj@*CwD(=lsmR2a;AS0R=mO%QXqd#b^Er&k*f1@5QRp
> > zE1#qa_VaWqE}!H=IC7mHXftCpr`KF31=~4|IsON`COWuCFqBfleh$@dgp$z
> > zs!&?t8N&}|D5jR$rv$?!tQHz6jjNzi?}b{eNf}N_0me)dgVHE6Xg~T8Pway#7z>!u
> > zD|V?_%H(j*g0_pYn>JcsS4b6{^ > zt_oYY;xzxqei9_?Xz??k2exEKiK7Hlah_qTiJ#Y~K1z(_B;JJp|NrtFK&YpCW-y?8
> > zhCUJm8qqYVgTc2yhnoNVM6H{tU{GP}mb7bjrhq|(C5Gn5Y74gFjRMy~PlMX>{o(hA
> > zOd*oHie0wr;nMt?1cN)JPMzd}SMZ7%*!jG(iRPVrb8bpu=roRRH0M+WyFu*pP`XJP
> > z3PAtU@$$LdYs-HzmqQ2cm8u(<5II6mc&x|t7*#{P4F#GT1}X
> > z4-mKeu9F=KxZV;N$d > zB`97KXbPd&u0}uwYfM06QP8|##S0>2or;7C)@cEqj)LA0tk`~stZF1wu*L;69tFKA
> > zSh;}mNT^`N#0wqJqM)|~YeGO1kx;>!6wqW8^tND~5zv`Ps9?ngF!Xgc3VKJd>H?}q
> > zLIrC|KvPlB_XX>mfX+oi1?#+k&PPEDf)z7A49kdw3f40MdL{~56s%_j^lT(lu$~jp
> > zb5T%Ju$~vt^N~=&dO<)hL_saVdQm_xMnVPaf`BeWK}&-54FP>45-M2V6wo)Lpv!{w
> > zEdhNi5-M0P3FxKMphkJYx?gKZy~J)C@6tW0b&mo;M**u@D5Oacnk5vlB88d`38YEx
> > zwl%zn_Z%!MLrEfOq*GAB;xN>at(zkq8N(w!)RDDYBORH;gF4iaz1yUl-ECua#am?0
> > z!2>5`yhl(z*5hrm=it#1dTbU@KGy3b;~^0GbRZw=mq^BR7Wz&gAM1mNF)@W6FOxmT
> > z9MeiGn0^Um-1d>aLo(`(OVoJae}w3#J#8W0bsKCru( > DEH%hI
> >
> > literal 0 HcmV?d1
> >
Re: [Qemu-devel] [PATCH] ACPI/unit-test: Add a testcase for RAM allocation in numa node
Hi Eduardo,
Thanks for your reply.
At 10/21/2017 03:15 AM, Eduardo Habkost wrote:
On Wed, Oct 18, 2017 at 11:50:32AM +0800, Dou Liyang wrote:
At 10/18/2017 11:48 AM, Dou Liyang wrote:
As QEMU supports the memory-less node, it is possible that there is
no RAM in the first numa node(also be called as node0). eg:
... \
-m 128,slots=3,maxmem=1G \
-numa node -numa node,mem=128M \
But, this makes it hard for QEMU to build a known-to-work ACPI SRAT
table. Only fixing it is not enough.
Add a testcase for this situation to make sure the ACPI table is
correct for guest.
Suggested-by: Eduardo Habkost
Signed-off-by: Dou Liyang
---
Changelog:
This patch can pass the 'make check' in Peter's machine,
Oops, s/can/can not :-).
Thanks,
dou.
But, in my own and Eduardo's machine, we can pass it.
So I rebase and spilt it independently.
Thanks!
Peter:
could you help me test it in your machine.
then, give me the output files, such as /tmp/asl-6QYK7Y.dsl and
/tmp/asl-1H1I7Y.dsl file.
As the failures reported by Peter were about the *.memhp test
case files, I think this patch is unlikely to be the cause of
those failures. I believe the problem is at:
From: Dou Liyang
Date: Fri, 1 Sep 2017 10:10:03 +0800
Message-ID: <1504231805-30957-3-git-send-email-douly.f...@cn.fujitsu.com>
Subject: [PATCH v6 2/4] hw/acpi-build: Make assignment statement
of next_base easy to read
This patch is just used for readability. so I discard it. and, we
need a test case for the commit
4926403c250f ("hw/acpi-build: Fix SRAT memory building in case of
node 0 without RAM")
So, I re-post this patch. ;-)
However, I would still like to get test results from Peter before
applying this patch again.
I see.
Thanks,
dou
