Searched hist:85 (Results 26 - 50 of 50) sorted by relevance
/gem5/src/mem/ | ||
H A D | noncoherent_xbar.cc | 10888:85a001f2193b Fri Jul 03 10:14:00 EDT 2015 Andreas Hansson <andreas.hansson@arm.com> mem: Delay responses in the crossbar before forwarding This patch changes how the crossbar classes deal with responses. Instead of forwarding responses directly and burdening the neighbouring modules in paying for the latency (through the pkt->headerDelay), we now queue them before sending them. The coherency protocol is not affected as requests and any snoop requests/responses are still passed on in zero time. Thus, the responses end up paying for any header delay accumulated when passing through the crossbar. Any latency incurred on the request path will be paid for on the response side, if no other module has dealt with it. As a result of this patch, responses are returned at a later point. This affects the number of outstanding transactions, and quite a few regressions see an impact in blocking due to no MSHRs, increased cache-miss latencies, etc. Going forward we should be able to use the same concept also for snoop responses, and any request that is not an express snoop. |
H A D | xbar.hh | 10888:85a001f2193b Fri Jul 03 10:14:00 EDT 2015 Andreas Hansson <andreas.hansson@arm.com> mem: Delay responses in the crossbar before forwarding This patch changes how the crossbar classes deal with responses. Instead of forwarding responses directly and burdening the neighbouring modules in paying for the latency (through the pkt->headerDelay), we now queue them before sending them. The coherency protocol is not affected as requests and any snoop requests/responses are still passed on in zero time. Thus, the responses end up paying for any header delay accumulated when passing through the crossbar. Any latency incurred on the request path will be paid for on the response side, if no other module has dealt with it. As a result of this patch, responses are returned at a later point. This affects the number of outstanding transactions, and quite a few regressions see an impact in blocking due to no MSHRs, increased cache-miss latencies, etc. Going forward we should be able to use the same concept also for snoop responses, and any request that is not an express snoop. |
H A D | coherent_xbar.hh | 10888:85a001f2193b Fri Jul 03 10:14:00 EDT 2015 Andreas Hansson <andreas.hansson@arm.com> mem: Delay responses in the crossbar before forwarding This patch changes how the crossbar classes deal with responses. Instead of forwarding responses directly and burdening the neighbouring modules in paying for the latency (through the pkt->headerDelay), we now queue them before sending them. The coherency protocol is not affected as requests and any snoop requests/responses are still passed on in zero time. Thus, the responses end up paying for any header delay accumulated when passing through the crossbar. Any latency incurred on the request path will be paid for on the response side, if no other module has dealt with it. As a result of this patch, responses are returned at a later point. This affects the number of outstanding transactions, and quite a few regressions see an impact in blocking due to no MSHRs, increased cache-miss latencies, etc. Going forward we should be able to use the same concept also for snoop responses, and any request that is not an express snoop. |
H A D | coherent_xbar.cc | 10888:85a001f2193b Fri Jul 03 10:14:00 EDT 2015 Andreas Hansson <andreas.hansson@arm.com> mem: Delay responses in the crossbar before forwarding This patch changes how the crossbar classes deal with responses. Instead of forwarding responses directly and burdening the neighbouring modules in paying for the latency (through the pkt->headerDelay), we now queue them before sending them. The coherency protocol is not affected as requests and any snoop requests/responses are still passed on in zero time. Thus, the responses end up paying for any header delay accumulated when passing through the crossbar. Any latency incurred on the request path will be paid for on the response side, if no other module has dealt with it. As a result of this patch, responses are returned at a later point. This affects the number of outstanding transactions, and quite a few regressions see an impact in blocking due to no MSHRs, increased cache-miss latencies, etc. Going forward we should be able to use the same concept also for snoop responses, and any request that is not an express snoop. |
H A D | port.cc | 5494:85c8d296c1cb Sat Jun 28 13:19:00 EDT 2008 Steve Reinhardt <stever@gmail.com> Backed out changeset 94a7bb476fca: caused memory leak. |
H A D | bridge.cc | 5494:85c8d296c1cb Sat Jun 28 13:19:00 EDT 2008 Steve Reinhardt <stever@gmail.com> Backed out changeset 94a7bb476fca: caused memory leak. |
H A D | port.hh | 5494:85c8d296c1cb Sat Jun 28 13:19:00 EDT 2008 Steve Reinhardt <stever@gmail.com> Backed out changeset 94a7bb476fca: caused memory leak. |
/gem5/src/cpu/checker/ | ||
H A D | thread_context.hh | 13693:85fa3a41014b Thu Feb 14 12:39:00 EST 2019 Giacomo Gabrielli <giacomo.gabrielli@arm.com> cpu: Add ISA* getter in Thread interface This patch is adding a ISA* getter to the TC interface Change-Id: Ib8ddc5d8fdd44e782f50a2ad15878a6bcf931e58 Reviewed-on: https://gem5-review.googlesource.com/c/16462 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com> Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com> 3548:85e64c82c522 Tue Nov 07 05:36:00 EST 2006 Gabe Black <gblack@eecs.umich.edu> Moved the switched version of kernel_stats.hh back to kern, and moved the base kernel_stats to base_kernel_stats |
/gem5/src/cpu/o3/ | ||
H A D | thread_context.hh | 13693:85fa3a41014b Thu Feb 14 12:39:00 EST 2019 Giacomo Gabrielli <giacomo.gabrielli@arm.com> cpu: Add ISA* getter in Thread interface This patch is adding a ISA* getter to the TC interface Change-Id: Ib8ddc5d8fdd44e782f50a2ad15878a6bcf931e58 Reviewed-on: https://gem5-review.googlesource.com/c/16462 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com> Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com> 3548:85e64c82c522 Tue Nov 07 05:36:00 EST 2006 Gabe Black <gblack@eecs.umich.edu> Moved the switched version of kernel_stats.hh back to kern, and moved the base kernel_stats to base_kernel_stats |
H A D | thread_context_impl.hh | 5494:85c8d296c1cb Sat Jun 28 13:19:00 EDT 2008 Steve Reinhardt <stever@gmail.com> Backed out changeset 94a7bb476fca: caused memory leak. 3548:85e64c82c522 Tue Nov 07 05:36:00 EST 2006 Gabe Black <gblack@eecs.umich.edu> Moved the switched version of kernel_stats.hh back to kern, and moved the base kernel_stats to base_kernel_stats |
H A D | lsq.hh | 5494:85c8d296c1cb Sat Jun 28 13:19:00 EDT 2008 Steve Reinhardt <stever@gmail.com> Backed out changeset 94a7bb476fca: caused memory leak. |
H A D | lsq_impl.hh | 5494:85c8d296c1cb Sat Jun 28 13:19:00 EDT 2008 Steve Reinhardt <stever@gmail.com> Backed out changeset 94a7bb476fca: caused memory leak. |
H A D | fetch.hh | 5494:85c8d296c1cb Sat Jun 28 13:19:00 EDT 2008 Steve Reinhardt <stever@gmail.com> Backed out changeset 94a7bb476fca: caused memory leak. |
H A D | fetch_impl.hh | 5494:85c8d296c1cb Sat Jun 28 13:19:00 EDT 2008 Steve Reinhardt <stever@gmail.com> Backed out changeset 94a7bb476fca: caused memory leak. |
/gem5/src/cpu/ | ||
H A D | simple_thread.cc | 5494:85c8d296c1cb Sat Jun 28 13:19:00 EDT 2008 Steve Reinhardt <stever@gmail.com> Backed out changeset 94a7bb476fca: caused memory leak. 3548:85e64c82c522 Tue Nov 07 05:36:00 EST 2006 Gabe Black <gblack@eecs.umich.edu> Moved the switched version of kernel_stats.hh back to kern, and moved the base kernel_stats to base_kernel_stats |
H A D | thread_state.cc | 5494:85c8d296c1cb Sat Jun 28 13:19:00 EDT 2008 Steve Reinhardt <stever@gmail.com> Backed out changeset 94a7bb476fca: caused memory leak. |
H A D | simple_thread.hh | 13693:85fa3a41014b Thu Feb 14 12:39:00 EST 2019 Giacomo Gabrielli <giacomo.gabrielli@arm.com> cpu: Add ISA* getter in Thread interface This patch is adding a ISA* getter to the TC interface Change-Id: Ib8ddc5d8fdd44e782f50a2ad15878a6bcf931e58 Reviewed-on: https://gem5-review.googlesource.com/c/16462 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com> Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com> 3548:85e64c82c522 Tue Nov 07 05:36:00 EST 2006 Gabe Black <gblack@eecs.umich.edu> Moved the switched version of kernel_stats.hh back to kern, and moved the base kernel_stats to base_kernel_stats |
H A D | thread_context.hh | 13693:85fa3a41014b Thu Feb 14 12:39:00 EST 2019 Giacomo Gabrielli <giacomo.gabrielli@arm.com> cpu: Add ISA* getter in Thread interface This patch is adding a ISA* getter to the TC interface Change-Id: Ib8ddc5d8fdd44e782f50a2ad15878a6bcf931e58 Reviewed-on: https://gem5-review.googlesource.com/c/16462 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com> Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com> 3548:85e64c82c522 Tue Nov 07 05:36:00 EST 2006 Gabe Black <gblack@eecs.umich.edu> Moved the switched version of kernel_stats.hh back to kern, and moved the base kernel_stats to base_kernel_stats |
H A D | thread_state.hh | 3548:85e64c82c522 Tue Nov 07 05:36:00 EST 2006 Gabe Black <gblack@eecs.umich.edu> Moved the switched version of kernel_stats.hh back to kern, and moved the base kernel_stats to base_kernel_stats |
/gem5/src/arch/arm/ | ||
H A D | tlb.hh | 7734:85a8198aa2ff Mon Nov 08 14:58:00 EST 2010 Ali Saidi <Ali.Saidi@ARM.com> ARM: Add some TLB statistics for ARM |
H A D | tlb.cc | 7734:85a8198aa2ff Mon Nov 08 14:58:00 EST 2010 Ali Saidi <Ali.Saidi@ARM.com> ARM: Add some TLB statistics for ARM |
/gem5/src/python/m5/ | ||
H A D | main.py | 2891:85ce5705650b Wed Jul 12 15:18:00 EDT 2006 Nathan Binkert <binkertn@umich.edu> Fix __file__ for scripts src/python/m5/main.py: set __file__ to the script, not the m5 binary. |
H A D | params.py | 7798:85e1847726e3 Mon Dec 20 04:20:00 EST 2010 Gabe Black <gblack@eecs.umich.edu> Params: Fix a broken error message in verifyIp. |
/gem5/src/arch/alpha/ | ||
H A D | ev5.cc | 3548:85e64c82c522 Tue Nov 07 05:36:00 EST 2006 Gabe Black <gblack@eecs.umich.edu> Moved the switched version of kernel_stats.hh back to kern, and moved the base kernel_stats to base_kernel_stats |
/gem5/src/mem/cache/ | ||
H A D | cache.hh | 5494:85c8d296c1cb Sat Jun 28 13:19:00 EDT 2008 Steve Reinhardt <stever@gmail.com> Backed out changeset 94a7bb476fca: caused memory leak. |
Completed in 466 milliseconds