Hi Linus,
please pull these changes to your tree. They are cleaning up the PCI bus support
which was pretty much c&p from PowerPC long time ago for one custom platform
which is not available for years. Finally Xilinx/AMD PCIe team tested Microblaze
with IP cores also used on ARM SOCs and clean up Microblaze code.
Thanks,
Michal
The following changes since commit 9abf2313adc1ca1b6180c508c25f22f9395cc780:
Linux 6.1-rc1 (2022-10-16 15:36:24 -0700)
are available in the Git repository at:
git://git.monstr.eu/linux-2.6-microblaze.git tags/microblaze-v6.2
for you to fetch changes up to 5cfe469c2654c3a4dda8504b1209abdd8b379e95:
microblaze/PCI: Moving PCI iounmap and dependent code (2022-11-25 11:39:23 +0100)
----------------------------------------------------------------
Microblaze patches for 6.2-rc1
- Cleanup PCI support
----------------------------------------------------------------
Thippeswamy Havalige (13):
microblaze/PCI: Remove unused early_read_config_byte() et al declarations
microblaze/PCI: Remove Null PCI config access unused functions
microblaze/PCI: Remove unused PCI bus scan if configured as a host
microblaze/PCI: Remove unused PCI legacy IO's access on a bus
microblaze/PCI: Remove unused device tree parsing for a host bridge resources
microblaze/PCI: Remove unused allocation & free of PCI host bridge structure
microblaze/PCI: Remove unused PCI BIOS resource allocation
microblaze/PCI: Remove unused PCI Indirect ops
microblaze/PCI: Remove unused pci_address_to_pio() conversion of CPU
address to I/O port
microblaze/PCI: Remove unused sys_pciconfig_iobase() and et al declaration
microblaze/PCI: Remove unused pci_iobar_pfn() and et al declarations
microblaze/PCI: Remove support for Xilinx PCI host bridge
microblaze/PCI: Moving PCI iounmap and dependent code
arch/microblaze/Kconfig | 8 -
arch/microblaze/configs/mmu_defconfig | 1 -
arch/microblaze/include/asm/pci-bridge.h | 92 --
arch/microblaze/include/asm/pci.h | 29 -
arch/microblaze/pci/Makefile | 3 +-
arch/microblaze/pci/indirect_pci.c | 158 ---
arch/microblaze/pci/iomap.c | 36 +
arch/microblaze/pci/pci-common.c | 1067 --------------------
arch/microblaze/pci/xilinx_pci.c | 170 ----
9 files changed, 37 insertions(+), 1527 deletions(-)
delete mode 100644 arch/microblaze/pci/indirect_pci.c
delete mode 100644 arch/microblaze/pci/pci-common.c
delete mode 100644 arch/microblaze/pci/xilinx_pci.c
--
Michal Simek, Ing. (M.Eng), OpenPGP -> KeyID: FE3D1F91
w: http://www.monstr.eu p: +42-0-721842854
Maintainer of Linux kernel - Xilinx Microblaze
Maintainer of Linux kernel - Xilinx Zynq ARM and ZynqMP ARM64 SoCs
U-Boot custodian - Xilinx Microblaze/Zynq/ZynqMP/Versal SoCs
On Thu, Dec 8, 2022 at 2:07 AM Michal Simek <[email protected]> wrote:
>
> Microblaze patches for 6.2-rc1
>
> - Cleanup PCI support
That tag looks normal in the email, but it's actually nastily encoded.
When I fetched it, I did this:
git cat-file tag FETCH_HEAD | hexdump -c
and it shows
...
0000080 9 3 5 4 6 + 0 1 0 0 \n \n M i c
0000090 r o b l a z e 302 240 p a t c h e s
00000a0 302 240 f o r 302 240 6 . 2 - r c 1 \n \n
00000b0 - C l e a n u p P C I s u
00000c0 p p o r t \n - - - - - B E G I N
...
Note how the "spaces" between "Microblaze", "patches", "for" and "6.2"
are not actually regular space characters, but \xc2\a0, which is utf-8
encoding for unicode character A0 ("no-break space").
Please don't do that. I have my editor show odd characters as hex
codes (I don't want any hidden information), which is why I noticed,
and I'm not sure why/how you did it.
Linus
The pull request you sent on Thu, 8 Dec 2022 11:07:27 +0100:
> git://git.monstr.eu/linux-2.6-microblaze.git tags/microblaze-v6.2
has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/164f59000c19fa1ee5d09327a8055ec9f9b9905a
Thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/prtracker.html
On 12/12/22 18:49, Linus Torvalds wrote:
> On Thu, Dec 8, 2022 at 2:07 AM Michal Simek <[email protected]> wrote:
>>
>> Microblaze patches for 6.2-rc1
>>
>> - Cleanup PCI support
>
> That tag looks normal in the email, but it's actually nastily encoded.
> When I fetched it, I did this:
>
> git cat-file tag FETCH_HEAD | hexdump -c
>
> and it shows
>
> ...
> 0000080 9 3 5 4 6 + 0 1 0 0 \n \n M i c
> 0000090 r o b l a z e 302 240 p a t c h e s
> 00000a0 302 240 f o r 302 240 6 . 2 - r c 1 \n \n
> 00000b0 - C l e a n u p P C I s u
> 00000c0 p p o r t \n - - - - - B E G I N
> ...
>
> Note how the "spaces" between "Microblaze", "patches", "for" and "6.2"
> are not actually regular space characters, but \xc2\a0, which is utf-8
> encoding for unicode character A0 ("no-break space").
>
> Please don't do that. I have my editor show odd characters as hex
> codes (I don't want any hidden information), which is why I noticed,
> and I'm not sure why/how you did it.
It is not done intentionally. I am looking at previous commit messages and I do
c&p the first line from firefox via gitweb interface. This is style I use for
years but I see when I do it now that there are added additional chars as you
mentioned above.
I tracked it down that it started to behave like this between Oct 11/Nov 22
(from my U-Boot tags) and it is actually firefox (which was upgraded) who is
adding additional chars there. When I do the same style via chrome I can't see
any issue.
I will start to write the whole message by hand without c&p to make sure that it
won't happen again.
Thanks,
Michal
--
Michal Simek, Ing. (M.Eng), OpenPGP -> KeyID: FE3D1F91
w: http://www.monstr.eu p: +42-0-721842854
Maintainer of Linux kernel - Xilinx Microblaze
Maintainer of Linux kernel - Xilinx Zynq ARM and ZynqMP/Versal ARM64 SoCs
U-Boot custodian - Xilinx Microblaze/Zynq/ZynqMP/Versal/Versal NET SoCs