linux-kernel - Re: Re: [PATCH v3] PCI: Add vendor ID for QUECTEL

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <Y2C7uhzVorUrfQA2@kroah.com>
Date:   Tue, 1 Nov 2022 07:24:58 +0100
From:   Greg KH <gregkh@...uxfoundation.org>
To:     Slark Xiao <slark_xiao@....com>
Cc:     mani@...nel.org, quic_hemantk@...cinc.com, bhelgaas@...gle.com,
        loic.poulain@...aro.org, dnlplm@...il.com, yonglin.tan@...look.com,
        mhi@...ts.linux.dev, linux-arm-msm@...r.kernel.org,
        linux-kernel@...r.kernel.org, linux-pci@...r.kernel.org
Subject: Re: Re: [PATCH v3] PCI: Add vendor ID for QUECTEL

On Tue, Nov 01, 2022 at 02:09:57PM +0800, Slark Xiao wrote:
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> At 2022-11-01 12:46:19, "Greg KH" <gregkh@...uxfoundation.org> wrote:
> >On Tue, Nov 01, 2022 at 10:10:52AM +0800, Slark Xiao wrote:
> >> n MHI driver, there are some companies' product still do not have their
> >> own PCI vendor macro. So we add it here to make the code neat. Ref ID
> >> could be found in link https://pcisig.com/membership/member-companies.
> >> 
> >> Signed-off-by: Slark Xiao <slark_xiao@....com>
> >> ---
> >> v3: Separate different vendors into different patch.
> >> 
> >> v2: Update vendor ID to the right location sorted by numeric value.
> >> ---
> >>  drivers/bus/mhi/host/pci_generic.c | 6 +++---
> >>  include/linux/pci_ids.h            | 2 ++
> >>  2 files changed, 5 insertions(+), 3 deletions(-)
> >> 
> >> diff --git a/drivers/bus/mhi/host/pci_generic.c b/drivers/bus/mhi/host/pci_generic.c
> >> index caa4ce28cf9e..81ae9c49ce2a 100644
> >> --- a/drivers/bus/mhi/host/pci_generic.c
> >> +++ b/drivers/bus/mhi/host/pci_generic.c
> >> @@ -555,11 +555,11 @@ static const struct pci_device_id mhi_pci_id_table[] = {
> >>  		.driver_data = (kernel_ulong_t) &mhi_telit_fn990_info },
> >>  	{ PCI_DEVICE(PCI_VENDOR_ID_QCOM, 0x0308),
> >>  		.driver_data = (kernel_ulong_t) &mhi_qcom_sdx65_info },
> >> -	{ PCI_DEVICE(0x1eac, 0x1001), /* EM120R-GL (sdx24) */
> >> +	{ PCI_DEVICE(PCI_VENDOR_ID_QUECTEL, 0x1001), /* EM120R-GL (sdx24) */
> >>  		.driver_data = (kernel_ulong_t) &mhi_quectel_em1xx_info },
> >> -	{ PCI_DEVICE(0x1eac, 0x1002), /* EM160R-GL (sdx24) */
> >> +	{ PCI_DEVICE(PCI_VENDOR_ID_QUECTEL, 0x1002), /* EM160R-GL (sdx24) */
> >>  		.driver_data = (kernel_ulong_t) &mhi_quectel_em1xx_info },
> >> -	{ PCI_DEVICE(0x1eac, 0x2001), /* EM120R-GL for FCCL (sdx24) */
> >> +	{ PCI_DEVICE(PCI_VENDOR_ID_QUECTEL, 0x2001), /* EM120R-GL for FCCL (sdx24) */
> >>  		.driver_data = (kernel_ulong_t) &mhi_quectel_em1xx_info },
> >>  	/* T99W175 (sdx55), Both for eSIM and Non-eSIM */
> >>  	{ PCI_DEVICE(PCI_VENDOR_ID_FOXCONN, 0xe0ab),
> >> diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
> >> index b362d90eb9b0..3c91461bcfe4 100644
> >> --- a/include/linux/pci_ids.h
> >> +++ b/include/linux/pci_ids.h
> >> @@ -2585,6 +2585,8 @@
> >>  #define PCI_VENDOR_ID_TEKRAM		0x1de1
> >>  #define PCI_DEVICE_ID_TEKRAM_DC290	0xdc29
> >>  
> >> +#define PCI_VENDOR_ID_QUECTEL		0x1eac
> >
> >Why did you ignore the comment at the top of this file saying that new
> >entries are not needed to be added, especially for just one user?
> >
> >thanks,
> >
> >greg k-h
> Hi Greg,
>  Actually I didn't see this notice before committing this patch. I even discussed 
> it with the maintainer for several times and nobody show me this rule.
> I have a concern, some IOT module vendors, like QUECTEL, CINTERION(THALES),
> SIERRA,ROLLING and so on, they only produce IOT modules without other 
> hardware with PCIe  interface, and they applied for their own VID. But they
> can't get a their own VENDOR MARCO? This seems unreasonable.
> This change should be harmless and  make the code neat.
> This is my opinion.

It causes a _LOT_ of churn and merge issues when everyone is adding new
entries to a single file.  Which is why, 15+ years ago, we made the
decision that if a vendor or device id is only needed in one file, then
it should not be added to the pci_ids.h file.

No need to change that now, please just put the vendor id in the single
driver that it is needed in.

thanks,

greg k-h