OpenCores

OpenCores

label in generated asm code

no use

no use

1/1

no use

no use

label in generated asm code by Unknown on Feb 10, 2004			Not available!
[q] <A HREF="http://www.opencores.org/cvsweb.shtml/or1k/docs/openrisc_arch3.doc">http://www.opencores.org/cvsweb.shtml/or1k/docs/openrisc_arch3.doc</A> [/q] And there is also a PDF version: <A HREF="http://www.opencores.org/cvsweb.shtml/or1k/docs/openrisc_arch3.pdf">http://www.opencores.org/cvsweb.shtml/or1k/docs/openrisc_arch3.pdf</A> regards, Damjan http://www.opencores.org/cvsweb.shtml/or1k/docs/openrisc_arch3.doc And there is also a PDF version: http://www.opencores.org/cvsweb.shtml/or1k/docs/openrisc_arch3.pdf regards, Damjan

label in generated asm code by Unknown on Feb 10, 2004			Not available!
Hi all, I have a question about the following code. l.slli r3,r11,2 l.movhi r4,hi(.L145) # move immediate (high) l.ori r4,r4,lo(.L145) # move immediate (low) l.add r3,r3,r4 l.lwz r4,0(r3) # SI load l.jr r4 l.nop # nop delay slot .section .rodata .align 4 .align 4 .L145: .word .L108 .word .L113 .word .L144 .word .L114 .word .L119 .word .L120 .word .L144 .word .L121 .word .L144 .word .L144 .word .L122 .word .L144 .word .L144 .word .L127 .word .L144 .word .L134 .word .L135 .word .L144 .word .L136 .word .L137 .word .L138 .word .L144 .word .L139 .text .L108: l.addi r4,r0,1 # move immediate l.movhi r3,hi(_Main_Bitstream_Flag) # move immediate (high) l.ori r3,r3,lo(_Main_Bitstream_Flag) # move immediate (low) l.sw 0(r3),r4 [q]From the above code, the pc goes to label .L145 with l.jr4. But I donot [/q] know what the following asm code will do. Does it will run the code, in label .L108, then label .L113....? Can you one give me some explaination? Thank you very much. Best Regards, Stephen Hi all, I have a question about the following code. l.slli r3,r11,2 l.movhi r4,hi(.L145) # move immediate (high) l.ori r4,r4,lo(.L145) # move immediate (low) l.add r3,r3,r4 l.lwz r4,0(r3) # SI load l.jr r4 l.nop # nop delay slot .section .rodata .align 4 .align 4 .L145: .word .L108 .word .L113 .word .L144 .word .L114 .word .L119 .word .L120 .word .L144 .word .L121 .word .L144 .word .L144 .word .L122 .word .L144 .word .L144 .word .L127 .word .L144 .word .L134 .word .L135 .word .L144 .word .L136 .word .L137 .word .L138 .word .L144 .word .L139 .text .L108: l.addi r4,r0,1 # move immediate l.movhi r3,hi(_Main_Bitstream_Flag) # move immediate (high) l.ori r3,r3,lo(_Main_Bitstream_Flag) # move immediate (low) l.sw 0(r3),r4 From the above code, the pc goes to label .L145 with l.jr4. But I donot know what the following asm code will do. Does it will run the code, in label .L108, then label .L113....? Can you one give me some explaination? Thank you very much. Best Regards, Stephen

label in generated asm code by Unknown on Feb 10, 2004			Not available!
Firstly, I would strongly suggest you spend some time reading the architecture manual to understand how the OR32 instruction set works. You can find this here: <A HREF="http://www.opencores.org/cvsweb.shtml/or1k/docs/openrisc_arch3.doc">http://www.opencores.org/cvsweb.shtml/or1k/docs/openrisc_arch3.doc</A> Further comments inline. [q] l.slli r3,r11,2[/q] Shift r11 left by two and put result in r3. This is creating a 32 bit offset (i.e. 0,4,8 bytes) from an ordinal index (i.e. 0,1,2 etc.). [q] l.movhi r4,hi(.L145) # move immediate (high) l.ori r4,r4,lo(.L145) # move immediate (low)[/q] Load address of .L145 into r4 [q] l.add r3,r3,r4[/q] Add offset created above to it and store result in r3 [q] l.lwz r4,0(r3) # SI load[/q] Load data at address pointed to by r3 into r4. [q] l.jr r4 l.nop # nop delay slot[/q] ...and jump to it [q] .section .rodata .align 4 .align 4 .L145: .word .L108 .word .L113 .word .L144[/q] [snip] [q] .word .L139 .text[/q] The above is simply the jump table (I've shortened it a bit). [q] .L108: l.addi r4,r0,1 # move immediate l.movhi r3,hi(_Main_Bitstream_Flag) # move immediate (high) l.ori r3,r3,lo(_Main_Bitstream_Flag) # move immediate (low) l.sw 0(r3),r4[/q] and this is the first bit of code it will execute via the jump table at index 0. I have annotated the code to show you in more detail: [q] l.slli r3,r11,2 # Let's says r11 = 2: r3 now contains 2 <<[/q] 2, i.e. 8 [q] l.movhi r4,hi(.L145) # l.ori r4,r4,lo(.L145) # r4 now contains the address .L145 l.add r3,r3,r4 # r3 now contains r4 + 8 = .L145 + 8 l.lwz r4,0(r3) # r4 contains the data at .L145 + 8, which is[/q] .L144 from the table below [q] l.jr r4 # So jump to .L144 l.nop # nop delay slot .section .rodata .align 4 .align 4 .L145: .word .L108 .word .L113[/q] #### .L145 + 8 is here [q] .word .L144[/q] [q]>From the above code, the pc goes to label .L145 with l.jr4.[/q] No - see my explanation. [q] But I donot know what the following asm code will do. Does it will run the code, in label .L108, then label .L113....? Can you one give me some explaination?[/q] You don't show enough of the code to know this - what happens at the end of the code after .L108? I would assume there is another jump here. Robert Cragie, Design Engineer _______________________________________________________________ Jennic Ltd, Furnival Street, Sheffield, S1 4QT, UK <A HREF="http://www.jennic.com">http://www.jennic.com</A> Tel: +44 (0) 114 281 2655 _______________________________________________________________ Firstly, I would strongly suggest you spend some time reading the architecture manual to understand how the OR32 instruction set works. You can find this here: http://www.opencores.org/cvsweb.shtml/or1k/docs/openrisc_arch3.doc Further comments inline. l.slli r3,r11,2 Shift r11 left by two and put result in r3. This is creating a 32 bit offset (i.e. 0,4,8 bytes) from an ordinal index (i.e. 0,1,2 etc.). l.movhi r4,hi(.L145) # move immediate (high) l.ori r4,r4,lo(.L145) # move immediate (low) Load address of .L145 into r4 l.add r3,r3,r4 Add offset created above to it and store result in r3 l.lwz r4,0(r3) # SI load Load data at address pointed to by r3 into r4. l.jr r4 l.nop # nop delay slot ...and jump to it .section .rodata .align 4 .align 4 .L145: .word .L108 .word .L113 .word .L144 [snip] .word .L139 .text The above is simply the jump table (I've shortened it a bit). .L108: l.addi r4,r0,1 # move immediate l.movhi r3,hi(_Main_Bitstream_Flag) # move immediate (high) l.ori r3,r3,lo(_Main_Bitstream_Flag) # move immediate (low) l.sw 0(r3),r4 and this is the first bit of code it will execute via the jump table at index 0. I have annotated the code to show you in more detail: l.slli r3,r11,2 # Let's says r11 = 2: r3 now contains 2 2, i.e. 8 l.movhi r4,hi(.L145) # l.ori r4,r4,lo(.L145) # r4 now contains the address .L145 l.add r3,r3,r4 # r3 now contains r4 + 8 = .L145 + 8 l.lwz r4,0(r3) # r4 contains the data at .L145 + 8, which is .L144 from the table below l.jr r4 # So jump to .L144 l.nop # nop delay slot .section .rodata .align 4 .align 4 .L145: .word .L108 .word .L113 #### .L145 + 8 is here .word .L144 >From the above code, the pc goes to label .L145 with l.jr4. No - see my explanation. But I donot know what the following asm code will do. Does it will run the code, in label .L108, then label .L113....? Can you one give me some explaination? You don't show enough of the code to know this - what happens at the end of the code after .L108? I would assume there is another jump here. Robert Cragie, Design Engineer _______________________________________________________________ Jennic Ltd, Furnival Street, Sheffield, S1 4QT, UK http://www.jennic.com Tel: +44 (0) 114 281 2655 _______________________________________________________________

label in generated asm code by Unknown on Feb 10, 2004			Not available!
Hi Robert, Thank you for your explaination. In my source code, the jump table is coded in the function(generated by complier). i.e. .proc _function _function: ..... .... jump table .... .... .endproc _function If I move the jump the table to the end of function manually. i.e. .proc _function _function: ..... .... .... .... .endproc _function jump table The result of the function will be changed ? Thank you very much. Best Regards, Stephen Robert Cragie <<A HREF="mailto:rcc@jennic.com">rcc@jennic.com</A>> wrote: Firstly, I would strongly suggest you spend some time reading the architecture manual to understand how the OR32 instruction set works. You can find this here: <A HREF="http://www.opencores.org/cvsweb.shtml/or1k/docs/openrisc_arch3.doc">http://www.opencores.org/cvsweb.shtml/or1k/docs/openrisc_arch3.doc</A> Further comments inline. [q] l.slli r3,r11,2[/q] Shift r11 left by two and put result in r3. This is creating a 32 bit offset (i.e. 0,4,8 bytes) from an ordinal index (i.e. 0,1,2 etc.). [q] l.movhi r4,hi(.L145) # move immediate (high) l.ori r4,r4,lo(.L145) # move immediate (low)[/q] Load address of .L145 into r4 [q] l.add r3,r3,r4[/q] Add offset created above to it and store result in r3 [q] l.lwz r4,0(r3) # SI load[/q] Load data at address pointed to by r3 into r4. [q] l.jr r4 l.nop # nop delay slot[/q] ...and jump to it [q] .section .rodata .align 4 .align 4 .L145: .word .L108 .word .L113 .word .L144[/q] [snip] [q] .word .L139 .text[/q] The above is simply the jump table (I've shortened it a bit). [q] .L108: l.addi r4,r0,1 # move immediate l.movhi r3,hi(_Main_Bitstream_Flag) # move immediate (high) l.ori r3,r3,lo(_Main_Bitstream_Flag) # move immediate (low) l.sw 0(r3),r4[/q] and this is the first bit of code it will execute via the jump table at index 0. I have annotated the code to show you in more detail: [q] l.slli r3,r11,2 # Let's says r11 = 2: r3 now contains 2 <<[/q] 2, i.e. 8 [q] l.movhi r4,hi(.L145) # l.ori r4,r4,lo(.L145) # r4 now contains the address .L145 l.add r3,r3,r4 # r3 now contains r4 + 8 = .L145 + 8 l.lwz r4,0(r3) # r4 contains the data at .L145 + 8, which is[/q] .L144 from the table below [q] l.jr r4 # So jump to .L144 l.nop # nop delay slot .section .rodata .align 4 .align 4 .L145: .word .L108 .word .L113[/q] #### .L145 + 8 is here [q] .word .L144[/q] [q]>From the above code, the pc goes to label .L145 with l.jr4.[/q] No - see my explanation. [q] But I donot know what the following asm code will do. Does it will run the code, in label .L108, then label .L113....? Can you one give me some explaination?[/q] You don't show enough of the code to know this - what happens at the end of the code after .L108? I would assume there is another jump here. Robert Cragie, Design Engineer _______________________________________________________________ Jennic Ltd, Furnival Street, Sheffield, S1 4QT, UK <A HREF="http://www.jennic.com">http://www.jennic.com</A> Tel: +44 (0) 114 281 2655 _______________________________________________________________ _______________________________________________ <A HREF="http://www.opencores.org/mailman/listinfo/openrisc">http://www.opencores.org/mailman/listinfo/openrisc</A> Â¥Â²Â±Ã¾Â§ÃžÂ¡BÂ¶Â¼ÂºqÂ¡BÂ¤pÂ¬PÂ¬P... Â®Ã¶ÂºÂ©Â¹aÃn Â±Â¡Â¤ÃŸÂ³sÃƒÂ´ <A HREF="http://ringtone.yahoo.com.hk/">http://ringtone.yahoo.com.hk/</A> -------------- next part -------------- An HTML attachment was scrubbed... URL: <A HREF="http://www.opencores.org/forums/openrisc/attachments/20040210/1691632b/attachment.htm">http://www.opencores.org/forums/openrisc/attachments/20040210/1691632b/attachment.htm</A> Hi Robert, Thank you for your explaination. In my source code, the jump table is coded in the function(generated by complier). i.e. .proc _function _function: ..... .... jump table .... .... .endproc _function If I move the jump the table to the end of function manually. i.e. .proc _function _function: ..... .... .... .... .endproc _function jump table The result of the function will be changed ? Thank you very much. Best Regards, Stephen Robert Cragie rcc@jennic.com> wrote: Firstly, I would strongly suggest you spend some time reading the architecture manual to understand how the OR32 instruction set works. You can find this here: http://www.opencores.org/cvsweb.shtml/or1k/docs/openrisc_arch3.doc Further comments inline. l.slli r3,r11,2 Shift r11 left by two and put result in r3. This is creating a 32 bit offset (i.e. 0,4,8 bytes) from an ordinal index (i.e. 0,1,2 etc.). l.movhi r4,hi(.L145) # move immediate (high) l.ori r4,r4,lo(.L145) # move immediate (low) Load address of .L145 into r4 l.add r3,r3,r4 Add offset created above to it and store result in r3 l.lwz r4,0(r3) # SI load Load data at address pointed to by r3 into r4. l.jr r4 l.nop # nop delay slot ...and jump to it .section .rodata .align 4 .align 4 .L145: .word .L108 .word .L113 .word .L144 [snip] .word .L139 .text The above is simply the jump table (I've shortened it a bit). .L108: l.addi r4,r0,1 # move immediate l.movhi r3,hi(_Main_Bitstream_Flag) # move immediate (high) l.ori r3,r3,lo(_Main_Bitstream_Flag) # move immediate (low) l.sw 0(r3),r4 and this is the first bit of code it will execute via the jump table at index 0. I have annotated the code to show you in more detail: l.slli r3,r11,2 # Let's says r11 = 2: r3 now contains 2 2, i.e. 8 l.movhi r4,hi(.L145) # l.ori r4,r4,lo(.L145) # r4 now contains the address .L145 l.add r3,r3,r4 # r3 now contains r4 + 8 = .L145 + 8 l.lwz r4,0(r3) # r4 contains the data at .L145 + 8, which is .L144 from the table below l.jr r4 # So jump to .L144 l.nop # nop delay slot .section .rodata .align 4 .align 4 .L145: .word .L108 .word .L113 #### .L145 + 8 is here .word .L144 >From the above code, the pc goes to label .L145 with l.jr4. No - see my explanation. But I donot know what the following asm code will do. Does it will run the code, in label .L108, then label .L113....? Can you one give me some explaination? You don't show enough of the code to know this - what happens at the end of the code after .L108? I would assume there is another jump here. Robert Cragie, Design Engineer _______________________________________________________________ Jennic Ltd, Furnival Street, Sheffield, S1 4QT, UK http://www.jennic.com Tel: +44 (0) 114 281 2655 _______________________________________________________________ _______________________________________________ http://www.opencores.org/mailman/listinfo/openrisc Â¥Â²Â±Ã¾Â§ÃžÂ¡BÂ¶Â¼ÂºqÂ¡BÂ¤pÂ¬PÂ¬P... Â®Ã¶ÂºÂ©Â¹aÃn Â±Â¡Â¤ÃŸÂ³sÃƒÂ´ http://ringtone.yahoo.com.hk/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.opencores.org/forums/openrisc/attachments/20040210/1691632b/attachment.htm

no use

no use

1/1

no use

no use

© copyright 1999-2025 OpenCores.org, equivalent to Oliscience, all rights reserved. OpenCores®, registered trademark.