Raspberry Pi4: add support for DFU over USB
Subsystems (2)
A message that is not in the archive (sent off-list or not yet received): CGME20191202111128eucas1p1398c5729ba6af763c7866cfd5462e3d1@eucas1p1.samsung.com
v4 cover Raspberry Pi4: add support for DFU over USB
Hi All!
This patchset enables support for DFU over USB protocol on Raspberry Pi4 board. The board has DWC2 UDC controller connected to the USB-C power connector. Enabling DFU on it, make the u-boot development much more convenient, as one no longer needs to swap SD-card between RPi4 board and host machine to update the u-boot binary.
Patches are based on current 'master' u-boot branch. They were tested on the 2019-07-10-raspbian-buster-lite.img sd-card image with the following lines added to config.txt:
dtoverlay=dwc2,dr_mode=peripheral
dtparam=i2c_arm=on
dtparam=spi=on
enable_uart=1
uart_2ndstage=1
kernel=u-boot.bin
To enable DFU, one has to enter follwing command:
# dfu 0 mmc 0
During the development of this feature I've encountered a serious bugs in FAT write code. Over-writing discontiguous files always caused serious filesystem corruption. This was especially anoying, because the system environment is kept on FAT volume in uboot.env file, so 'saveenv' basically corrupted the boot partiting on the second call. Another bunch of the issues in the FAT write code has been revealed while removing predefined file size limit in DFU MMC code and then running sandbox tests.
I hope that my fixes for FAT code will be helpful for non-RPi users too.
Best regards
Marek Szyprowski
Samsung R&D Institute Poland
Changelog:
v4:
- rechecked the FAT related fixes, it turned out that much simpler patch fixes both issues discovered while working on DFU support; added simple sandbox based tests reveleaing the issue and showing correctness of the fix
- rebased patches onto current u-boot's master branch: 4b19b89ca4a8 ("Merge tag 'rpi-next-2020.01' of https://github.com/mbgg/u-boot")
v3: https://patchwork.ozlabs.org/cover/1200793/
- fixed one more FAT issue revealed by sandbox tests
v3: (patch 6/6 posted separately): https://patchwork.ozlabs.org/patch/1195645/
- fixed non-RPi4 builds (missing #else ENV_DFU_SETTINGS def)
- removed config.txt entity (not needed in uboot-based boot)
- switched arm64 kernel filename to 'Image'
v2: https://patchwork.ozlabs.org/cover/1166589/
- added changes to rpi_4_defconfig too (arm64 version)
- extended DFU entity list by confix.txt, cmdline.txt and Image.gz
- fixed missing '\0' at the end of dfu_alt_info env
- added reviewed-by tags
- added patches for DFU MMC to remove file size limit
- added patch for fat write to fix issues on non-zero file offset (revealed by previous patch)
v1: https://patchwork.ozlabs.org/cover/1162770/
- initial version
Patch summary:
Marek Szyprowski (6):
fat: write: fix broken write to fragmented files
fat: write: adjust data written in each partial write
dfu: mmc: rearrange the code
dfu: mmc: remove file size limit for io operations
usb: dwc2_udc_otg: add bcm2835 SoC (Raspberry Pi4) support
config: enable DFU over USB on Raspberry Pi4 boards
configs/rpi_4_32b_defconfig | 11 +++ configs/rpi_4_defconfig | 11 +++ drivers/dfu/dfu_mmc.c | 93 +++++++++++++--------- drivers/usb/gadget/dwc2_udc_otg.c | 2 + drivers/usb/gadget/dwc2_udc_otg_xfer_dma.c | 12 +-- fs/fat/fat_write.c | 8 +- include/configs/rpi.h | 20 +++++ 7 files changed, 112 insertions(+), 45 deletions(-)
Signature · 1 line
2.17.1
PATCH v4 1/6 fat: write: fix broken write to fragmented files
The code for handing file overwrite incorrectly assumed that the file on disk is always contiguous. This resulted in corrupting disk structure every time when write to existing fragmented file happened. Fix this by adding proper check for cluster discontinuity and adjust chunk size on each partial write.
---
This patch partially fixes the issue revealed by the following test script:
--->8-fat_test1.sh---
#!/bin/bash
make sandbox_defconfig
make
dd if=/dev/zero of=/tmp/10M.img bs=1024 count=10k
mkfs.vfat -v /tmp/10M.img
cat >/tmp/cmds <<EOF
x
host bind 0 /tmp/10M.img
fatls host 0
mw 0x1000000 0x0a434241 0x1000 # "ABC\n"
mw 0x1100000 0x0a464544 0x8000 # "DEF\n"
fatwrite host 0 0x1000000 file0001.raw 0x1000
fatwrite host 0 0x1000000 file0002.raw 0x1000
fatwrite host 0 0x1000000 file0003.raw 0x1000
fatwrite host 0 0x1000000 file0004.raw 0x1000
fatwrite host 0 0x1000000 file0005.raw 0x1000
fatrm host 0 file0002.raw
fatrm host 0 file0004.raw
fatls host 0
fatwrite host 0 0x1100000 file0007.raw 0x4000
fatwrite host 0 0x1100000 file0007.raw 0x4000
reset
EOF
./u-boot </tmp/cmds
#verify
rm -r /tmp/result /tmp/model
mkdir /tmp/result
mkdir /tmp/model
yes ABC | head -c 4096 >/tmp/model/file0001.raw yes ABC | head -c 4096 >/tmp/model/file0003.raw yes ABC | head -c 4096 >/tmp/model/file0005.raw yes DEF | head -c 16384 >/tmp/model/file0007.raw
mcopy -n -i /tmp/10M.img ::file0001.raw /tmp/result mcopy -n -i /tmp/10M.img ::file0003.raw /tmp/result mcopy -n -i /tmp/10M.img ::file0005.raw /tmp/result mcopy -n -i /tmp/10M.img ::file0007.raw /tmp/result hd /tmp/10M.img
if diff -urq /tmp/model /tmp/result
then
echo Test okay
else
echo Test fail
fi
--->8---
Overwritting a discontiguous test file (file0007.raw) no longer causes corruption to file0003.raw, which's data lies between the chunks of the test file. The amount of data written to disk is still incorrect, what causes damage to the file (file0005.raw), which's data lies next to the test file. This will be fixed by the next patch.
Feel free to prepare a proper sandbox/py_test based tests based on the provided test scripts.
--- fs/fat/fat_write.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
fs/fat/fat_write.c +2 -2
@@ -794,6 +794,8 @@ set_contents(fsdata *mydata, dir_entry *dentptr, loff_t pos, __u8 *buffer, newclust =get_fatent (mydata, endclust); +if (newclust != endclust +1 ) +break ;if (IS_LAST_CLUST (newclust, mydata->fatsize))break ;if (CHECK_CLUST (newclust, mydata->fatsize)) {
@@ -824,8 +826,6 @@ set_contents(fsdata *mydata, dir_entry *dentptr, loff_t pos, __u8 *buffer,if (filesize <= cur_pos)break ; -/* CHECK: newclust = get_fatent(mydata, endclust); */ -if (IS_LAST_CLUST (newclust, mydata->fatsize))/* no more clusters */ break ;
Signature · 1 line
2.17.1
PATCH v4 1/6 fat: write: fix broken write to fragmented files
Hi Tom, Matthias,
Quoted 6 lines
The code for handing file overwrite incorrectly assumed that the file on disk is always contiguous. This resulted in corrupting disk structure every time when write to existing fragmented file happened. Fix this by adding proper check for cluster discontinuity and adjust chunk size on each partial write.
Quoted 6 lines
Signed-off-by Marek Szyprowski <[email protected]>---
This patch partially fixes the issue revealed by the following test script:
Tom could you pic this patch and the following one (2/6):
https://patchwork.ozlabs.org/patch/1203101/
to -master as a fix?
This seems like a _real_ fix for FAT.
The dfu part of this series IMHO shall be grabbed by Matthias or me into the -next branch of u-boot-dfu/rpi4 tree.
I do guess that Matthias shall fetch this series as he is assigned to it in the patchwork?
I'm fine for this as I've already acked / reviewed DFU part of this series.
Quoted 72 lines
--->8-fat_test1.sh---
#!/bin/bash
makesandbox_defconfig
make
dd if=/dev/zero of=/tmp/10M.img bs=1024 count=10k
mkfs.vfat -v/tmp/10M.img
cat >/tmp/cmds <<EOF
x
host bind 0/tmp/10M.img
fatls host 0
mw0x10000000x0a4342410x1000# "ABC\n"
mw0x11000000x0a4645440x8000# "DEF\n"
fatwrite host 00x1000000file0001.raw0x1000
fatwrite host 00x1000000file0002.raw0x1000
fatwrite host 00x1000000file0003.raw0x1000
fatwrite host 00x1000000file0004.raw0x1000
fatwrite host 00x1000000file0005.raw0x1000
fatrm host 0 file0002.raw
fatrm host 0 file0004.raw
fatls host 0
fatwrite host 00x1100000file0007.raw0x4000
fatwrite host 00x1100000file0007.raw0x4000
reset
EOF
./u-boot </tmp/cmds
#verify
rm -r/tmp/result/tmp/model
mkdir/tmp/result
mkdir/tmp/modelyes ABC | head -c 4096 >/tmp/model/file0001.raw yes ABC | head -c 4096 >/tmp/model/file0003.raw yes ABC | head -c 4096 >/tmp/model/file0005.raw yes DEF | head -c 16384 >/tmp/model/file0007.rawmcopy -n -i
/tmp/10M.img::file0001.raw/tmp/resultmcopy -n -i/tmp/10M.img::file0003.raw/tmp/resultmcopy -n -i/tmp/10M.img::file0005.raw/tmp/resultmcopy -n -i/tmp/10M.img::file0007.raw/tmp/resulthd/tmp/10M.img
if diff -urq/tmp/model/tmp/result
then
echo Test okay
else
echo Test fail
fi
--->8---Overwritting a discontiguous test file (file0007.raw) no longer causes corruption to file0003.raw, which's data lies between the chunks of the test file. The amount of data written to disk is still incorrect, what causes damage to the file (file0005.raw), which's data lies next to the test file. This will be fixed by the next patch.
Feel free to prepare a proper sandbox/py_test based tests based on the provided test scripts.
--- fs/fat/fat_write.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)fs/fat/fat_write.c +2
-0@@ -794,6 +794,8 @@ set_contents(fsdata *mydata, dir_entry *dentptr, loff_t pos, __u8 *buffer, newclust =get_fatent (mydata, endclust); +if (newclust != endclust +1 ) +break ;if (IS_LAST_CLUST (newclust, mydata->fatsize))break ;if (CHECK_CLUST (newclust, mydata->fatsize)) {
fs/fat/fat_write.c set_contents(fsdata *mydata, dir_entry *dentptr, · line 831
Quoted 8 lines
fs/fat/fat_write.c +0
-2@@ -824,8 +826,6 @@ set_contents(fsdata *mydata, dir_entry *dentptr, loff_t pos, __u8 *buffer, if (filesize <= cur_pos) break; -/* CHECK: newclust = get_fatent(mydata, endclust); */ -if (IS_LAST_CLUST (newclust, mydata->fatsize))/* no more clusters */ break ;
Best regards,
Lukasz Majewski
--
DENX Software Engineering GmbH, Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-59 Fax: (+49)-8142-66989-80 Email: lukma at denx.de -------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 488 bytes
Desc: OpenPGP digital signature
URL: <https://lists.denx.de/pipermail/u-boot/attachments/20191205/b2f0d7c5/attachment.sig>
PATCH v4 1/6 fat: write: fix broken write to fragmented files
Quoted 15 lines · Lukasz Majewski
Hi Tom, Matthias,
Quoted 6 lines
The code for handing file overwrite incorrectly assumed that the file on disk is always contiguous. This resulted in corrupting disk structure every time when write to existing fragmented file happened. Fix this by adding proper check for cluster discontinuity and adjust chunk size on each partial write.
Quoted 6 lines
Signed-off-by Marek Szyprowski <[email protected]>---
This patch partially fixes the issue revealed by the following test script:
Quoted 6 lines
Tom could you pic this patch and the following one (2/6):
https://patchwork.ozlabs.org/patch/1203101/to -master as a fix?
This seems like a _real_ fix for FAT.
Right, I think the first patches should go in for v2020.01. I can send them together with some fixes for RPi I'm working on.
Tom what do you think?
Regards,
Matthias
Quoted 95 lines
The dfu part of this series IMHO shall be grabbed by Matthias or me into the -next branch of u-boot-dfu/rpi4 tree.
I do guess that Matthias shall fetch this series as he is assigned to it in the patchwork?
I'm fine for this as I've already acked / reviewed DFU part of this series.Quoted 72 lines
--->8-fat_test1.sh---
#!/bin/bash
makesandbox_defconfig
make
dd if=/dev/zero of=/tmp/10M.img bs=1024 count=10k
mkfs.vfat -v/tmp/10M.img
cat >/tmp/cmds <<EOF
x
host bind 0/tmp/10M.img
fatls host 0
mw0x10000000x0a4342410x1000# "ABC\n"
mw0x11000000x0a4645440x8000# "DEF\n"
fatwrite host 00x1000000file0001.raw0x1000
fatwrite host 00x1000000file0002.raw0x1000
fatwrite host 00x1000000file0003.raw0x1000
fatwrite host 00x1000000file0004.raw0x1000
fatwrite host 00x1000000file0005.raw0x1000
fatrm host 0 file0002.raw
fatrm host 0 file0004.raw
fatls host 0
fatwrite host 00x1100000file0007.raw0x4000
fatwrite host 00x1100000file0007.raw0x4000
reset
EOF
./u-boot </tmp/cmds
#verify
rm -r/tmp/result/tmp/model
mkdir/tmp/result
mkdir/tmp/modelyes ABC | head -c 4096 >/tmp/model/file0001.raw yes ABC | head -c 4096 >/tmp/model/file0003.raw yes ABC | head -c 4096 >/tmp/model/file0005.raw yes DEF | head -c 16384 >/tmp/model/file0007.rawmcopy -n -i
/tmp/10M.img::file0001.raw/tmp/resultmcopy -n -i/tmp/10M.img::file0003.raw/tmp/resultmcopy -n -i/tmp/10M.img::file0005.raw/tmp/resultmcopy -n -i/tmp/10M.img::file0007.raw/tmp/resulthd/tmp/10M.img
if diff -urq/tmp/model/tmp/result
then
echo Test okay
else
echo Test fail
fi
--->8---Overwritting a discontiguous test file (file0007.raw) no longer causes corruption to file0003.raw, which's data lies between the chunks of the test file. The amount of data written to disk is still incorrect, what causes damage to the file (file0005.raw), which's data lies next to the test file. This will be fixed by the next patch.
Feel free to prepare a proper sandbox/py_test based tests based on the provided test scripts.
--- fs/fat/fat_write.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)fs/fat/fat_write.c +2
-0@@ -794,6 +794,8 @@ set_contents(fsdata *mydata, dir_entry *dentptr, loff_t pos, __u8 *buffer, newclust =get_fatent (mydata, endclust); +if (newclust != endclust +1 ) +break ;if (IS_LAST_CLUST (newclust, mydata->fatsize))break ;if (CHECK_CLUST (newclust, mydata->fatsize)) {fs/fat/fat_write.c set_contents(fsdata *mydata, dir_entry *dentptr, · line 831
Quoted 8 lines
fs/fat/fat_write.c +0
-2@@ -824,8 +826,6 @@ set_contents(fsdata *mydata, dir_entry *dentptr, loff_t pos, __u8 *buffer, if (filesize <= cur_pos) break; -/* CHECK: newclust = get_fatent(mydata, endclust); */ -if (IS_LAST_CLUST (newclust, mydata->fatsize))/* no more clusters */ break ;Best regards,
Quoted 8 lines
Lukasz Majewski
--
DENX Software Engineering GmbH, Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-59 Fax: (+49)-8142-66989-80 Email: lukma at denx.de
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <https://lists.denx.de/pipermail/u-boot/attachments/20191205/3761086d/attachment.sig>
PATCH v4 1/6 fat: write: fix broken write to fragmented files
Quoted 21 lines · Matthias Brugger
Quoted 18 lines · Lukasz Majewski
Hi Tom, Matthias,
Quoted 6 lines
The code for handing file overwrite incorrectly assumed that the file on disk is always contiguous. This resulted in corrupting disk structure every time when write to existing fragmented file happened. Fix this by adding proper check for cluster discontinuity and adjust chunk size on each partial write.
Quoted 6 lines
Signed-off-by Marek Szyprowski <[email protected]>---
This patch partially fixes the issue revealed by the following test script:
Tom could you pic this patch and the following one (2/6):
https://patchwork.ozlabs.org/patch/1203101/
Quoted 8 lines
Quoted 3 lines
to -master as a fix?
This seems like a _real_ fix for FAT.
Right, I think the first patches should go in for v2020.01. I can send them together with some fixes for RPi I'm working on.
Tom what do you think?
I'd _really_ like to see the test that's in the commit message turned in to a test we can run from CI as Travis will be doing those Soon(TM) and GitLab/Azure do. Then I'll pick them up (I'm testing some other fat fixes right now). Thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20191205/ca63e64d/attachment.sig>
PATCH v4 1/6 fat: write: fix broken write to fragmented files
In reply to Marek SzyprowskiQuoted 45 lines · Marek Szyprowski
The code for handing file overwrite incorrectly assumed that the file on disk is always contiguous. This resulted in corrupting disk structure every time when write to existing fragmented file happened. Fix this by adding proper check for cluster discontinuity and adjust chunk size on each partial write.
Signed-off-by Marek Szyprowski <[email protected]>This patch partially fixes the issue revealed by the following test script:
--->8-fat_test1.sh---
#!/bin/bash
makesandbox_defconfig
make
dd if=/dev/zero of=/tmp/10M.img bs=1024 count=10k
mkfs.vfat -v/tmp/10M.img
cat >/tmp/cmds <<EOF
x
host bind 0/tmp/10M.img
fatls host 0
mw0x10000000x0a4342410x1000# "ABC\n"
mw0x11000000x0a4645440x8000# "DEF\n"fatwrite host 0
0x1000000file0001.raw0x1000
fatwrite host 00x1000000file0002.raw0x1000
fatwrite host 00x1000000file0003.raw0x1000
fatwrite host 00x1000000file0004.raw0x1000
fatwrite host 00x1000000file0005.raw0x1000
fatrm host 0 file0002.raw
fatrm host 0 file0004.raw
fatls host 0
fatwrite host 00x1100000file0007.raw0x4000
fatwrite host 00x1100000file0007.raw0x4000
reset
EOF
./u-boot </tmp/cmds
#verify
rm -r/tmp/result/tmp/model
mkdir/tmp/result
mkdir/tmp/modelyes ABC | head -c 4096 >/tmp/model/file0001.raw yes ABC | head -c 4096 >/tmp/model/file0003.raw yes ABC | head -c 4096 >/tmp/model/file0005.raw yes DEF | head -c 16384 >/tmp/model/file0007.raw
Quoted 11 lines
mcopy -n -i
/tmp/10M.img::file0001.raw/tmp/resultmcopy -n -i/tmp/10M.img::file0003.raw/tmp/resultmcopy -n -i/tmp/10M.img::file0005.raw/tmp/resultmcopy -n -i/tmp/10M.img::file0007.raw/tmp/resulthd/tmp/10M.img
if diff -urq/tmp/model/tmp/result
then
echo Test okay
else
echo Test fail
fi
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200207/aceddb5d/attachment.sig>
PATCH v4 2/6 fat: write: adjust data written in each partial write
In reply to Marek SzyprowskiThe code for handing file overwrite incorrectly calculated the amount of data to write when writing to the last non-cluster aligned chunk. Fix this by ensuring that no more data than the 'filesize' is written to disk. While touching min()-based calculations, change it to type-safe min_t() function.
---
This patch finally fixes the issue revealed by the test script from the previous patch. The correctness of the change has been also verified by the following additional test scripts:
--->8-fat_test2.sh---
#!/bin/bash
make sandbox_defconfig
make
dd if=/dev/zero of=/tmp/10M.img bs=1024 count=10k mkfs.vfat -v /tmp/10M.img
cat >/tmp/cmds <<EOF
x
host bind 0 /tmp/10M.img
fatls host 0
mw 0x1000000 0x0a434241 0x1000 # "ABC\n"
mw 0x1100000 0x0a464544 0x8000 # "DEF\n" fatwrite host 0 0x1000000 file0001.raw 0x1000 fatwrite host 0 0x1000000 file0002.raw 0x1000 fatwrite host 0 0x1000000 file0003.raw 0x1000 fatwrite host 0 0x1000000 file0004.raw 0x1000 fatwrite host 0 0x1000000 file0005.raw 0x1000 fatrm host 0 file0002.raw
fatrm host 0 file0004.raw
fatls host 0
fatwrite host 0 0x1100000 file0007.raw 0x2000 fatwrite host 0 0x1100000 file0007.raw 0x1f00 reset
EOF
./u-boot </tmp/cmds
#verify
rm -r /tmp/result /tmp/model
mkdir /tmp/result
mkdir /tmp/model
yes ABC | head -c 4096 >/tmp/model/file0001.raw yes ABC | head -c 4096 >/tmp/model/file0003.raw yes ABC | head -c 4096 >/tmp/model/file0005.raw yes DEF | head -c 7936 >/tmp/model/file0007.raw
mcopy -n -i /tmp/10M.img ::file0001.raw /tmp/result mcopy -n -i /tmp/10M.img ::file0003.raw /tmp/result mcopy -n -i /tmp/10M.img ::file0005.raw /tmp/result mcopy -n -i /tmp/10M.img ::file0007.raw /tmp/result hd /tmp/10M.img
if diff -urq /tmp/model /tmp/result
then
echo Test okay
else
echo Test fail
fi
--->8-fat_test3.sh---
#!/bin/bash
make sandbox_defconfig
make
dd if=/dev/zero of=/tmp/10M.img bs=1024 count=10k mkfs.vfat -v /tmp/10M.img
cat >/tmp/cmds <<EOF
x
host bind 0 /tmp/10M.img
fatls host 0
mw 0x1000000 0x0a434241 0x1000 # "ABC\n"
mw 0x1100000 0x0a464544 0x8000 # "DEF\n" fatwrite host 0 0x1000000 file0001.raw 0x1000 fatwrite host 0 0x1000000 file0002.raw 0x1000 fatwrite host 0 0x1000000 file0003.raw 0x1000 fatwrite host 0 0x1000000 file0004.raw 0x1000 fatwrite host 0 0x1000000 file0005.raw 0x1000 fatrm host 0 file0002.raw
fatrm host 0 file0004.raw
fatls host 0
fatwrite host 0 0x1100000 file0007.raw 0x2000 fatwrite host 0 0x1100000 file0007.raw 0x2100 reset
EOF
./u-boot </tmp/cmds
#verify
rm -r /tmp/result /tmp/model
mkdir /tmp/result
mkdir /tmp/model
yes ABC | head -c 4096 >/tmp/model/file0001.raw yes ABC | head -c 4096 >/tmp/model/file0003.raw yes ABC | head -c 4096 >/tmp/model/file0005.raw yes DEF | head -c 8448 >/tmp/model/file0007.raw
mcopy -n -i /tmp/10M.img ::file0001.raw /tmp/result mcopy -n -i /tmp/10M.img ::file0003.raw /tmp/result mcopy -n -i /tmp/10M.img ::file0005.raw /tmp/result mcopy -n -i /tmp/10M.img ::file0007.raw /tmp/result hd /tmp/10M.img
if diff -urq /tmp/model /tmp/result
then
echo Test okay
else
echo Test fail
fi
--->8-fat_test4.sh---
#!/bin/bash
make sandbox_defconfig
make
dd if=/dev/zero of=/tmp/10M.img bs=1024 count=10k mkfs.vfat -v /tmp/10M.img
cat >/tmp/cmds <<EOF
x
host bind 0 /tmp/10M.img
fatls host 0
mw 0x1000000 0x0a434241 0x1000 # "ABC\n"
mw 0x1100000 0x0a464544 0x8000 # "DEF\n"
mw 0x1200000 0x0a494847 0x8000 # "GHI\n" fatwrite host 0 0x1000000 file0001.raw 0x1000 fatwrite host 0 0x1000000 file0002.raw 0x1000 fatwrite host 0 0x1000000 file0003.raw 0x1000 fatwrite host 0 0x1000000 file0004.raw 0x1000 fatwrite host 0 0x1000000 file0005.raw 0x1000 fatrm host 0 file0002.raw
fatrm host 0 file0004.raw
fatls host 0
fatwrite host 0 0x1100000 file0007.raw 0x900 fatwrite host 0 0x1200000 file0007.raw 0x900 0x900 fatwrite host 0 0x1100000 file0007.raw 0x900 0x1200 fatwrite host 0 0x1200000 file0007.raw 0x900 0x1b00 reset
EOF
./u-boot </tmp/cmds
#verify
rm -r /tmp/result /tmp/model
mkdir /tmp/result
mkdir /tmp/model
yes ABC | head -c 4096 >/tmp/model/file0001.raw yes ABC | head -c 4096 >/tmp/model/file0003.raw yes ABC | head -c 4096 >/tmp/model/file0005.raw yes DEF | head -c 2304 >/tmp/model/file0007.raw yes GHI | head -c 2304 >>/tmp/model/file0007.raw yes DEF | head -c 2304 >>/tmp/model/file0007.raw yes GHI | head -c 2304 >>/tmp/model/file0007.raw
mcopy -n -i /tmp/10M.img ::file0001.raw /tmp/result mcopy -n -i /tmp/10M.img ::file0003.raw /tmp/result mcopy -n -i /tmp/10M.img ::file0005.raw /tmp/result mcopy -n -i /tmp/10M.img ::file0007.raw /tmp/result
hd /tmp/10M.img
if diff -urq /tmp/model /tmp/result
then
echo Test okay
else
echo Test fail
fi
--->8---
Feel free to prepare a proper sandbox/py_test based tests based on the provided test scripts.
--- fs/fat/fat_write.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
fs/fat/fat_write.c +3 -1
@@ -813,7 +813,9 @@ set_contents(fsdata *mydata, dir_entry *dentptr, loff_t pos, __u8 *buffer, offset =0 ; else offset = pos - cur_pos; - wsize =min (cur_pos + actsize, filesize) - pos; + wsize = min_t(unsigned long long, actsize, filesize - cur_pos); + wsize -= offset; +if (get_set_cluster (mydata, curclust, offset, buffer, wsize, &actsize)) {printf ("Error get-and-setting cluster\n" );
Signature · 1 line
2.17.1
PATCH v4 2/6 fat: write: adjust data written in each partial write
Quoted 6 lines · Marek Szyprowski
The code for handing file overwrite incorrectly calculated the amount of data to write when writing to the last non-cluster aligned chunk. Fix this by ensuring that no more data than the 'filesize' is written to disk. While touching min()-based calculations, change it to type-safe
min_t()function.
Quoted 6 lines
Signed-off-by Marek Szyprowski <[email protected]>This patch finally fixes the issue revealed by the test script from the previous patch. The correctness of the change has been also verified by the following additional test scripts:
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200207/b188b2cd/attachment.sig>
PATCH v4 3/6 dfu: mmc: rearrange the code
In reply to Marek SzyprowskiRename functions for bufferred file io operations to make them easier to understand. Also add missing file offset argument to them (currently unused). All this is a preparation to remove predefined file size limit (CONFIG_SYS_DFU_MAX_FILE_SIZE) for DFU read/write operations.
--- drivers/dfu/dfu_mmc.c | 61 ++++++++++++++++++++++++------------------- 1 file changed, 34 insertions(+), 27 deletions(-)
drivers/dfu/dfu_mmc.c +34 -27
@@ -91,22 +91,8 @@ static int mmc_block_op(enum dfu_op op, struct dfu_entity *dfu,return 0 ; } -static int mmc_file_buffer (struct dfu_entity *dfu,void *buf,long *len) -{ -if (dfu_file_buf_len + *len >CONFIG_SYS_DFU_MAX_FILE_SIZE ) { - dfu_file_buf_len =0 ; -return -EINVAL ; - } - -/* Add to the current buffer. */ -memcpy (dfu_file_buf + dfu_file_buf_len, buf, *len); - dfu_file_buf_len += *len; - -return 0 ; -} -static int mmc_file_op (enum dfu_op op,struct dfu_entity *dfu, -void *buf,u64 *len) +u64 offset,void *buf,u64 *len) {char dev_part_str[8 ];int ret;
@@ -137,7 +123,7 @@ static int mmc_file_op(enum dfu_op op, struct dfu_entity *dfu,switch (op) {case DFU_OP_READ : - ret =fs_read (dfu->name, (size_t )buf,0 ,0 , &size); + ret =fs_read (dfu->name, (size_t )buf, offset,0 , &size);if (ret) {puts ("dfu: fs_read error!\n" );return ret;
@@ -145,7 +131,7 @@ static int mmc_file_op(enum dfu_op op, struct dfu_entity *dfu, *len = size;break ;case DFU_OP_WRITE : - ret =fs_write (dfu->name, (size_t )buf,0 , *len, &size); + ret =fs_write (dfu->name, (size_t )buf, offset, *len, &size);if (ret) {puts ("dfu: fs_write error!\n" );return ret;
@@ -166,6 +152,30 @@ static int mmc_file_op(enum dfu_op op, struct dfu_entity *dfu,return ret; } +static int mmc_file_buf_write (struct dfu_entity *dfu,u64 offset,void *buf,long *len) +{ +if (dfu_file_buf_len + *len >CONFIG_SYS_DFU_MAX_FILE_SIZE ) { + dfu_file_buf_len =0 ; +return -EINVAL ; + } + +/* Add to the current buffer. */ +memcpy (dfu_file_buf + dfu_file_buf_len, buf, *len); + dfu_file_buf_len += *len; + +return 0 ; +} + +static int mmc_file_buf_write_finish (struct dfu_entity *dfu) +{ +int ret =mmc_file_op (DFU_OP_WRITE , dfu,0 , dfu_file_buf, + &dfu_file_buf_len); + +/* Now that we're done */ + dfu_file_buf_len =0 ; +return ret; +} +int dfu_write_medium_mmc (struct dfu_entity *dfu,u64 offset,void *buf,long *len) {
@@ -177,7 +187,7 @@ int dfu_write_medium_mmc(struct dfu_entity *dfu,break ;case DFU_FS_FAT :case DFU_FS_EXT4 : - ret =mmc_file_buffer (dfu, buf, len); + ret =mmc_file_buf_write (dfu, offset, buf, len);break ;default : printf("%s: Layout (%s) not (yet) supported!\n", __func__,
@@ -193,11 +203,7 @@ int dfu_flush_medium_mmc(struct dfu_entity *dfu)if (dfu->layout !=DFU_RAW_ADDR ) {/* Do stuff here. */ - ret =mmc_file_op (DFU_OP_WRITE , dfu, dfu_file_buf, - &dfu_file_buf_len); - -/* Now that we're done */ - dfu_file_buf_len =0 ; + ret =mmc_file_buf_write_finish (dfu); }return ret;
@@ -214,7 +220,7 @@ int dfu_get_medium_size_mmc(struct dfu_entity *dfu, u64 *size)case DFU_FS_FAT :case DFU_FS_EXT4 : dfu_file_buf_filled =-1 ; - ret =mmc_file_op (DFU_OP_SIZE , dfu,NULL , size); + ret =mmc_file_op (DFU_OP_SIZE , dfu,0 ,NULL , size);if (ret <0 )return ret;if (*size >CONFIG_SYS_DFU_MAX_FILE_SIZE )
@@ -227,14 +233,15 @@ int dfu_get_medium_size_mmc(struct dfu_entity *dfu, u64 *size) } } -static int mmc_file_unbuffer (struct dfu_entity *dfu,u64 offset,void *buf, + +static int mmc_file_buf_read (struct dfu_entity *dfu,u64 offset,void *buf,long *len) {int ret;u64 file_len;if (dfu_file_buf_filled ==-1 ) { - ret =mmc_file_op (DFU_OP_READ , dfu, dfu_file_buf, &file_len); + ret =mmc_file_op (DFU_OP_READ , dfu,0 , dfu_file_buf, &file_len);if (ret <0 )return ret; dfu_file_buf_filled = file_len;
@@ -259,7 +266,7 @@ int dfu_read_medium_mmc(struct dfu_entity *dfu, u64 offset, void *buf,break ;case DFU_FS_FAT :case DFU_FS_EXT4 : - ret =mmc_file_unbuffer (dfu, offset, buf, len); + ret =mmc_file_buf_read (dfu, offset, buf, len);break ;default : printf("%s: Layout (%s) not (yet) supported!\n", __func__,
Signature · 1 line
2.17.1
PATCH v4 4/6 dfu: mmc: remove file size limit for io operations
In reply to Marek SzyprowskiAdd support for operations on files larger than CONFIG_SYS_DFU_MAX_FILE_SIZE. The buffered io mechanism is still used for aggregating io requests, so for files up to CONFIG_SYS_DFU_MAX_FILE_SIZE nothing is changed and they will be handled in a single filesystem call.
--- drivers/dfu/dfu_mmc.c | 46 ++++++++++++++++++++++++++++--------------- 1 file changed, 30 insertions(+), 16 deletions(-)
drivers/dfu/dfu_mmc.c +30 -16
@@ -17,7 +17,7 @@static unsigned char *dfu_file_buf;static u64 dfu_file_buf_len; -static long dfu_file_buf_filled; +static u64 dfu_file_buf_offset;static int mmc_block_op (enum dfu_op op,struct dfu_entity *dfu,u64 offset,void *buf,long *len)
@@ -123,7 +123,7 @@ static int mmc_file_op(enum dfu_op op, struct dfu_entity *dfu,switch (op) {case DFU_OP_READ : - ret =fs_read (dfu->name, (size_t )buf, offset,0 , &size); + ret =fs_read (dfu->name, (size_t )buf, offset, *len, &size);if (ret) {puts ("dfu: fs_read error!\n" );return ret;
@@ -154,25 +154,38 @@ static int mmc_file_op(enum dfu_op op, struct dfu_entity *dfu,static int mmc_file_buf_write (struct dfu_entity *dfu,u64 offset,void *buf,long *len) { -if (dfu_file_buf_len + *len >CONFIG_SYS_DFU_MAX_FILE_SIZE ) { +int ret =0 ; + +if (offset ==0 ) { dfu_file_buf_len =0 ; -return -EINVAL ; + dfu_file_buf_offset =0 ; }/* Add to the current buffer. */ +if (dfu_file_buf_len + *len >CONFIG_SYS_DFU_MAX_FILE_SIZE ) + *len =CONFIG_SYS_DFU_MAX_FILE_SIZE - dfu_file_buf_len;memcpy (dfu_file_buf + dfu_file_buf_len, buf, *len); dfu_file_buf_len += *len; -return 0 ; +if (dfu_file_buf_len ==CONFIG_SYS_DFU_MAX_FILE_SIZE ) { + ret =mmc_file_op (DFU_OP_WRITE , dfu, dfu_file_buf_offset, + dfu_file_buf, &dfu_file_buf_len); + dfu_file_buf_offset += dfu_file_buf_len; + dfu_file_buf_len =0 ; + } + +return ret; }static int mmc_file_buf_write_finish (struct dfu_entity *dfu) { -int ret =mmc_file_op (DFU_OP_WRITE , dfu,0 , dfu_file_buf, - &dfu_file_buf_len); +int ret =mmc_file_op (DFU_OP_WRITE , dfu, dfu_file_buf_offset, + dfu_file_buf, &dfu_file_buf_len);/* Now that we're done */ dfu_file_buf_len =0 ; + dfu_file_buf_offset =0 ; +return ret; }
@@ -219,12 +232,9 @@ int dfu_get_medium_size_mmc(struct dfu_entity *dfu, u64 *size)return 0 ;case DFU_FS_FAT :case DFU_FS_EXT4 : - dfu_file_buf_filled =-1 ; ret =mmc_file_op (DFU_OP_SIZE , dfu,0 ,NULL , size);if (ret <0 )return ret; -if (*size >CONFIG_SYS_DFU_MAX_FILE_SIZE ) -return -1 ;return 0 ;default : printf("%s: Layout (%s) not (yet) supported!\n", __func__,
@@ -238,19 +248,23 @@ static int mmc_file_buf_read(struct dfu_entity *dfu, u64 offset, void *buf, long *len) {int ret; -u64 file_len; -if (dfu_file_buf_filled ==-1 ) { - ret =mmc_file_op (DFU_OP_READ , dfu,0 , dfu_file_buf, &file_len); +if (offset ==0 || offset >= dfu_file_buf_offset + dfu_file_buf_len || + offset + *len < dfu_file_buf_offset) { +u64 file_len =CONFIG_SYS_DFU_MAX_FILE_SIZE ; + + ret =mmc_file_op (DFU_OP_READ , dfu, offset, dfu_file_buf, + &file_len);if (ret <0 )return ret; - dfu_file_buf_filled = file_len; + dfu_file_buf_len = file_len; + dfu_file_buf_offset = offset; } -if (offset + *len > dfu_file_buf_filled) +if (offset + *len > dfu_file_buf_offset + dfu_file_buf_len)return -EINVAL ;/* Add to the current buffer. */ -memcpy (buf, dfu_file_buf + offset, *len); +memcpy (buf, dfu_file_buf + offset - dfu_file_buf_offset, *len);return 0 ; }
Signature · 1 line
2.17.1
PATCH v4 5/6 usb: dwc2_udc_otg: add bcm2835 SoC (Raspberry Pi4) support
In reply to Marek SzyprowskiBroadcom 2835 SoC requires special conversion of physical memory addresses for DMA purpose, so add needed wrappers to dwc2_udc_otg driver. Also extend the list of compatible devices with 'brcm,bcm2835-usb' entry. This allows to use USB gadget drivers (i.e. DFU) on Raspberry Pi4 boards.
--- drivers/usb/gadget/dwc2_udc_otg.c | 2 ++ drivers/usb/gadget/dwc2_udc_otg_xfer_dma.c | 12 ++++++------ 2 files changed, 8 insertions(+), 6 deletions(-)
drivers/usb/gadget/dwc2_udc_otg.c +2 -0
@@ -31,6 +31,7 @@#include <linux/usb/otg.h> #include <linux/usb/gadget.h> +#include <phys2bus.h> #include <asm/byteorder.h> #include <asm/unaligned.h> #include <asm/io.h>
@@ -1213,6 +1214,7 @@ static int dwc2_udc_otg_remove(struct udevice *dev)static const struct udevice_id dwc2_udc_otg_ids[] = { { .compatible ="snps,dwc2" }, + { .compatible ="brcm,bcm2835-usb" }, { .compatible ="st,stm32mp1-hsotg" , .data = (ulong )dwc2_set_stm32mp1_hsotg_params }, {},
drivers/usb/gadget/dwc2_udc_otg_xfer_dma.c +6 -6
@@ -28,7 +28,7 @@ static inline void dwc2_udc_ep0_zlp(struct dwc2_udc *dev) {u32 ep_ctrl; -writel (usb_ctrl_dma_addr, ®->in_endp[EP0_CON ].diepdma); +writel (phys_to_bus ((unsigned long )usb_ctrl_dma_addr), ®->in_endp[EP0_CON ].diepdma);writel (DIEPT_SIZ_PKT_CNT (1 ), ®->in_endp[EP0_CON ].dieptsiz); ep_ctrl =readl (®->in_endp[EP0_CON ].diepctl);
@@ -49,7 +49,7 @@ static void dwc2_udc_pre_setup(void)writel (DOEPT_SIZ_PKT_CNT (1 ) |sizeof (struct usb_ctrlrequest ), ®->out_endp[EP0_CON ].doeptsiz); -writel (usb_ctrl_dma_addr, ®->out_endp[EP0_CON ].doepdma); +writel (phys_to_bus ((unsigned long )usb_ctrl_dma_addr), ®->out_endp[EP0_CON ].doepdma); ep_ctrl =readl (®->out_endp[EP0_CON ].doepctl);writel (ep_ctrl|DEPCTL_EPENA , ®->out_endp[EP0_CON ].doepctl);
@@ -75,7 +75,7 @@ static inline void dwc2_ep0_complete_out(void)writel (DOEPT_SIZ_PKT_CNT (1 ) |sizeof (struct usb_ctrlrequest ), ®->out_endp[EP0_CON ].doeptsiz); -writel (usb_ctrl_dma_addr, ®->out_endp[EP0_CON ].doepdma); +writel (phys_to_bus ((unsigned long )usb_ctrl_dma_addr), ®->out_endp[EP0_CON ].doepdma); ep_ctrl =readl (®->out_endp[EP0_CON ].doepctl); writel(ep_ctrl|DEPCTL_EPENA|DEPCTL_CNAK,
@@ -113,7 +113,7 @@ static int setdma_rx(struct dwc2_ep *ep, struct dwc2_request *req) (unsigned long ) ep->dma_buf + ROUND(ep->len, CONFIG_SYS_CACHELINE_SIZE)); -writel ((unsigned long ) ep->dma_buf, ®->out_endp[ep_num].doepdma); +writel (phys_to_bus ((unsigned long )ep->dma_buf), ®->out_endp[ep_num].doepdma);writel (DOEPT_SIZ_PKT_CNT (pktcnt) |DOEPT_SIZ_XFER_SIZE (length), ®->out_endp[ep_num].doeptsiz);writel (DEPCTL_EPENA |DEPCTL_CNAK |ctrl, ®->out_endp[ep_num].doepctl);
@@ -161,7 +161,7 @@ static int setdma_tx(struct dwc2_ep *ep, struct dwc2_request *req)while (readl (®->grstctl) &TX_FIFO_FLUSH ) ; -writel ((unsigned long ) ep->dma_buf, ®->in_endp[ep_num].diepdma); +writel (phys_to_bus ((unsigned long )ep->dma_buf), ®->in_endp[ep_num].diepdma);writel (DIEPT_SIZ_PKT_CNT (pktcnt) |DIEPT_SIZ_XFER_SIZE (length), ®->in_endp[ep_num].dieptsiz);
@@ -921,7 +921,7 @@ static int dwc2_udc_get_status(struct dwc2_udc *dev, (unsigned long ) usb_ctrl + ROUND(sizeof(g_status), CONFIG_SYS_CACHELINE_SIZE)); -writel (usb_ctrl_dma_addr, ®->in_endp[EP0_CON ].diepdma); +writel (phys_to_bus (usb_ctrl_dma_addr), ®->in_endp[EP0_CON ].diepdma);writel (DIEPT_SIZ_PKT_CNT (1 ) |DIEPT_SIZ_XFER_SIZE (2 ), ®->in_endp[EP0_CON ].dieptsiz);
Signature · 1 line
2.17.1
PATCH v4 6/6 config: enable DFU over USB on Raspberry Pi4 boards
In reply to Marek SzyprowskiEnable support for DFU over USB. This requires to enable USB gadget, DWC2 UDC OTG driver and DFU command. DFU entities are defined for the following firmware objects: u-boot.bin, uboot.env, config.txt, and zImage/Image.
--- configs/rpi_4_32b_defconfig | 11 +++++++++++ configs/rpi_4_defconfig | 11 +++++++++++ include/configs/rpi.h | 20 ++++++++++++++++++++ 3 files changed, 42 insertions(+)
configs/rpi_4_32b_defconfig +11 -0
@@ -12,6 +12,7 @@ CONFIG_MISC_INIT_R=y # CONFIG_DISPLAY_CPUINFO is not set # CONFIG_DISPLAY_BOARDINFO is not set CONFIG_SYS_PROMPT="U-Boot> " +CONFIG_CMD_DFU=y # CONFIG_CMD_FLASH is not set CONFIG_CMD_GPIO=y CONFIG_CMD_MMC=y
@@ -21,6 +22,7 @@ CONFIG_ENV_FAT_INTERFACE="mmc" CONFIG_ENV_FAT_DEVICE_AND_PART="0:1" CONFIG_SYS_RELOC_GD_ENV_ADDR=y CONFIG_ENV_VARS_UBOOT_RUNTIME_CONFIG=y +CONFIG_DFU_MMC=y CONFIG_DM_KEYBOARD=y CONFIG_DM_MMC=y CONFIG_MMC_SDHCI=y
@@ -28,6 +30,15 @@ CONFIG_MMC_SDHCI_BCM2835=y CONFIG_PINCTRL=y # CONFIG_PINCTRL_GENERIC is not set # CONFIG_REQUIRE_SERIAL_CONSOLE is not set +CONFIG_USB=y +CONFIG_DM_USB=y +CONFIG_DM_USB_GADGET=y +CONFIG_USB_GADGET=y +CONFIG_USB_GADGET_MANUFACTURER="FSL" +CONFIG_USB_GADGET_VENDOR_NUM=0x0525 +CONFIG_USB_GADGET_PRODUCT_NUM=0xa4a5 +CONFIG_USB_GADGET_DWC2_OTG=y +CONFIG_USB_GADGET_DOWNLOAD=y CONFIG_DM_VIDEO=y CONFIG_SYS_WHITE_ON_BLACK=y CONFIG_CONSOLE_SCROLL_LINES=10
configs/rpi_4_defconfig +11 -0
@@ -12,6 +12,7 @@ CONFIG_MISC_INIT_R=y # CONFIG_DISPLAY_CPUINFO is not set # CONFIG_DISPLAY_BOARDINFO is not set CONFIG_SYS_PROMPT="U-Boot> " +CONFIG_CMD_DFU=y # CONFIG_CMD_FLASH is not set CONFIG_CMD_GPIO=y CONFIG_CMD_MMC=y
@@ -21,6 +22,7 @@ CONFIG_ENV_FAT_INTERFACE="mmc" CONFIG_ENV_FAT_DEVICE_AND_PART="0:1" CONFIG_SYS_RELOC_GD_ENV_ADDR=y CONFIG_ENV_VARS_UBOOT_RUNTIME_CONFIG=y +CONFIG_DFU_MMC=y CONFIG_DM_KEYBOARD=y CONFIG_DM_MMC=y CONFIG_MMC_SDHCI=y
@@ -28,6 +30,15 @@ CONFIG_MMC_SDHCI_BCM2835=y CONFIG_PINCTRL=y # CONFIG_PINCTRL_GENERIC is not set # CONFIG_REQUIRE_SERIAL_CONSOLE is not set +CONFIG_USB=y +CONFIG_DM_USB=y +CONFIG_DM_USB_GADGET=y +CONFIG_USB_GADGET=y +CONFIG_USB_GADGET_MANUFACTURER="FSL" +CONFIG_USB_GADGET_VENDOR_NUM=0x0525 +CONFIG_USB_GADGET_PRODUCT_NUM=0xa4a5 +CONFIG_USB_GADGET_DWC2_OTG=y +CONFIG_USB_GADGET_DOWNLOAD=y CONFIG_DM_VIDEO=y CONFIG_SYS_WHITE_ON_BLACK=y CONFIG_CONSOLE_SCROLL_LINES=10
include/configs/rpi.h +20 -0
@@ -74,6 +74,25 @@#define CONFIG_TFTP_TSIZE #endif +/* DFU over USB/UDC */ +#ifdef CONFIG_CMD_DFU +#define CONFIG_SYS_DFU_DATA_BUF_SIZE SZ_1M +#define CONFIG_SYS_DFU_MAX_FILE_SIZE SZ_2M + +#ifdef CONFIG_ARM64 +#define KERNEL_FILENAME "Image" +#else +#define KERNEL_FILENAME "zImage" +#endif + +#define ENV_DFU_SETTINGS \ + "dfu_alt_info=u-boot.bin fat 0 1;uboot.env fat 0 1;" \ + "config.txt fat 0 1;" \ + KERNEL_FILENAME " fat 0 1\0" +#else +#define ENV_DFU_SETTINGS "" +#endif +/* Console configuration */ #define CONFIG_SYS_CBSIZE 1024
@@ -188,6 +207,7 @@#define CONFIG_EXTRA_ENV_SETTINGS \ "dhcpuboot=usb start; dhcp u-boot.uimg; bootm\0" \ ENV_DEVICE_SETTINGS \ + ENV_DFU_SETTINGS \ ENV_MEM_LAYOUT_SETTINGS \ BOOTENV
Signature · 1 line
2.17.1
PATCH v4 6/6 config: enable DFU over USB on Raspberry Pi4 boards
Quoted 10 lines · Marek Szyprowski
Enable support for DFU over USB. This requires to enable USB gadget, DWC2 UDC OTG driver and DFU command. DFU entities are defined for the following firmware objects: u-boot.bin, uboot.env, config.txt, and zImage/Image.
Signed-off-by Marek Szyprowski <[email protected]>Reviewed-by Lukasz Majewski <[email protected]>--- configs/rpi_4_32b_defconfig | 11 +++++++++++ configs/rpi_4_defconfig | 11 +++++++++++
In the meanwhile we have a third config rpi_arm64_defconfig
does it make sense to add DFU support there too?
I suppose it works on RPi3 as well. If so, would you mind to send a follow-up patch?
Regards,
Matthias
Quoted 114 lines
include/configs/rpi.h | 20 ++++++++++++++++++++ 3 files changed, 42 insertions(+)configs/rpi_4_32b_defconfig +11
-0@@ -12,6 +12,7 @@ CONFIG_MISC_INIT_R=y # CONFIG_DISPLAY_CPUINFO is not set # CONFIG_DISPLAY_BOARDINFO is not set CONFIG_SYS_PROMPT="U-Boot> " +CONFIG_CMD_DFU=y # CONFIG_CMD_FLASH is not set CONFIG_CMD_GPIO=y CONFIG_CMD_MMC=y@@ -21,6 +22,7 @@ CONFIG_ENV_FAT_INTERFACE="mmc" CONFIG_ENV_FAT_DEVICE_AND_PART="0:1" CONFIG_SYS_RELOC_GD_ENV_ADDR=y CONFIG_ENV_VARS_UBOOT_RUNTIME_CONFIG=y +CONFIG_DFU_MMC=y CONFIG_DM_KEYBOARD=y CONFIG_DM_MMC=y CONFIG_MMC_SDHCI=y@@ -28,6 +30,15 @@ CONFIG_MMC_SDHCI_BCM2835=y CONFIG_PINCTRL=y # CONFIG_PINCTRL_GENERIC is not set # CONFIG_REQUIRE_SERIAL_CONSOLE is not set +CONFIG_USB=y +CONFIG_DM_USB=y +CONFIG_DM_USB_GADGET=y +CONFIG_USB_GADGET=y +CONFIG_USB_GADGET_MANUFACTURER="FSL" +CONFIG_USB_GADGET_VENDOR_NUM=0x0525 +CONFIG_USB_GADGET_PRODUCT_NUM=0xa4a5 +CONFIG_USB_GADGET_DWC2_OTG=y +CONFIG_USB_GADGET_DOWNLOAD=y CONFIG_DM_VIDEO=y CONFIG_SYS_WHITE_ON_BLACK=y CONFIG_CONSOLE_SCROLL_LINES=10configs/rpi_4_defconfig +11
-0@@ -12,6 +12,7 @@ CONFIG_MISC_INIT_R=y # CONFIG_DISPLAY_CPUINFO is not set # CONFIG_DISPLAY_BOARDINFO is not set CONFIG_SYS_PROMPT="U-Boot> " +CONFIG_CMD_DFU=y # CONFIG_CMD_FLASH is not set CONFIG_CMD_GPIO=y CONFIG_CMD_MMC=y@@ -21,6 +22,7 @@ CONFIG_ENV_FAT_INTERFACE="mmc" CONFIG_ENV_FAT_DEVICE_AND_PART="0:1" CONFIG_SYS_RELOC_GD_ENV_ADDR=y CONFIG_ENV_VARS_UBOOT_RUNTIME_CONFIG=y +CONFIG_DFU_MMC=y CONFIG_DM_KEYBOARD=y CONFIG_DM_MMC=y CONFIG_MMC_SDHCI=y@@ -28,6 +30,15 @@ CONFIG_MMC_SDHCI_BCM2835=y CONFIG_PINCTRL=y # CONFIG_PINCTRL_GENERIC is not set # CONFIG_REQUIRE_SERIAL_CONSOLE is not set +CONFIG_USB=y +CONFIG_DM_USB=y +CONFIG_DM_USB_GADGET=y +CONFIG_USB_GADGET=y +CONFIG_USB_GADGET_MANUFACTURER="FSL" +CONFIG_USB_GADGET_VENDOR_NUM=0x0525 +CONFIG_USB_GADGET_PRODUCT_NUM=0xa4a5 +CONFIG_USB_GADGET_DWC2_OTG=y +CONFIG_USB_GADGET_DOWNLOAD=y CONFIG_DM_VIDEO=y CONFIG_SYS_WHITE_ON_BLACK=y CONFIG_CONSOLE_SCROLL_LINES=10include/configs/rpi.h +20
-0@@ -74,6 +74,25 @@#define CONFIG_TFTP_TSIZE #endif +/* DFU over USB/UDC */ +#ifdef CONFIG_CMD_DFU +#define CONFIG_SYS_DFU_DATA_BUF_SIZE SZ_1M +#define CONFIG_SYS_DFU_MAX_FILE_SIZE SZ_2M + +#ifdef CONFIG_ARM64 +#define KERNEL_FILENAME "Image" +#else +#define KERNEL_FILENAME "zImage" +#endif + +#define ENV_DFU_SETTINGS \ + "dfu_alt_info=u-boot.bin fat 0 1;uboot.env fat 0 1;" \ + "config.txt fat 0 1;" \ + KERNEL_FILENAME " fat 0 1\0" +#else +#define ENV_DFU_SETTINGS "" +#endif +/* Console configuration */ #define CONFIG_SYS_CBSIZE 1024@@ -188,6 +207,7 @@#define CONFIG_EXTRA_ENV_SETTINGS \ "dhcpuboot=usb start; dhcp u-boot.uimg; bootm\0" \ ENV_DEVICE_SETTINGS \ + ENV_DFU_SETTINGS \ ENV_MEM_LAYOUT_SETTINGS \ BOOTENV
PATCH v4 6/6 config: enable DFU over USB on Raspberry Pi4 boards
Quoted 14 lines · Matthias Brugger
Quoted 10 lines · Marek Szyprowski
Enable support for DFU over USB. This requires to enable USB gadget, DWC2 UDC OTG driver and DFU command. DFU entities are defined for the following firmware objects: u-boot.bin, uboot.env, config.txt, and zImage/Image.
Signed-off-by Marek Szyprowski <[email protected]>Reviewed-by Lukasz Majewski <[email protected]>--- configs/rpi_4_32b_defconfig | 11 +++++++++++ configs/rpi_4_defconfig | 11 +++++++++++
Quoted 4 lines
In the meanwhile we have a third config
rpi_arm64_defconfigdoes it make sense to add DFU support there too?
I suppose it works on RPi3 as well. If so, would you mind to send a follow-up patch?
As i know, RPI3's HW doesn't support. SO it doesn't need to apply this patch on RPI3.
Best Regards,
Jaehoon Chung
Quoted 120 lines
Regards,
MatthiasQuoted 114 lines
include/configs/rpi.h | 20 ++++++++++++++++++++ 3 files changed, 42 insertions(+)configs/rpi_4_32b_defconfig +11
-0@@ -12,6 +12,7 @@ CONFIG_MISC_INIT_R=y # CONFIG_DISPLAY_CPUINFO is not set # CONFIG_DISPLAY_BOARDINFO is not set CONFIG_SYS_PROMPT="U-Boot> " +CONFIG_CMD_DFU=y # CONFIG_CMD_FLASH is not set CONFIG_CMD_GPIO=y CONFIG_CMD_MMC=y@@ -21,6 +22,7 @@ CONFIG_ENV_FAT_INTERFACE="mmc" CONFIG_ENV_FAT_DEVICE_AND_PART="0:1" CONFIG_SYS_RELOC_GD_ENV_ADDR=y CONFIG_ENV_VARS_UBOOT_RUNTIME_CONFIG=y +CONFIG_DFU_MMC=y CONFIG_DM_KEYBOARD=y CONFIG_DM_MMC=y CONFIG_MMC_SDHCI=y@@ -28,6 +30,15 @@ CONFIG_MMC_SDHCI_BCM2835=y CONFIG_PINCTRL=y # CONFIG_PINCTRL_GENERIC is not set # CONFIG_REQUIRE_SERIAL_CONSOLE is not set +CONFIG_USB=y +CONFIG_DM_USB=y +CONFIG_DM_USB_GADGET=y +CONFIG_USB_GADGET=y +CONFIG_USB_GADGET_MANUFACTURER="FSL" +CONFIG_USB_GADGET_VENDOR_NUM=0x0525 +CONFIG_USB_GADGET_PRODUCT_NUM=0xa4a5 +CONFIG_USB_GADGET_DWC2_OTG=y +CONFIG_USB_GADGET_DOWNLOAD=y CONFIG_DM_VIDEO=y CONFIG_SYS_WHITE_ON_BLACK=y CONFIG_CONSOLE_SCROLL_LINES=10configs/rpi_4_defconfig +11
-0@@ -12,6 +12,7 @@ CONFIG_MISC_INIT_R=y # CONFIG_DISPLAY_CPUINFO is not set # CONFIG_DISPLAY_BOARDINFO is not set CONFIG_SYS_PROMPT="U-Boot> " +CONFIG_CMD_DFU=y # CONFIG_CMD_FLASH is not set CONFIG_CMD_GPIO=y CONFIG_CMD_MMC=y@@ -21,6 +22,7 @@ CONFIG_ENV_FAT_INTERFACE="mmc" CONFIG_ENV_FAT_DEVICE_AND_PART="0:1" CONFIG_SYS_RELOC_GD_ENV_ADDR=y CONFIG_ENV_VARS_UBOOT_RUNTIME_CONFIG=y +CONFIG_DFU_MMC=y CONFIG_DM_KEYBOARD=y CONFIG_DM_MMC=y CONFIG_MMC_SDHCI=y@@ -28,6 +30,15 @@ CONFIG_MMC_SDHCI_BCM2835=y CONFIG_PINCTRL=y # CONFIG_PINCTRL_GENERIC is not set # CONFIG_REQUIRE_SERIAL_CONSOLE is not set +CONFIG_USB=y +CONFIG_DM_USB=y +CONFIG_DM_USB_GADGET=y +CONFIG_USB_GADGET=y +CONFIG_USB_GADGET_MANUFACTURER="FSL" +CONFIG_USB_GADGET_VENDOR_NUM=0x0525 +CONFIG_USB_GADGET_PRODUCT_NUM=0xa4a5 +CONFIG_USB_GADGET_DWC2_OTG=y +CONFIG_USB_GADGET_DOWNLOAD=y CONFIG_DM_VIDEO=y CONFIG_SYS_WHITE_ON_BLACK=y CONFIG_CONSOLE_SCROLL_LINES=10include/configs/rpi.h +20
-0@@ -74,6 +74,25 @@#define CONFIG_TFTP_TSIZE #endif +/* DFU over USB/UDC */ +#ifdef CONFIG_CMD_DFU +#define CONFIG_SYS_DFU_DATA_BUF_SIZE SZ_1M +#define CONFIG_SYS_DFU_MAX_FILE_SIZE SZ_2M + +#ifdef CONFIG_ARM64 +#define KERNEL_FILENAME "Image" +#else +#define KERNEL_FILENAME "zImage" +#endif + +#define ENV_DFU_SETTINGS \ + "dfu_alt_info=u-boot.bin fat 0 1;uboot.env fat 0 1;" \ + "config.txt fat 0 1;" \ + KERNEL_FILENAME " fat 0 1\0" +#else +#define ENV_DFU_SETTINGS "" +#endif +/* Console configuration */ #define CONFIG_SYS_CBSIZE 1024@@ -188,6 +207,7 @@#define CONFIG_EXTRA_ENV_SETTINGS \ "dhcpuboot=usb start; dhcp u-boot.uimg; bootm\0" \ ENV_DEVICE_SETTINGS \ + ENV_DFU_SETTINGS \ ENV_MEM_LAYOUT_SETTINGS \ BOOTENV
v4 cover Raspberry Pi4: add support for DFU over USB
In reply to Marek SzyprowskiHi Matthias,
Do you plan to pull this patch series to RPI repository?
I'm asking as it contains some DFU related patches (Acked already by me), which I would like to see pulled in this merge window.
Thanks in advance for your help :-)
Quoted 80 lines
Hi All!
This patchset enables support for DFU over USB protocol on Raspberry Pi4 board. The board has DWC2 UDC controller connected to the USB-C power connector. Enabling DFU on it, make the u-boot development much more convenient, as one no longer needs to swap SD-card between RPi4 board and host machine to update the u-boot binary.
Patches are based on current 'master' u-boot branch. They were tested on the 2019-07-10-raspbian-buster-lite.img sd-card image with the following lines added to config.txt:
dtoverlay=dwc2,dr_mode=peripheral
dtparam=i2c_arm=on
dtparam=spi=on
enable_uart=1
uart_2ndstage=1
kernel=u-boot.binTo enable DFU, one has to enter follwing command:
# dfu 0 mmc 0During the development of this feature I've encountered a serious bugs in FAT write code. Over-writing discontiguous files always caused serious filesystem corruption. This was especially anoying, because the system environment is kept on FAT volume in uboot.env file, so 'saveenv' basically corrupted the boot partiting on the second call. Another bunch of the issues in the FAT write code has been revealed while removing predefined file size limit in DFU MMC code and then running sandbox tests.
I hope that my fixes for FAT code will be helpful for non-RPi users too.
Best regards
Marek Szyprowski
Samsung R&D Institute PolandChangelog:
v4:
- rechecked the FAT related fixes, it turned out that much simpler patch fixes both issues discovered while working on DFU support; added simple sandbox based tests reveleaing the issue and showing correctness of the fix
- rebased patches onto current u-boot's master branch:4b19b89ca4a8("Merge tag 'rpi-next-2020.01' of https://github.com/mbgg/u-boot")v3: https://patchwork.ozlabs.org/cover/1200793/
- fixed one more FAT issue revealed by sandbox testsv3: (patch 6/6 posted separately):
https://patchwork.ozlabs.org/patch/1195645/
- fixed non-RPi4 builds (missing #elseENV_DFU_SETTINGSdef)
- removed config.txt entity (not needed in uboot-based boot)
- switched arm64 kernel filename to 'Image'v2: https://patchwork.ozlabs.org/cover/1166589/
- added changes torpi_4_defconfigtoo (arm64 version)
- extended DFU entity list by confix.txt, cmdline.txt and Image.gz
- fixed missing '\0' at the end ofdfu_alt_infoenv
- added reviewed-by tags
- added patches for DFU MMC to remove file size limit
- added patch for fat write to fix issues on non-zero file offset (revealed by previous patch)v1: https://patchwork.ozlabs.org/cover/1162770/
- initial versionPatch summary:
Marek Szyprowski (6):
fat: write: fix broken write to fragmented files
fat: write: adjust data written in each partial write
dfu: mmc: rearrange the code
dfu: mmc: remove file size limit for io operations
usb:dwc2_udc_otg: add bcm2835 SoC (Raspberry Pi4) support
config: enable DFU over USB on Raspberry Pi4 boards
Quoted 9 lines
configs/rpi_4_32b_defconfig | 11 +++ configs/rpi_4_defconfig | 11 +++ drivers/dfu/dfu_mmc.c | 93 +++++++++++++--------- drivers/usb/gadget/dwc2_udc_otg.c | 2 + drivers/usb/gadget/dwc2_udc_otg_xfer_dma.c | 12 +-- fs/fat/fat_write.c | 8 +- include/configs/rpi.h | 20 +++++ 7 files changed, 112 insertions(+), 45 deletions(-)
Best regards,
Lukasz Majewski
--
DENX Software Engineering GmbH, Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-59 Fax: (+49)-8142-66989-80 Email: lukma at denx.de -------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 488 bytes
Desc: OpenPGP digital signature
URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200127/5d63a466/attachment.sig>
v4 cover Raspberry Pi4: add support for DFU over USB
Quoted 7 lines · Lukasz Majewski
Hi Matthias,
Do you plan to pull this patch series to RPI repository?
I'm asking as it contains some DFU related patches (Acked already by me), which I would like to see pulled in this merge window.
Yes I was waiting on the first to patches for FAT as Tom asked to add a test. So I'm reluctant to take these two. I can take the rest of the series though.
Regards,
Matthias
Quoted 105 lines
Thanks in advance for your help :-)
Quoted 80 lines
Hi All!
This patchset enables support for DFU over USB protocol on Raspberry Pi4 board. The board has DWC2 UDC controller connected to the USB-C power connector. Enabling DFU on it, make the u-boot development much more convenient, as one no longer needs to swap SD-card between RPi4 board and host machine to update the u-boot binary.
Patches are based on current 'master' u-boot branch. They were tested on the 2019-07-10-raspbian-buster-lite.img sd-card image with the following lines added to config.txt:
dtoverlay=dwc2,dr_mode=peripheral
dtparam=i2c_arm=on
dtparam=spi=on
enable_uart=1
uart_2ndstage=1
kernel=u-boot.binTo enable DFU, one has to enter follwing command:
# dfu 0 mmc 0During the development of this feature I've encountered a serious bugs in FAT write code. Over-writing discontiguous files always caused serious filesystem corruption. This was especially anoying, because the system environment is kept on FAT volume in uboot.env file, so 'saveenv' basically corrupted the boot partiting on the second call. Another bunch of the issues in the FAT write code has been revealed while removing predefined file size limit in DFU MMC code and then running sandbox tests.
I hope that my fixes for FAT code will be helpful for non-RPi users too.
Best regards
Marek Szyprowski
Samsung R&D Institute PolandChangelog:
v4:
- rechecked the FAT related fixes, it turned out that much simpler patch fixes both issues discovered while working on DFU support; added simple sandbox based tests reveleaing the issue and showing correctness of the fix
- rebased patches onto current u-boot's master branch:4b19b89ca4a8("Merge tag 'rpi-next-2020.01' of https://github.com/mbgg/u-boot")v3: https://patchwork.ozlabs.org/cover/1200793/
- fixed one more FAT issue revealed by sandbox testsv3: (patch 6/6 posted separately):
https://patchwork.ozlabs.org/patch/1195645/
- fixed non-RPi4 builds (missing #elseENV_DFU_SETTINGSdef)
- removed config.txt entity (not needed in uboot-based boot)
- switched arm64 kernel filename to 'Image'v2: https://patchwork.ozlabs.org/cover/1166589/
- added changes torpi_4_defconfigtoo (arm64 version)
- extended DFU entity list by confix.txt, cmdline.txt and Image.gz
- fixed missing '\0' at the end ofdfu_alt_infoenv
- added reviewed-by tags
- added patches for DFU MMC to remove file size limit
- added patch for fat write to fix issues on non-zero file offset (revealed by previous patch)v1: https://patchwork.ozlabs.org/cover/1162770/
- initial versionPatch summary:
Marek Szyprowski (6):
fat: write: fix broken write to fragmented files
fat: write: adjust data written in each partial write
dfu: mmc: rearrange the code
dfu: mmc: remove file size limit for io operations
usb:dwc2_udc_otg: add bcm2835 SoC (Raspberry Pi4) support
config: enable DFU over USB on Raspberry Pi4 boardsQuoted 9 lines
configs/rpi_4_32b_defconfig | 11 +++ configs/rpi_4_defconfig | 11 +++ drivers/dfu/dfu_mmc.c | 93 +++++++++++++--------- drivers/usb/gadget/dwc2_udc_otg.c | 2 + drivers/usb/gadget/dwc2_udc_otg_xfer_dma.c | 12 +-- fs/fat/fat_write.c | 8 +- include/configs/rpi.h | 20 +++++ 7 files changed, 112 insertions(+), 45 deletions(-)Best regards,
Lukasz Majewski
--
DENX Software Engineering GmbH, Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-59 Fax: (+49)-8142-66989-80 Email: lukma at denx.de