Re: [PATCH v4] [POWERPC] Move to runtime allocated exception stacks
On May 31, 2008, at 12:04 AM, Paul Mackerras wrote: Kumar Gala writes: For the additonal exception levels (critical, debug, machine check) on 40x/book-e we were using static allocations of the stack in the associated head.S. Move to a runtime allocation to make the code a bit easier to read as we mimic how we handle IRQ stacks. Its also a bit easier to setup the stack with a dummy thread_info in C code. Looks fine, with just one very minor comment: - addir11,r11,THREAD_SIZE; \ -1: subi r11,r11,INT_FRAME_SIZE; /* Allocate an exception frame */\ +1: addir11,r11,THREAD_SIZE; \ + subir11,r11,INT_FRAME_SIZE; /* Allocate an exception frame */\ You could make that a single addi r11,r11,THREAD_SIZE-INT_FRAME_SIZE now. I remember noticing that and thinking I should go back in fix. Will look at doing that and add the patches to my powerpc-next tree. - k ___ Linuxppc-dev mailing list Linuxppc-dev@ozlabs.org https://ozlabs.org/mailman/listinfo/linuxppc-dev
Re: [PATCH v4] [POWERPC] Move to runtime allocated exception stacks
Kumar Gala writes: For the additonal exception levels (critical, debug, machine check) on 40x/book-e we were using static allocations of the stack in the associated head.S. Move to a runtime allocation to make the code a bit easier to read as we mimic how we handle IRQ stacks. Its also a bit easier to setup the stack with a dummy thread_info in C code. Looks fine, with just one very minor comment: - addir11,r11,THREAD_SIZE; \ -1: subir11,r11,INT_FRAME_SIZE; /* Allocate an exception frame */\ +1: addir11,r11,THREAD_SIZE; \ + subir11,r11,INT_FRAME_SIZE; /* Allocate an exception frame */\ You could make that a single addi r11,r11,THREAD_SIZE-INT_FRAME_SIZE now. Paul. ___ Linuxppc-dev mailing list Linuxppc-dev@ozlabs.org https://ozlabs.org/mailman/listinfo/linuxppc-dev
Re: [PATCH v4] [POWERPC] Move to runtime allocated exception stacks
On Tue, 20 May 2008 00:02:45 -0500 Kumar Gala [EMAIL PROTECTED] wrote: For the additonal exception levels (critical, debug, machine check) on 40x/book-e we were using static allocations of the stack in the associated head.S. Move to a runtime allocation to make the code a bit easier to read as we mimic how we handle IRQ stacks. Its also a bit easier to setup the stack with a dummy thread_info in C code. Signed-off-by: Kumar Gala [EMAIL PROTECTED] --- Fixed 40x handling based on Josh's feedback and help. I tested this series on 44x and 40x today. Both boot up and function normally. GDB is able to set breakpoints, single step, and operate as well as it normally does. Again, no hard stress testing but it does appear to work. josh ___ Linuxppc-dev mailing list Linuxppc-dev@ozlabs.org https://ozlabs.org/mailman/listinfo/linuxppc-dev
