[SeaBIOS] Re: [PATCH 0/2] nvme fixes

[Cole Robinson]

On 5/26/21 3:49 AM, Gerd Hoffmann wrote:
> 
> 
> 
> 
> Gerd Hoffmann (2):
> 
>   nvme: improve namespace allocation
> 
>   nvme: drive desc should not include the newline
> 
> 
> 
>  src/hw/nvme-int.h |  1 -
> 
>  src/hw/nvme.c | 44 ++--
> 
>  2 files changed, 18 insertions(+), 27 deletions(-)
> 
> 
> 

Thanks, this fixes https://bugzilla.redhat.com/show_bug.cgi?id=1963255
for me

- Cole

___
SeaBIOS mailing list -- seabios@seabios.org
To unsubscribe send an email to seabios-le...@seabios.org

[SeaBIOS] Re: SeaBIOS fails to boot from NVMe controller with lots of namespaces

[Cole Robinson]

On 5/25/21 12:06 PM, Thanos Makatos wrote:
>> -Original Message-
>> From: Cole Robinson 
>> Sent: 25 May 2021 17:01
>> To: Gerd Hoffmann 
>> Cc: Thanos Makatos ; seabios@seabios.org;
>> John Levon ; Swapnil Ingle
>> ; Liu, Changpeng 
>> Subject: Re: [SeaBIOS] Re: SeaBIOS fails to boot from NVMe controller with
>> lots of namespaces
>>
>> On 5/25/21 9:23 AM, Gerd Hoffmann wrote:
>>>   Hi,
>>>
>>>> Where can I file this so it doesn't get lost?
>>>> Anyone know a qemu command line workaround?
>>>
>>> Does
>>> https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.coreboot.org
>>> _hyperkitty_list_seabios-
>> 40seabios.org_thread_2Q7NPH7TJNHK6JGPHQL7755H
>>>
>> ILO23ISN_=DwICAg=s883GpUCOChKOHiocYtGcg=XTpYsh5Ps2zJvtw6
>> ogtti46a
>>> tk736SI4vgsJiUKIyDE=mdjJ1-
>> 1RXDTbTR7f6XC83pvNxgTqd_Yz_1nmiNOsSqk=4E
>>> z7Bvnncx71uPWLUTdweByMiN5ypl3-FzhmrDT2ZcY=
>>> help?
>>>
>>
>> No, tested with qemu.git. Plus the code the patch touches in
>> nvme_controller_enable comes after where the alloc failure happens.
> 
> I haven't upstreamed the relevant QEMU patch yet as I'm waiting for the 
> SeaBIOS part to be reviewed first.
> 
> The alloc failure can be avoided be setting BUILD_MIN_BIOSTABLE to 32K.
> 

Okay I just tried that, on top of master. It's still not enough for
qemu's 256 namespace default, same reported error. If I override it with
`identify->nn = 128;` in nvme_controller_enable then things start working.

Thanks,
Cole

___
SeaBIOS mailing list -- seabios@seabios.org
To unsubscribe send an email to seabios-le...@seabios.org

[SeaBIOS] Re: SeaBIOS fails to boot from NVMe controller with lots of namespaces

[Cole Robinson]

On 5/25/21 9:23 AM, Gerd Hoffmann wrote:
>   Hi,
> 
>> Where can I file this so it doesn't get lost?
>> Anyone know a qemu command line workaround?
> 
> Does
> https://mail.coreboot.org/hyperkitty/list/seabios@seabios.org/thread/2Q7NPH7TJNHK6JGPHQL7755HILO23ISN/
> help?
> 

No, tested with qemu.git. Plus the code the patch touches in
nvme_controller_enable comes after where the alloc failure happens.

- Cole

___
SeaBIOS mailing list -- seabios@seabios.org
To unsubscribe send an email to seabios-le...@seabios.org

[SeaBIOS] Re: SeaBIOS fails to boot from NVMe controller with lots of namespaces

[Cole Robinson]

On 4/16/21 4:46 AM, Thanos Makatos wrote:
> 
> 
>> -Original Message-
>> From: Gerd Hoffmann 
>> Sent: 16 April 2021 08:15
>> To: Thanos Makatos 
>> Cc: Kevin O'Connor ; seabios@seabios.org; John
>> Levon ; Swapnil Ingle
>> ; Liu, Changpeng 
>> Subject: Re: [SeaBIOS] SeaBIOS fails to boot from NVMe controller with lots
>> of namespaces
>>
>>   Hi,
>>
>>> Regarding the failure cases, will things break during build
>> (BUILD_MIN_BIOSTABLE=16K), e.g:
>>>
>>> [seabios] Error!  ROM doesn't fit (135584 > 131072)
>>> [seabios]You have to either increase the size (CONFIG_ROM_SIZE)
>>> [seabios]or turn off some features (such as hardware support not
>>> [seabios]needed) to make it fit.  Trying a more recent gcc version
>>> [seabios]might work too.
>>> [seabios] make: *** [out/bios.bin.prep] Error 1
>>>
>>> Or do we expect undefined behavior at run time?
>>
>> Running out of memory can lead to undefined behavior at run time,
>> depending on which allocations fail.  Typical error pattern is that seabios 
>> can't
>> initialize all devices, leading to boot failures.
>>
>>> 256 namespaces is not an insanely huge number.
>>
>> Well, back in the 80ies when the BIOS interfaces have been created
>> 256 was an insanely huge number ...
>>
>> Given we have a number of real mode constrains for compatibility reasons
>> (like some data structures must live in the f segment) there is no easy way
>> out, we simply can't support an unlimited number of disks (which btw is one
>> of the reasons why the "skip non-bootable disks" code exists).
> 
> SeaBIOS only tries to boot from the first namespace; if that fails other 
> namespaces aren't tried 
> (https://mail.coreboot.org/hyperkitty/list/seabios@seabios.org/thread/72LFLT7KFMWE4GVZHWF4G34PKLVG5LRD/).
> If we stick to the current behavior, then there's no reason to probe any 
> other namespace apart from the first one. This greatly reduces memory 
> requirements so we should be able to support the max number of namespaces.
> 
>>
>> Of course there is the option to leave behind the 80ies and go for UEFI.
> 
> Or can just live with it as you suggest and simply document it .

Sorry for thread necro, but CoreOS devs are hitting this issue too, but
using plain emulated qemu nvme. Fedora bug:
https://bugzilla.redhat.com/show_bug.cgi?id=1963255

Originally a `-device nvme,...` would report only 1 namespace, and
seabios could boot from it fine. After '7f0f1acedf hw/block/nvme:
support multiple namespaces', the device now reports 256 namespaces, and
seabios fails like described in this thread.

Where can I file this so it doesn't get lost?
Anyone know a qemu command line workaround?

Thanks,
Cole

___
SeaBIOS mailing list -- seabios@seabios.org
To unsubscribe send an email to seabios-le...@seabios.org

[SeaBIOS] [PATCH v2] biostables: Support SMBIOS 2.6+ UUID format

[Cole Robinson]

SMBIOS 2.6+ stores the UUID in a different format, with the first 3
fields in little endian format. This is what modern qemu delivers
and what dmidecode also handles, so let's follow suit too. More
info at this thread:

http://www.seabios.org/pipermail/seabios/2015-November/010031.html

The only place this affects is when reporting the UUID at startup.

https://bugzilla.redhat.com/show_bug.cgi?id=1284259
Signed-off-by: Cole Robinson <crobi...@redhat.com>
---
v2:
Update the comment

 src/fw/biostables.c | 47 ---
 1 file changed, 36 insertions(+), 11 deletions(-)

diff --git a/src/fw/biostables.c b/src/fw/biostables.c
index cb74396..9fb9ff9 100644
--- a/src/fw/biostables.c
+++ b/src/fw/biostables.c
@@ -306,17 +306,42 @@ display_uuid(void)
 if (memcmp(uuid, empty_uuid, sizeof(empty_uuid)) == 0)
 return;
 
-printf("Machine UUID"
-   " %02x%02x%02x%02x"
-   "-%02x%02x"
-   "-%02x%02x"
-   "-%02x%02x"
-   "-%02x%02x%02x%02x%02x%02x\n"
-   , uuid[ 0], uuid[ 1], uuid[ 2], uuid[ 3]
-   , uuid[ 4], uuid[ 5]
-   , uuid[ 6], uuid[ 7]
-   , uuid[ 8], uuid[ 9]
-   , uuid[10], uuid[11], uuid[12], uuid[13], uuid[14], 
uuid[15]);
+/*
+ * According to SMBIOS v2.6 the first three fields are encoded in
+ * little-endian format.  Versions prior to v2.6 did not specify
+ * the encoding, but we follow dmidecode and assume big-endian
+ * encoding.
+ */
+if (SMBiosAddr->smbios_major_version > 2 ||
+(SMBiosAddr->smbios_major_version == 2 &&
+ SMBiosAddr->smbios_minor_version >= 6)) {
+printf("Machine UUID"
+   " %02x%02x%02x%02x"
+   "-%02x%02x"
+   "-%02x%02x"
+   "-%02x%02x"
+   "-%02x%02x%02x%02x%02x%02x\n"
+   , uuid[ 3], uuid[ 2], uuid[ 1], uuid[ 0]
+   , uuid[ 5], uuid[ 4]
+   , uuid[ 7], uuid[ 6]
+   , uuid[ 8], uuid[ 9]
+   , uuid[10], uuid[11], uuid[12]
+   , uuid[13], uuid[14], uuid[15]);
+} else {
+printf("Machine UUID"
+   " %02x%02x%02x%02x"
+   "-%02x%02x"
+   "-%02x%02x"
+   "-%02x%02x"
+   "-%02x%02x%02x%02x%02x%02x\n"
+   , uuid[ 0], uuid[ 1], uuid[ 2], uuid[ 3]
+   , uuid[ 4], uuid[ 5]
+   , uuid[ 6], uuid[ 7]
+   , uuid[ 8], uuid[ 9]
+   , uuid[10], uuid[11], uuid[12]
+   , uuid[13], uuid[14], uuid[15]);
+}
+
 return;
 }
 }
-- 
2.5.0


___
SeaBIOS mailing list
SeaBIOS@seabios.org
http://www.seabios.org/mailman/listinfo/seabios

Re: [SeaBIOS] [PATCH] biostables: Support SMBIOS 2.6+ UUID format

[Cole Robinson]

On 01/14/2016 12:22 PM, Kevin O'Connor wrote:
> On Tue, Jan 12, 2016 at 02:22:41PM -0500, Cole Robinson wrote:
>> SMBIOS 2.6+ stores the UUID in a different format, with the first 3
>> fields in little endian format. This is what modern qemu delivers
>> and what dmidecode also handles, so let's follow suit too. More
>> info at this thread:
>>
>> http://www.seabios.org/pipermail/seabios/2015-November/010031.html
>>
>> The only place this affects is when reporting the UUID at startup.
>>
>> https://bugzilla.redhat.com/show_bug.cgi?id=1284259
>> ---
>>  src/fw/biostables.c | 51 ---
>>  1 file changed, 40 insertions(+), 11 deletions(-)
>>
>> diff --git a/src/fw/biostables.c b/src/fw/biostables.c
>> index cb74396..2e15f82 100644
>> --- a/src/fw/biostables.c
>> +++ b/src/fw/biostables.c
>> @@ -306,17 +306,46 @@ display_uuid(void)
>>  if (memcmp(uuid, empty_uuid, sizeof(empty_uuid)) == 0)
>>  return;
>>  
>> -printf("Machine UUID"
>> -   " %02x%02x%02x%02x"
>> -   "-%02x%02x"
>> -   "-%02x%02x"
>> -   "-%02x%02x"
>> -   "-%02x%02x%02x%02x%02x%02x\n"
>> -   , uuid[ 0], uuid[ 1], uuid[ 2], uuid[ 3]
>> -   , uuid[ 4], uuid[ 5]
>> -   , uuid[ 6], uuid[ 7]
>> -   , uuid[ 8], uuid[ 9]
>> -   , uuid[10], uuid[11], uuid[12], uuid[13], uuid[14], 
>> uuid[15]);
>> +/*
>> +* comment borrowed from dmidecode:
>> +*
>> +* As of version 2.6 of the SMBIOS specification, the first 3
>> +* fields of the UUID are supposed to be encoded on 
>> little-endian.
>> +* The specification says that this is the defacto standard,
>> +* however I've seen systems following RFC 4122 instead and use
>> +* network byte order, so I am reluctant to apply the 
>> byte-swapping
>> +* for older versions.
>> +*/
> 
> I find this comment confusing - how about something like:
> 
> According to SMBIOS v2.6 the first three fields are encoded in
> little-endian format.  Versions prior to v2.6 did not specify the
> encoding, but we follow dmidecode and assume big-endian encoding.
> 

That would be fine with me. If Gerd gives his ACK do you want to fix up the
comment or should I post a v2?

Thanks,
Cole

___
SeaBIOS mailing list
SeaBIOS@seabios.org
http://www.seabios.org/mailman/listinfo/seabios

[SeaBIOS] [PATCH] biostables: Support SMBIOS 2.6+ UUID format

[Cole Robinson]

SMBIOS 2.6+ stores the UUID in a different format, with the first 3
fields in little endian format. This is what modern qemu delivers
and what dmidecode also handles, so let's follow suit too. More
info at this thread:

http://www.seabios.org/pipermail/seabios/2015-November/010031.html

The only place this affects is when reporting the UUID at startup.

https://bugzilla.redhat.com/show_bug.cgi?id=1284259
---
 src/fw/biostables.c | 51 ---
 1 file changed, 40 insertions(+), 11 deletions(-)

diff --git a/src/fw/biostables.c b/src/fw/biostables.c
index cb74396..2e15f82 100644
--- a/src/fw/biostables.c
+++ b/src/fw/biostables.c
@@ -306,17 +306,46 @@ display_uuid(void)
 if (memcmp(uuid, empty_uuid, sizeof(empty_uuid)) == 0)
 return;
 
-printf("Machine UUID"
-   " %02x%02x%02x%02x"
-   "-%02x%02x"
-   "-%02x%02x"
-   "-%02x%02x"
-   "-%02x%02x%02x%02x%02x%02x\n"
-   , uuid[ 0], uuid[ 1], uuid[ 2], uuid[ 3]
-   , uuid[ 4], uuid[ 5]
-   , uuid[ 6], uuid[ 7]
-   , uuid[ 8], uuid[ 9]
-   , uuid[10], uuid[11], uuid[12], uuid[13], uuid[14], 
uuid[15]);
+/*
+* comment borrowed from dmidecode:
+*
+* As of version 2.6 of the SMBIOS specification, the first 3
+* fields of the UUID are supposed to be encoded on little-endian.
+* The specification says that this is the defacto standard,
+* however I've seen systems following RFC 4122 instead and use
+* network byte order, so I am reluctant to apply the byte-swapping
+* for older versions.
+*/
+if (SMBiosAddr->smbios_major_version > 2 ||
+(SMBiosAddr->smbios_major_version == 2 &&
+ SMBiosAddr->smbios_minor_version >= 6)) {
+printf("Machine UUID"
+   " %02x%02x%02x%02x"
+   "-%02x%02x"
+   "-%02x%02x"
+   "-%02x%02x"
+   "-%02x%02x%02x%02x%02x%02x\n"
+   , uuid[ 3], uuid[ 2], uuid[ 1], uuid[ 0]
+   , uuid[ 5], uuid[ 4]
+   , uuid[ 7], uuid[ 6]
+   , uuid[ 8], uuid[ 9]
+   , uuid[10], uuid[11], uuid[12]
+   , uuid[13], uuid[14], uuid[15]);
+} else {
+printf("Machine UUID"
+   " %02x%02x%02x%02x"
+   "-%02x%02x"
+   "-%02x%02x"
+   "-%02x%02x"
+   "-%02x%02x%02x%02x%02x%02x\n"
+   , uuid[ 0], uuid[ 1], uuid[ 2], uuid[ 3]
+   , uuid[ 4], uuid[ 5]
+   , uuid[ 6], uuid[ 7]
+   , uuid[ 8], uuid[ 9]
+   , uuid[10], uuid[11], uuid[12]
+   , uuid[13], uuid[14], uuid[15]);
+}
+
 return;
 }
 }
-- 
2.5.0


___
SeaBIOS mailing list
SeaBIOS@seabios.org
http://www.seabios.org/mailman/listinfo/seabios

[SeaBIOS] bug: incorrect uuid in seabios output

[Cole Robinson]

Hi,

The UUID seabios reports in its boot output doesn't match what is passed via
qemu -uuid option. An example is reported here:

https://bugzilla.redhat.com/show_bug.cgi?id=1284259

This is due to:

commit caad057bb6ce86a9cb71520af395fd0bd04a659f
Author: Eduardo Habkost 
Date:   Wed Oct 29 11:26:08 2014 -0200

smbios: Encode UUID according to SMBIOS specification

Differently from older versions, SMBIOS version 2.6 is explicit about
the encoding of UUID fields:

> Although RFC 4122 recommends network byte order for all fields, the PC
> industry (including the ACPI, UEFI, and Microsoft specifications) has
> consistently used little-endian byte encoding for the first three fields:
> time_low, time_mid, time_hi_and_version. The same encoding, also known as
> wire format, should also be used for the SMBIOS representation of the 
UUID.
>
> The UUID {00112233-4455-6677-8899-AABBCCDDEEFF} would thus be represented
> as 33 22 11 00 55 44 77 66 88 99 AA BB CC DD EE FF.

The dmidecode tool implements this and decodes the above "wire format"
when SMBIOS version >= 2.6. We moved from SMBIOS version 2.4 to 2.8 when
we started building the SMBIOS entry point inside QEMU, on commit
c97294ec1b9e36887e119589d456557d72ab37b5.


seabios doesn't seem to handle this special UUID format when reading from 
smbios.

Thanks,
Cole

___
SeaBIOS mailing list
SeaBIOS@seabios.org
http://www.seabios.org/mailman/listinfo/seabios

Re: [SeaBIOS] [Qemu-devel] seabios 1.7.3 breaks booting windows 7?

[Cole Robinson]

On 07/08/2013 08:41 PM, Kevin O'Connor wrote:
 On Mon, Jul 08, 2013 at 08:18:42PM -0400, Cole Robinson wrote:
 On 07/08/2013 07:18 PM, Kevin O'Connor wrote:
 On Mon, Jul 08, 2013 at 05:23:40PM -0400, Cole Robinson wrote:
 On 07/07/2013 11:41 PM, Kevin O'Connor wrote:
 The 1.7.3 version of SeaBIOS has now been released.  For more
 information on the release, please see:

 Just gave it a spin, breaks booting a Windows 7 VM for me. Using qemu-1.4.2
 from Fedora 19 as well as qemu.git.

 Here's the BSOD (F19 qemu and seabios 1.7.3):
 http://crobinso.fedorapeople.org/misc/seabios_1.7.3_windows_err.png

 Thanks for testing.  I have a Windows 7 Ultimate beta (build 7100)
 that I tested with and I don't see this problem.  (Nor do I see the
 problem on WinXP.)

 Can you try without smp and with a smaller memory size (eg, 512M) to
 see if the behavior changes.


 Hmm, my winxp 32 guest is reproducing as well. And I can reproduce with just
 the windows 7 media (mine is ultimate N, came off MSDN years ago):

 ~/src/qemu/x86_64-softmmu/qemu-system-x86_64 -m 512 -sdl -cdrom
 /mnt/data/devel/media/win7_64.iso -boot d -bios ~/src/seabios/out/bios.bin

 vista and windows 8 preview media are reproducing for me too.

 Maybe it's a toolchain thing, anyone else on Fedora 19 that can independently
 confirm?
 
 Ugh - make sure you aren't using an old acpi-dsdt.aml file.  I suggest
 following the steps outlined in the SeaBIOS README file for creating a
 mybiosdir/ directory and use the QEMU -L option to point to that
 directory.
 

Indeed that was it, thanks Kevin, sorry for the noise.

- Cole


___
SeaBIOS mailing list
SeaBIOS@seabios.org
http://www.seabios.org/mailman/listinfo/seabios

[SeaBIOS] seabios 1.7.3 breaks booting windows 7? (was: [ANNOUNCE] SeaBIOS 1.7.3)

[Cole Robinson]

On 07/07/2013 11:41 PM, Kevin O'Connor wrote:
 The 1.7.3 version of SeaBIOS has now been released.  For more
 information on the release, please see:
 
 http://seabios.org/Releases
 
 
 New in this release:
 
 * Initial support for using SeaBIOS as a UEFI Compatibility Support
   Module (CSM)
 * Support for detecting and using ACPI reboot ports.
 * By default, all 16bit entry points now use an internal stack to
   reduce stack footprint.
 * Floppy controller code has been rewritten to improve
   compatibility. Non-standard floppy sizes now work again with recent
   QEMU versions.
 * Several bug fixes and code cleanups
 
 

Just gave it a spin, breaks booting a Windows 7 VM for me. Using qemu-1.4.2
from Fedora 19 as well as qemu.git.

Here's the BSOD (F19 qemu and seabios 1.7.3):
http://crobinso.fedorapeople.org/misc/seabios_1.7.3_windows_err.png

I bisected to:

commit 76e58028d28e78431f9de3cee0b3c88d807fa39d
Author: Kevin O'Connor ke...@koconnor.net
Date:   Wed Mar 6 21:50:09 2013 -0500

acpi: Eliminate BDAT parameter passing to DSDT code.


Reproducing command line:

x86_64-softmmu/qemu-system-x86_64 \
-bios /usr/share/seabios/bios.bin \
-machine pc-i440fx-1.4,accel=kvm,usb=off \
-m 4096 -smp 4 \
-sdl -vga std \
win7.qcow2

Let me know if there's any more info I can provide.

Thanks,
Cole

___
SeaBIOS mailing list
SeaBIOS@seabios.org
http://www.seabios.org/mailman/listinfo/seabios

Re: [SeaBIOS] [Qemu-devel] seabios 1.7.3 breaks booting windows 7?

[Cole Robinson]

On 07/08/2013 07:18 PM, Kevin O'Connor wrote:
 On Mon, Jul 08, 2013 at 05:23:40PM -0400, Cole Robinson wrote:
 On 07/07/2013 11:41 PM, Kevin O'Connor wrote:
 The 1.7.3 version of SeaBIOS has now been released.  For more
 information on the release, please see:

 Just gave it a spin, breaks booting a Windows 7 VM for me. Using qemu-1.4.2
 from Fedora 19 as well as qemu.git.

 Here's the BSOD (F19 qemu and seabios 1.7.3):
 http://crobinso.fedorapeople.org/misc/seabios_1.7.3_windows_err.png
 
 Thanks for testing.  I have a Windows 7 Ultimate beta (build 7100)
 that I tested with and I don't see this problem.  (Nor do I see the
 problem on WinXP.)
 
 Can you try without smp and with a smaller memory size (eg, 512M) to
 see if the behavior changes.
 

Hmm, my winxp 32 guest is reproducing as well. And I can reproduce with just
the windows 7 media (mine is ultimate N, came off MSDN years ago):

~/src/qemu/x86_64-softmmu/qemu-system-x86_64 -m 512 -sdl -cdrom
/mnt/data/devel/media/win7_64.iso -boot d -bios ~/src/seabios/out/bios.bin

vista and windows 8 preview media are reproducing for me too.

Maybe it's a toolchain thing, anyone else on Fedora 19 that can independently
confirm?

 Otherwise, we'll need to iterate through each change in the 76e58028
 commit and find out which part of the DSDT that your copy of Windows
 doesn't like.


I'll probably need some hand holding through that, I can't tell from looking
at the patch what parts are independent.

- Cole

___
SeaBIOS mailing list
SeaBIOS@seabios.org
http://www.seabios.org/mailman/listinfo/seabios