See patch.
This is the third iteration, now all code is in print.c together with the other printing code. The output now also contains chip size (in KB) and type (parallel, LPC, FWH, SPI).
The page at http://www.coreboot.org/Flashrom already contains the current -z output (as of yesterday) including new fields for chip size and type.
Uwe.
On 17.06.2009 16:10, Uwe Hermann wrote:
Add a --list-supported-wiki / -z option which outputs the currently supported flash chips (and their status, size, and type), chipsets (plus status), mainboards (plus status), and external PCI devices usable as programmer to stdout.
This allows for very easy pasting into the http://coreboot.org/Flashrom page, so we can keep that page up-to-date without much hassle.
The list of boards is mostly new (known good ones which don't need write-enable code, and known-bad ones) and also lists URLs to the vendor's mainboard pages.
Signed-off-by: Uwe Hermann uwe@hermann-uwe.de
Acked-by: Carl-Daniel Hailfinger c-d.hailfinger.devel.2006@gmx.net but see the comments below.
Index: flashrom.8
--- flashrom.8 (Revision 601) +++ flashrom.8 (Arbeitskopie)
The --list-supported-wiki is missing in the synopsis section.
@@ -123,6 +123,12 @@ to test an ERASE and/or WRITE operation, so make sure you only do that if you have proper means to recover from failure! .TP +.B "-z, --list-supported-wiki"
It would be really great if we could avoid short options for such seldom-used features. However, that's a general TODO for some options and we can try that later.
Index: flashrom.c
--- flashrom.c (Revision 601) +++ flashrom.c (Arbeitskopie) @@ -483,8 +483,8 @@ void usage(const char *name) { printf("usage: %s [-VfLhR] [-E|-r file|-w file|-v file] [-c chipname] [-s addr]\n"
" [-e addr] [-m [vendor:]part] [-l file] [-i image] [-p programmer] [file]\n\n",name);
" [-e addr] [-m [vendor:]part] [-l file] [-i image] ""[-p programmer] [file]\n\n", name);
[file] at the end should disappear. It's not your fault, but I just noticed it. Will send a patch.
@@ -505,12 +505,14 @@ " -l | --layout <file.layout>: read ROM layout from file\n" " -i | --image <name>: only flash image name from flash layout\n" " -L | --list-supported: print supported devices\n"
" -z | --list-supported-wiki: print supported devices in wiki syntax\n" " -p | --programmer <name>: specify the programmer device\n"
" (internal, dummy, nic3com, satasii, it87spi, ft2232spi)\n"
" (internal, dummy, nic3com, satasii,\n"" it87spi, ft2232spi)\n" " -h | --help: print this help text\n" " -R | --version: print the version (release)\n"
"\nYou can specify one of -E, -r, -w, -v or no operation.\n""If no operation is specified, then all that happens"
"\nYou can specify one of -E, -r, -w, -v or no operation. ""If no operation is\nspecified, then all that happens" " is that flash info is dumped.\n\n");
Can you kill the two-line change above? \n in the middle of a string seriously hampers readability. Since we violate the 80-column restriction in the code (but not the messages) above anyway, you could split the strings at the \n marker if you really want to perform cosmetic changes.
Index: print.c
--- print.c (Revision 601) +++ print.c (Arbeitskopie) @@ -222,3 +222,340 @@ "(total: %d):\n\n", boardcount); print_supported_boards_helper(boards_bad); }
+#define CHIPSET_TH "{| border="0" style="font-size: smaller"\n\ +|- bgcolor="#6699dd"\n! align="left" | Vendor\n\ +! align="left" | Southbridge\n! align="left" | PCI IDs\n\ +! align="left" | Status\n\n"
I'd really like to see const char *chipset_th instead of the #define. First, it decreases the size of the binary. Second, we are safe against unintended format strings in chipset_th (and others) if we specify them as printk arguments after the format string.
- { "ASUS", "A8NE-FM/S", "http://www.hardwareschotte.de/hardware/preise/proid_1266090/preis_ASUS+A8NE-..." },
Is that link to a pricing comparison website intentional?
I have to admit that I have a hard time parsing all the wiki syntax and table emitting code, but I trust you on this.
Regards, Carl-Daniel
On Fri, Jun 19, 2009 at 12:06:18PM +0200, Carl-Daniel Hailfinger wrote:
Acked-by: Carl-Daniel Hailfinger c-d.hailfinger.devel.2006@gmx.net
Thanks, r607.
Index: flashrom.c
--- flashrom.c (Revision 601) +++ flashrom.c (Arbeitskopie) @@ -483,8 +483,8 @@ void usage(const char *name) { printf("usage: %s [-VfLhR] [-E|-r file|-w file|-v file] [-c chipname] [-s addr]\n"
" [-e addr] [-m [vendor:]part] [-l file] [-i image] [-p programmer] [file]\n\n",name);
" [-e addr] [-m [vendor:]part] [-l file] [-i image] ""[-p programmer] [file]\n\n", name);[file] at the end should disappear. It's not your fault, but I just noticed it. Will send a patch.
OK. Not touching it in this patch then.
+#define CHIPSET_TH "{| border="0" style="font-size: smaller"\n\ +|- bgcolor="#6699dd"\n! align="left" | Vendor\n\ +! align="left" | Southbridge\n! align="left" | PCI IDs\n\ +! align="left" | Status\n\n"
I'd really like to see const char *chipset_th instead of the #define. First, it decreases the size of the binary. Second, we are safe against unintended format strings in chipset_th (and others) if we specify them as printk arguments after the format string.
Hm, I'll try to fix that in another patch. Last time I think I ran into some weird compiler issues and gave up.
- { "ASUS", "A8NE-FM/S", "http://www.hardwareschotte.de/hardware/preise/proid_1266090/preis_ASUS+A8NE-..." },
Is that link to a pricing comparison website intentional?
Well, sort of. It was the most usable (as in: contains board specs) I was able to find back then. This is an OEM board and I couldn't find a real "vendor page" for it.
Uwe.
-
Carl-Daniel Hailfinger -
Uwe Hermann