public inbox for ecos-patches@sourceware.org
 help / color / mirror / Atom feed
* [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
@ 2011-05-03 22:11 bugzilla-daemon
  2011-05-04  7:41 ` [Bug 1001219] " bugzilla-daemon
                   ` (67 more replies)
  0 siblings, 68 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-05-03 22:11 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

           Summary: Ethernet driver for STM32 connectivity line with port
                    on MMstm32f107 board.
           Product: eCos
           Version: CVS
          Platform: Other (please specify)
        OS/Version: ARM
            Status: UNCONFIRMED
          Severity: enhancement
          Priority: low
         Component: Patches and contributions
        AssignedTo: unassigned@bugs.ecos.sourceware.org
        ReportedBy: jerzdy@gmail.com
                CC: ecos-patches@ecos.sourceware.org
             Class: Advice Request


Created an attachment (id=1218)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1218)
Ethernet driver for STM32 connectivity line with port on MMstm32f107 board

Hello all,

I would like to contribute Ethernet driver for STM32 connectivity line but due
to fact that the some support from HAL package is also required. I provide port
on MMstm32f107 board.

Beside of extending stm32/var include files to provide description of Ethernet
module and adding support of new model of clock distribution in STM32
connectivity line I would like to point functionality which cause me a lot of
problem to start Ethernet driver i.e.
stm32_misc.c in function hal_variant_init() is enabled clocks for all APB
peripherals. In case of Ethernet driver it should be done :
1) after PLL configuration (switching it on this stage cause hang up of MDIO
bus) 
2) only used peripherals (  switching on all APB module cause hang up of RMII
bus)

My suggestion is to move it to board dependent area i.e. hal_platform_init()
and switch only those modules which are really added to eCos image.

Best regards,
jerzy

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
@ 2011-05-04  7:41 ` bugzilla-daemon
  2011-05-12  9:18 ` bugzilla-daemon
                   ` (66 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-05-04  7:41 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Ilija Kocho <ilijak@siva.com.mk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |ilijak@siva.com.mk

--- Comment #1 from Ilija Kocho <ilijak@siva.com.mk> 2011-05-04 08:41:11 BST ---
(In reply to comment #0)
> Created an attachment (id=1218)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1218) [details]
> Ethernet driver for STM32 connectivity line with port on MMstm32f107 board
> 

Hi Jerzy

Thanks for contribution. Just a hint, i found it's better to submit diff in
plain text format (not compressed) since Bugzilla will recognize it as a patch
(or you can additionally mark it "patch"). Bugzilla can present diff in user
friendly side-by side format.

Ilija

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
  2011-05-04  7:41 ` [Bug 1001219] " bugzilla-daemon
@ 2011-05-12  9:18 ` bugzilla-daemon
  2011-10-08 14:02 ` bugzilla-daemon
                   ` (65 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-05-12  9:18 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1218|0                           |1
        is obsolete|                            |

--- Comment #2 from Jerzy Dyrda <jerzdy@gmail.com> 2011-05-12 10:17:56 BST ---
Created an attachment (id=1230)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1230)
Ethernet driver for STM32 connectivity line with port on MMstm32f107 board -
not gzipped

Plain text - not compressed.

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
  2011-05-04  7:41 ` [Bug 1001219] " bugzilla-daemon
  2011-05-12  9:18 ` bugzilla-daemon
@ 2011-10-08 14:02 ` bugzilla-daemon
  2011-10-08 14:27 ` bugzilla-daemon
                   ` (64 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-08 14:02 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #3 from Ilija Kocho <ilijak@siva.com.mk> 2011-10-08 15:02:21 BST ---
Hi Jerzy

Thank you for your contribution. Here are my first remarks:

1. Some unresolved conflicts remain upon Propox board selection in Configtool
and it may mislead the user that something is broken. It seems that Configtool
can't resolve the CYGHWR_HAL_CORTEXM_STM32_CLOCK_PLL_SOURCE on it's own so
let's make it little-bit easier:
  -  Is it possible to re-arrange legal_values expression in order to make it
easier for Configtool?
  - Or it is better to calculate default_value conditionally
(CYGINT_HAL_CORTEXM_STM32_CL==0 ?)

2. CYGPKG_IO_ETH_DRIVERS is normally not included in the target. Not everybody
would use Ethernet. Please remove it.

3. FYI, the PHY driver DP8348 has been resolved in meantime by Bug 1001235.
Please synchronize your code with it.

Now some general hints: It usually takes a number of iterations before code is
fit for commit and it would be easier for me (hopefully also for you) if you
break the patch in several diffs (divide and conquer). Ideally - a diff for
every affected package, in this case: /eth driver/, /variant/, /platform/ and
eventually /phy driver/. As exception ecos.db is better not sent as diff,
instead put your ecos.db entries in a plain file and name it ecos_db.txt .

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (2 preceding siblings ...)
  2011-10-08 14:02 ` bugzilla-daemon
@ 2011-10-08 14:27 ` bugzilla-daemon
  2011-10-10 20:51 ` bugzilla-daemon
                   ` (63 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-08 14:27 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #4 from Ilija Kocho <ilijak@siva.com.mk> 2011-10-08 15:27:25 BST ---
(In reply to comment #3)
> Hi Jerzy
> 
> Thank you for your contribution. Here are my first remarks:
> 
> 1. Some unresolved conflicts remain upon Propox board selection in Configtool
> and it may mislead the user that something is broken. It seems that Configtool
> can't resolve the CYGHWR_HAL_CORTEXM_STM32_CLOCK_PLL_SOURCE on it's own so
> let's make it little-bit easier:
>   -  Is it possible to re-arrange legal_values expression in order to make it
> easier for Configtool?
>   - Or it is better to calculate default_value conditionally
> (CYGINT_HAL_CORTEXM_STM32_CL==0 ?)
> 
> 2. CYGPKG_IO_ETH_DRIVERS is normally not included in the target. Not everybody
> would use Ethernet. Please remove it.
> 
> 3. FYI, the PHY driver DP8348 has been resolved in meantime by Bug 1001235.
> Please synchronize your code with it.

4. Just one more thing: Won't it be better if Propox on-board chip member
appears (CYGHWR_HAL_CORTEXM_STM32) once this board is selected? 

> 
> Now some general hints: It usually takes a number of iterations before code is
> fit for commit and it would be easier for me (hopefully also for you) if you
> break the patch in several diffs (divide and conquer). Ideally - a diff for
> every affected package, in this case: /eth driver/, /variant/, /platform/ and
> eventually /phy driver/. As exception ecos.db is better not sent as diff,
> instead put your ecos.db entries in a plain file and name it ecos_db.txt .

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (3 preceding siblings ...)
  2011-10-08 14:27 ` bugzilla-daemon
@ 2011-10-10 20:51 ` bugzilla-daemon
  2011-10-10 20:54 ` bugzilla-daemon
                   ` (62 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-10 20:51 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #5 from Jerzy Dyrda <jerzdy@gmail.com> 2011-10-10 21:50:42 BST ---
Created an attachment (id=1394)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1394)
Ethernet driver for STM32 CL

Ethernet driver for STM32 Connectivity Line

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (4 preceding siblings ...)
  2011-10-10 20:51 ` bugzilla-daemon
@ 2011-10-10 20:54 ` bugzilla-daemon
  2011-10-10 20:56 ` bugzilla-daemon
                   ` (61 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-10 20:54 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1230|0                           |1
        is obsolete|                            |

--- Comment #6 from Jerzy Dyrda <jerzdy@gmail.com> 2011-10-10 21:54:08 BST ---
Created an attachment (id=1395)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1395)
Platform for Propox MMstm32f107 module

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (5 preceding siblings ...)
  2011-10-10 20:54 ` bugzilla-daemon
@ 2011-10-10 20:56 ` bugzilla-daemon
  2011-10-10 21:13 ` bugzilla-daemon
                   ` (60 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-10 20:56 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #7 from Jerzy Dyrda <jerzdy@gmail.com> 2011-10-10 21:55:57 BST ---
Created an attachment (id=1396)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1396)
Variant modification needed by STM32 CL

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (6 preceding siblings ...)
  2011-10-10 20:56 ` bugzilla-daemon
@ 2011-10-10 21:13 ` bugzilla-daemon
  2011-10-10 21:50 ` bugzilla-daemon
                   ` (59 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-10 21:13 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #8 from Jerzy Dyrda <jerzdy@gmail.com> 2011-10-10 22:12:55 BST ---
Created an attachment (id=1397)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1397)
ecos.db for Ethernet driver for STM32 connectivity line with port on
MMstm32f107 board

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (7 preceding siblings ...)
  2011-10-10 21:13 ` bugzilla-daemon
@ 2011-10-10 21:50 ` bugzilla-daemon
  2011-10-11 19:53 ` bugzilla-daemon
                   ` (58 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-10 21:50 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #9 from Jerzy Dyrda <jerzdy@gmail.com> 2011-10-10 22:49:41 BST ---
Hello Ilija,

(In reply to comment #3)

> 1. Some unresolved conflicts remain upon Propox board selection in Configtool
> and it may mislead the user that something is broken. It seems that Configtool
> can't resolve the CYGHWR_HAL_CORTEXM_STM32_CLOCK_PLL_SOURCE on it's own so
> let's make it little-bit easier:
>   -  Is it possible to re-arrange legal_values expression in order to make it
> easier for Configtool?
>   - Or it is better to calculate default_value conditionally
> (CYGINT_HAL_CORTEXM_STM32_CL==0 ?)
Done according to second hint.

> 2. CYGPKG_IO_ETH_DRIVERS is normally not included in the target. Not everybody
> would use Ethernet. Please remove it.
OK.

> 3. FYI, the PHY driver DP8348 has been resolved in meantime by Bug 1001235.
> Please synchronize your code with it.
Done.

> Now some general hints: It usually takes a number of iterations before code is
> fit for commit and it would be easier for me (hopefully also for you) if you
> break the patch in several diffs (divide and conquer). Ideally - a diff for
> every affected package, in this case: /eth driver/, /variant/, /platform/ and
> eventually /phy driver/. As exception ecos.db is better not sent as diff,
> instead put your ecos.db entries in a plain file and name it ecos_db.txt .
OK. I split patch into three parts but please consider that initially I would
like to contribute only eth driver. However STM32 Connectivity Line introduces
so many modification that even platform package is required. Summarizing eth
driver can't be introduced without rest of stuff.

Best regards,
jerzy

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (8 preceding siblings ...)
  2011-10-10 21:50 ` bugzilla-daemon
@ 2011-10-11 19:53 ` bugzilla-daemon
  2011-10-11 20:39 ` bugzilla-daemon
                   ` (57 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-11 19:53 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #10 from Ilija Kocho <ilijak@siva.com.mk> 2011-10-11 20:53:18 BST ---
(In reply to comment #8)
> Created an attachment (id=1397)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1397) [details]
> ecos.db for Ethernet driver for STM32 connectivity line with port on
> MMstm32f107 board

Some lines are missing spaces and configtool can't read the repository.

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (9 preceding siblings ...)
  2011-10-11 19:53 ` bugzilla-daemon
@ 2011-10-11 20:39 ` bugzilla-daemon
  2011-10-11 21:08 ` bugzilla-daemon
                   ` (56 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-11 20:39 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #11 from Ilija Kocho <ilijak@siva.com.mk> 2011-10-11 21:39:04 BST ---
(In reply to comment #7)
> Created an attachment (id=1396)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1396) [details]
> Variant modification needed by STM32 CL

http://bugs.ecos.sourceware.org/attachment.cgi?id=1396&action=diff#ecos_clear/packages/hal/cortexm/stm32/var/current/include/var_io.h_sec7

Ethernet definitions are usually stored with Ethernet driver, either with C
source or in a header file (I prefer the second). If you move it there than
that would imply that names do not contain "_HAL_".


http://bugs.ecos.sourceware.org/attachment.cgi?id=1396&action=diff#ecos_clear/packages/hal/cortexm/stm32/var/current/src/hal_diag.c_sec1

What is the reason for renaming of hal_stm32_serial_init() into
hal_stm32_serial_init_channel() 


http://bugs.ecos.sourceware.org/attachment.cgi?id=1396&action=diff#ecos_clear/packages/hal/cortexm/stm32/var/current/src/stm32_misc.c_sec2

Changes like this may look attractive but (as well as previous one) may (will)
break other user's applications. Conditional compilation may help but in this
case I suggest to keep old code, it's simple and provides working conditions to
all peripherals.

http://bugs.ecos.sourceware.org/attachment.cgi?id=1396&action=diff#ecos_clear/packages/hal/cortexm/stm32/var/current/src/stm32_misc.c_sec3

Is this necessary?

http://bugs.ecos.sourceware.org/attachment.cgi?id=1396&action=diff#ecos_clear/packages/hal/cortexm/stm32/var/current/src/stm32_misc.c_sec4

I'm referring to following:
-    hal_stm32_sysclk /= 2;
+    hal_stm32_sysclk = CYGARC_HAL_CORTEXM_STM32_INTERNAL_CLOCK / 2;

If it works don't fix it :)

Now some minor remarks:
Comments like 
#endif // CYGINT_HAL_CORTEXM_STM32_CL>0
on some places are not consistent with their respective #ifs.
And: (this is my personal only)
#if CYGINT_HAL_CORTEXM_STM32_CL != 0
or
#if CYGINT_HAL_CORTEXM_STM32_CL
is more common than >0.

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (10 preceding siblings ...)
  2011-10-11 20:39 ` bugzilla-daemon
@ 2011-10-11 21:08 ` bugzilla-daemon
  2011-10-11 21:45 ` bugzilla-daemon
                   ` (55 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-11 21:08 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #12 from Ilija Kocho <ilijak@siva.com.mk> 2011-10-11 22:08:06 BST ---
(In reply to comment #6)
> Created an attachment (id=1395)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1395) [details]
> Platform for Propox MMstm32f107 module

First to remind you to my remark in comment #4

At present the user will see F103ZE which is unlikely to be the target device.
Platform may require the "STM variany in use" (CYGHWR_HAL_CORTEXM_STM32) to be
a specific device, or a set of eligible devices. If you opt for second case
just make sure that configtool resolves with the one found on Propox board.

Why JTAG startup? AFAIK Connectivity line desn't feature FSMC (or am I out of
date?). I guess SRAM startup should work (but test it first).

Regarding ROM startup there's one issue: RAM region is required in order to
build RedBoot. True, there's not much use of RedBoot on device with 64KiB RAM
but somebody would like to try. Solution is simple rename SRAM region (nor
section) into RAM.

Last but not least, please spend some time on Copyright banners (all files that
need it). They should for instance show correct year, author/contributor, etc.

Ilija

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (11 preceding siblings ...)
  2011-10-11 21:08 ` bugzilla-daemon
@ 2011-10-11 21:45 ` bugzilla-daemon
  2011-10-12 11:25 ` bugzilla-daemon
                   ` (54 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-11 21:45 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #13 from Jerzy Dyrda <jerzdy@gmail.com> 2011-10-11 22:44:36 BST ---

(In reply to comment #11)
> (In reply to comment #7)
> > Created an attachment (id=1396)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1396) [details]
[details]
> > Variant modification needed by STM32 CL
> 
> http://bugs.ecos.sourceware.org/attachment.cgi?id=1396&action=diff#ecos_clear/packages/hal/cortexm/stm32/var/current/include/var_io.h_sec7
> 
> Ethernet definitions are usually stored with Ethernet driver, either with C
> source or in a header file (I prefer the second). If you move it there than
> that would imply that names do not contain "_HAL_".
I would like to follow convention e.g. ADC registers with pin description are
stored in this header. I thought that any new peripheral description will be
stored in var_io.h

> 
> http://bugs.ecos.sourceware.org/attachment.cgi?id=1396&action=diff#ecos_clear/packages/hal/cortexm/stm32/var/current/src/hal_diag.c_sec1
> 
> What is the reason for renaming of hal_stm32_serial_init() into
> hal_stm32_serial_init_channel() 
I'm not sure now :) but as I suppose it doesn't work.
BTW. 
Definition of function is static void hal_stm32_serial_init(void) but function
is called with argument
&stm32_ser_channels[CYGNUM_HAL_VIRTUAL_VECTOR_CONSOLE_CHANNEL]
IMHO in case of non virtual vector is initialized other serial port than I
expect 

> 
> http://bugs.ecos.sourceware.org/attachment.cgi?id=1396&action=diff#ecos_clear/packages/hal/cortexm/stm32/var/current/src/stm32_misc.c_sec2
> 
> Changes like this may look attractive but (as well as previous one) may (will)
> break other user's applications. Conditional compilation may help but in this
> case I suggest to keep old code, it's simple and provides working conditions 
> to all peripherals.
I agree I have just added it because in case of non virtual vector isn't such
function.

> http://bugs.ecos.sourceware.org/attachment.cgi?id=1396&action=diff#ecos_clear/packages/hal/cortexm/stm32/var/current/src/stm32_misc.c_sec3
> 
> Is this necessary?
Yes RCC (clock generation) module for CL has another PLL stage which is
required to achieve proper clock for eth module.    

BTW
I have further bad news : in STM32F2xx family RCC module is different than
STM32f105/7 family.

> 
> http://bugs.ecos.sourceware.org/attachment.cgi?id=1396&action=diff#ecos_clear/packages/hal/cortexm/stm32/var/current/src/stm32_misc.c_sec4
> 
> I'm referring to following:
> -    hal_stm32_sysclk /= 2;
> +    hal_stm32_sysclk = CYGARC_HAL_CORTEXM_STM32_INTERNAL_CLOCK / 2;
> 
> If it works don't fix it :)
It was made assumption that internal clock is the same as input clock but it's
wrong for CL controller.

> Now some minor remarks:
> Comments like 
> #endif // CYGINT_HAL_CORTEXM_STM32_CL>0
> on some places are not consistent with their respective #ifs.
> And: (this is my personal only)
> #if CYGINT_HAL_CORTEXM_STM32_CL != 0
> or
> #if CYGINT_HAL_CORTEXM_STM32_CL
> is more common than >0.
OK I check it.

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (12 preceding siblings ...)
  2011-10-11 21:45 ` bugzilla-daemon
@ 2011-10-12 11:25 ` bugzilla-daemon
  2011-10-12 20:34 ` bugzilla-daemon
                   ` (53 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-12 11:25 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #14 from Jerzy Dyrda <jerzdy@gmail.com> 2011-10-12 12:24:35 BST ---
Hello Ilija,

(In reply to comment #12)
> (In reply to comment #6)
> > Created an attachment (id=1395)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1395) [details]
[details]
> > Platform for Propox MMstm32f107 module
> 
> First to remind you to my remark in comment #4
> 
> At present the user will see F103ZE which is unlikely to be the target device.
Please consider that such situation will always happen if further target will
be introduced.

> Platform may require the "STM variany in use" (CYGHWR_HAL_CORTEXM_STM32) to be
> a specific device, or a set of eligible devices. If you opt for second case
> just make sure that configtool resolves with the one found on Propox board.
IMHO Either platform should provide proper STM32 variant or this option should
be remove - nobody is interested in this option.

Please review this option considering targets existing STM32F100/1/2/3 further
like my STM32F105/7 and possible STM32F2xx, STM32F4xx and give me "order" how I
should introduce STM32F105/7 family.


> Why JTAG startup? AFAIK Connectivity line desn't feature FSMC (or am I out of
> date?). I guess SRAM startup should work (but test it first).
You are absolute right. JTAG startup doesn't have sens and I remove it.

> Regarding ROM startup there's one issue: RAM region is required in order to
> build RedBoot. True, there's not much use of RedBoot on device with 64KiB RAM
> but somebody would like to try. Solution is simple rename SRAM region (nor
> section) into RAM.
I follow other cortex-m target like lm3s where is:
- only one startup ROM
- only SRAM section

BTW. I've built RedBoot successful with only SRAM section. 

If it's possible I would like to shrink amount of startups only to ROM.

> 
> Last but not least, please spend some time on Copyright banners 
> (all files that need it). 
> They should for instance show correct year, author/contributor, etc.
Of course I correct it.

Best regards,
jerzy

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (13 preceding siblings ...)
  2011-10-12 11:25 ` bugzilla-daemon
@ 2011-10-12 20:34 ` bugzilla-daemon
  2011-10-12 21:15 ` bugzilla-daemon
                   ` (52 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-12 20:34 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #15 from Ilija Kocho <ilijak@siva.com.mk> 2011-10-12 21:33:58 BST ---
(In reply to comment #13)
> (In reply to comment #11)
> > (In reply to comment #7)
> > > Created an attachment (id=1396)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1396) [details]
[details] [details]
> > > Variant modification needed by STM32 CL
> > 
> > http://bugs.ecos.sourceware.org/attachment.cgi?id=1396&action=diff#ecos_clear/packages/hal/cortexm/stm32/var/current/include/var_io.h_sec7
> > 
> > Ethernet definitions are usually stored with Ethernet driver, either with C
> > source or in a header file (I prefer the second). If you move it there than
> > that would imply that names do not contain "_HAL_".
> I would like to follow convention e.g. ADC registers with pin description are
> stored in this header. I thought that any new peripheral description will be
> stored in var_io.h
> 

Obviously there's more than one convention. However I think that driver is
better place for Ethernet #defines than HAL. I can imagine somebody using ADC
without driver so it would be convenient having defines handy in HAL. On the
other hand using Ethernet without driver is unlikely. Also, if for some reason,
in future new variant is added (this is not a proposal merely possibility) it
would reuse the header.


> > 
> > http://bugs.ecos.sourceware.org/attachment.cgi?id=1396&action=diff#ecos_clear/packages/hal/cortexm/stm32/var/current/src/hal_diag.c_sec1
> > 
> > What is the reason for renaming of hal_stm32_serial_init() into
> > hal_stm32_serial_init_channel() 
> I'm not sure now :) but as I suppose it doesn't work.
> BTW. 
> Definition of function is static void hal_stm32_serial_init(void) but function
> is called with argument
> &stm32_ser_channels[CYGNUM_HAL_VIRTUAL_VECTOR_CONSOLE_CHANNEL]
> IMHO in case of non virtual vector is initialized other serial port than I
> expect 

I am referring (and link points) to changes in hal_stm32_diag_init() 
See discussion on issue below.

> 
> > 
> > http://bugs.ecos.sourceware.org/attachment.cgi?id=1396&action=diff#ecos_clear/packages/hal/cortexm/stm32/var/current/src/stm32_misc.c_sec2
> > 
> > Changes like this may look attractive but (as well as previous one) may (will)
> > break other user's applications. Conditional compilation may help but in this
> > case I suggest to keep old code, it's simple and provides working conditions 
> > to all peripherals.
> I agree I have just added it because in case of non virtual vector isn't such
> function.

I am referring (and link points) to changes in hal_variant_init(). Once these
functions were published they are being (as are) used in applications not known
to me or you. Such changes may make many users unhappy and are unacceptable.


> 
> > http://bugs.ecos.sourceware.org/attachment.cgi?id=1396&action=diff#ecos_clear/packages/hal/cortexm/stm32/var/current/src/stm32_misc.c_sec3
> > 
> > Is this necessary?
> Yes RCC (clock generation) module for CL has another PLL stage which is
> required to achieve proper clock for eth module.    
> 

My question is referring to cyg_uint32 hal_arch_default_isr(). Why do you need
it?

> BTW
> I have further bad news : in STM32F2xx family RCC module is different than
> STM32f105/7 family.
> 

Regarding this (and in general), take in consideration future extensions when
you design system software.


> > 
> > http://bugs.ecos.sourceware.org/attachment.cgi?id=1396&action=diff#ecos_clear/packages/hal/cortexm/stm32/var/current/src/stm32_misc.c_sec4
> > 
> > I'm referring to following:
> > -    hal_stm32_sysclk /= 2;
> > +    hal_stm32_sysclk = CYGARC_HAL_CORTEXM_STM32_INTERNAL_CLOCK / 2;
> > 
> > If it works don't fix it :)
> It was made assumption that internal clock is the same as input clock but it's
> wrong for CL controller.

OK so this is needed (sorry :).
Then you can add it, but again, take care not to break backward compatibility.
Preprocessor may help you.

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (14 preceding siblings ...)
  2011-10-12 20:34 ` bugzilla-daemon
@ 2011-10-12 21:15 ` bugzilla-daemon
  2011-10-16 14:12 ` bugzilla-daemon
                   ` (51 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-12 21:15 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #16 from Ilija Kocho <ilijak@siva.com.mk> 2011-10-12 22:15:09 BST ---
(In reply to comment #14)
> Hello Ilija,
> 
> (In reply to comment #12)
> > (In reply to comment #6)
> > > Created an attachment (id=1395)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1395) [details]
[details] [details]
> > > Platform for Propox MMstm32f107 module
> > 
> > First to remind you to my remark in comment #4
> > 
> > At present the user will see F103ZE which is unlikely to be the target device.
> Please consider that such situation will always happen if further target will
> be introduced.
> 
> > Platform may require the "STM variany in use" (CYGHWR_HAL_CORTEXM_STM32) to be
> > a specific device, or a set of eligible devices. If you opt for second case
> > just make sure that configtool resolves with the one found on Propox board.
> IMHO Either platform should provide proper STM32 variant or this option should
> be remove - nobody is interested in this option.
> 
> Please review this option considering targets existing STM32F100/1/2/3 further
> like my STM32F105/7 and possible STM32F2xx, STM32F4xx and give me "order" how I
> should introduce STM32F105/7 family.
> 

I am not in a position to "order" you but I can give you some hints. Here is a
hal_cortexm_stm32_mmstm32f107.cdl diff snippet.

     implements  CYGINT_DEVS_ETH_CORTEXM_STM32_REMAP_PINS
     implements  CYGINT_DEVS_SPI_CORTEXM_STM32_BUS3_REMAP_PINS
+
+    requires {
+        (CYGHWR_HAL_CORTEXM_STM32 == "F107VB") ||
+        (CYGHWR_HAL_CORTEXM_STM32 == "F107VC") ||
+        (CYGHWR_HAL_CORTEXM_STM32 == "F105VB") ||
+        (CYGHWR_HAL_CORTEXM_STM32 == "F105VC")
+    }
+
     requires      { is_active(CYGPKG_DEVS_ETH_PHY) implies
                     (1 == CYGHWR_DEVS_ETH_PHY_DP8384X) }

You can at your option put a single or multiple devices in "requires".
Configtool will resolve (conflict) with first one. Other listed devices shall
be "allowed" and rest of devices (off the list) shall raise conflict.

Note: If you opt for multiple devices they should be ones that the controller
on Propox board can emulate.

Regarding F2 devices let's leave them for the time being. It will be another
platform anyway.

> 
> > Why JTAG startup? AFAIK Connectivity line desn't feature FSMC (or am I out of
> > date?). I guess SRAM startup should work (but test it first).
> You are absolute right. JTAG startup doesn't have sens and I remove it.
> 
> > Regarding ROM startup there's one issue: RAM region is required in order to
> > build RedBoot. True, there's not much use of RedBoot on device with 64KiB RAM
> > but somebody would like to try. Solution is simple rename SRAM region (nor
> > section) into RAM.
> I follow other cortex-m target like lm3s where is:
> - only one startup ROM
> - only SRAM section

I am refering to SRAM memory region, not .sram section.

> 
> BTW. I've built RedBoot successful with only SRAM section. 

You probably do it without GDB support. Then I was unclear, sorry, I should
have said "RAM is required by GDB support".

When I reconsider, this is a small memory device, perhaps we could live without
it.

> 
> If it's possible I would like to shrink amount of startups only to ROM.

ROM startup is enough.


P.S.
As a general advice, please take care about backward compatibility - it is very
important. If able, please do regressions on "STM3210E EVAL board" (or
equivalent). If you don't have hardware at least check the building process.

Ilija

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (16 preceding siblings ...)
  2011-10-16 14:12 ` bugzilla-daemon
@ 2011-10-16 14:12 ` bugzilla-daemon
  2011-10-22 12:22 ` bugzilla-daemon
                   ` (49 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-16 14:12 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Ilija Kocho <ilijak@siva.com.mk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         OS/Version|ARM                         |Cortex-M

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (15 preceding siblings ...)
  2011-10-12 21:15 ` bugzilla-daemon
@ 2011-10-16 14:12 ` bugzilla-daemon
  2011-10-16 14:12 ` bugzilla-daemon
                   ` (50 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-16 14:12 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #17 from Ilija Kocho <ilijak@siva.com.mk> 2011-10-16 15:11:36 BST ---
(In reply to comment #13)
> (In reply to comment #11)
> > (In reply to comment #7)
> > And: (this is my personal only)
> > #if CYGINT_HAL_CORTEXM_STM32_CL != 0
> > or
> > #if CYGINT_HAL_CORTEXM_STM32_CL
> > is more common than >0.
> OK I check it.

Even better would be to change cdl_interface CYGINT_HAL_CORTEXM_STM32_CL flavor
to bool and then use

#ifdef CYGINT_HAL_CORTEXM_STM32_CL
#ifndef CYGINT_HAL_CORTEXM_STM32_CL

The advantage is that for non-Connectivity-Line devices
CYGINT_HAL_CORTEXM_STM32_CL shall not exist.

Ilija

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (17 preceding siblings ...)
  2011-10-16 14:12 ` bugzilla-daemon
@ 2011-10-22 12:22 ` bugzilla-daemon
  2011-10-23 21:27 ` bugzilla-daemon
                   ` (48 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-22 12:22 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #19 from Ilija Kocho <ilijak@siva.com.mk> 2011-10-22 13:21:29 BST ---
(In reply to comment #5)
> Created an attachment (id=1394)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1394) [details]
> Ethernet driver for STM32 CL
> 
> Ethernet driver for STM32 Connectivity Line

Some comments on Ethernet driver:

MII and RMII interfaces are mutually exclusive and CDL should reflect this. A
cdl_option with legal values "RMII" "MII" would provide it and in addition make
it visible in configtool.


"Remap pins" could be also visible if it were cdl_option.


Regarding pins, some addition to my statement in Comment #11. Since pins are
being provided by HAL, they should be defined in HAL (unlike other Ethernet
definitions such as registers, etc.). Preferable place is plf_io.h rather than
var_io.h. because other chips (present or future) may have different pin
mapping.
On the other hand, the pin functions - once assigned to Ethernet (although pins
are provided provided by HAL) belong to Ethernet so their naming should reflect
this Here is a plf_io.h snuppet:

plf_io.h snippet --------------------------------

#define CYGHWR_IO_ETH_STM32MAC_MII_COL \
        CYGHWR_HAL_STM32_GPIO(A, 3,  IN , FLOATING)
...
-------------------------------------------------

Also if_stm32.c
Could CYGHWR_HAL_STM32_GPIO_SET(CYGHWR_...); lines be replaced by a loop?
And in order to avoid specifying HAL specific macros in a device driver, a new
macro can be defined CYGHWR_IO_ETH_STM32MAC_PIN(...).

Note: In macro names above I arbitrarily put "STM32MAC" segment. Probably there
is a more appropriate name for this Ethernet controller.

CYGNUM_DEVS_ETH_CORTEXM_STM32_RX_BUFS: Is there a range of legal values?
BTW other Ethernet devices that I have seen also provide configuration option
for TX_BUFFS.Is this fixed on STM32 Ethernet controller?


TCP/IP Checksum generation and check.
FYI lwIP is aware of such hardware features
http://sourceware.org/ml/ecos-discuss/2011-07/msg00017.html
and it would be good if they are implemented.

Ilija

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (18 preceding siblings ...)
  2011-10-22 12:22 ` bugzilla-daemon
@ 2011-10-23 21:27 ` bugzilla-daemon
  2011-10-24 15:55 ` bugzilla-daemon
                   ` (47 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-23 21:27 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         OS/Version|Cortex-M                    |ARM

--- Comment #20 from Jerzy Dyrda <jerzdy@gmail.com> 2011-10-23 22:26:20 BST ---
Hello Ilja,

(In reply to comment #19)
> (In reply to comment #5)
> > Created an attachment (id=1394)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1394) [details]
[details]
> > Ethernet driver for STM32 CL
> > 
> > Ethernet driver for STM32 Connectivity Line
> 
> Some comments on Ethernet driver:
> 
> MII and RMII interfaces are mutually exclusive and CDL should reflect this. A
> cdl_option with legal values "RMII" "MII" would provide it and in addition 
> make it visible in configtool.
> 
> "Remap pins" could be also visible if it were cdl_option.
OK.  


> Regarding pins, some addition to my statement in Comment #11. Since pins are
> being provided by HAL, they should be defined in HAL (unlike other Ethernet
> definitions such as registers, etc.). Preferable place is plf_io.h rather than
> var_io.h. because other chips (present or future) may have different pin
> mapping.
Yes you are right. Despite of promised pin compatibility by ST pins assignment
between F105/7 CL and F2xx family differs.

> On the other hand, the pin functions - once assigned to Ethernet 
> (although pins are provided provided by HAL) belong to Ethernet 
> so their naming should reflect this Here is a plf_io.h snuppet:
> 
> plf_io.h snippet --------------------------------
> 
> #define CYGHWR_IO_ETH_STM32MAC_MII_COL \
>         CYGHWR_HAL_STM32_GPIO(A, 3,  IN , FLOATING)
> ...
> -------------------------------------------------
OK.


> Also if_stm32.c
> Could CYGHWR_HAL_STM32_GPIO_SET(CYGHWR_...); lines be replaced by a loop?
Could you explain me what you mean?

> And in order to avoid specifying HAL specific macros in a device driver, a new
> macro can be defined CYGHWR_IO_ETH_STM32MAC_PIN(...).
> 
> Note: In macro names above I arbitrarily put "STM32MAC" segment. 
> Probably there is a more appropriate name for this Ethernet controller.
Maybe just CYGHWR_IO_ETH_STM32_MII_(pin_name)? 

> CYGNUM_DEVS_ETH_CORTEXM_STM32_RX_BUFS: Is there a range of legal values?
Amount of available RAM ;)

> BTW other Ethernet devices that I have seen also provide configuration option
> for TX_BUFFS.Is this fixed on STM32 Ethernet controller?
Driver based on size of SG list : 
#define TDES_NUM (CYGNUM_IO_ETH_DRIVERS_SG_LIST_SIZE >> 1)

Instead of copying data buffer driver uses it - just attach buffer to TX
descriptor list.

> 
> TCP/IP Checksum generation and check.
> FYI lwIP is aware of such hardware features
> http://sourceware.org/ml/ecos-discuss/2011-07/msg00017.html
> and it would be good if they are implemented.
I know. Driver provides such functionality :
CYGNUM_DEVS_ETH_CORTEXM_STM32_TX_CHECKSUM_GEN
CYGNUM_DEVS_ETH_CORTEXM_STM32_RX_CHECKSUM_VER
but without splitting it in IP checksum and UDP/TCP/ICMP checksum (MAC can't
calculates checksum only for UDP or TCP).

Best regards,
jerzy

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (19 preceding siblings ...)
  2011-10-23 21:27 ` bugzilla-daemon
@ 2011-10-24 15:55 ` bugzilla-daemon
  2011-12-04 21:22 ` bugzilla-daemon
                   ` (46 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-10-24 15:55 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #21 from Ilija Kocho <ilijak@siva.com.mk> 2011-10-24 16:54:17 BST ---
(In reply to comment #20)

> 
> > Regarding pins, some addition to my statement in Comment #11. Since pins are
> > being provided by HAL, they should be defined in HAL (unlike other Ethernet
> > definitions such as registers, etc.). Preferable place is plf_io.h rather than
> > var_io.h. because other chips (present or future) may have different pin
> > mapping.
> Yes you are right. Despite of promised pin compatibility by ST pins assignment
> between F105/7 CL and F2xx family differs.
> 

Pining, in general is something that can change when new chips introduced.

> > On the other hand, the pin functions - once assigned to Ethernet 
> > (although pins are provided provided by HAL) belong to Ethernet 
> > so their naming should reflect this Here is a plf_io.h snuppet:
> > 
> > plf_io.h snippet --------------------------------
> > 
> > #define CYGHWR_IO_ETH_STM32MAC_MII_COL \
> >         CYGHWR_HAL_STM32_GPIO(A, 3,  IN , FLOATING)
> > ...
> > -------------------------------------------------
> OK.
> 
> 
> > Also if_stm32.c
> > Could CYGHWR_HAL_STM32_GPIO_SET(CYGHWR_...); lines be replaced by a loop?
> Could you explain me what you mean?

The idea is that, since there are many pins, a loop might slightly reduce the 
size, although I might be wrong.

> 
> > And in order to avoid specifying HAL specific macros in a device driver, a new
> > macro can be defined CYGHWR_IO_ETH_STM32MAC_PIN(...).
> > 
> > Note: In macro names above I arbitrarily put "STM32MAC" segment. 
> > Probably there is a more appropriate name for this Ethernet controller.
> Maybe just CYGHWR_IO_ETH_STM32_MII_(pin_name)? 

I would like to avoid references to architecture. If this Ethernet controller
has some name it would be suitable.

> 
> > CYGNUM_DEVS_ETH_CORTEXM_STM32_RX_BUFS: Is there a range of legal values?
> Amount of available RAM ;)

You may introduce some reasonable boundaries anyway. Also, usually there is a
minimal number of buffers (1 or 2) necessary for the driver to work.

> 
> > BTW other Ethernet devices that I have seen also provide configuration option
> > for TX_BUFFS.Is this fixed on STM32 Ethernet controller?
> Driver based on size of SG list : 
> #define TDES_NUM (CYGNUM_IO_ETH_DRIVERS_SG_LIST_SIZE >> 1)
> 
> Instead of copying data buffer driver uses it - just attach buffer to TX
> descriptor list.

Than, is there possibility for configuring the number of TX buffer descriptors?

> 
> > 
> > TCP/IP Checksum generation and check.
> > FYI lwIP is aware of such hardware features
> > http://sourceware.org/ml/ecos-discuss/2011-07/msg00017.html
> > and it would be good if they are implemented.
> I know. Driver provides such functionality :
> CYGNUM_DEVS_ETH_CORTEXM_STM32_TX_CHECKSUM_GEN
> CYGNUM_DEVS_ETH_CORTEXM_STM32_RX_CHECKSUM_VER
> but without splitting it in IP checksum and UDP/TCP/ICMP checksum (MAC can't
> calculates checksum only for UDP or TCP).

IMO it should be used if possible. The gain is considerable according to:
http://sourceware.org/ml/ecos-discuss/2011-06/msg00029.html

You can put some CDL that enables checksum by HW if lwIP is included and then
HW checksum should be the default.

Ilija

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (20 preceding siblings ...)
  2011-10-24 15:55 ` bugzilla-daemon
@ 2011-12-04 21:22 ` bugzilla-daemon
  2011-12-04 21:23 ` bugzilla-daemon
                   ` (45 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-12-04 21:22 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1394|0                           |1
        is obsolete|                            |

--- Comment #22 from Jerzy Dyrda <jerzdy@gmail.com> 2011-12-04 21:22:14 GMT ---
Created an attachment (id=1446)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1446)
Ethernet driver for STM32 CL

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (21 preceding siblings ...)
  2011-12-04 21:22 ` bugzilla-daemon
@ 2011-12-04 21:23 ` bugzilla-daemon
  2011-12-04 21:24 ` bugzilla-daemon
                   ` (44 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-12-04 21:23 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #23 from Jerzy Dyrda <jerzdy@gmail.com> 2011-12-04 21:23:27 GMT ---
Created an attachment (id=1447)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1447)
SPI pins remap for bus 1 and 3

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (22 preceding siblings ...)
  2011-12-04 21:23 ` bugzilla-daemon
@ 2011-12-04 21:24 ` bugzilla-daemon
  2011-12-04 21:25 ` bugzilla-daemon
                   ` (43 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-12-04 21:24 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1395|0                           |1
        is obsolete|                            |

--- Comment #24 from Jerzy Dyrda <jerzdy@gmail.com> 2011-12-04 21:24:10 GMT ---
Created an attachment (id=1448)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1448)
Platform for Propox MMstm32f107 module

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (23 preceding siblings ...)
  2011-12-04 21:24 ` bugzilla-daemon
@ 2011-12-04 21:25 ` bugzilla-daemon
  2011-12-04 21:35 ` bugzilla-daemon
                   ` (42 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-12-04 21:25 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1396|0                           |1
        is obsolete|                            |

--- Comment #25 from Jerzy Dyrda <jerzdy@gmail.com> 2011-12-04 21:24:55 GMT ---
Created an attachment (id=1449)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1449)
Variant modification needed by STM32 CL

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (24 preceding siblings ...)
  2011-12-04 21:25 ` bugzilla-daemon
@ 2011-12-04 21:35 ` bugzilla-daemon
  2011-12-21  9:33 ` bugzilla-daemon
                   ` (41 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-12-04 21:35 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1397|0                           |1
        is obsolete|                            |

--- Comment #26 from Jerzy Dyrda <jerzdy@gmail.com> 2011-12-04 21:35:29 GMT ---
Created an attachment (id=1450)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1450)
ecos.db for Ethernet driver for STM32 connectivity line with port on
MMstm32f107 board

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (25 preceding siblings ...)
  2011-12-04 21:35 ` bugzilla-daemon
@ 2011-12-21  9:33 ` bugzilla-daemon
  2011-12-26 21:47 ` bugzilla-daemon
                   ` (40 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-12-21  9:33 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #27 from Ilija Kocho <ilijak@siva.com.mk> 2011-12-21 09:33:14 GMT ---
Created an attachment (id=1475)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1475)
CDL Snippet: Eth/lwIP Hardware TCP/IP checksums

Hi Jerzy

Thank you for update.

I I haven't reviewed complete code yet, but I tried HW checksum gen/check part.
The behavior of HW checksum CDL is not as expected:
    - Following addition of lwIP and Common Ethernet Support HW checksum
remains unselected. It seems that configtool conflict resolution in this case
produces an opposite result than what we want.
    - When HW checksums are manually selected lwIP checksums are cleared as
they should but the opposite action is missing:
      - Unselect of HW checksum does not re-activate lwIP checksums;
      - Activation of some lwIP checksum de-selects HW checksum, but other lwIP
checksums remain unselected.

Attached CDL snippet should provide correct behavior for both. It may look
complex, but it seem there are some aspects of configtool's conflict resolution
machine that we have to circumvent somehow.

The submitted code is not fully tested, but stems from code that works for
other hardware so I expect it to be close to what we need for this driver.
Following test cases should be executed as a minimum:
    - eCos
        - lwIP default settings (should use HW checksum)
        - lwIP HW checksum disabled (should use lwIP checksum)
    - RedBoot - HW checksum should be inactive -> SW checksum.

It would be also good to test BSDstack but I guess that memory resources are
too low.

I'm looking forward to some feedback.

Ilija

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (26 preceding siblings ...)
  2011-12-21  9:33 ` bugzilla-daemon
@ 2011-12-26 21:47 ` bugzilla-daemon
  2012-01-02  0:35 ` bugzilla-daemon
                   ` (39 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2011-12-26 21:47 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1446|0                           |1
        is obsolete|                            |

--- Comment #28 from Jerzy Dyrda <jerzdy@gmail.com> 2011-12-26 21:47:23 GMT ---
Created an attachment (id=1483)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1483)
Ethernet driver for STM32 CL

Hello Ilija,

> I I haven't reviewed complete code yet, but I tried HW checksum gen/check
> part. The behavior of HW checksum CDL is not as expected:
(...)
> Attached CDL snippet should provide correct behavior for both. It may look
> complex, but it seem there are some aspects of configtool's conflict
> resolution machine that we have to circumvent somehow.
Thank you. I've just copied it into cdf file. I hope it's ok and it was your
intention to use your code.

> The submitted code is not fully tested, but stems from code that works for
> other hardware so I expect it to be close to what we need for this driver.
> Following test cases should be executed as a minimum:
To test ethernet driver I used a simple HTTP server from test module of lwIP
stack and several concurrent ping command with different payload sizes and
intervals
>     - eCos
>         - lwIP default settings (should use HW checksum)
>         - lwIP HW checksum disabled (should use lwIP checksum)
and in both cases eCos works stable and fine. 

>     - RedBoot - HW checksum should be inactive -> SW checksum.
Due to small amount of RAM >64k and no memory bus to connect external memory I
skipped RedBoot. 

> It would be also good to test BSDstack but I guess that memory resources
> are too low.
Exactly. 

Best regards,
jerzy

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (27 preceding siblings ...)
  2011-12-26 21:47 ` bugzilla-daemon
@ 2012-01-02  0:35 ` bugzilla-daemon
  2012-01-06 23:53 ` bugzilla-daemon
                   ` (38 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-01-02  0:35 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Ilija Kocho <ilijak@siva.com.mk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1475|0                           |1
        is obsolete|                            |

--- Comment #29 from Ilija Kocho <ilijak@siva.com.mk> 2012-01-02 00:34:50 GMT ---
(From update of attachment 1475)
Thank you for the update Jerzy.

I have checked building for default template with added Ethernet and lwIP -
seems proper.

Note: Just to state, there are some warnings that seem to be from old code -
not a matter for this bug.

Now we need little finish:

Missing ChangeLog entries for Variant and SPI remapping patches. Please add.

Copyright banners for driver and platform probably go too far in the past. I
take it that this is essentially new code (especially the driver) and copyright
message should reflect the fact.

And little clean up. There is a little trailing whitespace and some macros in
Variant have trailing semicolons ";" (look arround line 415). I guess they
shouldn't be.

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (28 preceding siblings ...)
  2012-01-02  0:35 ` bugzilla-daemon
@ 2012-01-06 23:53 ` bugzilla-daemon
  2012-01-06 23:56 ` bugzilla-daemon
                   ` (37 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-01-06 23:53 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1447|0                           |1
        is obsolete|                            |

--- Comment #30 from Jerzy Dyrda <jerzdy@gmail.com> 2012-01-06 23:53:21 GMT ---
Created an attachment (id=1488)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1488)
SPI pins remap for bus 1 and 3

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (29 preceding siblings ...)
  2012-01-06 23:53 ` bugzilla-daemon
@ 2012-01-06 23:56 ` bugzilla-daemon
  2012-01-06 23:57 ` bugzilla-daemon
                   ` (36 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-01-06 23:56 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1448|0                           |1
        is obsolete|                            |

--- Comment #31 from Jerzy Dyrda <jerzdy@gmail.com> 2012-01-06 23:55:52 GMT ---
Created an attachment (id=1489)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1489)
Platform for Propox MMstm32f107 module

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (31 preceding siblings ...)
  2012-01-06 23:57 ` bugzilla-daemon
@ 2012-01-06 23:57 ` bugzilla-daemon
  2012-01-07  0:02 ` bugzilla-daemon
                   ` (34 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-01-06 23:57 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1483|0                           |1
        is obsolete|                            |

--- Comment #33 from Jerzy Dyrda <jerzdy@gmail.com> 2012-01-06 23:57:22 GMT ---
Created an attachment (id=1491)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1491)
Ethernet driver for STM32 CL

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (30 preceding siblings ...)
  2012-01-06 23:56 ` bugzilla-daemon
@ 2012-01-06 23:57 ` bugzilla-daemon
  2012-01-06 23:57 ` bugzilla-daemon
                   ` (35 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-01-06 23:57 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1449|0                           |1
        is obsolete|                            |

--- Comment #32 from Jerzy Dyrda <jerzdy@gmail.com> 2012-01-06 23:56:37 GMT ---
Created an attachment (id=1490)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1490)
Variant modification needed by STM32 CL

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (32 preceding siblings ...)
  2012-01-06 23:57 ` bugzilla-daemon
@ 2012-01-07  0:02 ` bugzilla-daemon
  2012-01-07 22:31 ` bugzilla-daemon
                   ` (33 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-01-07  0:02 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #34 from Jerzy Dyrda <jerzdy@gmail.com> 2012-01-07 00:02:10 GMT ---
Hello Ilija,

(In reply to comment #29)
> (From update of attachment 1475 [details])
> Thank you for the update Jerzy.
> 
> Now we need little finish:
> 
> Missing ChangeLog entries for Variant and SPI remapping patches. Please add.
> 
> Copyright banners for driver and platform probably go too far in the past. I
> take it that this is essentially new code (especially the driver) and copyright
> message should reflect the fact.
> 
> And little clean up. There is a little trailing whitespace and some macros in
> Variant have trailing semicolons ";" (look arround line 415). I guess they
> shouldn't be.
All above done.

Thank you for your patient during code cleaning.

Best regards,
jerzy

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (33 preceding siblings ...)
  2012-01-07  0:02 ` bugzilla-daemon
@ 2012-01-07 22:31 ` bugzilla-daemon
  2012-01-09 14:19 ` bugzilla-daemon
                   ` (32 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-01-07 22:31 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #35 from Ilija Kocho <ilijak@siva.com.mk> 2012-01-07 22:31:15 GMT ---
(In reply to comment #32)
> Created an attachment (id=1490)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1490) [details]
> Variant modification needed by STM32 CL

Jerzy

ChangeLog seem to be missing for Variant modification.

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (34 preceding siblings ...)
  2012-01-07 22:31 ` bugzilla-daemon
@ 2012-01-09 14:19 ` bugzilla-daemon
  2012-02-02 21:44 ` bugzilla-daemon
                   ` (31 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-01-09 14:19 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1490|0                           |1
        is obsolete|                            |

--- Comment #36 from Jerzy Dyrda <jerzdy@gmail.com> 2012-01-09 14:19:00 GMT ---
Created an attachment (id=1494)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1494)
Variant modification needed by STM32 CL

Hello Ilja,
(In reply to comment #35)
> (In reply to comment #32)
> > Created an attachment (id=1490)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1490) [details]
[details]
> > Variant modification needed by STM32 CL

> ChangeLog seem to be missing for Variant modification.
Sorry, I attached old one patch instead on corrected.

Now should be ok. 

Best regards,
jerzy

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (35 preceding siblings ...)
  2012-01-09 14:19 ` bugzilla-daemon
@ 2012-02-02 21:44 ` bugzilla-daemon
  2012-02-03 13:20 ` bugzilla-daemon
                   ` (30 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-02-02 21:44 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #37 from Ilija Kocho <ilijak@siva.com.mk> 2012-02-02 21:44:01 GMT ---
Hi Jerzy

I notice that you have commented out great deal of hal_variant_init()
(stm32_misc.c). This can't be accepted since it breaks STM3210E. Simply doesn't
work.
The rule is simple, whatever you do, present code must work.

Selective clocking is a good so you can choose to use it with STM32CL but
untill it is (hopefully) implemented on STM3210E we must keep the code in
hal_variant_init()

Ilija

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (36 preceding siblings ...)
  2012-02-02 21:44 ` bugzilla-daemon
@ 2012-02-03 13:20 ` bugzilla-daemon
  2012-12-10 22:33 ` bugzilla-daemon
                   ` (29 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-02-03 13:20 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1494|0                           |1
        is obsolete|                            |

--- Comment #38 from Jerzy Dyrda <jerzdy@gmail.com> 2012-02-03 13:19:53 GMT ---
Created an attachment (id=1548)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1548)
Variant modification needed by STM32 CL

Hello Ilja,

(In reply to comment #37)
> I notice that you have commented out great deal of hal_variant_init()
> (stm32_misc.c). This can't be accepted since it breaks STM3210E. Simply doesn't
> work.
> The rule is simple, whatever you do, present code must work.
Of course.

> Selective clocking is a good so you can choose to use it with STM32CL but
> untill it is (hopefully) implemented on STM3210E we must keep the code in
> hal_variant_init()
OK. In new patch I'm using preprocessor to omit global clock enabling in cas of
STM32 CL family.

Best regards,
jerzy

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (37 preceding siblings ...)
  2012-02-03 13:20 ` bugzilla-daemon
@ 2012-12-10 22:33 ` bugzilla-daemon
  2012-12-21 20:02 ` bugzilla-daemon
                   ` (28 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-12-10 22:33 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1450|0                           |1
        is obsolete|                            |
   Attachment #1488|0                           |1
        is obsolete|                            |
   Attachment #1489|0                           |1
        is obsolete|                            |
   Attachment #1491|0                           |1
        is obsolete|                            |
   Attachment #1548|0                           |1
        is obsolete|                            |

--- Comment #39 from Jerzy Dyrda <jerzdy@gmail.com> 2012-12-10 22:33:15 GMT ---
Created an attachment (id=1997)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1997)
Eth driver for STM32 family. 

Hello all,

After long time I've gone back to eth driver and now it's works as well on
STM32F4 controller. 
Patch consists only from driver, few extra definitions in 
packages/hal/cortexm/stm32/var/current/include/var_io_eth and entry in ecos.db
Disclaimer:
Driver was tested only on Olimex board STM32-E407 what means different
connections
between PHY and MAC (RMII mode and clock 50 MHz from external generator) than
on STM3240G-EVAL.

Best regards,
jerzy

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (38 preceding siblings ...)
  2012-12-10 22:33 ` bugzilla-daemon
@ 2012-12-21 20:02 ` bugzilla-daemon
  2012-12-24 17:31 ` bugzilla-daemon
                   ` (27 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-12-21 20:02 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #40 from Ilija Kocho <ilijak@siva.com.mk> 2012-12-21 20:01:48 GMT ---
(In reply to comment #39)

Hi Jerzy

Thank you for the upgrade. I have STM3240G-EVAL handy and I am going to try it
soon.

Ilija


> Created an attachment (id=1997)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1997) [details]
> Eth driver for STM32 family. 
> 
> Hello all,
> 
> After long time I've gone back to eth driver and now it's works as well on
> STM32F4 controller. 
> Patch consists only from driver, few extra definitions in 
> packages/hal/cortexm/stm32/var/current/include/var_io_eth and entry in ecos.db
> Disclaimer:
> Driver was tested only on Olimex board STM32-E407 what means different
> connections
> between PHY and MAC (RMII mode and clock 50 MHz from external generator) than
> on STM3240G-EVAL.
> 
> Best regards,
> jerzy

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (39 preceding siblings ...)
  2012-12-21 20:02 ` bugzilla-daemon
@ 2012-12-24 17:31 ` bugzilla-daemon
  2012-12-24 21:46 ` bugzilla-daemon
                   ` (26 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-12-24 17:31 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #41 from Ilija Kocho <ilijak@siva.com.mk> 2012-12-24 17:31:27 GMT ---
(In reply to comment #39)
> Created an attachment (id=1997)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1997) [details]
> Eth driver for STM32 family. 
> 
> Hello all,
> 
> After long time I've gone back to eth driver and now it's works as well on
> STM32F4 controller. 
> Patch consists only from driver, few extra definitions in 
> packages/hal/cortexm/stm32/var/current/include/var_io_eth and entry in ecos.db
> Disclaimer:
> Driver was tested only on Olimex board STM32-E407 what means different
> connections
> between PHY and MAC (RMII mode and clock 50 MHz from external generator) than
> on STM3240G-EVAL.
> 

Jerzy, it would be of great help for testing if you provide some HAL patch for
STM3240G-EVAL

Thanks
Ilija

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (40 preceding siblings ...)
  2012-12-24 17:31 ` bugzilla-daemon
@ 2012-12-24 21:46 ` bugzilla-daemon
  2012-12-25 11:34 ` bugzilla-daemon
                   ` (25 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-12-24 21:46 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1997|0                           |1
        is obsolete|                            |

--- Comment #42 from Jerzy Dyrda <jerzdy@gmail.com> 2012-12-24 21:45:55 GMT ---
Created an attachment (id=2007)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=2007)
Eth driver for STM32 family.

Hello Ilija,

(In reply to comment #41)
> (In reply to comment #39)
> > Created an attachment (id=1997)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1997) [details]
[details]
> > Eth driver for STM32 family. 
> > After long time I've gone back to eth driver and now it's works as well on
> > STM32F4 controller. 
> > Patch consists only from driver, few extra definitions in 
> > packages/hal/cortexm/stm32/var/current/include/var_io_eth and entry in ecos.db
> > Disclaimer:
> > Driver was tested only on Olimex board STM32-E407 what means different
> > connections
> > between PHY and MAC (RMII mode and clock 50 MHz from external generator) than
> > on STM3240G-EVAL.
> > 
> 
> Jerzy, it would be of great help for testing if you provide some HAL patch for
> STM3240G-EVAL
I hope that I didn’t miss anything. At now I couldn't find no more that type of
MII interface requested by ETH driver - rest stuff like pins configuration,
clocks enabling and so on is done by driver.
Regarding CYGNUM_DEVS_ETH_CORTEXM_STM32_REMAP_PINS option is valid only for F1
family while AFAIK it isn't any stm32_f1 board with Ethernet support in eCos
repo. Thus I left it in driver's cdl without implementing inside the driver. If
you think that the driver should configure it not board dependent part I will
add it.

Best regards,
jerzy

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (41 preceding siblings ...)
  2012-12-24 21:46 ` bugzilla-daemon
@ 2012-12-25 11:34 ` bugzilla-daemon
  2012-12-27 17:45 ` bugzilla-daemon
                   ` (24 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-12-25 11:34 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #43 from Ilija Kocho <ilijak@siva.com.mk> 2012-12-25 11:33:45 GMT ---
(In reply to comment #42)
> Created an attachment (id=2007)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=2007) [details]
> Eth driver for STM32 family.
> 
> Hello Ilija,
> 
> (In reply to comment #41)
> > (In reply to comment #39)
> > > Created an attachment (id=1997)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1997) [details]
[details] [details]
> > > Eth driver for STM32 family. 
> > > After long time I've gone back to eth driver and now it's works as well on
> > > STM32F4 controller. 
> > > Patch consists only from driver, few extra definitions in 
> > > packages/hal/cortexm/stm32/var/current/include/var_io_eth and entry in ecos.db
> > > Disclaimer:
> > > Driver was tested only on Olimex board STM32-E407 what means different
> > > connections
> > > between PHY and MAC (RMII mode and clock 50 MHz from external generator) than
> > > on STM3240G-EVAL.
> > > 
> > 
> > Jerzy, it would be of great help for testing if you provide some HAL patch for
> > STM3240G-EVAL
> I hope that I didn’t miss anything. At now I couldn't find no more that type of
> MII interface requested by ETH driver - rest stuff like pins configuration,
> clocks enabling and so on is done by driver.
> Regarding CYGNUM_DEVS_ETH_CORTEXM_STM32_REMAP_PINS option is valid only for F1
> family while AFAIK it isn't any stm32_f1 board with Ethernet support in eCos
> repo. Thus I left it in driver's cdl without implementing inside the driver. If
> you think that the driver should configure it not board dependent part I will
> add it.

Thank you for the update.

I am able to run httpd_sequential on STM3240G-EVAL after I have:

1. Added CYGPKG_DEVS_ETH_CORTEXM_STM32 and CYGPKG_DEVS_ETH_PHY to target
stm3240g_eval (ecos.db)

2. Changed CYGHWR_DEVS_ETH_PHY_DP83847 to CYGHWR_DEVS_ETH_PHY_DP8384X
(hal_cortexm_stm32_stm32x0g_eval.cdl)

3. Changed J5 to select external clock.
Note: Selecting the MCO option produces conflict - needs fix.

The good thing is that I did not have to touch the driver, but the MCO conflict
implies that some fix may be needed.
On the formal side - pay attention to trailing spaces.

When posting the updated code please split the driver, HAL and ecos.db in
separate patches.

Ilija

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (42 preceding siblings ...)
  2012-12-25 11:34 ` bugzilla-daemon
@ 2012-12-27 17:45 ` bugzilla-daemon
  2012-12-28 14:15 ` bugzilla-daemon
                   ` (23 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-12-27 17:45 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #44 from Jerzy Dyrda <jerzdy@gmail.com> 2012-12-27 17:45:12 GMT ---
Hello Ilija,

(In reply to comment #43)
> (In reply to comment #42)
> > Created an attachment (id=2007)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=2007) [details]
[details]
> > Eth driver for STM32 family.
> > (In reply to comment #41)
> > > (In reply to comment #39)
> > > 
> > > Jerzy, it would be of great help for testing if you provide some HAL patch for
> > > STM3240G-EVAL
> > I hope that I didn’t miss anything. At now I couldn't find no more that type of
> > MII interface requested by ETH driver - rest stuff like pins configuration,
> > clocks enabling and so on is done by driver.
> > Regarding CYGNUM_DEVS_ETH_CORTEXM_STM32_REMAP_PINS option is valid only for F1
> > family while AFAIK it isn't any stm32_f1 board with Ethernet support in eCos
> > repo. Thus I left it in driver's cdl without implementing inside the driver. If
> > you think that the driver should configure it not board dependent part I will
> > add it.
> 
> Thank you for the update.
> 
> I am able to run httpd_sequential on STM3240G-EVAL after I have:
> 
> 1. Added CYGPKG_DEVS_ETH_CORTEXM_STM32 and CYGPKG_DEVS_ETH_PHY to target
> stm3240g_eval (ecos.db)
> 
> 2. Changed CYGHWR_DEVS_ETH_PHY_DP83847 to CYGHWR_DEVS_ETH_PHY_DP8384X
> (hal_cortexm_stm32_stm32x0g_eval.cdl)
Exactly. I forgot.

> 3. Changed J5 to select external clock.
> Note: Selecting the MCO option produces conflict - needs fix.
> 
> The good thing is that I did not have to touch the driver, but the MCO 
> conflict implies that some fix may be needed.
This was my source of worryings - MCO clock for PHY. According to the conflict
in deed I faced such problem but I ignored it.

> On the formal side - pay attention to trailing spaces.
> When posting the updated code please split the driver, HAL and ecos.db in
> separate patches.
Of course.

Should I correct something or regenerate patch split into parts or
patch in current state is acceptable?

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (43 preceding siblings ...)
  2012-12-27 17:45 ` bugzilla-daemon
@ 2012-12-28 14:15 ` bugzilla-daemon
  2013-01-01 21:09 ` bugzilla-daemon
                   ` (22 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2012-12-28 14:15 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #45 from Ilija Kocho <ilijak@siva.com.mk> 2012-12-28 14:15:27 GMT ---
Jerzy

(In reply to comment #44)

[snip]

> 
> Should I correct something or regenerate patch split into parts or
> patch in current state is acceptable?

In addition to points 1..3 mentioned in comment 43:

4. Check naming of CDLs. For instance CYGNUM_DEVS_ETH_CORTEXM_STM32_INTF is not
numerical. Likewise CYGNUM_DEVS_ETH_CORTEXM_STM32_REMAP_PINS and some others.
You shall find CDL naming convention at
http://ecos.sourceware.org/docs-latest/cdl-guide/language.naming.html.

5. There are some lines longer than 78 char. Not too much longer (couple of
char is not a problem, but check them.

6. Clean up trailing spaces.

7. Update the copyright banners (now is 2012 and soon 2013).
8. Likewise ChangeLog(s).

Happy New 2013
Ilija

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (44 preceding siblings ...)
  2012-12-28 14:15 ` bugzilla-daemon
@ 2013-01-01 21:09 ` bugzilla-daemon
  2013-01-01 21:10 ` bugzilla-daemon
                   ` (21 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-01-01 21:09 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #2007|0                           |1
        is obsolete|                            |

--- Comment #46 from Jerzy Dyrda <jerzdy@gmail.com> 2013-01-01 21:08:56 GMT ---
Created an attachment (id=2011)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=2011)
Eth driver for STM32 family.

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (45 preceding siblings ...)
  2013-01-01 21:09 ` bugzilla-daemon
@ 2013-01-01 21:10 ` bugzilla-daemon
  2013-01-01 21:11 ` bugzilla-daemon
                   ` (20 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-01-01 21:10 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #47 from Jerzy Dyrda <jerzdy@gmail.com> 2013-01-01 21:10:46 GMT ---
Created an attachment (id=2012)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=2012)
Adjustments to STM32 HAL required by eth driver.

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (46 preceding siblings ...)
  2013-01-01 21:10 ` bugzilla-daemon
@ 2013-01-01 21:11 ` bugzilla-daemon
  2013-01-01 21:17 ` bugzilla-daemon
                   ` (19 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-01-01 21:11 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #48 from Jerzy Dyrda <jerzdy@gmail.com> 2013-01-01 21:11:27 GMT ---
Created an attachment (id=2013)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=2013)
Introduced eth driver.

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (47 preceding siblings ...)
  2013-01-01 21:11 ` bugzilla-daemon
@ 2013-01-01 21:17 ` bugzilla-daemon
  2013-01-01 21:22 ` bugzilla-daemon
                   ` (18 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-01-01 21:17 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #2012|0                           |1
        is obsolete|                            |

--- Comment #49 from Jerzy Dyrda <jerzdy@gmail.com> 2013-01-01 21:17:08 GMT ---
Created an attachment (id=2014)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=2014)
Adjustments to STM32 HAL required by eth driver.

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (48 preceding siblings ...)
  2013-01-01 21:17 ` bugzilla-daemon
@ 2013-01-01 21:22 ` bugzilla-daemon
  2013-01-15 21:26 ` bugzilla-daemon
                   ` (17 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-01-01 21:22 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #50 from Jerzy Dyrda <jerzdy@gmail.com> 2013-01-01 21:22:11 GMT ---
Hello Ilija,

(In reply to comment #45)
> (In reply to comment #44)
> 
> [snip]
> 
> > 
> > Should I correct something or regenerate patch split into parts or
> > patch in current state is acceptable?
> 
> In addition to points 1..3 mentioned in comment 43:
> 
> 4. Check naming of CDLs. For instance CYGNUM_DEVS_ETH_CORTEXM_STM32_INTF is not
> numerical. Likewise CYGNUM_DEVS_ETH_CORTEXM_STM32_REMAP_PINS and some others.
> You shall find CDL naming convention at
> http://ecos.sourceware.org/docs-latest/cdl-guide/language.naming.html.
> 
> 5. There are some lines longer than 78 char. Not too much longer (couple of
> char is not a problem, but check them.
> 
> 6. Clean up trailing spaces.
> 
> 7. Update the copyright banners (now is 2012 and soon 2013).
> 8. Likewise ChangeLog(s).
Done.

According to point 3. I put MCO setting into board dependent part. IMHO driver
has to be generic and MCO configuration like source of clock and divider is up
to board dependent set up.

Best regards,
jerzy

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (49 preceding siblings ...)
  2013-01-01 21:22 ` bugzilla-daemon
@ 2013-01-15 21:26 ` bugzilla-daemon
  2013-01-15 21:27 ` bugzilla-daemon
                   ` (16 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-01-15 21:26 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #51 from Ilija Kocho <ilijak@siva.com.mk> 2013-01-15 21:25:23 GMT ---
Created an attachment (id=2027)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=2027)
TCP/IP checksum CDL fix 130115 (Incremental to attachment 2013)

Jerzy

There is a bug in IP/TCP/UDP checksum generation/checking CDL. The attached
patch should fix. Check if it works for your target.

Note: The patch is incremental to Attachment 2013.

Ilija

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (50 preceding siblings ...)
  2013-01-15 21:26 ` bugzilla-daemon
@ 2013-01-15 21:27 ` bugzilla-daemon
  2013-01-15 21:29 ` bugzilla-daemon
                   ` (15 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-01-15 21:27 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Ilija Kocho <ilijak@siva.com.mk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #2027|TCP/IP checksum CDL fix     |TCP/IP checksum CDL fix
        description|130115 (Incremental to      |130115 (Incremental to
                   |attachment 2013)            |attachment 2011)

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (51 preceding siblings ...)
  2013-01-15 21:27 ` bugzilla-daemon
@ 2013-01-15 21:29 ` bugzilla-daemon
  2013-01-20 21:47 ` bugzilla-daemon
                   ` (14 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-01-15 21:29 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #52 from Ilija Kocho <ilijak@siva.com.mk> 2013-01-15 21:28:49 GMT ---
(In reply to comment #51)
> Created an attachment (id=2027)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=2027) [details]
> TCP/IP checksum CDL fix 130115 (Incremental to attachment 2013 [details])
> 
> Jerzy
> 
> There is a bug in IP/TCP/UDP checksum generation/checking CDL. The attached
> patch should fix. Check if it works for your target.
> 
> Note: The patch is incremental to Attachment 2013 [details].

Oops, actually to Attachment 2011, sorry :(.

> 
> Ilija

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (52 preceding siblings ...)
  2013-01-15 21:29 ` bugzilla-daemon
@ 2013-01-20 21:47 ` bugzilla-daemon
  2013-02-07 15:06 ` bugzilla-daemon
                   ` (13 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-01-20 21:47 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #2011|0                           |1
        is obsolete|                            |

--- Comment #53 from Jerzy Dyrda <jerzdy@gmail.com> 2013-01-20 21:47:21 GMT ---
Created an attachment (id=2036)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=2036)
Eth driver for STM32 family.

Hello Ilija,

(In reply to comment #51)
> Created an attachment (id=2027)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=2027) [details]
> TCP/IP checksum CDL fix 130115 (Incremental to attachment 2013 [details])
> 
> There is a bug in IP/TCP/UDP checksum generation/checking CDL. The attached
> patch should fix. Check if it works for your target.
Ups. Now should be better and I hope that it's last issue to fix.

Best regards,
jerzy

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (53 preceding siblings ...)
  2013-01-20 21:47 ` bugzilla-daemon
@ 2013-02-07 15:06 ` bugzilla-daemon
  2013-02-07 18:05 ` bugzilla-daemon
                   ` (12 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-02-07 15:06 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Ilija Kocho <ilijak@siva.com.mk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #2027|0                           |1
        is obsolete|                            |

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (54 preceding siblings ...)
  2013-02-07 15:06 ` bugzilla-daemon
@ 2013-02-07 18:05 ` bugzilla-daemon
  2013-02-16 21:55 ` bugzilla-daemon
                   ` (11 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-02-07 18:05 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #54 from Ilija Kocho <ilijak@siva.com.mk> ---
(In reply to comment #53)
> Created attachment 2036 [details]
> Eth driver for STM32 family.
> 
> Hello Ilija,
> 
> (In reply to comment #51)
> > Created an attachment (id=2027) [details]
> > TCP/IP checksum CDL fix 130115 (Incremental to attachment 2013 [details])
> > 
> > There is a bug in IP/TCP/UDP checksum generation/checking CDL. The attached
> > patch should fix. Check if it works for your target.
> Ups. Now should be better and I hope that it's last issue to fix.

There's more. Have you tested with hardware checksum enabled? It doesn't work
on my target. Problem are the same macros, only in C source.

So check for other similar issues, than fix and than test, test and test. Until
you are sure that it works with both hw and sw checksum.
Then test again before you post the patch.
And for the end, don't forget to test.

Ilija

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (55 preceding siblings ...)
  2013-02-07 18:05 ` bugzilla-daemon
@ 2013-02-16 21:55 ` bugzilla-daemon
  2013-04-01 17:28 ` bugzilla-daemon
                   ` (10 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-02-16 21:55 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Jerzy Dyrda <jerzdy@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #2036|0                           |1
        is obsolete|                            |

--- Comment #55 from Jerzy Dyrda <jerzdy@gmail.com> ---
Created attachment 2093
  --> http://bugs.ecos.sourceware.org/attachment.cgi?id=2093&action=edit
Eth driver for STM32 family.

Hello Ilija,

(In reply to comment #54)
> (In reply to comment #53)
> > Created attachment 2036 [details]
> > Eth driver for STM32 family.
> > 
> > Hello Ilija,
> > 
> > (In reply to comment #51)
> > > Created an attachment (id=2027) [details]
> > > TCP/IP checksum CDL fix 130115 (Incremental to attachment 2013 [details])
> > > 
> > > There is a bug in IP/TCP/UDP checksum generation/checking CDL. The attached
> > > patch should fix. Check if it works for your target.
> > Ups. Now should be better and I hope that it's last issue to fix.
> 
> There's more. Have you tested with hardware checksum enabled? It doesn't
> work on my target. Problem are the same macros, only in C source.
> 
> So check for other similar issues, than fix and than test, test and test.
> Until you are sure that it works with both hw and sw checksum.
> Then test again before you post the patch.
> And for the end, don't forget to test.
Yes. You are absolutely right but my wiling to test becomes lower and lower
during so long time.

I hope that we are finalizing release of driver.

Best regards,
jerzy

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (56 preceding siblings ...)
  2013-02-16 21:55 ` bugzilla-daemon
@ 2013-04-01 17:28 ` bugzilla-daemon
  2013-04-01 18:25 ` bugzilla-daemon
                   ` (9 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-04-01 17:28 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Mike Jones <mjones@linear.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mjones@linear.com

--- Comment #56 from Mike Jones <mjones@linear.com> ---
I would like to try this on my STM3240G-EVAL Board. A couple of questions:

1) Should it work on this board? Need changes? No chance?
2) Do I apply all three patches? Last patch?
3) Will there be a new patch near term that it would be better to wait for?

Once it works on my eval board, I have a LwIP application with SNMP, FTP,
Telnet from my Kinetis project and I can develop on this board for a while to
help make sure it is robust.

Mike

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (57 preceding siblings ...)
  2013-04-01 17:28 ` bugzilla-daemon
@ 2013-04-01 18:25 ` bugzilla-daemon
  2013-04-06 22:43 ` bugzilla-daemon
                   ` (8 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-04-01 18:25 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #57 from Ilija Kocho <ilijak@siva.com.mk> ---
Mike

(In reply to comment #56)

Thank you for jumping in.
I just tried some basic tests: nc_test_slave, httpd_sequential and redboot. It
works for me when I apply all 3 patches to current SVN.

Probably no changes to the code for now (see my reply to Jerzy below). However
there's no guaranty :).


Jerzy
(In reply to comment #55)

Please excuse me for delay, I am in the middle of high priority job.

We're almost there, the driver seems ready for check-in. However ChangeLogs for
HAL are missing. Please update ChangeLog entries for all affected HAL packages.
To make it easier and cleaner for me, the patch(es) should be incremental (i.e.
only containing ChangeLog diffs).

Ilija


> I would like to try this on my STM3240G-EVAL Board. A couple of questions:
> 
> 1) Should it work on this board? Need changes? No chance?
> 2) Do I apply all three patches? Last patch?
> 3) Will there be a new patch near term that it would be better to wait for?
> 
> Once it works on my eval board, I have a LwIP application with SNMP, FTP,
> Telnet from my Kinetis project and I can develop on this board for a while
> to help make sure it is robust.
> 
> Mike

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (58 preceding siblings ...)
  2013-04-01 18:25 ` bugzilla-daemon
@ 2013-04-06 22:43 ` bugzilla-daemon
  2013-04-08 18:48 ` bugzilla-daemon
                   ` (7 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-04-06 22:43 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #58 from Jerzy Dyrda <jerzdy@gmail.com> ---
Created attachment 2165
  --> http://bugs.ecos.sourceware.org/attachment.cgi?id=2165&action=edit
Updated ChangeLog for patch "Adjustments to STM32 HAL required by eth driver."

Hello Ilija,

> Please excuse me for delay, I am in the middle of high priority job.
Of course. No problem.

> Please update ChangeLog entries for all affected HAL packages.
Done.

Best regards,
jerzy

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (59 preceding siblings ...)
  2013-04-06 22:43 ` bugzilla-daemon
@ 2013-04-08 18:48 ` bugzilla-daemon
  2013-04-09 11:03 ` bugzilla-daemon
                   ` (6 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-04-08 18:48 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #59 from Ilija Kocho <ilijak@siva.com.mk> ---
Jerzy

I was about to commit but in the last check I noticed something that needs
explanetion.

The function hal_stm32x0_led(). I assume there is sime conflict wiyh ethernet.
Please explain.

There must be better solution that deletting 3 leds.

1. Make compilation conditional, so LEDS are in use when possible and ommitted
otherwise.

2. Remove this function completely. or replace it with new. (I wouldn't
reccommend this).

If above assumption is not correct, there may be other solutions as well.

Ilija.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (60 preceding siblings ...)
  2013-04-08 18:48 ` bugzilla-daemon
@ 2013-04-09 11:03 ` bugzilla-daemon
  2013-04-09 12:14 ` bugzilla-daemon
                   ` (5 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-04-09 11:03 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #60 from Jerzy Dyrda <jerzdy@gmail.com> ---
Hello Ilija,

(In reply to comment #59)
> I was about to commit but in the last check I noticed something that needs
> explanation.
> 
> The function hal_stm32x0_led(). I assume there is some conflict with
> ethernet. Please explain.
> 
> There must be better solution that deletting 3 leds.
> 
> 1. Make compilation conditional, so LEDS are in use when possible and
> ommitted otherwise.
> 
> 2. Remove this function completely. or replace it with new. (I wouldn't
> reccommend this).
> 
> If above assumption is not correct, there may be other solutions as well.
Sorry I omitted it during checking. This was used by me only for my debug
purpose. Development board, which I used, has only one LED. 
Please skip this change it's absolutely not needed by driver.

Best regards,
jerzy

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (61 preceding siblings ...)
  2013-04-09 11:03 ` bugzilla-daemon
@ 2013-04-09 12:14 ` bugzilla-daemon
  2013-04-12 11:12 ` bugzilla-daemon
                   ` (4 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-04-09 12:14 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #61 from Ilija Kocho <ilijak@siva.com.mk> ---
(In reply to comment #60)
> Hello Ilija,
> 
> (In reply to comment #59)
> > I was about to commit but in the last check I noticed something that needs
> > explanation.
> > 
> > The function hal_stm32x0_led(). I assume there is some conflict with
> > ethernet. Please explain.
> > 
> > There must be better solution that deletting 3 leds.
> > 
> > 1. Make compilation conditional, so LEDS are in use when possible and
> > ommitted otherwise.
> > 
> > 2. Remove this function completely. or replace it with new. (I wouldn't
> > reccommend this).
> > 
> > If above assumption is not correct, there may be other solutions as well.
> Sorry I omitted it during checking. This was used by me only for my debug
> purpose. Development board, which I used, has only one LED. 
> Please skip this change it's absolutely not needed by driver.
> 

For avoidance of doubt: I revert the code for this function to the state before
adding of Ethernet driver.

Ilija

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (62 preceding siblings ...)
  2013-04-09 12:14 ` bugzilla-daemon
@ 2013-04-12 11:12 ` bugzilla-daemon
  2013-04-14 23:40 ` bugzilla-daemon
                   ` (3 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-04-12 11:12 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #62 from Jerzy Dyrda <jerzdy@gmail.com> ---
Hello Ilija,

(In reply to comment #61)
> (In reply to comment #60)
> > (In reply to comment #59)
> > > I was about to commit but in the last check I noticed something that needs
> > > explanation.
> > > 
> > > The function hal_stm32x0_led(). I assume there is some conflict with
> > > ethernet. Please explain.
> > > 
> > > There must be better solution that deletting 3 leds.
> > > 
> > > 1. Make compilation conditional, so LEDS are in use when possible and
> > > ommitted otherwise.
> > > 
> > > 2. Remove this function completely. or replace it with new. (I wouldn't
> > > reccommend this).
> > > 
> > > If above assumption is not correct, there may be other solutions as well.
> > Sorry I omitted it during checking. This was used by me only for my debug
> > purpose. Development board, which I used, has only one LED. 
> > Please skip this change it's absolutely not needed by driver.
> > 
> 
> For avoidance of doubt: I revert the code for this function to the state
> before adding of Ethernet driver.
Yes. Please.

Best regards,
jerzy

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (63 preceding siblings ...)
  2013-04-12 11:12 ` bugzilla-daemon
@ 2013-04-14 23:40 ` bugzilla-daemon
  2013-04-17 21:34 ` bugzilla-daemon
                   ` (2 subsequent siblings)
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-04-14 23:40 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #63 from Ilija Kocho <ilijak@siva.com.mk> ---
Checked in.
Jerzy, thank you for the contribution. Please check out and try.

Ilija

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (64 preceding siblings ...)
  2013-04-14 23:40 ` bugzilla-daemon
@ 2013-04-17 21:34 ` bugzilla-daemon
  2013-04-17 21:48 ` bugzilla-daemon
  2017-02-15  7:40 ` bugzilla-daemon
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-04-17 21:34 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

--- Comment #64 from Jerzy Dyrda <jerzdy@gmail.com> ---
Hello Ilija,

(In reply to comment #63)
> Jerzy, thank you for the contribution. Please check out and try.
It works. In case of any problems/questions from community I'm always ready
to support.

I would like to say thank you Ilija for your patient and comprehensive support.
It was pleasure to work with you.

Best regards,
jerzy

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219] Ethernet driver for STM32 connectivity line with port on MMstm32f107 board.
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (65 preceding siblings ...)
  2013-04-17 21:34 ` bugzilla-daemon
@ 2013-04-17 21:48 ` bugzilla-daemon
  2017-02-15  7:40 ` bugzilla-daemon
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2013-04-17 21:48 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Ilija Kocho <ilijak@siva.com.mk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|---                         |CURRENTRELEASE

--- Comment #65 from Ilija Kocho <ilijak@siva.com.mk> ---
(In reply to comment #64)
> Hello Ilija,
> 
> (In reply to comment #63)
> > Jerzy, thank you for the contribution. Please check out and try.
> It works. In case of any problems/questions from community I'm always ready
> to support.
> 
> I would like to say thank you Ilija for your patient and comprehensive
> support.
> It was pleasure to work with you.
> 
> Best regards,
> jerzy

Then this is resolved.
Jerzy, once again thank you for your contribution.

Ilija

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [Bug 1001219]   Ethernet driver for STM32 connectivity line with port on MMstm32f107 board  .
  2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
                   ` (66 preceding siblings ...)
  2013-04-17 21:48 ` bugzilla-daemon
@ 2017-02-15  7:40 ` bugzilla-daemon
  67 siblings, 0 replies; 69+ messages in thread
From: bugzilla-daemon @ 2017-02-15  7:40 UTC (permalink / raw)
  To: ecos-patches

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001219

Ilija Kocho [Илија Кочо] <ilijak@siva.com.mk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |CLOSED
      Custom Target|                            |STM32

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 69+ messages in thread

end of thread, other threads:[~2017-02-15  7:40 UTC | newest]

Thread overview: 69+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2011-05-03 22:11 [Bug 1001219] New: Ethernet driver for STM32 connectivity line with port on MMstm32f107 board bugzilla-daemon
2011-05-04  7:41 ` [Bug 1001219] " bugzilla-daemon
2011-05-12  9:18 ` bugzilla-daemon
2011-10-08 14:02 ` bugzilla-daemon
2011-10-08 14:27 ` bugzilla-daemon
2011-10-10 20:51 ` bugzilla-daemon
2011-10-10 20:54 ` bugzilla-daemon
2011-10-10 20:56 ` bugzilla-daemon
2011-10-10 21:13 ` bugzilla-daemon
2011-10-10 21:50 ` bugzilla-daemon
2011-10-11 19:53 ` bugzilla-daemon
2011-10-11 20:39 ` bugzilla-daemon
2011-10-11 21:08 ` bugzilla-daemon
2011-10-11 21:45 ` bugzilla-daemon
2011-10-12 11:25 ` bugzilla-daemon
2011-10-12 20:34 ` bugzilla-daemon
2011-10-12 21:15 ` bugzilla-daemon
2011-10-16 14:12 ` bugzilla-daemon
2011-10-16 14:12 ` bugzilla-daemon
2011-10-22 12:22 ` bugzilla-daemon
2011-10-23 21:27 ` bugzilla-daemon
2011-10-24 15:55 ` bugzilla-daemon
2011-12-04 21:22 ` bugzilla-daemon
2011-12-04 21:23 ` bugzilla-daemon
2011-12-04 21:24 ` bugzilla-daemon
2011-12-04 21:25 ` bugzilla-daemon
2011-12-04 21:35 ` bugzilla-daemon
2011-12-21  9:33 ` bugzilla-daemon
2011-12-26 21:47 ` bugzilla-daemon
2012-01-02  0:35 ` bugzilla-daemon
2012-01-06 23:53 ` bugzilla-daemon
2012-01-06 23:56 ` bugzilla-daemon
2012-01-06 23:57 ` bugzilla-daemon
2012-01-06 23:57 ` bugzilla-daemon
2012-01-07  0:02 ` bugzilla-daemon
2012-01-07 22:31 ` bugzilla-daemon
2012-01-09 14:19 ` bugzilla-daemon
2012-02-02 21:44 ` bugzilla-daemon
2012-02-03 13:20 ` bugzilla-daemon
2012-12-10 22:33 ` bugzilla-daemon
2012-12-21 20:02 ` bugzilla-daemon
2012-12-24 17:31 ` bugzilla-daemon
2012-12-24 21:46 ` bugzilla-daemon
2012-12-25 11:34 ` bugzilla-daemon
2012-12-27 17:45 ` bugzilla-daemon
2012-12-28 14:15 ` bugzilla-daemon
2013-01-01 21:09 ` bugzilla-daemon
2013-01-01 21:10 ` bugzilla-daemon
2013-01-01 21:11 ` bugzilla-daemon
2013-01-01 21:17 ` bugzilla-daemon
2013-01-01 21:22 ` bugzilla-daemon
2013-01-15 21:26 ` bugzilla-daemon
2013-01-15 21:27 ` bugzilla-daemon
2013-01-15 21:29 ` bugzilla-daemon
2013-01-20 21:47 ` bugzilla-daemon
2013-02-07 15:06 ` bugzilla-daemon
2013-02-07 18:05 ` bugzilla-daemon
2013-02-16 21:55 ` bugzilla-daemon
2013-04-01 17:28 ` bugzilla-daemon
2013-04-01 18:25 ` bugzilla-daemon
2013-04-06 22:43 ` bugzilla-daemon
2013-04-08 18:48 ` bugzilla-daemon
2013-04-09 11:03 ` bugzilla-daemon
2013-04-09 12:14 ` bugzilla-daemon
2013-04-12 11:12 ` bugzilla-daemon
2013-04-14 23:40 ` bugzilla-daemon
2013-04-17 21:34 ` bugzilla-daemon
2013-04-17 21:48 ` bugzilla-daemon
2017-02-15  7:40 ` bugzilla-daemon

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).