* [ECOS] Possible typo in hal/arm/arm9/var/current/include/hal_cache.h @ 2003-01-07 18:53 Patrick Doyle 2003-01-07 20:18 ` Jonathan Larmour 0 siblings, 1 reply; 9+ messages in thread From: Patrick Doyle @ 2003-01-07 18:53 UTC (permalink / raw) To: eCos Can somebody (jskov perhaps, since your name is in the ChangeLog) tell me the source of the data for the ARM925T cache configuration? I am asking because the #defines in this file don't match the documentation I have from TI. I will ask TI as well, but in the mean time, here is what the file says: # define HAL_ICACHE_SIZE 0x4000 # define HAL_ICACHE_LINE_SIZE 32 # define HAL_ICACHE_WAYS 2 # define HAL_ICACHE_SETS (HAL_ICACHE_SIZE/(HAL_ICACHE_LINE_SIZE*HAL_ICACHE_WAYS)) And here is what the TI documentation (somewhat ambiguously) says: "The 16K-byte instruction cache (I-cache) has 1024 lines of 16 bytes arranged as a two-way set-associative cache." These two pieces of information clearly differ in the definition of the line size. Also, my first read of the TI documentation made me think there were 1024 lines in each set, but the math doesn't work out for that. Instead, I think there are 1024 lines total, 512 in each set. Regardless, these don't match the 256 sets defined by HAL_ICACHE_SETS. Anyway, I thought I would ask here as well as at TI. All comments are welcome. --wpd -- Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos and search the list archive: http://sources.redhat.com/ml/ecos-discuss ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [ECOS] Possible typo in hal/arm/arm9/var/current/include/hal_cache.h 2003-01-07 18:53 [ECOS] Possible typo in hal/arm/arm9/var/current/include/hal_cache.h Patrick Doyle @ 2003-01-07 20:18 ` Jonathan Larmour 2003-01-07 20:24 ` Gary D. Thomas 0 siblings, 1 reply; 9+ messages in thread From: Jonathan Larmour @ 2003-01-07 20:18 UTC (permalink / raw) To: Patrick Doyle; +Cc: eCos Patrick Doyle wrote: > Can somebody (jskov perhaps, since your name is in the ChangeLog) I'm not sure Jesper still reads ecos-discuss. I think he doesn't any more unfortunately. > tell me > the source of the data for the ARM925T cache configuration? I am asking > because the #defines in this file don't match the documentation I have from > TI. I will ask TI as well, but in the mean time, here is what the file > says: > > # define HAL_ICACHE_SIZE 0x4000 > # define HAL_ICACHE_LINE_SIZE 32 > # define HAL_ICACHE_WAYS 2 > # define HAL_ICACHE_SETS > (HAL_ICACHE_SIZE/(HAL_ICACHE_LINE_SIZE*HAL_ICACHE_WAYS)) > > And here is what the TI documentation (somewhat ambiguously) says: > > "The 16K-byte instruction cache (I-cache) has 1024 lines of 16 bytes > arranged as a two-way set-associative cache." I notice hal_arm_arm9.cdl says: The ARM925T has 8k data cache, 16k instruction cache, 16 word write buffer and an MMU." > These two pieces of information clearly differ in the definition of the line > size. Also, my first read of the TI documentation made me think there were > 1024 lines in each set, but the math doesn't work out for that. Instead, I > think there are 1024 lines total, 512 in each set. Regardless, these don't > match the 256 sets defined by HAL_ICACHE_SETS. The board this was written for was never released publically, but was from TI. I believe this is the CPU: http://www-s.ti.com/sc/ds/omap5910.pdf and more relevantly <http://focus.ti.com/docs/prod/productfolder.jhtml?genericPartNumber=OMAP5910&pfsection=user_man> which includes the very paragraph you mention! So I'm inclined to believe it's wrong, and a patch is welcome :-). > Anyway, I thought I would ask here as well as at TI. All comments are > welcome. If you've got the hardware you could play around! At least to verify that it's wrong. Jifl -- eCosCentric http://www.eCosCentric.com/ <info@eCosCentric.com> --[ "You can complain because roses have thorns, or you ]-- --[ can rejoice because thorns have roses." -Lincoln ]-- Opinions==mine -- Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos and search the list archive: http://sources.redhat.com/ml/ecos-discuss ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [ECOS] Possible typo in hal/arm/arm9/var/current/include/hal_cache.h 2003-01-07 20:18 ` Jonathan Larmour @ 2003-01-07 20:24 ` Gary D. Thomas 2003-01-07 21:00 ` Patrick Doyle 0 siblings, 1 reply; 9+ messages in thread From: Gary D. Thomas @ 2003-01-07 20:24 UTC (permalink / raw) To: Jonathan Larmour; +Cc: Patrick Doyle, eCos Discussion On Tue, 2003-01-07 at 13:17, Jonathan Larmour wrote: > Patrick Doyle wrote: > > Can somebody (jskov perhaps, since your name is in the ChangeLog) > > I'm not sure Jesper still reads ecos-discuss. I think he doesn't any more > unfortunately. > > > tell me > > the source of the data for the ARM925T cache configuration? I am asking > > because the #defines in this file don't match the documentation I have from > > TI. I will ask TI as well, but in the mean time, here is what the file > > says: > > > > # define HAL_ICACHE_SIZE 0x4000 > > # define HAL_ICACHE_LINE_SIZE 32 > > # define HAL_ICACHE_WAYS 2 > > # define HAL_ICACHE_SETS > > (HAL_ICACHE_SIZE/(HAL_ICACHE_LINE_SIZE*HAL_ICACHE_WAYS)) > > > > And here is what the TI documentation (somewhat ambiguously) says: > > > > "The 16K-byte instruction cache (I-cache) has 1024 lines of 16 bytes > > arranged as a two-way set-associative cache." > > I notice hal_arm_arm9.cdl says: > The ARM925T has 8k data cache, 16k instruction cache, 16 word > write buffer and an MMU." > > > These two pieces of information clearly differ in the definition of the line > > size. Also, my first read of the TI documentation made me think there were > > 1024 lines in each set, but the math doesn't work out for that. Instead, I > > think there are 1024 lines total, 512 in each set. Regardless, these don't > > match the 256 sets defined by HAL_ICACHE_SETS. > > The board this was written for was never released publically, but was from > TI. I believe this is the CPU: http://www-s.ti.com/sc/ds/omap5910.pdf and > more relevantly > <http://focus.ti.com/docs/prod/productfolder.jhtml?genericPartNumber=OMAP5910&pfsection=user_man> > which includes the very paragraph you mention! > IIRC, Jesper never got the DATA cache working on that platform anyway - maybe this is why :-) > So I'm inclined to believe it's wrong, and a patch is welcome :-). > > > Anyway, I thought I would ask here as well as at TI. All comments are > > welcome. > > If you've got the hardware you could play around! At least to verify that > it's wrong. > Indeed, and then please share the results! -- .--------------------------------------------------------. | Mind: Embedded Linux and eCos Development | |--------------------------------------------------------| | Gary Thomas email: gary.thomas@mind.be | | Mind ( http://mind.be ) tel: +1 (970) 229-1963 | | gpg: http://www.chez-thomas.org/gary/gpg_key.asc | '--------------------------------------------------------' -- Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos and search the list archive: http://sources.redhat.com/ml/ecos-discuss ^ permalink raw reply [flat|nested] 9+ messages in thread
* RE: [ECOS] Possible typo in hal/arm/arm9/var/current/include/hal_cache.h 2003-01-07 20:24 ` Gary D. Thomas @ 2003-01-07 21:00 ` Patrick Doyle 2003-01-07 21:14 ` Gary D. Thomas 0 siblings, 1 reply; 9+ messages in thread From: Patrick Doyle @ 2003-01-07 21:00 UTC (permalink / raw) To: Gary D. Thomas, Jonathan Larmour; +Cc: eCos Discussion > > The board this was written for was never released publically, Watch this space for an announcement real soon about RedBoot running just such an OMAP5910 board! > IIRC, Jesper never got the DATA cache working on that platform > anyway - maybe this is why :-) > > So I'm inclined to believe it's wrong, and a patch is welcome :-). > > If you've got the hardware you could play around! At least to verify that > > it's wrong. Hmmm... that opens up a very interesting question, how would I "verify that it's wrong"? I'll have to think about a test case on the drive home tonight that would verify one cache line size vs. another. > Indeed, and then please share the results! But, of course! --wpd -- Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos and search the list archive: http://sources.redhat.com/ml/ecos-discuss ^ permalink raw reply [flat|nested] 9+ messages in thread
* RE: [ECOS] Possible typo in hal/arm/arm9/var/current/include/hal_cache.h 2003-01-07 21:00 ` Patrick Doyle @ 2003-01-07 21:14 ` Gary D. Thomas 2003-01-07 21:31 ` [ECOS] Possible typo inhal/arm/arm9/var/current/include/hal_cache.h Patrick Doyle 0 siblings, 1 reply; 9+ messages in thread From: Gary D. Thomas @ 2003-01-07 21:14 UTC (permalink / raw) To: Patrick Doyle; +Cc: Jonathan Larmour, eCos Discussion On Tue, 2003-01-07 at 13:57, Patrick Doyle wrote: > > > The board this was written for was never released publically, > > Watch this space for an announcement real soon about RedBoot running just > such an OMAP5910 board! > > > IIRC, Jesper never got the DATA cache working on that platform > > anyway - maybe this is why :-) > > > > So I'm inclined to believe it's wrong, and a patch is welcome :-). > > > > > If you've got the hardware you could play around! At least to verify > that > > > it's wrong. > > Hmmm... that opens up a very interesting question, how would I "verify that > it's wrong"? I'll have to think about a test case on the drive home tonight > that would verify one cache line size vs. another. > I just added some test code to kernel/tests/kcache2.c which is supposed to be able to check this sort of thing. You have to enable it [by hand], but it might be useful. > > Indeed, and then please share the results! > But, of course! > > --wpd -- .--------------------------------------------------------. | Mind: Embedded Linux and eCos Development | |--------------------------------------------------------| | Gary Thomas email: gary.thomas@mind.be | | Mind ( http://mind.be ) tel: +1 (970) 229-1963 | | gpg: http://www.chez-thomas.org/gary/gpg_key.asc | '--------------------------------------------------------' -- Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos and search the list archive: http://sources.redhat.com/ml/ecos-discuss ^ permalink raw reply [flat|nested] 9+ messages in thread
* RE: [ECOS] Possible typo inhal/arm/arm9/var/current/include/hal_cache.h 2003-01-07 21:14 ` Gary D. Thomas @ 2003-01-07 21:31 ` Patrick Doyle 2003-01-07 21:36 ` Gary D. Thomas 0 siblings, 1 reply; 9+ messages in thread From: Patrick Doyle @ 2003-01-07 21:31 UTC (permalink / raw) To: Gary D. Thomas; +Cc: Jonathan Larmour, eCos Discussion OK, I'll have a look at that. So far I have avoided running any of the kernel tests because I have been focusing very intently on getting RedBoot up and running, and have ignored anything not directly related to RedBoot. Thus, I will feel comfortable saying at the end of the week that "I have ported RedBoot to the OMAP", but there is still somme work to be done to claim that I have ported eCos to the OMAP (most noticably, I have not done any work at all with interrupts, consequently, I have not done any work at all with the real time clock, consequently, I have not tried running any of the "kernel specific" tests). Anyway, I'll look at it. Perhaps it will run fine even if I don't have interrupts and the real time clock enabled. Thanks. --wpd > -----Original Message----- > From: Gary D. Thomas [mailto:gary.thomas@mind.be] > Sent: Tuesday, January 07, 2003 4:12 PM > To: Patrick Doyle > Cc: Jonathan Larmour; eCos Discussion > Subject: RE: [ECOS] Possible typo > inhal/arm/arm9/var/current/include/hal_cache.h > > > On Tue, 2003-01-07 at 13:57, Patrick Doyle wrote: > > > > The board this was written for was never released publically, > > > > Watch this space for an announcement real soon about RedBoot > running just > > such an OMAP5910 board! > > > > > IIRC, Jesper never got the DATA cache working on that platform > > > anyway - maybe this is why :-) > > > > > > So I'm inclined to believe it's wrong, and a patch is welcome :-). > > > > > > > > If you've got the hardware you could play around! At least to verify > > that > > > > it's wrong. > > > > Hmmm... that opens up a very interesting question, how would I > "verify that > > it's wrong"? I'll have to think about a test case on the drive > home tonight > > that would verify one cache line size vs. another. > > > > I just added some test code to kernel/tests/kcache2.c which is > supposed to be able to check this sort of thing. You have to > enable it [by hand], but it might be useful. > > > > Indeed, and then please share the results! > > But, of course! > > > > --wpd > -- > .--------------------------------------------------------. > | Mind: Embedded Linux and eCos Development | > |--------------------------------------------------------| > | Gary Thomas email: gary.thomas@mind.be | > | Mind ( http://mind.be ) tel: +1 (970) 229-1963 | > | gpg: http://www.chez-thomas.org/gary/gpg_key.asc | > '--------------------------------------------------------' > -- Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos and search the list archive: http://sources.redhat.com/ml/ecos-discuss ^ permalink raw reply [flat|nested] 9+ messages in thread
* RE: [ECOS] Possible typo inhal/arm/arm9/var/current/include/hal_cache.h 2003-01-07 21:31 ` [ECOS] Possible typo inhal/arm/arm9/var/current/include/hal_cache.h Patrick Doyle @ 2003-01-07 21:36 ` Gary D. Thomas 2003-01-07 21:39 ` Jonathan Larmour 0 siblings, 1 reply; 9+ messages in thread From: Gary D. Thomas @ 2003-01-07 21:36 UTC (permalink / raw) To: Patrick Doyle; +Cc: Jonathan Larmour, eCos Discussion On Tue, 2003-01-07 at 14:28, Patrick Doyle wrote: > OK, I'll have a look at that. So far I have avoided running any of the > kernel tests because I have been focusing very intently on getting RedBoot > up and running, and have ignored anything not directly related to RedBoot. > Thus, I will feel comfortable saying at the end of the week that "I have > ported RedBoot to the OMAP", but there is still somme work to be done to > claim that I have ported eCos to the OMAP (most noticably, I have not done > any work at all with interrupts, consequently, I have not done any work at > all with the real time clock, consequently, I have not tried running any of > the "kernel specific" tests). > > Anyway, I'll look at it. Perhaps it will run fine even if I don't have > interrupts and the real time clock enabled. Thanks. > In the worst case, just take the code and run it via some other technique. It's quite easy to pull out. > --wpd > > > > -----Original Message----- > > From: Gary D. Thomas [mailto:gary.thomas@mind.be] > > Sent: Tuesday, January 07, 2003 4:12 PM > > To: Patrick Doyle > > Cc: Jonathan Larmour; eCos Discussion > > Subject: RE: [ECOS] Possible typo > > inhal/arm/arm9/var/current/include/hal_cache.h > > > > > > On Tue, 2003-01-07 at 13:57, Patrick Doyle wrote: > > > > > The board this was written for was never released publically, > > > > > > Watch this space for an announcement real soon about RedBoot > > running just > > > such an OMAP5910 board! > > > > > > > IIRC, Jesper never got the DATA cache working on that platform > > > > anyway - maybe this is why :-) > > > > > > > > So I'm inclined to believe it's wrong, and a patch is welcome :-). > > > > > > > > > > > If you've got the hardware you could play around! At least to verify > > > that > > > > > it's wrong. > > > > > > Hmmm... that opens up a very interesting question, how would I > > "verify that > > > it's wrong"? I'll have to think about a test case on the drive > > home tonight > > > that would verify one cache line size vs. another. > > > > > > > I just added some test code to kernel/tests/kcache2.c which is > > supposed to be able to check this sort of thing. You have to > > enable it [by hand], but it might be useful. > > > > > > Indeed, and then please share the results! > > > But, of course! > > > -- .--------------------------------------------------------. | Mind: Embedded Linux and eCos Development | |--------------------------------------------------------| | Gary Thomas email: gary.thomas@mind.be | | Mind ( http://mind.be ) tel: +1 (970) 229-1963 | | gpg: http://www.chez-thomas.org/gary/gpg_key.asc | '--------------------------------------------------------' -- Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos and search the list archive: http://sources.redhat.com/ml/ecos-discuss ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [ECOS] Possible typo inhal/arm/arm9/var/current/include/hal_cache.h 2003-01-07 21:36 ` Gary D. Thomas @ 2003-01-07 21:39 ` Jonathan Larmour 2003-01-08 14:31 ` [ECOS] Possible typo in hal/arm/arm9/var/current/include/hal_cache.h Patrick Doyle 0 siblings, 1 reply; 9+ messages in thread From: Jonathan Larmour @ 2003-01-07 21:39 UTC (permalink / raw) To: Patrick Doyle; +Cc: eCos Discussion Gary D. Thomas wrote: > On Tue, 2003-01-07 at 14:28, Patrick Doyle wrote: > >>OK, I'll have a look at that. So far I have avoided running any of the >>kernel tests because I have been focusing very intently on getting RedBoot >>up and running, and have ignored anything not directly related to RedBoot. >>Thus, I will feel comfortable saying at the end of the week that "I have >>ported RedBoot to the OMAP", but there is still somme work to be done to >>claim that I have ported eCos to the OMAP (most noticably, I have not done >>any work at all with interrupts, consequently, I have not done any work at >>all with the real time clock, consequently, I have not tried running any of >>the "kernel specific" tests). Patrick, one other thing you could do is ask Red Hat to release their OMAP port. I'm fairly sure that included the support required for eCOs (ints, rtc, etc.) too, not just redboot, but I could be wrong. <msalter@redhat.com> is probably the best starting point. Jifl -- eCosCentric http://www.eCosCentric.com/ <info@eCosCentric.com> --[ "You can complain because roses have thorns, or you ]-- --[ can rejoice because thorns have roses." -Lincoln ]-- Opinions==mine -- Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos and search the list archive: http://sources.redhat.com/ml/ecos-discuss ^ permalink raw reply [flat|nested] 9+ messages in thread
* RE: [ECOS] Possible typo in hal/arm/arm9/var/current/include/hal_cache.h 2003-01-07 21:39 ` Jonathan Larmour @ 2003-01-08 14:31 ` Patrick Doyle 0 siblings, 0 replies; 9+ messages in thread From: Patrick Doyle @ 2003-01-08 14:31 UTC (permalink / raw) To: Jonathan Larmour; +Cc: eCos Discussion Thanks... I'll do that right now. BTW, I didn't mean to imply that full support for eCos would not be forthcoming, it's just not on the roadmap for the quickest route to our end goal. (Getting a bootloader that we knew, such as RedBoot, was on that roadmap. Getting a full version of eCos up and running is also on the roadmap, just not as soon). --wpd > -----Original Message----- > From: Jonathan Larmour [mailto:jifl@eCosCentric.com] > Sent: Tuesday, January 07, 2003 4:38 PM > To: Patrick Doyle > Cc: eCos Discussion > Subject: Re: [ECOS] Possible typo > inhal/arm/arm9/var/current/include/hal_cache.h specific" tests). > > Patrick, one other thing you could do is ask Red Hat to release > their OMAP > port. I'm fairly sure that included the support required for eCOs (ints, > rtc, etc.) too, not just redboot, but I could be wrong. > <msalter@redhat.com> is probably the best starting point. > -- Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos and search the list archive: http://sources.redhat.com/ml/ecos-discuss ^ permalink raw reply [flat|nested] 9+ messages in thread
end of thread, other threads:[~2003-01-08 14:31 UTC | newest] Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2003-01-07 18:53 [ECOS] Possible typo in hal/arm/arm9/var/current/include/hal_cache.h Patrick Doyle 2003-01-07 20:18 ` Jonathan Larmour 2003-01-07 20:24 ` Gary D. Thomas 2003-01-07 21:00 ` Patrick Doyle 2003-01-07 21:14 ` Gary D. Thomas 2003-01-07 21:31 ` [ECOS] Possible typo inhal/arm/arm9/var/current/include/hal_cache.h Patrick Doyle 2003-01-07 21:36 ` Gary D. Thomas 2003-01-07 21:39 ` Jonathan Larmour 2003-01-08 14:31 ` [ECOS] Possible typo in hal/arm/arm9/var/current/include/hal_cache.h Patrick Doyle
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).