Re: [PATCH V2] ubifs: Call ubifs_iput when ubifs_iget is used

Hello Micahel, On 10.08.24 14:57, Michael Trimarchi wrote: The inode should be freed after a reference is get to avoid memory leak Tested-by: Alexander Dahl Link: https://lore.kernel.org/u-boot/b698ec3e-d857-6512-8cc9-4edcab0a4...@denx.de/T/#t Link: https://lore.kernel.org/all/8f3a7059-6330-

Re: [PATCH 07/21] Revert "net: wget: Support retransmission a dropped packet"

Dear Simon, Could you inform me how the wget test was broken? As I explained in the commit log, I fixed the bug in specific conditions. Without the details of how the break happened, it is difficult to find a proper patch. Best regards, Yasuharu Shibata On Sun, 11 Aug 2024 at 05:52, Simon G

[PATCH 00/21] Tidy up console recording in tests

This series started as a small fix for checking for an empty line, but in the process several other problems were found and fixed: - fix tests which use console recording but don't set the flag - fix wget test and make it run on sandbox - drop unnecessary resetting of the console in tests - drop u

Re: [PATCH v2 2/2] i2c: samsung: Support platforms other than EXYNOS4 and EXYNOS5

Hi David, On Sat, Aug 10, 2024 at 01:05:06AM +0200, David Virag wrote: > Hi Henrik, > > On Fri, 2024-08-09 at 23:08 +0200, Henrik Grimler wrote: > > Hi David, > > > > Thinking about this a bit more... > > > > On Fri, Aug 02, 2024 at 09:19:16PM +0200, David Virag wrote: > [snip] > > > @@ -9,11 +

[PATCH 20/21] test: Tidy up checking for console end

Use the ut_assert_console_end() function provided, rather than doing it separately. Signed-off-by: Simon Glass --- test/cmd/bdinfo.c | 8 +- test/cmd/exit.c | 24 +++--- test/cmd/fdt.c | 178 test/cmd/font.c | 8 +- te

[PATCH 21/21] doc: Add a few notes about how to use console checking

Tidy up the existing docs in line with current conventions. Signed-off-by: Simon Glass --- doc/develop/commands.rst | 9 + doc/develop/tests_writing.rst | 1 - 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/doc/develop/commands.rst b/doc/develop/commands.rst index 6

[PATCH 19/21] test: Use UTF_CONSOLE in remaining tests

Set this flag rather than doing things manually in the test. Signed-off-by: Simon Glass --- test/common/cread.c | 4 +--- test/common/test_autoboot.c | 3 +-- test/lib/test_print.c | 10 ++ test/print_ut.c | 15 --- 4 files changed, 4 insertions(+

[PATCH 18/21] test: log: Use UTF_CONSOLE in tests

Set this flag rather than doing things manually in the test. Signed-off-by: Simon Glass --- include/test/log.h | 5 +++-- test/log/cont_test.c| 1 - test/log/log_filter.c | 3 --- test/log/log_test.c | 19 --- test/log/nolog_ndebug.c | 1 - test/log/nolog_test

[PATCH 17/21] test: hush: Use UTF_CONSOLE in tests

Set this flag rather than doing things manually in the test. Signed-off-by: Simon Glass --- test/hush/dollar.c | 10 +++--- test/hush/loop.c | 11 +++ 2 files changed, 6 insertions(+), 15 deletions(-) diff --git a/test/hush/dollar.c b/test/hush/dollar.c index 4caa07c192a..91b0970

[PATCH 16/21] test: dm: Use UTF_CONSOLE in tests

Set this flag rather than doing things manually in the test. Signed-off-by: Simon Glass --- test/dm/acpi.c| 15 --- test/dm/blkmap.c | 4 +--- test/dm/cros_ec.c | 14 -- test/dm/host.c| 4 +--- test/dm/mux-cmd.c | 2 +- test/dm/rng.c | 2 -- test/dm/rtc.

[PATCH 15/21] test: cmd: Drop unnecessary console_record_reset_enable()

It is seldom necessary to call this function. Drop its use in the command tests. Add a few extra checks to the wget test so that resetting is not needed. Signed-off-by: Simon Glass --- test/cmd/addrmap.c | 1 - test/cmd/bdinfo.c | 4 -- test/cmd/exit.c| 12 -- test/cmd/f

[PATCH 14/21] test: cmd: Use UTF_CONSOLE in tests

Set this flag rather than doing things manually in the test. Signed-off-by: Simon Glass --- test/cmd/pinmux.c | 3 ++- test/cmd/wget.c | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/test/cmd/pinmux.c b/test/cmd/pinmux.c index 0b4e001fa1e..7ee92690756 100644 --- a/test/

[PATCH 13/21] test: fdt: Move common code into the setup functions

Quite a lot of tests have the same two lines of code at the start. Move this into the two setup functions to reduce redundancy. Add a line to check the output from set_working_fdt_addr() since this is always emitted. Signed-off-by: Simon Glass --- test/cmd/fdt.c | 92 +-

[PATCH 12/21] test: fdt: Check internal-function return values

Some functions are using asserts but the result of the functions themselves is not checked. This means that if a test fails, the result is not noticed until later, which can be confusing to debug. Add the missing asserts. Signed-off-by: Simon Glass --- test/cmd/fdt.c | 153

[PATCH 11/21] test: boot: Use UTF_CONSOLE in tests

Set this flag rather than doing things manually in the test. Drop unnecessary calls to console_record_reset_enable() Signed-off-by: Simon Glass --- test/boot/bootdev.c | 31 ++ test/boot/bootflow.c | 53 ++-- test/boot/bootmeth.c

[PATCH 10/21] test: bloblist: Use UTF_CONSOLE in tests

Set this flag rather than doing things manually in the test. Drop the code which is now unnecessary. Signed-off-by: Simon Glass --- test/bloblist.c | 14 ++ 1 file changed, 2 insertions(+), 12 deletions(-) diff --git a/test/bloblist.c b/test/bloblist.c index 7c63682908a..fd85c7ab7

[PATCH 09/21] test: Update NAND test to avoid extra macros

Use a single test function with a for() loop instead of using macros to create multiple test functions. Add a message so it is clear what piece the test is up to, if it fails. Signed-off-by: Simon Glass --- test/dm/nand.c | 30 +++--- 1 file changed, 15 insertions(+), 15

[PATCH 08/21] sandbox: Enable wget command

Enable this so that the tests run. Fix a few warnings in the code so that CI passes. Signed-off-by: Simon Glass --- cmd/Kconfig | 1 + net/wget.c | 7 +++ 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/cmd/Kconfig b/cmd/Kconfig index 8af136d2bd4..43f78a5aeb1 100644 --- a/c

[PATCH 07/21] Revert "net: wget: Support retransmission a dropped packet"

This breaks the wget tests, so revert it until a proper fix can be figured out. This reverts commit cab7867cff33f9c21025102ea9ca3b26e362fb52. Signed-off-by: Simon Glass --- net/wget.c | 16 +--- 1 file changed, 5 insertions(+), 11 deletions(-) diff --git a/net/wget.c b/net/wget.c

[PATCH 05/21] test: Rename UTF_CONSOLE_REC to UTF_CONSOLE

The _REC suffix doesn't add much. Really what we want to know is whether the test uses the console, so rename this flag. Signed-off-by: Simon Glass --- doc/develop/tests_writing.rst | 4 ++-- include/dm/test.h | 2 +- include/test/test.h | 2 +- test/boot/bootflow.c

[PATCH 04/21] test: Drop the blank line before test macros

Most tests don't have this. It helps to keep the test declaration clearly associated with the function it relates to, rather than the next one in the file. Remove the extra blank line and mention this in the docs. Signed-off-by: Simon Glass --- doc/develop/tests_writing.rst | 3 +++ test/cmd/ar

[PATCH 06/21] mmc: Drop the blank line before accesses

Several mmc subcommand print a blank line before starting and after finishing. It isn't necessary to do both, so drop the first one. It is questionable whether these command should produce any output at all, but leave it for now. Signed-off-by: Simon Glass --- cmd/mmc.c | 12 ++-- 1 fi

[PATCH 02/21] test: Fail when an empty line is expected but not present

The existing implementation of ut_assert_nextline_empty() cannot distinguish between an empty line and no line at all. It can in fact be called at the end of the recorded output and will happily return success. Adjust the logic so that this condition is detected. Show a failure message in this cas

[PATCH 01/21] buildman: Make test_process_limit handle time.monotonic()

Newer versions of filelock use time.monotonic() instead of time.time(). Update the test the handle this. It would be better if filelock had support for writing unit tests which use locking. Signed-off-by: Simon Glass --- tools/buildman/test.py | 1 + 1 file changed, 1 insertion(+) diff --git

Aw: [PATCH 13/13] clk: mediatek: mt7988: rename CK to CLK

Hi, tested on Banana Pi R4 does not break downstream DTS, upstream dts is nearly empty, have tried with our internal dts(i)...so we srtil nee some driver patches to add/fix compatibles and work without syscon. tested: mmc, usb, network, spi, pci/nvme Tested-by: Frank Wunderlich regards Fran

Re: [PATCH v4 00/21] Universal Payload initial series

On Wed, 07 Aug 2024 16:47:18 -0600, Simon Glass wrote: > Universal Payload (UPL) is an Industry Standard for firmware > components[1]. UPL is designed to improve interoperability within the > firmware industry, allowing mixing and matching of projects with less > friction and fewer project-specifi

[PATCH] pinmux: generic: Use ENOENT instead of ENOSYS

ENOSYS should only be used when a subsystem is completely absent. Convert its use in pinctrl-generic to ENOENT, which better reflects the error condition (that a function/group/pin is missing). Signed-off-by: Sean Anderson --- drivers/pinctrl/pinctrl-generic.c | 26 +- 1

[PATCH] sandbox: Fix pinmux warnings with non-test devicetrees

The sandbox pinmux driver is used in the non-test devicetree as well as the test one. I didn't realize this when I modified the driver for tests, and so broke the regular use case (which only resulted in warnings). First, making the pinmux and the UART group available in all boot phases to avoid EN

I think I found a typographical error

Hi, I was looking in the 'arch > arm > cpu > armv7 > start.S' and I think I found a typographical error. On line 45, 'This allows U-Boot to loaded to ...' should be 'This allows U-Boot to *be* loaded to ...' Please check if it's right or not. Best regards, Tae Hyun Kim

Re: Selecting serial device for console

Hi Simon, Mark, thanks for your replies! Am 08.08.24 um 16:28 schrieb Simon Glass: Hi Mark, On Tue, 6 Aug 2024 at 16:33, Mark Kettenis wrote: From: Simon Glass Date: Tue, 6 Aug 2024 15:50:44 -0600 Hi Simon, Hi Fiona, On Thu, 1 Aug 2024 at 08:07, Fiona Klute wrote: Hi everyone, ho

Aw: [PATCH 15/15] clk: mediatek: mt7986: rename CK to CLK

Hi tested on r3 and r3mini does not break downstream dts and works with slightly modified upstream dts. tested mmc,usb,network,spi,pci/nvme Tested-by: Frank Wunderlich regards Frank > Gesendet: Samstag, 03. August 2024 um 10:40 Uhr > Von: "Christian Marangi" > An: "Tom Rini" , "Lukasz Maje

Re: [PATCH] ubifs: Call ubifs_iput when ubifs_iget is used

Hi Heiko On Sat, Aug 10, 2024 at 1:11 PM Heiko Schocher wrote: > > Hello Michael, > > On 07.08.24 17:42, Michael Trimarchi wrote: > > The inode should be freed after a reference is get to avoid > > memory leak > > > > Tested-by: Alexander Dahl > > Link: > > https://lore.kernel.org/u-boot/b698ec

[PATCH V2] ubifs: Call ubifs_iput when ubifs_iget is used

The inode should be freed after a reference is get to avoid memory leak Tested-by: Alexander Dahl Link: https://lore.kernel.org/u-boot/b698ec3e-d857-6512-8cc9-4edcab0a4...@denx.de/T/#t Link: https://lore.kernel.org/all/8f3a7059-6330-f332-8e9f-729b853e0...@denx.de/T/ Co-Developed-by: Heiko Schoc

Aw: [PATCH 8/8] clk: mediatek: mt7622: add missing A1/2SYS clock ID

Hi tested full series on bananapi r64 does not break downstream DTS and works with modified upstream dts tested mmc, pci, sata, ethernet, usb Tested-by: Frank Wunderlich regards Frank > Gesendet: Freitag, 02. August 2024 um 15:48 Uhr > Von: "Christian Marangi" > An: "Lukasz Majewski" , "Se

Aw: [PATCH 5/5] clk: mediatek: mt7623: remap peri clock ID and add MUX

Hi, does not break downstream dts (still have some issues with ethernet without the series, which i cannot nail down) so only tested: usb,mmc,pci,sata also works with upstream dts and a bunch of modifications (i still need to clean) Tested-by: Frank Wunderlich regards Frank > Gesendet: Fre

Re: [PATCH] ubifs: Call ubifs_iput when ubifs_iget is used

Hello Michael, On 07.08.24 17:42, Michael Trimarchi wrote: The inode should be freed after a reference is get to avoid memory leak Tested-by: Alexander Dahl Link: https://lore.kernel.org/u-boot/b698ec3e-d857-6512-8cc9-4edcab0a4...@denx.de/T/#t Link: https://lore.kernel.org/all/8f3a7059-6330-