Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Memory configuration hardcoded at 512MB DDR3 #1

Closed
hno opened this issue May 19, 2012 · 7 comments
Closed

Memory configuration hardcoded at 512MB DDR3 #1

hno opened this issue May 19, 2012 · 7 comments

Comments

@hno
Copy link
Member

hno commented May 19, 2012

The u-boot spl loader currently have a hardcoded memory configuration of 512MB DDR3 at some fixed speed. Needs to either parameterize the DRAM settings somehow, or extend the code to probe for common configurations.

This involves both configuring the SDRAM controller proper, and to pass correct memory region information to the kernel on bootm.

@amery
Copy link
Member

amery commented May 25, 2012

you mentioned spl could boot linux directly, that means reading fat. which also means it could load script.bin... and so parse the dram data out of it. no?

example from the mele a1000.

[dram_para]
dram_baseaddr   = 0x40000000
dram_clk    = 360
dram_type   = 3
dram_rank_num   = 1
dram_chip_density   = 2048
dram_io_width   = 16
dram_bus_width  = 32
dram_cas    = 6
dram_zq = 0x7b
dram_odt_en = 0
dram_size   = 512
dram_tpr0   = 0x30926692
dram_tpr1   = 0x1090
dram_tpr2   = 0x1a0c8
dram_tpr3   = 0x0
dram_tpr4   = 0x0
dram_tpr5   = 0x0
dram_emr1   = 0x0
dram_emr2   = 0x0
dram_emr3   = 0x0

@hno
Copy link
Member Author

hno commented May 25, 2012

fre 2012-05-25 klockan 13:48 -0700 skrev Alejandro Mery:

you mentioned spl could boot linux directly, that means reading fat. which also means it could load script.bin... and so parse the dram data out of it. no?

spl could be made to read fat partitions.

it may be possible to add script.bin parsing as well, but memory is very
limited (24K) and I doubt it will all fit including script.bin.

In allwinner original bootloader the DRAM settings and a couple of other
parameters is stored in the header of boot0 (similar to spl).

Regards
Henrik

@hipboi
Copy link

hipboi commented May 30, 2012

spl could be made to read fat partitions.

it may be possible to add script.bin parsing as well, but memory is very
limited (24K) and I doubt it will all fit including script.bin.
yes, i have tried. The spl will be bigger than 24K if fat is enabled.

In allwinner original bootloader the DRAM settings and a couple of other
parameters is stored in the header of boot0 (similar to spl).

yes, the boot0 is handled by a pc program to get the right dram config from
script.bin.

One way to solve the problem is to try to get the real dram size by trying different paras, i will look into this.

@amery
Copy link
Member

amery commented May 30, 2012

One way to solve the problem is to try to get the real dram size by trying different paras, i will look into this.

you mean trial&error?

@hipboi
Copy link

hipboi commented May 30, 2012

yes, try to init dram for some times with different paras.

On Wed, May 30, 2012 at 4:03 PM, Alejandro Mery <
reply@reply.github.com

wrote:

One way to solve the problem is to try to get the real dram size by
trying different paras, i will look into this.

you mean trial&error?


Reply to this email directly or view it on GitHub:
https://github.com/hno/uboot-allwinner/issues/1#issuecomment-6004455

Keep simple, stay foolish.

@hipboi
Copy link

hipboi commented May 31, 2012

only the following parameters is configurable usually.

[dram_para]
dram_baseaddr   = 0x40000000
dram_clk    = 360
dram_type   = 3
dram_rank_num   = 1
dram_chip_density   = 2048
dram_io_width   = 16
dram_bus_width  = 32
dram_size   = 512

corresponding the following definition

#define DCR_TYPE_DDR2             (0)
#define DCR_TYPE_DDR3             (1)
#define DCR_IO_WIDTH_8            (1)
#define DCR_IO_WIDTH_16           (2)
#define DCR_CHIP_DENSITY_256Mb    (0)
#define DCR_CHIP_DENSITY_512Mb    (1)
#define DCR_CHIP_DENSITY_1Gb      (2)
#define DCR_CHIP_DENSITY_2Gb      (3)
#define DCR_CHIP_DENSITY_4Gb      (4)
#define DCR_CHIP_DENSITY_8Gb      (5)
#define DCR_BUS_WIDTH_16          (1)
#define DCR_BUS_WIDTH_32          (3)
#define DCR_ONE_RANK              (0)
#define DCR_TWO_RANKS             (1)

hno pushed a commit that referenced this issue Jul 2, 2012
This reverts commit 69d26d0.

Apparently, this commit got mainline only because of out-of-tree
port and causes breakage on board that is mainline. Revert.

Reason:
* The OOT board has 512MB of DRAM, enabling this additional address
  line enabled it to work fine with 512MB of RAM.
* Every mainline port has max. 256MB of DRAM, therefore this revert
  has no impact on any mainline port
* Though this caused a problem with new M28 board with 256MB of DRAM
  where the chips are wired differently. The patch-to-be-reverted
  caused the DRAM to behave like this:

  [128MB chunk #1][128MB chunk #1 again][128MB chunk #2][128MB chunk #2 again]

Therefore to retain the current one-memory-init-rules-them-all situation,
revert this patch until another board emerges and will actually be pushed
mainline that needs different setup.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Wolfgang Denk <wd@denx.de>
Cc: Detlev Zundel <dzu@denx.de>
Cc: Stefano Babic <sbabic@denx.de>
Cc: Fabio Estevam <festevam@gmail.com>
@hno
Copy link
Member Author

hno commented Oct 10, 2012

Memory code has been reworked based on standby code, and is now parameterized with per board configuration.

@hno hno closed this as completed Oct 10, 2012
slapin referenced this issue in slapin/uboot-allwinner Nov 14, 2012
ARM sunxi: Preliminary MK802 clone (Android-mini-pc) SPL support
slapin referenced this issue in slapin/uboot-allwinner Jan 14, 2013
Our `ops' information was converted to a local variable recently, and
apparently, old code relied on the fact that the global version was
often left in a valid mode. We can't make this assumption on local
structs, and we shouldn't be relying on a previous state anyway.

Instead, we initialize mode to 0 for don't-care situations (i.e., the
operation does not use OOB anyway) and MTD_OPS_PLACE_OOB when we want to
place OOB data.

This fixes a bug with nand_default_block_markbad(), where we catch on
the BUG() call in nand_fill_oob():

Kernel bug detected[hno#1]:
...
Call Trace:
[<80307350>] nand_fill_oob.clone.5+0xa4/0x15c
[<803075d8>] nand_do_write_oob+0x1d0/0x260
[<803077c4>] nand_default_block_markbad+0x15c/0x1a8
[<802e8c2c>] part_block_markbad+0x80/0x98
[<802ebc74>] mtd_ioctl+0x6d8/0xbd0
[<802ec1a4>] mtd_unlocked_ioctl+0x38/0x5c
[<800d9c60>] do_vfs_ioctl+0xa4/0x6e4
[<800da2e4>] sys_ioctl+0x44/0xa0
[<8001381c>] stack_done+0x20/0x40

Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
jwrdegoede referenced this issue in jwrdegoede/u-boot-sunxi May 26, 2014
Since B&R boards uses only MMC-Controller #1, it only
wastes time if we initialize #0 first to see that there is nothing.

Cc: <trini@ti.com>
Acked-by: Pantelis Antoniou <panto@antoniou-consulting.com>
Signed-off-by: Hannes Petermaier <oe5hpm@oevsv.at>
jwrdegoede referenced this issue in jwrdegoede/u-boot-sunxi Jul 26, 2014
Use get_device_and_partition() is better since:
1. It will call the device initialize function internally. So we can
remove the mmc intialization code to save many lines.
2. It is used by fatls/fatload/fatwrite. So saveenv & load env should
use it too.
3. It can parse the "D:P", "D", "D:", "D:auto" string to get correct
device and partition information by run-time.

Also we remove the FAT_ENV_DEVICE and FAT_ENV_PART. We use a string:
FAT_ENV_DEVICE_AND_PART.
For at91sam9m10g45ek, it is "0". That means use device 0 and if:
a)device 0 has no partition table, use the whole device as a FAT file
system.
b)device 0 has partittion table, use the partition #1.

Refer to the commit: 10a37fd for details of device & partition string.

Signed-off-by: Josh Wu <josh.wu@atmel.com>
Reviewed-by: Stephen Warren <swarren@nvidia.com>
wens pushed a commit to wens/u-boot-sunxi that referenced this issue Nov 28, 2014
Attempting to run the sandbox leads to a segfault, because some dynamic
libraries (outside of u-boot) attempt to use malloc() to allocate memory
before u-boot's gd variable is initialized.

Check for gd not being NULL in the SYS_MALLOC_F_LEN handling, so that
malloc() doesn't crash when called at this point.

 $ gdb -q --args ./u-boot
 (gdb) r
 Program received signal SIGSEGV, Segmentation fault.
 0x0000000000412b9b in malloc (bytes=bytes@entry=37) at common/dlmalloc.c:2184
 2184		if (!(gd->flags & GD_FLG_RELOC)) {
 (gdb) p gd
 $1 = (gd_t *) 0x0
 (gdb) bt
 #0  0x0000000000412b9b in malloc (bytes=bytes@entry=37) at common/dlmalloc.c:2184
 linux-sunxi#1  0x00007ffff75bf8e1 in set_binding_values (domainname=0x7ffff11f4f12 "libgpg-error", dirnamep=0x7fffffffe168, codesetp=0x0)
     at bindtextdom.c:228
 linux-sunxi#2  0x00007ffff75bfb4c in set_binding_values (codesetp=0x0, dirnamep=0x7fffffffe168, domainname=<optimized out>) at bindtextdom.c:350
 linux-sunxi#3  __bindtextdomain (domainname=<optimized out>, dirname=0x7ffff11f4f00 "/usr/share/locale") at bindtextdom.c:348
 linux-sunxi#4  0x00007ffff11eca17 in ?? () from /lib/x86_64-linux-gnu/libgpg-error.so.0
 linux-sunxi#5  0x00007ffff7dea9fa in call_init (l=<optimized out>, argc=argc@entry=1, argv=argv@entry=0x7fffffffe208,
     env=env@entry=0x7fffffffe218) at dl-init.c:78
 linux-sunxi#6  0x00007ffff7deaae3 in call_init (env=0x7fffffffe218, argv=0x7fffffffe208, argc=1, l=<optimized out>) at dl-init.c:36
 linux-sunxi#7  _dl_init (main_map=0x7ffff7ffe1a8, argc=1, argv=0x7fffffffe208, env=0x7fffffffe218) at dl-init.c:126
 linux-sunxi#8  0x00007ffff7ddd1ca in _dl_start_user () from /lib64/ld-linux-x86-64.so.2

Signed-off-by: Rabin Vincent <rabin@rab.in>
Acked-by: Simon Glass <sjg@chromium.org>
ssvb added a commit that referenced this issue Sep 17, 2015
Currently some uninitialized padding bytes are written to the output
file, as can be confirmed with valgrind:

$ valgrind tools/mksunxiboot spl/u-boot-spl.bin spl/sunxi-spl.bin

==5581== Syscall param write(buf) points to uninitialised byte(s)
==5581==    at 0x4F0F940: __write_nocancel (in /lib64/libc-2.20.so)
==5581==    by 0x400839: main (in /tmp/u-boot/tools/mksunxiboot)
==5581==  Address 0xffeff5d3c is on thread 1's stack
==5581==  in frame #1, created by main (???)

This patch fixes the problem by clearing the whole structure instead
of just a portion of it.

Signed-off-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>
Acked-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
amery pushed a commit that referenced this issue Nov 5, 2015
>From source code comments:
"x0: 0 flush & invalidate, 1 invalidate only"

Current value 0xffff can make invalidate work, since we only judge whether
input value is 0 or not, see following code:
"
    tbz     w1, #0, 1f
    dc      isw, x9
    b       2f
1:  dc      cisw, x9      /* clean & invalidate by set/way */
2:  subs    x6, x6, #1    /* decrement the way */
"

Later we may add "2 clean only" support. So following the comments,
correct value from 0xffff to 1.

Signed-off-by: Peng Fan <Peng.Fan@freescale.com>
Cc: York Sun <yorksun@freescale.com>
Cc: Albert Aribaud <albert.u.boot@aribaud.net>
wens pushed a commit to wens/u-boot-sunxi that referenced this issue May 26, 2016
After power-on, both LAPIC and I/O APIC appear with the same APIC ID
zero, which creates an ID conflict. When generating MP table, U-Boot
reports zero as the LAPIC ID in the processor entry, and zero as the
I/O APIC ID in the I/O APIC as well as the I/O interrupt assignment
entries. Such MP table confuses Linux kernel and finally a kernel
panic is seen during boot:

  BUG: unable to handle kernel paging request at ffff9000
  IP: [<c101d462>] native_io_apic_write+0x22/0x30
  *pdpt = 00000000014fb001 *pde = 00000000014ff067 *pte = 0000000000000000
  Oops: 0002 [linux-sunxi#1]
  Modules linked in:
  Pid: 1, comm: swapper Tainted: G        W    3.8.7 linux-sunxi#3 intel galileo/galileo
  EIP: 0060:[<c101d462>] EFLAGS: 00010086 CPU: 0
  EIP is at native_io_apic_write+0x22/0x30
  ...
  Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000009

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
amery pushed a commit that referenced this issue Jul 29, 2016
Booting a NXP kernel with mainline U-boot leads to the following kernel
crash:

caam: probe of 30900000.caam failed with error -11
Unable to handle kernel NULL pointer dereference at virtual address 00000004
pgd = 80004000
[00000004] *pgd=00000000
Internal error: Oops: 805 [#1] PREEMPT SMP ARM

This happens because NXP kernel expects MX7 to boot in secure mode,
so introduce mx7dsabresd_secure_defconfig that selects CONFIG_MX7_SEC
and allows booting a NXP provided kernel successfully.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
amery pushed a commit that referenced this issue Dec 4, 2016
On DRA7xx platform, CPU Core 1 is not used in u-boot. However, in some
cases it is need to make secure API calls from Core 1. This patch adds
an assembly function to make a secure (SMC) call from CPU Core #1.

Signed-off-by: Harinarayan Bhatta <harinarayan@ti.com>
Signed-off-by: Andrew F. Davis <afd@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
amery pushed a commit that referenced this issue Jul 30, 2017
It was observed that on Intel MinnowMax board, when xHCI is enabled
in the BayTrail SoC, with a USB 3.0 device connected to the bottom
USB 3.0 port (mapped to xHCI root port #7), its PORTSC register is
always 0x201203 (CCS = 1, CSC = 0). The root cause of such behavior
is unknown yet. Connect status change bit is set on the same port
with a USB 2.0 device (mapped to xHCI port #1, which is a different
port on the root hub).

With current logic in usb_scan_port(), the enumeration process will
abort if it does not detect a connect status change on a hub port.
However since a device connection status is correctly reported, the
enumeration process can still continue.

With this change, USB device connected to the bottom blue port on
MinnowMax board can be enumerated under either SS or HS mode.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Stefan Roese <sr@denx.de>
Tested-by: Stefan Roese <sr@denx.de>
Tested-by: Dinh Nguyen <dinguyen@kernel.org>
amery pushed a commit that referenced this issue Aug 7, 2017
Strings read from devices may sometimes fail to be
NULL-terminated.   The functions in lib/string.c are subject to
failure in this case.   Protect against observed failures in
set_gpt_info() by switching to length-checking variants with a length
limit of the maximum possible partition table length.  At the same
time, add a few checks for NULL string pointers.

Here is an example as observed in sandbox under GDB:

    => gpt verify host 0 $partitions
    Program received signal SIGSEGV, Segmentation fault.
    0x0000000000477747 in strlen (s=0x0) at lib/string.c:267
    267             for (sc = s; *sc != '\0'; ++sc)
    (gdb) bt
    #0  0x0000000000477747 in strlen (s=0x0) at lib/string.c:267
    #1  0x00000000004140b2 in set_gpt_info (str_part=<optimized out>,
    str_disk_guid=str_disk_guid@entry=0x7fffffffdbe8, partitions=partitions@entry=0x7fffffffdbd8,
    parts_count=parts_count@entry=0x7fffffffdbcf "", dev_desc=<optimized out>) at cmd/gpt.c:415
    #2  0x00000000004145b9 in gpt_verify (str_part=<optimized out>, blk_dev_desc=0x7fffef09a9d0) at cmd/gpt.c:580
    #3  do_gpt (cmdtp=<optimized out>, flag=<optimized out>, argc=<optimized out>, argv=0x7fffef09a8f0)
    at cmd/gpt.c:783
    #4  0x00000000004295b0 in cmd_call (argv=0x7fffef09a8f0, argc=0x5, flag=<optimized out>,
    cmdtp=0x714e20 <_u_boot_list_2_cmd_2_gpt>) at common/command.c:500
    #5  cmd_process (flag=<optimized out>, argc=0x5, argv=0x7fffef09a8f0,
    repeatable=repeatable@entry=0x726c04 <flag_repeat>, ticks=ticks@entry=0x0) at common/command.c:539

Suggested-by: Lothar Waßmann <LW@karo-electronics.de>
Signed-off-by: Alison Chaiken <alison@peloton-tech.com>
amery pushed a commit that referenced this issue Apr 29, 2018
If generating a script image and no datafile has been passed in, mkimage
dies with SIGSEGV:

  #0  __strchr_sse2 () at ../sysdeps/x86_64/multiarch/../strchr.S:32
  #1  0x0000000000403818 in main
      at tools/mkimage.c:503

Add explicit test for datafile to fix this.

Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
amery pushed a commit that referenced this issue Apr 30, 2018
Booting the following image with tip-of-tree we get a CAAM DECO error (and
subsequent crash due to a kernel bug in 4.1).

http://freescale.github.io/#download -> BoardsWaRPboard community - WaRP -
Wearable Reference PlatformFSL Community BSP 2.3fsl-image-multimediawayland

Image: fsl-image-multimedia-imx7s-warp-20180323-90.rootfs.sdcard

Error:
caam 30900000.caam: Entropy delay = 3200
caam 30900000.caam: failed to acquire DECO 0
<snip>
caam 30900000.caam: failed to acquire DECO 0
caam 30900000.caam: Entropy delay = 12400
caam 30900000.caam: failed to acquire DECO 0
caam 30900000.caam: failed to instantiate RNG
------------[ cut here ]------------
WARNING: CPU: 0 PID: 1 at
/home/jenkins/workspace/fsl-community-bsp-pyro_xwayland_2/build/tmp/work-shared/imx7s-warp/kernel-source/mm/vmalloc.c:1465
caam_remove+0x6)
Trying to vfree() nonexistent vm area (88047000)
Modules linked in:
CPU: 0 PID: 1 Comm: swapper/0 Not tainted
4.1.36-4.1-1.0.x-imx-warp7+ga543d1b #1
Hardware name: Freescale i.MX7 Dual (Device Tree)
[<80015d54>] (unwind_backtrace) from [<80012688>] (show_stack+0x10/0x14)
[<80012688>] (show_stack) from [<8076e810>] (dump_stack+0x78/0x8c)
[<8076e810>] (dump_stack) from [<800346a0>]
(warn_slowpath_common+0x80/0xb0)
[<800346a0>] (warn_slowpath_common) from [<80034700>]
(warn_slowpath_fmt+0x30/0x40)
[<80034700>] (warn_slowpath_fmt) from [<8054c278>] (caam_remove+0x6c/0x3f4)
[<8054c278>] (caam_remove) from [<8054ce74>] (caam_probe+0x874/0xfa8)
[<8054ce74>] (caam_probe) from [<80382a7c>] (platform_drv_probe+0x48/0xa4)
[<80382a7c>] (platform_drv_probe) from [<80381328>]
(driver_probe_device+0x174/0x2a8)
[<80381328>] (driver_probe_device) from [<8038152c>]
(__driver_attach+0x8c/0x90)
[<8038152c>] (__driver_attach) from [<8037f9d4>]
(bus_for_each_dev+0x68/0x9c)
[<8037f9d4>] (bus_for_each_dev) from [<80380a68>]
(bus_add_driver+0xf4/0x1e8)
[<80380a68>] (bus_add_driver) from [<80381b38>] (driver_register+0x78/0xf4)
[<80381b38>] (driver_register) from [<80009738>]
(do_one_initcall+0x8c/0x1d0)
[<80009738>] (do_one_initcall) from [<80a66dac>]
(kernel_init_freeable+0x140/0x1d0)
[<80a66dac>] (kernel_init_freeable) from [<8076aa38>]
(kernel_init+0x8/0xe8)
[<8076aa38>] (kernel_init) from [<8000f468>] (ret_from_fork+0x14/0x2c)
---[ end trace d5f941204ed8cb28 ]---
caam: probe of 30900000.caam failed with error -11
Unable to handle kernel NULL pointer dereference at virtual address
00000004
pgd = 80004000
[00000004] *pgd=00000000
Internal error: Oops: 805 [#1] PREEMPT SMP ARM
<snip>
[<8055cdf8>] (caam_sm_startup) from [<80aa83c8>] (caam_sm_init+0x50/0x58)
[<80aa83c8>] (caam_sm_init) from [<80009738>] (do_one_initcall+0x8c/0x1d0)
[<80009738>] (do_one_initcall) from [<80a66dac>]
(kernel_init_freeable+0x140/0x1d0)
[<80a66dac>] (kernel_init_freeable) from [<8076aa38>]
(kernel_init+0x8/0xe8)
[<8076aa38>] (kernel_init) from [<8000f468>] (ret_from_fork+0x14/0x2c)
Code: e59d300c e2832010 e5843008 e5834068 (e58a2004)
---[ end trace d5f941204ed8cb29 ]---
Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b

Fix: Enable the CAAM correctly by setting CONFIG_ARMV7_BOOT_SEC_DEFAULT=y
in the upstream defconfig.

Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Cc: Fabio Estevam <fabio.estevam@nxp.com>
Cc: Breno Lima <breno.lima@nxp.com>
Reviewed-by: Fabio Estevam <fabio.estevam@nxp.com>
Tested-by: Breno Lima <breno.lima@nxp.com>
amery pushed a commit that referenced this issue Aug 12, 2018
Commit 5d8fae7 ("dfu: avoid memory leak") brings a regression which
described below. This patch is effectively reverting that commit, adding
corresponding comment to avoid such regressions in future.

In case of error in dfu_config_entities(), it frees "dfu" array, which
leads to "data abort" in dfu_free_entities(), which tries to free the
same array (and even tries to access it from linked list first). The
issue occurs e.g. when partition table on device does not match
$dfu_alt_info layout:

    => dfu 0 mmc 1
    Couldn't find part #2 on mmc device #1
    DFU entities configuration failed!
    data abort

To fix this issue, do not free "dfu" array in dfu_config_entities(). It
will be freed later in dfu_free_entities().

Tested on BeagleBone Black (where this regression was originally found).

Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
amery pushed a commit that referenced this issue Aug 26, 2018
Compiling U-Boot with ubsan/asan libraries and running it in sandbox
may lead to below backtrace:

 => avb init 0
 => avb verify
 ## Android Verified Boot 2.0 version 1.1.0
read_is_device_unlocked not supported yet
common/avb_verify.c:407:31: runtime error: division by zero
AddressSanitizer:DEADLYSIGNAL
Reviewed-by: Igor Opaniuk <igor.opaniuk@linaro.org>

=================================================================
==9388==ERROR: AddressSanitizer: FPE on unknown address 0x0000004b467f \
    (pc 0x0000004b467f bp 0x000000000000 sp 0x7ffd899fe150 T0)
    #0 0x4b467e in mmc_byte_io common/avb_verify.c:407
    #1 0x4b4c47 in mmc_byte_io common/avb_verify.c:532
    #2 0x4b4c47 in read_from_partition common/avb_verify.c:533
    #3 0x69dc0d in load_and_verify_vbmeta lib/libavb/avb_slot_verify.c:560
    #4 0x6a1ee6 in avb_slot_verify lib/libavb/avb_slot_verify.c:1139
    #5 0x45dabd in do_avb_verify_part cmd/avb.c:245
    #6 0x4af77c in cmd_call common/command.c:499
    #7 0x4af77c in cmd_process common/command.c:538
    #8 0x46bafc in run_pipe_real common/cli_hush.c:1677
    #9 0x46bafc in run_list_real common/cli_hush.c:1875
    #10 0x46c780 in run_list common/cli_hush.c:2024
    #11 0x46c780 in parse_stream_outer common/cli_hush.c:3216
    #12 0x46d34b in parse_file_outer common/cli_hush.c:3299
    #13 0x4ad609 in cli_loop common/cli.c:217
    #14 0x4625ae in main_loop common/main.c:65
    #15 0x46f2d1 in run_main_loop common/board_r.c:648
    #16 0x640253 in initcall_run_list lib/initcall.c:30
    #17 0x46f9d0 in board_init_r common/board_r.c:879
    #18 0x40539b in main arch/sandbox/cpu/start.c:321
    #19 0x7fa94925f82f in __libc_start_main (/lib/x86_64-linux-gnu/libc.so.6+0x2082f)
    #20 0x408908 in _start (/srv/R/u-boot-master/u-boot+0x408908)

AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: FPE common/avb_verify.c:407 in mmc_byte_io
==9388==ABORTING

Signed-off-by: Eugeniu Rosca <erosca@de.adit-jv.com>
mrnuke pushed a commit to mrnuke/u-boot that referenced this issue Sep 7, 2021
Enable RTC command to be able to check available.
And also enable ZynqMP RTC driver to be possible to use by default.

Here is the list when both drivers are enabled:
ZynqMP> rtc list
RTC #0 - rtc_emul
RTC linux-sunxi#1 - rtc@ffa60000

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
mrnuke pushed a commit to mrnuke/u-boot that referenced this issue Jan 24, 2022
Lot of PCI and PCIe controllers are using standard Config Address for PCI
Configuration Mechanism linux-sunxi#1 or its extended version.

So add PCI_CONF1_ADDRESS() and PCI_CONF1_EXT_ADDRESS() macros into U-Boot's
pci.h header file which can be suitable for most PCI and PCIe controller
drivers. Drivers do not have to invent their own macros and can use these
new U-Boot macros.

Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
mrnuke pushed a commit to mrnuke/u-boot that referenced this issue Jan 24, 2022
PCI gt64120 driver uses standard format of Config Address for PCI
Configuration Mechanism linux-sunxi#1.

So use new U-Boot macro PCI_CONF1_ADDRESS() and remove old custom driver
address macros.

Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
mrnuke pushed a commit to mrnuke/u-boot that referenced this issue Jan 24, 2022
PCI mpc85xx driver uses extended format of Config Address for PCI
Configuration Mechanism linux-sunxi#1.

So use new U-Boot macro PCI_CONF1_EXT_ADDRESS().

Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
mrnuke pushed a commit to mrnuke/u-boot that referenced this issue Jan 24, 2022
PCI msc01 driver uses standard format of Config Address for PCI
Configuration Mechanism linux-sunxi#1 but with cleared Enable bit.

So use new U-Boot macro PCI_CONF1_ADDRESS() with clearing PCI_CONF1_ENABLE
bit and remove old custom driver address macros.

Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
mrnuke pushed a commit to mrnuke/u-boot that referenced this issue Jan 24, 2022
PCI mvebu driver uses extended format of Config Address for PCI
Configuration Mechanism linux-sunxi#1.

So use new U-Boot macro PCI_CONF1_EXT_ADDRESS() and remove old custom
driver address macros.

Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
mrnuke pushed a commit to mrnuke/u-boot that referenced this issue Jan 24, 2022
PCI tegra driver uses extended format of Config Address for PCI
Configuration Mechanism linux-sunxi#1 but with cleared Enable bit.

So use new U-Boot macro PCI_CONF1_EXT_ADDRESS() with clearing
PCI_CONF1_ENABLE bit and remove old custom driver address function.

Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
mrnuke pushed a commit to mrnuke/u-boot that referenced this issue Jan 24, 2022
PCI fsl driver uses extended format of Config Address for PCI
Configuration Mechanism linux-sunxi#1.

So use new U-Boot macro PCI_CONF1_EXT_ADDRESS().

Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
mrnuke pushed a commit to mrnuke/u-boot that referenced this issue Jan 24, 2022
PCI mediatek driver uses extended format of Config Address for PCI
Configuration Mechanism linux-sunxi#1 but with cleared Enable bit.

So use new U-Boot macro PCI_CONF1_EXT_ADDRESS() with clearing
PCI_CONF1_ENABLE bit and remove old custom driver address macros.

Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
mrnuke pushed a commit to mrnuke/u-boot that referenced this issue Jan 24, 2022
PCI sh7780 driver uses standard format of Config Address for PCI
Configuration Mechanism linux-sunxi#1.

So use new U-Boot macro PCI_CONF1_ADDRESS().

Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
mrnuke pushed a commit to mrnuke/u-boot that referenced this issue Jan 24, 2022
x86 platform uses standard format of Config Address for PCI Configuration
Mechanism linux-sunxi#1. So use new U-Boot macro PCI_CONF1_ADDRESS().

Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
mrnuke pushed a commit to mrnuke/u-boot that referenced this issue Jan 24, 2022
mcf5445x platform uses standard format of Config Address for PCI
Configuration Mechanism linux-sunxi#1. So use new U-Boot macro PCI_CONF1_ADDRESS().

Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
mrnuke pushed a commit to mrnuke/u-boot that referenced this issue Jan 24, 2022
sh7751 platform uses standard format of Config Address for PCI
Configuration Mechanism linux-sunxi#1.

Commit 72c2f4a ("pci: sh7751: Convert to DM and DT probing") which did
conversion of PCI sh7751 driver to DM, broke access to config space as that
commit somehow swapped device and function bits in config address.

Fix all these issues by using new U-Boot macro PCI_CONF1_ADDRESS() which
calculates Config Address correctly.

Also remove nonsense function sh7751_pci_addr_valid() which was introduced
in commit 72c2f4a ("pci: sh7751: Convert to DM and DT probing")
probably due to workarounded issues with mixing/swapping device and
function bits of config address which probably resulted in non-working
access to some devices. With correct composing of config address there
should not be such issue anymore.

Signed-off-by: Pali Rohár <pali@kernel.org>
Fixes: 72c2f4a ("pci: sh7751: Convert to DM and DT probing")
Cc: Marek Vasut <marek.vasut+renesas@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants