From: Philipp K. K. <pk...@sp...> - 2012-01-10 13:49:26
|
Is it time to finally get rid of at, bit, code, critical, data, far, eeprom, fixed16x16 flash idata, interrupt, nonbanked, banked, near, pdata, reentrant, shadowregs, wparam, sfr, sfr16, sfr32, sbit, sram, using, _naked, xdata, _overlay? They have been deprected for a while (sometime in between 2.9.0 and 3.0.0), giving a warning recommending to use the standard-compliant variants. They are still a source of annoying problems (people using them as identifiers), especially for newbies and when compiling third-party code not originally written for sdcc. If no one speaks up to save them, I'd remove them soon, so they are gone for the 3.2.0 release. Philipp |
From: Maarten B. <sou...@ds...> - 2012-01-10 14:30:58
|
They can go as far as I'm concerned. How about _asm and _endasm? I'd say remove them too. Maarten > Is it time to finally get rid of at, bit, code, critical, data, far, > eeprom, fixed16x16 flash idata, interrupt, nonbanked, banked, near, > pdata, reentrant, shadowregs, wparam, sfr, sfr16, sfr32, sbit, sram, > using, _naked, xdata, _overlay? They have been deprected for a while > (sometime in between 2.9.0 and 3.0.0), giving a warning recommending to > use the standard-compliant variants. They are still a source of annoying > problems (people using them as identifiers), especially for newbies and > when compiling third-party code not originally written for sdcc. If no > one speaks up to save them, I'd remove them soon, so they are gone for > the 3.2.0 release. > > Philipp |
From: Philipp K. K. <pk...@sp...> - 2012-01-10 15:07:07
|
Am 10.01.2012 14:59, schrieb Maarten Brock: > They can go as far as I'm concerned. > > How about _asm and _endasm? I'd say remove them too. I agree, I just forgot them when I made the list (since they are handled differently in SDCC.lex). Philipp |
From: Borut R. <bor...@gm...> - 2012-01-10 19:20:35
|
On 10. 01. 2012 16:07, Philipp Klaus Krause wrote: > Am 10.01.2012 14:59, schrieb Maarten Brock: >> They can go as far as I'm concerned. >> >> How about _asm and _endasm? I'd say remove them too. > I agree, I just forgot them when I made the list (since they are handled > differently in SDCC.lex). Don't forget that they are also handled in sdcpp. Borut |
From: Philipp K. K. <pk...@sp...> - 2012-01-25 11:02:53
|
Am 10.01.2012 20:20, schrieb Borut Razem: > On 10. 01. 2012 16:07, Philipp Klaus Krause wrote: >> Am 10.01.2012 14:59, schrieb Maarten Brock: >>> They can go as far as I'm concerned. >>> >>> How about _asm and _endasm? I'd say remove them too. >> I agree, I just forgot them when I made the list (since they are handled >> differently in SDCC.lex). > > Don't forget that they are also handled in sdcpp. I've never made a change top sdcpp. Could you point me to a location in the source that handles them? grep sfr, etc find nothing. Philipp |