Menu
▾
▴
tpmdd-devel
|
From: mipsan.K <mi...@gm...> - 2015-10-27 02:29:47
|
Dear, Intel *NUC5i5MYHE *has TPM 2.0 chip from Infineon(Infineon *SLB9665TT2.0 *TPM version 2.0). I tested various kernel to activate it, however failed. Even linux-tpmdd(4.3-rc4) as well. That may support only SLB9635/9645 and not SLB9665. Any kinds of advice will be appreciated. Thanks, mipsan -- Everything around me is GREAT. |
|
From: Jarkko S. <jar...@li...> - 2015-10-27 08:03:22
|
On Tue, 2015-10-27 at 11:29 +0900, mipsan.K wrote: > Dear, > > Intel NUC5i5MYHE has TPM 2.0 chip from Infineon(Infineon SLB9665TT2.0 > TPM version 2.0). > > I tested various kernel to activate it, however failed. > Even linux-tpmdd(4.3-rc4) as well. > That may support only SLB9635/9645 and not SLB9665. > > Any kinds of advice will be appreciated. Can you please write plain text emails. Please read http://kernelnewbies.org/mailinglistguidelines. Just a friendly complain. Can you retry with 4.4-rc1 ones it comes out? It has a critical bug fix tpm_tis driver (used for discrete TPM chips). This is a known bug: https://bugzilla.kernel.org/show_bug.cgi?id=98181 > Thanks, > /Jarkko > mipsan |
|
From: Peter H. <pet...@gm...> - 2015-10-27 09:20:46
|
Hi Am 27. Oktober 2015 03:29:22 MEZ, schrieb "mipsan.K" <mi...@gm...>: >Dear, > >Intel *NUC5i5MYHE *has TPM 2.0 chip from Infineon(Infineon >*SLB9665TT2.0 *TPM >version 2.0). > >I tested various kernel to activate it, however failed. >Even linux-tpmdd(4.3-rc4) as well. >That may support only SLB9635/9645 and not SLB9665. > >Any kinds of advice will be appreciated. > >Thanks, > >mipsan As Jarkko said, that will be fixed in 4.4 In the meantime you can tryout the security-next tree, or my tpm tree. Or try loading the tpm_tis driver with force=1 Peter -- Sent from my mobile |
|
From: mipsan.K <mi...@gm...> - 2015-11-04 06:43:19
|
> > There is TSS2 based tpm2-tools: https://github.com/01org/tpm2.0-tools > Very thanks. I will test it immediately. > > > /Jarkko youngwha |
|
From: Jarkko S. <jar...@li...> - 2015-11-04 08:03:32
|
On Wed, Nov 04, 2015 at 03:42:52PM +0900, mipsan.K wrote: > > There is TSS2 based tpm2-tools: https://github.com/01org/tpm2.0-tools > > > > Very thanks. > I will test it immediately. Sorry I didn't mention about it earlier because I wasn't sure whether it was already in public or not. > youngwha /Jarkko |
|
From: mipsan.K <mi...@gm...> - 2015-11-04 08:20:38
|
> > Sorry I didn't mention about it earlier because I wasn't sure whether > it was already in public or not. > Thanks, I cloned the tree and tested it. As I execute a command to read PCRs, it works pretty well. I'm testing other commands. > > /Jarkko youngwha -- ///////////////////////////////////////////////// // // // WHY NOT !!! // // LET'S DO IT !!! // // // // mi...@gm... // // // ///////////////////////////////////////////////// |
|
From: mipsan.K <mi...@gm...> - 2015-11-05 04:38:01
|
> > Sorry I didn't mention about it earlier because I wasn't sure whether > it was already in public or not. > Can I introduce that to my acquaintances to test TPM 2.0 chip on their cases? youngwha |
|
From: mipsan.K <mi...@gm...> - 2015-10-29 10:16:09
|
Hi, I send reply-mesg to Peter only. Apologize. On Tue, Oct 27, 2015 at 6:20 PM, Peter Huewe <pet...@gm...> wrote: > Hi > > Am 27. Oktober 2015 03:29:22 MEZ, schrieb "mipsan.K" <mi...@gm...>: > >Dear, > > > >Intel *NUC5i5MYHE *has TPM 2.0 chip from Infineon(Infineon > >*SLB9665TT2.0 *TPM > >version 2.0). > > > >I tested various kernel to activate it, however failed. > >Even linux-tpmdd(4.3-rc4) as well. > >That may support only SLB9635/9645 and not SLB9665. > > > >Any kinds of advice will be appreciated. > > > >Thanks, > > > >mipsan > > > As Jarkko said, that will be fixed in 4.4 > In the meantime you can tryout the security-next tree, or my tpm tree. > > Would you let me know when the 4.4 kernel is available? And, does linux-tpmdd(4.3-rc4) tree still have same problem? Or try loading the tpm_tis driver with force=1 > > Peter > -- > Sent from my mobile > -- |
|
From: Peter H. <Pet...@gm...> - 2015-10-29 18:36:22
|
Hi, sorry for the late reply I was traveling the last days. >Am 27. Oktober 2015 03:29:22 MEZ, schrieb "mipsan.K" <mi...@gm...[mi...@gm...]>: >>Dear, >> >>Intel *NUC5i5MYHE *has TPM 2.0 chip from Infineon(Infineon >>*SLB9665TT2.0 *TPM >>version 2.0). >> >>I tested various kernel to activate it, however failed. >>Even linux-tpmdd(4.3-rc4) as well. >>That may support only SLB9635/9645 and not SLB9665. >> >>Any kinds of advice will be appreciated. >> >>Thanks, >> >>mipsan >> As Jarkko said, that will be fixed in 4.4 >> In the meantime you can tryout the security-next tree, or my tpm tree. > Would you let me know when the 4.4 kernel is available? > And, does linux-tpmdd(4.3-rc4) tree still have same problem? until 4.4 is really available it will take quite a while. Please try either (linux-tpmdd) https://github.com/PeterHuewe/linux-tpmdd/tree/for-james Or https://git.kernel.org/cgit/linux/kernel/git/jmorris/linux-security.git/log/?h=next (this is what I meant by security-next) >> Or try loading the tpm_tis driver with force=1 Thanks, Peter |
|
From: Jarkko S. <jar...@li...> - 2015-10-30 10:25:30
|
Hi On Thu, Oct 29, 2015 at 07:15:44PM +0900, mipsan.K wrote: > Hi, > > I send reply-mesg to Peter only. > Apologize. > On Tue, Oct 27, 2015 at 6:20 PM, Peter Huewe <pet...@gm...> wrote: > > Hi > > Am 27. Oktober 2015 03:29:22 MEZ, schrieb "mipsan.K" <mi...@gm...>: > >Dear, > > > >Intel *NUC5i5MYHE *has TPM 2.0 chip from Infineon(Infineon > >*SLB9665TT2.0 *TPM > >version 2.0). > > > >I tested various kernel to activate it, however failed. > >Even linux-tpmdd(4.3-rc4) as well. > >That may support only SLB9635/9645 and not SLB9665. > > > >Any kinds of advice will be appreciated. > > > >Thanks, > > > >mipsan > > > As Jarkko said, that will be fixed in 4.4 > In the meantime you can tryout the security-next tree, or my tpm tree. > > > > Would you let me know when the 4.4 kernel is available? > And, does linux-tpmdd(4.3-rc4) tree still have same problem? Could you not use HTML for emails? These are pretty difficult to follow especially when you are using text-only email client. This the second time I'm asking for this. To know when the next release is available you should follow http://vger.kernel.org/vger-lists.html#linux-kernel-announce The fixes that you require should be in 4.4-rc1 release and you get the annoucement from that list. In the meanwhile you can try to compile kernel from this branch: https://github.com/jsakkine/linux-tpm2/tree/for-peter-v44 /Jarkko |
|
From: mipsan.K <mi...@gm...> - 2015-11-05 08:05:00
|
Hi, > > Sorry I didn't mention about it earlier because I wasn't sure whether > it was already in public or not. > When I build tpm2.0-tools, I met following error messages: ./configure: line 14222: syntax error near unexpected token `,' ./configure: line 14222: `AX_PTHREAD(, AC_MSG_ERROR([requires pthread]))' Although I could build this tool with simple job, is there another primitive action to resolve it? (simply, comment that line not to check AX_PTHREAD). This requires another modification in the tss/Makefile regard to pthread library. youngwha |
|
From: Jarkko S. <jar...@li...> - 2015-11-05 11:07:54
|
On Thu, Nov 05, 2015 at 05:04:35PM +0900, mipsan.K wrote: > Hi, > > > > > Sorry I didn't mention about it earlier because I wasn't sure whether > > it was already in public or not. > > > When I build tpm2.0-tools, I met following error messages: > > ./configure: line 14222: syntax error near unexpected token `,' > ./configure: line 14222: `AX_PTHREAD(, AC_MSG_ERROR([requires pthread]))' > > Although I could build this tool with simple job, is there another > primitive action to resolve it? > (simply, comment that line not to check AX_PTHREAD). > > This requires another modification in the tss/Makefile regard to > pthread library. Please report an issue directly github about this: https://github.com/01org/tpm2.0-tools/issues/new Thanks. > youngwha /Jarkko |
|
From: mipsan.K <mi...@gm...> - 2015-11-05 11:50:59
|
> > Please report an issue directly github about this: > > https://github.com/01org/tpm2.0-tools/issues/new Thanks, I did. > > /Jarkko youngwha -- ///////////////////////////////////////////////// // // // WHY NOT !!! // // LET'S DO IT !!! // // // // mi...@gm... // // // ///////////////////////////////////////////////// |
|
From: mipsan.K <mi...@gm...> - 2015-11-02 01:56:13
|
Hi On Fri, Oct 30, 2015 at 7:25 PM, Jarkko Sakkinen <jar...@li...> wrote: > Hi > > On Thu, Oct 29, 2015 at 07:15:44PM +0900, mipsan.K wrote: >> Hi, >> >> I send reply-mesg to Peter only. >> Apologize. >> On Tue, Oct 27, 2015 at 6:20 PM, Peter Huewe <pet...@gm...> wrote: >> >> Hi >> >> Am 27. Oktober 2015 03:29:22 MEZ, schrieb "mipsan.K" <mi...@gm...>: >> >Dear, >> > >> >Intel *NUC5i5MYHE *has TPM 2.0 chip from Infineon(Infineon >> >*SLB9665TT2.0 *TPM >> >version 2.0). >> > >> >I tested various kernel to activate it, however failed. >> >Even linux-tpmdd(4.3-rc4) as well. >> >That may support only SLB9635/9645 and not SLB9665. >> > >> >Any kinds of advice will be appreciated. >> > >> >Thanks, >> > >> >mipsan >> >> >> As Jarkko said, that will be fixed in 4.4 >> In the meantime you can tryout the security-next tree, or my tpm tree. >> >> >> >> Would you let me know when the 4.4 kernel is available? >> And, does linux-tpmdd(4.3-rc4) tree still have same problem? > > Could you not use HTML for emails? These are pretty difficult to follow > especially when you are using text-only email client. This the second > time I'm asking for this. > > To know when the next release is available you should follow > http://vger.kernel.org/vger-lists.html#linux-kernel-announce > > The fixes that you require should be in 4.4-rc1 release and you get the > annoucement from that list. > > In the meanwhile you can try to compile kernel from this branch: > https://github.com/jsakkine/linux-tpm2/tree/for-peter-v44 > > /Jarkko Sorry for disturbing you by message format. I changed formatting into plain text mode at GMAIL. I will try to build that kernel you suggested. Thanks, youngwha -- |
|
From: Jarkko S. <jar...@li...> - 2015-11-02 11:38:36
|
On Mon, 2015-11-02 at 10:55 +0900, mipsan.K wrote: > Hi > > On Fri, Oct 30, 2015 at 7:25 PM, Jarkko Sakkinen > <jar...@li...> wrote: > > Hi > > > > On Thu, Oct 29, 2015 at 07:15:44PM +0900, mipsan.K wrote: > > > Hi, > > > > > > I send reply-mesg to Peter only. > > > Apologize. > > > On Tue, Oct 27, 2015 at 6:20 PM, Peter Huewe <pet...@gm...> > > > wrote: > > > > > > Hi > > > > > > Am 27. Oktober 2015 03:29:22 MEZ, schrieb "mipsan.K" < > > > mi...@gm...>: > > > >Dear, > > > > > > > >Intel *NUC5i5MYHE *has TPM 2.0 chip from Infineon(Infineon > > > >*SLB9665TT2.0 *TPM > > > >version 2.0). > > > > > > > >I tested various kernel to activate it, however failed. > > > >Even linux-tpmdd(4.3-rc4) as well. > > > >That may support only SLB9635/9645 and not SLB9665. > > > > > > > >Any kinds of advice will be appreciated. > > > > > > > >Thanks, > > > > > > > >mipsan > > > > > > > > > As Jarkko said, that will be fixed in 4.4 > > > In the meantime you can tryout the security-next tree, or my > > > tpm tree. > > > > > > > > > > > > Would you let me know when the 4.4 kernel is available? > > > And, does linux-tpmdd(4.3-rc4) tree still have same problem? > > > > Could you not use HTML for emails? These are pretty difficult to > > follow > > especially when you are using text-only email client. This the > > second > > time I'm asking for this. > > > > To know when the next release is available you should follow > > http://vger.kernel.org/vger-lists.html#linux-kernel-announce > > > > The fixes that you require should be in 4.4-rc1 release and you get > > the > > annoucement from that list. > > > > In the meanwhile you can try to compile kernel from this branch: > > https://github.com/jsakkine/linux-tpm2/tree/for-peter-v44 > > > > /Jarkko > > Sorry for disturbing you by message format. > > I changed formatting into plain text mode at GMAIL. No worries, it's just a practice that we use here and is very common in open source project mailing lists in general. > I will try to build that kernel you suggested. Alright, thank you for your effort to test this feature. Keep me posted and we can sort out the possible issues. > Thanks, > > youngwha /Jarkko |
|
From: mipsan.K <mi...@gm...> - 2015-11-03 05:34:21
|
Hi, > > No worries, it's just a practice that we use here and is very common > in open source project mailing lists in general. > Thank you for your understanding! >> I will try to build that kernel you suggested. > > Alright, thank you for your effort to test this feature. Keep me posted > and we can sort out the possible issues. > >> Thanks, >> >> youngwha > > /Jarkko Both of linux-tpm2(for-peter-v44 branch) and linux-tpmdd(for-james branch) kernel tree returned same log as follow: ------------------------- debian@debian:~$ dmesg | grep -i tpm [ 0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-4.3.0-rc4+ root=UUID=a5e5e761-c2f2-41c3-9dfd-800a73e7a3e6 ro quiet tpm_tis.force=1 [ 0.000000] ACPI: TPM2 0x00000000A1F64EF8 000034 (v03 INTEL NUC5i5MY 00000001 AMI 00000000) [ 0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-4.3.0-rc4+ root=UUID=a5e5e761-c2f2-41c3-9dfd-800a73e7a3e6 ro quiet tpm_tis.force=1 [ 0.844214] tpm_tis tpm_tis: 2.0 TPM (device-id 0x1A, rev-id 16) [ 1.026107] genirq: Flags mismatch irq 8. 00000080 (tpm0) vs. 00000000 (rtc0) [ 1.026110] tpm_tis tpm_tis: Unable to request irq: 8 for probe ------------------------- I ADDed tpm_tis.force=1 to kernel boot command. At this status, if I run getcapability command of IBM TPM 2.0 TSS, the machine returns some data. Additionally, tpm2-util(by jethro G. B) can read vendor/chip name. However, trousers and tpm-tool return error as follow: ------------------------- root@debian:/home/debian# /usr/local/sbin/tcsd -f -e TCSD TDDL ioctl: (25) Inappropriate ioctl for device TCSD TDDL Falling back to Read/Write device support. TCSD TCS ERROR: TCS GetCapability failed with result = 0x1e ------------------------- May I think that Kernel supports TPM 2.0 but TrouSers TSS does not? youngwha |
|
From: Peter H. <pet...@gm...> - 2015-11-03 05:36:29
|
Hi, trousers/tpm tools is 1.2 only Maybe have a look into the tss 2.0 stacks by tcg and ibm Peter -- Sent from my mobile |
|
From: Peter H. <pet...@gm...> - 2015-11-03 07:28:14
|
Hi, > Both of linux-tpm2(for-peter-v44 branch) and linux-tpmdd(for-james branch) > kernel tree returned same log as follow: > > ------------------------- > debian@debian:~$ dmesg | grep -i tpm > [ 0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-4.3.0-rc4+ > root=UUID=a5e5e761-c2f2-41c3-9dfd-800a73e7a3e6 ro quiet > tpm_tis.force=1 > [ 0.000000] ACPI: TPM2 0x00000000A1F64EF8 000034 (v03 INTEL > NUC5i5MY 00000001 AMI 00000000) > [ 0.000000] Kernel command line: > BOOT_IMAGE=/boot/vmlinuz-4.3.0-rc4+ > root=UUID=a5e5e761-c2f2-41c3-9dfd-800a73e7a3e6 ro quiet > tpm_tis.force=1 > [ 0.844214] tpm_tis tpm_tis: 2.0 TPM (device-id 0x1A, rev-id 16) > [ 1.026107] genirq: Flags mismatch irq 8. 00000080 (tpm0) vs. 00000000 (rtc0) > [ 1.026110] tpm_tis tpm_tis: Unable to request irq: 8 for probe > ------------------------- > I ADDed tpm_tis.force=1 to kernel boot command. Does it show the same result, with and without the tpm_tis.force=1? But at least the driver is correctly loaded I guess - /dev/tpm0 is available? > At this status, if I run getcapability command of IBM TPM 2.0 TSS, the > machine returns some data. Good. > Additionally, tpm2-util(by jethro G. B) can read vendor/chip name. Great. Out of curiosity, where can I find that tool? But judging from this information the driver is up and running and your device is good to go. > However, trousers and tpm-tool return error as follow: > ------------------------- > root@debian:/home/debian# /usr/local/sbin/tcsd -f -e > TCSD TDDL ioctl: (25) Inappropriate ioctl for device > TCSD TDDL Falling back to Read/Write device support. > TCSD TCS ERROR: TCS GetCapability failed with result = 0x1e > ------------------------- > May I think that Kernel supports TPM 2.0 but TrouSers TSS does not? Yes - trousers/tpm tools is 1.2 only Maybe have a look into the tss 2.0 stacks by tcg and ibm Thanks, Peter -- Peter Huewe <pet...@gm...> |
|
From: Jarkko S. <jar...@li...> - 2015-11-03 08:02:59
|
On Tue, Nov 03, 2015 at 08:27:59AM +0100, Peter Huewe wrote: > Hi, > > > Both of linux-tpm2(for-peter-v44 branch) and linux-tpmdd(for-james branch) > > kernel tree returned same log as follow: > > > > ------------------------- > > debian@debian:~$ dmesg | grep -i tpm > > [ 0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-4.3.0-rc4+ > > root=UUID=a5e5e761-c2f2-41c3-9dfd-800a73e7a3e6 ro quiet > > tpm_tis.force=1 > > [ 0.000000] ACPI: TPM2 0x00000000A1F64EF8 000034 (v03 INTEL > > NUC5i5MY 00000001 AMI 00000000) > > [ 0.000000] Kernel command line: > > BOOT_IMAGE=/boot/vmlinuz-4.3.0-rc4+ > > root=UUID=a5e5e761-c2f2-41c3-9dfd-800a73e7a3e6 ro quiet > > tpm_tis.force=1 > > [ 0.844214] tpm_tis tpm_tis: 2.0 TPM (device-id 0x1A, rev-id 16) > > [ 1.026107] genirq: Flags mismatch irq 8. 00000080 (tpm0) vs. 00000000 (rtc0) > > [ 1.026110] tpm_tis tpm_tis: Unable to request irq: 8 for probe > > ------------------------- > > I ADDed tpm_tis.force=1 to kernel boot command. > > Does it show the same result, with and without the tpm_tis.force=1? The above is not a TPM driver error. It's a side effect how it probes the IRQ. /Jarkko |
|
From: mipsan.K <mi...@gm...> - 2015-11-03 08:09:10
|
Hi Peter, On Tue, Nov 3, 2015 at 4:27 PM, Peter Huewe <pet...@gm...> wrote: > Hi, > > Does it show the same result, with and without the tpm_tis.force=1? > But at least the driver is correctly loaded I guess - /dev/tpm0 is available? > I added the parameter to test by default. I will test it again without. >> At this status, if I run getcapability command of IBM TPM 2.0 TSS, the >> machine returns some data. > Good. > >> Additionally, tpm2-util(by jethro G. B) can read vendor/chip name. > Great. > Out of curiosity, where can I find that tool? You can find it here: https://github.com/jethrogb/tpm2-utils > > But judging from this information the driver is up and running and your device is good to go. > Thanks for your helpful mention! > > > Yes - trousers/tpm tools is 1.2 only > Maybe have a look into the tss 2.0 stacks by tcg and ibm > Thanks, I will dive into IBM's TSS for TPM 2.0 much more. > > Thanks, > Peter > > -- > Peter Huewe <pet...@gm...> Thanks a lot, youngwha |
|
From: Jarkko S. <jar...@li...> - 2015-11-03 07:36:21
|
On Tue, Nov 03, 2015 at 02:33:56PM +0900, mipsan.K wrote: > Both of linux-tpm2(for-peter-v44 branch) and linux-tpmdd(for-james branch) > kernel tree returned same log as follow: > > ------------------------- > debian@debian:~$ dmesg | grep -i tpm > [ 0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-4.3.0-rc4+ > root=UUID=a5e5e761-c2f2-41c3-9dfd-800a73e7a3e6 ro quiet > tpm_tis.force=1 > [ 0.000000] ACPI: TPM2 0x00000000A1F64EF8 000034 (v03 INTEL > NUC5i5MY 00000001 AMI 00000000) > [ 0.000000] Kernel command line: > BOOT_IMAGE=/boot/vmlinuz-4.3.0-rc4+ > root=UUID=a5e5e761-c2f2-41c3-9dfd-800a73e7a3e6 ro quiet > tpm_tis.force=1 > [ 0.844214] tpm_tis tpm_tis: 2.0 TPM (device-id 0x1A, rev-id 16) > [ 1.026107] genirq: Flags mismatch irq 8. 00000080 (tpm0) vs. 00000000 (rtc0) > [ 1.026110] tpm_tis tpm_tis: Unable to request irq: 8 for probe This is OK, not a bug. When you use the force parameter the TPM driver does a probing process finding a free interrupt vector that it can set for a locality. It's just a warning that kernel interrupt code gives when IRQ cannot be shared. The driver works as expected. > ------------------------- > I ADDed tpm_tis.force=1 to kernel boot command. Starting from v4.4-rc1 you should not need force anymore with tpm_tis. > At this status, if I run getcapability command of IBM TPM 2.0 TSS, the > machine returns some data. > Additionally, tpm2-util(by jethro G. B) can read vendor/chip name. > > However, trousers and tpm-tool return error as follow: > ------------------------- > root@debian:/home/debian# /usr/local/sbin/tcsd -f -e > TCSD TDDL ioctl: (25) Inappropriate ioctl for device > TCSD TDDL Falling back to Read/Write device support. > TCSD TCS ERROR: TCS GetCapability failed with result = 0x1e > ------------------------- > May I think that Kernel supports TPM 2.0 but TrouSers TSS does not? TrouSerS supports only TPM 1.2 and what I've heard there is not interest to add a translation layer (compability mode) to it. > youngwha /Jarkko |
|
From: mipsan.K <mi...@gm...> - 2015-11-03 08:13:43
|
Hi Jarkko, On Tue, Nov 3, 2015 at 4:36 PM, Jarkko Sakkinen <jar...@li...> wrote: > > This is OK, not a bug. When you use the force parameter the TPM driver > does a probing process finding a free interrupt vector that it can set > for a locality. It's just a warning that kernel interrupt code gives > when IRQ cannot be shared. > > The driver works as expected. > Thanks, I'm hesitated by warning message. >> ------------------------- >> I ADDed tpm_tis.force=1 to kernel boot command. > > Starting from v4.4-rc1 you should not need force anymore with tpm_tis. > Okay. > > TrouSerS supports only TPM 1.2 and what I've heard there is not interest > to add a translation layer (compability mode) to it. > TrouSerS was very helpful so far. But I have to find other TSS. >> youngwha > > /Jarkko Thank you very much, youngwha |
|
From: Jarkko S. <jar...@li...> - 2015-11-03 08:18:35
|
On Tue, Nov 03, 2015 at 05:13:18PM +0900, mipsan.K wrote: > Hi Jarkko, > > On Tue, Nov 3, 2015 at 4:36 PM, Jarkko Sakkinen > <jar...@li...> wrote: > > > > This is OK, not a bug. When you use the force parameter the TPM driver > > does a probing process finding a free interrupt vector that it can set > > for a locality. It's just a warning that kernel interrupt code gives > > when IRQ cannot be shared. > > > > The driver works as expected. > > > > Thanks, I'm hesitated by warning message. Yeah, undestandable.It's basically saying that your are trying to request an IRQ that cannot be shared and then the loop continues to the next IRQ number. /Jarkko |
|
From: mipsan.K <mi...@gm...> - 2015-11-04 02:12:59
|
Hi Peter, On Tue, Nov 3, 2015 at 4:27 PM, Peter Huewe <pet...@gm...> wrote: > Hi, > Does it show the same result, with and without the tpm_tis.force=1? > But at least the driver is correctly loaded I guess - /dev/tpm0 is available? > without force=1 parameter, and at linux-tpm2 4.4-rc1, returned as follow: ---------------------------- root@debian:/home/debian# dmesg | grep -i tpm [ 0.000000] Linux version 4.3.0-rc4-tpm2+ (root@debian) (gcc version 4.9.2 (Debian 4.9.2-10) ) #1 SMP Tue Nov 3 20:31:35 EST 2015 [ 0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-4.3.0-rc4-tpm2+ root=UUID=a5e5e761-c2f2-41c3-9dfd-800a73e7a3e6 ro quiet [ 0.000000] ACPI: TPM2 0x00000000A1F64EF8 000034 (v03 INTEL NUC5i5MY 00000001 AMI 00000000) [ 0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-4.3.0-rc4-tpm2+ root=UUID=a5e5e761-c2f2-41c3-9dfd-800a73e7a3e6 ro quiet [ 0.417763] usb usb1: Manufacturer: Linux 4.3.0-rc4-tpm2+ ehci_hcd [ 0.839882] tpm_tis MSFT0101:00: 2.0 TPM (device-id 0x1A, rev-id 16) root@debian:/home/debian# lsmod Module Size Used by tpm_crb 3676 0 x86_pkg_temp_thermal 4727 0 tpm_tis 10223 0 tpm 28797 2 tpm_crb,tpm_tis ---------------------------- As you can find, no IRQ related warning message and tpm module is added(linux-tpmdd does not have tpm module). > > Yes - trousers/tpm tools is 1.2 only > Maybe have a look into the tss 2.0 stacks by tcg and ibm > I tested TPM 2.0 supporting TSS, from IBM and Intel. - IBM: http://sourceforge.net/projects/ibmtpm20tss/ - Intel: https://github.com/01org/TPM2.0-TSS I could get some result only at getcapability command in IBM's TSS as below: ---------------------------- root@debian:/home/debian/Downloads/ibmtpm2/utils# ./getcapability -cap 6 | less moreData: 1 63 properties TPM_PT 00000100 value 322e3000 TPM_PT_FAMILY_INDICATOR - a 4-octet character string containing the TPM Family value (TPM_SPEC_FAMILY) TPM_PT 00000101 value 00000000 TPM_PT_LEVEL - the level of the specification TPM_PT 00000102 value 00000074 TPM_PT_REVISION - the specification Revision times 100 TPM_PT 00000103 value 0000012f TPM_PT_DAY_OF_YEAR - the specification day of year using TCG calendar TPM_PT 00000104 value 000007de TPM_PT_YEAR - the specification year using the CE TPM_PT 00000105 value 49465800 TPM_PT_MANUFACTURER - the vendor ID unique to each TPM manufacturer TPM_PT 00000106 value 534c4239 TPM_PT_VENDOR_STRING_1 - the first four characters of the vendor ID string TPM_PT 00000107 value 36363500 TPM_PT_VENDOR_STRING_2 - the second four characters of the vendor ID string ... ---------------------------- > -- > Peter Huewe <pet...@gm...> It's very hard to test TPM 2.0 TSS like tpm-tools of TPM 1.2. Thanks, youngwha |
|
From: Jarkko S. <jar...@li...> - 2015-11-04 05:37:15
|
On Wed, Nov 04, 2015 at 11:12:32AM +0900, mipsan.K wrote: > Hi Peter, > > On Tue, Nov 3, 2015 at 4:27 PM, Peter Huewe <pet...@gm...> wrote: > > Hi, > > Does it show the same result, with and without the tpm_tis.force=1? > > But at least the driver is correctly loaded I guess - /dev/tpm0 is available? > > > without force=1 parameter, and at linux-tpm2 4.4-rc1, returned as follow: > ---------------------------- > root@debian:/home/debian# dmesg | grep -i tpm > [ 0.000000] Linux version 4.3.0-rc4-tpm2+ (root@debian) (gcc > version 4.9.2 (Debian 4.9.2-10) ) #1 SMP Tue Nov 3 20:31:35 EST 2015 > [ 0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-4.3.0-rc4-tpm2+ > root=UUID=a5e5e761-c2f2-41c3-9dfd-800a73e7a3e6 ro quiet > [ 0.000000] ACPI: TPM2 0x00000000A1F64EF8 000034 (v03 INTEL > NUC5i5MY 00000001 AMI 00000000) > [ 0.000000] Kernel command line: > BOOT_IMAGE=/boot/vmlinuz-4.3.0-rc4-tpm2+ > root=UUID=a5e5e761-c2f2-41c3-9dfd-800a73e7a3e6 ro quiet > [ 0.417763] usb usb1: Manufacturer: Linux 4.3.0-rc4-tpm2+ ehci_hcd > [ 0.839882] tpm_tis MSFT0101:00: 2.0 TPM (device-id 0x1A, rev-id 16) > > root@debian:/home/debian# lsmod > Module Size Used by > tpm_crb 3676 0 > x86_pkg_temp_thermal 4727 0 > tpm_tis 10223 0 > tpm 28797 2 tpm_crb,tpm_tis > ---------------------------- > > As you can find, no IRQ related warning message and tpm module is > added(linux-tpmdd does not have tpm module). > > > > > Yes - trousers/tpm tools is 1.2 only > > Maybe have a look into the tss 2.0 stacks by tcg and ibm > > > > I tested TPM 2.0 supporting TSS, from IBM and Intel. > - IBM: http://sourceforge.net/projects/ibmtpm20tss/ > - Intel: https://github.com/01org/TPM2.0-TSS > > I could get some result only at getcapability command in IBM's TSS as below: > ---------------------------- > root@debian:/home/debian/Downloads/ibmtpm2/utils# ./getcapability -cap 6 | less > moreData: 1 > 63 properties > TPM_PT 00000100 value 322e3000 TPM_PT_FAMILY_INDICATOR - a 4-octet > character string containing the TPM Family value (TPM_SPEC_FAMILY) > TPM_PT 00000101 value 00000000 TPM_PT_LEVEL - the level of the specification > TPM_PT 00000102 value 00000074 TPM_PT_REVISION - the specification > Revision times 100 > TPM_PT 00000103 value 0000012f TPM_PT_DAY_OF_YEAR - the specification > day of year using TCG calendar > TPM_PT 00000104 value 000007de TPM_PT_YEAR - the specification year using the CE > TPM_PT 00000105 value 49465800 TPM_PT_MANUFACTURER - the vendor ID > unique to each TPM manufacturer > TPM_PT 00000106 value 534c4239 TPM_PT_VENDOR_STRING_1 - the first four > characters of the vendor ID string > TPM_PT 00000107 value 36363500 TPM_PT_VENDOR_STRING_2 - the second > four characters of the vendor ID string > ... > ---------------------------- > > -- > > Peter Huewe <pet...@gm...> > > It's very hard to test TPM 2.0 TSS like tpm-tools of TPM 1.2. There is TSS2 based tpm2-tools: https://github.com/01org/tpm2.0-tools > Thanks, > youngwha /Jarkko |
Thanks for helping keep SourceForge clean.
X