All of lore.kernel.org
 help / color / mirror / Atom feed
* Many CVE_PRODUCTS patches
@ 2025-12-30 15:51 Gyorgy Sarvari
  2025-12-31  0:59 ` [oe] " Khem Raj
  2025-12-31  8:12 ` ChenQi
  0 siblings, 2 replies; 4+ messages in thread
From: Gyorgy Sarvari @ 2025-12-30 15:51 UTC (permalink / raw)
  To: openembedded-devel@lists.openembedded.org

Apologies for this CVE_PRODUCT patch-tsunami. 

While reviewing an earlier submitted CVE patch, I was wondering why that
CVE didn't show up in the CVE report - it turned out that almost all
recipes using the pypi class use the default CVE_PRODUCT, however there
are quite a few where it is not appropriate, making the cve-checker to
miss relevant CVEs.

For today I don't plan to spam further, but I'm only about halfway
through my list for the master branch, about the same amount left for
tomorrow-ish.



^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [oe] Many CVE_PRODUCTS patches
  2025-12-30 15:51 Many CVE_PRODUCTS patches Gyorgy Sarvari
@ 2025-12-31  0:59 ` Khem Raj
  2025-12-31  8:12 ` ChenQi
  1 sibling, 0 replies; 4+ messages in thread
From: Khem Raj @ 2025-12-31  0:59 UTC (permalink / raw)
  To: skandigraun; +Cc: openembedded-devel@lists.openembedded.org

[-- Attachment #1: Type: text/plain, Size: 1212 bytes --]

such patches are quite useful as they improve the recipe quality, so keep
it up.

On Tue, Dec 30, 2025 at 7:51 AM Gyorgy Sarvari via lists.openembedded.org
<skandigraun=gmail.com@lists.openembedded.org> wrote:

> Apologies for this CVE_PRODUCT patch-tsunami.
>
> While reviewing an earlier submitted CVE patch, I was wondering why that
> CVE didn't show up in the CVE report - it turned out that almost all
> recipes using the pypi class use the default CVE_PRODUCT, however there
> are quite a few where it is not appropriate, making the cve-checker to
> miss relevant CVEs.
>
> For today I don't plan to spam further, but I'm only about halfway
> through my list for the master branch, about the same amount left for
> tomorrow-ish.
>
>
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
> View/Reply Online (#123037):
> https://lists.openembedded.org/g/openembedded-devel/message/123037
> Mute This Topic: https://lists.openembedded.org/mt/116999150/1997914
> Group Owner: openembedded-devel+owner@lists.openembedded.org
> Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [
> raj.khem@gmail.com]
> -=-=-=-=-=-=-=-=-=-=-=-
>
>

[-- Attachment #2: Type: text/html, Size: 2076 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [oe] Many CVE_PRODUCTS patches
  2025-12-30 15:51 Many CVE_PRODUCTS patches Gyorgy Sarvari
  2025-12-31  0:59 ` [oe] " Khem Raj
@ 2025-12-31  8:12 ` ChenQi
  2025-12-31 10:16   ` Gyorgy Sarvari
  1 sibling, 1 reply; 4+ messages in thread
From: ChenQi @ 2025-12-31  8:12 UTC (permalink / raw)
  To: skandigraun, openembedded-devel@lists.openembedded.org

I think your finding is important.

Yet, I'd suggest we change the default value in pypi.bbclass, removing 
the 'python:' prefix.
That would set a more reasonable value.

Take NVD data as an example:
$ grep 'cpe:2.3:a:python:' ~/.cvedb/nvdcve-2.0-* | grep -v 
':python:python:' | wc -l
159

In all NVD data, from 2002 to 2025 (+ Modified, Recent), there are only 
159 lines that have 'python:' prefix.
That's a very small percentage. To be honest, the number is way below my 
initial guess.

So, could you please try changing the default value in pypi.bbclass to 
see how things work?

Regards,
Qi


On 12/30/25 23:51, Gyorgy Sarvari via lists.openembedded.org wrote:
> Apologies for this CVE_PRODUCT patch-tsunami.
>
> While reviewing an earlier submitted CVE patch, I was wondering why that
> CVE didn't show up in the CVE report - it turned out that almost all
> recipes using the pypi class use the default CVE_PRODUCT, however there
> are quite a few where it is not appropriate, making the cve-checker to
> miss relevant CVEs.
>
> For today I don't plan to spam further, but I'm only about halfway
> through my list for the master branch, about the same amount left for
> tomorrow-ish.
>
>
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
> View/Reply Online (#123037): https://lists.openembedded.org/g/openembedded-devel/message/123037
> Mute This Topic: https://lists.openembedded.org/mt/116999150/7304865
> Group Owner: openembedded-devel+owner@lists.openembedded.org
> Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [Qi.Chen@eng.windriver.com]
> -=-=-=-=-=-=-=-=-=-=-=-
>



^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [oe] Many CVE_PRODUCTS patches
  2025-12-31  8:12 ` ChenQi
@ 2025-12-31 10:16   ` Gyorgy Sarvari
  0 siblings, 0 replies; 4+ messages in thread
From: Gyorgy Sarvari @ 2025-12-31 10:16 UTC (permalink / raw)
  To: ChenQi, openembedded-devel@lists.openembedded.org

On 12/31/25 09:12, ChenQi wrote:
> I think your finding is important.
>
> Yet, I'd suggest we change the default value in pypi.bbclass, removing 
> the 'python:' prefix.
> That would set a more reasonable value.
>
> Take NVD data as an example:
> $ grep 'cpe:2.3:a:python:' ~/.cvedb/nvdcve-2.0-* | grep -v 
> ':python:python:' | wc -l
> 159
>
> In all NVD data, from 2002 to 2025 (+ Modified, Recent), there are only 
> 159 lines that have 'python:' prefix.
> That's a very small percentage. To be honest, the number is way below my 
> initial guess.
>
> So, could you please try changing the default value in pypi.bbclass to 
> see how things work?

I think you are correct, and changing the default to
CVE_PRODUCT ?= "${PYPI_PACKAGE}"
would simplify things a lot, and probably would improve the status quo.
That would make about 80% of these patches obsolete, and I think the
introduced incorrect changes would be minimal (I know that it would
mis-identify as python a few wordpress and jenkins plugins and also 1 or
2 rust projects, but I am talking about ~10 recipes altogether in
meta-oe. All can be trivially corrected with specific CVE_PRODUCT) -
overall I think it would do more good than harm.
Though I haven't checked specifically, but I think there are recipes
with the same issue in oe-core also, which would be helped too.

With that said I am not on oe-core list, and I don't plan to go back
voluntarily- if you (or anyone else) would like to propose it however,
please feel free.

> Regards,
> Qi
>
>
> On 12/30/25 23:51, Gyorgy Sarvari via lists.openembedded.org wrote:
>> Apologies for this CVE_PRODUCT patch-tsunami.
>>
>> While reviewing an earlier submitted CVE patch, I was wondering why that
>> CVE didn't show up in the CVE report - it turned out that almost all
>> recipes using the pypi class use the default CVE_PRODUCT, however there
>> are quite a few where it is not appropriate, making the cve-checker to
>> miss relevant CVEs.
>>
>> For today I don't plan to spam further, but I'm only about halfway
>> through my list for the master branch, about the same amount left for
>> tomorrow-ish.
>>
>>
>> -=-=-=-=-=-=-=-=-=-=-=-
>> Links: You receive all messages sent to this group.
>> View/Reply Online (#123037): https://lists.openembedded.org/g/openembedded-devel/message/123037
>> Mute This Topic: https://lists.openembedded.org/mt/116999150/7304865
>> Group Owner: openembedded-devel+owner@lists.openembedded.org
>> Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [Qi.Chen@eng.windriver.com]
>> -=-=-=-=-=-=-=-=-=-=-=-
>>



^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2025-12-31 10:16 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-12-30 15:51 Many CVE_PRODUCTS patches Gyorgy Sarvari
2025-12-31  0:59 ` [oe] " Khem Raj
2025-12-31  8:12 ` ChenQi
2025-12-31 10:16   ` Gyorgy Sarvari

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.